diff --git a/LICENSE b/LICENSE new file mode 100644 index 0000000000000000000000000000000000000000..5e997469c03267dd816016d614230b95ab4f5f20 --- /dev/null +++ b/LICENSE @@ -0,0 +1,40 @@ +Attribution 4.0 International (CC BY 4.0) + +The ActiveVision benchmark dataset and accompanying generation pipeline are +licensed under the Creative Commons Attribution 4.0 International License. + +You are free to: + + Share — copy and redistribute the material in any medium or format + Adapt — remix, transform, and build upon the material for any purpose, + even commercially. + +Under the following terms: + + Attribution — You must give appropriate credit, provide a link to the + license, and indicate if changes were made. You may do so in any reasonable + manner, but not in any way that suggests the licensor endorses you or your + use. + + No additional restrictions — You may not apply legal terms or technological + measures that legally restrict others from doing anything the license permits. + +Notices: + + You do not have to comply with the license for elements of the material in + the public domain or where your use is permitted by an applicable exception + or limitation. + + No warranties are given. The license may not give you all of the permissions + necessary for your intended use. For example, other rights such as publicity, + privacy, or moral rights may limit how you use the material. + +Full legal code: https://creativecommons.org/licenses/by/4.0/legalcode +Human-readable summary: https://creativecommons.org/licenses/by/4.0/ + +──────────────────────────────────────────────────────────────────────────── + +Citation: + + Anonymous. ActiveVision: An Exam for Active Observers. + NeurIPS 2026 Datasets and Benchmarks Track (under review). diff --git a/README.md b/README.md new file mode 100644 index 0000000000000000000000000000000000000000..b1192e0a09cd8ae3d2761cab1453d0c5d07d5073 --- /dev/null +++ b/README.md @@ -0,0 +1,106 @@ +--- +license: cc-by-4.0 +pretty_name: ActiveVision +size_categories: + - n<1K +task_categories: + - visual-question-answering + - image-classification +language: + - en +tags: + - active-perception + - active-vision + - multimodal + - benchmark + - synthetic + - mllm-evaluation + - visual-reasoning +configs: + - config_name: default + data_files: + - split: test + path: data/manifest.json +--- + +# ActiveVision + +An exam for active observers — a benchmark for diagnosing whether multimodal large language models can iteratively look at an image during reasoning, instead of compressing it into a fixed embedding once. + +## What's in this archive + +| Path | Contents | +|---|---| +| `data/images/` | 85 photorealistic PNGs, the released benchmark images | +| `data/manifest.json` | Canonical index — one record per instance with `id`, `task`, `category`, `image`, `image_sha256`, `image_source_filename`, `question`, `answer` | +| `data/annotations/.jsonl` | Per-task verification metadata (5 records per file × 17 tasks = 85). Includes the structural ground truth used to compute each answer (region adjacency, arrow chains, traversal paths, Hausdorff distances, etc.) | +| `code///creation.py` | Seedable, deterministic generator. The released images at v0.4 are produced at `--difficulty 4`. | +| `code///creation.md` | Per-task design and anti-shortcut spec (where present). | +| `code///data.json` | Per-task definition: shared question text, answer format. | +| `code/gpt_image_prompts.json` | One gpt-image-2 image-edit prompt per task, used to re-render the matplotlib structural draft as a photorealistic variant while preserving the discriminative structure. | +| `code/scope.md` | Project specification: the three task families and the six shortcut classes the design defeats. | +| `croissant.json` | Croissant 1.0 + Croissant-RAI 1.0 metadata for this dataset. | +| `LICENSE` | CC BY 4.0. | + +## Statistics + +- **85 instances**, 17 tasks, 3 task families. +- **Distributed Scanning** (25 instances, 5 tasks): attribute_group_counting, bounded_faces_counting, counting_connected_components, counting_regions, tangled_loops. +- **Sequential Traversal** (25 instances, 5 tasks): arrow_chain, color_zone_sequence, line_intersections, maze, traverse_ordering. +- **Visual Attribute Transfer** (35 instances, 7 tasks): constellation_match_count, contour_silhouette_count, spot_the_contour_diff, spot_the_field_diff, spot_the_signal_diff, spot_the_stroke_diff, stroke_gesture_count. + +## Loading + +```python +import json, pathlib +root = pathlib.Path("data_neurips2026") +manifest = json.loads((root / "data" / "manifest.json").read_text()) +for item in manifest: + image_path = root / item["image"] + question = item["question"] + gold = item["answer"] +``` + +## Generation pipeline + +The pipeline is the artifact. Every benchmark image is produced in two deterministic stages: + +1. **Geometric draft (matplotlib).** `creation.py --seed S --difficulty 4` lays out a structural specification (region partition, arrow positions, maze graph, brush-stroke field, etc.) and computes the answer in closed form. Output: a plain matplotlib PNG. +2. **Photorealistic re-render (gpt-image-2).** The matplotlib draft is sent to OpenAI gpt-image-2 via the image-edit endpoint, with a per-task prompt from `code/gpt_image_prompts.json` that preserves silhouettes, positions, counts, and labels but replaces the surface material with a photorealistic style (stones on sand, hedge maze from above, starfield, etc.). + +The released benchmark contains only the Stage-2 images. Held-out splits with unpublished seeds and additional difficulties can be regenerated from the included generators. + +## Anti-shortcut design + +Every task is constructed to defeat six shortcut classes (full discussion in `code/scope.md`): + +1. One-shot symbolic perception, then text reasoning. +2. End-to-end CV scripting (OCR / edge / blob / vector tracing). +3. Prior leakage / answer-distribution memorisation. +4. Gestalt heuristics that bypass step-by-step tracing. +5. Memorisation of the released benchmark (mitigated because the *pipeline* is released, not just the data). +6. Tool-using zoom/crop/OCR loops. + +## Responsible AI + +See `croissant.json` for the full RAI block. Headlines: + +- **Synthetic only**: 100% synthetic. No human subjects, no PII, no real-world events. +- **Use cases**: testing and validation. **Not for training.** +- **Limitations**: small evaluation set; adversarial-by-design (not predictive of general vision-language ability); photorealistic re-renders depend on a closed-source service. +- **License**: [CC BY 4.0](https://creativecommons.org/licenses/by/4.0/). + +## Validating the Croissant file + +Before submission, validate `croissant.json` with the official Croissant validator at: + +> https://huggingface.co/spaces/JoaquinVanschoren/croissant-checker + +(Run well in advance of any submission deadline — the doc warns of heavy load near deadlines.) + +## Citation + +``` +Anonymous. ActiveVision: An Exam for Active Observers. +NeurIPS 2026 Datasets and Benchmarks Track (under review). +``` diff --git a/code/distributed_scanning/attribute_group_counting/creation.md b/code/distributed_scanning/attribute_group_counting/creation.md new file mode 100644 index 0000000000000000000000000000000000000000..51526e2ff6efeb852e66747c0f02773607e87102 --- /dev/null +++ b/code/distributed_scanning/attribute_group_counting/creation.md @@ -0,0 +1,28 @@ +# Attribute Group Counting + +## Goal + +Scatter many small visual elements across the canvas. All elements are the same color — groups are distinguished **only by shape**. Each shape is an **irregular random blob** generated via Fourier harmonics in polar coordinates. They cannot be described by simple geometric names (circle, square, etc.), forcing the model to perform visual pairing rather than textual identification. The model must scan the entire image, visually compare every element's silhouette, and count how many distinct groups have at least 2 members. + +## Element library + +- Shapes: procedurally generated irregular blobs (random Fourier harmonics, 8-14 lobes, unique per group) +- Color: uniform dark gray for all elements (no color differentiation) +- Texture: solid fill for all elements (no texture differentiation) + +A group identity is determined solely by shape. Each group has 2-5 elements, all sharing the exact same silhouette. There are no decoys (every element belongs to a group of size ≥2). + +## Question + +"How many distinct groups of matching shapes are in the image? All shapes are the same color. Two shapes belong to the same group if and only if they have exactly the same silhouette. The shapes are irregular and cannot be described by simple geometric names — you must visually compare them. Count the number of distinct shape groups and report the total." + +## Generation + +1. Pick `K` (number of groups), e.g. 4-9. +2. Generate `K` distinct random blob shapes using Fourier harmonics in polar coordinates. +3. For each group, sample size in 2-5. +4. Place all elements at random positions with minimum spacing so they don't overlap. +5. Render using matplotlib for smooth, anti-aliased outlines. +6. Store annotations. + +GT = `K`. diff --git a/code/distributed_scanning/attribute_group_counting/creation.py b/code/distributed_scanning/attribute_group_counting/creation.py new file mode 100644 index 0000000000000000000000000000000000000000..5ea6cd5967664aaade36938b602fc89d14d0d4d4 --- /dev/null +++ b/code/distributed_scanning/attribute_group_counting/creation.py @@ -0,0 +1,419 @@ +from __future__ import annotations + +import argparse +import json +import math +import random +from pathlib import Path +from typing import Dict, List, Tuple + +import matplotlib +matplotlib.use("Agg") +import matplotlib.pyplot as plt +import matplotlib.patches as mpatches +from matplotlib.path import Path as MplPath +import numpy as np + + +# --------------------------------------------------------------------------- +# Random irregular shape generation +# --------------------------------------------------------------------------- + +def generate_blob_vertices(rng: random.Random, + fourier_perturbation_amplitude: float = 0.20, + num_samples: int = 80) -> List[Tuple[float, float]]: + """Generate a random irregular closed shape as unit-radius vertices. + + ``fourier_perturbation_amplitude`` scales the harmonic amplitude budget. + At the default of 0.20 we match the original shape character; smaller + values produce blobs that are closer to a smooth near-circular profile, + making silhouettes harder to distinguish visually. + """ + # Vary structure dramatically across shapes + num_lobes = rng.randint(3, 12) + base_radius = rng.uniform(0.5, 0.8) + + # Random harmonics with wide amplitude range. Scale amplitudes by the + # perturbation factor so that smaller perturbation => gentler bumps. + amp_scale = max(fourier_perturbation_amplitude, 1e-6) / 0.20 + harmonics = [] + for k in range(2, num_lobes + 1): + amp = rng.uniform(0.08, 0.45) / (k ** rng.uniform(0.3, 0.7)) + harmonics.append((k, amp * amp_scale, rng.uniform(0, 2 * math.pi))) + + # Optionally add one dominant low-frequency lobe for variety + if rng.random() < 0.5: # unrelated coin (kept as-is) + dom_freq = rng.randint(2, 4) + dom_amp = rng.uniform(0.15, 0.35) * amp_scale + dom_phase = rng.uniform(0, 2 * math.pi) + harmonics.append((dom_freq, dom_amp, dom_phase)) + + # Build polar radius function + angles = [2 * math.pi * i / num_samples for i in range(num_samples)] + radii = [] + for a in angles: + r = base_radius + for freq, amp, phase in harmonics: + r += amp * math.sin(freq * a + phase) + radii.append(max(r, 0.12)) + + # Convert to cartesian + pts = [(r * math.cos(a), r * math.sin(a)) for r, a in zip(radii, angles)] + + # Random aspect stretch + rotation for more variety + stretch_x = rng.uniform(0.6, 1.0) + stretch_y = rng.uniform(0.6, 1.0) + rot = rng.uniform(0, 2 * math.pi) + cos_r, sin_r = math.cos(rot), math.sin(rot) + stretched = [] + for x, y in pts: + sx, sy = x * stretch_x, y * stretch_y + rx = sx * cos_r - sy * sin_r + ry = sx * sin_r + sy * cos_r + stretched.append((rx, ry)) + pts = stretched + + # Normalise so max extent = 1.0 + max_ext = max(max(abs(x), abs(y)) for x, y in pts) + if max_ext > 0: + pts = [(x / max_ext, y / max_ext) for x, y in pts] + + return pts + + +def _silhouette_distance(a: List[Tuple[float, float]], + b: List[Tuple[float, float]], + num_samples: int = 64) -> float: + """Rotation/reflection-invariant silhouette distance between two unit blobs. + + We compare the two shapes via their polar radius signatures sampled on + a common angular grid. We minimise over cyclic rotations and a reflection + to account for the shapes' arbitrary orientation. + """ + def radial_signature(pts: List[Tuple[float, float]]) -> np.ndarray: + xs = np.asarray([p[0] for p in pts]) + ys = np.asarray([p[1] for p in pts]) + # Centre at centroid + xs = xs - xs.mean() + ys = ys - ys.mean() + angs = np.arctan2(ys, xs) % (2 * np.pi) + rads = np.hypot(xs, ys) + grid = np.linspace(0, 2 * np.pi, num_samples, endpoint=False) + order = np.argsort(angs) + angs_sorted = angs[order] + rads_sorted = rads[order] + # Extend for wrap-around interpolation + ang_ext = np.concatenate([angs_sorted - 2 * np.pi, angs_sorted, + angs_sorted + 2 * np.pi]) + rad_ext = np.concatenate([rads_sorted, rads_sorted, rads_sorted]) + sig = np.interp(grid, ang_ext, rad_ext) + # Normalise + m = sig.max() + if m > 0: + sig = sig / m + return sig + + sa = radial_signature(a) + sb = radial_signature(b) + sb_rev = sb[::-1] + + best = float("inf") + for sig_b in (sb, sb_rev): + for shift in range(num_samples): + rolled = np.roll(sig_b, shift) + d = float(np.mean(np.abs(sa - rolled))) + if d < best: + best = d + return best + + +def generate_distinct_shapes(rng: random.Random, n: int, + fourier_perturbation_amplitude: float, + min_pairwise_silhouette_distance: float, + max_attempts_per_shape: int = 60, + ) -> List[List[Tuple[float, float]]]: + """Generate ``n`` blob shapes whose pairwise silhouette distances all exceed + ``min_pairwise_silhouette_distance``. + + Falls back to the best-available shape after ``max_attempts_per_shape`` + retries to avoid pathological infinite loops at tight thresholds. + """ + shapes: List[List[Tuple[float, float]]] = [] + for _ in range(n): + best_candidate = None + best_min_dist = -1.0 + for _attempt in range(max_attempts_per_shape): + verts = generate_blob_vertices(rng, fourier_perturbation_amplitude) + if not shapes: + shapes.append(verts) + break + min_d = min(_silhouette_distance(verts, s) for s in shapes) + if min_d >= min_pairwise_silhouette_distance: + shapes.append(verts) + break + if min_d > best_min_dist: + best_min_dist = min_d + best_candidate = verts + else: + # Couldn't satisfy threshold; use best candidate we found. + shapes.append(best_candidate if best_candidate is not None + else generate_blob_vertices(rng, fourier_perturbation_amplitude)) + return shapes + + +# --------------------------------------------------------------------------- +# Position sampling +# --------------------------------------------------------------------------- + +def sample_positions(rng: random.Random, n: int, width: int, height: int, + margin: int, min_dist: float, max_attempts: int = 8000) -> List[Tuple[float, float]]: + pts: List[Tuple[float, float]] = [] + for _ in range(max_attempts): + if len(pts) == n: + break + x = rng.uniform(margin, width - margin) + y = rng.uniform(margin, height - margin) + if all(math.hypot(x - px, y - py) >= min_dist for px, py in pts): + pts.append((x, y)) + return pts + + +# --------------------------------------------------------------------------- +# Replica sampling +# --------------------------------------------------------------------------- + +def _sample_replicas(rng: random.Random, d_val: int) -> int: + """Replica count from a flatter geometric: continue with prob 0.8 each + step, capped at max(2, d). P(k=1)=0.2, P(k=2)=0.16, P(k=3)=0.128, …, + with the residual mass piling at k=cap. Distribution is much more + spread-out than p=0.5 → meaningful chance of seeing replica counts + near the cap, while small counts are still common. + """ + cap = max(2, d_val) + k = 1 + while k < cap and rng.random() < 0.8: + k += 1 + return k + + +# --------------------------------------------------------------------------- +# Sample generation +# --------------------------------------------------------------------------- + +ELEMENT_COLOR = "#303030" + + +def sample_instance(rng: random.Random, width: int, height: int, + answer_lo: int, answer_hi: int, + d_val: int, + fourier_perturbation_amplitude: float, + min_pairwise_silhouette_distance: float, + radius: int) -> Dict[str, object] | None: + num_groups = rng.randint(answer_lo, answer_hi) + + blob_shapes = generate_distinct_shapes( + rng, + num_groups, + fourier_perturbation_amplitude=fourier_perturbation_amplitude, + min_pairwise_silhouette_distance=min_pairwise_silhouette_distance, + ) + + elements: List[Dict[str, object]] = [] + group_records: List[Dict[str, object]] = [] + for gid, blob in enumerate(blob_shapes): + size = _sample_replicas(rng, d_val) + for _ in range(size): + elements.append({ + "shape_id": gid, + "group": gid, + }) + group_records.append({ + "id": gid, + "shape_id": gid, + "size": size, + "shape_vertices": [[round(x, 4), round(y, 4)] for x, y in blob], + }) + + n = len(elements) + min_dist = radius * 2.4 + positions = sample_positions(rng, n, width, height, margin=radius + 6, min_dist=min_dist) + if len(positions) < n: + return None + + rng.shuffle(positions) + for el, (x, y) in zip(elements, positions): + el["x"] = round(x, 2) + el["y"] = round(y, 2) + + return { + "width": width, + "height": height, + "num_groups": num_groups, + "num_elements": n, + "answer": num_groups, + "elements": elements, + "groups": group_records, + "radius": radius, + "fourier_perturbation_amplitude": round(fourier_perturbation_amplitude, 5), + "min_pairwise_silhouette_distance": round(min_pairwise_silhouette_distance, 5), + "blob_shapes": [[[round(x, 4), round(y, 4)] for x, y in b] for b in blob_shapes], + } + + +def render_instance(out_path: Path, record: Dict[str, object]) -> None: + width = int(record["width"]) + height = int(record["height"]) + radius = float(record["radius"]) + blob_shapes = record["blob_shapes"] + + fig = plt.figure(figsize=(width / 100, height / 100), dpi=100) + ax = fig.add_axes([0, 0, 1, 1]) + ax.set_xlim(0, width) + ax.set_ylim(height, 0) + ax.axis("off") + ax.set_facecolor("#faf6ed") + + for el in record["elements"]: + cx = float(el["x"]) + cy = float(el["y"]) + blob = blob_shapes[el["shape_id"]] + + # Scale blob vertices to canvas coordinates + verts = [(cx + vx * radius, cy + vy * radius) for vx, vy in blob] + verts.append(verts[0]) + + codes = [MplPath.MOVETO] + [MplPath.LINETO] * (len(verts) - 2) + [MplPath.CLOSEPOLY] + path = MplPath(verts, codes) + patch = mpatches.PathPatch( + path, + facecolor=ELEMENT_COLOR, + edgecolor=ELEMENT_COLOR, + linewidth=1.8, + alpha=0.92, + joinstyle="round", + capstyle="round", + zorder=2, + ) + ax.add_patch(patch) + + fig.savefig(out_path, dpi=100, bbox_inches="tight", pad_inches=0) + plt.close(fig) + + +QUESTION = ( + "How many distinct shapes are in the image? " + "The image contains many small irregular shapes (blobs) scattered across the canvas. " + "All shapes are the same color. Two shapes belong to the same type if and only if " + "they have exactly the same silhouette. The shapes are irregular and cannot be described " + "by simple geometric names — you must visually compare them. " + "Some shapes may appear only once while others appear multiple times. " + "Count the total number of distinct shape types and report the count as a positive integer. " + "Provide your final answer enclosed in ... tags." +) + + +def generate_dataset(rng: random.Random, count: int, output_dir: Path, + width: int, height: int, + answer_lo: int, answer_hi: int, + d_val: int, + fourier_perturbation_amplitude: float, + min_pairwise_silhouette_distance: float, + radius: int) -> None: + images_dir = output_dir / "images" + images_dir.mkdir(parents=True, exist_ok=True) + + # Force evenly-spaced answers across [answer_lo, answer_hi]. + if count > 1: + forced_targets = [ + int(round(answer_lo + i * (answer_hi - answer_lo) / (count - 1))) + for i in range(count) + ] + else: + forced_targets = [answer_lo] + print(f"forced group counts: {forced_targets}") + + records: List[Dict[str, object]] = [] + data_records: List[Dict[str, object]] = [] + for idx in range(count): + target = forced_targets[idx] + for _ in range(2000): + rec = sample_instance( + rng, width, height, + answer_lo=target, answer_hi=target, + d_val=d_val, + fourier_perturbation_amplitude=fourier_perturbation_amplitude, + min_pairwise_silhouette_distance=min_pairwise_silhouette_distance, + radius=radius, + ) + if rec is not None and rec.get("answer") == target: + break + else: + print(f"Skip {idx} (could not hit target={target})") + continue + name = f"attribute_group_counting_{idx:05d}.png" + render_instance(images_dir / name, rec) + rec["image"] = f"images/{name}" + rec["question"] = QUESTION + records.append(rec) + data_records.append({"image": rec["image"], "question": QUESTION, "gt": rec["answer"]}) + print(f" [{idx+1}/{count}] groups={rec['answer']} elements={rec['num_elements']}") + + (output_dir / "annotations.jsonl").write_text( + "\n".join(json.dumps(r) for r in records) + "\n" + ) + (output_dir / "data.json").write_text(json.dumps(data_records, indent=4)) + + +def parse_args() -> argparse.Namespace: + p = argparse.ArgumentParser() + p.add_argument("--output-root", type=Path, required=True) + p.add_argument("--count", type=int, default=30) + p.add_argument("--width", type=int, default=900) + p.add_argument("--height", type=int, default=900) + p.add_argument("--radius", type=int, default=36) + p.add_argument("--seed", type=int, default=42) + p.add_argument("--difficulty", type=int, default=5, + help="Integer difficulty >=0; scales distinct-shape count, " + "replica geometric cap, blob perturbation, silhouette " + "separation, and canvas area.") + return p.parse_args() + + +def main() -> None: + args = parse_args() + rng = random.Random(args.seed) + + d = max(0, int(args.difficulty)) + + # Difficulty formulas. + answer_lo = 3 + answer_hi = 5 + d + fourier_perturbation_amplitude = 0.20 / (1 + 0.2 * d) + min_pairwise_silhouette_distance = 0.22 / (1 + 0.2 * d) + + # Canvas scaling: area ~ total shape instances (answer_hi * E[replicas=2]). + N_d = (5 + d) * 2 + N_0 = 5 * 2 + s = math.sqrt(max(1.0, N_d / N_0)) + args.width = int(round(args.width * s)) + args.height = int(round(args.height * s)) + + print(f"difficulty={d} answer_range=[{answer_lo},{answer_hi}] " + f"fourier_amp={fourier_perturbation_amplitude:.4f} " + f"silhouette_min={min_pairwise_silhouette_distance:.4f} " + f"canvas={args.width}x{args.height}") + + generate_dataset( + rng=rng, count=args.count, output_dir=args.output_root, + width=args.width, height=args.height, + answer_lo=answer_lo, answer_hi=answer_hi, + d_val=d, + fourier_perturbation_amplitude=fourier_perturbation_amplitude, + min_pairwise_silhouette_distance=min_pairwise_silhouette_distance, + radius=args.radius, + ) + print(f"Saved to {args.output_root}") + + +if __name__ == "__main__": + main() diff --git a/code/distributed_scanning/attribute_group_counting/data.json b/code/distributed_scanning/attribute_group_counting/data.json new file mode 100644 index 0000000000000000000000000000000000000000..a68de27f02229c6cda646b4027353ea7072c61b7 --- /dev/null +++ b/code/distributed_scanning/attribute_group_counting/data.json @@ -0,0 +1,27 @@ +[ + { + "image": "images/attribute_group_counting_00000.png", + "question": "How many distinct shapes are in the image? The image contains many small pieces scattered across the canvas. Two pieces belong to the same type if and only if they have exactly the same silhouette. Pieces may differ in surface texture or colour tone \u2014 only the silhouette outline matters. Count the total number of distinct silhouette types and report the count as a positive integer. Provide your final answer enclosed in ... tags.", + "gt": 3 + }, + { + "image": "images/attribute_group_counting_00001.png", + "question": "How many distinct shapes are in the image? The image contains many small pieces scattered across the canvas. Two pieces belong to the same type if and only if they have exactly the same silhouette. Pieces may differ in surface texture or colour tone \u2014 only the silhouette outline matters. Count the total number of distinct silhouette types and report the count as a positive integer. Provide your final answer enclosed in ... tags.", + "gt": 5 + }, + { + "image": "images/attribute_group_counting_00002.png", + "question": "How many distinct shapes are in the image? The image contains many small pieces scattered across the canvas. Two pieces belong to the same type if and only if they have exactly the same silhouette. Pieces may differ in surface texture or colour tone \u2014 only the silhouette outline matters. Count the total number of distinct silhouette types and report the count as a positive integer. Provide your final answer enclosed in ... tags.", + "gt": 6 + }, + { + "image": "images/attribute_group_counting_00003.png", + "question": "How many distinct shapes are in the image? The image contains many small pieces scattered across the canvas. Two pieces belong to the same type if and only if they have exactly the same silhouette. Pieces may differ in surface texture or colour tone \u2014 only the silhouette outline matters. Count the total number of distinct silhouette types and report the count as a positive integer. Provide your final answer enclosed in ... tags.", + "gt": 8 + }, + { + "image": "images/attribute_group_counting_00004.png", + "question": "How many distinct shapes are in the image? The image contains many small pieces scattered across the canvas. Two pieces belong to the same type if and only if they have exactly the same silhouette. Pieces may differ in surface texture or colour tone \u2014 only the silhouette outline matters. Count the total number of distinct silhouette types and report the count as a positive integer. Provide your final answer enclosed in ... tags.", + "gt": 10 + } +] \ No newline at end of file diff --git a/code/distributed_scanning/bounded_faces_counting/creation.md b/code/distributed_scanning/bounded_faces_counting/creation.md new file mode 100644 index 0000000000000000000000000000000000000000..3b787290d86851b193917e6a42aa11b90af67721 --- /dev/null +++ b/code/distributed_scanning/bounded_faces_counting/creation.md @@ -0,0 +1,30 @@ +# Bounded Faces Counting + +## Goal + +Render several closed curves (ellipses) on a canvas. The curves overlap to form a planar arrangement that subdivides the interior into one or more *bounded faces*. The model must scan the entire image and count how many distinct enclosed regions exist. + +This is a *distributed scanning* task — there is no path to follow. The model has to look at the whole arrangement and tally every enclosed region. + +## Avoiding ambiguity + +The user requested that there be no very small or very thin regions, and that curve overlaps should not create ambiguous regions. The generator enforces this by: + +- Stroking each loop with a generous thickness (visually clear). +- Rasterizing the figure and using a flood fill to find all interior connected components. +- Rejecting any sample where the smallest interior region is below a minimum area threshold (so no tiny slivers). +- Rejecting any sample where the bounding box of an interior region is unusually thin (so no needle-like regions). + +## Question + +"How many distinct enclosed regions (bounded faces) are visible in this image? An enclosed region is a maximal area that is fully surrounded by the drawn curves on every side, with no opening to the outside background. The unbounded outside area does not count. Each enclosed region should be counted exactly once, even if it is shaped irregularly. Count every enclosed region in the entire image and report the total as a positive integer." + +## Generation + +1. Pick `K` (number of loops), e.g. 3-6. +2. For each loop, sample random center, semi-axes, and rotation. +3. Stroke each loop as a thick outline on the canvas. +4. Flood-fill the background from the canvas border. +5. Find all interior connected components (regions not reached by the outer flood). +6. Reject the sample if any region has area below `min_area` or has a thin bounding box. +7. GT = number of interior components. diff --git a/code/distributed_scanning/bounded_faces_counting/creation.py b/code/distributed_scanning/bounded_faces_counting/creation.py new file mode 100644 index 0000000000000000000000000000000000000000..49d67ba72c7d08ac5b8b150b314ab2297dffff99 --- /dev/null +++ b/code/distributed_scanning/bounded_faces_counting/creation.py @@ -0,0 +1,473 @@ +from __future__ import annotations + +import argparse +import json +import math +import random +from pathlib import Path +from typing import Dict, List, Tuple + +import matplotlib +matplotlib.use("Agg") +import matplotlib.pyplot as plt +import numpy as np + + +Cell = Tuple[int, int] + + +# --------------------------------------------------------------------------- +# Geometry helpers +# --------------------------------------------------------------------------- + + +def _cross(ox: float, oy: float, px: float, py: float, qx: float, qy: float) -> float: + return (px - ox) * (qy - oy) - (py - oy) * (qx - ox) + + +def segments_intersect_properly( + ax: float, ay: float, bx: float, by: float, + cx: float, cy: float, dx: float, dy: float, +) -> bool: + """True if segment AB *properly* crosses segment CD (shared endpoints don't count).""" + d1 = _cross(cx, cy, dx, dy, ax, ay) + d2 = _cross(cx, cy, dx, dy, bx, by) + d3 = _cross(ax, ay, bx, by, cx, cy) + d4 = _cross(ax, ay, bx, by, dx, dy) + if ((d1 > 0 and d2 < 0) or (d1 < 0 and d2 > 0)) and \ + ((d3 > 0 and d4 < 0) or (d3 < 0 and d4 > 0)): + return True + return False + + +def point_seg_dist(px: float, py: float, ax: float, ay: float, bx: float, by: float) -> float: + dx = bx - ax + dy = by - ay + len_sq = dx * dx + dy * dy + if len_sq < 1e-12: + return math.hypot(px - ax, py - ay) + t = max(0.0, min(1.0, ((px - ax) * dx + (py - ay) * dy) / len_sq)) + return math.hypot(px - (ax + t * dx), py - (ay + t * dy)) + + +# --------------------------------------------------------------------------- +# Union-Find +# --------------------------------------------------------------------------- + + +class UnionFind: + def __init__(self, n: int) -> None: + self.parent = list(range(n)) + self.rank = [0] * n + self.num_sets = n + + def find(self, x: int) -> int: + while self.parent[x] != x: + self.parent[x] = self.parent[self.parent[x]] + x = self.parent[x] + return x + + def union(self, a: int, b: int) -> bool: + ra, rb = self.find(a), self.find(b) + if ra == rb: + return False + if self.rank[ra] < self.rank[rb]: + ra, rb = rb, ra + self.parent[rb] = ra + if self.rank[ra] == self.rank[rb]: + self.rank[ra] += 1 + self.num_sets -= 1 + return True + + +# --------------------------------------------------------------------------- +# Graph construction — planar, no-dot-crossing edge set +# --------------------------------------------------------------------------- + + +def place_dots( + rng: random.Random, + grid_rows: int, + grid_cols: int, + num_dots: int, + min_gap: float, + border_margin: int = 5, + max_attempts: int = 8000, +) -> List[Cell]: + cells: List[Cell] = [] + lo_r, hi_r = border_margin, grid_rows - border_margin + lo_c, hi_c = border_margin, grid_cols - border_margin + for _ in range(max_attempts): + if len(cells) == num_dots: + break + r = rng.randint(lo_r, hi_r - 1) + c = rng.randint(lo_c, hi_c - 1) + if all(math.hypot(r - er, c - ec) >= min_gap for er, ec in cells): + cells.append((r, c)) + return cells + + +def build_planar_edge_set( + dots: List[Cell], + dot_radius: float, + max_edge_len: float, +) -> List[Tuple[int, int, float]]: + n = len(dots) + candidates: List[Tuple[float, int, int]] = [] + for i in range(n): + ri, ci = dots[i] + for j in range(i + 1, n): + rj, cj = dots[j] + d = math.hypot(ri - rj, ci - cj) + if d > max_edge_len: + continue + clear = True + for k in range(n): + if k == i or k == j: + continue + if point_seg_dist(dots[k][0], dots[k][1], ri, ci, rj, cj) < dot_radius + 0.8: + clear = False + break + if clear: + candidates.append((d, i, j)) + candidates.sort() + + accepted: List[Tuple[int, int, float]] = [] + seg_coords: List[Tuple[float, float, float, float]] = [] + for dist, i, j in candidates: + ri, ci = dots[i] + rj, cj = dots[j] + crosses = False + for ax, ay, bx, by in seg_coords: + if (ri == ax and ci == ay) or (ri == bx and ci == by) or \ + (rj == ax and cj == ay) or (rj == bx and cj == by): + continue + if segments_intersect_properly(ri, ci, rj, cj, ax, ay, bx, by): + crosses = True + break + if not crosses: + accepted.append((i, j, dist)) + seg_coords.append((float(ri), float(ci), float(rj), float(cj))) + + return accepted + + +# --------------------------------------------------------------------------- +# Graph construction for bounded faces +# --------------------------------------------------------------------------- + + +def build_graph_with_faces( + rng: random.Random, + n: int, + planar_edges: List[Tuple[int, int, float]], + target_faces: int, +) -> Tuple[List[Tuple[int, int]], int, int] | None: + """Build a planar graph targeting a specific number of bounded faces. + + Strategy: + 1. Build a spanning forest from shuffled edges (connecting as many dots + as possible into one component). + 2. Each additional intra-component edge beyond the spanning forest creates + exactly one new bounded face (Euler: F_bounded = E - V + C). + 3. Add extra edges until we reach the target face count. + + Returns (edges, bounded_faces, num_components) or None if infeasible. + """ + uf = UnionFind(n) + + # Shuffle edges, biased toward shorter ones + mid = len(planar_edges) // 2 + short = list(planar_edges[:mid]) + long = list(planar_edges[mid:]) + rng.shuffle(short) + rng.shuffle(long) + shuffled = short + long + + # Phase 1: build spanning forest (connect everything into one component ideally) + tree_edges: List[Tuple[int, int]] = [] + extra_edges: List[Tuple[int, int]] = [] + + for i, j, d in shuffled: + if uf.find(i) != uf.find(j): + uf.union(i, j) + tree_edges.append((i, j)) + else: + extra_edges.append((i, j)) + + num_components = uf.num_sets + + # Phase 2: add extra edges to create faces + # bounded_faces = E - V + C = (tree + extra) - V + C + # spanning forest has V - C edges, so faces = num_extra_added + rng.shuffle(extra_edges) + + if len(extra_edges) < target_faces: + return None + + selected_edges = tree_edges + extra_edges[:target_faces] + bounded_faces = target_faces + + return selected_edges, bounded_faces, num_components + + +# --------------------------------------------------------------------------- +# Instance sampling +# --------------------------------------------------------------------------- + +QUESTION = ( + "How many distinct enclosed regions (bounded faces) are visible in this image? " + "An enclosed region is a maximal area that is fully surrounded by the drawn " + "line segments on every side, with no opening to the outside background. The " + "unbounded outside area does not count as an enclosed region. Each enclosed " + "region should be counted exactly once, regardless of its shape. Count every " + "enclosed region in the entire image and report the total as a positive integer. " + "Provide your final answer enclosed in ... tags." +) + + +def sample_instance( + rng: random.Random, + width: int, + height: int, + grid_rows: int, + grid_cols: int, + min_faces: int, + max_faces: int, + num_dots_min: int, + num_dots_max: int, + min_gap: float, + dot_radius: float, + max_edge_len: float, +) -> Dict[str, object] | None: + target_faces = rng.randint(min_faces, max_faces) + num_dots = rng.randint(num_dots_min, num_dots_max) + + dots = place_dots(rng, grid_rows, grid_cols, num_dots, min_gap) + if len(dots) < 10: + return None + + planar_edges = build_planar_edge_set(dots, dot_radius, max_edge_len) + + result = build_graph_with_faces(rng, len(dots), planar_edges, target_faces) + if result is None: + return None + + edges, bounded_faces, num_components = result + + if bounded_faces < min_faces: + return None + + margin = int(min(width, height) * 0.10) + square_size = min(width, height) - 2 * margin + square_left = (width - square_size) / 2.0 + square_top = (height - square_size) / 2.0 + + return { + "width": width, + "height": height, + "grid_rows": grid_rows, + "grid_cols": grid_cols, + "square_left": round(square_left, 2), + "square_top": round(square_top, 2), + "square_size": round(square_size, 2), + "num_dots": len(dots), + "num_edges": len(edges), + "num_components": num_components, + "question": QUESTION, + "answer": bounded_faces, + "dots": [[r, c] for r, c in dots], + "edges": [[i, j] for i, j in edges], + "dot_radius": dot_radius, + } + + +# --------------------------------------------------------------------------- +# Rendering (matplotlib — smooth anti-aliased output) +# --------------------------------------------------------------------------- + +LINE_COLOR = "#2f2f2f" +DOT_COLOR = "#1d1916" + + +def render_instance(out_path: Path, record: Dict[str, object], noise_seed: int = 0) -> None: + width = int(record["width"]) + height = int(record["height"]) + grid_rows = int(record["grid_rows"]) + grid_cols = int(record["grid_cols"]) + square_left = float(record["square_left"]) + square_top = float(record["square_top"]) + square_size = float(record["square_size"]) + dots: List[List[int]] = record["dots"] # type: ignore[assignment] + edges: List[List[int]] = record["edges"] # type: ignore[assignment] + dot_radius_grid = float(record["dot_radius"]) + + cell_w = square_size / grid_cols + cell_h = square_size / grid_rows + + def to_pixel(r: float, c: float) -> Tuple[float, float]: + px = square_left + (c + 0.5) * cell_w + py = square_top + (r + 0.5) * cell_h + return px, py + + pixel_dot_radius = dot_radius_grid * min(cell_w, cell_h) * 0.5 + edge_thickness = max(1.5, pixel_dot_radius * 0.3) + + fig = plt.figure(figsize=(width / 100, height / 100), dpi=100) + ax = fig.add_axes([0, 0, 1, 1]) + ax.set_xlim(0, width) + ax.set_ylim(height, 0) + ax.axis("off") + ax.set_facecolor("#f8f6f0") + + # Subtle noise background + nrng = np.random.default_rng(noise_seed) + noise = nrng.normal(0.0, 1.0, size=(height, width)) + noise = (noise - noise.min()) / max(noise.max() - noise.min(), 1e-6) + ax.imshow(noise, cmap="Greys", alpha=0.05, extent=(0, width, height, 0), + interpolation="bilinear") + + # White square background + ax.fill_between( + [square_left, square_left + square_size], + [square_top, square_top], + [square_top + square_size, square_top + square_size], + color="#fffdf8", zorder=0.5, + ) + + # Border + bx = [square_left, square_left + square_size, square_left + square_size, square_left, square_left] + by = [square_top, square_top, square_top + square_size, square_top + square_size, square_top] + ax.plot(bx, by, color="#2d2720", linewidth=2.0, solid_capstyle="round", zorder=1.0) + + # Plain (v4_plain): solid edges. No dashed-line anti-shortcut. + for i, j in edges: + px1, py1 = to_pixel(dots[i][0], dots[i][1]) + px2, py2 = to_pixel(dots[j][0], dots[j][1]) + ax.plot([px1, px2], [py1, py2], + color=LINE_COLOR, linewidth=edge_thickness, + solid_capstyle="round", alpha=0.92, zorder=2.0) + + # Dots on top + for r, c in dots: + px, py = to_pixel(r, c) + circle = plt.Circle((px, py), pixel_dot_radius, color=DOT_COLOR, zorder=3.0) + ax.add_patch(circle) + + fig.savefig(out_path, dpi=100, bbox_inches="tight", pad_inches=0) + plt.close(fig) + + +# --------------------------------------------------------------------------- +# Dataset generation +# --------------------------------------------------------------------------- + + +def main() -> None: + parser = argparse.ArgumentParser() + parser.add_argument("--output-root", type=Path, required=True) + parser.add_argument("--count", type=int, default=30) + parser.add_argument("--seed", type=int, default=42) + parser.add_argument("--width", type=int, default=1024) + parser.add_argument("--height", type=int, default=1024) + parser.add_argument("--grid-rows", type=int, default=100) + parser.add_argument("--grid-cols", type=int, default=100) + parser.add_argument("--min-faces", type=int, default=4) + parser.add_argument("--max-faces", type=int, default=15) + parser.add_argument("--num-dots-min", type=int, default=60) + parser.add_argument("--num-dots-max", type=int, default=120) + parser.add_argument("--min-gap", type=float, default=5.0) + parser.add_argument("--dot-radius", type=float, default=1.5) + parser.add_argument("--max-edge-len", type=float, default=25.0) + parser.add_argument("--difficulty", type=int, default=5, + help="Integer difficulty >=0; scales faces and dot count.") + args = parser.parse_args() + + d = max(0, int(args.difficulty)) + # Difficulty scaling per spec + args.min_faces = 5 + args.max_faces = 5 + 2 * d + args.num_dots_min = 10 * d + args.num_dots_max = 20 + 10 * d + base_max_edge_len = args.max_edge_len + args.max_edge_len = base_max_edge_len / (1.0 + 0.08 * d) + + # Canvas scaling based on num_dots_max growth + N_d = 20 + 10 * d + N_0 = 20 + s = math.sqrt(max(1.0, N_d / N_0)) + args.width = int(round(args.width * s)) + args.height = int(round(args.height * s)) + + out_root: Path = args.output_root + img_dir = out_root / "images" + img_dir.mkdir(parents=True, exist_ok=True) + ann_path = out_root / "annotations.jsonl" + + rng = random.Random(args.seed) + records = [] + # Force evenly-spaced answers across [min_faces, max_faces]. + if args.count > 1: + forced_targets = [ + int(round(args.min_faces + i * (args.max_faces - args.min_faces) / (args.count - 1))) + for i in range(args.count) + ] + else: + forced_targets = [args.min_faces] + print(f"forced face counts: {forced_targets}") + with ann_path.open("w") as f: + for i in range(args.count): + sub_seed = rng.randint(0, 2**31 - 1) + tgt = forced_targets[i] + for _ in range(2000): + record = sample_instance( + rng=rng, + width=args.width, + height=args.height, + grid_rows=args.grid_rows, + grid_cols=args.grid_cols, + min_faces=tgt, + max_faces=tgt, + num_dots_min=args.num_dots_min, + num_dots_max=args.num_dots_max, + min_gap=args.min_gap, + dot_radius=args.dot_radius, + max_edge_len=args.max_edge_len, + ) + if record is not None and record.get("answer") == tgt: + break + else: + print(f" [{i+1}/{args.count}] SKIP (failed to generate)") + continue + + name = f"bounded_faces_counting_{i:05d}.png" + render_instance(img_dir / name, record, noise_seed=sub_seed) + print(f" [{i+1}/{args.count}] faces={record['answer']} dots={record['num_dots']} edges={record['num_edges']}") + + rec = { + "image": f"images/{name}", + "question": QUESTION, + "answer": record["answer"], + "metadata": { + "bounded_faces": record["answer"], + "num_dots": record["num_dots"], + "num_edges": record["num_edges"], + "num_components": record["num_components"], + "seed": sub_seed, + }, + } + f.write(json.dumps(rec) + "\n") + records.append(rec) + + data_json = { + "task": "bounded_faces_counting", + "category": "distributed_scanning", + "count": len(records), + "items": records, + } + (out_root / "data.json").write_text(json.dumps(data_json, indent=2)) + print(f"Saved to {out_root}") + + +if __name__ == "__main__": + main() diff --git a/code/distributed_scanning/bounded_faces_counting/data.json b/code/distributed_scanning/bounded_faces_counting/data.json new file mode 100644 index 0000000000000000000000000000000000000000..dbbb88fe52db831d40044c158ede7e385d6fb8b4 --- /dev/null +++ b/code/distributed_scanning/bounded_faces_counting/data.json @@ -0,0 +1,67 @@ +{ + "task": "bounded_faces_counting", + "category": "distributed_scanning", + "count": 5, + "items": [ + { + "image": "images/bounded_faces_counting_00000.png", + "question": "How many distinct enclosed regions (bounded faces) are visible in this image? An enclosed region is a maximal area that is fully surrounded by drawn lines or strands on every side, with no opening to the outside background. The unbounded outside area does not count. Each enclosed region should be counted exactly once. Provide your final answer enclosed in ... tags.", + "answer": 5, + "metadata": { + "bounded_faces": 5, + "num_dots": 58, + "num_edges": 62, + "num_components": 1, + "seed": 478163327 + } + }, + { + "image": "images/bounded_faces_counting_00001.png", + "question": "How many distinct enclosed regions (bounded faces) are visible in this image? An enclosed region is a maximal area that is fully surrounded by drawn lines or strands on every side, with no opening to the outside background. The unbounded outside area does not count. Each enclosed region should be counted exactly once. Provide your final answer enclosed in ... tags.", + "answer": 8, + "metadata": { + "bounded_faces": 8, + "num_dots": 69, + "num_edges": 76, + "num_components": 1, + "seed": 798112150 + } + }, + { + "image": "images/bounded_faces_counting_00002.png", + "question": "How many distinct enclosed regions (bounded faces) are visible in this image? An enclosed region is a maximal area that is fully surrounded by drawn lines or strands on every side, with no opening to the outside background. The unbounded outside area does not count. Each enclosed region should be counted exactly once. Provide your final answer enclosed in ... tags.", + "answer": 10, + "metadata": { + "bounded_faces": 10, + "num_dots": 68, + "num_edges": 77, + "num_components": 1, + "seed": 110287971 + } + }, + { + "image": "images/bounded_faces_counting_00003.png", + "question": "How many distinct enclosed regions (bounded faces) are visible in this image? An enclosed region is a maximal area that is fully surrounded by drawn lines or strands on every side, with no opening to the outside background. The unbounded outside area does not count. Each enclosed region should be counted exactly once. Provide your final answer enclosed in ... tags.", + "answer": 12, + "metadata": { + "bounded_faces": 12, + "num_dots": 60, + "num_edges": 70, + "num_components": 2, + "seed": 1411178445 + } + }, + { + "image": "images/bounded_faces_counting_00004.png", + "question": "How many distinct enclosed regions (bounded faces) are visible in this image? An enclosed region is a maximal area that is fully surrounded by drawn lines or strands on every side, with no opening to the outside background. The unbounded outside area does not count. Each enclosed region should be counted exactly once. Provide your final answer enclosed in ... tags.", + "answer": 15, + "metadata": { + "bounded_faces": 15, + "num_dots": 61, + "num_edges": 75, + "num_components": 1, + "seed": 963124353 + } + } + ] +} \ No newline at end of file diff --git a/code/distributed_scanning/counting_connected_components/creation.md b/code/distributed_scanning/counting_connected_components/creation.md new file mode 100644 index 0000000000000000000000000000000000000000..af19104e348811997010f977253cd91dd04a547d --- /dev/null +++ b/code/distributed_scanning/counting_connected_components/creation.md @@ -0,0 +1,13 @@ +# Counting Connected Components Dataset Generation + +## Goal + +Generate images of a large grid (e.g., 80x80 to 150x150) where scattered dots are connected by lines to form several distinct connected components — groups of dots linked together by drawn edges. + +The diameter of each dot should be about 3 unit lengths. Enforce a minimum gap of at least distance of 5 unit length between any two dots so they are unambiguously separated. + +Each component does not necessary stay in the same region. Dots from different components can intersect with each other, but are connected with other dots under the same components. + +The connection edge should not get across any other dots to avoid ambiguity. + +Render PNG files and the question is asking how many connected components are there. \ No newline at end of file diff --git a/code/distributed_scanning/counting_connected_components/creation.py b/code/distributed_scanning/counting_connected_components/creation.py new file mode 100644 index 0000000000000000000000000000000000000000..9f1a1e10992905afb2ef54a3d0bdf2cb268d1970 --- /dev/null +++ b/code/distributed_scanning/counting_connected_components/creation.py @@ -0,0 +1,574 @@ +from __future__ import annotations + +import argparse +import json +import math +import random +from pathlib import Path +from typing import Dict, List, Tuple + +import matplotlib +matplotlib.use("Agg") +import matplotlib.pyplot as plt +import numpy as np + + +Cell = Tuple[int, int] + + +# --------------------------------------------------------------------------- +# Geometry helpers +# --------------------------------------------------------------------------- + + +def _cross(ox: float, oy: float, px: float, py: float, qx: float, qy: float) -> float: + return (px - ox) * (qy - oy) - (py - oy) * (qx - ox) + + +def segments_intersect_properly( + ax: float, ay: float, bx: float, by: float, + cx: float, cy: float, dx: float, dy: float, +) -> bool: + """True if segment AB *properly* crosses segment CD (shared endpoints don't count).""" + d1 = _cross(cx, cy, dx, dy, ax, ay) + d2 = _cross(cx, cy, dx, dy, bx, by) + d3 = _cross(ax, ay, bx, by, cx, cy) + d4 = _cross(ax, ay, bx, by, dx, dy) + if ((d1 > 0 and d2 < 0) or (d1 < 0 and d2 > 0)) and \ + ((d3 > 0 and d4 < 0) or (d3 < 0 and d4 > 0)): + return True + return False + + +def point_seg_dist(px: float, py: float, ax: float, ay: float, bx: float, by: float) -> float: + dx = bx - ax + dy = by - ay + len_sq = dx * dx + dy * dy + if len_sq < 1e-12: + return math.hypot(px - ax, py - ay) + t = max(0.0, min(1.0, ((px - ax) * dx + (py - ay) * dy) / len_sq)) + return math.hypot(px - (ax + t * dx), py - (ay + t * dy)) + + +# --------------------------------------------------------------------------- +# Union-Find +# --------------------------------------------------------------------------- + + +class UnionFind: + def __init__(self, n: int) -> None: + self.parent = list(range(n)) + self.rank = [0] * n + self.size = [1] * n + self.num_sets = n + + def find(self, x: int) -> int: + while self.parent[x] != x: + self.parent[x] = self.parent[self.parent[x]] + x = self.parent[x] + return x + + def set_size(self, x: int) -> int: + return self.size[self.find(x)] + + def union(self, a: int, b: int) -> bool: + ra, rb = self.find(a), self.find(b) + if ra == rb: + return False + if self.rank[ra] < self.rank[rb]: + ra, rb = rb, ra + self.parent[rb] = ra + self.size[ra] += self.size[rb] + if self.rank[ra] == self.rank[rb]: + self.rank[ra] += 1 + self.num_sets -= 1 + return True + + +# --------------------------------------------------------------------------- +# Graph construction — planar, no-dot-crossing edge set +# --------------------------------------------------------------------------- + + +def place_dots( + rng: random.Random, + grid_rows: int, + grid_cols: int, + num_dots: int, + min_gap: float, + border_margin: int = 5, + max_attempts: int = 8000, +) -> List[Cell]: + cells: List[Cell] = [] + lo_r, hi_r = border_margin, grid_rows - border_margin + lo_c, hi_c = border_margin, grid_cols - border_margin + for _ in range(max_attempts): + if len(cells) == num_dots: + break + r = rng.randint(lo_r, hi_r - 1) + c = rng.randint(lo_c, hi_c - 1) + if all(math.hypot(r - er, c - ec) >= min_gap for er, ec in cells): + cells.append((r, c)) + return cells + + +def build_planar_edge_set( + dots: List[Cell], + dot_radius: float, + max_edge_len: float, +) -> List[Tuple[int, int, float]]: + """ + Build a set of edges that: + 1. Are shorter than max_edge_len + 2. Don't pass within dot_radius of any other dot + 3. Don't cross each other (planar) + Returns list of (i, j, dist) sorted by dist, with the planar filter applied. + """ + n = len(dots) + + # Step 1: candidate edges sorted by length, filtered by dot clearance + candidates: List[Tuple[float, int, int]] = [] + for i in range(n): + ri, ci = dots[i] + for j in range(i + 1, n): + rj, cj = dots[j] + d = math.hypot(ri - rj, ci - cj) + if d > max_edge_len: + continue + # Check this edge doesn't pass near any other dot + clear = True + for k in range(n): + if k == i or k == j: + continue + if point_seg_dist(dots[k][0], dots[k][1], ri, ci, rj, cj) < dot_radius + 0.8: + clear = False + break + if clear: + candidates.append((d, i, j)) + candidates.sort() + + # Step 2: greedily add edges, skip if they cross an already-added edge + accepted: List[Tuple[int, int, float]] = [] + # For fast crossing checks, store segment coords + seg_coords: List[Tuple[float, float, float, float]] = [] + + for dist, i, j in candidates: + ri, ci = dots[i] + rj, cj = dots[j] + crosses = False + for ax, ay, bx, by in seg_coords: + # Skip if shared endpoint + if (ri == ax and ci == ay) or (ri == bx and ci == by) or \ + (rj == ax and cj == ay) or (rj == bx and cj == by): + continue + if segments_intersect_properly(ri, ci, rj, cj, ax, ay, bx, by): + crosses = True + break + if not crosses: + accepted.append((i, j, dist)) + seg_coords.append((float(ri), float(ci), float(rj), float(cj))) + + return accepted + + +def build_spanning_forest( + rng: random.Random, + n: int, + planar_edges: List[Tuple[int, int, float]], + num_components: int, +) -> Tuple[List[Tuple[int, int]], List[List[int]]]: + """ + From the planar edge set, build a spanning forest with exactly + num_components trees using Union-Find. + + Strategy: shuffle edges, greedily merge until we reach the target + number of components. Then collect extra intra-component edges. + """ + uf = UnionFind(n) + + # We need to reduce n sets down to num_components, so we need n - num_components merges. + target_merges = n - num_components + + # Cap: no single component should exceed ~2x the ideal even share. + max_component_size = max(4, (n // num_components) * 2) + + # Shuffle edges but bias toward shorter ones: split into short/long halves, + # shuffle each, concatenate. + mid = len(planar_edges) // 2 + short = list(planar_edges[:mid]) + long = list(planar_edges[mid:]) + rng.shuffle(short) + rng.shuffle(long) + shuffled = short + long + + tree_edges: List[Tuple[int, int]] = [] + deferred: List[Tuple[int, int, float]] = [] + extra_edges: List[Tuple[int, int]] = [] + + for i, j, d in shuffled: + if uf.find(i) != uf.find(j): + if len(tree_edges) < target_merges: + merged_size = uf.set_size(i) + uf.set_size(j) + if merged_size <= max_component_size: + uf.union(i, j) + tree_edges.append((i, j)) + else: + deferred.append((i, j, d)) + else: + extra_edges.append((i, j)) + else: + extra_edges.append((i, j)) + + # Second pass: use deferred edges if we still need merges + for i, j, _d in deferred: + if len(tree_edges) >= target_merges: + break + if uf.find(i) != uf.find(j): + uf.union(i, j) + tree_edges.append((i, j)) + + # Add some extra intra-component edges for visual richness + # Only pick edges where both endpoints are already in the same component + intra_edges = [(i, j) for i, j in extra_edges if uf.find(i) == uf.find(j)] + rng.shuffle(intra_edges) + bonus = min(len(intra_edges), max(n // 6, 3)) + tree_edges.extend(intra_edges[:bonus]) + + # Build component membership + comp_map: Dict[int, List[int]] = {} + for node in range(n): + root = uf.find(node) + comp_map.setdefault(root, []).append(node) + components = list(comp_map.values()) + + return tree_edges, components + + +# --------------------------------------------------------------------------- +# Instance sampling +# --------------------------------------------------------------------------- + + +def sample_instance( + rng: random.Random, + width: int, + height: int, + grid_rows: int, + grid_cols: int, + min_components: int, + max_components: int, + num_dots_min: int, + num_dots_max: int, + min_gap: float, + dot_radius: float, + max_edge_len: float, + close_strand_tolerance: float = 0.0, +) -> Dict[str, object] | None: + num_components = rng.randint(min_components, max_components) + num_dots = rng.randint(max(num_dots_min, num_components * 2), num_dots_max) + + dots = place_dots(rng, grid_rows, grid_cols, num_dots, min_gap) + if len(dots) < num_components * 2: + return None + + planar_edges = build_planar_edge_set(dots, dot_radius, max_edge_len) + + # Check we have enough edges to connect dots into num_components trees + # (need at least len(dots) - num_components edges in a spanning forest) + if len(planar_edges) < len(dots) - num_components: + return None + + edges, components = build_spanning_forest(rng, len(dots), planar_edges, num_components) + + # Reject if any component has fewer than 2 dots + if any(len(c) < 2 for c in components): + return None + + actual_components = len(components) + + if actual_components < min_components: + return None + + # Enforce close_strand_tolerance: dots from different components must be + # at least this many grid units apart (visual separation). + if close_strand_tolerance > 0: + node_to_comp = {} + for ci, comp in enumerate(components): + for node in comp: + node_to_comp[node] = ci + for a in range(len(dots)): + ra, ca = dots[a] + for b in range(a + 1, len(dots)): + if node_to_comp[a] == node_to_comp[b]: + continue + rb, cb = dots[b] + if math.hypot(ra - rb, ca - cb) < close_strand_tolerance: + return None + + margin = int(min(width, height) * 0.10) + square_size = min(width, height) - 2 * margin + square_left = (width - square_size) / 2.0 + square_top = (height - square_size) / 2.0 + + return { + "width": width, + "height": height, + "grid_rows": grid_rows, + "grid_cols": grid_cols, + "square_left": round(square_left, 2), + "square_top": round(square_top, 2), + "square_size": round(square_size, 2), + "num_components": actual_components, + "num_dots": len(dots), + "question": ( + "How many connected components are there in the image? " + "A connected component is a maximal group of dots such that any two dots " + "in the group are linked by a path of one or more drawn line segments " + "(directly or through other dots in the same group). " + "Every component contains at least two dots. " + "Two dots that are not linked by any chain of line segments " + "belong to different components, even if they appear visually close. " + "Count every connected component and report the total. " + "Provide your final answer enclosed in ... tags." + ), + "answer": actual_components, + "dots": [[r, c] for r, c in dots], + "components": components, + "edges": [[i, j] for i, j in edges], + "dot_radius": dot_radius, + } + + +# --------------------------------------------------------------------------- +# Rendering (matplotlib — smooth anti-aliased output) +# --------------------------------------------------------------------------- + +LINE_COLOR = "#2f2f2f" +DOT_COLOR = "#1d1916" + + +def render_instance(out_path: Path, record: Dict[str, object], noise_seed: int = 0) -> None: + width = int(record["width"]) + height = int(record["height"]) + grid_rows = int(record["grid_rows"]) + grid_cols = int(record["grid_cols"]) + square_left = float(record["square_left"]) + square_top = float(record["square_top"]) + square_size = float(record["square_size"]) + dots: List[List[int]] = record["dots"] # type: ignore[assignment] + edges: List[List[int]] = record["edges"] # type: ignore[assignment] + dot_radius_grid = float(record["dot_radius"]) + + cell_w = square_size / grid_cols + cell_h = square_size / grid_rows + + def to_pixel(r: float, c: float) -> Tuple[float, float]: + px = square_left + (c + 0.5) * cell_w + py = square_top + (r + 0.5) * cell_h + return px, py + + pixel_dot_radius = dot_radius_grid * min(cell_w, cell_h) * 0.5 + edge_thickness = max(1.5, pixel_dot_radius * 0.3) + + fig = plt.figure(figsize=(width / 100, height / 100), dpi=100) + ax = fig.add_axes([0, 0, 1, 1]) + ax.set_xlim(0, width) + ax.set_ylim(height, 0) + ax.axis("off") + ax.set_facecolor("#f8f6f0") + + # Subtle noise background + nrng = np.random.default_rng(noise_seed) + noise = nrng.normal(0.0, 1.0, size=(height, width)) + noise = (noise - noise.min()) / max(noise.max() - noise.min(), 1e-6) + ax.imshow(noise, cmap="Greys", alpha=0.05, extent=(0, width, height, 0), + interpolation="bilinear") + + # White square background + ax.fill_between( + [square_left, square_left + square_size], + [square_top, square_top], + [square_top + square_size, square_top + square_size], + color="#fffdf8", zorder=0.5, + ) + + # Border + border_lw = 2.0 + bx = [square_left, square_left + square_size, square_left + square_size, square_left, square_left] + by = [square_top, square_top, square_top + square_size, square_top + square_size, square_top] + ax.plot(bx, by, color="#2d2720", linewidth=border_lw, solid_capstyle="round", zorder=1.0) + + # Plain (v4_plain): solid edges. No dashed-line anti-shortcut. + for i, j in edges: + px1, py1 = to_pixel(dots[i][0], dots[i][1]) + px2, py2 = to_pixel(dots[j][0], dots[j][1]) + ax.plot([px1, px2], [py1, py2], + color=LINE_COLOR, linewidth=edge_thickness, + solid_capstyle="round", alpha=0.92, zorder=2.0) + + # Dots on top + for r, c in dots: + px, py = to_pixel(r, c) + circle = plt.Circle((px, py), pixel_dot_radius, color=DOT_COLOR, zorder=3.0) + ax.add_patch(circle) + + fig.savefig(out_path, dpi=100, bbox_inches="tight", pad_inches=0) + plt.close(fig) + + +# --------------------------------------------------------------------------- +# Dataset generation +# --------------------------------------------------------------------------- + + +def ensure_output_dir(root: Path) -> Tuple[Path, Path]: + root.mkdir(parents=True, exist_ok=True) + images_dir = root / "images" + images_dir.mkdir(exist_ok=True) + return root, images_dir + + +def generate_dataset( + rng: random.Random, + count: int, + output_dir: Path, + images_dir: Path, + width: int, + height: int, + grid_rows: int, + grid_cols: int, + min_components: int, + max_components: int, + num_dots_min: int, + num_dots_max: int, + min_gap: float, + dot_radius: float, + max_edge_len: float, + close_strand_tolerance: float = 0.0, +) -> None: + records: List[Dict[str, object]] = [] + data_records: List[Dict[str, object]] = [] + + # Force evenly-spaced answers across [min_components, max_components]. + if count > 1: + forced_targets = [ + int(round(min_components + i * (max_components - min_components) / (count - 1))) + for i in range(count) + ] + else: + forced_targets = [min_components] + print(f"forced component counts: {forced_targets}") + + for idx in range(count): + sub_seed = rng.randint(0, 2**31 - 1) + tgt = forced_targets[idx] + for _ in range(2000): + record = sample_instance( + rng=rng, + width=width, + height=height, + grid_rows=grid_rows, + grid_cols=grid_cols, + min_components=tgt, + max_components=tgt, + num_dots_min=num_dots_min, + num_dots_max=num_dots_max, + min_gap=min_gap, + dot_radius=dot_radius, + max_edge_len=max_edge_len, + close_strand_tolerance=close_strand_tolerance, + ) + if record is not None and record.get("answer") == tgt: + break + else: + print(f"Warning: could not generate sample {idx}, skipping") + continue + + image_name = f"counting_connected_components_{idx:05d}.png" + render_instance(images_dir / image_name, record, noise_seed=sub_seed) + record["image"] = f"images/{image_name}" + records.append(record) + data_records.append({ + "image": record["image"], + "question": record["question"], + "answer": record["answer"], + }) + print(f" [{idx+1}/{count}] components={record['answer']} dots={record['num_dots']}") + + with (output_dir / "annotations.jsonl").open("w", encoding="utf-8") as fh: + for record in records: + fh.write(json.dumps(record) + "\n") + + data_json = { + "task": "counting_connected_components", + "category": "distributed_scanning", + "count": len(data_records), + "items": data_records, + } + with (output_dir / "data.json").open("w", encoding="utf-8") as fh: + json.dump(data_json, fh, indent=2) + + +def parse_args() -> argparse.Namespace: + parser = argparse.ArgumentParser(description="Generate a counting-connected-components dataset.") + parser.add_argument("--output-root", type=Path, required=True, help="Dataset root directory.") + parser.add_argument("--count", type=int, default=30) + parser.add_argument("--width", type=int, default=1024) + parser.add_argument("--height", type=int, default=1024) + parser.add_argument("--grid-rows", type=int, default=100) + parser.add_argument("--grid-cols", type=int, default=100) + parser.add_argument("--min-components", type=int, default=4) + parser.add_argument("--max-components", type=int, default=12) + parser.add_argument("--num-dots-min", type=int, default=60) + parser.add_argument("--num-dots-max", type=int, default=120) + parser.add_argument("--min-gap", type=float, default=5.0) + parser.add_argument("--dot-radius", type=float, default=1.5) + parser.add_argument("--max-edge-len", type=float, default=25.0) + parser.add_argument("--seed", type=int, default=42) + parser.add_argument("--difficulty", type=int, default=5, + help="Integer difficulty >=0; scales components and dot count.") + return parser.parse_args() + + +def main() -> None: + args = parse_args() + rng = random.Random(args.seed) + output_dir, images_dir = ensure_output_dir(args.output_root) + d = max(0, int(args.difficulty)) + # Difficulty scaling per spec + min_components = 10 + max_components = 10 + 2 * d + num_dots_min = 40 + num_dots_max = 40 + 20 * d + close_strand_tolerance = float(max(3, 8 - d)) + + # Canvas scaling based on num_dots_max growth + N_d = 20 + 10 * d + N_0 = 20 + s = math.sqrt(max(1.0, N_d / N_0)) + args.width = int(round(args.width * s)) + args.height = int(round(args.height * s)) + + generate_dataset( + rng=rng, + count=args.count, + output_dir=output_dir, + images_dir=images_dir, + width=args.width, + height=args.height, + grid_rows=args.grid_rows, + grid_cols=args.grid_cols, + min_components=min_components, + max_components=max_components, + num_dots_min=num_dots_min, + num_dots_max=num_dots_max, + min_gap=args.min_gap, + dot_radius=args.dot_radius, + max_edge_len=args.max_edge_len, + close_strand_tolerance=close_strand_tolerance, + ) + print(f"Saved dataset to {args.output_root}") + + +if __name__ == "__main__": + main() diff --git a/code/distributed_scanning/counting_connected_components/data.json b/code/distributed_scanning/counting_connected_components/data.json new file mode 100644 index 0000000000000000000000000000000000000000..aa66e5294aaed85f9bcbf13f06a4d1a7a81d8b67 --- /dev/null +++ b/code/distributed_scanning/counting_connected_components/data.json @@ -0,0 +1,32 @@ +{ + "task": "counting_connected_components", + "category": "distributed_scanning", + "count": 5, + "items": [ + { + "image": "images/counting_connected_components_00000.png", + "question": "How many connected components are there in the image? A connected component is a maximal group of anchor points such that any two are linked by a path of one or more strands (directly or through other anchors in the same group). Anchors not linked by any chain belong to different components, even if they appear visually close. Count every connected component and report the total. Provide your final answer enclosed in ... tags.", + "answer": 10 + }, + { + "image": "images/counting_connected_components_00001.png", + "question": "How many connected components are there in the image? A connected component is a maximal group of anchor points such that any two are linked by a path of one or more strands (directly or through other anchors in the same group). Anchors not linked by any chain belong to different components, even if they appear visually close. Count every connected component and report the total. Provide your final answer enclosed in ... tags.", + "answer": 12 + }, + { + "image": "images/counting_connected_components_00002.png", + "question": "How many connected components are there in the image? A connected component is a maximal group of anchor points such that any two are linked by a path of one or more strands (directly or through other anchors in the same group). Anchors not linked by any chain belong to different components, even if they appear visually close. Count every connected component and report the total. Provide your final answer enclosed in ... tags.", + "answer": 15 + }, + { + "image": "images/counting_connected_components_00003.png", + "question": "How many connected components are there in the image? A connected component is a maximal group of anchor points such that any two are linked by a path of one or more strands (directly or through other anchors in the same group). Anchors not linked by any chain belong to different components, even if they appear visually close. Count every connected component and report the total. Provide your final answer enclosed in ... tags.", + "answer": 18 + }, + { + "image": "images/counting_connected_components_00004.png", + "question": "How many connected components are there in the image? A connected component is a maximal group of anchor points such that any two are linked by a path of one or more strands (directly or through other anchors in the same group). Anchors not linked by any chain belong to different components, even if they appear visually close. Count every connected component and report the total. Provide your final answer enclosed in ... tags.", + "answer": 20 + } + ] +} \ No newline at end of file diff --git a/code/distributed_scanning/counting_regions/creation.md b/code/distributed_scanning/counting_regions/creation.md new file mode 100644 index 0000000000000000000000000000000000000000..5dd356e7ad2ea9f0a635cc2a831c2178a725dda9 --- /dev/null +++ b/code/distributed_scanning/counting_regions/creation.md @@ -0,0 +1,170 @@ +# Counting Regions Dataset Generation + +## Goal + +Create tasks where the model counts separated regions in an image. Each +region inside the central square is filled with a distinctive +gradient + texture combination, and adjacent regions are guaranteed to +look visually different so a human can read them apart. + +The image always contains a square in the central area. Inside the +square the canvas is partitioned into irregular regions; outside the +square stays plain. + +## Core Task + +Given an image containing one central square, answer: + +- `How many separated regions are inside the square?` + +The answer is a positive integer. + +## Visual Structure + +- Single square placed near the centre of the canvas with a thin dark + border. Everything outside is a plain off-white background. +- Inside the square the area is partitioned into irregular regions. +- Each region is rendered with two cues at once: + 1. A linear two-colour gradient. The two endpoint colours are drawn + from a fixed palette but each region perturbs them in HSV space + so no two regions render identical colours globally — colour + histograms cannot recover the count. + 2. A texture pattern (speckle, horizontal / vertical / diagonal + stripes, dots, or perlin-like band) modulated multiplicatively on + top of the gradient. Adjacent regions are forced to use different + texture styles where possible. +- Adjacency constraints: for every pair of touching regions, their + unordered colour pairs differ AND there is at least ~55° of hue + separation at the closest pairing. This guarantees a human-readable + contrast across every shared boundary. +- A low-amplitude global speckle is added across the whole canvas so + Canny / Sobel edge detectors fire uniformly inside regions and not + only at boundaries — this defeats the simple "boundary-detect → + flood-fill" attack. +- No drawn boundary lines. Region discrimination relies on colour + discontinuity and texture-style change, not strokes. + +Recommended defaults: + +- Image canvas `1024×1024` +- Painted square `820×820` centred +- Region-synthesis grid `50×50` (low resolution → upsampled with smooth + per-region masks) +- Number of final regions in the range `6-12` + +## Generation Procedure + +### 1. Sample target count and partition + +- Sample target `K` in `[min_regions, max_regions]`. +- Choose `K` spaced seed cells on a 50×50 lattice. +- Grow connected regions from the seeds via priority-queue expansion, + weighted by per-region noise fields so boundaries are organic. + +### 2. Upsample and clean + +- Upsample the label map from 50×50 to canvas resolution by + per-region soft-mask interpolation + argmax (smooth boundaries + without aliasing). +- Absorb tiny connected-component slivers (< 0.2 % of canvas) into + their dominant neighbour label. +- Relabel region ids contiguously after cleanup. +- **Reject** the sample if the smallest remaining region covers less + than `min_region_frac` (default 2.5 %) of the canvas — this avoids + tiny near-invisible regions slipping through. + +### 3. Assign gradient + texture per region + +- Build the region adjacency graph at canvas resolution. +- Greedy assignment over regions (descending degree first): + - Pick `(colour_a, colour_b)` from the palette such that no + neighbour shares the same unordered pair AND `pair_hue_gap` to + every neighbour is at least `hue_gap_min` (default 55°). + - Apply per-region HSV jitter (≈±18° hue, ±0.12 sat, ±0.10 val) to + the two endpoints. + - Pick a random gradient angle. +- Assign a texture style + parameters per region, preferring styles + not used by already-assigned neighbours. + +### 4. Render + +- For each region, paint the linear gradient between its two jittered + endpoint colours along the chosen angle. +- Multiply by `(1 + amp · texture)` per pixel to add the per-region + texture modulation. +- Multiply by `(1 + 0.05 · global_speckle)` to add the canvas-wide + high-frequency noise that drowns Canny boundary detection. +- Composite the painted square onto the off-white canvas with a thin + dark border. + +## Quality Checks + +Reject or regenerate samples if: + +- after cleanup the actual region count drops below 2 +- any region is smaller than `min_region_frac` of the canvas +- gradient assignment fails to satisfy the adjacency constraints + +The final `answer` field always reflects the post-cleanup region +count, which may differ from the sampled target `K`. + +## Anti-Shortcut Notes + +The previous version used dashed boundary lines on a uniform fill, +which was defeated in one step by `cv2.dilate(boundary, k)` followed +by `cv2.connectedComponents`. The new rendering blocks several attack +families simultaneously: + +- **Boundary edge detection (Canny / Sobel)**: drowned by global + speckle + per-region texture so edges fire uniformly across the + whole image rather than only at region boundaries. +- **Colour histogram / k-means clustering**: the per-region HSV + jitter ensures each region renders unique pixel colours even when + two regions share the same palette anchors, so cluster counts have + no clean correspondence to region counts. +- **LAB-space quantisation + connected components**: same — the + jitter scatters pixels across many quantisation bins per region. +- **Smooth-then-detect (Gaussian / median / bilateral filter + + Canny)**: smoothing strong enough to kill the texture also blurs + region boundaries enough that small regions merge. + +Empirically, the strongest CV attack tested (`median13 + Canny`) gives +MAE ~3.5 with 0/6 exact across a held-out set of v7 prototypes. + +## Annotation Format + +Each sample stores the partition metadata required to reproduce or +verify the answer: + +```json +{ + "image": "images/counting_regions_00000.png", + "width": 1024, + "height": 1024, + "grid_rows": 50, + "grid_cols": 50, + "square_left": 102.0, + "square_top": 102.0, + "square_size": 820.0, + "num_regions": 7, + "question": "How many separated regions are inside the square? ...", + "answer": 7, + "difficulty": "medium", + "region_seed_cells": [...], + "region_cell_counts": [...], + "region_adjacency": [[0, 2], [0, 3], ...] +} +``` + +## Output Organization + +```text +counting_regions/ + creation.py + creation.md + annotations.jsonl + data.json + images/ + counting_regions_00000.png + ... +``` diff --git a/code/distributed_scanning/counting_regions/creation.py b/code/distributed_scanning/counting_regions/creation.py new file mode 100644 index 0000000000000000000000000000000000000000..5ea16546b7c65437f031f100120bccc7e4554976 --- /dev/null +++ b/code/distributed_scanning/counting_regions/creation.py @@ -0,0 +1,783 @@ +from __future__ import annotations + +import argparse +import colorsys +import heapq +import json +import math +import random +from collections import deque +from pathlib import Path +from typing import Dict, List, Sequence, Tuple + +import cv2 +import numpy as np +from PIL import Image +from scipy.ndimage import zoom +from tqdm import tqdm + + +Cell = Tuple[int, int] +Point = Tuple[int, int] +Grid = List[List[int]] + + +# ── Palette + colour helpers ─────────────────────────────────────────── + +PALETTE = [ + (236, 152, 99), # peach + (111, 163, 220), # clear blue + (140, 197, 113), # grass green + (223, 127, 186), # pink + (194, 160, 85), # ochre + (157, 134, 212), # purple + (237, 211, 81), # yellow + (75, 184, 173), # teal + (224, 122, 112), # coral red + (122, 160, 106), # sage +] + +PALETTE_HUE = [ + colorsys.rgb_to_hsv(r / 255, g / 255, b / 255)[0] * 360 + for r, g, b in PALETTE +] + + +def hue_gap(h1: float, h2: float) -> float: + d = abs(h1 - h2) % 360 + return min(d, 360 - d) + + +def pair_hue_gap(pair_a: Tuple[int, int], pair_b: Tuple[int, int]) -> float: + """Maximum, over each colour in pair_a, of the nearest hue distance to + any colour in pair_b. Tells us how visually contrastable the two pairs + are at their closest point.""" + best = 0.0 + for ca in pair_a: + nearest = min(hue_gap(PALETTE_HUE[ca], PALETTE_HUE[cb]) for cb in pair_b) + best = max(best, nearest) + return best + + +def jitter_color( + rng: random.Random, + rgb: Tuple[int, int, int], + hue_jitter: float = 18.0, + sat_jitter: float = 0.12, + val_jitter: float = 0.10, +) -> Tuple[float, float, float]: + """Apply small HSV perturbation to a palette anchor so each region's + rendered endpoints are unique (defeats colour-histogram attacks).""" + r, g, b = [c / 255.0 for c in rgb] + h, s, v = colorsys.rgb_to_hsv(r, g, b) + h = (h + rng.uniform(-hue_jitter, hue_jitter) / 360.0) % 1.0 + s = max(0.0, min(1.0, s + rng.uniform(-sat_jitter, sat_jitter))) + v = max(0.0, min(1.0, v + rng.uniform(-val_jitter, val_jitter))) + rr, gg, bb = colorsys.hsv_to_rgb(h, s, v) + return (rr * 255.0, gg * 255.0, bb * 255.0) + + +# ── Partition (connected-component growing on a small grid) ──────────── + +def sample_spaced_cells( + rng: random.Random, + rows: int, + cols: int, + count: int, + min_dist: float, + max_attempts: int = 800, +) -> List[Cell]: + cells: List[Cell] = [] + for _ in range(max_attempts): + if len(cells) == count: + break + candidate = (rng.randrange(rows), rng.randrange(cols)) + if all(math.hypot(candidate[0] - r, candidate[1] - c) >= min_dist for r, c in cells): + cells.append(candidate) + while len(cells) < count: + candidate = (rng.randrange(rows), rng.randrange(cols)) + if candidate not in cells: + cells.append(candidate) + return cells + + +def _make_noise_field(rng: random.Random, rows: int, cols: int, scale: int = 15) -> np.ndarray: + lo_r = max(2, rows // scale) + lo_c = max(2, cols // scale) + lo = np.array([[rng.gauss(0, 1) for _ in range(lo_c)] for _ in range(lo_r)]) + field = zoom(lo, (rows / lo_r, cols / lo_c), order=1) + return field[:rows, :cols] + + +def make_connected_partition( + rng: random.Random, + rows: int, + cols: int, + num_regions: int, +) -> Tuple[np.ndarray, List[Cell]]: + """Two-phase partition. + + Phase 1 (sequential round-robin): each region in turn claims one + cell from its current frontier, repeated until every region has at + least `min_cells_per_region` cells. This guarantees a minimum size + for every region and prevents tiny / vanishing regions. + + Phase 2 (priority-queue Voronoi-like): the remaining unclaimed cells + are distributed using the original distance-to-centroid + noise-field + cost so the final boundaries look organic. + """ + labels = np.full((rows, cols), -1, dtype=np.int32) + seeds = sample_spaced_cells( + rng=rng, + rows=rows, + cols=cols, + count=num_regions, + min_dist=max(2.2, min(rows, cols) / max(3.4, math.sqrt(num_regions) + 0.5)), + ) + dirs = [(-1, 0), (1, 0), (0, -1), (0, 1)] + cr = [float(r) for r, c in seeds] + cc = [float(c) for r, c in seeds] + size = [1] * num_regions + + # ── Phase 1: sequential round-robin grow ── + fair_share = (rows * cols) // num_regions + min_cells_per_region = max(8, (fair_share * 7) // 10) # 70% of fair share floor + frontiers: List[List[Cell]] = [[] for _ in range(num_regions)] + for rid, (r, c) in enumerate(seeds): + labels[r, c] = rid + for dr, dc in dirs: + nr, nc = r + dr, c + dc + if 0 <= nr < rows and 0 <= nc < cols and labels[nr, nc] == -1: + frontiers[rid].append((nr, nc)) + + while True: + # Stop when every region reached the minimum, OR no region can grow. + if all(s >= min_cells_per_region for s in size): + break + progress = False + for rid in range(num_regions): + if size[rid] >= min_cells_per_region: + continue + # Pull a random claimable cell from this region's frontier. + f = frontiers[rid] + while f: + idx = rng.randrange(len(f)) + nr, nc = f.pop(idx) + if labels[nr, nc] != -1: + continue + labels[nr, nc] = rid + n = size[rid] + cr[rid] = (cr[rid] * n + nr) / (n + 1) + cc[rid] = (cc[rid] * n + nc) / (n + 1) + size[rid] += 1 + for dr, dc in dirs: + rr, ccc = nr + dr, nc + dc + if 0 <= rr < rows and 0 <= ccc < cols and labels[rr, ccc] == -1: + f.append((rr, ccc)) + progress = True + break + if not progress: + break # all frontiers exhausted (only possible at very high n) + + # ── Phase 2: priority-queue grow on remaining cells ── + fields = [_make_noise_field(rng, rows, cols, scale=15) for _ in range(num_regions)] + noise_strength = rows * cols * 0.03 + heap: list = [] + for rid in range(num_regions): + for nr, nc in frontiers[rid]: + if labels[nr, nc] == -1: + d = (nr - cr[rid]) ** 2 + (nc - cc[rid]) ** 2 + d += fields[rid][nr, nc] * noise_strength + d += rng.random() * 5.0 + heapq.heappush(heap, (d, rid, nr, nc)) + while heap: + _, rid, nr, nc = heapq.heappop(heap) + if labels[nr, nc] != -1: + continue + labels[nr, nc] = rid + n = size[rid] + cr[rid] = (cr[rid] * n + nr) / (n + 1) + cc[rid] = (cc[rid] * n + nc) / (n + 1) + size[rid] += 1 + for dr, dc in dirs: + rr, ccc = nr + dr, nc + dc + if 0 <= rr < rows and 0 <= ccc < cols and labels[rr, ccc] == -1: + d = (rr - cr[rid]) ** 2 + (ccc - cc[rid]) ** 2 + d += fields[rid][rr, ccc] * noise_strength + d += rng.random() * 5.0 + heapq.heappush(heap, (d, rid, rr, ccc)) + return labels, seeds + + +def upsample_labels( + grid_labels: np.ndarray, + canvas_w: int, + canvas_h: int, + smooth_sigma: float = 4.5, +) -> np.ndarray: + """Upsample a low-resolution label map to canvas size with smooth + boundaries by per-region soft-mask interpolation + argmax.""" + num_regions = int(grid_labels.max()) + 1 + soft = np.empty((num_regions, canvas_h, canvas_w), dtype=np.float32) + for rid in range(num_regions): + mask = (grid_labels == rid).astype(np.float32) + big = cv2.resize(mask, (canvas_w, canvas_h), interpolation=cv2.INTER_LINEAR) + if smooth_sigma > 0: + big = cv2.GaussianBlur(big, (0, 0), sigmaX=smooth_sigma, sigmaY=smooth_sigma) + soft[rid] = big + return np.argmax(soft, axis=0).astype(np.int32) + + +def clean_tiny_components(labels: np.ndarray, min_frac: float = 0.002) -> np.ndarray: + """Absorb tiny connected-component specks into the dominant neighbour + label. Small ``min_frac`` keeps the cleanup conservative.""" + h, w = labels.shape + threshold = int(h * w * min_frac) + out = labels.copy() + num_regions = int(labels.max()) + 1 + for rid in range(num_regions): + mask = (out == rid).astype(np.uint8) + num, comp = cv2.connectedComponents(mask) + sizes = [int((comp == i).sum()) for i in range(num)] + if num <= 2: + continue + keep = max(range(1, num), key=lambda i: sizes[i]) + for i in range(1, num): + if i == keep: + continue + if sizes[i] < threshold: + small = (comp == i) + ys, xs = np.where(small) + nbrs = [] + for dy, dx in [(-1, 0), (1, 0), (0, -1), (0, 1)]: + ny, nx = np.clip(ys + dy, 0, h - 1), np.clip(xs + dx, 0, w - 1) + nbrs.append(out[ny, nx]) + nbr_labels = np.concatenate(nbrs) + nbr_labels = nbr_labels[nbr_labels != rid] + if len(nbr_labels) == 0: + continue + replacement = int(np.bincount(nbr_labels).argmax()) + out[small] = replacement + return out + + +def relabel_contiguous(labels: np.ndarray) -> Tuple[np.ndarray, int]: + unique = np.unique(labels) + remap = -np.ones(int(unique.max()) + 1, dtype=np.int32) + for new_id, old_id in enumerate(unique): + remap[old_id] = new_id + return remap[labels], len(unique) + + +def adjacency_from_labels(labels: np.ndarray) -> Dict[int, set]: + n = int(labels.max()) + 1 + adj: Dict[int, set] = {rid: set() for rid in range(n)} + diff_h = labels[:, :-1] != labels[:, 1:] + a_h = labels[:, :-1][diff_h] + b_h = labels[:, 1:][diff_h] + diff_v = labels[:-1, :] != labels[1:, :] + a_v = labels[:-1, :][diff_v] + b_v = labels[1:, :][diff_v] + for a, b in zip(a_h.tolist(), b_h.tolist()): + adj[a].add(b) + adj[b].add(a) + for a, b in zip(a_v.tolist(), b_v.tolist()): + adj[a].add(b) + adj[b].add(a) + return adj + + +def adjacency_pairs(labels: np.ndarray) -> List[Tuple[int, int]]: + adj = adjacency_from_labels(labels) + edges: set[Tuple[int, int]] = set() + for a, neighbours in adj.items(): + for b in neighbours: + if a < b: + edges.add((a, b)) + return sorted(edges) + + +# ── Gradient assignment with adjacency constraints ──────────────────── + +def assign_gradients( + rng: random.Random, + num_regions: int, + adj: Dict[int, set], + hue_gap_min: float = 55.0, + max_attempts: int = 400, +) -> List[Tuple[Tuple[float, float, float], Tuple[float, float, float], float]] | None: + """Greedy assignment with restart. Adjacent regions must differ in + their unordered colour pair AND have at least ``hue_gap_min`` degrees + of hue separation at their closest pairing.""" + palette_size = len(PALETTE) + order = sorted(range(num_regions), key=lambda r: -len(adj[r])) + + # Since pure-flat fill uses only the FIRST colour of each pair, enforce + # that the FIRST hues of adjacent regions are well separated, in addition + # to the original pair-hue-gap constraint. + PRIMARY_HUE_GAP_MIN = 60.0 + + for _ in range(max_attempts): + pairs: List[Tuple[int, int] | None] = [None] * num_regions + ok = True + for rid in order: + cand = [] + for a in range(palette_size): + for b in range(a + 1, palette_size): + cand.append((a, b)) + rng.shuffle(cand) + chosen = None + for a, b in cand: + valid = True + for n in adj[rid]: + if pairs[n] is None: + continue + na, nb = pairs[n][0], pairs[n][1] + if {a, b} == {na, nb}: + valid = False + break + if pair_hue_gap((a, b), (na, nb)) < hue_gap_min: + valid = False + break + if hue_gap(PALETTE_HUE[a], PALETTE_HUE[na]) < PRIMARY_HUE_GAP_MIN: + valid = False + break + if valid: + chosen = (a, b) + break + if chosen is None: + ok = False + break + pairs[rid] = chosen + if ok: + angles = [rng.uniform(0, 2 * math.pi) for _ in range(num_regions)] + jittered = [] + for p, ang in zip(pairs, angles): + ca = jitter_color(rng, PALETTE[p[0]]) + cb = jitter_color(rng, PALETTE[p[1]]) + jittered.append((ca, cb, ang)) + return jittered + return None + + +# ── Per-region textures (defeat canny + colour-cluster attacks) ─────── + +TEXTURE_STYLES = ["speckle", "stripes_h", "stripes_v", "stripes_d1", + "stripes_d2", "dots", "perlin"] + + +def assign_textures( + rng: random.Random, + num_regions: int, + adj: Dict[int, set], +) -> List[Tuple[str, Dict[str, float]]]: + """Pick a (style, params) per region. Adjacent regions get different + styles where possible so texture acts as a region cue for humans.""" + out: List[Tuple[str, Dict[str, float]]] = [] + for rid in range(num_regions): + used_by_neighbours = {out[n][0] for n in adj[rid] if n < rid} + candidates = [s for s in TEXTURE_STYLES if s not in used_by_neighbours] + if not candidates: + candidates = TEXTURE_STYLES + style = rng.choice(candidates) + params = { + "freq": rng.uniform(0.08, 0.22), + "amp": rng.uniform(0.08, 0.16), + "orient": rng.uniform(0, math.pi), + "speckle_seed": rng.randint(0, 2**31 - 1), + } + out.append((style, params)) + return out + + +def make_texture(style: str, params: Dict[str, float], h: int, w: int, rng_seed: int) -> np.ndarray: + rng = np.random.default_rng(rng_seed) + ys, xs = np.mgrid[0:h, 0:w].astype(np.float32) + freq = params["freq"] + orient = params["orient"] + if style == "speckle": + n = rng.standard_normal((h, w)).astype(np.float32) + return cv2.GaussianBlur(n, (0, 0), sigmaX=0.7) * 1.4 + if style.startswith("stripes"): + if style == "stripes_h": + theta = 0.0 + elif style == "stripes_v": + theta = math.pi / 2 + elif style == "stripes_d1": + theta = math.pi / 4 + else: + theta = -math.pi / 4 + u = xs * math.cos(theta) + ys * math.sin(theta) + return np.sin(u * freq * 2 * math.pi).astype(np.float32) + if style == "dots": + u = xs * freq * 2 * math.pi + v = ys * freq * 2 * math.pi + return (np.sin(u) * np.sin(v)).astype(np.float32) * 1.2 + if style == "perlin": + a = np.sin((xs * math.cos(orient) + ys * math.sin(orient)) * freq * 2 * math.pi) + b = np.cos((xs * math.cos(orient + 1.1) + ys * math.sin(orient + 1.1)) * freq * 1.7 * 2 * math.pi) + return ((a + b) * 0.5).astype(np.float32) + return np.zeros((h, w), dtype=np.float32) + + +# ── Renderer ────────────────────────────────────────────────────────── + +def render_region_canvas( + labels: np.ndarray, + assignments: List[Tuple], + textures: List[Tuple[str, Dict[str, float]]], + global_speckle: float = 0.05, +) -> np.ndarray: + """Paint each region with its (gradient + texture). No drawn boundary + lines — region discrimination relies on colour discontinuity and + texture-style change.""" + h, w = labels.shape + # SINGLE global base colour (warm light grey) across the whole canvas. + # Region labels and per-region palette assignments are intentionally + # ignored: the goal is one uniform texture that flows continuously + # across the entire image, so the only visible cue for the partition + # is the boundary line drawn on top by render_instance. + base = np.array([220.0, 215.0, 205.0], dtype=np.float32) + img = np.broadcast_to(base, (h, w, 3)).copy() + + # Two-octave smooth noise field across the whole canvas, plus fine grain. + rng_np = np.random.default_rng(0xC0FFEE) + big = rng_np.standard_normal((h, w)).astype(np.float32) + big = cv2.GaussianBlur(big, (0, 0), sigmaX=h / 6.0, sigmaY=h / 6.0) + med = rng_np.standard_normal((h, w)).astype(np.float32) + med = cv2.GaussianBlur(med, (0, 0), sigmaX=h / 18.0, sigmaY=h / 18.0) + noise = big / (big.std() + 1e-6) + 0.7 * med / (med.std() + 1e-6) + noise /= (noise.std() + 1e-6) + # Multiplicative brightness modulation across the whole canvas: ±25%. + img = img * (1.0 + 0.25 * noise[..., None]) + # Per-channel fine grain (±8 brightness units). + grain = rng_np.standard_normal((h, w, 3)).astype(np.float32) * 8.0 + img = img + grain + return np.clip(img, 0, 255).astype(np.uint8) + + +BG_COLOR = (248, 246, 240) +BORDER_COLOR = (45, 39, 32) + + +def composite_full_image(painted: np.ndarray, width: int, height: int) -> np.ndarray: + """Paste the painted region square onto the full canvas with a thin + dark border around it.""" + canvas_size = painted.shape[0] + margin = (min(width, height) - canvas_size) // 2 + full = np.full((height, width, 3), BG_COLOR, dtype=np.uint8) + + border = max(2, canvas_size // 256) + y0 = margin + x0 = (width - canvas_size) // 2 + full[y0:y0 + canvas_size, x0:x0 + canvas_size] = painted + full[y0 - border:y0, x0 - border:x0 + canvas_size + border] = BORDER_COLOR + full[y0 + canvas_size:y0 + canvas_size + border, x0 - border:x0 + canvas_size + border] = BORDER_COLOR + full[y0 - border:y0 + canvas_size + border, x0 - border:x0] = BORDER_COLOR + full[y0 - border:y0 + canvas_size + border, x0 + canvas_size:x0 + canvas_size + border] = BORDER_COLOR + return full + + +# ── Sampling: build one valid instance ──────────────────────────────── + +def difficulty_for_regions(num_regions: int) -> str: + if num_regions <= 6: + return "easy" + if num_regions <= 10: + return "medium" + return "hard" + + +def count_region_cells(labels: np.ndarray, num_regions: int) -> List[int]: + return [int((labels == rid).sum()) for rid in range(num_regions)] + + +def sample_instance( + rng: random.Random, + width: int, + height: int, + min_regions: int, + max_regions: int, + grid_rows: int, + grid_cols: int, + canvas_size: int, + min_region_frac: float, + hue_gap_min: float, + forced_target: int | None = None, +) -> Dict[str, object] | None: + """Build one valid instance or return None if rejected. If forced_target + is given, only accept samples whose actual region count equals it.""" + max_feasible_regions = min(max_regions, max(2, (grid_rows * grid_cols) // 6)) + min_feasible_regions = min(min_regions, max_feasible_regions) + if forced_target is not None: + target_n = max(2, min(max_feasible_regions, forced_target)) + else: + target_n = rng.randint(max(2, min_feasible_regions), max_feasible_regions) + + grid_labels, seeds = make_connected_partition(rng, grid_rows, grid_cols, target_n) + + if forced_target is not None: + # 200x200 grid + 50%-fair-share phase-1 floor makes every region + # big enough to survive heavy smoothing. Use the smoothed argmax + # path so boundaries are smooth curves instead of NEAREST staircase. + canvas_labels = upsample_labels(grid_labels, canvas_size, canvas_size, smooth_sigma=14.0) + else: + canvas_labels = upsample_labels(grid_labels, canvas_size, canvas_size, smooth_sigma=9.0) + if forced_target is None: + # Normal path: clean tiny specks and apply min-area floor. + canvas_labels = clean_tiny_components(canvas_labels, min_frac=0.002) + canvas_labels, actual_n = relabel_contiguous(canvas_labels) + if actual_n < 2: + return None + region_pixels = count_region_cells(canvas_labels, actual_n) + min_pixels = int(canvas_size * canvas_size * min_region_frac) + if min(region_pixels) < min_pixels: + return None + else: + # Forced path: trust the seeded grow — the partition produced exactly + # forced_target connected regions. Skip clean_tiny_components and the + # min-area floor so the count is honoured exactly. + canvas_labels, actual_n = relabel_contiguous(canvas_labels) + if actual_n != forced_target: + return None + region_pixels = count_region_cells(canvas_labels, actual_n) + + adj = adjacency_from_labels(canvas_labels) + assignments = assign_gradients(rng, actual_n, adj, hue_gap_min=hue_gap_min) + if assignments is None: + return None + textures = assign_textures(rng, actual_n, adj) + + margin = int(min(width, height) * 0.12) + square_size = canvas_size + square_left = (width - square_size) // 2 + square_top = (height - square_size) // 2 + + return { + "width": width, + "height": height, + "grid_rows": grid_rows, + "grid_cols": grid_cols, + "square_left": float(square_left), + "square_top": float(square_top), + "square_size": float(square_size), + "num_regions": actual_n, + "question": ( + "How many separated regions are inside the square? " + "A region is a maximal area inside the square that is filled with " + "one continuous colour pattern. Two locations belong to the same " + "region if you can travel between them without crossing into a " + "differently-coloured area. Count every distinct region inside the " + "square and report the total as a positive integer. " + "Provide your final answer enclosed in ... tags." + ), + "answer": actual_n, + "difficulty": difficulty_for_regions(actual_n), + "region_seed_cells": [[int(r), int(c)] for r, c in seeds], + "region_cell_counts": region_pixels, + "region_adjacency": [[int(a), int(b)] for a, b in adjacency_pairs(canvas_labels)], + "_canvas_labels": canvas_labels, + "_assignments": assignments, + "_textures": textures, + } + + +def render_instance(out_path: Path, record: Dict[str, object]) -> None: + """Coloured fill renderer: each region is filled with its assigned + gradient + texture, plus a thin dark boundary line drawn along the + label-difference set so triple junctions never have hollow gaps.""" + canvas_labels: np.ndarray = record.pop("_canvas_labels") # type: ignore[assignment] + assignments = record.pop("_assignments") + textures = record.pop("_textures") + width = int(record["width"]) + height = int(record["height"]) + + h, w = canvas_labels.shape + painted = render_region_canvas(canvas_labels, assignments, textures) + line_color = (32, 32, 32) + line_thickness = max(1, min(h, w) // 700) + + # Smooth labels first via per-region one-hot → blur → argmax. This + # removes the staircase from the original integer upsample but, crucially, + # gives a SINGLE consistent label map (no triple-junction gaps). + num_regions = int(canvas_labels.max()) + 1 + smooth_sigma = max(2.5, min(h, w) / 140.0) + soft = np.empty((num_regions, h, w), dtype=np.float32) + for rid in range(num_regions): + mask = (canvas_labels == rid).astype(np.float32) + soft[rid] = cv2.GaussianBlur(mask, (0, 0), sigmaX=smooth_sigma, + sigmaY=smooth_sigma) + smoothed_labels = np.argmax(soft, axis=0).astype(np.int32) + + # Boundary mask: a pixel is on a boundary iff any of its 4-neighbours has + # a different label. This produces a single closed curve per junction + # with no hollow gap at triple points. + boundary = np.zeros((h, w), dtype=np.uint8) + diff_h = smoothed_labels[:, :-1] != smoothed_labels[:, 1:] + diff_v = smoothed_labels[:-1, :] != smoothed_labels[1:, :] + boundary[:, :-1][diff_h] = 1 + boundary[:, 1:][diff_h] = 1 + boundary[:-1, :][diff_v] = 1 + boundary[1:, :][diff_v] = 1 + if line_thickness > 1: + kernel = cv2.getStructuringElement(cv2.MORPH_ELLIPSE, (line_thickness, line_thickness)) + boundary = cv2.dilate(boundary, kernel) + painted[boundary > 0] = line_color + + full = composite_full_image(painted, width, height) + Image.fromarray(full).save(out_path) + + +# ── Dataset generation ──────────────────────────────────────────────── + +def ensure_output_dir(root: Path) -> Tuple[Path, Path]: + root.mkdir(parents=True, exist_ok=True) + images_dir = root / "images" + images_dir.mkdir(exist_ok=True) + return root, images_dir + + +def generate_dataset( + rng: random.Random, + count: int, + output_dir: Path, + images_dir: Path, + width: int, + height: int, + min_regions: int, + max_regions: int, + grid_rows: int, + grid_cols: int, + canvas_size: int, + min_region_frac: float, + hue_gap_min: float, + forced_targets: List[int] | None = None, +) -> None: + records: List[Dict[str, object]] = [] + data_records: List[Dict[str, object]] = [] + pbar = tqdm(total=count, desc="counting_regions") + idx = 0 + rejects = 0 + while idx < count: + target = (forced_targets[idx] if forced_targets and idx < len(forced_targets) else None) + record = sample_instance( + rng=rng, + width=width, + height=height, + min_regions=min_regions, + max_regions=max_regions, + grid_rows=grid_rows, + grid_cols=grid_cols, + canvas_size=canvas_size, + min_region_frac=min_region_frac, + hue_gap_min=hue_gap_min, + forced_target=target, + ) + if record is None: + rejects += 1 + continue + image_name = f"counting_regions_{idx:05d}.png" + render_instance(images_dir / image_name, record) + record["image"] = f"images/{image_name}" + records.append(record) + data_records.append({ + "image": record["image"], + "question": record["question"], + "answer": record["answer"], + }) + idx += 1 + pbar.update(1) + pbar.set_postfix(answer=record["answer"], rejects=rejects) + pbar.close() + + with (output_dir / "annotations.jsonl").open("w", encoding="utf-8") as fh: + for record in records: + fh.write(json.dumps(record) + "\n") + + data_json = { + "task": "counting_regions", + "category": "distributed_scanning", + "count": len(data_records), + "items": data_records, + } + with (output_dir / "data.json").open("w", encoding="utf-8") as fh: + json.dump(data_json, fh, indent=2) + + +def parse_args() -> argparse.Namespace: + parser = argparse.ArgumentParser(description="Generate a counting-regions dataset.") + parser.add_argument("--output-root", type=Path, required=True) + parser.add_argument("--count", type=int, default=36) + parser.add_argument("--width", type=int, default=1024) + parser.add_argument("--height", type=int, default=1024) + parser.add_argument("--min-regions", type=int, default=6) + parser.add_argument("--max-regions", type=int, default=12) + parser.add_argument("--grid-rows", type=int, default=200) + parser.add_argument("--grid-cols", type=int, default=200) + parser.add_argument("--canvas-size", type=int, default=820, + help="Pixel side of the painted square inside the image.") + parser.add_argument("--min-region-frac", type=float, default=0.025, + help="Reject samples whose smallest region covers " + "less than this fraction of the canvas.") + parser.add_argument("--hue-gap-min", type=float, default=55.0) + parser.add_argument("--seed", type=int, default=23) + parser.add_argument("--difficulty", type=int, default=5, + help="Integer difficulty >=0; scales region count.") + return parser.parse_args() + + +def main() -> None: + args = parse_args() + d = max(0, int(args.difficulty)) + + # Canvas scaling: N_d = 10 + d, N_0 = 10. + N_d = 10 + d + N_0 = 10 + s = math.sqrt(max(1.0, N_d / N_0)) + args.width = int(round(args.width * s)) + args.height = int(round(args.height * s)) + args.canvas_size = int(round(args.canvas_size * s)) + + rng = random.Random(args.seed) + output_dir, images_dir = ensure_output_dir(args.output_root) + + # num_regions ∈ [5, 10 + d] + min_regions = 10 + max_regions = 10 + 2 * d + # Auto-scale: min_region_frac = min(0.04, 1 / (2.5 * max_regions)) + # Use max_regions (upper bound) as a conservative frac floor so any + # sampled count satisfies the constraint. + min_region_frac = min(0.04, 1.0 / (2.5 * max_regions)) + + # Force per-instance region counts evenly spaced across [min_regions, + # max_regions]. With count=5 and range=[10, 20] this yields [10, 13, 15, 18, 20]. + if args.count > 1: + forced_targets = [ + int(round(min_regions + i * (max_regions - min_regions) / (args.count - 1))) + for i in range(args.count) + ] + else: + forced_targets = [min_regions] + + generate_dataset( + rng=rng, + count=args.count, + output_dir=output_dir, + images_dir=images_dir, + width=args.width, + height=args.height, + min_regions=min_regions, + max_regions=max_regions, + grid_rows=args.grid_rows, + grid_cols=args.grid_cols, + canvas_size=args.canvas_size, + min_region_frac=min_region_frac, + hue_gap_min=args.hue_gap_min, + forced_targets=forced_targets, + ) + print(f"forced region counts: {forced_targets}") + print(f"Saved dataset to {args.output_root} (canvas {args.width}x{args.height})") + + +if __name__ == "__main__": + main() diff --git a/code/distributed_scanning/counting_regions/data.json b/code/distributed_scanning/counting_regions/data.json new file mode 100644 index 0000000000000000000000000000000000000000..b985a4b3cb9e83ce60cb5f7d6bff0563e8b88370 --- /dev/null +++ b/code/distributed_scanning/counting_regions/data.json @@ -0,0 +1,32 @@ +{ + "task": "counting_regions", + "category": "distributed_scanning", + "count": 5, + "items": [ + { + "image": "images/counting_regions_00000.png", + "question": "How many separated regions are inside the image? A region is a maximal area filled with one continuous fill. Boundaries between regions may appear as continuous lines OR as a chain of small individual markers \u2014 discrete dots, footprints, dashes, or short segments aligned along a curve \u2014 that you must visually link together to recognize as a single boundary. Two locations belong to the same region if you can travel between them without crossing such a boundary (continuous or chained). Count every distinct region and report the total as a positive integer. Provide your final answer enclosed in ... tags.", + "answer": 10 + }, + { + "image": "images/counting_regions_00001.png", + "question": "How many separated regions are inside the image? A region is a maximal area filled with one continuous fill. Boundaries between regions may appear as continuous lines OR as a chain of small individual markers \u2014 discrete dots, footprints, dashes, or short segments aligned along a curve \u2014 that you must visually link together to recognize as a single boundary. Two locations belong to the same region if you can travel between them without crossing such a boundary (continuous or chained). Count every distinct region and report the total as a positive integer. Provide your final answer enclosed in ... tags.", + "answer": 12 + }, + { + "image": "images/counting_regions_00002.png", + "question": "How many separated regions are inside the image? A region is a maximal area filled with one continuous fill. Boundaries between regions may appear as continuous lines OR as a chain of small individual markers \u2014 discrete dots, footprints, dashes, or short segments aligned along a curve \u2014 that you must visually link together to recognize as a single boundary. Two locations belong to the same region if you can travel between them without crossing such a boundary (continuous or chained). Count every distinct region and report the total as a positive integer. Provide your final answer enclosed in ... tags.", + "answer": 15 + }, + { + "image": "images/counting_regions_00003.png", + "question": "How many separated regions are inside the image? A region is a maximal area filled with one continuous fill. Boundaries between regions may appear as continuous lines OR as a chain of small individual markers \u2014 discrete dots, footprints, dashes, or short segments aligned along a curve \u2014 that you must visually link together to recognize as a single boundary. Two locations belong to the same region if you can travel between them without crossing such a boundary (continuous or chained). Count every distinct region and report the total as a positive integer. Provide your final answer enclosed in ... tags.", + "answer": 18 + }, + { + "image": "images/counting_regions_00004.png", + "question": "How many separated regions are inside the image? A region is a maximal area filled with one continuous fill. Boundaries between regions may appear as continuous lines OR as a chain of small individual markers \u2014 discrete dots, footprints, dashes, or short segments aligned along a curve \u2014 that you must visually link together to recognize as a single boundary. Two locations belong to the same region if you can travel between them without crossing such a boundary (continuous or chained). Count every distinct region and report the total as a positive integer. Provide your final answer enclosed in ... tags.", + "answer": 20 + } + ] +} \ No newline at end of file diff --git a/code/distributed_scanning/tangled_loops/creation.md b/code/distributed_scanning/tangled_loops/creation.md new file mode 100644 index 0000000000000000000000000000000000000000..b2b8b79cac3dbf1e4817bd8c733baccc6c6c888e --- /dev/null +++ b/code/distributed_scanning/tangled_loops/creation.md @@ -0,0 +1,107 @@ +# Tangled Closed-Loop Counting + +## Goal + +Render a tangle of smooth **closed loops** on a single canvas. Every +loop is drawn in the **same** dark colour, so the model cannot recover +the count by colour-segmenting the image. Loops cross each other and +themselves freely, but never run parallel for long stretches — every +loop remains individually traceable. + +This is a *distributed scanning* task: there is no single starting +point. The model must look across the whole image, follow each loop +all the way around, and report how many distinct closed loops exist. + +## Differs from `sequential_traversal/tube_connection` + +`tube_connection` labels endpoints and asks "which top label connects +to which bottom label", which is solved by tracing one path at a time. +Here there are no labels, no endpoints at all, and no per-loop tracing +question. The model needs the global *count* of distinct closed loops. + +## Differs from `sequential_traversal/line_intersections` + +`line_intersections` uses perimeter-anchored open curves and asks for +crossing sequences along one labelled string. Here every curve is a +closed loop (no endpoints anywhere), every loop is anonymous, all in +the same colour, and the only ground truth is the number of loops. + +## Question + +> How many distinct closed loops are tangled together in this image? +> Each loop is a single continuous curve that closes back on itself — +> there are no loose endpoints anywhere. All loops are drawn in the +> same colour and may cross other loops or themselves freely. Count +> the total number of distinct closed loops and report the count as a +> positive integer. + +## Generation Procedure + +1. **Sample N** in `[min_loops, max_loops]` (default 5–11). +2. **For each loop**, build a smooth closed curve: + - Pick a random interior centre so the loop fits inside + `interior_margin` (default 55 px) of the canvas edge. + - Sample 6–10 waypoints on a jittered ring around that centre + (angle jitter ≈ 0.42 rad, radius jitter ≈ 32 % of the mean + radius, mean radius sampled from 150–275 px). + - Fit a periodic cubic B-spline through the waypoints + (`scipy.interpolate.splprep(..., per=True, k=3)`) and + dense-sample 900 points along it. The last point is snapped to + equal the first so the renderer draws a genuine closed curve. +3. **Find all crossings** (self and pair) — used only by the + close-approach validator below. +4. **Reject** the sample if any two curves come within ~9 px of each + other anywhere except at a real crossing point. For self-closeness + the segment-index gap is computed **circularly** (with wraparound), + since closed curves have no linear endpoint ordering. +5. **Render** all loops in the same dark colour onto an off-white + canvas with subtle background noise. + +## Anti-Shortcut Notes + +The task has been through two shortcut-plugging rewrites: + +1. **Unique-colour shortcut → single stroke colour.** An earlier + version assigned each string a unique colour, collapsing the task + to "count distinct colours". Fixed by drawing every curve in the + same dark colour. +2. **Skeleton-endpoint shortcut → closed loops.** The previous + perimeter-anchored design had each string as an *open* curve with + two endpoints on the border. A ten-line attack recovered the + answer on 28 / 30 samples: `skeletonize(gray < 110)` → + count pixels with exactly one 8-neighbour → divide by two. + Crossings are X/T junctions (degree ≥ 3), so only the two terminal + endpoints of each string have degree 1, and endpoints ÷ 2 is the + string count. Moving to closed curves eliminates the attack + entirely — a closed loop's skeleton has no degree-1 pixels, so the + attack returns 0 regardless of loop count. Verified on the + regenerated dataset. +3. **No-near-parallel validation** (retained) prevents two loops that + run parallel for a long stretch from reading as one fat loop, + which would also confuse a human counter. + +## Annotation Format + +```json +{ + "image": "images/tangled_loops_00000.png", + "width": 1024, + "height": 1024, + "num_loops": 7, + "question": "How many distinct closed loops are tangled together ...", + "answer": 7 +} +``` + +## Output Organization + +```text +tangled_loops/ + creation.py + creation.md + annotations.jsonl + data.json + images/ + tangled_loops_00000.png + ... +``` diff --git a/code/distributed_scanning/tangled_loops/creation.py b/code/distributed_scanning/tangled_loops/creation.py new file mode 100644 index 0000000000000000000000000000000000000000..c75d8f2b91fcbef7a4a7cd33273b798c6f5edcee --- /dev/null +++ b/code/distributed_scanning/tangled_loops/creation.py @@ -0,0 +1,531 @@ +"""Tangled Closed-Loop Counting. + +Each string is a smooth **closed curve** living entirely in the interior +of the canvas. Loops are generated by sampling interior waypoints on a +jittered ring around a random centre and fitting a periodic cubic +B-spline through them — the resulting curve has no endpoints at all, +which plugs the "skeletonize → count degree-1 pixels → divide by 2" +shortcut that beat the previous perimeter-anchored design. + +All loops render in the same dark colour. Loops may cross other loops +or themselves, but long parallel runs are rejected so every loop stays +individually traceable. + +The model is asked to count the total number of distinct closed loops. +""" +from __future__ import annotations + +import argparse +import json +import math +import os +import random +import sys +from collections import defaultdict +from pathlib import Path +from typing import Dict, List, Tuple + +import matplotlib +matplotlib.use("Agg") +import matplotlib.pyplot as plt +import numpy as np +from scipy.interpolate import splev, splprep +from tqdm import tqdm + + +LINE_COLOR = "#2f2f2f" + + +# ── Closed-loop construction ─────────────────────────────────────── + +def build_closed_loop( + rng: random.Random, + width: int, + height: int, + interior_margin: int = 55, + num_waypoints_range: Tuple[int, int] = (6, 10), + radius_range: Tuple[float, float] = (130.0, 215.0), + radius_jitter: float = 0.30, + angle_jitter: float = 0.42, + num_samples: int = 520, + existing_centers: List[Tuple[float, float]] | None = None, + min_center_gap: float = 190.0, + center_placement_attempts: int = 80, +) -> Tuple[np.ndarray, Tuple[float, float]] | None: + """Build one smooth closed curve through jittered ring waypoints. + + Returns ``(polyline, (cx, cy))`` where the polyline's last point + equals its first, or ``None`` if no valid placement was found. + """ + num_wp = rng.randint(*num_waypoints_range) + r_mean = rng.uniform(*radius_range) + max_r = r_mean * (1 + radius_jitter) + + low_x = interior_margin + max_r + high_x = width - interior_margin - max_r + low_y = interior_margin + max_r + high_y = height - interior_margin - max_r + if low_x >= high_x or low_y >= high_y: + return None + + cx = cy = None + centers = existing_centers or [] + for _ in range(center_placement_attempts): + cand_x = rng.uniform(low_x, high_x) + cand_y = rng.uniform(low_y, high_y) + ok = True + for ex_x, ex_y in centers: + if (cand_x - ex_x) ** 2 + (cand_y - ex_y) ** 2 < min_center_gap ** 2: + ok = False + break + if ok: + cx, cy = cand_x, cand_y + break + if cx is None: + return None + + base_angles = np.linspace(0.0, 2 * math.pi, num_wp, endpoint=False) + phase = rng.uniform(0.0, 2 * math.pi) + + xs: List[float] = [] + ys: List[float] = [] + for base in base_angles: + ang = base + phase + rng.uniform(-angle_jitter, angle_jitter) + r = r_mean * (1.0 + rng.uniform(-radius_jitter, radius_jitter)) + xs.append(cx + r * math.cos(ang)) + ys.append(cy + r * math.sin(ang)) + + # splprep with per=True expects the input to already be closed. + xs.append(xs[0]) + ys.append(ys[0]) + + try: + tck, _ = splprep([xs, ys], s=0.0, per=True, k=3) + except (TypeError, ValueError): + return None + + u_dense = np.linspace(0.0, 1.0, num_samples) + x_dense, y_dense = splev(u_dense, tck) + poly = np.column_stack([np.asarray(x_dense), np.asarray(y_dense)]) + + if (poly[:, 0].min() < interior_margin - 5 + or poly[:, 0].max() > width - interior_margin + 5 + or poly[:, 1].min() < interior_margin - 5 + or poly[:, 1].max() > height - interior_margin + 5): + return None + + poly[-1] = poly[0] + return poly, (cx, cy) + + +# ── Crossing detection (used for the close-approach validation) ──── + +def _segments_cross(p0, p1, q0, q1) -> bool: + eps = 1e-8 + o1 = (p1[0]-p0[0])*(q0[1]-p0[1]) - (p1[1]-p0[1])*(q0[0]-p0[0]) + o2 = (p1[0]-p0[0])*(q1[1]-p0[1]) - (p1[1]-p0[1])*(q1[0]-p0[0]) + o3 = (q1[0]-q0[0])*(p0[1]-q0[1]) - (q1[1]-q0[1])*(p0[0]-q0[0]) + o4 = (q1[0]-q0[0])*(p1[1]-q0[1]) - (q1[1]-q0[1])*(p1[0]-q0[0]) + return ((o1 > eps and o2 < -eps) or (o1 < -eps and o2 > eps)) and \ + ((o3 > eps and o4 < -eps) or (o3 < -eps and o4 > eps)) + + +def _circular_seg_gap(i: int, j: int, n: int) -> int: + d = abs(i - j) + return min(d, n - d) + + +def _find_crossings( + poly_a: np.ndarray, + poly_b: np.ndarray, + same_curve: bool = False, + min_seg_gap: int = 10, +) -> List[Dict]: + na = len(poly_a) - 1 + nb = len(poly_b) - 1 + + a_min_x = np.minimum(poly_a[:-1, 0], poly_a[1:, 0]) + a_max_x = np.maximum(poly_a[:-1, 0], poly_a[1:, 0]) + a_min_y = np.minimum(poly_a[:-1, 1], poly_a[1:, 1]) + a_max_y = np.maximum(poly_a[:-1, 1], poly_a[1:, 1]) + + b_min_x = np.minimum(poly_b[:-1, 0], poly_b[1:, 0]) + b_max_x = np.maximum(poly_b[:-1, 0], poly_b[1:, 0]) + b_min_y = np.minimum(poly_b[:-1, 1], poly_b[1:, 1]) + b_max_y = np.maximum(poly_b[:-1, 1], poly_b[1:, 1]) + + cell_size = max(np.median(np.concatenate([a_max_x - a_min_x, + b_max_x - b_min_x])), 1.0) * 3 + + grid_b = defaultdict(list) + for j in range(nb): + cx0 = int(b_min_x[j] / cell_size); cx1 = int(b_max_x[j] / cell_size) + cy0 = int(b_min_y[j] / cell_size); cy1 = int(b_max_y[j] / cell_size) + for gx in range(cx0, cx1 + 1): + for gy in range(cy0, cy1 + 1): + grid_b[(gx, gy)].append(j) + + checked = set() + details: List[Dict] = [] + for i in range(na): + cx0 = int(a_min_x[i] / cell_size); cx1 = int(a_max_x[i] / cell_size) + cy0 = int(a_min_y[i] / cell_size); cy1 = int(a_max_y[i] / cell_size) + for gx in range(cx0, cx1 + 1): + for gy in range(cy0, cy1 + 1): + if (gx, gy) not in grid_b: + continue + for j in grid_b[(gx, gy)]: + if same_curve: + ii, jj = min(i, j), max(i, j) + # Closed curve: neighbours wrap around. + if _circular_seg_gap(ii, jj, na) < min_seg_gap: + continue + key = (ii, jj) + else: + key = (i, j) + if key in checked: + continue + checked.add(key) + if same_curve: + si, sj = key + else: + si, sj = i, j + p0, p1 = poly_a[si], poly_a[si + 1] + q0, q1 = poly_b[sj], poly_b[sj + 1] + if not _segments_cross(p0, p1, q0, q1): + continue + d1 = p1 - p0 + d2 = q1 - q0 + denom = d1[0] * d2[1] - d1[1] * d2[0] + if abs(denom) < 1e-12: + continue + ti = ((q0[0] - p0[0]) * d2[1] - (q0[1] - p0[1]) * d2[0]) / denom + px = float(p0[0] + ti * d1[0]) + py = float(p0[1] + ti * d1[1]) + details.append({"px": px, "py": py}) + return details + + +def _details_to_points(details: List[Dict]) -> np.ndarray: + if not details: + return np.zeros((0, 2)) + return np.array([[d["px"], d["py"]] for d in details]) + + +def _curves_too_close( + poly_a: np.ndarray, + poly_b: np.ndarray, + same_curve: bool = False, + min_dist: float = 7.0, + sample_step: int = 3, + self_index_gap: int = 30, + known_crossings: np.ndarray | None = None, + crossing_exclude_radius: float = 55.0, +) -> bool: + nb = len(poly_b) - 1 + b_pts = poly_b[:-1] + b_vecs = poly_b[1:] - poly_b[:-1] + b_lens_sq = np.maximum((b_vecs ** 2).sum(axis=1), 1e-12) + + has_crossings = known_crossings is not None and len(known_crossings) > 0 + + for idx in range(0, len(poly_a), sample_step): + px, py = poly_a[idx] + p = np.array([px, py]) + dp = p - b_pts + t = (dp * b_vecs).sum(axis=1) / b_lens_sq + t = np.clip(t, 0.0, 1.0) + proj = b_pts + t[:, None] * b_vecs + dists = np.sqrt(((p - proj) ** 2).sum(axis=1)) + if same_curve: + # Closed curve: mask neighbours with wraparound distance. + seg_idx = np.arange(nb) + lin = np.abs(seg_idx - idx) + circ = np.minimum(lin, nb - lin) + mask = circ < self_index_gap + dists[mask] = 9999.0 + if dists.min() < min_dist: + if has_crossings: + cross_dists = np.sqrt(((p - known_crossings) ** 2).sum(axis=1)) + if cross_dists.min() < crossing_exclude_radius: + continue + return True + return False + + +# ── Instance sampling ────────────────────────────────────────────── + +QUESTION = ( + "How many distinct closed loops are tangled together in this image? " + "Each loop is a single continuous curve that closes back on itself — " + "there are no loose endpoints anywhere. All loops are drawn in the " + "same colour and may cross other loops or themselves freely. Count " + "the total number of distinct closed loops and report the count as " + "a positive integer. " + "Provide your final answer enclosed in ... tags." +) + + +def _min_crossing_angle_deg(poly_a: np.ndarray, poly_b: np.ndarray, + details: List[Dict], same_curve: bool = False) -> float: + """Return the smallest crossing angle (deg) across all crossings, or + 180.0 if there are no crossings.""" + if not details: + return 180.0 + # Re-detect with segment indices for angle computation. + na = len(poly_a) - 1 + nb = len(poly_b) - 1 + min_angle = 180.0 + for i in range(na): + p0, p1 = poly_a[i], poly_a[i + 1] + for j in range(nb): + if same_curve: + ii, jj = min(i, j), max(i, j) + if _circular_seg_gap(ii, jj, na) < 10: + continue + q0, q1 = poly_b[j], poly_b[j + 1] + if not _segments_cross(p0, p1, q0, q1): + continue + d1 = p1 - p0 + d2 = q1 - q0 + n1 = math.hypot(d1[0], d1[1]) + n2 = math.hypot(d2[0], d2[1]) + if n1 < 1e-9 or n2 < 1e-9: + continue + cos_a = (d1[0] * d2[0] + d1[1] * d2[1]) / (n1 * n2) + cos_a = max(-1.0, min(1.0, cos_a)) + ang = math.degrees(math.acos(abs(cos_a))) + if ang < min_angle: + min_angle = ang + return min_angle + + +def sample_instance( + rng: random.Random, + width: int, + height: int, + num_loops: int, + interior_margin: int = 55, + max_attempts: int = 600, + min_inter_crossings: int = 0, + max_self_crossings_per_loop: int = 0, + min_crossing_angle_deg: float = 30.0, +) -> Dict | None: + for _ in range(max_attempts): + polylines: List[np.ndarray] = [] + centers: List[Tuple[float, float]] = [] + build_failed = False + for _ in range(num_loops): + result = None + for _ in range(40): + result = build_closed_loop( + rng, width, height, + interior_margin=interior_margin, + existing_centers=centers, + ) + if result is not None: + break + if result is None: + build_failed = True + break + poly, centre = result + polylines.append(poly) + centers.append(centre) + if build_failed: + continue + + self_details: List[List[Dict]] = [] + pair_details: Dict[Tuple[int, int], List[Dict]] = {} + for a in range(num_loops): + self_details.append(_find_crossings(polylines[a], polylines[a], + same_curve=True)) + for b in range(a + 1, num_loops): + pair_details[(a, b)] = _find_crossings(polylines[a], polylines[b]) + + # Enforce: no self-crossings beyond allowed limit. + if any(len(sd) > max_self_crossings_per_loop for sd in self_details): + continue + + # Enforce: total inter-loop crossings >= min_inter_crossings. + total_inter = sum(len(v) for v in pair_details.values()) + if total_inter < min_inter_crossings: + continue + + # Enforce: every crossing has angle >= min_crossing_angle_deg. + bad_angle = False + for a in range(num_loops): + if _min_crossing_angle_deg(polylines[a], polylines[a], + self_details[a], same_curve=True) < min_crossing_angle_deg: + bad_angle = True + break + for b in range(a + 1, num_loops): + if _min_crossing_angle_deg(polylines[a], polylines[b], + pair_details[(a, b)]) < min_crossing_angle_deg: + bad_angle = True + break + if bad_angle: + break + if bad_angle: + continue + + too_close = False + for a in range(num_loops): + if _curves_too_close(polylines[a], polylines[a], same_curve=True, + known_crossings=_details_to_points(self_details[a])): + too_close = True + break + for b in range(a + 1, num_loops): + if _curves_too_close(polylines[a], polylines[b], + known_crossings=_details_to_points(pair_details[(a, b)])): + too_close = True + break + if too_close: + break + if too_close: + continue + + return { + "width": width, + "height": height, + "num_loops": num_loops, + "polylines": polylines, + "inter_loop_crossings": int(total_inter), + "question": QUESTION, + "answer": num_loops, + } + return None + + +# ── Rendering ────────────────────────────────────────────────────── + +def render_instance(out_path: Path, record: Dict, noise_seed: int, + thickness: float) -> None: + width = int(record["width"]) + height = int(record["height"]) + polylines = record["polylines"] + + fig = plt.figure(figsize=(width / 100, height / 100), dpi=100) + ax = fig.add_axes([0, 0, 1, 1]) + ax.set_xlim(0, width) + ax.set_ylim(height, 0) + ax.axis("off") + ax.set_facecolor("#f8f6f0") + + nrng = np.random.default_rng(noise_seed) + noise = nrng.normal(0.0, 1.0, size=(height, width)) + noise = (noise - noise.min()) / max(noise.max() - noise.min(), 1e-6) + ax.imshow(noise, cmap="Greys", alpha=0.05, extent=(0, width, height, 0), + interpolation="bilinear") + + for poly in polylines: + ax.plot(poly[:, 0], poly[:, 1], + color=LINE_COLOR, linewidth=thickness, alpha=0.92, + solid_capstyle="round", solid_joinstyle="round", + zorder=2.0) + + fig.savefig(out_path, dpi=100, bbox_inches="tight", pad_inches=0) + plt.close(fig) + + +# ── Main ─────────────────────────────────────────────────────────── + +def main() -> None: + parser = argparse.ArgumentParser() + parser.add_argument("--output-root", type=Path, required=True) + parser.add_argument("--count", type=int, default=30) + parser.add_argument("--seed", type=int, default=42) + parser.add_argument("--width", type=int, default=1024) + parser.add_argument("--height", type=int, default=1024) + parser.add_argument("--min-loops", type=int, default=3) + parser.add_argument("--max-loops", type=int, default=5) + parser.add_argument("--thickness", type=float, default=2.0, + help="Absolute pixel thickness; never scaled.") + parser.add_argument("--difficulty", type=int, default=5, + help="Integer difficulty >=0; scales loop count.") + parser.add_argument("--workers", type=int, default=8, + help="Parallel worker processes for sampling. 1 = serial.") + args = parser.parse_args() + + d = max(0, int(args.difficulty)) + + # Canvas scaling: N_d = 5 + d, N_0 = 5. + N_d = 5 + d + N_0 = 5 + s = math.sqrt(max(1.0, N_d / N_0)) + args.width = int(round(args.width * s)) + args.height = int(round(args.height * s)) + + # num_loops ∈ [3, 5 + d] + args.min_loops = 10 + args.max_loops = 10 + 2 * d + + # Fixed constraints. + max_self_crossings_per_loop = 0 + min_inter_crossings = 10 + 2 * d + min_crossing_angle_deg = 30.0 + loop_thickness_px = 4.0 # absolute; do not scale + + out_root: Path = args.output_root + img_dir = out_root / "images" + img_dir.mkdir(parents=True, exist_ok=True) + + sys.path.insert(0, str(Path(__file__).resolve().parents[3])) + from _sample_pool import parallel_sample_records # noqa: E402 + + # Force evenly-spaced answer counts across [min_loops, max_loops]. + if args.count > 1: + forced_targets = [ + int(round(args.min_loops + i * (args.max_loops - args.min_loops) / (args.count - 1))) + for i in range(args.count) + ] + else: + forced_targets = [args.min_loops] + print(f"forced loop counts: {forced_targets}") + + records_raw = [] + for ti, tgt in enumerate(forced_targets): + def _attempt(rng, _tgt=tgt): + rec = sample_instance( + rng, args.width, args.height, num_loops=_tgt, + min_inter_crossings=min_inter_crossings, + max_self_crossings_per_loop=max_self_crossings_per_loop, + min_crossing_angle_deg=min_crossing_angle_deg, + max_attempts=50, + ) + return rec + sub = parallel_sample_records( + _attempt, count=1, workers=args.workers, + seed_base=args.seed + ti * 977, + ) + records_raw.extend(sub) + rng_render = random.Random(args.seed ^ 0xA5A5) + records = [] + for idx, record in enumerate(records_raw): + name = f"tangled_loops_{idx:05d}.png" + ns = rng_render.randint(0, 10**9) + render_instance(img_dir / name, record, noise_seed=ns, + thickness=loop_thickness_px) + record.pop("polylines") + record["image"] = f"images/{name}" + records.append(record) + print(f" {len(records)}/{args.count} valid samples (workers={args.workers})") + + with (out_root / "annotations.jsonl").open("w") as fh: + for r in records: + fh.write(json.dumps(r) + "\n") + + data_json = { + "task": "tangled_loops", + "category": "distributed_scanning", + "count": len(records), + "items": [ + {"image": r["image"], "question": r["question"], "answer": r["answer"]} + for r in records + ], + } + (out_root / "data.json").write_text(json.dumps(data_json, indent=2)) + print(f"Saved {len(records)} items to {out_root}") + + +if __name__ == "__main__": + main() diff --git a/code/distributed_scanning/tangled_loops/data.json b/code/distributed_scanning/tangled_loops/data.json new file mode 100644 index 0000000000000000000000000000000000000000..52eb4e22de04423875f0e3547d436c8a950b4192 --- /dev/null +++ b/code/distributed_scanning/tangled_loops/data.json @@ -0,0 +1,32 @@ +{ + "task": "tangled_loops", + "category": "distributed_scanning", + "count": 5, + "items": [ + { + "image": "images/tangled_loops_00000.png", + "question": "How many distinct closed loops are tangled together in this image? Each loop is a continuous closed curve (rope, cord, cable, etc.) with no loose endpoints anywhere. Loops may cross other loops or themselves freely. Count the total number of distinct closed loops and report the count as a positive integer. Provide your final answer enclosed in ... tags.", + "answer": 10 + }, + { + "image": "images/tangled_loops_00001.png", + "question": "How many distinct closed loops are tangled together in this image? Each loop is a continuous closed curve (rope, cord, cable, etc.) with no loose endpoints anywhere. Loops may cross other loops or themselves freely. Count the total number of distinct closed loops and report the count as a positive integer. Provide your final answer enclosed in ... tags.", + "answer": 12 + }, + { + "image": "images/tangled_loops_00002.png", + "question": "How many distinct closed loops are tangled together in this image? Each loop is a continuous closed curve (rope, cord, cable, etc.) with no loose endpoints anywhere. Loops may cross other loops or themselves freely. Count the total number of distinct closed loops and report the count as a positive integer. Provide your final answer enclosed in ... tags.", + "answer": 15 + }, + { + "image": "images/tangled_loops_00003.png", + "question": "How many distinct closed loops are tangled together in this image? Each loop is a continuous closed curve (rope, cord, cable, etc.) with no loose endpoints anywhere. Loops may cross other loops or themselves freely. Count the total number of distinct closed loops and report the count as a positive integer. Provide your final answer enclosed in ... tags.", + "answer": 18 + }, + { + "image": "images/tangled_loops_00004.png", + "question": "How many distinct closed loops are tangled together in this image? Each loop is a continuous closed curve (rope, cord, cable, etc.) with no loose endpoints anywhere. Loops may cross other loops or themselves freely. Count the total number of distinct closed loops and report the count as a positive integer. Provide your final answer enclosed in ... tags.", + "answer": 20 + } + ] +} \ No newline at end of file diff --git a/code/gpt_image_prompts.json b/code/gpt_image_prompts.json new file mode 100644 index 0000000000000000000000000000000000000000..45bb8233af7b8d8c96ab9b48eac7e242529b4661 --- /dev/null +++ b/code/gpt_image_prompts.json @@ -0,0 +1,175 @@ +{ + "_comment": "One brief gpt-image-2 image-edit prompt per task. Prompts re-render the v4_plain matplotlib source as a photorealistic / natural variant while preserving the task's discriminative structure (positions, counts, labels). One sample (#00000) generated per task for review.", + "tasks": { + "distributed_scanning/attribute_group_counting": { + "src": "images/attribute_group_counting_00000.png", + "prompts": [ + "Re-render as a photorealistic top-down photograph of stones laid on a NOISY natural surface — pick ONE: damp riverbed gravel, mossy forest floor with twigs and leaves, dark wet beach sand with shell fragments, weathered wooden plank with knots and scuffs.\n\nEach blob in the source becomes ONE stone with the EXACT same silhouette, size, and position. ALL stones share the SAME shading, SAME colour, and SAME surface texture (one stone material across the whole image — e.g. uniformly dark slate, or uniformly pale river pebble). Stones differ ONLY in OUTLINE SHAPE; nothing else distinguishes them. Subtle natural lighting micro-variation per stone is fine, but no recolouring, no different materials, no patterned tops.\n\nThe stones must contrast STRONGLY with the surface (uniform dark stones on a pale sand surface, or uniform pale stones on dark moss) so each silhouette outline is unambiguously visible to a human at a glance. Background clutter (loose grain, dust, twigs) is allowed but must NEVER produce another stone-shaped blob and must NEVER intrude on the stone outlines.\n\nDo not add or remove any stone. No text, no labels, no overlays.", + "Re-render as a photorealistic overhead photograph of cookies laid out on a NOISY baker's surface — pick ONE: flour-dusted dark wooden board with crumbs, parchment-lined baking tray with butter smears, scratched dark stone counter with chocolate flakes, worn metal sheet with sugar specks.\n\nEach blob in the source becomes ONE cookie with the EXACT same silhouette, size, and position. ALL cookies share the SAME shading, SAME dough colour, and SAME surface texture (one cookie type across the whole image — e.g. uniformly golden shortbread, or uniformly dark chocolate cookie). Cookies differ ONLY in OUTLINE SHAPE; nothing else distinguishes them. Subtle bake-variation in tone is fine, but no different cookie types, no contrast garnishes inside.\n\nThe cookies must contrast STRONGLY with the surface (golden cookies on a dark counter, or dark cookies on a pale parchment) so every silhouette outline is unambiguously visible. Background clutter (crumbs, flour patches, butter smears) is allowed but must NEVER produce another cookie-shaped blob and must NEVER intrude on the cookie outlines.\n\nDo not add or remove any cookie. No text, no labels, no overlays.", + "Re-render as a photorealistic close-up overhead photograph of leaves scattered on a NOISY forest floor — pick ONE: damp dark soil with twigs and pine needles, mossy bark with lichen patches, dry oak-leaf litter, fallen autumn leaves with bark fragments.\n\nEach blob in the source becomes ONE leaf with the EXACT same silhouette, size, and position. ALL leaves share the SAME shading, SAME colour, and SAME surface texture (one leaf species across the whole image — e.g. uniformly bright yellow ginkgo, or uniformly dark green ivy). Leaves differ ONLY in OUTLINE SHAPE; nothing else distinguishes them. Subtle natural weathering per leaf is fine, but no different species, no veining variation that creates internal categories.\n\nThe leaves must contrast STRONGLY with the surface (bright yellow leaves on dark damp soil, or dark green leaves on pale dry mulch) so each silhouette outline is unambiguously visible. Background natural clutter (twigs, needles, dirt, lichen) is allowed but must NEVER produce another leaf-shaped blob and must NEVER intrude on the leaf outlines.\n\nDo not add or remove any leaf. No text, no labels, no overlays.", + "Re-render as a photorealistic top-down photograph of porcelain tea cups laid on a NOISY domestic surface — pick ONE: a worn wooden table with crumbs and a faint coffee ring, a marble counter with sugar specks, a chipped enamel tray with droplets, an old slate counter with smudges and dust.\n\nEach blob in the source becomes ONE porcelain tea cup with the EXACT same silhouette, size, and position. ALL tea cups share the SAME shading, SAME glaze colour (e.g. uniformly white-bone porcelain, OR uniformly pale-blue glaze), and SAME surface texture across the whole image. Cups differ ONLY in their OUTLINE SHAPE (the silhouette of the rim seen from above varies — round, oval, slightly squared); nothing else distinguishes them. Subtle natural lighting micro-variation per cup is fine, but no painted designs, no different glazes, no decorative rims that create internal categories.\n\nThe cups must contrast STRONGLY with the surface (white porcelain on dark wood, or dark glaze on pale marble) so each silhouette outline is unambiguously visible to a human at a glance. Background clutter (crumbs, dust, smudges) is allowed but must NEVER produce another cup-shaped blob and must NEVER intrude on the cup outlines.\n\nDo not add or remove any cup. No text, no labels, no overlays.", + "Re-render as a photorealistic top-down photograph of hand-pressed clay tiles laid on a NOISY workshop surface — pick ONE: a dusty potter's wooden bench with stray clay scraps, a stone-flagged kiln floor with ash flecks, a worn metal tray with terracotta dust, a chipped concrete slab with tool marks.\n\nEach blob in the source becomes ONE hand-pressed clay tile with the EXACT same silhouette, size, and position. ALL tiles share the SAME shading, SAME terracotta colour, and SAME surface texture (one clay material across the whole image — uniformly warm terracotta, slightly matte). Tiles differ ONLY in OUTLINE SHAPE; nothing else distinguishes them. Subtle firing variation in tone is fine, but no glazes, no painted patterns, no inlays that create internal categories.\n\nThe tiles must contrast STRONGLY with the surface (terracotta on dark stone, or dark tiles on pale concrete) so each silhouette outline is unambiguously visible to a human at a glance. Background clutter (clay scraps, dust, tool marks) is allowed but must NEVER produce another tile-shaped blob and must NEVER intrude on the tile outlines.\n\nDo not add or remove any tile. No text, no labels, no overlays." + ] + }, + "distributed_scanning/bounded_faces_counting": { + "src": "images/bounded_faces_counting_00000.png", + "prompts": [ + "Re-render as a wild, weathered, real-world top-down photograph. Each dot in the source becomes ONE driftwood stake hammered into a salt flat at dawn at the EXACT same position. Each line segment becomes ONE old rope strung taut between the SAME pair of stakes, following the line exactly with no slack, no extras, no missing strands. NOISY surface: cracked salt crusts, wet rust streaks, scattered shell fragments, drift-debris, faint footprints. Stakes and ropes sit visibly on top, never hidden by clutter. No text, no labels, no overlays.", + "Re-render as a wild, weathered, real-world top-down photograph. Each dot in the source becomes ONE rusted iron rebar peg driven into dry cracked earth at the EXACT same position. Each line segment becomes ONE weathered hemp twine knotted taut between the SAME pair of pegs, following the line exactly with no slack, no extras, no missing strands. NOISY surface: dry red dust, scattered pebbles, clumps of dead grass, tyre tracks, paint flecks, occasional metal shavings. Rebar and twine sit visibly on top, never hidden by clutter. No text, no labels, no overlays.", + "Re-render as a wild, weathered, real-world top-down photograph. Each dot in the source becomes ONE mossy stone bollard on a damp riverbank at the EXACT same position. Each line segment becomes ONE braided seaweed strand stretched taut between the SAME pair of bollards, following the line exactly with no slack, no extras, no missing strands. NOISY surface: damp pebbles, fallen leaves, lichen crust, twigs, water-pooled grit, micro-puddles. Stones and seaweed sit visibly on top, never hidden by clutter. No text, no labels, no overlays.", + "Re-render as a wild, weathered, real-world top-down photograph. Each dot in the source becomes ONE wooden tent stake hammered into a dry meadow at the EXACT same position. Each line segment becomes ONE taut nylon guy-line strung between the SAME pair of stakes, following the line exactly with no slack, no extras, no missing strands. NOISY surface: dry yellow grass tufts, scattered wildflowers, twigs, faint footpaths, occasional pebbles. Stakes and guy-lines sit visibly on top, never hidden by clutter. No text, no labels, no overlays.", + "Re-render as a wild, weathered, real-world top-down photograph. Each dot in the source becomes ONE weathered wooden telegraph pole stub seen from directly above standing in cracked earth at the EXACT same position. Each line segment becomes ONE sagging black copper wire stretched between the SAME pair of pole tops, following the line exactly with no slack, no extras, no missing strands. NOISY surface: dry red soil with deep cracks, scattered scrub-grass tufts, pebbles, tyre tracks, rust flecks. Pole stubs and wires sit visibly on top, never hidden by clutter. No text, no labels, no overlays." + ] + }, + "distributed_scanning/counting_connected_components": { + "src": "images/counting_connected_components_00000.png", + "prompts": [ + "Re-render as a wild, weathered, real-world top-down photograph. Each dot in the source becomes ONE driftwood stake hammered into a salt flat at dawn at the EXACT same position. Each line segment becomes ONE old rope strung taut between the SAME pair of stakes, following the line exactly with no slack, no extras, no missing strands. NOISY surface: cracked salt crusts, wet rust streaks, scattered shell fragments, drift-debris, faint footprints. Stakes and ropes sit visibly on top, never hidden by clutter. No text, no labels, no overlays.", + "Re-render as a wild, weathered, real-world top-down photograph. Each dot in the source becomes ONE rusted iron rebar peg driven into dry cracked earth at the EXACT same position. Each line segment becomes ONE weathered hemp twine knotted taut between the SAME pair of pegs, following the line exactly with no slack, no extras, no missing strands. NOISY surface: dry red dust, scattered pebbles, clumps of dead grass, tyre tracks, paint flecks, occasional metal shavings. Rebar and twine sit visibly on top, never hidden by clutter. No text, no labels, no overlays.", + "Re-render as a wild, weathered, real-world top-down photograph. Each dot in the source becomes ONE mossy stone bollard on a damp riverbank at the EXACT same position. Each line segment becomes ONE braided seaweed strand stretched taut between the SAME pair of bollards, following the line exactly with no slack, no extras, no missing strands. NOISY surface: damp pebbles, fallen leaves, lichen crust, twigs, water-pooled grit, micro-puddles. Stones and seaweed sit visibly on top, never hidden by clutter. No text, no labels, no overlays.", + "Re-render as a wild, weathered, real-world top-down photograph. Each dot in the source becomes ONE wooden tent stake hammered into a dry meadow at the EXACT same position. Each line segment becomes ONE taut nylon guy-line strung between the SAME pair of stakes, following the line exactly with no slack, no extras, no missing strands. NOISY surface: dry yellow grass tufts, scattered wildflowers, twigs, faint footpaths, occasional pebbles. Stakes and guy-lines sit visibly on top, never hidden by clutter. No text, no labels, no overlays.", + "Re-render as a wild, weathered, real-world top-down photograph. Each dot in the source becomes ONE weathered wooden telegraph pole stub seen from directly above standing in cracked earth at the EXACT same position. Each line segment becomes ONE sagging black copper wire stretched between the SAME pair of pole tops, following the line exactly with no slack, no extras, no missing strands. NOISY surface: dry red soil with deep cracks, scattered scrub-grass tufts, pebbles, tyre tracks, rust flecks. Pole stubs and wires sit visibly on top, never hidden by clutter. No text, no labels, no overlays." + ] + }, + "distributed_scanning/counting_regions": { + "src": "images/counting_regions_00000.png", + "prompts": [ + "Re-render as a photorealistic top-down aerial photograph of a vast UNIFORM grass meadow, where every region shares the EXACT same tall green grass texture. Inside every region the ground is identical mature meadow grass with no biome variation, no colour variation, no density variation, no patches of any kind.\n\nBoundaries are the ONLY feature separating regions, and they are drawn as ANIMAL TRAILS — thin worn paths threaded through the grass where countless small animals (deer, foxes, badgers) have walked over time. Each trail follows the source partition curve EXACTLY. The trails are deliberately HARD for naive computer-vision tools to detect:\n • The grass on BOTH sides of a trail is the same height, same colour, same species — only the trail itself shows slightly trampled grass and bare soil glimpses.\n • Trail width is irregular along its length — sometimes 4 cm, sometimes 8 cm, occasionally pinched almost shut where grass has grown back.\n • Trails are NOT smooth painted lines: they are noisy, broken, partly overgrown, with stray grass blades crossing them.\n • There are NO hard high-contrast edges anywhere: the soil shows through only in faint patches.\n • The grass texture must remain dominant across the entire image; the trails are subtle, narrow, and nearly camouflaged — a careful human can trace them by following the slightly trampled line, but a thresholding/edge-detection algorithm would see only one big texture.\n\nTrails must trace the source partition curve faithfully on every twist and turn, with NO gaps, NO breaks, NO doubled lines, NO new trails inside a region. Adjacent regions are distinguishable ONLY by spotting these subtle trails between them.\n\nNo text, no labels, no compass, no overlays, no animals visible, no people, no fences, no other features.", + "Re-render as a photorealistic top-down aerial photograph of a UNIFORM expanse of fresh white snow at dawn, where every region shares the EXACT same fresh-snow texture. Inside every region the surface is identical undisturbed powdery snow — no biome variation, no colour variation, no patches of any kind.\n\nBoundaries are the ONLY feature separating regions, and they are drawn as SINGLE-FILE FOOTPRINT TRAILS — a line of HIGHLY IRREGULAR footprints (mix of adult boots of different sizes, smaller children's boots, bare-foot prints, animal paw-prints from foxes/deer/rabbits, snowshoe imprints) where various walkers crossed along each boundary. Each trail follows the source partition curve EXACTLY. The trails are deliberately HARD for naive computer-vision tools to detect:\n • The snow on BOTH sides of a footprint trail is the same fresh untouched powder — only the prints themselves show as small dimpled depressions in the snow with soft blue-grey shadows.\n • The footprints are DISCRETE — small rounded boot shapes spaced ~25–40 cm apart along the trail, NOT a continuous line. To follow a trail a viewer must mentally CONNECT THE DOTS along the source curve.\n • Footprint SHAPE varies along each trail (different boots, different paws, even occasional bare-foot or snowshoe), and depth, orientation, and stride length vary too. The trail meanders slightly off the source curve (within ~3 cm) before snapping back (some prints sharper, some softer where wind has dusted snow back into them).\n • There are NO hard linear edges anywhere; each footprint is an isolated small feature, only their collective alignment along the source curve reveals the boundary.\n • The snow surface ALSO contains MANY NOISY DISTRACTORS unrelated to any boundary: scatter ~80–150 random isolated features across the entire canvas — stray paw-prints in random positions, dropped twigs, fallen pine cones, small icy lumps, dimpled wind hollows, scattered red berries, fallen brown leaves blown over the snow, occasional small rabbit-droppings. These distractors are randomly placed and NOT aligned with any partition curve. The vast majority of the image is still uniform fresh snow — a careful human can trace each boundary by visually chaining its footprints, but a basic edge-detector would just register thousands of unrelated little blobs.\n\nFootprint trails must trace the source partition curve faithfully on every twist and turn, with NO gaps too long for a human to chain (max ~50 cm between successive prints), NO doubled trails, NO new trails inside a region. Adjacent regions are distinguishable ONLY by spotting these subtle dotted lines.\n\nNo text, no labels, no compass, no overlays, no people visible, no animals, no other features.", + "Re-render as a photorealistic top-down close-up photograph of one large continuous patch of dry beige sand on a beach, covering the ENTIRE canvas. The sand has natural organic variation: subtle ripples from wind, scattered grain-size differences, micro-shadow patterns, occasional tiny pebbles, broken shell fragments, mica flakes, dropped sesame-sized debris, beetle exoskeletons, dried seaweed bits, salt crystals — strew ~150–250 small dark distractor specks randomly across the canvas, NOT aligned with any partition curve — these texture features flow continuously across the whole image with NO RESPECT for the partition.\n\nThe texture variation is INDEPENDENT of the regions: a wind-ripple band can sit anywhere on the sand, span across multiple regions, or fall entirely inside one region. Inside every region the sand has the EXACT same character (same beige hue, same grain texture, same ambient ripple distribution).\n\nBoundaries are the ONLY feature separating regions, drawn as TINY ANT MARCH LINES — a single-file column of small black ants moving along the boundary. Each line follows the source partition curve EXACTLY. The trails are deliberately HARD for naive computer-vision tools to detect:\n • The sand on BOTH sides of every ant march is identical — only the ants themselves break the uniform sand texture.\n • The ants are DISCRETE small black specks (1–2 mm visible diameter from above) spaced 2–4 mm apart along the line. They do NOT form a continuous edge — to read a boundary the viewer must mentally CHAIN successive ants along the source curve.\n • Some ants are crisp, some blurry from motion; gaps occasionally widen up to 1 cm where the column thins; occasional outlier ants stray briefly off the line and back.\n • The image must look CHAOTICALLY NOISY — hundreds of small dark dots scattered everywhere, only a small fraction of them being ants on the partition curves. There is NO continuous high-contrast edge anywhere — only thousands of small dark specks. A basic edge-detector returns just blob noise without alignment reasoning; a careful human reads the line by visual gestalt.\n\nAnt march lines must trace the source partition curve faithfully on every twist and turn, with NO gaps too long for a human to chain (max ~1 cm between successive ants), NO doubled lines, NO new lines inside a region. Adjacent regions are distinguishable ONLY by chaining these dotted ant trails.\n\nNo text, no labels, no compass, no overlays, no other features.", + "Re-render as a photorealistic top-down macro photograph of a single sheet of aged cream-coloured parchment paper covering the ENTIRE canvas. The parchment surface has natural variation across the whole image — subtle tea-staining, faint fold creases, water spots, and aged fibres flowing continuously, with NO RESPECT for the partition.\n\nCRITICAL — the distractors and the boundary are the SAME material (dark brown/black ink) but DIFFERENT SHAPE CATEGORIES:\n • Scatter 30–50 SHORT INK MARKS across the parchment, randomly positioned and randomly oriented: small ink blots, brief signatures, single-letter scribbles, smudges, fingerprint smears, short underline strokes (each 0.5–4 cm long). They look like accidental marks left by hand. They are SHORT and DO NOT trace the partition curve.\n • Add 10–20 SLIGHTLY LONGER ink doodles — small loops, simple flourishes, abstract squiggles (each 4–8 cm). Still localised; do NOT span across the canvas.\n • All marks are SAME ink colour and SAME ink texture as the boundary (so a shape-classifier sees them as the same object class).\n\nBoundaries must trace the source partition curve, but the BOUNDARY ITSELF MUST BE NOISY (not a crisp continuous line) so that classical edge / Hough / line-detection CV pipelines fail:\n • The boundary is rendered as a SERIES OF DISJOINT INK DASHES of irregular length (each 1–3 cm), with small visible GAPS between dashes (0.3–1 cm gaps), so the line is BROKEN as a pixel-string.\n • Dash thickness VARIES along the curve (1–4 mm), and ink OPACITY varies (some dashes dark and saturated, others faint and washed-out as if the quill was running dry).\n • Occasional ink BLOTS, small smudges, and tiny accidental flicks deviate slightly off the partition curve (within 1–2 mm of the true curve), so the boundary has a hand-drawn, noisy feel rather than a clean drafted line.\n • Despite the noise, every dash sits ON the source partition curve so a human can still TRACE the curve by chaining the dashes visually. Wherever a partition curve touches the canvas border in the source, the dashed boundary in the output must EXTEND ALL THE WAY TO THE CORRESPONDING IMAGE EDGE — do NOT inset, crop, fade out, or stop the dashes before the edge. Dashes continue right up to the page edge.\n • The boundary is the SAME ink colour and texture as the distractor scribbles, distinguishable from them ONLY by the dashes' COLLECTIVE alignment along the partition curve through every twist and turn.\n\nThe boundary must follow the source partition faithfully on every twist with NO doubled lines and NO new boundaries inside any region. Distractor scribbles must NOT trace the partition.\n\nA naive edge-detector finds a noisy mess of unconnected ink fragments. A careful human reads the boundary by gestalt continuity along each curve.\n\nNo text other than the ink scribbles, no labels, no compass, no overlays, no other features.", + "Re-render as a photorealistic top-down close-up photograph of a wet sandy beach at low tide. The sand — pale beige, faint wind ripples, micro shadows, continuous grain texture — fills the ENTIRE canvas, flowing continuously with NO RESPECT for the partition.\n\nCRITICAL — the beach is strewn with many ROPE-LIKE LINEAR distractors that look visually similar to the real boundary rope:\n • Scatter 30–60 discarded SHORT ROPE SCRAPS of varied lengths (10 cm – 1.5 m), varied colours and weaves (manila / sisal / nylon / faded synthetic), randomly oriented across the canvas. They are short and DO NOT trace the partition curve.\n • Scatter 30–60 long pieces of SEAWEED, KELP STRANDS, DRIFTWOOD, AND DRIED VINES — all narrow linear features, looking ropy at first glance, randomly oriented across the canvas, NOT aligned with the partition.\n • Add 20–40 ropy crab-track grooves and worm-trail marks in the sand — narrow squiggly lines randomly placed, NOT aligned with the partition.\n • Add 60–120 scattered shells, pebbles, and small drift debris for general clutter.\n\nSo the image contains DOZENS of rope-like and linear-strand-like features, none of which — except one — trace the partition. Edge-detection / line-detection / shape-classifier CV will find HUNDREDS of false positives.\n\nBoundaries are the ONLY feature that traces the partition: a CHAIN OF MEDIUM-LENGTH MANILA ROPE SEGMENTS (each 25–60 cm long, ~2 cm thick, weathered golden-brown, twisted-fibre texture) lying flat on the sand, end-to-end with VERY SMALL gaps (5–15 mm) between successive segments. The boundary reads as a CLEARLY CONTINUOUS rope trail to a human — each segment touches or nearly touches the next, with only a hairline gap. The gaps are small enough that a human eye instantly chains them as one rope, while still being technically broken to defeat continuous-line CV detectors. Together the segments trace the source partition curve EXACTLY through every twist and turn. The boundary rope segments are visually nearly identical to the many short rope-scrap distractors (same material, same thickness, same texture) — distinguishable from them ONLY by their COLLECTIVE alignment along the partition curve.\n\nA human can trace the boundary easily because the rope segments are long and barely separated, reading as a near-continuous rope line. A naive line/shape detector cannot, because dozens of similar-looking ropy features lie everywhere.\n\nThe disjoint rope segments must collectively trace the source partition curve faithfully on every twist; gaps between segments are intentional but very small (5–15 mm) so the rope reads as continuous to humans. NO doubled lines. There must be NO new boundary ropes inside any region. The rope must NEVER be hidden under clutter — it sits clearly visible on top.\n\nNo text, no labels, no compass, no overlays, no people, no boats." + ] + }, + "distributed_scanning/tangled_loops": { + "src": "images/tangled_loops_00000.png", + "prompts": [ + "Re-render as a photorealistic top-down photograph of ropes lying tangled on a NOISY real-world surface — pick ONE: an old workshop floor with scattered sawdust, screws, and grease patches; a damp forest floor with fallen leaves, twigs, moss, and small pebbles; a worn warehouse concrete floor stained with paint splatters, grime, and dust; a low-tide beach with shells, kelp scraps, and wet sand ripples. The surface should look genuinely textured and lived-in (real-world camera grain noise, varied lighting), not a clean studio backdrop.\n\nALL loops must be the SAME uniform colour and SAME material (one type of rope) chosen to CONTRAST sharply with the surface so every strand stays unambiguously readable on top. Each loop follows the EXACT same path as in the source, with proper over/under crossings at every intersection. Loops must NEVER be hidden, broken, or interrupted by background clutter — they sit visibly on top.\n\nPreserve every loop. Do not add or remove any loop. No text, no labels, no overlays.", + "Re-render as a photorealistic top-down photograph of tangled cords lying on a NOISY cluttered real-world surface — pick ONE: a workshop floor strewn with sawdust, screws, and metal shavings; a damp forest floor with fallen leaves, twigs, and moss; an attic floor with dust, old papers, and crumpled boxes; a beach at low tide with shells, kelp, and pebbles. ALL loops must be the SAME uniform colour and SAME material (e.g., bright orange paracord, glossy black rubber cable, white nylon rope) chosen to CONTRAST sharply with the noisy background so each strand remains unambiguously readable on top. Each loop follows the EXACT same path as in the source, with proper over/under crossings at every intersection. Cords must NEVER be hidden or interrupted by background clutter. Preserve every loop. Do not add or remove any loop. No text, no labels.", + "Re-render as a photorealistic top-down photograph of a CHAOTIC pile of tangled identical objects on a busy real-world setting — pick ONE: bright orange extension cords on a workshop bench scattered with tools, gaffer tape, and dust; identical silver chains spilled across a black velvet jewellery tray cluttered with mixed beads, buttons, and clasps; coiled garden hoses on a patio strewn with leaves, sprinkler heads, and gravel; black stage cables on a backstage floor with crumpled tape, paper, and shoe scuffs. ALL loops must be the SAME uniform colour and SAME material; the surrounding clutter contains many similar-looking but DIFFERENT short bits (frayed ends, smaller cords, scrap pieces) that are NOT part of any loop. Each real loop follows the EXACT same path as in the source, with proper over/under crossings; the loops always sit visibly on top of the clutter and are never broken by it. Preserve every loop. Do not add or remove any loop. No text, no labels.", + "Re-render as a photorealistic top-down photograph of bright climbing rope tangled on a NOISY rocky alpine ledge — scattered carabiners, chalk dust, small pebbles, lichen patches, snow-melt streaks. ALL loops must be SAME uniform colour and material (e.g. high-vis orange dynamic climbing rope, OR royal-blue accessory cord — pick ONE per image, same across every loop) chosen to CONTRAST sharply with the rock and clutter so each strand stays unambiguously readable on top. Each loop follows the EXACT same path as in the source, with proper over/under crossings at every intersection. Loops must NEVER be hidden, broken, or interrupted by background clutter — they sit visibly on top.\n\nPreserve every loop. Do not add or remove any loop. No text, no labels, no overlays.", + "Re-render as a photorealistic top-down photograph of dark electrical cables piled on a NOISY backstage floor — strewn gaffer-tape scraps, set-piece edges, tape gel fragments, scuff marks, leftover rosin dust. ALL loops must be SAME uniform colour and material (e.g. matte-black rubber-jacketed XLR cable, OR thick yellow stage extension cord — pick ONE per image, same across every loop) chosen to CONTRAST sharply with the floor and clutter so each strand stays unambiguously readable on top. Each loop follows the EXACT same path as in the source, with proper over/under crossings at every intersection. Cables must NEVER be hidden, broken, or interrupted by background clutter — they sit visibly on top.\n\nPreserve every loop. Do not add or remove any loop. No text, no labels, no overlays." + ] + }, + "sequential_traversal/arrow_chain": { + "src": "images/arrow_chain_00000.png", + "prompts": [ + "⚠️ PIXEL-EXACT (NON-NEGOTIABLE):\n • Every footprint stays at the EXACT same (x, y) within ~2 px AND points in the EXACT same direction within ~3° — do NOT move, rotate, mirror, scale, or re-orient any footprint.\n • Every labeled circle keeps the SAME letter and the SAME (x, y) position. Letter glyphs unchanged.\n • Object COUNT equals the source EXACTLY — never add, skip, or merge.\n • Position and direction are the load-bearing signal of this image; nothing about geometry may change.\n • The single chain cell with a GREEN ring around it is the START marker. Preserve the green ring CLEARLY in the rendered image — a bright vivid green circle of the SAME size and at the SAME position as in the source. Do NOT remove, recolour, or fade this green start ring; it is critical to the question.\n\nRe-render as a photorealistic top-down photograph of damp beach sand. Every footprint in the source becomes ONE realistic shoe/boot footprint pressed into the sand, in the SAME position, pointing the SAME direction (toes = direction). Each labeled circle becomes a flat, smooth wet pebble with the SAME letter painted in white. Sand has natural ripples and grain texture; lighting is even and soft. No text other than the letters on pebbles.", + "⚠️ PIXEL-EXACT (NON-NEGOTIABLE):\n • Every fish stays at the EXACT same (x, y) within ~2 px AND points in the EXACT same direction within ~3° — do NOT move, rotate, mirror, scale, or re-orient any fish.\n • Every labeled circle keeps the SAME letter and the SAME (x, y) position. Letter glyphs unchanged.\n • Object COUNT equals the source EXACTLY — never add, skip, or merge.\n • Position and direction are the load-bearing signal of this image; nothing about geometry may change.\n • The single chain cell with a GREEN ring around it is the START marker. Preserve the green ring CLEARLY in the rendered image — a bright vivid green circle of the SAME size and at the SAME position as in the source. Do NOT remove, recolour, or fade this green start ring; it is critical to the question.\n\nRe-render as a photorealistic top-down photograph of clear shallow tropical water over white sand. Every fish in the source becomes ONE small tropical fish swimming flat over the sand at the SAME position, pointing the SAME direction (head = direction). Each labeled circle becomes a flat round coral disc bearing the SAME letter. Water shows soft caustic ripples; sand has natural grain. No text other than the letters on coral.", + "⚠️ PIXEL-EXACT (NON-NEGOTIABLE):\n • Every bird stays at the EXACT same (x, y) within ~2 px AND points in the EXACT same direction within ~3° — do NOT move, rotate, mirror, scale, or re-orient any bird.\n • Every labeled circle keeps the SAME letter and the SAME (x, y) position. Letter glyphs unchanged.\n • Object COUNT equals the source EXACTLY — never add, skip, or merge.\n • Position and direction are the load-bearing signal of this image; nothing about geometry may change.\n • The single chain cell with a GREEN ring around it is the START marker. Preserve the green ring CLEARLY in the rendered image — a bright vivid green circle of the SAME size and at the SAME position as in the source. Do NOT remove, recolour, or fade this green start ring; it is critical to the question.\n\nRe-render as a photorealistic top-down photograph of a flock of small birds flying over a pale overcast sky. Every bird in the source becomes ONE small bird in flight at the SAME position, beak/head pointing the SAME direction. Each labeled circle becomes a small painted wooden disc bearing the SAME letter, suspended in the same plane. Sky has subtle cloud variation; lighting is even. No text other than the letters on discs.", + "⚠️ PIXEL-EXACT (NON-NEGOTIABLE):\n • Every key stays at the EXACT same (x, y) within ~2 px AND points in the EXACT same direction within ~3° — do NOT move, rotate, mirror, scale, or re-orient any key.\n • Every labeled circle keeps the SAME letter and the SAME (x, y) position. Letter glyphs unchanged.\n • Object COUNT equals the source EXACTLY — never add, skip, or merge.\n • Position and direction are the load-bearing signal of this image; nothing about geometry may change.\n • The single chain cell with a GREEN ring around it is the START marker. Preserve the green ring CLEARLY in the rendered image — a bright vivid green circle of the SAME size and at the SAME position as in the source. Do NOT remove, recolour, or fade this green start ring; it is critical to the question.\n\nRe-render as a photorealistic top-down photograph of an old wooden workbench. Every key in the source becomes ONE antique brass key lying flat on the wood at the SAME position, with the bow end pointing the SAME direction (the pointy/shaft end). Each labeled circle becomes a small leather tag stamped with the SAME letter. Wood has natural grain and minor scratches; lighting is warm and soft. No text other than the letters on tags.", + "⚠️ PIXEL-EXACT (NON-NEGOTIABLE):\n • Every leaf stays at the EXACT same (x, y) within ~2 px AND points in the EXACT same direction within ~3° — do NOT move, rotate, mirror, scale, or re-orient any leaf.\n • Every labeled circle keeps the SAME letter and the SAME (x, y) position. Letter glyphs unchanged.\n • Object COUNT equals the source EXACTLY — never add, skip, or merge.\n • Position and direction are the load-bearing signal of this image; nothing about geometry may change.\n • The single chain cell with a GREEN ring around it is the START marker. Preserve the green ring CLEARLY in the rendered image — a bright vivid green circle of the SAME size and at the SAME position as in the source. Do NOT remove, recolour, or fade this green start ring; it is critical to the question.\n\nRe-render as a photorealistic top-down photograph of forest floor with damp earth and moss. Every leaf in the source becomes ONE real fallen leaf lying flat on the soil at the SAME position, with its TIP pointing the SAME direction as in the source (stem opposite the tip). Each labeled circle becomes a small smooth river stone bearing the SAME letter painted in white. The soil has scattered fine debris (twig fragments, soil grains); lighting is overcast and even. No text other than the letters on stones." + ] + }, + "sequential_traversal/color_zone_sequence": { + "src": "images/color_zone_00000.png", + "prompts": [ + "Re-render as a hand-drawn stylised tourist map (think: fantasy / RPG world map, or vintage national-park trail brochure). The output should look LIKE A DRAWN MAP — flat colour regions, clean borders, decorative shading — not a photorealistic aerial photo. Stylisation makes precise geometry easier to preserve.\n\nGEOMETRY MUST BE PRESERVED EXACTLY:\n • Each coloured region stays at the EXACT same shape, position, and adjacency. Replace each region's flat colour with one biome icon-fill (forest dotted with little trees / lake with light wave hatching / meadow with grass tufts / desert with sand dots / mountains with peak symbols). Choose ONE biome per region; do NOT mix biomes inside a region; do NOT bleed a biome across a boundary.\n • Boundaries between regions are thin dark hand-drawn ink lines, traced EXACTLY along the source partition curves.\n • Every black curve in the source becomes a SINGLE UNIFORM-STYLE trail in the output. ALL curves — the query curve AND every distractor curve — must be rendered IDENTICALLY: same colour, same dash pattern, same line weight, same texture. The viewer must NOT be able to tell which curve is the query just by looking at the lines themselves; the only distinguishing feature allowed is the small black starting dot on the query curve. Trace each source curve EXACTLY (do NOT smooth, shorten, reroute, or break any curve). • Each single-letter region label stays at the EXACT same position with the SAME glyph, drawn in a small map-label box.\n\nNo new regions, no new trails, no new labels. No photorealism, no 3D perspective. Pure flat illustrated map look. Do not colour-code or style-code the trails differently from each other.\n\nThe source image contains a literal 'Start' text label inside a small white box with a thin black border, placed at the beginning of the query curve. PRESERVE THIS 'Start' LABEL EXACTLY AS-IS in the output: same wording (\"Start\"), same position, same small white box with thin black border, same legibility. Do NOT replace the label with any icon, do NOT translate the word, do NOT restyle the box. The 'Start' label is the ONLY distinguishing mark between the query curve and the distractor curves; every curve must otherwise look identical (same color, same dash pattern, same weight, same texture).", + "Re-render as a vintage maritime nautical chart on aged parchment (think: 18th-century sailor's chart with calligraphic flourishes). Flat-style cartography — NOT a photorealistic ocean photo. Stylisation makes precise geometry easier to preserve.\n\nGEOMETRY MUST BE PRESERVED EXACTLY:\n • Each coloured region stays at the EXACT same shape, position, and adjacency. Replace each region's flat colour with ONE distinctive island-or-sea texture: deep ocean with tiny waves / sandbar with fine stippling / coral reef with cross-hatching / mangrove with leaf flecks / kelp forest with curling strokes / open beach with dotted sand. ONE texture per region; do NOT mix textures inside a region; do NOT bleed a texture across a boundary.\n • Boundaries between regions are thin sepia-ink hand-drawn coastlines, traced EXACTLY along the source partition curves.\n • Every black curve in the source becomes a SINGLE UNIFORM-STYLE rhumb line in the output. ALL curves — the query curve AND every distractor curve — must be rendered IDENTICALLY: same sepia ink colour, same uniform thin solid stroke, same weight, same texture. The viewer must NOT be able to tell which curve is the query just by looking at the lines themselves; the only distinguishing feature allowed is the literal text 'Start' label sitting on the query curve. Trace each source curve EXACTLY (do NOT smooth, shorten, reroute, or break any curve). • Each single-letter region label stays at the EXACT same position with the SAME glyph, drawn in a small calligraphic cartouche.\n\nNo new regions, no new lines, no new labels. No photorealism, no 3D perspective. Pure flat antique chart look. Do not colour-code or style-code the rhumb lines differently from each other.\n\nThe source image contains a literal 'Start' text label inside a small white box with a thin black border, placed at the beginning of the query curve. PRESERVE THIS 'Start' LABEL EXACTLY AS-IS in the output: same wording (\"Start\"), same position, same small white box with thin black border, same legibility. Do NOT replace the label with any icon, do NOT translate the word, do NOT restyle the box. The 'Start' label is the ONLY distinguishing mark between the query curve and the distractor curves; every curve must otherwise look identical (same color, same dash pattern, same weight, same texture).", + "Re-render as a flat children's-storybook watercolour map (think: a friendly hand-painted nursery storybook). Flat illustrated style — NOT a photorealistic aerial photo. Stylisation makes precise geometry easier to preserve.\n\nGEOMETRY MUST BE PRESERVED EXACTLY:\n • Each coloured region stays at the EXACT same shape, position, and adjacency. Replace each region's flat colour with ONE storybook biome wash (a meadow of tiny flowers / a cosy forest of round trees / a candy-pink desert with cacti / a pastel pond with lily pads / a soft mountain pass with tiny rocks / a snow patch with little snowflakes). ONE biome per region; do NOT mix biomes inside a region; do NOT bleed a biome across a boundary.\n • Boundaries between regions are thin charcoal hand-drawn outlines, traced EXACTLY along the source partition curves.\n • Every black curve in the source becomes a SINGLE UNIFORM-STYLE pencil dotted path in the output. ALL curves — the query curve AND every distractor curve — must be rendered IDENTICALLY: same charcoal colour, same dotted pattern, same dot spacing, same dot size. The viewer must NOT be able to tell which curve is the query just by looking at the lines themselves; the only distinguishing feature allowed is the literal text 'Start' label sitting on the query curve. Trace each source curve EXACTLY (do NOT smooth, shorten, reroute, or break any curve). • Each single-letter region label stays at the EXACT same position with the SAME glyph, drawn in a small painted ribbon banner.\n\nNo new regions, no new paths, no new labels. No photorealism, no 3D perspective. Pure flat storybook watercolour look. Do not colour-code or style-code the paths differently from each other.\n\nThe source image contains a literal 'Start' text label inside a small white box with a thin black border, placed at the beginning of the query curve. PRESERVE THIS 'Start' LABEL EXACTLY AS-IS in the output: same wording (\"Start\"), same position, same small white box with thin black border, same legibility. Do NOT replace the label with any icon, do NOT translate the word, do NOT restyle the box. The 'Start' label is the ONLY distinguishing mark between the query curve and the distractor curves; every curve must otherwise look identical (same color, same dash pattern, same weight, same texture).", + "Re-render as a flat vintage transit map (think: classic mid-century subway diagram printed on cream paper). The output should look LIKE A PRINTED MAP — flat colour districts, clean borders, decorative typography — not a photorealistic aerial photo. Stylisation makes precise geometry easier to preserve.\n\nGEOMETRY MUST BE PRESERVED EXACTLY:\n • Each coloured region stays at the EXACT same shape, position, and adjacency. Replace each region's flat colour with ONE solid printed-ink district fill (mauve, ochre, mint, rose, navy, sand, slate, terracotta) — choose ONE colour per region; do NOT mix colours inside a region; do NOT bleed colour across a boundary.\n • Inside each district, scatter 3–6 tiny printed landmark icons appropriate to a transit map (small church spire, tree, square dot for park, fountain symbol, monument, library glyph, fountain, plus-sign for hospital, train station icon). Vary icons across districts so each district has a slightly different mix of small landmarks. NEVER let an icon bleed across a boundary.\n • Boundaries between regions are thin matte black printed-ink lines, traced EXACTLY along the source partition curves.\n • Every black curve in the source becomes a SINGLE UNIFORM-STYLE printed metro line in the output. ALL curves — the query curve AND every distractor curve — must be rendered IDENTICALLY: same dark-grey ink colour, same uniform thin solid stroke, same weight, same texture. The viewer must NOT be able to tell which curve is the query just by looking at the lines themselves; the only distinguishing feature allowed is the literal text 'Start' label sitting on the query curve. Trace each source curve EXACTLY (do NOT smooth, shorten, reroute, or break any curve).\n • Each single-letter region label stays at the EXACT same position with the SAME glyph, set in a small printed serif tag.\n\nNo new regions, no new lines, no new labels. No photorealism, no 3D perspective. Pure flat printed transit-map look. Do not colour-code or style-code the lines differently from each other.\n\nThe source image contains a literal 'Start' text label inside a small white box with a thin black border, placed at the beginning of the query curve. PRESERVE THIS 'Start' LABEL EXACTLY AS-IS in the output: same wording (\"Start\"), same position, same small white box with thin black border, same legibility. Do NOT replace the label with any icon, do NOT translate the word, do NOT restyle the box. The 'Start' label is the ONLY distinguishing mark between the query curve and the distractor curves; every curve must otherwise look identical (same color, same dash pattern, same weight, same texture).", + "Re-render as a flat architectural floor plan (think: clean blueprint or museum gallery plan, drawn on light-grey paper). The output should look LIKE A DRAFTED PLAN — flat coloured rooms, fine pen borders, technical clarity — not a photorealistic aerial photo. Stylisation makes precise geometry easier to preserve.\n\nGEOMETRY MUST BE PRESERVED EXACTLY:\n • Each coloured region stays at the EXACT same shape, position, and adjacency. Replace each region's flat colour with ONE muted architectural fill (parquet wood, polished concrete tone, carpet hatch, terrazzo speckle, marble, wood-deck) — choose ONE fill per region; do NOT mix fills inside a region; do NOT bleed a fill across a boundary.\n • Inside each room, draft 4–8 small simplified-symbol furniture pieces appropriate to a floor plan (rectangle = table or sofa, oval = rug, small circle = chair, square = stool, U-shape = sink, narrow rectangle = bed, slim rectangle = bookcase, dashed-rectangle = doorway). Vary the furniture mix across rooms so each room reads as a different room type. NEVER let a piece of furniture bleed across a boundary.\n • Boundaries between regions are thin uniform-weight black ink lines (the \"wall\" symbol of a floor plan), traced EXACTLY along the source partition curves.\n • Every black curve in the source becomes a SINGLE UNIFORM-STYLE walking-path overlay in the output. ALL curves — the query curve AND every distractor curve — must be rendered IDENTICALLY: same dark-grey colour, same uniform dotted pattern, same dot spacing, same dot size. The viewer must NOT be able to tell which curve is the query just by looking at the lines themselves; the only distinguishing feature allowed is the literal text 'Start' label sitting on the query curve. Trace each source curve EXACTLY (do NOT smooth, shorten, reroute, or break any curve).\n • Each single-letter region label stays at the EXACT same position with the SAME glyph, set in a small drafted-style label box.\n\nNo new regions, no new paths, no new labels. No photorealism, no 3D perspective. Pure flat floor-plan look. Do not colour-code or style-code the paths differently from each other.\n\nThe source image contains a literal 'Start' text label inside a small white box with a thin black border, placed at the beginning of the query curve. PRESERVE THIS 'Start' LABEL EXACTLY AS-IS in the output: same wording (\"Start\"), same position, same small white box with thin black border, same legibility. Do NOT replace the label with any icon, do NOT translate the word, do NOT restyle the box. The 'Start' label is the ONLY distinguishing mark between the query curve and the distractor curves; every curve must otherwise look identical (same color, same dash pattern, same weight, same texture)." + ] + }, + "sequential_traversal/line_intersections": { + "src": "images/line_intersections_00000.png", + "prompts": [ + "Re-render as a photorealistic top-down photograph of curving strands on damp wet sand at low tide. ALL strands are SAME uniform colour and material (e.g. dark green-brown braided seaweed, OR pale bleached driftwood twine, OR crimson kelp fronds — pick ONE per image, same across every strand). Each strand follows the source path EXACTLY; at every crossing, render an unambiguous over/under (one strand clearly on top of the other), never fuse or branch. NOISY surface: shell fragments, scattered pebbles, ribbons of foam, faint footprints, tide-line debris, ripple marks; the strands sit visibly on top and are NEVER hidden by clutter. Soft directional morning light, real shadows, slight specular wet-sand sheen. Each endpoint label (A, B, C, …) stays at the EXACT same position with the SAME letter glyph, fully readable. No text other than these endpoint labels. No overlays.", + "Re-render as a photorealistic top-down photograph of curving cables on a worn concrete workshop floor. ALL strands are SAME uniform colour and material (e.g. bright orange paracord, OR matte black rubber cable, OR coiled copper wire — pick ONE per image, same across every strand). Each strand follows the source path EXACTLY; at every crossing, render an unambiguous over/under, never fuse or branch. NOISY surface: scattered sawdust, metal shavings, rust patches, oil stains, dropped screws, gaffer tape scraps; the cables sit visibly on top, never hidden. Hard directional shop-lamp light, sharp shadows, slight grease sheen on the cables. Each endpoint label (A, B, C, …) stays at the EXACT same position with the SAME letter glyph, fully readable. No text other than these endpoint labels. No overlays.", + "Re-render as a photorealistic top-down photograph of curving slender objects on a damp dappled forest floor. ALL strands are SAME uniform colour and material (e.g. weathered grey twine, OR thin twisted vine stalks, OR pale tan jute rope — pick ONE per image, same across every strand). Each strand follows the source path EXACTLY; at every crossing, render an unambiguous over/under, never fuse or branch. NOISY surface: fallen leaves of mixed colour, pine needles, twigs, moss patches, damp soil with bark fragments; the strands sit visibly on top, never hidden. Dappled sunlight through canopy gives uneven highlights and soft shadows. Each endpoint label (A, B, C, …) stays at the EXACT same position with the SAME letter glyph, fully readable. No text other than these endpoint labels. No overlays.", + "Re-render as a photorealistic top-down photograph of curving garden hoses lying on a sunlit patio. ALL strands are SAME uniform colour and material (e.g. dark-green rubber hose, OR yellow-coiled garden hose — pick ONE per image, same across every strand). Each strand follows the source path EXACTLY; at every crossing, render an unambiguous over/under crossing (one strand clearly on top of the other), never fuse or branch. NOISY surface: scattered leaves, pebbles, watering cans on the edge, soil flecks, water drops, faint footprints; the hoses sit visibly on top, never hidden. Soft afternoon sunlight, real shadows, slight rubber sheen. Each endpoint label (A, B, C, …) stays at the EXACT same position with the SAME letter glyph, fully readable. No text other than these endpoint labels. No overlays.", + "Re-render as a photorealistic top-down photograph of curving ivy vines lying on a damp moss-covered stone path. ALL strands are SAME uniform colour and material (e.g. fresh dark-green ivy vines, OR weathered grey-brown bramble shoots — pick ONE per image, same across every strand). Each strand follows the source path EXACTLY; at every crossing, render an unambiguous over/under (one vine clearly resting on top of the other), never fuse or branch. NOISY surface: damp stone slabs, scattered fallen leaves, moss patches, twigs, small puddles, lichen flecks; the vines sit visibly on top, never hidden. Soft directional daylight, real shadows, slight wet sheen on the stone. Each endpoint label (A, B, C, …) stays at the EXACT same position with the SAME letter glyph, fully readable. No text other than these endpoint labels. No overlays." + ] + }, + "sequential_traversal/maze": { + "src": "images/maze_00000.png", + "prompts": [ + "Re-render as a photorealistic top-down aerial photograph of a real green hedge maze in a country garden. Every wall in the source becomes a thick well-trimmed hedge tracing the EXACT same path; every walkable corridor becomes mown grass.\n\nCRITICAL — preserve every border opening:\n • The maze has EXACTLY {n} labeled openings on its border, with letters: {labels}. Each opening is a clear gap in the perimeter hedge that allows entry/exit at the EXACT same border position as in the source. The output MUST contain ALL {n} openings in their EXACT source positions, each clearly marked by its corresponding letter ({labels}). Do NOT miss, merge, relocate, add, or rename any opening.\n • Each letter is rendered next to its opening as a small wooden signpost so the letter is unambiguously readable.\n\nWalls (hedges) follow source paths exactly — never add, remove, thicken beyond uniform width, or break a wall. No new openings anywhere except those {n} listed. No people, no animals, no fountains, no decorative plantings inside the corridors. No text other than the {n} letters {labels}.", + "Re-render as a photorealistic top-down aerial photograph of a stone-walled medieval castle dungeon, lit by torches mounted along the walls. Every wall in the source becomes a thick weathered grey-stone block wall tracing the EXACT same path; every walkable corridor becomes a flagstone floor with subtle wear.\n\nCRITICAL — preserve every border opening:\n • The dungeon has EXACTLY {n} labeled archway openings on its outer wall, with letters: {labels}. Each archway is a clear breach in the perimeter stone wall at the EXACT same border position as in the source. The output MUST contain ALL {n} archways in their EXACT source positions, each clearly marked by its corresponding letter ({labels}). Do NOT miss, merge, relocate, add, or rename any archway.\n • Each letter is rendered above its archway as a small carved-stone keystone plaque so the letter is unambiguously readable.\n\nWalls (stone blocks) follow source paths exactly — never add, remove, thicken, or break a wall. No new archways anywhere except those {n} listed. No furniture, treasure chests, or props inside the corridors. No text other than the {n} letters {labels}.\n\n⚠️ WALL-PRESENCE PRECISION (HIGHEST PRIORITY): EVERY single wall segment in the source MUST be present in the output — including short stubs, dead-ends, T-junctions, and isolated single-cell walls. Do NOT skip, omit, shorten, merge, fuse, or thicken-away any wall. Every wall pixel-path in the source must have a corresponding wall in the rendered scene. A wall that exists in the source but is missing or smoothed-over in the output is a critical failure. Equally: do NOT add any new wall not present in the source. The maze WALL TOPOLOGY must be identical to the source within ~3 px on every wall path.", + "Re-render as a photorealistic top-down aerial photograph of a real autumn cornfield maze cut into a working farm field. Every wall in the source becomes a tall standing block of corn stalks tracing the EXACT same path; every walkable corridor becomes a flat cleared dirt path through the field.\n\nCRITICAL — preserve every border opening:\n • The cornfield maze has EXACTLY {n} labeled border openings, with letters: {labels}. Each opening is a clear gap in the perimeter corn block that allows entry/exit at the EXACT same border position as in the source. The output MUST contain ALL {n} openings in their EXACT source positions, each clearly marked by its corresponding letter ({labels}). Do NOT miss, merge, relocate, add, or rename any opening.\n • Each letter is rendered next to its opening as a painted wooden farm sign so the letter is unambiguously readable.\n\nWalls (corn stalks) follow source paths exactly — never add, remove, thicken beyond uniform width, or break a wall. No new openings anywhere except those {n} listed. No farm equipment, scarecrows, or visitors inside the corridors. No text other than the {n} letters {labels}.\n\n⚠️ WALL-PRESENCE PRECISION (HIGHEST PRIORITY): EVERY single wall segment in the source MUST be present in the output — including short stubs, dead-ends, T-junctions, and isolated single-cell walls. Do NOT skip, omit, shorten, merge, fuse, or thicken-away any wall. Every wall pixel-path in the source must have a corresponding wall in the rendered scene. A wall that exists in the source but is missing or smoothed-over in the output is a critical failure. Equally: do NOT add any new wall not present in the source. The maze WALL TOPOLOGY must be identical to the source within ~3 px on every wall path.", + "Re-render as a photorealistic top-down aerial photograph of a vast library seen from directly above. Every wall in the source becomes a tall double-sided wooden bookshelf packed with books, tracing the EXACT same path; every walkable corridor becomes a polished hardwood floor.\n\nCRITICAL — preserve every border opening:\n • The library has EXACTLY {n} labeled border openings, with letters: {labels}. Each opening is a clear gap in the perimeter shelving at the EXACT same border position as in the source. The output MUST contain ALL {n} openings in their EXACT source positions, each clearly marked by its corresponding letter ({labels}). Do NOT miss, merge, relocate, add, or rename any opening.\n • Each letter is rendered above its opening as a small brass directory plate so the letter is unambiguously readable.\n\nWalls (bookshelves) follow source paths exactly — never add, remove, thicken beyond uniform width, or break a wall. No new openings anywhere except those {n} listed. No people, no chairs, no reading lamps, no carts inside the corridors. No text other than the {n} letters {labels}.\n\n⚠️ WALL-PRESENCE PRECISION (HIGHEST PRIORITY): EVERY single wall segment in the source MUST be present in the output — including short stubs, dead-ends, T-junctions, and isolated single-cell walls. Do NOT skip, omit, shorten, merge, fuse, or thicken-away any wall. Every wall pixel-path in the source must have a corresponding wall in the rendered scene. A wall that exists in the source but is missing or smoothed-over in the output is a critical failure. Equally: do NOT add any new wall not present in the source. The maze WALL TOPOLOGY must be identical to the source within ~3 px on every wall path.", + "Re-render as a photorealistic top-down aerial photograph of a snow-path maze cut into deep arctic snowfields. Every wall in the source becomes a tall packed snowbank tracing the EXACT same path; every walkable corridor becomes pressed flat snow with subtle shoe scuffs.\n\nCRITICAL — preserve every border opening:\n • The maze has EXACTLY {n} labeled border openings, with letters: {labels}. Each opening is a clear gap in the perimeter snowbank at the EXACT same border position as in the source. The output MUST contain ALL {n} openings in their EXACT source positions, each clearly marked by its corresponding letter ({labels}). Do NOT miss, merge, relocate, add, or rename any opening.\n • Each letter is rendered next to its opening as a small wooden trail-marker post so the letter is unambiguously readable.\n\nWalls (snowbanks) follow source paths exactly — never add, remove, thicken beyond uniform width, or break a wall. No new openings anywhere except those {n} listed. No people, no sleds, no animals, no equipment inside the corridors. No text other than the {n} letters {labels}." + ] + }, + "sequential_traversal/traverse_ordering": { + "src": "images/traverse_ordering_00000.png", + "prompts": [ + "Re-render as a photorealistic top-down photograph of curving strands on damp wet sand at low tide. ALL strands are SAME uniform colour and material (e.g. dark green-brown braided seaweed, OR pale bleached driftwood twine, OR crimson kelp fronds — pick ONE per image, same across every strand). Each strand follows the source path EXACTLY; at every crossing, render an unambiguous over/under (one strand clearly on top of the other), never fuse or branch. NOISY surface: shell fragments, scattered pebbles, ribbons of foam, faint footprints, tide-line debris, ripple marks; the strands sit visibly on top and are NEVER hidden by clutter. Soft directional morning light, real shadows, slight specular wet-sand sheen. Every labelled crossing point and the green 'S' start marker stay at the EXACT same position with the SAME letter glyph and the SAME green colour for 'S'. No text other than these existing labels and the 'S'. No overlays.", + "Re-render as a photorealistic top-down photograph of curving cables on a worn concrete workshop floor. ALL strands are SAME uniform colour and material (e.g. bright orange paracord, OR matte black rubber cable, OR coiled copper wire — pick ONE per image, same across every strand). Each strand follows the source path EXACTLY; at every crossing, render an unambiguous over/under, never fuse or branch. NOISY surface: scattered sawdust, metal shavings, rust patches, oil stains, dropped screws, gaffer tape scraps; the cables sit visibly on top, never hidden. Hard directional shop-lamp light, sharp shadows, slight grease sheen on the cables. Every labelled crossing point and the green 'S' start marker stay at the EXACT same position with the SAME letter glyph and the SAME green colour for 'S'. No text other than these existing labels and the 'S'. No overlays.", + "Re-render as a photorealistic top-down photograph of curving slender objects on a damp dappled forest floor. ALL strands are SAME uniform colour and material (e.g. weathered grey twine, OR thin twisted vine stalks, OR pale tan jute rope — pick ONE per image, same across every strand). Each strand follows the source path EXACTLY; at every crossing, render an unambiguous over/under, never fuse or branch. NOISY surface: fallen leaves of mixed colour, pine needles, twigs, moss patches, damp soil with bark fragments; the strands sit visibly on top, never hidden. Dappled sunlight through canopy gives uneven highlights and soft shadows. Every labelled crossing point and the green 'S' start marker stay at the EXACT same position with the SAME letter glyph and the SAME green colour for 'S'. No text other than these existing labels and the 'S'. No overlays.", + "Re-render as a photorealistic top-down photograph of curving garden hoses lying on a sunlit patio. ALL strands are SAME uniform colour and material (e.g. dark-green rubber hose, OR yellow-coiled garden hose — pick ONE per image, same across every strand). Each strand follows the source path EXACTLY; at every crossing, render an unambiguous over/under, never fuse or branch. NOISY surface: scattered leaves, pebbles, watering cans on the edge, soil flecks, water drops, faint footprints; the hoses sit visibly on top, never hidden. Soft afternoon sunlight, real shadows, slight rubber sheen. Every labelled crossing point and the green 'S' start marker stay at the EXACT same position with the SAME letter glyph and the SAME green colour for 'S'. No text other than these existing labels and the 'S'. No overlays.", + "Re-render as a photorealistic top-down photograph of curving ivy vines lying on a damp moss-covered stone path. ALL strands are SAME uniform colour and material (e.g. fresh dark-green ivy vines, OR weathered grey-brown bramble shoots — pick ONE per image, same across every strand). Each strand follows the source path EXACTLY; at every crossing, render an unambiguous over/under (one vine clearly resting on top of the other), never fuse or branch. NOISY surface: damp stone slabs, scattered fallen leaves, moss patches, twigs, small puddles, lichen flecks; the vines sit visibly on top, never hidden. Soft directional daylight, real shadows, slight wet sheen on the stone. Every labelled crossing point and the green 'S' start marker stay at the EXACT same position with the SAME letter glyph and the SAME green colour for 'S'. No text other than these existing labels and the 'S'. No overlays." + ] + }, + "visual_attribute_transfer/constellation_match_count": { + "src": "images/constellation_match_count_00000.png", + "prompts": [ + "Re-render as a deep-space astrophotograph. Preserve EXACTLY: the LEFT 'TEMPLATE' bordered box (with its label and frame), the vertical divider, and the position of every white dot on both sides (within ~2 px).\n\nLEFT (template): each dot → one identical small white star, all matching, on a uniform dark sky.\n\nRIGHT (field): each dot → ONE star, but EVERY star is rendered DIFFERENTLY — vary colour (white, blue, yellow, orange, red), brightness, diffraction-spike pattern, halo, hue, faint nebula tint. Sizes vary slightly but stay small (similar order to template stars). Positions must be pixel-exact — no shifting, no adding, no removing.\n\nNo text overlays beyond the existing TEMPLATE label.", + "Re-render as a top-down photograph on black velvet. Preserve EXACTLY: the LEFT 'TEMPLATE' bordered box (with its label and frame), the vertical divider, and the position of every dot on both sides (within ~2 px).\n\nLEFT (template): each dot → one identical small white pearl, all matching.\n\nRIGHT (field): each dot → ONE small gemstone, but EVERY stone is rendered DIFFERENTLY — vary colour (ruby, sapphire, emerald, citrine, amethyst, opal, pearl, garnet), cut (round, oval, faceted, cabochon), material (gem, polished bead, glass), and surface sheen. Keep all stones small (no larger than ~1.5× a template pearl). Positions must be pixel-exact — no shifting, no adding, no removing.\n\nNo text overlays beyond the existing TEMPLATE label.", + "Re-render as a top-down photograph at night. Preserve EXACTLY: the LEFT 'TEMPLATE' bordered box (with its label and frame), the vertical divider, and the position of every dot on both sides (within ~2 px).\n\nLEFT (template): each dot → one identical small warm tealight flame, all matching, on dark wood.\n\nRIGHT (field): each dot → ONE point of light, but EVERY light is rendered DIFFERENTLY — vary type (candle flame, lantern bulb, fairy LED, bokeh glint, fibre-optic tip, glowstick dot), colour (warm white, gold, blue, green, magenta, red), bloom shape, and halo size. Keep each light small (no larger than ~1.5× a template flame). Positions must be pixel-exact — no shifting, no adding, no removing.\n\nNo text overlays beyond the existing TEMPLATE label.", + "Re-render as a top-down photograph on a dark slate plate. Preserve EXACTLY: the LEFT 'TEMPLATE' bordered box (with its label and frame), the vertical divider, and the position of every dot on both sides (within ~2 px).\n\nLEFT (template): each dot → one identical small white peppermint candy, all matching.\n\nRIGHT (field): each dot → ONE small confection, but EVERY one is rendered DIFFERENTLY — vary type (gumdrop, jelly bean, hard candy, chocolate ball, marshmallow, pill, dragee, m&m), colour, gloss, texture (smooth, sugar-dusted, crystallised, matte). Keep each piece small (no larger than ~1.5× a template mint). Positions must be pixel-exact — no shifting, no adding, no removing.\n\nNo text overlays beyond the existing TEMPLATE label.", + "Re-render as a top-down photograph on wet dark sand. Preserve EXACTLY: the LEFT 'TEMPLATE' bordered box (with its label and frame), the vertical divider, and the position of every dot on both sides (within ~2 px).\n\nLEFT (template): each dot → one identical small white shell, all matching.\n\nRIGHT (field): each dot → ONE small sea object, but EVERY one is rendered DIFFERENTLY — vary type (cowrie, scallop, cone shell, sea-glass chip, polished pebble, coral fragment, starfish point, mother-of-pearl chip), colour, texture (glossy, ridged, matte, translucent), and material. Keep each object small (no larger than ~1.5× a template shell). Positions must be pixel-exact — no shifting, no adding, no removing.\n\nNo text overlays beyond the existing TEMPLATE label." + ] + }, + "visual_attribute_transfer/contour_silhouette_count": { + "src": "images/contour_silhouette_count_00000.png", + "prompts": [ + "GEOMETRY IS A PRECISION TEMPLATE — DO NOT CHANGE IT. Re-render as a photorealistic top-down photograph. The source defines exact pixel positions and sizes; copy them.\n\n⚠️ PIXEL-EXACT (within ~2 px, NON-NEGOTIABLE):\n • LEFT TEMPLATE SIZE & POSITION: the single contour keeps its EXACT bounding box (width × height) and EXACT (x, y) from the source. If the source template is small, render it small. Do NOT enlarge, do NOT recentre, do NOT zoom in, do NOT make it fill the LEFT panel.\n • RIGHT FIELD SIZE & POSITION: every source contour → ONE object at the same bounding box and same (x, y). Same count as source — never add, skip, merge, or split.\n • RELATIVE SIZES: template-vs-field and field-vs-field ratios match the source pixel-for-pixel.\n • DIVIDER: vertical separator at the source x-coordinate; LEFT/RIGHT panel widths match the source.\n\nSTRUCTURE: LEFT → ONE class-A object on LEFT surface. RIGHT → many class-B objects on RIGHT surface.\n\nPick TWO different items — one for LEFT (class A), a different one for RIGHT (class B):\n pebbles, river stones, autumn leaves, pottery shards, mushrooms, bread rolls, cookies, pancakes, pastries, dumplings, silk pouches, beach shells, driftwood pieces, clay tiles, wax seals, sponge pieces.\n\nLEFT surface: worn wooden cutting board. RIGHT surface: moss-covered stone slab. Each panel has its own directional lighting. Per-object micro-variation in shade/sheen. Background must not create new object-like blobs.\n\nNo text, no labels, no overlays.", + "⚠️ CRITICAL CONSTRAINTS — HIGHEST PRIORITY (do not violate):\n 1. SCALE IS LOCKED. Every rendered object must occupy the EXACT same pixel area as its source contour/stroke. Do NOT enlarge, shrink, or normalise. The single template object on the LEFT must match the source template's pixel size exactly. Each field object on the RIGHT must match the pixel size of its corresponding source contour/stroke exactly. Relative sizes between all objects (left vs right and right-vs-right) must be preserved within ~5 px. The model must NOT make the LEFT template object larger or smaller than its source, and must NOT rescale the RIGHT field objects to fill space.\n 2. DIVIDER POSITION IS LOCKED. The vertical separator line between LEFT and RIGHT panels must be at the EXACT same x-coordinate as in the source image. Do NOT shift, widen, narrow, or recentre the divider. The LEFT panel pixel width and RIGHT panel pixel width must each match the source exactly. The single template object on the LEFT must remain at the SAME (x, y) position as the source template stroke/contour — do NOT recentre it within the LEFT panel.\n\nRe-render the two panels as a photorealistic top-down photograph. The two panels are separated by the SAME vertical divider, share the SAME overall lighting bias, but use TWO DIFFERENT object types and TWO DIFFERENT surfaces.\n\nPANEL STRUCTURE — read carefully:\n • The LEFT panel is the TEMPLATE: it contains EXACTLY ONE closed contour in the source. Render this single template contour as ONE object of class A on the LEFT surface, preserving the EXACT silhouette outline, size, and position of that single contour.\n • The RIGHT panel is the FIELD: it contains MANY closed contours in the source. Render EVERY contour in the right-panel source as ONE object of class B on the RIGHT surface, preserving the EXACT silhouette outline, size, and position of each contour. The number of objects in the RIGHT panel must equal the source's count exactly.\n\nPick TWO different items at random from this pool — one for the LEFT (class A, the single template) and a different one for the RIGHT (class B, the many field objects):\n vintage coins, brass buttons, sea glass, polished gemstones, wax seals, glass beads, metal tokens, bottle caps, mosaic chips, glass marbles, mother-of-pearl discs, enamel buttons.\n\nThe LEFT surface is a velvet jeweller's tray; the RIGHT surface is a slate counter top. Each panel has its own directional lighting (slightly different angle / colour temperature).\n\nACCURACY IS CRITICAL:\n • LEFT panel: exactly ONE class-A object whose silhouette matches the single template contour outline within ~5 px. No drift, no rescale, no rotation.\n • RIGHT panel: EVERY source contour becomes exactly ONE class-B object whose silhouette matches its source contour within ~5 px. Same OBJECT COUNT per panel as the source. Never add, skip, merge, or split contours.\n • Object TYPE differs left vs right, but each rendered object's silhouette is identical to its source contour. A silhouette overlay should align almost perfectly within each panel.\n • Background fills empty space only — must not create new object-like blobs.\n\nANTI-PROGRAM-COMPARISON HARDENING:\n • Each panel has a NOISY, lived-in surface — micro-scratches, dust, subtle wear, slight colour drift; same genre of noise across both panels but DIFFERENT realisations on each side.\n • Per-object micro-variation (texture / shade / sheen) so corresponding objects across panels never share a pixel-level signature.\n\nNo text, no labels, no overlays.", + "⚠️ CRITICAL CONSTRAINTS — HIGHEST PRIORITY (do not violate):\n 1. SCALE IS LOCKED. Every rendered object must occupy the EXACT same pixel area as its source contour/stroke. Do NOT enlarge, shrink, or normalise. The single template object on the LEFT must match the source template's pixel size exactly. Each field object on the RIGHT must match the pixel size of its corresponding source contour/stroke exactly. Relative sizes between all objects (left vs right and right-vs-right) must be preserved within ~5 px. The model must NOT make the LEFT template object larger or smaller than its source, and must NOT rescale the RIGHT field objects to fill space.\n 2. DIVIDER POSITION IS LOCKED. The vertical separator line between LEFT and RIGHT panels must be at the EXACT same x-coordinate as in the source image. Do NOT shift, widen, narrow, or recentre the divider. The LEFT panel pixel width and RIGHT panel pixel width must each match the source exactly. The single template object on the LEFT must remain at the SAME (x, y) position as the source template stroke/contour — do NOT recentre it within the LEFT panel.\n\nRe-render the two panels as a photorealistic top-down photograph. The two panels are separated by the SAME vertical divider, share the SAME overall lighting bias, but use TWO DIFFERENT object types and TWO DIFFERENT surfaces.\n\nPANEL STRUCTURE — read carefully:\n • The LEFT panel is the TEMPLATE: it contains EXACTLY ONE closed contour in the source. Render this single template contour as ONE object of class A on the LEFT surface, preserving the EXACT silhouette outline, size, and position of that single contour.\n • The RIGHT panel is the FIELD: it contains MANY closed contours in the source. Render EVERY contour in the right-panel source as ONE object of class B on the RIGHT surface, preserving the EXACT silhouette outline, size, and position of each contour. The number of objects in the RIGHT panel must equal the source's count exactly.\n\nPick TWO different items at random from this pool — one for the LEFT (class A, the single template) and a different one for the RIGHT (class B, the many field objects):\n grapes, cherries, olives, dried apricots, macarons, chocolate truffles, hard candies, dried figs, kumquats, lychees, almonds, blueberries, dried cranberries, raisins, gummies.\n\nThe LEFT surface is a marble pastry slab; the RIGHT surface is a raw linen tablecloth. Each panel has its own directional lighting (slightly different angle / colour temperature).\n\nACCURACY IS CRITICAL:\n • LEFT panel: exactly ONE class-A object whose silhouette matches the single template contour outline within ~5 px. No drift, no rescale, no rotation.\n • RIGHT panel: EVERY source contour becomes exactly ONE class-B object whose silhouette matches its source contour within ~5 px. Same OBJECT COUNT per panel as the source. Never add, skip, merge, or split contours.\n • Object TYPE differs left vs right, but each rendered object's silhouette is identical to its source contour. A silhouette overlay should align almost perfectly within each panel.\n • Background fills empty space only — must not create new object-like blobs.\n\nANTI-PROGRAM-COMPARISON HARDENING:\n • Each panel has a NOISY, lived-in surface — micro-scratches, dust, subtle wear, slight colour drift; same genre of noise across both panels but DIFFERENT realisations on each side.\n • Per-object micro-variation (texture / shade / sheen) so corresponding objects across panels never share a pixel-level signature.\n\nNo text, no labels, no overlays.", + "⚠️ CRITICAL CONSTRAINTS — HIGHEST PRIORITY (do not violate):\n 1. SCALE IS LOCKED. Every rendered object must occupy the EXACT same pixel area as its source contour/stroke. Do NOT enlarge, shrink, or normalise. The single template object on the LEFT must match the source template's pixel size exactly. Each field object on the RIGHT must match the pixel size of its corresponding source contour/stroke exactly. Relative sizes between all objects (left vs right and right-vs-right) must be preserved within ~5 px. The model must NOT make the LEFT template object larger or smaller than its source, and must NOT rescale the RIGHT field objects to fill space.\n 2. DIVIDER POSITION IS LOCKED. The vertical separator line between LEFT and RIGHT panels must be at the EXACT same x-coordinate as in the source image. Do NOT shift, widen, narrow, or recentre the divider. The LEFT panel pixel width and RIGHT panel pixel width must each match the source exactly. The single template object on the LEFT must remain at the SAME (x, y) position as the source template stroke/contour — do NOT recentre it within the LEFT panel.\n\nRe-render the two panels as a photorealistic top-down photograph. The two panels are separated by the SAME vertical divider, share the SAME overall lighting bias, but use TWO DIFFERENT object types and TWO DIFFERENT surfaces.\n\nPANEL STRUCTURE — read carefully:\n • The LEFT panel is the TEMPLATE: it contains EXACTLY ONE closed contour in the source. Render this single template contour as ONE object of class A on the LEFT surface, preserving the EXACT silhouette outline, size, and position of that single contour.\n • The RIGHT panel is the FIELD: it contains MANY closed contours in the source. Render EVERY contour in the right-panel source as ONE object of class B on the RIGHT surface, preserving the EXACT silhouette outline, size, and position of each contour. The number of objects in the RIGHT panel must equal the source's count exactly.\n\nPick TWO different items at random from this pool — one for the LEFT (class A, the single template) and a different one for the RIGHT (class B, the many field objects):\n rubber washers, metal nuts, plastic discs, silicone gaskets, marble cabochons, wooden game pieces, cardboard punch-outs, ceramic tiles, linoleum scraps, cork rounds, leather punches, felted-wool dots.\n\nThe LEFT surface is a galvanised steel workbench; the RIGHT surface is a weathered concrete floor. Each panel has its own directional lighting (slightly different angle / colour temperature).\n\nACCURACY IS CRITICAL:\n • LEFT panel: exactly ONE class-A object whose silhouette matches the single template contour outline within ~5 px. No drift, no rescale, no rotation.\n • RIGHT panel: EVERY source contour becomes exactly ONE class-B object whose silhouette matches its source contour within ~5 px. Same OBJECT COUNT per panel as the source. Never add, skip, merge, or split contours.\n • Object TYPE differs left vs right, but each rendered object's silhouette is identical to its source contour. A silhouette overlay should align almost perfectly within each panel.\n • Background fills empty space only — must not create new object-like blobs.\n\nANTI-PROGRAM-COMPARISON HARDENING:\n • Each panel has a NOISY, lived-in surface — micro-scratches, dust, subtle wear, slight colour drift; same genre of noise across both panels but DIFFERENT realisations on each side.\n • Per-object micro-variation (texture / shade / sheen) so corresponding objects across panels never share a pixel-level signature.\n\nNo text, no labels, no overlays.", + "⚠️ CRITICAL CONSTRAINTS — HIGHEST PRIORITY (do not violate):\n 1. SCALE IS LOCKED. Every rendered object must occupy the EXACT same pixel area as its source contour/stroke. Do NOT enlarge, shrink, or normalise. The single template object on the LEFT must match the source template's pixel size exactly. Each field object on the RIGHT must match the pixel size of its corresponding source contour/stroke exactly. Relative sizes between all objects (left vs right and right-vs-right) must be preserved within ~5 px. The model must NOT make the LEFT template object larger or smaller than its source, and must NOT rescale the RIGHT field objects to fill space.\n 2. DIVIDER POSITION IS LOCKED. The vertical separator line between LEFT and RIGHT panels must be at the EXACT same x-coordinate as in the source image. Do NOT shift, widen, narrow, or recentre the divider. The LEFT panel pixel width and RIGHT panel pixel width must each match the source exactly. The single template object on the LEFT must remain at the SAME (x, y) position as the source template stroke/contour — do NOT recentre it within the LEFT panel.\n\nRe-render the two panels as a photorealistic top-down photograph. The two panels are separated by the SAME vertical divider, share the SAME overall lighting bias, but use TWO DIFFERENT object types and TWO DIFFERENT surfaces.\n\nPANEL STRUCTURE — read carefully:\n • The LEFT panel is the TEMPLATE: it contains EXACTLY ONE closed contour in the source. Render this single template contour as ONE object of class A on the LEFT surface, preserving the EXACT silhouette outline, size, and position of that single contour.\n • The RIGHT panel is the FIELD: it contains MANY closed contours in the source. Render EVERY contour in the right-panel source as ONE object of class B on the RIGHT surface, preserving the EXACT silhouette outline, size, and position of each contour. The number of objects in the RIGHT panel must equal the source's count exactly.\n\nPick TWO different items at random from this pool — one for the LEFT (class A, the single template) and a different one for the RIGHT (class B, the many field objects):\n silk pouches, fabric patches, felt cutouts, paper origami pieces, leather scraps, lace pieces, suede swatches, linen squares, vellum cutouts, cork punches, foam-board chips.\n\nThe LEFT surface is a craft-paper sheet with creases; the RIGHT surface is a denim-blue cotton sheet. Each panel has its own directional lighting (slightly different angle / colour temperature).\n\nACCURACY IS CRITICAL:\n • LEFT panel: exactly ONE class-A object whose silhouette matches the single template contour outline within ~5 px. No drift, no rescale, no rotation.\n • RIGHT panel: EVERY source contour becomes exactly ONE class-B object whose silhouette matches its source contour within ~5 px. Same OBJECT COUNT per panel as the source. Never add, skip, merge, or split contours.\n • Object TYPE differs left vs right, but each rendered object's silhouette is identical to its source contour. A silhouette overlay should align almost perfectly within each panel.\n • Background fills empty space only — must not create new object-like blobs.\n\nANTI-PROGRAM-COMPARISON HARDENING:\n • Each panel has a NOISY, lived-in surface — micro-scratches, dust, subtle wear, slight colour drift; same genre of noise across both panels but DIFFERENT realisations on each side.\n • Per-object micro-variation (texture / shade / sheen) so corresponding objects across panels never share a pixel-level signature.\n\nNo text, no labels, no overlays." + ] + }, + "visual_attribute_transfer/spot_the_contour_diff": { + "src": "images/contour_diff_00000.png", + "prompts": [ + "Re-render the two panels as a photorealistic top-down photograph using TWO DIFFERENT object types for left vs right — same vertical divider, same overall lighting bias. Pick TWO different items at random from this pool, one for the LEFT panel and a different one for the RIGHT panel:\n pebbles, river stones, autumn leaves, pottery shards, mushrooms, bread rolls, cookies, pancakes, pastries, dumplings, silk pouches, leather scraps, fabric patches, beach shells, driftwood pieces, clay tiles, wax seals, sponge pieces.\n\nThe LEFT panel sits on a worn wooden cutting board; the RIGHT panel sits on a moss-covered stone slab. The two surfaces share the same lighting but their materials and colours are completely different.\n\nACCURACY IS CRITICAL:\n • EVERY closed contour in the source becomes EXACTLY ONE object on the matching side, at the SAME silhouette outline, SAME size, SAME (x, y) — within ~5 px. No drift, no rescale, no rotation away from the source.\n • Same OBJECT COUNT per panel as the source. Never add, skip, merge, or split contours.\n • Object TYPE differs left vs right, but each silhouette is identical to the source contour on its side. A silhouette overlay should align almost perfectly.\n • Background fills empty space only — must not create new object-like blobs.\n\nANTI-PROGRAM-COMPARISON HARDENING:\n • Each panel has a NOISY, lived-in surface — micro-scratches, dust, subtle wear, slight colour drift; same genre of noise across both panels but DIFFERENT realisations on each side.\n • Each panel has its own directional lighting (slightly different angle and colour temperature), so the same silhouette renders with different shadow direction and brightness on the two sides.\n • Per-object micro-variation (texture / shade / sheen) so corresponding objects across panels never share a pixel-level signature.\n\nNo text, no labels, no overlays.", + "Re-render the two panels as a photorealistic top-down photograph using TWO DIFFERENT object types for left vs right — same vertical divider, same overall lighting bias. Pick TWO different items at random from this pool, one for the LEFT panel and a different one for the RIGHT panel:\n vintage coins, brass buttons, sea glass, polished gemstones, wax seals, glass beads, metal tokens, bottle caps, mosaic chips, glass marbles, mother-of-pearl discs, enamel buttons.\n\nThe LEFT panel sits on a velvet jeweller's tray; the RIGHT panel sits on a slate counter top. The two surfaces share the same lighting but their materials and colours are completely different.\n\nACCURACY IS CRITICAL:\n • EVERY closed contour in the source becomes EXACTLY ONE object on the matching side, at the SAME silhouette outline, SAME size, SAME (x, y) — within ~5 px. No drift, no rescale, no rotation away from the source.\n • Same OBJECT COUNT per panel as the source. Never add, skip, merge, or split contours.\n • Object TYPE differs left vs right, but each silhouette is identical to the source contour on its side. A silhouette overlay should align almost perfectly.\n • Background fills empty space only — must not create new object-like blobs.\n\nANTI-PROGRAM-COMPARISON HARDENING:\n • Each panel has a NOISY, lived-in surface — micro-scratches, dust, subtle wear, slight colour drift; same genre of noise across both panels but DIFFERENT realisations on each side.\n • Each panel has its own directional lighting (slightly different angle and colour temperature), so the same silhouette renders with different shadow direction and brightness on the two sides.\n • Per-object micro-variation (texture / shade / sheen) so corresponding objects across panels never share a pixel-level signature.\n\nNo text, no labels, no overlays.", + "Re-render the two panels as a photorealistic top-down photograph using TWO DIFFERENT object types for left vs right — same vertical divider, same overall lighting bias. Pick TWO different items at random from this pool, one for the LEFT panel and a different one for the RIGHT panel:\n grapes, cherries, olives, dried apricots, macarons, chocolate truffles, hard candies, dried figs, kumquats, lychees, almonds, blueberries, dried cranberries, raisins, gummies.\n\nThe LEFT panel sits on a marble pastry slab; the RIGHT panel sits on a raw linen tablecloth. The two surfaces share the same lighting but their materials and colours are completely different.\n\nACCURACY IS CRITICAL:\n • EVERY closed contour in the source becomes EXACTLY ONE object on the matching side, at the SAME silhouette outline, SAME size, SAME (x, y) — within ~5 px. No drift, no rescale, no rotation away from the source.\n • Same OBJECT COUNT per panel as the source. Never add, skip, merge, or split contours.\n • Object TYPE differs left vs right, but each silhouette is identical to the source contour on its side. A silhouette overlay should align almost perfectly.\n • Background fills empty space only — must not create new object-like blobs.\n\nANTI-PROGRAM-COMPARISON HARDENING:\n • Each panel has a NOISY, lived-in surface — micro-scratches, dust, subtle wear, slight colour drift; same genre of noise across both panels but DIFFERENT realisations on each side.\n • Each panel has its own directional lighting (slightly different angle and colour temperature), so the same silhouette renders with different shadow direction and brightness on the two sides.\n • Per-object micro-variation (texture / shade / sheen) so corresponding objects across panels never share a pixel-level signature.\n\nNo text, no labels, no overlays.", + "Re-render the two panels as a photorealistic top-down photograph using TWO DIFFERENT object types for left vs right — same vertical divider, same overall lighting bias. Pick TWO different items at random from this pool, one for the LEFT panel and a different one for the RIGHT panel:\n rubber washers, metal nuts, plastic discs, silicone gaskets, marble cabochons, wooden game pieces, cardboard punch-outs, ceramic tiles, linoleum scraps, cork rounds, leather punches, felted-wool dots.\n\nThe LEFT panel sits on a galvanised steel workbench; the RIGHT panel sits on a weathered concrete floor. The two surfaces share the same lighting but their materials and colours are completely different.\n\nACCURACY IS CRITICAL:\n • EVERY closed contour in the source becomes EXACTLY ONE object on the matching side, at the SAME silhouette outline, SAME size, SAME (x, y) — within ~5 px. No drift, no rescale, no rotation away from the source.\n • Same OBJECT COUNT per panel as the source. Never add, skip, merge, or split contours.\n • Object TYPE differs left vs right, but each silhouette is identical to the source contour on its side. A silhouette overlay should align almost perfectly.\n • Background fills empty space only — must not create new object-like blobs.\n\nANTI-PROGRAM-COMPARISON HARDENING:\n • Each panel has a NOISY, lived-in surface — micro-scratches, dust, subtle wear, slight colour drift; same genre of noise across both panels but DIFFERENT realisations on each side.\n • Each panel has its own directional lighting (slightly different angle and colour temperature), so the same silhouette renders with different shadow direction and brightness on the two sides.\n • Per-object micro-variation (texture / shade / sheen) so corresponding objects across panels never share a pixel-level signature.\n\nNo text, no labels, no overlays.", + "Re-render the two panels as a photorealistic top-down photograph using TWO DIFFERENT object types for left vs right — same vertical divider, same overall lighting bias. Pick TWO different items at random from this pool, one for the LEFT panel and a different one for the RIGHT panel:\n silk pouches, fabric patches, felt cutouts, paper origami pieces, leather scraps, lace pieces, suede swatches, linen squares, vellum cutouts, cork punches, foam-board chips.\n\nThe LEFT panel sits on a craft-paper sheet with creases; the RIGHT panel sits on a denim-blue cotton sheet. The two surfaces share the same lighting but their materials and colours are completely different.\n\nACCURACY IS CRITICAL:\n • EVERY closed contour in the source becomes EXACTLY ONE object on the matching side, at the SAME silhouette outline, SAME size, SAME (x, y) — within ~5 px. No drift, no rescale, no rotation away from the source.\n • Same OBJECT COUNT per panel as the source. Never add, skip, merge, or split contours.\n • Object TYPE differs left vs right, but each silhouette is identical to the source contour on its side. A silhouette overlay should align almost perfectly.\n • Background fills empty space only — must not create new object-like blobs.\n\nANTI-PROGRAM-COMPARISON HARDENING:\n • Each panel has a NOISY, lived-in surface — micro-scratches, dust, subtle wear, slight colour drift; same genre of noise across both panels but DIFFERENT realisations on each side.\n • Each panel has its own directional lighting (slightly different angle and colour temperature), so the same silhouette renders with different shadow direction and brightness on the two sides.\n • Per-object micro-variation (texture / shade / sheen) so corresponding objects across panels never share a pixel-level signature.\n\nNo text, no labels, no overlays." + ] + }, + "visual_attribute_transfer/spot_the_field_diff": { + "src": "images/field_diff_00000.png", + "prompts": [ + "⚠️ PIXEL-EXACT (NON-NEGOTIABLE):\n • The TWO-PANEL CONTENT is sacred. Every cell, every grid line, every cell value, every cell colour, every divider, every panel border MUST be reproduced EXACTLY — do NOT alter, recolour, redraw, blur, smooth, re-rasterise, or shift any cell within the panels.\n • The vertical divider between LEFT and RIGHT panels stays at the EXACT same x-coordinate.\n • Object COUNT inside the panels equals the source EXACTLY — never add, skip, merge, or split anything inside a panel.\n • No signal-level modifications inside the panels: no extra noise, paper grain, blur, JPEG artefacts, or colour shifts on the panel pixels (these have already been baked in upstream). The panels are passed through UNCHANGED.\n • You may ONLY add real-world context AROUND the panels (outside their bounding boxes). Inside the panels: ZERO change.\n\nPlace the unchanged two-panel sheet pinned to a corkboard. Add visible cork texture surrounding the sheet, two coloured push-pins at the top corners, and gentle warm lighting from a desk lamp at upper left. The sheet sits flat against the cork; do NOT crop or rotate the panels. Use a 2:1 wide framing.", + "⚠️ PIXEL-EXACT (NON-NEGOTIABLE):\n • The TWO-PANEL CONTENT is sacred. Every cell, every grid line, every cell value, every cell colour, every divider, every panel border MUST be reproduced EXACTLY — do NOT alter, recolour, redraw, blur, smooth, re-rasterise, or shift any cell within the panels.\n • The vertical divider between LEFT and RIGHT panels stays at the EXACT same x-coordinate.\n • Object COUNT inside the panels equals the source EXACTLY — never add, skip, merge, or split anything inside a panel.\n • No signal-level modifications inside the panels: no extra noise, paper grain, blur, JPEG artefacts, or colour shifts on the panel pixels (these have already been baked in upstream). The panels are passed through UNCHANGED.\n • You may ONLY add real-world context AROUND the panels (outside their bounding boxes). Inside the panels: ZERO change.\n\nPlace the unchanged two-panel sheet lying on a worn wooden desk. Surround the sheet with a few real-world objects positioned OUTSIDE the panels: a closed pen, a coffee-ring stain on the wood, scattered paper-clips, and a corner of an open notebook. Lighting is even daylight. Use a 2:1 wide framing.", + "⚠️ PIXEL-EXACT (NON-NEGOTIABLE):\n • The TWO-PANEL CONTENT is sacred. Every cell, every grid line, every cell value, every cell colour, every divider, every panel border MUST be reproduced EXACTLY — do NOT alter, recolour, redraw, blur, smooth, re-rasterise, or shift any cell within the panels.\n • The vertical divider between LEFT and RIGHT panels stays at the EXACT same x-coordinate.\n • Object COUNT inside the panels equals the source EXACTLY — never add, skip, merge, or split anything inside a panel.\n • No signal-level modifications inside the panels: no extra noise, paper grain, blur, JPEG artefacts, or colour shifts on the panel pixels (these have already been baked in upstream). The panels are passed through UNCHANGED.\n • You may ONLY add real-world context AROUND the panels (outside their bounding boxes). Inside the panels: ZERO change.\n\nPlace the unchanged two-panel sheet held against a clipboard, with the metal clip visible at the top edge OUTSIDE the panels. Show the clipboard's plywood grain and a darker shadow under the right edge. Lighting is fluorescent overhead, mildly cool. Use a 2:1 wide framing.", + "⚠️ PIXEL-EXACT (NON-NEGOTIABLE):\n • The TWO-PANEL CONTENT is sacred. Every cell, every grid line, every cell value, every cell colour, every divider, every panel border MUST be reproduced EXACTLY — do NOT alter, recolour, redraw, blur, smooth, re-rasterise, or shift any cell within the panels.\n • The vertical divider between LEFT and RIGHT panels stays at the EXACT same x-coordinate.\n • Object COUNT inside the panels equals the source EXACTLY — never add, skip, merge, or split anything inside a panel.\n • No signal-level modifications inside the panels: no extra noise, paper grain, blur, JPEG artefacts, or colour shifts on the panel pixels (these have already been baked in upstream). The panels are passed through UNCHANGED.\n • You may ONLY add real-world context AROUND the panels (outside their bounding boxes). Inside the panels: ZERO change.\n\nPlace the unchanged two-panel sheet sitting on a darker textile background — a folded grey wool blanket or felt mat. Add a faint vignette and a small folded paper triangle at the top-right corner OUTSIDE the panels. Lighting is overcast daylight from above. Use a 2:1 wide framing.", + "⚠️ PIXEL-EXACT (NON-NEGOTIABLE):\n • The TWO-PANEL CONTENT is sacred. Every cell, every grid line, every cell value, every cell colour, every divider, every panel border MUST be reproduced EXACTLY — do NOT alter, recolour, redraw, blur, smooth, re-rasterise, or shift any cell within the panels.\n • The vertical divider between LEFT and RIGHT panels stays at the EXACT same x-coordinate.\n • Object COUNT inside the panels equals the source EXACTLY — never add, skip, merge, or split anything inside a panel.\n • No signal-level modifications inside the panels: no extra noise, paper grain, blur, JPEG artefacts, or colour shifts on the panel pixels (these have already been baked in upstream). The panels are passed through UNCHANGED.\n • You may ONLY add real-world context AROUND the panels (outside their bounding boxes). Inside the panels: ZERO change.\n\nPlace the unchanged two-panel sheet in a manila folder, with the folder's tan edge visible at the bottom and a small yellow sticky note in the top-left margin OUTSIDE the panels. The sheet sits inside the folder, slightly raised. Lighting is warm desk lamp from upper right. Use a 2:1 wide framing." + ] + }, + "visual_attribute_transfer/spot_the_signal_diff": { + "src": "images/signal_diff_00000.png", + "prompts": [ + "⚠️ PIXEL-EXACT (NON-NEGOTIABLE):\n • The TWO-PANEL CONTENT is sacred. Every cell, every grid line, every waveform sample, every gridline, every axis tick, every label, every divider, every panel border MUST be reproduced EXACTLY — do NOT alter, recolour, redraw, blur, smooth, re-rasterise, or shift any cell within the panels.\n • The vertical divider between LEFT and RIGHT panels stays at the EXACT same x-coordinate.\n • Object COUNT inside the panels equals the source EXACTLY — never add, skip, merge, or split anything inside a panel.\n • No signal-level modifications inside the panels: no extra noise, paper grain, blur, JPEG artefacts, or colour shifts on the panel pixels (these have already been baked in upstream). The panels are passed through UNCHANGED.\n • You may ONLY add real-world context AROUND the panels (outside their bounding boxes). Inside the panels: ZERO change.\n\nPlace the unchanged two-signal-trace panel as if it were a printed lab strip lying on a metal lab bench. Around the strip add: a partial view of an oscilloscope chassis dial at top-right, a coiled BNC cable at bottom-left, and a printed-circuit board corner at lower-right — ALL outside the panel area. Lighting is bright fluorescent. Use the wide 2:1 ratio of the original strip.", + "⚠️ PIXEL-EXACT (NON-NEGOTIABLE):\n • The TWO-PANEL CONTENT is sacred. Every cell, every grid line, every waveform sample, every gridline, every axis tick, every label, every divider, every panel border MUST be reproduced EXACTLY — do NOT alter, recolour, redraw, blur, smooth, re-rasterise, or shift any cell within the panels.\n • The vertical divider between LEFT and RIGHT panels stays at the EXACT same x-coordinate.\n • Object COUNT inside the panels equals the source EXACTLY — never add, skip, merge, or split anything inside a panel.\n • No signal-level modifications inside the panels: no extra noise, paper grain, blur, JPEG artefacts, or colour shifts on the panel pixels (these have already been baked in upstream). The panels are passed through UNCHANGED.\n • You may ONLY add real-world context AROUND the panels (outside their bounding boxes). Inside the panels: ZERO change.\n\nPlace the unchanged two-signal-trace panel pinned to a lab whiteboard. Show whiteboard texture surrounding the strip, two magnetic clips at top corners, and faint marker scribbles in the visible whiteboard area OUTSIDE the panel. Lighting is even cool office daylight. Use a 2:1 ratio.", + "⚠️ PIXEL-EXACT (NON-NEGOTIABLE):\n • The TWO-PANEL CONTENT is sacred. Every cell, every grid line, every waveform sample, every gridline, every axis tick, every label, every divider, every panel border MUST be reproduced EXACTLY — do NOT alter, recolour, redraw, blur, smooth, re-rasterise, or shift any cell within the panels.\n • The vertical divider between LEFT and RIGHT panels stays at the EXACT same x-coordinate.\n • Object COUNT inside the panels equals the source EXACTLY — never add, skip, merge, or split anything inside a panel.\n • No signal-level modifications inside the panels: no extra noise, paper grain, blur, JPEG artefacts, or colour shifts on the panel pixels (these have already been baked in upstream). The panels are passed through UNCHANGED.\n • You may ONLY add real-world context AROUND the panels (outside their bounding boxes). Inside the panels: ZERO change.\n\nPlace the unchanged two-signal-trace panel as a printout sitting on an electronics workbench. Around the printout (NOT on it) add scattered components: a few resistors, a multi-meter probe tip, and a small breadboard at bottom-right. The printout itself is flat and uncropped. Lighting is warm bench-lamp from above. Use a 2:1 ratio.", + "⚠️ PIXEL-EXACT (NON-NEGOTIABLE):\n • The TWO-PANEL CONTENT is sacred. Every cell, every grid line, every waveform sample, every gridline, every axis tick, every label, every divider, every panel border MUST be reproduced EXACTLY — do NOT alter, recolour, redraw, blur, smooth, re-rasterise, or shift any cell within the panels.\n • The vertical divider between LEFT and RIGHT panels stays at the EXACT same x-coordinate.\n • Object COUNT inside the panels equals the source EXACTLY — never add, skip, merge, or split anything inside a panel.\n • No signal-level modifications inside the panels: no extra noise, paper grain, blur, JPEG artefacts, or colour shifts on the panel pixels (these have already been baked in upstream). The panels are passed through UNCHANGED.\n • You may ONLY add real-world context AROUND the panels (outside their bounding boxes). Inside the panels: ZERO change.\n\nPlace the unchanged two-signal-trace panel as if displayed on a flat-panel monitor seen at slight glancing angle (so the bezel and minor screen reflections frame the panel). Critically, the panel content is rendered crisp and unmodified — only the surrounding monitor bezel + faint room reflection is added. Use a 2:1 ratio.", + "⚠️ PIXEL-EXACT (NON-NEGOTIABLE):\n • The TWO-PANEL CONTENT is sacred. Every cell, every grid line, every waveform sample, every gridline, every axis tick, every label, every divider, every panel border MUST be reproduced EXACTLY — do NOT alter, recolour, redraw, blur, smooth, re-rasterise, or shift any cell within the panels.\n • The vertical divider between LEFT and RIGHT panels stays at the EXACT same x-coordinate.\n • Object COUNT inside the panels equals the source EXACTLY — never add, skip, merge, or split anything inside a panel.\n • No signal-level modifications inside the panels: no extra noise, paper grain, blur, JPEG artefacts, or colour shifts on the panel pixels (these have already been baked in upstream). The panels are passed through UNCHANGED.\n • You may ONLY add real-world context AROUND the panels (outside their bounding boxes). Inside the panels: ZERO change.\n\nPlace the unchanged two-signal-trace panel as a strip lying on a darker fabric background (denim or felt), with a small ruler placed alongside (parallel, OUTSIDE the panel) and a metal washer near the corner. Lighting is even daylight. Use a 2:1 ratio." + ] + }, + "visual_attribute_transfer/spot_the_stroke_diff": { + "src": "images/stroke_diff_00000.png", + "prompts": [ + "Re-render the two panels as a photorealistic top-down photograph using TWO DIFFERENT object types for left vs right — same vertical divider, same overall lighting bias. Pick TWO different items at random from this pool of curve-shaped objects, one for the LEFT panel and a different one for the RIGHT panel:\n thin twigs, dry pine needles, blades of grass, fallen leaf-stems, pieces of twine, paracord segments, ribbons, leather thongs, cooked spaghetti, ramen noodles, copper wire snippets, hairpins, pencil shavings, dried herb stalks, fish bones, eel grass.\n\nThe LEFT panel sits on damp moss with twigs; the RIGHT panel sits on a polished slate plate. The two surfaces share the same lighting but their materials and colours are completely different.\n\nACCURACY IS CRITICAL:\n • EVERY brush stroke in the source becomes EXACTLY ONE elongated object on the matching side, following the SAME curve path, SAME length, SAME (x, y) — within ~5 px. No drift, no rescale, no flip.\n • Same OBJECT COUNT per panel as the source. Never add, skip, merge, or split strokes.\n • Object TYPE differs left vs right, but each curve is identical to the source stroke on its side. A curve overlay should align almost perfectly.\n • Background fills empty space only — must not create new stroke-like shapes.\n\nANTI-PROGRAM-COMPARISON HARDENING:\n • Each panel has a NOISY, lived-in surface — micro-scratches, dust, subtle wear, slight colour drift; same genre of noise across both panels but DIFFERENT realisations on each side.\n • Each panel has its own directional lighting (slightly different angle and colour temperature), so the same curve renders with different shadow direction and brightness on the two sides.\n • Per-object micro-variation (texture / shade / sheen) so corresponding curves across panels never share a pixel-level signature.\n\nNo text, no labels, no overlays.", + "Re-render the two panels as a photorealistic top-down photograph using TWO DIFFERENT object types for left vs right — same vertical divider, same overall lighting bias. Pick TWO different items at random from this pool of curve-shaped objects, one for the LEFT panel and a different one for the RIGHT panel:\n yarn strands, embroidery thread, wool roving, cotton thread, silk thread, sinew, horsehair, fishing line, dental floss, jute twine, hemp string, linen thread.\n\nThe LEFT panel sits on a tailor's wooden cutting mat; the RIGHT panel sits on a denim cotton sheet. The two surfaces share the same lighting but their materials and colours are completely different.\n\nACCURACY IS CRITICAL:\n • EVERY brush stroke in the source becomes EXACTLY ONE elongated object on the matching side, following the SAME curve path, SAME length, SAME (x, y) — within ~5 px. No drift, no rescale, no flip.\n • Same OBJECT COUNT per panel as the source. Never add, skip, merge, or split strokes.\n • Object TYPE differs left vs right, but each curve is identical to the source stroke on its side. A curve overlay should align almost perfectly.\n • Background fills empty space only — must not create new stroke-like shapes.\n\nANTI-PROGRAM-COMPARISON HARDENING:\n • Each panel has a NOISY, lived-in surface — micro-scratches, dust, subtle wear, slight colour drift; same genre of noise across both panels but DIFFERENT realisations on each side.\n • Each panel has its own directional lighting (slightly different angle and colour temperature), so the same curve renders with different shadow direction and brightness on the two sides.\n • Per-object micro-variation (texture / shade / sheen) so corresponding curves across panels never share a pixel-level signature.\n\nNo text, no labels, no overlays.", + "Re-render the two panels as a photorealistic top-down photograph using TWO DIFFERENT object types for left vs right — same vertical divider, same overall lighting bias. Pick TWO different items at random from this pool of curve-shaped objects, one for the LEFT panel and a different one for the RIGHT panel:\n spaghetti, udon, soba, rice vermicelli, lasagna strips, churros, breadsticks, pretzel sticks, liquorice sticks, chow-mein noodles, fettuccine, linguine.\n\nThe LEFT panel sits on a marble pastry slab; the RIGHT panel sits on a dark slate plate. The two surfaces share the same lighting but their materials and colours are completely different.\n\nACCURACY IS CRITICAL:\n • EVERY brush stroke in the source becomes EXACTLY ONE elongated object on the matching side, following the SAME curve path, SAME length, SAME (x, y) — within ~5 px. No drift, no rescale, no flip.\n • Same OBJECT COUNT per panel as the source. Never add, skip, merge, or split strokes.\n • Object TYPE differs left vs right, but each curve is identical to the source stroke on its side. A curve overlay should align almost perfectly.\n • Background fills empty space only — must not create new stroke-like shapes.\n\nANTI-PROGRAM-COMPARISON HARDENING:\n • Each panel has a NOISY, lived-in surface — micro-scratches, dust, subtle wear, slight colour drift; same genre of noise across both panels but DIFFERENT realisations on each side.\n • Each panel has its own directional lighting (slightly different angle and colour temperature), so the same curve renders with different shadow direction and brightness on the two sides.\n • Per-object micro-variation (texture / shade / sheen) so corresponding curves across panels never share a pixel-level signature.\n\nNo text, no labels, no overlays.", + "Re-render the two panels as a photorealistic top-down photograph using TWO DIFFERENT object types for left vs right — same vertical divider, same overall lighting bias. Pick TWO different items at random from this pool of curve-shaped objects, one for the LEFT panel and a different one for the RIGHT panel:\n copper wire, aluminium wire, solder strips, steel cable, nylon zip-ties, fibre-optic strands, jumper wires, leather laces, electric ribbon cable, brass rod offcuts.\n\nThe LEFT panel sits on a galvanised steel workbench; the RIGHT panel sits on a weathered concrete floor. The two surfaces share the same lighting but their materials and colours are completely different.\n\nACCURACY IS CRITICAL:\n • EVERY brush stroke in the source becomes EXACTLY ONE elongated object on the matching side, following the SAME curve path, SAME length, SAME (x, y) — within ~5 px. No drift, no rescale, no flip.\n • Same OBJECT COUNT per panel as the source. Never add, skip, merge, or split strokes.\n • Object TYPE differs left vs right, but each curve is identical to the source stroke on its side. A curve overlay should align almost perfectly.\n • Background fills empty space only — must not create new stroke-like shapes.\n\nANTI-PROGRAM-COMPARISON HARDENING:\n • Each panel has a NOISY, lived-in surface — micro-scratches, dust, subtle wear, slight colour drift; same genre of noise across both panels but DIFFERENT realisations on each side.\n • Each panel has its own directional lighting (slightly different angle and colour temperature), so the same curve renders with different shadow direction and brightness on the two sides.\n • Per-object micro-variation (texture / shade / sheen) so corresponding curves across panels never share a pixel-level signature.\n\nNo text, no labels, no overlays.", + "Re-render the two panels as a photorealistic top-down photograph using TWO DIFFERENT object types for left vs right — same vertical divider, same overall lighting bias. Pick TWO different items at random from this pool of curve-shaped objects, one for the LEFT panel and a different one for the RIGHT panel:\n twigs, dried pine needles, fern fronds, vine tendrils, fallen leaf veins, herb stalks, dried wheat stalks, split bamboo strips, dried lavender stems, juniper sprigs.\n\nThe LEFT panel sits on damp dark forest soil; the RIGHT panel sits on a pale weathered birch plank. The two surfaces share the same lighting but their materials and colours are completely different.\n\nACCURACY IS CRITICAL:\n • EVERY brush stroke in the source becomes EXACTLY ONE elongated object on the matching side, following the SAME curve path, SAME length, SAME (x, y) — within ~5 px. No drift, no rescale, no flip.\n • Same OBJECT COUNT per panel as the source. Never add, skip, merge, or split strokes.\n • Object TYPE differs left vs right, but each curve is identical to the source stroke on its side. A curve overlay should align almost perfectly.\n • Background fills empty space only — must not create new stroke-like shapes.\n\nANTI-PROGRAM-COMPARISON HARDENING:\n • Each panel has a NOISY, lived-in surface — micro-scratches, dust, subtle wear, slight colour drift; same genre of noise across both panels but DIFFERENT realisations on each side.\n • Each panel has its own directional lighting (slightly different angle and colour temperature), so the same curve renders with different shadow direction and brightness on the two sides.\n • Per-object micro-variation (texture / shade / sheen) so corresponding curves across panels never share a pixel-level signature.\n\nNo text, no labels, no overlays." + ] + }, + "visual_attribute_transfer/stroke_gesture_count": { + "src": "images/stroke_gesture_count_00000.png", + "prompts": [ + "Re-render as a photorealistic top-down photograph. Same vertical divider as source.\n\n⚠️ LOCKED CONSTRAINTS:\n • Scale: every rendered curve occupies the EXACT same pixel area as its source stroke. No rescaling, no normalising.\n • Divider: vertical separator stays at the source's exact x-coordinate. LEFT and RIGHT panel widths match the source.\n • Position: every curve stays at the same (x, y) and follows the same path/length/curvature within ~5 px.\n\nPANEL STRUCTURE:\n • LEFT (template): the ONE source stroke → ONE elongated object of class A on the LEFT surface.\n • RIGHT (field): every source stroke → ONE elongated object of class B on the RIGHT surface. Object count must equal source exactly — never add, skip, merge, or split.\n\nPick TWO different items from this pool — one for LEFT (class A), a different one for RIGHT (class B):\n thin twigs, dry pine needles, blades of grass, twine, paracord, ribbon, leather thong, spaghetti, ramen, copper wire, hairpin, herb stalk, fish bone, eel grass.\n\nLEFT surface: damp moss with twigs. RIGHT surface: polished slate plate. Each panel has its own directional lighting (slightly different angle / colour temperature). Per-object micro-variation in shade/sheen so corresponding curves never share a pixel-level signature. Background must not create new stroke-like shapes.\n\nNo text, no labels, no overlays.\n\n⚠️ LEFT-PANEL PRECISION (HIGHEST PRIORITY): The single template stroke on the LEFT must be reproduced with PIXEL-EXACT precision — same start point, same end point, same path, same length, same curvature, same thickness, same (x, y) position as the source stroke within ~2 px. Do NOT move it, do NOT recentre it, do NOT lengthen or shorten it, do NOT smooth or restyle the curve. The LEFT stroke is a precision reference — a curve overlay must align almost perfectly.", + "Re-render as a photorealistic top-down photograph. Same vertical divider as source.\n\n⚠️ LOCKED CONSTRAINTS:\n • Scale: every rendered curve occupies the EXACT same pixel area as its source stroke. No rescaling, no normalising.\n • Divider: vertical separator stays at the source's exact x-coordinate. LEFT and RIGHT panel widths match the source.\n • Position: every curve stays at the same (x, y) and follows the same path/length/curvature within ~5 px.\n\nPANEL STRUCTURE:\n • LEFT (template): the ONE source stroke → ONE elongated object of class A on the LEFT surface.\n • RIGHT (field): every source stroke → ONE elongated object of class B on the RIGHT surface. Object count must equal source exactly — never add, skip, merge, or split.\n\nPick TWO different items from this pool — one for LEFT (class A), a different one for RIGHT (class B):\n yarn, embroidery thread, wool roving, cotton thread, silk thread, sinew, horsehair, fishing line, dental floss, jute twine, hemp string, linen thread.\n\nLEFT surface: tailor's wooden cutting mat. RIGHT surface: denim cotton sheet. Each panel has its own directional lighting (slightly different angle / colour temperature). Per-object micro-variation in shade/sheen so corresponding curves never share a pixel-level signature. Background must not create new stroke-like shapes.\n\nNo text, no labels, no overlays.\n\n⚠️ LEFT-PANEL PRECISION (HIGHEST PRIORITY): The single template stroke on the LEFT must be reproduced with PIXEL-EXACT precision — same start point, same end point, same path, same length, same curvature, same thickness, same (x, y) position as the source stroke within ~2 px. Do NOT move it, do NOT recentre it, do NOT lengthen or shorten it, do NOT smooth or restyle the curve. The LEFT stroke is a precision reference — a curve overlay must align almost perfectly.", + "Re-render as a photorealistic top-down photograph. Same vertical divider as source.\n\n⚠️ LOCKED CONSTRAINTS:\n • Scale: every rendered curve occupies the EXACT same pixel area as its source stroke. No rescaling, no normalising.\n • Divider: vertical separator stays at the source's exact x-coordinate. LEFT and RIGHT panel widths match the source.\n • Position: every curve stays at the same (x, y) and follows the same path/length/curvature within ~5 px.\n\nPANEL STRUCTURE:\n • LEFT (template): the ONE source stroke → ONE elongated object of class A on the LEFT surface.\n • RIGHT (field): every source stroke → ONE elongated object of class B on the RIGHT surface. Object count must equal source exactly — never add, skip, merge, or split.\n\nPick TWO different items from this pool — one for LEFT (class A), a different one for RIGHT (class B):\n spaghetti, udon, soba, rice vermicelli, lasagna strips, churros, breadsticks, pretzel sticks, liquorice sticks, chow-mein noodles, fettuccine, linguine.\n\nLEFT surface: marble pastry slab. RIGHT surface: dark slate plate. Each panel has its own directional lighting (slightly different angle / colour temperature). Per-object micro-variation in shade/sheen so corresponding curves never share a pixel-level signature. Background must not create new stroke-like shapes.\n\nNo text, no labels, no overlays.\n\n⚠️ LEFT-PANEL PRECISION (HIGHEST PRIORITY): The single template stroke on the LEFT must be reproduced with PIXEL-EXACT precision — same start point, same end point, same path, same length, same curvature, same thickness, same (x, y) position as the source stroke within ~2 px. Do NOT move it, do NOT recentre it, do NOT lengthen or shorten it, do NOT smooth or restyle the curve. The LEFT stroke is a precision reference — a curve overlay must align almost perfectly.", + "Re-render as a photorealistic top-down photograph. Same vertical divider as source.\n\n⚠️ LOCKED CONSTRAINTS:\n • Scale: every rendered curve occupies the EXACT same pixel area as its source stroke. No rescaling, no normalising.\n • Divider: vertical separator stays at the source's exact x-coordinate. LEFT and RIGHT panel widths match the source.\n • Position: every curve stays at the same (x, y) and follows the same path/length/curvature within ~5 px.\n\nPANEL STRUCTURE:\n • LEFT (template): the ONE source stroke → ONE elongated object of class A on the LEFT surface.\n • RIGHT (field): every source stroke → ONE elongated object of class B on the RIGHT surface. Object count must equal source exactly — never add, skip, merge, or split.\n\nPick TWO different items from this pool — one for LEFT (class A), a different one for RIGHT (class B):\n copper wire, aluminium wire, solder strips, steel cable, nylon zip-ties, fibre-optic strands, jumper wires, leather laces, ribbon cable, brass rod offcuts.\n\nLEFT surface: galvanised steel workbench. RIGHT surface: weathered concrete floor. Each panel has its own directional lighting (slightly different angle / colour temperature). Per-object micro-variation in shade/sheen so corresponding curves never share a pixel-level signature. Background must not create new stroke-like shapes.\n\nNo text, no labels, no overlays.", + "Re-render as a photorealistic top-down photograph. Same vertical divider as source.\n\n⚠️ LOCKED CONSTRAINTS:\n • Scale: every rendered curve occupies the EXACT same pixel area as its source stroke. No rescaling, no normalising.\n • Divider: vertical separator stays at the source's exact x-coordinate. LEFT and RIGHT panel widths match the source.\n • Position: every curve stays at the same (x, y) and follows the same path/length/curvature within ~5 px.\n\nPANEL STRUCTURE:\n • LEFT (template): the ONE source stroke → ONE elongated object of class A on the LEFT surface.\n • RIGHT (field): every source stroke → ONE elongated object of class B on the RIGHT surface. Object count must equal source exactly — never add, skip, merge, or split.\n\nPick TWO different items from this pool — one for LEFT (class A), a different one for RIGHT (class B):\n twigs, dried pine needles, fern fronds, vine tendrils, leaf veins, herb stalks, dried wheat stalks, split bamboo strips, dried lavender stems, juniper sprigs.\n\nLEFT surface: damp dark forest soil. RIGHT surface: pale weathered birch plank. Each panel has its own directional lighting (slightly different angle / colour temperature). Per-object micro-variation in shade/sheen so corresponding curves never share a pixel-level signature. Background must not create new stroke-like shapes.\n\nNo text, no labels, no overlays." + ] + } + } +} \ No newline at end of file diff --git a/code/scope.md b/code/scope.md new file mode 100644 index 0000000000000000000000000000000000000000..23e2984b1a2d0047edfb3c1c6a3d0e3f4eefafce --- /dev/null +++ b/code/scope.md @@ -0,0 +1,38 @@ +# Active Vision Benchmark + +Today's multimodal large language models (MLLMs) achieve strong performance on many vision-language tasks, but they typically process images as fixed embeddings. Human reasoning, by contrast, is often active: perception is continuously guided by intermediate reasoning. In psychology, active observers can readily solve tasks that are ill-posed for passive observers. In this paper, we investigate whether MLLMs can exhibit a similar form of active observation. We introduce a benchmark that requires iterative visual inspection, including distributed scanning, sequential traversal, and visual attribute transfer, where evidence must be accumulated across spatial locations and reasoning steps. State-of-the-art MLLMs show substantial performance drops on these tasks, and attention analysis reveals unstable and limited use of visual tokens during reasoning. These results suggest that current MLLMs lack robust active visual observation, motivating new methods and architectures for iterative, perception-driven reasoning. + + +### Distributed Scanning. + +Tasks that require the model to inspect multiple spatially separated regions and aggregate locally identifiable evidence across the image. The main challenge is exhaustive visual coverage rather than global structural inference, as the answer is obtained by repeatedly finding and accumulating relevant local signals. + +### Sequential Traversal. + +Tasks that require the model to follow a path, line, wire, or other connected structure step by step while maintaining intermediate state. The answer depends on ordered inspection along the structure, such as tracing a route, identifying visited elements, or counting events encountered during traversal. + +### Visual Attribute Transfer. + +Tasks that require the model to extract a fine-grained visual property from one region and match or apply it to another region. The transferred property is primarily visual rather than linguistic, such as length, curvature, thickness, or spacing, and the task tests whether the model can preserve and compare such information across separated parts of the image. + +### Requirement + +A task qualifies for the benchmark only if it forces the model to *actively look at the image during reasoning*. Concretely, every task must defeat the following two shortcuts. + +**Shortcut 1 — One-shot perception, then pure text reasoning.** +The model summarises the entire image in a single pass into a compact symbolic description (e.g. an adjacency list, a coordinate table, a grid of arrows), then solves the task by reasoning over that description without looking again. To block this, an instance must contain too much fine-grained visual state to be losslessly extracted in one pass at the resolutions models actually see. For example, a *Color Zone Sequence* image carries a continuous smooth curve whose region membership changes dozens of times along its length; verbalising every crossing in advance is itself the task. A *Connectivity Spotting* graph is dense enough that the model would have to verbalise the entire connectivity structure — at which point it is solving the task, not pre-extracting it. Difficulty should scale with image complexity (more zones, more arrows, more crossings) so that the one-shot description grows past what the model can reliably hold. + +**Shortcut 2 — Write code to solve the task end-to-end.** +The model emits Python that runs OCR, edge detection, blob counting, or vector tracing on the raw image and returns the answer without further reasoning. To block this, the visual primitives must require human-style perception that off-the-shelf CV libraries do not solve cleanly: smoothly anti-aliased turtle curves rather than crisp lines, color-only region boundaries with no contour to detect, hand-placed arrows pointing in arbitrary directions, weighted-graph layouts whose edges are styled rather than thresholded. The answer should also depend on a continuous tracing or counting decision (which arrow comes next, which region is visited 4th) that scripted pipelines get wrong on the soft, irregular renderings used here. + +**Shortcut 3 — Statistical priors / answer-distribution leakage.** +Without looking, the model guesses the modal answer for the task type (e.g. "shortest paths are usually 3 hops", "arrow chains usually terminate at A"). To block this, per-task answer distributions should be flat, with no correlation between trivially extractable image features (region count, canvas size, color palette) and the answer. Each task's annotations should be auditable for this. + +**Shortcut 4 — Gestalt heuristics instead of step-by-step tracing.** +The model glances at the image and uses a learned visual prior — "the green arrow's circle is on the left, the red terminus is on the right, so go right" — without actually executing the traversal. This is especially dangerous on Sequential Traversal: the model can short-circuit a 6-hop arrow chain by interpolating from start-position and end-position. To block this, insert decoys and detours so the geometric "obvious" path is wrong, the answer requires more than half the canvas to be inspected, and the traversal length is long enough that intermediate state is required (mental running count, current-region tracking). + +**Shortcut 5 — Memorisation of the released benchmark.** +Once the dataset is public, models can be fine-tuned on it directly. To block this, the generation pipeline is the artifact. We release `creation.py` with a seed protocol so that *new* unseen instances can be regenerated by evaluators, and we keep a small held-out split with seeds never published. Difficulty scaling (more zones, longer chains) also means evaluators can request harder splits than the released set. + +**Shortcut 6 — Agent-tool zoom / crop / OCR loops.** +A tool-using agent can repeatedly crop, upscale, or OCR sub-regions until the answer falls out, without exercising the model's own active visual reasoning. To block this in tool-using settings, the canvas must already be at a resolution where a single zoom does not localise the answer, the answer should depend on integrating evidence across non-adjacent regions, and any single crop must omit information that another crop also needs. diff --git a/code/sequential_traversal/arrow_chain/airplane_stamps/00.png b/code/sequential_traversal/arrow_chain/airplane_stamps/00.png new file mode 100644 index 0000000000000000000000000000000000000000..fee1c78d148b8b10eb0a10f7b9a6498a4f0e07b6 --- /dev/null +++ b/code/sequential_traversal/arrow_chain/airplane_stamps/00.png @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:65e81a4d15048673df92691fde9477d59930353aaab26e5a8b2736f4a45ccb35 +size 187850 diff --git a/code/sequential_traversal/arrow_chain/airplane_stamps/01.png b/code/sequential_traversal/arrow_chain/airplane_stamps/01.png new file mode 100644 index 0000000000000000000000000000000000000000..3740c4d6558a9a72b8b041071c861a1d2a628a15 --- /dev/null +++ b/code/sequential_traversal/arrow_chain/airplane_stamps/01.png @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:4754b2e0047a1b3b75ce89e19043764aa1dd0e607facc8f172fb30c2d81c45c0 +size 240704 diff --git a/code/sequential_traversal/arrow_chain/airplane_stamps/02.png b/code/sequential_traversal/arrow_chain/airplane_stamps/02.png new file mode 100644 index 0000000000000000000000000000000000000000..9848c883ed545c375f10842f7a2cbfeff9cbdf98 --- /dev/null +++ b/code/sequential_traversal/arrow_chain/airplane_stamps/02.png @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:e76a08f4791bf4f099947d08d5505f8b807bedd1c2ae7e27ea99af51deadf54d +size 252483 diff --git a/code/sequential_traversal/arrow_chain/airplane_stamps/03.png b/code/sequential_traversal/arrow_chain/airplane_stamps/03.png new file mode 100644 index 0000000000000000000000000000000000000000..04e4ca1c72fc97049ae0daf4b7f396cf2b3a8cbc --- /dev/null +++ b/code/sequential_traversal/arrow_chain/airplane_stamps/03.png @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:df27f038e07dde9d4cd70c8b9ac0c864a647ce37535c5ac5325772e16f70a868 +size 225121 diff --git a/code/sequential_traversal/arrow_chain/airplane_stamps/04.png b/code/sequential_traversal/arrow_chain/airplane_stamps/04.png new file mode 100644 index 0000000000000000000000000000000000000000..54c1499d2b7e4dbad2eefe19def9d69089b3215a --- /dev/null +++ b/code/sequential_traversal/arrow_chain/airplane_stamps/04.png @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:77bc096953bf5fb7d17798ef81324c577e36c7b8d0496e11844af6881560074b +size 250649 diff --git a/code/sequential_traversal/arrow_chain/airplane_stamps/05.png b/code/sequential_traversal/arrow_chain/airplane_stamps/05.png new file mode 100644 index 0000000000000000000000000000000000000000..7e03d626b0799e569a7cf589c9eda2641487fa08 --- /dev/null +++ b/code/sequential_traversal/arrow_chain/airplane_stamps/05.png @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:0a94ef932f63a8b11c0d7aa76ea5b9bf61af36c3490c52add62f4930fe10da21 +size 257202 diff --git a/code/sequential_traversal/arrow_chain/airplane_template_grid.png b/code/sequential_traversal/arrow_chain/airplane_template_grid.png new file mode 100644 index 0000000000000000000000000000000000000000..ac789daf69a86637252e3805afcce8fd98fe6402 --- /dev/null +++ b/code/sequential_traversal/arrow_chain/airplane_template_grid.png @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:46b906cb5e01ab367ba753cea9cc59557ea09c305ece7b85022dc74b9138051d +size 1586346 diff --git a/code/sequential_traversal/arrow_chain/bird_stamps/00.png b/code/sequential_traversal/arrow_chain/bird_stamps/00.png new file mode 100644 index 0000000000000000000000000000000000000000..e31ee3fe54b0b879d6ee35af33fb5c17fde3d0ff --- /dev/null +++ b/code/sequential_traversal/arrow_chain/bird_stamps/00.png @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:464045407bf489582f02dcc6085494798b3a906e1c5654e7cb02b96c5e8c31fe +size 222164 diff --git a/code/sequential_traversal/arrow_chain/bird_stamps/01.png b/code/sequential_traversal/arrow_chain/bird_stamps/01.png new file mode 100644 index 0000000000000000000000000000000000000000..70de1593aa2d6ce9aa7ad58088500bc33f730d1d --- /dev/null +++ b/code/sequential_traversal/arrow_chain/bird_stamps/01.png @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:0e3d7ce827b8a6dd3aae97db73783f964656fc20f1a5a869ece9c32f43418b8d +size 212048 diff --git a/code/sequential_traversal/arrow_chain/bird_stamps/02.png b/code/sequential_traversal/arrow_chain/bird_stamps/02.png new file mode 100644 index 0000000000000000000000000000000000000000..86468bc9ecf964a006d15ff25dad850c8548e3d5 --- /dev/null +++ b/code/sequential_traversal/arrow_chain/bird_stamps/02.png @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:f0340e997718051b0f215b0f8fe867be3ab5269f6e31302bdcae2add0b000457 +size 246276 diff --git a/code/sequential_traversal/arrow_chain/bird_stamps/03.png b/code/sequential_traversal/arrow_chain/bird_stamps/03.png new file mode 100644 index 0000000000000000000000000000000000000000..a5d86c43b0e461ad15814f35534e777e4f7e1563 --- /dev/null +++ b/code/sequential_traversal/arrow_chain/bird_stamps/03.png @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:c868d092934c54c2ab724a9b039b7778649937095224e12dabfac9e13f17f44c +size 208566 diff --git a/code/sequential_traversal/arrow_chain/bird_stamps/04.png b/code/sequential_traversal/arrow_chain/bird_stamps/04.png new file mode 100644 index 0000000000000000000000000000000000000000..429f9877a67bc087008935429648b83cce9fa102 --- /dev/null +++ b/code/sequential_traversal/arrow_chain/bird_stamps/04.png @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:8db690c98014b1114f32a4e009e4f1a6f6c9f05a763de34d222383352c9e288c +size 195166 diff --git a/code/sequential_traversal/arrow_chain/bird_stamps/05.png b/code/sequential_traversal/arrow_chain/bird_stamps/05.png new file mode 100644 index 0000000000000000000000000000000000000000..4802566023ae12383d784a3f355bd76dd3c761db --- /dev/null +++ b/code/sequential_traversal/arrow_chain/bird_stamps/05.png @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:2a688c490a6d1b25d7fe827aebb3dcb55b63ba0596de25df6a767d47148c189a +size 220098 diff --git a/code/sequential_traversal/arrow_chain/bird_template_grid.png b/code/sequential_traversal/arrow_chain/bird_template_grid.png new file mode 100644 index 0000000000000000000000000000000000000000..3c11bb7c6fa29f532e4c18eabe70cdc4be746e3d --- /dev/null +++ b/code/sequential_traversal/arrow_chain/bird_template_grid.png @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:415caaffc60d3e2bf88542faf422b15515be39a7e1a504476a0879bfd4e54364 +size 1457722 diff --git a/code/sequential_traversal/arrow_chain/creation.py b/code/sequential_traversal/arrow_chain/creation.py new file mode 100644 index 0000000000000000000000000000000000000000..8eac4d14d3b5e01dc478fc05375df63357070481 --- /dev/null +++ b/code/sequential_traversal/arrow_chain/creation.py @@ -0,0 +1,700 @@ +"""Arrow Chain Traversal (irregular scattered arrows). + +Small arrows are scattered across an open canvas. Each arrow is enclosed +in its own circle, and points at the circle's center — so a ray extended +from the arrow through the centre exits the circle on the opposite side. +A handful of larger, labeled terminus circles sit among the arrows. + +The traversal rule is pure first-hit ray casting: + + - Start at the green arrow S. + - Draw a ray from the current arrow's centre along its pointing + direction. The next step is the FIRST other circle the ray enters. + - Continue until the ray enters a labeled terminus circle; report its + label. + +The design deliberately avoids a grid. The "next element" is a global, +ray-dependent matching problem over every circle on the canvas, so the +transition table can only be reconstructed by doing the same geometric +work as the task itself — for every arrow. Decoys are placed everywhere +except in corridors that would interfere with the intended chain. +""" +from __future__ import annotations + +import argparse +import json +import math +import os +import random +import string +from pathlib import Path +from typing import Dict, List, Tuple + +import matplotlib +matplotlib.use("Agg") +import matplotlib.image as mpimg +import matplotlib.patches as mpatches +import matplotlib.pyplot as plt +import numpy as np +from scipy.ndimage import rotate as _ndimage_rotate +from tqdm import tqdm + +# Stamp pools: each is a list of (template-image, default-rotation-offset-deg) +# where default-rotation tells the renderer how to interpret the template's +# natural orientation. The original foot stamp uses an offset of +50° because +# the photo's toes are tilted ~50° CCW from straight-up. The fish and key +# grids were generated with stamps facing straight up, so offset = 0°. +_FOOT_TEMPLATE: np.ndarray | None = None +_FOOT_TEMPLATE_PATH = ( + Path(__file__).resolve().parents[2] + / "visual_attribute_transfer/constellation_match_count/image.png" +) +_STAMP_POOLS: dict[str, tuple[list[np.ndarray], list[float]]] = {} + +# Pool name read from $ARROW_CHAIN_STAMP (default "foot"). Options: foot, fish, key. +_STAMP_POOL_NAME = os.environ.get("ARROW_CHAIN_STAMP", "foot") + + +def _foot_template() -> np.ndarray: + global _FOOT_TEMPLATE + if _FOOT_TEMPLATE is None: + _FOOT_TEMPLATE = mpimg.imread(str(_FOOT_TEMPLATE_PATH)) + return _FOOT_TEMPLATE + + +_BG_RGB = (0xf3 / 255.0, 0xef / 255.0, 0xe8 / 255.0) # matches scene BG + + +def _chroma_key_to_alpha(rgb: np.ndarray, tol: float = 0.16) -> np.ndarray: + """Convert HxWx3 to HxWx4 with the stamp's dominant CORNER colour + keyed to alpha=0. Auto-detects whether the background is light (fish + paper) or dark (key leather). Critical so the stamp doesn't carry a + visible halo into the scene that gpt-image-2 would render as a + container.""" + if rgb.ndim != 3 or rgb.shape[2] not in (3, 4): + return rgb + if rgb.shape[2] == 4: + return rgb + h, w = rgb.shape[:2] + # Sample 4 corners to estimate background colour. + corners = np.stack([rgb[0, 0], rgb[0, w-1], rgb[h-1, 0], rgb[h-1, w-1]]) + bg = corners.mean(axis=0) + # Per-pixel distance to background colour (Euclidean in RGB). + dist = np.linalg.norm(rgb - bg, axis=2) + # Smooth alpha: fully transparent at dist <= tol*0.5, fully opaque at dist >= tol*1.5. + lo = tol * 0.5 + hi = tol * 1.5 + alpha = np.clip((dist - lo) / max(hi - lo, 1e-6), 0.0, 1.0).astype(rgb.dtype) + rgba = np.concatenate([rgb, alpha[..., None]], axis=2) + return rgba + + +def _pad_to_square(arr: np.ndarray, fill: tuple[float, float, float] | None = None) -> np.ndarray: + """Pad an HxWxC array to a square so scipy.ndimage.rotate operates + isotropically. With RGBA arrays, padding is fully transparent.""" + h, w = arr.shape[:2] + n = max(h, w) + if h == w == n: + return arr + if arr.ndim == 3: + c = arr.shape[2] + pad = np.zeros((n, n, c), dtype=arr.dtype) + if c == 4: + pad[..., 3] = 0.0 # transparent + elif fill is None: + for k in range(3): + pad[..., k] = _BG_RGB[k] + else: + for k in range(3): + pad[..., k] = fill[k] + else: + pad = np.full((n, n), fill[0] if fill else 1.0, dtype=arr.dtype) + y0 = (n - h) // 2 + x0 = (n - w) // 2 + pad[y0:y0+h, x0:x0+w] = arr + return pad + + +def _stamp_pool(name: str) -> tuple[list[np.ndarray], list[float]]: + """Return (list-of-templates, list-of-rotation-offsets-deg) for the pool.""" + if name in _STAMP_POOLS: + return _STAMP_POOLS[name] + base = Path(__file__).resolve().parent + if name == "foot": + templates = [_pad_to_square(_foot_template())] + offsets = [50.0] # foot photo tilted ~50° CCW vs straight-up + elif name in ("fish", "key", "airplane", "bird", "leaf"): + d = base / f"{name}_stamps" + files = sorted(d.glob("*.png")) + if not files: + raise FileNotFoundError(f"no stamps in {d}") + templates = [ + _pad_to_square(_chroma_key_to_alpha(mpimg.imread(str(f)))) + for f in files + ] + # Per-stamp offsets stored by the calibration page. + offsets_path = d / "offsets.json" + per_stamp = {} + if offsets_path.exists(): + try: + per_stamp = json.loads(offsets_path.read_text()) + except Exception: + per_stamp = {} + offsets = [float(per_stamp.get(f"{i:02d}", 0.0)) for i in range(len(templates))] + else: + raise ValueError(f"unknown stamp pool: {name}") + _STAMP_POOLS[name] = (templates, offsets) + return _STAMP_POOLS[name] + + +ARROW_COLOR = "#2d2d2d" +START_COLOR = "#1f9d55" +TERMINUS_COLORS = [ + "#c23030", "#1a6dba", "#c47a18", "#7a35a0", + "#2d8e2d", "#b83280", "#0f766e", "#b45309", +] + + +def _wrap(a: float) -> float: + return (a + math.pi) % (2 * math.pi) - math.pi + + +def _ray_circle_entry( + origin: np.ndarray, + direction_unit: np.ndarray, + center: np.ndarray, + radius: float, +) -> float | None: + """Return the entry t along the ray into a circle, or None if it misses. + + t is the signed distance along the (unit) ray direction at which the + ray first crosses the circle boundary. t < 0 or None → the circle is + not hit forward from the origin. + """ + to_c = center - origin + proj = float(to_c[0] * direction_unit[0] + to_c[1] * direction_unit[1]) + to_c_sq = float(to_c[0] ** 2 + to_c[1] ** 2) + perp_sq = max(0.0, to_c_sq - proj * proj) + r_sq = radius * radius + if perp_sq > r_sq: + return None + offset = math.sqrt(r_sq - perp_sq) + t_entry = proj - offset + if t_entry < 0.0: + t_exit = proj + offset + if t_exit < 0.0: + return None + return 0.0 # origin is already inside the circle + return t_entry + + +def _ray_min_gap_to_circles( + origin: np.ndarray, + direction_angle: float, + t_end: float, + other_circles: List[Tuple[np.ndarray, float, object]], +) -> float: + """Minimum gap between the ray segment [0, t_end] and each circle boundary. + + For each circle, computes the closest distance from the segment to the + circle's center, subtracts the radius. Negative means the segment enters + the circle. Returns the minimum over all circles (most threatening near- + miss). + """ + dv = np.array([math.cos(direction_angle), math.sin(direction_angle)]) + best = float("inf") + for center, radius, _tag in other_circles: + to_c = center - origin + proj = float(to_c[0] * dv[0] + to_c[1] * dv[1]) + if proj < 0.0: + t_clamp = 0.0 + elif proj > t_end: + t_clamp = t_end + else: + t_clamp = proj + closest = origin + t_clamp * dv + d = float(math.hypot(center[0] - closest[0], center[1] - closest[1])) + gap = d - radius + if gap < best: + best = gap + return best + + +def _first_circle_hit( + origin: np.ndarray, + direction_angle: float, + circles: List[Tuple[np.ndarray, float, object]], + exclude_tag: object | None = None, +) -> Tuple[float, object] | None: + """Return (t_entry, tag) of the first circle entered along the ray.""" + dv = np.array([math.cos(direction_angle), math.sin(direction_angle)]) + best = None + for center, radius, tag in circles: + if tag == exclude_tag: + continue + t = _ray_circle_entry(origin, dv, center, radius) + if t is None: + continue + if best is None or t < best[0]: + best = (t, tag) + return best + + +def sample_instance( + rng: random.Random, + width: int, + height: int, + min_hops: int = 10, + max_hops: int = 13, + num_termini: int = 10, + num_decoys_target: int = 28, + arrow_radius: float = 22.0, + terminus_radius: float = 30.0, + step_length: float = 300.0, + step_jitter: float = 200.0, + max_attempts: int = 600, +) -> Dict | None: + """Build a chain + decoys + termini; return a record dict or None.""" + edge_margin = 110 + min_center_gap = 2 * arrow_radius + 14 # min distance between arrow centres + term_gap = arrow_radius + terminus_radius + 12 + term_term_gap = 2 * terminus_radius + 60 + # Required clearance (px) between any ray segment [origin → correct next + # circle's boundary] and every OTHER circle on the canvas. This prevents + # wrong circles from sitting ambiguously close to a ray that isn't meant + # to hit them. + clearance_px = 25.0 + + for _ in range(max_attempts): + # ── 1. Place termini around the canvas (spaced apart) ── + termini_pts: List[np.ndarray] = [] + t_attempts = 0 + while len(termini_pts) < num_termini and t_attempts < 600: + t_attempts += 1 + x = rng.uniform(edge_margin, width - edge_margin) + y = rng.uniform(edge_margin, height - edge_margin) + p = np.array([x, y]) + if all(float(np.linalg.norm(p - q)) > term_term_gap for q in termini_pts): + termini_pts.append(p) + if len(termini_pts) < num_termini: + continue + + num_hops = rng.randint(min_hops, max_hops) + + # ── 2. Build chain of arrow (pos, direction) entries ── + start_pos = np.array([ + rng.uniform(edge_margin + 40, width - edge_margin - 40), + rng.uniform(edge_margin + 40, height - edge_margin - 40), + ]) + start_dir = rng.uniform(0, 2 * math.pi) + chain: List[Tuple[np.ndarray, float]] = [(start_pos, start_dir)] + + ok = True + move_heading = start_dir + for _hop in range(num_hops - 1): + cur_pos, _ = chain[-1] + + cx_margin = min(cur_pos[0] - edge_margin, width - edge_margin - cur_pos[0]) + cy_margin = min(cur_pos[1] - edge_margin, height - edge_margin - cur_pos[1]) + bias_strength = 0.0 + bias_dir = 0.0 + if cx_margin < 220 or cy_margin < 220: + to_center = np.array([width / 2 - cur_pos[0], height / 2 - cur_pos[1]]) + bias_dir = math.atan2(to_center[1], to_center[0]) + bias_strength = max(0.0, 1.0 - min(cx_margin, cy_margin) / 220.0) + + placed = False + for _retry in range(30): + turn = rng.uniform(-math.pi / 4, math.pi / 4) + candidate_dir = move_heading + turn + if bias_strength > 0: + cx = math.cos(candidate_dir) * (1 - 0.5 * bias_strength) + \ + math.cos(bias_dir) * (0.5 * bias_strength) + cy = math.sin(candidate_dir) * (1 - 0.5 * bias_strength) + \ + math.sin(bias_dir) * (0.5 * bias_strength) + candidate_dir = math.atan2(cy, cx) + dvec = np.array([math.cos(candidate_dir), math.sin(candidate_dir)]) + pvec = np.array([-math.sin(candidate_dir), math.cos(candidate_dir)]) + step_len = step_length + rng.uniform(-step_jitter, step_jitter) + perp = rng.uniform(-step_jitter, step_jitter) + next_pos = cur_pos + step_len * dvec + perp * pvec + if not (edge_margin < next_pos[0] < width - edge_margin and + edge_margin < next_pos[1] < height - edge_margin): + continue + if any(float(np.linalg.norm(next_pos - t)) < term_gap for t in termini_pts): + continue + if any(float(np.linalg.norm(next_pos - p)) < min_center_gap for p, _ in chain): + continue + + actual_dir = math.atan2( + next_pos[1] - cur_pos[1], next_pos[0] - cur_pos[0], + ) + chain[-1] = (cur_pos, actual_dir) + chain.append((next_pos, actual_dir)) + move_heading = actual_dir + placed = True + break + if not placed: + ok = False + break + + if not ok or len(chain) < num_hops: + continue + + # ── 3. Pick a terminus the last arrow can point at without blockage ── + last_pos = chain[-1][0] + other_arrow_circles = [ + (chain[j][0], arrow_radius, ("A", j)) for j in range(num_hops - 1) + ] + terminus_circles = [ + (termini_pts[k], terminus_radius, ("T", k)) for k in range(num_termini) + ] + feasible = [] + for k in range(num_termini): + to_t = termini_pts[k] - last_pos + dist = float(math.hypot(to_t[0], to_t[1])) + if dist < arrow_radius + terminus_radius + 10: + continue + last_dir_k = math.atan2(to_t[1], to_t[0]) + hit = _first_circle_hit( + last_pos, last_dir_k, + other_arrow_circles + terminus_circles, + exclude_tag=("A", num_hops - 1), + ) + if hit is not None and hit[1] == ("T", k): + feasible.append(k) + if not feasible: + continue + chosen_term_idx = rng.choice(feasible) + last_dir = math.atan2( + termini_pts[chosen_term_idx][1] - last_pos[1], + termini_pts[chosen_term_idx][0] - last_pos[0], + ) + chain[-1] = (last_pos, last_dir) + + # ── 4. Verify chain integrity under the first-hit ray rule ── + arrow_positions = [c[0] for c in chain] + all_arrow_circles = [ + (arrow_positions[j], arrow_radius, ("A", j)) for j in range(num_hops) + ] + chain_valid = True + for i in range(num_hops): + hit = _first_circle_hit( + arrow_positions[i], chain[i][1], + all_arrow_circles + terminus_circles, + exclude_tag=("A", i), + ) + if hit is None: + chain_valid = False + break + expected = ("T", chosen_term_idx) if i == num_hops - 1 else ("A", i + 1) + if hit[1] != expected: + chain_valid = False + break + + # Clearance check: every OTHER circle must stay ≥ clearance_px + # away from the ray segment [0, t_correct]. + t_correct = hit[0] + other_circles = [ + c for c in (all_arrow_circles + terminus_circles) + if c[2] not in (("A", i), expected) + ] + if other_circles: + gap = _ray_min_gap_to_circles( + arrow_positions[i], chain[i][1], t_correct, other_circles, + ) + if gap < clearance_px: + chain_valid = False + break + if not chain_valid: + continue + + # ── 5. Add decoys that do not break any chain transition ── + decoys: List[Tuple[np.ndarray, float]] = [] + add_attempts = 0 + while len(decoys) < num_decoys_target and add_attempts < num_decoys_target * 30: + add_attempts += 1 + dpos = np.array([ + rng.uniform(edge_margin, width - edge_margin), + rng.uniform(edge_margin, height - edge_margin), + ]) + if any(float(np.linalg.norm(dpos - p)) < min_center_gap for p in arrow_positions): + continue + if any(float(np.linalg.norm(dpos - t)) < term_gap for t in termini_pts): + continue + if any(float(np.linalg.norm(dpos - p)) < min_center_gap for p, _ in decoys): + continue + + # A decoy circle is safe iff, for every chain ray, the ray + # segment up to the correct next circle's entry stays at least + # `clearance_px` away from the decoy boundary. + broken = False + for i in range(num_hops): + origin = arrow_positions[i] + dir_angle = chain[i][1] + dv_unit = np.array([math.cos(dir_angle), math.sin(dir_angle)]) + if i < num_hops - 1: + correct_c = arrow_positions[i + 1] + correct_r = arrow_radius + else: + correct_c = termini_pts[chosen_term_idx] + correct_r = terminus_radius + t_correct = _ray_circle_entry(origin, dv_unit, correct_c, correct_r) + if t_correct is None: + broken = True + break + gap = _ray_min_gap_to_circles( + origin, dir_angle, t_correct, + [(dpos, arrow_radius, ("D", len(decoys)))], + ) + if gap < clearance_px: + broken = True + break + if broken: + continue + + ddir = rng.uniform(0, 2 * math.pi) + decoys.append((dpos, ddir)) + + if len(decoys) < max(18, num_decoys_target - 8): + continue + + # ── 6. Assemble record ── + terminus_labels = list(string.ascii_uppercase[:num_termini]) + rng.shuffle(terminus_labels) + answer = terminus_labels[chosen_term_idx] + + question = ( + f"The image shows many small footprints scattered across the canvas, " + f"each footprint enclosed in its own circle, plus {num_termini} " + f"larger labeled terminus circles " + f"({', '.join(sorted(terminus_labels))}). The footprint inside the " + f"GREEN circle is the starting point. From that footprint, follow " + f"the direction its toes point: cast an infinitely thin ray (a " + f"mathematical half-line with zero width) from the footprint's " + f"circle centre along the toe-to-heel pointing direction, and the " + f"next step is the FIRST other circle this zero-width ray enters. A " + f"circle only counts if the ray actually crosses its boundary — " + f"grazing nearby without entering does not count. Continue until " + f"the ray enters a labeled terminus circle and report its label. " + f"Answer with a single letter. " + f"Provide your final answer enclosed in ... tags." + ) + + return { + "width": width, + "height": height, + "num_hops": num_hops, + "num_decoys": len(decoys), + "arrow_radius": arrow_radius, + "terminus_radius": terminus_radius, + "chain": [{"x": float(p[0]), "y": float(p[1]), "dir": float(d)} + for p, d in chain], + "decoys": [{"x": float(p[0]), "y": float(p[1]), "dir": float(d)} + for p, d in decoys], + "termini": [{"x": float(t[0]), "y": float(t[1]), + "label": terminus_labels[i]} + for i, t in enumerate(termini_pts)], + "chosen_terminus_label": answer, + "question": question, + "answer": answer, + } + return None + + +# ── Rendering ────────────────────────────────────────────────────── + +def _draw_arrow_in_circle( + ax, + cx: float, + cy: float, + direction: float, + arrow_radius: float, + circle_color: str, + arrow_color: str, + zorder: float = 2.0, + circle_lw: float = 1.2, + arrow_lw: float = 1.8, + circle_fill: str = "none", +) -> None: + """Stamp the foot template at (cx, cy), rotated so the toes point in + ``direction`` (radians, screen convention: 0=right, π/2=down, -π/2=up). + + The template's natural orientation has toes "up" (data direction -π/2), + so the rotation needed is direction + π/2 in display coords. scipy's + ndimage.rotate uses degrees, positive = counter-clockwise in array + coordinates (y-down). With imshow on inverted-y axes that visually + matches counter-clockwise on screen, so we negate. + """ + pool, offsets = _stamp_pool(_STAMP_POOL_NAME) + # Pick a stamp from the pool. Use a process-stable hash of (cx, cy) so the + # same cell always gets the same stamp across regenerations (within one + # difficulty/seed combo). + pick = int((cx * 91 + cy * 53)) % len(pool) + template = pool[pick] + rot_deg = -(math.degrees(direction) + 90.0) + offsets[pick] + rotated = _ndimage_rotate( + template, rot_deg, reshape=False, order=1, mode="constant", cval=1.0 + ) + # Match stamp visual radius to arrow_radius (the cell circle radius). + stamp_radius = arrow_radius + extent = (cx - stamp_radius, cx + stamp_radius, + cy + stamp_radius, cy - stamp_radius) + img_artist = ax.imshow(rotated, extent=extent, zorder=zorder + 0.1, + interpolation="bilinear") + clip = mpatches.Circle((cx, cy), radius=stamp_radius, transform=ax.transData) + img_artist.set_clip_path(clip) + + # Cell outline: only for the foot pool (fish/key cells get no outline so + # the stamp is not visually cropped or boxed-in). + if _STAMP_POOL_NAME == "foot": + circle = mpatches.Circle( + (cx, cy), radius=arrow_radius, + facecolor="none", edgecolor=circle_color, + linewidth=circle_lw, zorder=zorder + 0.2, + ) + ax.add_patch(circle) + elif circle_color != "#6b6b6b": + # Non-foot pool but a special cell (e.g. green start ring): still + # draw the outline so the start cell is identifiable. + circle = mpatches.Circle( + (cx, cy), radius=arrow_radius, + facecolor="none", edgecolor=circle_color, + linewidth=circle_lw, zorder=zorder + 0.2, + ) + ax.add_patch(circle) + + +def render_instance(out_path: Path, record: Dict, noise_seed: int) -> None: + width = int(record["width"]) + height = int(record["height"]) + arrow_radius = float(record["arrow_radius"]) + terminus_radius = float(record["terminus_radius"]) + + fig = plt.figure(figsize=(width / 100, height / 100), dpi=100) + ax = fig.add_axes([0, 0, 1, 1]) + ax.set_xlim(0, width) + ax.set_ylim(height, 0) + ax.axis("off") + ax.set_facecolor("#f3efe8") + + nrng = np.random.default_rng(noise_seed) + noise = nrng.normal(0.0, 1.0, size=(height, width)) + noise = (noise - noise.min()) / max(noise.max() - noise.min(), 1e-6) + ax.imshow(noise, cmap="Greys", alpha=0.06, extent=(0, width, height, 0), + interpolation="bilinear") + + for d in record["decoys"]: + _draw_arrow_in_circle( + ax, d["x"], d["y"], d["dir"], arrow_radius, + circle_color="#6b6b6b", arrow_color=ARROW_COLOR, + zorder=2.0, circle_lw=1.1, arrow_lw=1.7, + ) + + for i, c in enumerate(record["chain"]): + if i == 0: + # Start cell: foot stamp inside a thicker GREEN circle (no "S" label). + _draw_arrow_in_circle( + ax, c["x"], c["y"], c["dir"], arrow_radius, + circle_color=START_COLOR, arrow_color=START_COLOR, + zorder=3.6, circle_lw=2.6, arrow_lw=2.4, + ) + else: + _draw_arrow_in_circle( + ax, c["x"], c["y"], c["dir"], arrow_radius, + circle_color="#6b6b6b", arrow_color=ARROW_COLOR, + zorder=2.3, circle_lw=1.1, arrow_lw=1.8, + ) + + for i, t in enumerate(record["termini"]): + color = TERMINUS_COLORS[i % len(TERMINUS_COLORS)] + circle = mpatches.Circle( + (t["x"], t["y"]), radius=terminus_radius, + facecolor=color, edgecolor="white", linewidth=2.0, zorder=5.0, + ) + ax.add_patch(circle) + ax.text(t["x"], t["y"], t["label"], + fontsize=20, fontweight="bold", color="white", + ha="center", va="center", zorder=5.5) + + fig.savefig(out_path, dpi=100, bbox_inches="tight", pad_inches=0) + plt.close(fig) + + +def main() -> None: + parser = argparse.ArgumentParser() + parser.add_argument("--output-root", type=Path, required=True) + parser.add_argument("--count", type=int, default=20) + parser.add_argument("--seed", type=int, default=42) + parser.add_argument("--width", type=int, default=512) + parser.add_argument("--height", type=int, default=512) + parser.add_argument("--difficulty", type=int, default=5, + help="Integer difficulty >=0; scales termini/hops/decoys.") + args = parser.parse_args() + + def _canvas_scale(n_d, n_0): + import math + return math.sqrt(max(1.0, n_d / n_0)) + d = max(0, int(args.difficulty)) + N_d = 15 + 8 * d + N_0 = 15 + s = _canvas_scale(N_d, N_0) + args.width = int(round(args.width * s)) + args.height = int(round(args.height * s)) + + out_root = args.output_root + img_dir = out_root / "images" + img_dir.mkdir(parents=True, exist_ok=True) + + rng = random.Random(args.seed) + records = [] + + _num_termini = min(12, 5 + d) + _min_hops = 5 + 2 * d + _max_hops = 5 + 2 * d + 2 + _num_decoys_target = 10 + 6 * d + + pbar = tqdm(range(args.count), desc="Generating", unit="img") + for idx in pbar: + record = sample_instance( + rng, args.width, args.height, + min_hops=_min_hops, max_hops=_max_hops, + num_termini=_num_termini, + num_decoys_target=_num_decoys_target, + step_length=200.0, + step_jitter=100.0, # ray step range = [100, 300] + arrow_radius=26.0, + terminus_radius=26.0, + ) + if record is None: + pbar.set_postfix(status="FAILED") + continue + + name = f"arrow_chain_{idx:05d}.png" + ns = rng.randint(0, 10 ** 9) + render_instance(img_dir / name, record, noise_seed=ns) + + record["image"] = f"images/{name}" + records.append(record) + pbar.set_postfix(ok=len(records), hops=record["num_hops"], + decoys=record["num_decoys"], ans=record["answer"]) + + with (out_root / "annotations.jsonl").open("w") as fh: + for r in records: + fh.write(json.dumps(r) + "\n") + + data_json = { + "task": "arrow_chain", + "category": "sequential_traversal", + "count": len(records), + "items": [ + {"image": r["image"], "question": r["question"], "answer": r["answer"]} + for r in records + ], + } + (out_root / "data.json").write_text(json.dumps(data_json, indent=2)) + print(f"Saved {len(records)} items to {out_root}") + + +if __name__ == "__main__": + main() diff --git a/code/sequential_traversal/arrow_chain/data.json b/code/sequential_traversal/arrow_chain/data.json new file mode 100644 index 0000000000000000000000000000000000000000..e8b9a436dba09e4ae2ea4398f39332aa5568c06d --- /dev/null +++ b/code/sequential_traversal/arrow_chain/data.json @@ -0,0 +1,32 @@ +{ + "task": "arrow_chain", + "category": "sequential_traversal", + "count": 5, + "items": [ + { + "image": "images/arrow_chain_00000.png", + "question": "The image shows many small pieces scattered across the canvas, each clearly oriented in a specific direction, plus several larger labeled terminus markers (each marker bears a distinct letter). The piece highlighted by a green ring is the starting point. From that piece, cast an infinitely thin ray (a mathematical half-line with zero width) from its centre along the direction it points; the next step is the FIRST other marker this ray enters. A marker only counts if the ray actually crosses its boundary \u2014 grazing nearby without entering does not count. Continue until the ray enters a labeled terminus marker and report its letter. Answer with a single letter. Provide your final answer enclosed in ... tags.", + "answer": "H" + }, + { + "image": "images/arrow_chain_00001.png", + "question": "The image shows many small pieces scattered across the canvas, each clearly oriented in a specific direction, plus several larger labeled terminus markers (each marker bears a distinct letter). The piece highlighted by a green ring is the starting point. From that piece, cast an infinitely thin ray (a mathematical half-line with zero width) from its centre along the direction it points; the next step is the FIRST other marker this ray enters. A marker only counts if the ray actually crosses its boundary \u2014 grazing nearby without entering does not count. Continue until the ray enters a labeled terminus marker and report its letter. Answer with a single letter. Provide your final answer enclosed in ... tags.", + "answer": "B" + }, + { + "image": "images/arrow_chain_00002.png", + "question": "The image shows many small pieces scattered across the canvas, each clearly oriented in a specific direction, plus several larger labeled terminus markers (each marker bears a distinct letter). The piece highlighted by a green ring is the starting point. From that piece, cast an infinitely thin ray (a mathematical half-line with zero width) from its centre along the direction it points; the next step is the FIRST other marker this ray enters. A marker only counts if the ray actually crosses its boundary \u2014 grazing nearby without entering does not count. Continue until the ray enters a labeled terminus marker and report its letter. Answer with a single letter. Provide your final answer enclosed in ... tags.", + "answer": "B" + }, + { + "image": "images/arrow_chain_00003.png", + "question": "The image shows many small pieces scattered across the canvas, each clearly oriented in a specific direction, plus several larger labeled terminus markers (each marker bears a distinct letter). The piece highlighted by a green ring is the starting point. From that piece, cast an infinitely thin ray (a mathematical half-line with zero width) from its centre along the direction it points; the next step is the FIRST other marker this ray enters. A marker only counts if the ray actually crosses its boundary \u2014 grazing nearby without entering does not count. Continue until the ray enters a labeled terminus marker and report its letter. Answer with a single letter. Provide your final answer enclosed in ... tags.", + "answer": "J" + }, + { + "image": "images/arrow_chain_00004.png", + "question": "The image shows many small pieces scattered across the canvas, each clearly oriented in a specific direction, plus several larger labeled terminus markers (each marker bears a distinct letter). The piece highlighted by a green ring is the starting point. From that piece, cast an infinitely thin ray (a mathematical half-line with zero width) from its centre along the direction it points; the next step is the FIRST other marker this ray enters. A marker only counts if the ray actually crosses its boundary \u2014 grazing nearby without entering does not count. Continue until the ray enters a labeled terminus marker and report its letter. Answer with a single letter. Provide your final answer enclosed in ... tags.", + "answer": "C" + } + ] +} \ No newline at end of file diff --git a/code/sequential_traversal/arrow_chain/fish_stamps/00.png b/code/sequential_traversal/arrow_chain/fish_stamps/00.png new file mode 100644 index 0000000000000000000000000000000000000000..0e9d9c2f98646ed0620415846ddf647bab62f80f --- /dev/null +++ b/code/sequential_traversal/arrow_chain/fish_stamps/00.png @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:7c38c4a8ed2f007a96beefec37450756a07cdddca3930d6839add6c6733be11d +size 233869 diff --git a/code/sequential_traversal/arrow_chain/fish_stamps/01.png b/code/sequential_traversal/arrow_chain/fish_stamps/01.png new file mode 100644 index 0000000000000000000000000000000000000000..5c6807fb26f14e08cb03e0c7584eef2fe64f5e15 --- /dev/null +++ b/code/sequential_traversal/arrow_chain/fish_stamps/01.png @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:50d615e2eece2aef314e2611fc1d11dfd517c64fd0fb99ccd46442274cb812f4 +size 233146 diff --git a/code/sequential_traversal/arrow_chain/fish_stamps/02.png b/code/sequential_traversal/arrow_chain/fish_stamps/02.png new file mode 100644 index 0000000000000000000000000000000000000000..609c1d8d08890077bfbd09ca39d0ccab9699360f --- /dev/null +++ b/code/sequential_traversal/arrow_chain/fish_stamps/02.png @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:14adc8f86f52da854904292d71815abac4bb2d35b582a5907f5a0389602812b8 +size 235687 diff --git a/code/sequential_traversal/arrow_chain/fish_stamps/03.png b/code/sequential_traversal/arrow_chain/fish_stamps/03.png new file mode 100644 index 0000000000000000000000000000000000000000..9f7b5c512737780a40ac16f27a38ff438a7ee0ac --- /dev/null +++ b/code/sequential_traversal/arrow_chain/fish_stamps/03.png @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:41443fea877af7142e32ba1ce2d1b8fd3f643c99119145b062f5f95c1e3cb238 +size 230012 diff --git a/code/sequential_traversal/arrow_chain/fish_stamps/04.png b/code/sequential_traversal/arrow_chain/fish_stamps/04.png new file mode 100644 index 0000000000000000000000000000000000000000..95708bb1433f78778319838c515db749a101f039 --- /dev/null +++ b/code/sequential_traversal/arrow_chain/fish_stamps/04.png @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:476242649ef535e97160e057dc4f287439de65d05b4304e42dde8588bb40b1d0 +size 256895 diff --git a/code/sequential_traversal/arrow_chain/fish_stamps/05.png b/code/sequential_traversal/arrow_chain/fish_stamps/05.png new file mode 100644 index 0000000000000000000000000000000000000000..ac2eb1b2234fe2a255d1f93621865f7025b04ec3 --- /dev/null +++ b/code/sequential_traversal/arrow_chain/fish_stamps/05.png @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:6b8e3f5bb88ab602210ffe242323804059f9ffc0fe613325fbcb92e40e6510c4 +size 292111 diff --git a/code/sequential_traversal/arrow_chain/fish_stamps/offsets.json b/code/sequential_traversal/arrow_chain/fish_stamps/offsets.json new file mode 100644 index 0000000000000000000000000000000000000000..62d965696cf73cadeef7da1df2285e34f047856c --- /dev/null +++ b/code/sequential_traversal/arrow_chain/fish_stamps/offsets.json @@ -0,0 +1,8 @@ +{ + "02": -1, + "01": -1, + "05": -3, + "00": 0, + "03": 1, + "04": -1 +} \ No newline at end of file diff --git a/code/sequential_traversal/arrow_chain/fish_template_grid.png b/code/sequential_traversal/arrow_chain/fish_template_grid.png new file mode 100644 index 0000000000000000000000000000000000000000..8484d7e4ad66f23568289dbe31fc612690f730d8 --- /dev/null +++ b/code/sequential_traversal/arrow_chain/fish_template_grid.png @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:90ce4c1fc25d2231b2951a0c92de4680b06685c00219e945b67a5e0d1454c538 +size 1648151 diff --git a/code/sequential_traversal/arrow_chain/key_stamps/00.png b/code/sequential_traversal/arrow_chain/key_stamps/00.png new file mode 100644 index 0000000000000000000000000000000000000000..25f929d2d31ad1235a0a6a6930f67729fe2a49c2 --- /dev/null +++ b/code/sequential_traversal/arrow_chain/key_stamps/00.png @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:718f6a909567c51cda9c718487568655aa0edc80289be118ec3b3d22b768c75b +size 353767 diff --git a/code/sequential_traversal/arrow_chain/key_stamps/01.png b/code/sequential_traversal/arrow_chain/key_stamps/01.png new file mode 100644 index 0000000000000000000000000000000000000000..9055f8c8055fac305288887d01c23b333c6cbc61 --- /dev/null +++ b/code/sequential_traversal/arrow_chain/key_stamps/01.png @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:1387ac1ba5da1187a11d3a2805237faeefd281c15850c3cd847468900a58c0a7 +size 355619 diff --git a/code/sequential_traversal/arrow_chain/key_stamps/02.png b/code/sequential_traversal/arrow_chain/key_stamps/02.png new file mode 100644 index 0000000000000000000000000000000000000000..5bb367fa8d1f3d5f700b74c67d56407dfb56b18f --- /dev/null +++ b/code/sequential_traversal/arrow_chain/key_stamps/02.png @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:19f258a1619c1bd7c8a7a7152e5413b5743c5ec9183f64e92ffffb5ad10a6dd9 +size 348061 diff --git a/code/sequential_traversal/arrow_chain/key_stamps/03.png b/code/sequential_traversal/arrow_chain/key_stamps/03.png new file mode 100644 index 0000000000000000000000000000000000000000..e75bee6886e518c86ae28fe6bb5e3feed69b9350 --- /dev/null +++ b/code/sequential_traversal/arrow_chain/key_stamps/03.png @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:35f58cd24dc3aa5a69bca602b1bc7c9d2a8149b8b55ef3c8fc069909322b445c +size 338103 diff --git a/code/sequential_traversal/arrow_chain/key_stamps/04.png b/code/sequential_traversal/arrow_chain/key_stamps/04.png new file mode 100644 index 0000000000000000000000000000000000000000..244d3b583de5790348106796855f25cc8dc8cb21 --- /dev/null +++ b/code/sequential_traversal/arrow_chain/key_stamps/04.png @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:8e9ecc7691cccf2e48486456ad8c224e2592644e15bea8d6e6fad7f9eeffdd58 +size 351904 diff --git a/code/sequential_traversal/arrow_chain/key_stamps/05.png b/code/sequential_traversal/arrow_chain/key_stamps/05.png new file mode 100644 index 0000000000000000000000000000000000000000..67f355b1d97e19e8af4cfaeadd54a96871000258 --- /dev/null +++ b/code/sequential_traversal/arrow_chain/key_stamps/05.png @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:580046fd343fa85559e51a4c9162fee743aaa6bf2f8fe5abb877281a58d71233 +size 340712 diff --git a/code/sequential_traversal/arrow_chain/key_stamps/offsets.json b/code/sequential_traversal/arrow_chain/key_stamps/offsets.json new file mode 100644 index 0000000000000000000000000000000000000000..14da3ed0b063801c0be91821a6d13dd1ca8354d0 --- /dev/null +++ b/code/sequential_traversal/arrow_chain/key_stamps/offsets.json @@ -0,0 +1,6 @@ +{ + "02": 0, + "01": 0, + "00": 0, + "05": 0 +} \ No newline at end of file diff --git a/code/sequential_traversal/arrow_chain/key_template_grid.png b/code/sequential_traversal/arrow_chain/key_template_grid.png new file mode 100644 index 0000000000000000000000000000000000000000..19f39115f0466989fccbd4cc20b7bcb996dddd29 --- /dev/null +++ b/code/sequential_traversal/arrow_chain/key_template_grid.png @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:c64f69e57f58a934833f234164997bbac5a87d908f872a411c95962175013d1d +size 2409233 diff --git a/code/sequential_traversal/arrow_chain/leaf_stamps/00.png b/code/sequential_traversal/arrow_chain/leaf_stamps/00.png new file mode 100644 index 0000000000000000000000000000000000000000..9567a85ad54f3a67dba0efb526473aac6847b629 --- /dev/null +++ b/code/sequential_traversal/arrow_chain/leaf_stamps/00.png @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:8953d17afb96b3dec794d10a37492ee2668afadf84af6b2a53891ec6d1bf254f +size 227886 diff --git a/code/sequential_traversal/arrow_chain/leaf_stamps/01.png b/code/sequential_traversal/arrow_chain/leaf_stamps/01.png new file mode 100644 index 0000000000000000000000000000000000000000..099ac42308043c3840fa47fa74f1b76df90841c5 --- /dev/null +++ b/code/sequential_traversal/arrow_chain/leaf_stamps/01.png @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:19a45db8cf6750a903090e9b66498e563a0b0efcedfe3267ae2b3401bdb135dc +size 221224 diff --git a/code/sequential_traversal/arrow_chain/leaf_stamps/02.png b/code/sequential_traversal/arrow_chain/leaf_stamps/02.png new file mode 100644 index 0000000000000000000000000000000000000000..dc9adf906ffdac9819acaf0986fee9d0992eab9f --- /dev/null +++ b/code/sequential_traversal/arrow_chain/leaf_stamps/02.png @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:335f5f10eeb998bb43a667f4255812830d3bad682234b4394ece89ee485830bb +size 242252 diff --git a/code/sequential_traversal/arrow_chain/leaf_stamps/03.png b/code/sequential_traversal/arrow_chain/leaf_stamps/03.png new file mode 100644 index 0000000000000000000000000000000000000000..f4893bb11b52b8e7da7331ebfec529c588113ce0 --- /dev/null +++ b/code/sequential_traversal/arrow_chain/leaf_stamps/03.png @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:42438744c6fc84b52e90cb6e4840b12b6ac3fd516af274970113d2c74714cb5a +size 237067 diff --git a/code/sequential_traversal/arrow_chain/leaf_stamps/04.png b/code/sequential_traversal/arrow_chain/leaf_stamps/04.png new file mode 100644 index 0000000000000000000000000000000000000000..59ad88c0a3326a4549366063d27b81c68e22b27a --- /dev/null +++ b/code/sequential_traversal/arrow_chain/leaf_stamps/04.png @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:c854b2fa454cc8fb79d1e19d6b1ccf6335a5baadaa0b06f2f5ad1a76a985c5c1 +size 229205 diff --git a/code/sequential_traversal/arrow_chain/leaf_stamps/05.png b/code/sequential_traversal/arrow_chain/leaf_stamps/05.png new file mode 100644 index 0000000000000000000000000000000000000000..6564a4a0a31db699e478da605a339eadc49db24e --- /dev/null +++ b/code/sequential_traversal/arrow_chain/leaf_stamps/05.png @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:185f5518700dcb8ea9d0a62ba7947b1582aeaa74be145df94c0884681f133095 +size 219915 diff --git a/code/sequential_traversal/color_zone_sequence/creation.py b/code/sequential_traversal/color_zone_sequence/creation.py new file mode 100644 index 0000000000000000000000000000000000000000..545d7b3384b5a7643c0a2ec9f1a8a005d39fa806 --- /dev/null +++ b/code/sequential_traversal/color_zone_sequence/creation.py @@ -0,0 +1,1015 @@ +"""Color Zone Sequence (irregular regions). + +Colored irregular regions fill the canvas — no boundary lines are drawn, +only colored fills. A single smooth black curve starts at a marked dot +and travels through a sequence of regions (possibly revisiting). Report +the ordered list of region labels the curve visits. + +Regions are built with the same voronoi-partition-with-noise algorithm +as Counting Regions, then merely color-filled. The black curve is a +smooth turtle walk routed through deep-interior waypoints of a chosen +visit sequence, so every visit spends a substantial arc length well +inside the region, and every boundary crossing is a clean transverse +crossing (not a graze). Self-intersections are allowed but constrained +(sharp crossing angles, crossings well separated in space) following +the same recipe as Line Intersections. +""" +from __future__ import annotations + +import argparse +import heapq +import json +import math +import os +import random +import string +import sys +from collections import defaultdict +from pathlib import Path +from typing import Dict, List, Optional, Tuple + +import cv2 +import matplotlib +matplotlib.use("Agg") +import matplotlib.pyplot as plt +import numpy as np +from scipy.ndimage import zoom +from tqdm import tqdm + + +# ── Palette ──────────────────────────────────────────────────────── + +REGION_COLORS = [ + "#ec9863", # saturated peach + "#6fa3dc", # clear blue + "#8cc571", # grass green + "#df7fba", # pink + "#c2a055", # ochre + "#9d86d4", # purple + "#edd351", # yellow + "#4bb8ad", # teal + "#e07a70", # coral red + "#7aa06a", # sage green + "#d9a5e0", # lavender + "#c9d36b", # chartreuse + "#e8b08a", # light peach + "#7ec4c1", # light teal + "#b49278", # taupe + "#e2c28e", # light ochre +] + +LINE_COLOR = "#111111" +BG_COLOR = "#f7f3ea" + + +# ── Partition (voronoi-with-noise, adapted from counting_regions) ── + +def _sample_seeds(rng: random.Random, rows: int, cols: int, + count: int, min_dist: float, + max_attempts: int = 800) -> List[Tuple[int, int]]: + cells: List[Tuple[int, int]] = [] + for _ in range(max_attempts): + if len(cells) == count: + break + cand = (rng.randrange(rows), rng.randrange(cols)) + if all(math.hypot(cand[0] - r, cand[1] - c) >= min_dist for r, c in cells): + cells.append(cand) + while len(cells) < count: + cand = (rng.randrange(rows), rng.randrange(cols)) + if cand not in cells: + cells.append(cand) + return cells + + +def _noise_field(rng: random.Random, rows: int, cols: int, + scale: int = 15) -> np.ndarray: + lo_r = max(2, rows // scale) + lo_c = max(2, cols // scale) + lo = np.array([[rng.gauss(0, 1) for _ in range(lo_c)] for _ in range(lo_r)]) + field = zoom(lo, (rows / lo_r, cols / lo_c), order=1) + return field[:rows, :cols] + + +def make_partition(rng: random.Random, rows: int, cols: int, + num_regions: int) -> np.ndarray: + labels = np.full((rows, cols), -1, dtype=np.int32) + seeds = _sample_seeds( + rng, rows, cols, num_regions, + min_dist=max(2.2, min(rows, cols) / + max(3.4, math.sqrt(num_regions) + 0.5)), + ) + dirs = [(-1, 0), (1, 0), (0, -1), (0, 1)] + cr = [float(r) for r, c in seeds] + cc = [float(c) for r, c in seeds] + size = [1] * num_regions + fields = [_noise_field(rng, rows, cols, scale=15) for _ in range(num_regions)] + noise_strength = rows * cols * 0.015 + + heap: list = [] + for rid, (r, c) in enumerate(seeds): + labels[r, c] = rid + for dr, dc in dirs: + nr, nc = r + dr, c + dc + if 0 <= nr < rows and 0 <= nc < cols and labels[nr, nc] == -1: + d = (nr - cr[rid]) ** 2 + (nc - cc[rid]) ** 2 + d += fields[rid][nr, nc] * noise_strength + d += rng.random() * 5.0 + heapq.heappush(heap, (d, rid, nr, nc)) + while heap: + _, rid, nr, nc = heapq.heappop(heap) + if labels[nr, nc] != -1: + continue + labels[nr, nc] = rid + n = size[rid] + cr[rid] = (cr[rid] * n + nr) / (n + 1) + cc[rid] = (cc[rid] * n + nc) / (n + 1) + size[rid] += 1 + for dr, dc in dirs: + rr, ccc = nr + dr, nc + dc + if 0 <= rr < rows and 0 <= ccc < cols and labels[rr, ccc] == -1: + d = (rr - cr[rid]) ** 2 + (ccc - cc[rid]) ** 2 + d += fields[rid][rr, ccc] * noise_strength + d += rng.random() * 5.0 + heapq.heappush(heap, (d, rid, rr, ccc)) + return labels + + +def regions_thick_enough(labels: np.ndarray, num_regions: int, + min_thick: int = 6) -> bool: + for rid in range(num_regions): + mask = (labels == rid).astype(np.uint8) + if int(mask.sum()) == 0: + return False + kernel = cv2.getStructuringElement( + cv2.MORPH_ELLIPSE, (min_thick * 2 + 1, min_thick * 2 + 1)) + eroded = cv2.erode(mask, kernel) + if int(eroded.sum()) == 0: + return False + nc, _ = cv2.connectedComponents(eroded) + if nc > 2: + return False + return True + + +# ── Canvas label raster & distance transforms ───────────────────── + +def upsample_labels(grid_labels: np.ndarray, canvas_w: int, + canvas_h: int, smooth_sigma: float = 5.5) -> np.ndarray: + """Upsample the small region grid to canvas resolution with smooth + boundaries — for speed, do the blur at an intermediate 512² resolution + and then bilinearly upsample to the final canvas before argmax.""" + num_regions = int(grid_labels.max()) + 1 + inter = min(512, max(canvas_w, canvas_h)) + inter_w = min(canvas_w, inter) + inter_h = min(canvas_h, inter) + inter_sigma = smooth_sigma * inter_w / canvas_w + soft = np.empty((num_regions, canvas_h, canvas_w), dtype=np.float32) + for rid in range(num_regions): + mask = (grid_labels == rid).astype(np.float32) + mid = cv2.resize(mask, (inter_w, inter_h), + interpolation=cv2.INTER_LINEAR) + if inter_sigma > 0: + mid = cv2.GaussianBlur(mid, (0, 0), sigmaX=inter_sigma, + sigmaY=inter_sigma) + big = cv2.resize(mid, (canvas_w, canvas_h), + interpolation=cv2.INTER_LINEAR) + soft[rid] = big + return np.argmax(soft, axis=0).astype(np.int32) + + +def region_at(canvas_labels: np.ndarray, x: float, y: float) -> int: + h, w = canvas_labels.shape + ix, iy = int(x), int(y) + if 0 <= ix < w and 0 <= iy < h: + return int(canvas_labels[iy, ix]) + return -1 + + +def boundary_distance_transform(canvas_labels: np.ndarray) -> np.ndarray: + """Distance from each canvas pixel to the nearest inter-region + boundary pixel. Used to detect curve samples that graze a boundary + without actually crossing it (the analogue of ``_curves_too_close`` + in Line Intersections).""" + h, w = canvas_labels.shape + boundary = np.zeros((h, w), dtype=np.uint8) + diff_h = canvas_labels[:, :-1] != canvas_labels[:, 1:] + diff_v = canvas_labels[:-1, :] != canvas_labels[1:, :] + boundary[:, :-1][diff_h] = 1 + boundary[:, 1:][diff_h] = 1 + boundary[:-1, :][diff_v] = 1 + boundary[1:, :][diff_v] = 1 + inv = (1 - boundary).astype(np.uint8) + return cv2.distanceTransform(inv, cv2.DIST_L2, 5) + + +def curve_grazes_boundary( + curve: np.ndarray, + dt_boundary: np.ndarray, + point_ids: np.ndarray, + min_dist: float = 10.0, + window: int = 8, + sample_step: int = 2, +) -> bool: + """True iff the curve tangents any region boundary — i.e., a sample + comes within ``min_dist`` of the boundary but the region id does not + change within a ±``window`` neighborhood (meaning the curve + approached the boundary and veered back without crossing it). + + This is the direct analogue of ``_curves_too_close`` in Line + Intersections. There, a "real crossing" is an actual curve-B + intersection; here, it is a region-id change on the curve itself. + """ + h, w = dt_boundary.shape + n = len(curve) + for idx in range(0, n, sample_step): + x, y = float(curve[idx, 0]), float(curve[idx, 1]) + ix, iy = int(x), int(y) + if not (0 <= ix < w and 0 <= iy < h): + continue + if float(dt_boundary[iy, ix]) >= min_dist: + continue + lo = max(0, idx - window) + hi = min(n, idx + window + 1) + local = point_ids[lo:hi] + # If region id changes within the window, this is a real crossing. + if np.any(local[:-1] != local[1:]): + continue + return True + return False + + +def region_distance_transforms(canvas_labels: np.ndarray, + num_regions: int, + edge_margin: int = 18) -> List[np.ndarray]: + """Distance transform per region. Computed at a 512² intermediate + resolution and bilinearly upsampled (with distance values rescaled) + to keep cost roughly constant as canvas grows with difficulty.""" + h, w = canvas_labels.shape + inter = 512 + if max(h, w) <= inter: + small = canvas_labels + sh, sw = h, w + scale = 1.0 + em = edge_margin + else: + sw = min(inter, w) + sh = int(round(h * sw / w)) + small = cv2.resize(canvas_labels.astype(np.int32), (sw, sh), + interpolation=cv2.INTER_NEAREST) + scale = w / sw + em = max(1, int(round(edge_margin / scale))) + dts: List[np.ndarray] = [] + for rid in range(num_regions): + mask = (small == rid).astype(np.uint8) + mask[:em, :] = 0 + mask[-em:, :] = 0 + mask[:, :em] = 0 + mask[:, -em:] = 0 + dt_small = cv2.distanceTransform(mask, cv2.DIST_L2, 5) + if scale != 1.0: + dt_big = cv2.resize(dt_small, (w, h), + interpolation=cv2.INTER_LINEAR) * scale + else: + dt_big = dt_small + dts.append(dt_big) + return dts + + +def _dt_lookup(dt: np.ndarray, x: float, y: float) -> float: + h, w = dt.shape + ix, iy = int(x), int(y) + if 0 <= ix < w and 0 <= iy < h: + return float(dt[iy, ix]) + return 0.0 + + +def sample_deep_point(rng: random.Random, dt: np.ndarray, + min_depth: float = 45.0) -> Tuple[float, float]: + mask = dt >= min_depth + ys, xs = np.where(mask) + if len(xs) == 0: + iy, ix = np.unravel_index(int(np.argmax(dt)), dt.shape) + return float(ix), float(iy) + k = rng.randrange(len(xs)) + return float(xs[k]), float(ys[k]) + + +# ── Turtle curve ────────────────────────────────────────────────── + +def _angle_diff(a: float, b: float) -> float: + return (b - a + math.pi) % (2 * math.pi) - math.pi + + +def free_turtle(rng: random.Random, width: int, height: int, + start: Tuple[float, float], heading: float, + num_steps: int, step_size: float = 3.0, + max_turn: float = 0.048, + drift_rate: float = 0.013) -> np.ndarray: + """Free-drifting turtle, no waypoints. Mirrors the body-drift loop + in line_intersections' ``build_anchored_curve``: Gaussian drift, + clamped turns, gentle push away from canvas edges.""" + m = 24.0 + x, y = float(start[0]), float(start[1]) + theta = float(heading) + drift = 0.0 + pts: List[Tuple[float, float]] = [(x, y)] + for _ in range(num_steps): + drift += rng.gauss(0, drift_rate) + drift = max(-max_turn * 0.7, min(max_turn * 0.7, drift)) + turn = drift + rng.gauss(0, max_turn * 0.12) + + if x < m * 2: + turn += _angle_diff(theta, 0.0) * 0.22 + elif x > width - m * 2: + turn += _angle_diff(theta, math.pi) * 0.22 + if y < m * 2: + turn += _angle_diff(theta, math.pi / 2) * 0.22 + elif y > height - m * 2: + turn += _angle_diff(theta, -math.pi / 2) * 0.22 + + turn = max(-max_turn, min(max_turn, turn)) + theta += turn + x += step_size * math.cos(theta) + y += step_size * math.sin(theta) + x = max(m, min(width - m, x)) + y = max(m, min(height - m, y)) + pts.append((x, y)) + + return np.array(pts) + + +# ── Run-length sequence extraction ──────────────────────────────── + +def extract_sequence(curve: np.ndarray, canvas_labels: np.ndarray, + min_run: int = 22 + ) -> Tuple[np.ndarray, List[Tuple[int, int, int]], + List[Tuple[int, int, int]]]: + """Return (per-point region ids, all runs, stable runs). + + Each run is (region_id, start_index, end_index_exclusive). + Stable runs are those with length >= min_run and region_id >= 0. + """ + n = len(curve) + ids = np.empty(n, dtype=np.int32) + for i in range(n): + ids[i] = region_at(canvas_labels, curve[i, 0], curve[i, 1]) + + runs: List[Tuple[int, int, int]] = [] + cur = int(ids[0]) + start = 0 + for i in range(1, n): + if int(ids[i]) != cur: + runs.append((cur, start, i)) + cur = int(ids[i]) + start = i + runs.append((cur, start, n)) + + stable = [(rid, s, e) for rid, s, e in runs + if (e - s) >= min_run and rid >= 0] + return ids, runs, stable + + +# ── Self-intersection (from line_intersections) ─────────────────── + +def _segments_cross(p0, p1, q0, q1) -> bool: + eps = 1e-8 + o1 = (p1[0] - p0[0]) * (q0[1] - p0[1]) - (p1[1] - p0[1]) * (q0[0] - p0[0]) + o2 = (p1[0] - p0[0]) * (q1[1] - p0[1]) - (p1[1] - p0[1]) * (q1[0] - p0[0]) + o3 = (q1[0] - q0[0]) * (p0[1] - q0[1]) - (q1[1] - q0[1]) * (p0[0] - q0[0]) + o4 = (q1[0] - q0[0]) * (p1[1] - q0[1]) - (q1[1] - q0[1]) * (p1[0] - q0[0]) + return ((o1 > eps and o2 < -eps) or (o1 < -eps and o2 > eps)) and \ + ((o3 > eps and o4 < -eps) or (o3 < -eps and o4 > eps)) + + +def _crossing_angle(p0, p1, q0, q1) -> float: + d1 = p1 - p0 + d2 = q1 - q0 + cos_a = float(np.dot(d1, d2)) / ( + float(np.linalg.norm(d1)) * float(np.linalg.norm(d2)) + 1e-12) + return math.degrees(math.acos(min(1.0, abs(cos_a)))) + + +def find_self_crossings(curve: np.ndarray, + min_seg_gap: int = 30) -> List[Tuple[int, int, float, float, float]]: + n = len(curve) - 1 + if n <= 0: + return [] + xs_min = np.minimum(curve[:-1, 0], curve[1:, 0]) + xs_max = np.maximum(curve[:-1, 0], curve[1:, 0]) + ys_min = np.minimum(curve[:-1, 1], curve[1:, 1]) + ys_max = np.maximum(curve[:-1, 1], curve[1:, 1]) + cell = 60.0 + grid: Dict[Tuple[int, int], List[int]] = defaultdict(list) + for i in range(n): + gx0, gx1 = int(xs_min[i] / cell), int(xs_max[i] / cell) + gy0, gy1 = int(ys_min[i] / cell), int(ys_max[i] / cell) + for gx in range(gx0, gx1 + 1): + for gy in range(gy0, gy1 + 1): + grid[(gx, gy)].append(i) + + checked: set = set() + results: List[Tuple[int, int, float, float, float]] = [] + for cell_key, segs in grid.items(): + for ai in range(len(segs)): + for bi in range(ai + 1, len(segs)): + i, j = segs[ai], segs[bi] + ii, jj = min(i, j), max(i, j) + if jj - ii < min_seg_gap: + continue + key = (ii, jj) + if key in checked: + continue + checked.add(key) + p0, p1 = curve[ii], curve[ii + 1] + q0, q1 = curve[jj], curve[jj + 1] + if not _segments_cross(p0, p1, q0, q1): + continue + d1 = p1 - p0 + d2 = q1 - q0 + denom = d1[0] * d2[1] - d1[1] * d2[0] + if abs(denom) < 1e-9: + continue + ti = ((q0[0] - p0[0]) * d2[1] - (q0[1] - p0[1]) * d2[0]) / denom + px = float(p0[0] + ti * d1[0]) + py = float(p0[1] + ti * d1[1]) + ang = _crossing_angle(p0, p1, q0, q1) + results.append((ii, jj, ang, px, py)) + return results + + +# ── Label placement (deep in region, far from curve) ─────────────── + +def find_label_position(dt: np.ndarray, curve: np.ndarray, + canvas_w: int, canvas_h: int, + line_band: int = 14 + ) -> Tuple[Tuple[float, float], float]: + mask = np.ones((canvas_h, canvas_w), dtype=np.uint8) + pts = curve.astype(np.int32).reshape(-1, 1, 2) + cv2.polylines(mask, [pts], False, 0, thickness=line_band) + dt_curve = cv2.distanceTransform(mask, cv2.DIST_L2, 5) + + score = np.minimum(dt, dt_curve) + score[dt <= 0] = -1.0 + flat_idx = int(np.argmax(score)) + iy, ix = np.unravel_index(flat_idx, score.shape) + return (float(ix), float(iy)), float(score[iy, ix]) + + +# ── Instance sampling ───────────────────────────────────────────── + +def _distractor_too_close(poly_a: np.ndarray, poly_b: np.ndarray, + min_dist: float = 10.0, + sample_step: int = 3, + known_crossings: Optional[np.ndarray] = None, + crossing_exclude_radius: float = 30.0, + same_curve: bool = False, + self_index_gap: int = 50) -> bool: + """Return True if any point on poly_a comes within ``min_dist`` of poly_b. + + If ``known_crossings`` is given, samples within ``crossing_exclude_radius`` + of any crossing point are ignored — close approach there is the crossing + itself, which is allowed. Used to enforce "cross transversely, else stay + apart" hygiene between distractor curves and the query curve (and between + distractors). + """ + nb = len(poly_b) - 1 + if nb <= 0 or len(poly_a) == 0: + return False + b_pts = poly_b[:-1] + b_vecs = poly_b[1:] - poly_b[:-1] + b_lens_sq = np.maximum((b_vecs ** 2).sum(axis=1), 1e-12) + has_crossings = known_crossings is not None and len(known_crossings) > 0 + for idx in range(0, len(poly_a), sample_step): + p = poly_a[idx] + dp = p - b_pts + t = (dp * b_vecs).sum(axis=1) / b_lens_sq + t = np.clip(t, 0.0, 1.0) + proj = b_pts + t[:, None] * b_vecs + dists = np.sqrt(((p - proj) ** 2).sum(axis=1)) + if same_curve: + seg_idx = np.arange(nb) + mask = np.abs(seg_idx - idx) < self_index_gap + dists[mask] = 9999.0 + if dists.min() < min_dist: + if has_crossings: + cross_dists = np.sqrt(((p - known_crossings) ** 2).sum(axis=1)) + if cross_dists.min() < crossing_exclude_radius: + continue + return True + return False + + +def _segments_cross(p0, p1, q0, q1) -> bool: + eps = 1e-8 + o1 = (p1[0]-p0[0])*(q0[1]-p0[1]) - (p1[1]-p0[1])*(q0[0]-p0[0]) + o2 = (p1[0]-p0[0])*(q1[1]-p0[1]) - (p1[1]-p0[1])*(q1[0]-p0[0]) + o3 = (q1[0]-q0[0])*(p0[1]-q0[1]) - (q1[1]-q0[1])*(p0[0]-q0[0]) + o4 = (q1[0]-q0[0])*(p1[1]-q0[1]) - (q1[1]-q0[1])*(p1[0]-q0[0]) + return ((o1 > eps and o2 < -eps) or (o1 < -eps and o2 > eps)) and \ + ((o3 > eps and o4 < -eps) or (o3 < -eps and o4 > eps)) + + +def _crossing_angle(p0, p1, q0, q1) -> float: + d1 = p1 - p0 + d2 = q1 - q0 + cos_a = float(np.dot(d1, d2)) / (float(np.linalg.norm(d1)) * + float(np.linalg.norm(d2)) + 1e-12) + return math.degrees(math.acos(min(1.0, abs(cos_a)))) + + +def _find_pair_crossings(poly_a: np.ndarray, poly_b: np.ndarray, + step: int = 1, same_curve: bool = False, + min_seg_gap: int = 10) -> Tuple[int, float, np.ndarray]: + """Return (count, worst_angle_deg, crossing_points[N,2]) for all + intersections between poly_a and poly_b. Linear scan with bbox pre-filter. + If ``same_curve`` is True, treat poly_b as poly_a and only count + crossings between segments separated by at least ``min_seg_gap`` + indices (i < j enforced). + """ + na = len(poly_a) - 1 + nb = len(poly_b) - 1 + if na <= 0 or nb <= 0: + return 0, 180.0, np.zeros((0, 2)) + pts: List[Tuple[float, float]] = [] + worst = 180.0 + a_min_x = np.minimum(poly_a[:-1, 0], poly_a[1:, 0]) + a_max_x = np.maximum(poly_a[:-1, 0], poly_a[1:, 0]) + a_min_y = np.minimum(poly_a[:-1, 1], poly_a[1:, 1]) + a_max_y = np.maximum(poly_a[:-1, 1], poly_a[1:, 1]) + b_min_x = np.minimum(poly_b[:-1, 0], poly_b[1:, 0]) + b_max_x = np.maximum(poly_b[:-1, 0], poly_b[1:, 0]) + b_min_y = np.minimum(poly_b[:-1, 1], poly_b[1:, 1]) + b_max_y = np.maximum(poly_b[:-1, 1], poly_b[1:, 1]) + for i in range(0, na, step): + j_start = i + min_seg_gap if same_curve else 0 + for j in range(j_start, nb, step): + if a_max_x[i] < b_min_x[j] or b_max_x[j] < a_min_x[i] or \ + a_max_y[i] < b_min_y[j] or b_max_y[j] < a_min_y[i]: + continue + p0, p1 = poly_a[i], poly_a[i + 1] + q0, q1 = poly_b[j], poly_b[j + 1] + if not _segments_cross(p0, p1, q0, q1): + continue + d1 = p1 - p0 + d2 = q1 - q0 + denom = d1[0] * d2[1] - d1[1] * d2[0] + if abs(denom) < 1e-12: + continue + ti = ((q0[0] - p0[0]) * d2[1] - (q0[1] - p0[1]) * d2[0]) / denom + px = float(p0[0] + ti * d1[0]) + py = float(p0[1] + ti * d1[1]) + ang = _crossing_angle(p0, p1, q0, q1) + if ang < worst: + worst = ang + pts.append((px, py)) + if not pts: + return 0, 180.0, np.zeros((0, 2)) + return len(pts), worst, np.array(pts, dtype=np.float64) + + +def sample_instance( + rng: random.Random, + width: int, + height: int, + min_zones: int = 10, + max_zones: int = 10, + grid_size: int = 140, + min_stable_seq_len: int = 10, + max_stable_seq_len: int = 16, + min_distinct_regions: int = 8, + min_run: int = 22, + min_run_depth: float = 22.0, + min_label_score: float = 18.0, + min_cross_angle: float = 46.0, + min_cross_gap: float = 34.0, + max_attempts: int = 2000, + num_distractor_curves: int = 0, +) -> Optional[Dict]: + for _ in range(max_attempts): + num_zones = rng.randint(min_zones, max_zones) + + # Keep regenerating partition until regions are nicely thick. + part_ok = False + for _ in range(1): + grid_labels = make_partition(rng, grid_size, grid_size, num_zones) + # Reject if any region tiny + counts = np.bincount(grid_labels.ravel(), minlength=num_zones) + if counts.min() < grid_size * grid_size // (num_zones * 4): + continue + if regions_thick_enough(grid_labels, num_zones, min_thick=6): + part_ok = True + break + if not part_ok: + continue + + canvas_labels = upsample_labels(grid_labels, width, height) + dts = region_distance_transforms(canvas_labels, num_zones) + max_depths = [float(dt.max()) for dt in dts] + if min(max_depths) < 55.0: + continue + + # Pre-compute adjacency just for sanity checking the extracted + # sequence (consecutive stable regions must share a boundary). + adj_pairs: set = set() + for dr, dc in [(0, 1), (1, 0)]: + a = canvas_labels[:-dr if dr else None, :-dc if dc else None] + b = canvas_labels[dr:, dc:] + diff = a != b + for u, v in zip(a[diff].ravel().tolist(), b[diff].ravel().tolist()): + if u != v: + adj_pairs.add((min(u, v), max(u, v))) + + # Try many curves on this same partition — region setup (~0.5s) + # is far more expensive than a curve attempt (~25ms), so amortize. + seq_ok = False + curve = None; ids = None; runs = None; stable = None; stable_ids = None + for _seq_try in range(20): + start_region = rng.randrange(num_zones) + start = sample_deep_point( + rng, dts[start_region], + min_depth=min(40.0, max_depths[start_region] * 0.5), + ) + heading = rng.uniform(0.0, 2 * math.pi) + # Scale num_steps with canvas linear size so curve path length + # stays proportional to canvas diagonal as difficulty grows. + step_scale = max(1.0, min(width, height) / 1024.0) + curve = free_turtle( + rng, width, height, start, heading, + num_steps=int(rng.randint(800, 1400) * step_scale), + step_size=rng.uniform(2.8, 3.3), + max_turn=rng.uniform(0.042, 0.054), + drift_rate=rng.uniform(0.011, 0.016), + ) + if len(curve) < 60: + continue + + ids, runs, stable = extract_sequence(curve, canvas_labels, min_run=min_run) + stable_ids = [rid for rid, _, _ in stable] + if not (min_stable_seq_len <= len(stable_ids) <= max_stable_seq_len): + continue + if len(set(stable_ids)) < min_distinct_regions: + continue + adj_ok = True + for a, b in zip(stable_ids[:-1], stable_ids[1:]): + if a == b: + adj_ok = False; break + if (min(a, b), max(a, b)) not in adj_pairs: + adj_ok = False; break + if not adj_ok: + continue + + depth_ok = True + for rid, s, e in stable: + dt = dts[rid] + max_d = 0.0 + for k in range(s, e, 2): + v = _dt_lookup(dt, curve[k, 0], curve[k, 1]) + if v > max_d: + max_d = v + if max_d < min_run_depth: + depth_ok = False + break + if not depth_ok: + continue + + stable_points = sum(e - s for _, s, e in stable) + if stable_points < len(curve) * 0.96: + continue + + dt_boundary = boundary_distance_transform(canvas_labels) + if curve_grazes_boundary( + curve, dt_boundary, ids, + min_dist=10.0, window=8, sample_step=2, + ): + continue + + if not runs or (runs[0][2] - runs[0][1]) < min_run or runs[0][0] < 0: + continue + if (runs[-1][2] - runs[-1][1]) < min_run or runs[-1][0] < 0: + continue + if runs[0][0] != stable[0][0] or runs[-1][0] != stable[-1][0]: + continue + + crossings = find_self_crossings(curve, min_seg_gap=40) + if any(c[2] < min_cross_angle for c in crossings): + continue + if len(crossings) >= 2: + cps = np.array([[c[3], c[4]] for c in crossings]) + diffs = cps[:, None, :] - cps[None, :, :] + dmat = np.sqrt((diffs ** 2).sum(axis=-1)) + np.fill_diagonal(dmat, np.inf) + if dmat.min() < min_cross_gap: + continue + + seq_ok = True + break + if not seq_ok: + continue + + # Distractor curves (optional, scales with difficulty). Per reviewer + # feedback, each distractor MUST be generated with the same + # ``free_turtle`` path-generator as the query curve AND match the + # query curve's stable-region statistics for this sample: + # - same number of stable-region runs, + # - same (sorted) distribution of stable-run lengths (±tolerance), + # - same total arc length (same number of turtle steps). + # Distractors are rendered with the same stroke/colour as the query + # curve but WITHOUT the green S start marker. They must not hug the + # query curve (close-strand hygiene) and they do NOT contribute to + # the answer sequence. + distractor_curves: List[np.ndarray] = [] + if num_distractor_curves > 0: + main_stable_count = len(stable) + main_num_steps = len(curve) - 1 + # Decimate long curves before O(N²) pairwise crossing checks. + # Caps work at ~800² regardless of difficulty. Continuity is + # preserved (unlike a `step=k` skip), so all crossings remain + # detectable; only their pixel coords are slightly coarser, + # which is fine for both the angle test and the 30 px + # crossing-exclude radius used by `_distractor_too_close`. + dec = max(1, len(curve) // 800) + curve_dec = curve[::dec] if dec > 1 else curve + distractor_ok = True + for _di in range(int(num_distractor_curves)): + placed = False + for _dtry in range(10): + d_start_region = rng.randrange(num_zones) + if max_depths[d_start_region] < 45.0: + continue + d_start = sample_deep_point( + rng, dts[d_start_region], + min_depth=min(40.0, max_depths[d_start_region] * 0.5), + ) + d_heading = rng.uniform(0.0, 2 * math.pi) + d_curve = free_turtle( + rng, width, height, d_start, d_heading, + num_steps=main_num_steps, # identical arc length + step_size=rng.uniform(2.8, 3.3), + max_turn=rng.uniform(0.042, 0.054), + drift_rate=rng.uniform(0.011, 0.016), + ) + if len(d_curve) < 60: + continue + d_dec = d_curve[::dec] if dec > 1 else d_curve + MIN_DISTRACTOR_CROSS_ANGLE = 35.0 + sc_count, sc_worst, sc_pts = _find_pair_crossings( + d_dec, d_dec, same_curve=True) + if sc_worst < MIN_DISTRACTOR_CROSS_ANGLE: + continue + if _distractor_too_close(d_dec, d_dec, min_dist=14.0, + sample_step=3, same_curve=True, + known_crossings=sc_pts, + crossing_exclude_radius=30.0): + continue + pc_count, pc_worst, pc_pts = _find_pair_crossings( + d_dec, curve_dec) + if pc_worst < MIN_DISTRACTOR_CROSS_ANGLE: + continue + if _distractor_too_close(d_dec, curve_dec, min_dist=14.0, + sample_step=3, + known_crossings=pc_pts, + crossing_exclude_radius=30.0): + continue + hug_prev = False + for prev in distractor_curves: + prev_dec = prev[::dec] if dec > 1 else prev + pp_count, pp_worst, pp_pts = _find_pair_crossings( + d_dec, prev_dec) + if pp_worst < MIN_DISTRACTOR_CROSS_ANGLE: + hug_prev = True + break + if _distractor_too_close(d_dec, prev_dec, min_dist=14.0, + sample_step=3, + known_crossings=pp_pts, + crossing_exclude_radius=30.0): + hug_prev = True + break + if hug_prev: + continue + distractor_curves.append(d_curve) + placed = True + break + if not placed: + distractor_ok = False + break + if not distractor_ok: + continue + + # Label positions (deep in region, away from curve). + label_positions: List[Tuple[float, float]] = [] + label_scores: List[float] = [] + for rid in range(num_zones): + pos, score = find_label_position(dts[rid], curve, width, height) + label_positions.append(pos) + label_scores.append(score) + if min(label_scores) < min_label_score: + continue + + # Build record. + region_labels = list(string.ascii_uppercase[:num_zones]) + palette = list(REGION_COLORS) + rng.shuffle(palette) + region_colors = palette[:num_zones] + + answer_labels = [region_labels[rid] for rid in stable_ids] + answer = ", ".join(answer_labels) + + question = ( + f"The image shows {num_zones} colored irregular regions (no " + f"boundary lines are drawn — regions are distinguished only by " + f"their fill color). Each region carries a single-letter label " + f"({', '.join(region_labels)}). A single smooth black curve " + f"starts at the position marked with a 'Start' label and travels " + f"through several regions, potentially re-entering earlier " + f"ones. Follow the curve from start to end and list the region " + f"labels it visits in order, separated by commas (for example: " + f"A, C, B, A). " + f"Provide your final answer enclosed in ... tags." + ) + + return { + "width": width, + "height": height, + "num_zones": num_zones, + "region_labels": region_labels, + "region_colors": region_colors, + "grid_labels": grid_labels.tolist(), + "curve": curve, + "distractor_curves": distractor_curves, + "label_positions": [[float(x), float(y)] for x, y in label_positions], + "sequence_length": len(stable_ids), + "sequence": answer_labels, + "visit_ids": [int(x) for x in stable_ids], + "question": question, + "answer": answer, + } + + return None + + +# ── Rendering ───────────────────────────────────────────────────── + +def render_instance(out_path: Path, record: Dict, noise_seed: int) -> None: + w = int(record["width"]) + h = int(record["height"]) + grid_labels = np.array(record["grid_labels"], dtype=np.int32) + num_zones = int(record["num_zones"]) + region_colors: List[str] = record["region_colors"] + region_labels: List[str] = record["region_labels"] + label_positions = record["label_positions"] + curve = np.array(record["curve"]) if not isinstance(record["curve"], np.ndarray) else record["curve"] + + canvas_labels = upsample_labels(grid_labels, w, h) + + # Build RGB image from label map. + color_rgb = np.zeros((num_zones, 3), dtype=np.float32) + for i, cs in enumerate(region_colors): + color_rgb[i] = np.array([ + int(cs[1:3], 16), int(cs[3:5], 16), int(cs[5:7], 16) + ]) / 255.0 + img = color_rgb[canvas_labels] + + fig = plt.figure(figsize=(w / 100, h / 100), dpi=100) + ax = fig.add_axes([0, 0, 1, 1]) + ax.set_xlim(0, w) + ax.set_ylim(h, 0) + ax.axis("off") + ax.set_facecolor(BG_COLOR) + + ax.imshow(img, extent=(0, w, h, 0), zorder=0.5, interpolation="nearest") + + # Faint noise to avoid a too-flat aesthetic. + nrng = np.random.default_rng(noise_seed) + noise = nrng.normal(0.0, 1.0, size=(h, w)) + noise = (noise - noise.min()) / max(noise.max() - noise.min(), 1e-6) + ax.imshow(noise, cmap="Greys", alpha=0.035, extent=(0, w, h, 0), + zorder=1.0, interpolation="bilinear") + + # Distractor curves (drawn first, underneath the query curve but above + # the background). Same colour and stroke as the query curve so they + # are visually indistinguishable at a glance — the dot marks which + # curve is the query curve. + distractor_curves = record.get("distractor_curves") or [] + for dc in distractor_curves: + dc_arr = np.array(dc) if not isinstance(dc, np.ndarray) else dc + if len(dc_arr) < 2: + continue + ax.plot(dc_arr[:, 0], dc_arr[:, 1], color=LINE_COLOR, linewidth=3.4, + solid_capstyle="round", solid_joinstyle="round", zorder=2.8) + + # Black curve. + ax.plot(curve[:, 0], curve[:, 1], color=LINE_COLOR, linewidth=3.4, + solid_capstyle="round", solid_joinstyle="round", zorder=3.0) + + # Start marker: a small "Start" text label in a white box with black border, + # placed right at the curve's first point. + sx, sy = curve[0] + ax.text( + sx, sy, "Start", + fontsize=14, fontweight="bold", color="#111111", + ha="center", va="center", zorder=4.0, + bbox=dict(boxstyle="round,pad=0.28", facecolor="white", + edgecolor="#111111", linewidth=1.6, alpha=0.95), + ) + + # Region labels. + for rid in range(num_zones): + lx, ly = label_positions[rid] + ax.text( + lx, ly, region_labels[rid], + fontsize=24, fontweight="bold", color="#111111", + ha="center", va="center", zorder=5.0, + bbox=dict(boxstyle="round,pad=0.28", facecolor="white", + edgecolor="#555", linewidth=1.2, alpha=0.88), + ) + + fig.savefig(out_path, dpi=100, bbox_inches="tight", pad_inches=0) + plt.close(fig) + + +# ── Main ───────────────────────────────────────────────────────── + +def main() -> None: + parser = argparse.ArgumentParser() + parser.add_argument("--output-root", type=Path, required=True) + parser.add_argument("--count", type=int, default=20) + parser.add_argument("--seed", type=int, default=42) + parser.add_argument("--width", type=int, default=1024) + parser.add_argument("--height", type=int, default=1024) + parser.add_argument("--difficulty", type=int, default=5, + help="Integer difficulty >=0; scales zones, seq " + "lengths, distinct regions, and adds distractor curves.") + parser.add_argument("--workers", type=int, default=8) + args = parser.parse_args() + + out = args.output_root + img_dir = out / "images" + img_dir.mkdir(parents=True, exist_ok=True) + + rng = random.Random(args.seed) + records: List[Dict] = [] + + d = max(0, int(args.difficulty)) + num_zones = 5 + d + min_stable_seq_len = 6 + d + max_stable_seq_len = 8 + 2 * d + min_distinct_regions = num_zones // 2 + 1 + num_distractor_curves = d // 2 + + # Canvas scaling: grow with zones + distractor curves proxy so that + # absolute-pixel constants (stroke width, S marker radius, label font) + # stay legible as content density increases. + N_d = 5 + int(1.5 * d) + N_0 = 5 + s = math.sqrt(max(1.0, N_d / N_0)) + args.width = int(round(args.width * s)) + args.height = int(round(args.height * s)) + + sys.path.insert(0, str(Path(__file__).resolve().parents[3])) + from _sample_pool import parallel_sample_records_persistent # noqa: E402 + + def _attempt(rng): + return sample_instance( + rng, args.width, args.height, + min_zones=num_zones, max_zones=num_zones, + min_stable_seq_len=min_stable_seq_len, + max_stable_seq_len=max_stable_seq_len, + min_distinct_regions=min_distinct_regions, + num_distractor_curves=num_distractor_curves, + max_attempts=1, + ) + + records_raw = parallel_sample_records_persistent( + _attempt, count=args.count, workers=args.workers, + seed_base=args.seed, + ) + rng_render = random.Random(args.seed ^ 0xA5A5) + for idx, rec in enumerate(records_raw): + name = f"color_zone_{idx:05d}.png" + ns = rng_render.randint(0, 10 ** 9) + render_instance(img_dir / name, rec, noise_seed=ns) + rec.pop("curve") + rec.pop("grid_labels") + rec.pop("distractor_curves", None) + rec["image"] = f"images/{name}" + records.append(rec) + with (out / "annotations.jsonl").open("w") as fh: + for r in records: + fh.write(json.dumps(r) + "\n") + print(f" {len(records)}/{args.count} valid samples (workers={args.workers})") + + data_json = { + "task": "color_zone_sequence", + "category": "sequential_traversal", + "count": len(records), + "items": [ + {"image": r["image"], "question": r["question"], "answer": r["answer"]} + for r in records + ], + } + (out / "data.json").write_text(json.dumps(data_json, indent=2)) + print(f"Saved {len(records)} items to {out}") + + +if __name__ == "__main__": + main() diff --git a/code/sequential_traversal/color_zone_sequence/data.json b/code/sequential_traversal/color_zone_sequence/data.json new file mode 100644 index 0000000000000000000000000000000000000000..53393750eb962565bbf5d39bba343c7c54212724 --- /dev/null +++ b/code/sequential_traversal/color_zone_sequence/data.json @@ -0,0 +1,32 @@ +{ + "task": "color_zone_sequence", + "category": "sequential_traversal", + "count": 5, + "items": [ + { + "image": "images/color_zone_00000.png", + "question": "The image shows several distinct regions, each carrying a single-letter label and distinguished by its own fill (colour, texture, or material). A single smooth curve starts at the position marked 'Start' and travels through several regions, potentially re-entering earlier ones. Follow the curve from start to end and list the region labels it visits in order, separated by commas. Provide your final answer enclosed in ... tags.", + "answer": "H, D, F, B, I, A, C, J, A, J, H" + }, + { + "image": "images/color_zone_00001.png", + "question": "The image shows several distinct regions, each carrying a single-letter label and distinguished by its own fill (colour, texture, or material). A single smooth curve starts at the position marked 'Start' and travels through several regions, potentially re-entering earlier ones. Follow the curve from start to end and list the region labels it visits in order, separated by commas. Provide your final answer enclosed in ... tags.", + "answer": "B, D, C, G, C, E, I, A, J, F, C" + }, + { + "image": "images/color_zone_00002.png", + "question": "The image shows several distinct regions, each carrying a single-letter label and distinguished by its own fill (colour, texture, or material). A single smooth curve starts at the position marked 'Start' and travels through several regions, potentially re-entering earlier ones. Follow the curve from start to end and list the region labels it visits in order, separated by commas. Provide your final answer enclosed in ... tags.", + "answer": "F, E, F, I, C, D, A, J, H, G, E" + }, + { + "image": "images/color_zone_00003.png", + "question": "The image shows several distinct regions, each carrying a single-letter label and distinguished by its own fill (colour, texture, or material). A single smooth curve starts at the position marked 'Start' and travels through several regions, potentially re-entering earlier ones. Follow the curve from start to end and list the region labels it visits in order, separated by commas. Provide your final answer enclosed in ... tags.", + "answer": "G, B, D, G, A, I, C, I, A, G, D, F" + }, + { + "image": "images/color_zone_00004.png", + "question": "The image shows several distinct regions, each carrying a single-letter label and distinguished by its own fill (colour, texture, or material). A single smooth curve starts at the position marked 'Start' and travels through several regions, potentially re-entering earlier ones. Follow the curve from start to end and list the region labels it visits in order, separated by commas. Provide your final answer enclosed in ... tags.", + "answer": "C, A, G, A, F, E, D, J, B, C, H, I, G" + } + ] +} \ No newline at end of file diff --git a/code/sequential_traversal/line_intersections/creation.py b/code/sequential_traversal/line_intersections/creation.py new file mode 100644 index 0000000000000000000000000000000000000000..50c65ad2aed9d18e8b9eb88cb47ba4e639792e24 --- /dev/null +++ b/code/sequential_traversal/line_intersections/creation.py @@ -0,0 +1,782 @@ +"""Line Intersections (turtle-based, anchored endpoints). + +Each curve has two preset endpoints placed on the perimeter of the canvas. +A straight "leader" segment connects each endpoint to an interior point +(~150px inward), keeping the perimeter band free of crossings. The middle +body of the curve is a turtle-style walk through waypoints distributed +across a generous interior zone — tangled, but not concentrated at the +very center. + +Only one of the two endpoints of each curve is labeled (A, B, C, ...). +The model starts tracing from the labeled endpoint and follows the +queried curve through the interior tangle. At every crossing it passes +through, it records the label of the OTHER curve involved (for a +self-crossing, that label is the queried curve's own letter). The +answer is the ordered sequence of labels. +""" +from __future__ import annotations + +import argparse +import json +import math +import os +import random +import string +import sys +from collections import defaultdict +from pathlib import Path +from typing import Dict, List, Tuple + +import matplotlib +matplotlib.use("Agg") +import matplotlib.pyplot as plt +import numpy as np +from tqdm import tqdm + +LINE_COLOR = "#2f2f2f" +LABEL_COLORS = [ + "#c23030", + "#1a6dba", + "#2d8e2d", + "#c47a18", + "#7a35a0", +] + + +# ── Perimeter anchors ────────────────────────────────────────────── + +def _perimeter_point(s: float, width: int, height: int, margin: int + ) -> Tuple[np.ndarray, np.ndarray]: + """Map s ∈ [0, 4) to (position, inward_unit_vector) on rect perimeter.""" + side = int(s) % 4 + t = s - int(s) + if side == 0: # top edge, left -> right + x = margin + (width - 2 * margin) * t + y = margin + inward = np.array([0.0, 1.0]) + elif side == 1: # right edge, top -> bottom + x = width - margin + y = margin + (height - 2 * margin) * t + inward = np.array([-1.0, 0.0]) + elif side == 2: # bottom edge, right -> left + x = width - margin - (width - 2 * margin) * t + y = height - margin + inward = np.array([0.0, -1.0]) + else: # left edge, bottom -> top + x = margin + y = height - margin - (height - 2 * margin) * t + inward = np.array([1.0, 0.0]) + return np.array([x, y]), inward + + +def sample_perimeter_anchors( + rng: random.Random, + num_anchors: int, + width: int, + height: int, + margin: int, + min_gap: float | None = None, + jitter_deg: float = 10.0, + max_attempts: int = 300, +) -> List[Tuple[np.ndarray, np.ndarray]]: + """Sample `num_anchors` positions around the canvas perimeter, enforcing + a minimum gap (in perimeter units, 1 unit = one side).""" + if min_gap is None: + # 50% of perfectly even spacing + min_gap = (4.0 / num_anchors) * 0.5 + for _ in range(max_attempts): + positions = sorted(rng.uniform(0.0, 4.0) for _ in range(num_anchors)) + ok = True + for i in range(num_anchors): + gap = (positions[(i + 1) % num_anchors] - positions[i]) % 4.0 + if gap < min_gap: + ok = False + break + if ok: + anchors = [] + for s in positions: + pt, inward = _perimeter_point(s, width, height, margin) + # small angular jitter of the inward direction + ang = math.atan2(inward[1], inward[0]) + ang += math.radians(rng.uniform(-jitter_deg, jitter_deg)) + inward = np.array([math.cos(ang), math.sin(ang)]) + anchors.append((pt, inward)) + return anchors + return None + + +# ── Turtle body with anchored start/end ──────────────────────────── + +def _angle_diff(current: float, target: float) -> float: + d = target - current + return (d + math.pi) % (2 * math.pi) - math.pi + + +def build_anchored_curve( + rng: random.Random, + width: int, + height: int, + start_pos: np.ndarray, + start_inward: np.ndarray, + end_pos: np.ndarray, + end_inward: np.ndarray, + leader_length: float = 150.0, + step_size: float = 3.3, + max_turn: float = 0.028, + drift_rate: float = 0.007, + num_waypoints: int = 2, + interior_margin: int = 260, + max_steps: int = 1800, +) -> np.ndarray: + """Single-turtle curve from start_pos to a point near end_pos. + + Phase-dependent turning: + - Entry leader (first `leader_length` of path): max_turn = 0 (straight). + - Body (interior): waypoint steering within max_turn clamp. + - Exit leader (within `leader_length` of end_pos perimeter plane): + head-alignment only (no waypoint steering), rotating toward + -end_inward at up to max_turn/step. Then drives straight. + + Because the entire curve is a single turtle walk, there are no splice + discontinuities. The curve terminates when it crosses the perimeter plane + at end_pos; the final point is wherever it actually crosses (close to + end_pos when head-alignment locks in time). + """ + pts: List[np.ndarray] = [start_pos.copy()] + x, y = float(start_pos[0]), float(start_pos[1]) + theta = math.atan2(start_inward[1], start_inward[0]) + + # Waypoints in interior, ordered by progress from start→end. + raw_wps: List[np.ndarray] = [] + for _ in range(num_waypoints): + wx = rng.uniform(interior_margin, width - interior_margin) + wy = rng.uniform(interior_margin, height - interior_margin) + raw_wps.append(np.array([wx, wy])) + direction = end_pos - start_pos + dir_norm = direction / (np.linalg.norm(direction) + 1e-9) + raw_wps.sort(key=lambda p: float(np.dot(p - start_pos, dir_norm))) + + # Funnel waypoint deep on the exit axis pulls the turtle onto the axis + # before the exit leader. Virtual lookahead past the exit then keeps + # heading aligned with -end_inward as we approach end_pos. + funnel = end_pos + end_inward * 400.0 + lookahead = end_pos + (-end_inward) * 240.0 + waypoints: List[np.ndarray] = raw_wps + [funnel, lookahead] + + exit_theta = math.atan2(-end_inward[1], -end_inward[0]) + entry_steps = max(2, int(leader_length / step_size)) + + wp_idx = 0 + drift = 0.0 + + # Perp to end_inward (for lateral-offset measurement along perimeter edge). + perp = np.array([-end_inward[1], end_inward[0]]) + + for step_count in range(max_steps): + # Signed perpendicular distance from turtle to end_pos perimeter plane. + # end_inward points from perimeter INTO canvas, so on interior side + # dot(pos - end_pos, end_inward) > 0; crossing the plane → <= 0. + sd = ((x - end_pos[0]) * end_inward[0] + + (y - end_pos[1]) * end_inward[1]) + # Lateral offset along the perimeter edge from end_pos. + lat = abs((x - end_pos[0]) * perp[0] + (y - end_pos[1]) * perp[1]) + + # Termination: turtle has reached/crossed the perimeter. + if sd <= 0.0: + break + + if step_count < entry_steps: + # Entry leader: no turn (straight). + turn = 0.0 + elif sd < leader_length and lat < 40.0: + # Exit leader: pure head-alignment, no positional steering. + head_err = _angle_diff(theta, exit_theta) + turn = max(-max_turn, min(max_turn, head_err)) + else: + # Body: waypoint steering. + is_final = (wp_idx == len(waypoints) - 1) + if is_final: + wx, wy = waypoints[-1] + pull = 0.18 + else: + wx, wy = waypoints[wp_idx] + steer_dist = math.hypot(wx - x, wy - y) + # reach must exceed turning diameter (2 * step / max_turn) + if steer_dist < 140.0: + wp_idx += 1 + continue + pull = 0.09 + + drift += rng.gauss(0, drift_rate) + drift = max(-max_turn * 0.7, min(max_turn * 0.7, drift)) + turn = drift + rng.gauss(0, max_turn * 0.12) + target_ang = math.atan2(wy - y, wx - x) + steer = _angle_diff(theta, target_ang) + turn += steer * pull + turn = max(-max_turn, min(max_turn, turn)) + + theta += turn + x += step_size * math.cos(theta) + y += step_size * math.sin(theta) + pts.append(np.array([x, y])) + + return np.array(pts) + + +# ── Intersection detection ───────────────────────────────────────── + +def _segments_cross(p0, p1, q0, q1) -> bool: + eps = 1e-8 + o1 = (p1[0]-p0[0])*(q0[1]-p0[1]) - (p1[1]-p0[1])*(q0[0]-p0[0]) + o2 = (p1[0]-p0[0])*(q1[1]-p0[1]) - (p1[1]-p0[1])*(q1[0]-p0[0]) + o3 = (q1[0]-q0[0])*(p0[1]-q0[1]) - (q1[1]-q0[1])*(p0[0]-q0[0]) + o4 = (q1[0]-q0[0])*(p1[1]-q0[1]) - (q1[1]-q0[1])*(p1[0]-q0[0]) + return ((o1 > eps and o2 < -eps) or (o1 < -eps and o2 > eps)) and \ + ((o3 > eps and o4 < -eps) or (o3 < -eps and o4 > eps)) + + +def _crossing_angle(p0, p1, q0, q1) -> float: + d1 = p1 - p0 + d2 = q1 - q0 + cos_a = np.dot(d1, d2) / (np.linalg.norm(d1) * np.linalg.norm(d2) + 1e-12) + return math.degrees(math.acos(min(1.0, abs(cos_a)))) + + +def _find_crossings( + poly_a: np.ndarray, + poly_b: np.ndarray, + same_curve: bool = False, + min_seg_gap: int = 10, +): + """Returns (count, worst_angle, details). + + `details` is a list of dicts, one per crossing, with keys: + i, j — segment indices on poly_a and poly_b + ti, tj — crossing parameters ∈ (0, 1) along each segment + px, py — exact crossing point + angle — crossing angle in degrees + For same_curve=True, i < j is enforced so each crossing appears once. + """ + na = len(poly_a) - 1 + nb = len(poly_b) - 1 + + a_min_x = np.minimum(poly_a[:-1, 0], poly_a[1:, 0]) + a_max_x = np.maximum(poly_a[:-1, 0], poly_a[1:, 0]) + a_min_y = np.minimum(poly_a[:-1, 1], poly_a[1:, 1]) + a_max_y = np.maximum(poly_a[:-1, 1], poly_a[1:, 1]) + + b_min_x = np.minimum(poly_b[:-1, 0], poly_b[1:, 0]) + b_max_x = np.maximum(poly_b[:-1, 0], poly_b[1:, 0]) + b_min_y = np.minimum(poly_b[:-1, 1], poly_b[1:, 1]) + b_max_y = np.maximum(poly_b[:-1, 1], poly_b[1:, 1]) + + all_min_x = np.concatenate([a_min_x, b_min_x]) + all_max_x = np.concatenate([a_max_x, b_max_x]) + cell_size = max(np.median(all_max_x - all_min_x), 1.0) * 3 + + grid_a = defaultdict(list) + for i in range(na): + cx0 = int(a_min_x[i] / cell_size); cx1 = int(a_max_x[i] / cell_size) + cy0 = int(a_min_y[i] / cell_size); cy1 = int(a_max_y[i] / cell_size) + for gx in range(cx0, cx1 + 1): + for gy in range(cy0, cy1 + 1): + grid_a[(gx, gy)].append(i) + + grid_b = defaultdict(list) + for j in range(nb): + cx0 = int(b_min_x[j] / cell_size); cx1 = int(b_max_x[j] / cell_size) + cy0 = int(b_min_y[j] / cell_size); cy1 = int(b_max_y[j] / cell_size) + for gx in range(cx0, cx1 + 1): + for gy in range(cy0, cy1 + 1): + grid_b[(gx, gy)].append(j) + + checked = set() + worst_angle = 180.0 + details: List[Dict] = [] + for cell_key in grid_a: + if cell_key not in grid_b: + continue + for i in grid_a[cell_key]: + for j in grid_b[cell_key]: + if same_curve: + ii, jj = min(i, j), max(i, j) + if jj - ii < min_seg_gap: + continue + key = (ii, jj) + else: + key = (i, j) + if key in checked: + continue + checked.add(key) + + if a_max_x[i] < b_min_x[j] or b_max_x[j] < a_min_x[i] or \ + a_max_y[i] < b_min_y[j] or b_max_y[j] < a_min_y[i]: + continue + + if same_curve: + si, sj = key # enforce i < j for self-crossings + p0, p1 = poly_a[si], poly_a[si + 1] + q0, q1 = poly_b[sj], poly_b[sj + 1] + else: + si, sj = i, j + p0, p1 = poly_a[si], poly_a[si + 1] + q0, q1 = poly_b[sj], poly_b[sj + 1] + if not _segments_cross(p0, p1, q0, q1): + continue + d1 = p1 - p0 + d2 = q1 - q0 + denom = d1[0] * d2[1] - d1[1] * d2[0] + if abs(denom) < 1e-12: + continue + ti = ((q0[0] - p0[0]) * d2[1] - (q0[1] - p0[1]) * d2[0]) / denom + tj = ((q0[0] - p0[0]) * d1[1] - (q0[1] - p0[1]) * d1[0]) / denom + px = float(p0[0] + ti * d1[0]) + py = float(p0[1] + ti * d1[1]) + angle = _crossing_angle(p0, p1, q0, q1) + worst_angle = min(worst_angle, angle) + details.append({ + "i": int(si), "j": int(sj), + "ti": float(ti), "tj": float(tj), + "px": px, "py": py, "angle": float(angle), + }) + return len(details), worst_angle, details + + +def _details_to_points(details: List[Dict]) -> np.ndarray: + if not details: + return np.zeros((0, 2)) + return np.array([[d["px"], d["py"]] for d in details]) + + +def _curves_too_close( + poly_a: np.ndarray, + poly_b: np.ndarray, + same_curve: bool = False, + min_dist: float = 7.0, + sample_step: int = 3, + self_index_gap: int = 50, + known_crossings: np.ndarray | None = None, + crossing_exclude_radius: float = 30.0, +) -> bool: + """Flag if ANY sample point on A comes within `min_dist` of B without + being attributable to a real crossing. Samples within + `crossing_exclude_radius` of any point in `known_crossings` are ignored + (that's the crossing region itself, where near-zero distance is expected). + """ + nb = len(poly_b) - 1 + if nb <= 0 or len(poly_a) == 0: + return False + b_pts = poly_b[:-1] + b_vecs = poly_b[1:] - poly_b[:-1] + b_lens_sq = np.maximum((b_vecs ** 2).sum(axis=1), 1e-12) + + has_crossings = known_crossings is not None and len(known_crossings) > 0 + + for idx in range(0, len(poly_a), sample_step): + px, py = poly_a[idx] + p = np.array([px, py]) + dp = p - b_pts + t = (dp * b_vecs).sum(axis=1) / b_lens_sq + t = np.clip(t, 0.0, 1.0) + proj = b_pts + t[:, None] * b_vecs + dists = np.sqrt(((p - proj) ** 2).sum(axis=1)) + if same_curve: + mask = np.abs(np.arange(nb) - idx) < self_index_gap + dists[mask] = 9999.0 + if dists.min() < min_dist: + if has_crossings: + cross_dists = np.sqrt(((p - known_crossings) ** 2).sum(axis=1)) + if cross_dists.min() < crossing_exclude_radius: + continue # close approach explained by a real crossing + return True + return False + + +# ── Instance sampling ────────────────────────────────────────────── + +def sample_instance( + rng: random.Random, + width: int, + height: int, + min_lines: int = 4, + max_lines: int = 4, + min_total_crossings: int = 12, + max_total_crossings: int = 60, + perimeter_margin: int = 70, + leader_length: float = 150.0, + interior_margin: int = 260, + max_attempts: int = 2500, +) -> Dict | None: + labels = list(string.ascii_uppercase) + + for _ in range(max_attempts): + num_lines = rng.randint(min_lines, max_lines) + + anchors = sample_perimeter_anchors( + rng, num_lines * 2, width, height, + margin=perimeter_margin, + ) + if anchors is None: + continue + + # Pair anchor i with anchor i+num_lines: opposite-side pairing to + # force curves through the interior. + pairs = [(anchors[i], anchors[i + num_lines]) for i in range(num_lines)] + rng.shuffle(pairs) + + polylines = [] + for (sp, sd), (ep, ed) in pairs: + poly = build_anchored_curve( + rng, width, height, + start_pos=sp, start_inward=sd, + end_pos=ep, end_inward=ed, + leader_length=leader_length, + step_size=rng.uniform(3.0, 3.8), + max_turn=rng.uniform(0.034, 0.044), + drift_rate=rng.uniform(0.010, 0.016), + num_waypoints=rng.randint(4, 6), + interior_margin=interior_margin, + ) + polylines.append(poly) + + total = 0 + global_min_angle = 180.0 + self_details: Dict[int, List[Dict]] = {} + pair_details: Dict[Tuple[int, int], List[Dict]] = {} + + for a in range(num_lines): + sc, ang, det = _find_crossings(polylines[a], polylines[a], + same_curve=True) + self_details[a] = det + total += sc + global_min_angle = min(global_min_angle, ang) + for b in range(a + 1, num_lines): + cc, ang, det = _find_crossings(polylines[a], polylines[b]) + pair_details[(a, b)] = det + total += cc + global_min_angle = min(global_min_angle, ang) + + if total < min_total_crossings or total > max_total_crossings: + continue + if global_min_angle < 45.0: + continue + + # 1% of the canvas's short side keeps curves visibly separated + # at any resolution while still letting the sampler converge at + # higher difficulties. + min_dist_px = 0.01 * float(min(width, height)) + too_close = False + for a in range(num_lines): + if _curves_too_close(polylines[a], polylines[a], same_curve=True, + min_dist=min_dist_px, + known_crossings=_details_to_points(self_details[a])): + too_close = True; break + for b in range(a + 1, num_lines): + if _curves_too_close(polylines[a], polylines[b], + min_dist=min_dist_px, + known_crossings=_details_to_points(pair_details[(a, b)])): + too_close = True; break + if too_close: + break + if too_close: + continue + + # Reject if any two crossing points are too close to each other — + # that would let multiple intersections be mistaken for one. + all_cross_pts = [] + for a in range(num_lines): + pts = _details_to_points(self_details[a]) + if len(pts) > 0: + all_cross_pts.append(pts) + for b in range(a + 1, num_lines): + pts = _details_to_points(pair_details[(a, b)]) + if len(pts) > 0: + all_cross_pts.append(pts) + if all_cross_pts: + stacked = np.concatenate(all_cross_pts, axis=0) + if len(stacked) >= 2: + diffs = stacked[:, None, :] - stacked[None, :, :] + dmat = np.sqrt((diffs ** 2).sum(axis=-1)) + np.fill_diagonal(dmat, np.inf) + if dmat.min() < 20.0: + continue + + # Pick label positions: one of the two endpoints per curve. + # "label_at_start" picks the first endpoint (index 0) when True. + label_starts = [rng.random() < 0.5 for _ in range(num_lines)] + + # Per-curve sequence length: each self-crossing is hit twice during + # the trace (entering + leaving), each pair-crossing once. + seq_lens = [] + for q in range(num_lines): + sl = 2 * len(self_details[q]) + for b in range(num_lines): + if b == q: + continue + key = (min(q, b), max(q, b)) + sl += len(pair_details[key]) + seq_lens.append(sl) + + # Prefer a query line whose sequence length is near ~10. + TARGET = 10 + candidates = sorted( + range(num_lines), key=lambda i: abs(seq_lens[i] - TARGET) + ) + query_line = candidates[0] + seq_len = seq_lens[query_line] + if seq_len < 8 or seq_len > 13: + continue + + query_label = labels[query_line] + line_labels = labels[:num_lines] + + # Build the ordered crossing sequence along the query curve. + query_poly = polylines[query_line] + qdiffs = np.diff(query_poly, axis=0) + qseg_lens = np.sqrt((qdiffs ** 2).sum(axis=1)) + qcum = np.concatenate([[0.0], np.cumsum(qseg_lens)]) + + entries: List[Tuple[float, str]] = [] + # Self-crossings contribute two entries (both passes), labeled Q. + for d in self_details[query_line]: + pos_a = qcum[d["i"]] + d["ti"] * qseg_lens[d["i"]] + pos_b = qcum[d["j"]] + d["tj"] * qseg_lens[d["j"]] + entries.append((pos_a, query_label)) + entries.append((pos_b, query_label)) + # Pair-crossings contribute one entry, labeled with the other curve. + for other in range(num_lines): + if other == query_line: + continue + if query_line < other: + dets = pair_details[(query_line, other)] + for d in dets: + pos = qcum[d["i"]] + d["ti"] * qseg_lens[d["i"]] + entries.append((pos, labels[other])) + else: + dets = pair_details[(other, query_line)] + for d in dets: + # query curve is poly_b in this pair → use j, tj + pos = qcum[d["j"]] + d["tj"] * qseg_lens[d["j"]] + entries.append((pos, labels[other])) + + # Orientation: labeled endpoint at poly[0] → forward arc order; + # labeled endpoint at poly[-1] → reverse arc order. + trace_forward = label_starts[query_line] + entries.sort(key=lambda e: e[0] if trace_forward else -e[0]) + answer_sequence = [label for _, label in entries] + answer = ", ".join(answer_sequence) + + question = ( + f"The image shows {num_lines} curves. Each curve has one labeled " + f"endpoint on the perimeter (labels: {', '.join(line_labels)}). " + f"Starting from the labeled endpoint of curve {query_label}, follow " + f"curve {query_label} through the interior tangle. At every " + f"crossing you pass through, record the label of the OTHER curve " + f"involved — when curve {query_label} crosses itself, record " + f"{query_label}. Report the ordered sequence of labels, separated " + f"by commas (for example: A, C, B, C, A). " + f"Provide your final answer enclosed in ... tags." + ) + + return { + "width": width, + "height": height, + "num_lines": num_lines, + "query_line": query_line, + "query_label": query_label, + "line_labels": line_labels, + "total_crossings": total, + "sequence_length": seq_len, + "polylines": polylines, + "label_starts": label_starts, + "question": question, + "answer": answer, + } + return None + + +# ── Rendering ────────────────────────────────────────────────────── + +def label_anchor_for_endpoint( + polyline: np.ndarray, + at_start: bool, + width: int, + height: int, + offset: float = 46.0, + tangent_span: int = 10, +) -> Tuple[float, float]: + """Offset a label perpendicular to the leader direction at a labeled + curve endpoint. Prefer the side that keeps the label inside the canvas + and farther from the polyline.""" + if at_start: + pt = polyline[0] + tan = polyline[min(tangent_span, len(polyline) - 1)] - polyline[0] + else: + pt = polyline[-1] + tan = polyline[-1] - polyline[max(-tangent_span - 1, -len(polyline))] + tnorm = math.hypot(float(tan[0]), float(tan[1])) + if tnorm < 1e-8: + ux, uy = 0.0, -1.0 + else: + ux = float(tan[0]) / tnorm + uy = float(tan[1]) / tnorm + perps = [(-uy, ux), (uy, -ux)] + best = None + best_score = -1e9 + for px, py in perps: + cx = float(pt[0]) + px * offset + cy = float(pt[1]) + py * offset + score = 0.0 + if 30 < cx < width - 30 and 30 < cy < height - 30: + score += 200.0 + diffs = polyline - np.array([cx, cy]) + dists = np.sqrt((diffs ** 2).sum(axis=1)) + score += float(dists.min()) + if score > best_score: + best_score = score + best = (cx, cy) + cx, cy = best + return ( + float(np.clip(cx, 28, width - 28)), + float(np.clip(cy, 28, height - 28)), + ) + + +def render_instance(out_path: Path, record: Dict, noise_seed: int) -> None: + width = int(record["width"]) + height = int(record["height"]) + polylines = record["polylines"] + num_lines = record["num_lines"] + line_labels = record["line_labels"] + label_starts = record["label_starts"] + + fig = plt.figure(figsize=(width / 100, height / 100), dpi=100) + ax = fig.add_axes([0, 0, 1, 1]) + ax.set_xlim(0, width) + ax.set_ylim(height, 0) + ax.axis("off") + ax.set_facecolor("#f3efe8") + + nrng = np.random.default_rng(noise_seed) + noise = nrng.normal(0.0, 1.0, size=(height, width)) + noise = (noise - noise.min()) / max(noise.max() - noise.min(), 1e-6) + ax.imshow(noise, cmap="Greys", alpha=0.06, extent=(0, width, height, 0), + interpolation="bilinear") + + for i in range(num_lines): + poly = polylines[i] + ax.plot( + poly[:, 0], poly[:, 1], + color=LINE_COLOR, linewidth=2.8, + solid_capstyle="round", solid_joinstyle="round", + zorder=2.0 + i * 0.05, + ) + + for i in range(num_lines): + poly = polylines[i] + labeled_pt = poly[0] if label_starts[i] else poly[-1] + lc = LABEL_COLORS[i % len(LABEL_COLORS)] + lx, ly = label_anchor_for_endpoint( + poly, at_start=label_starts[i], width=width, height=height, + ) + # Dashed connector from endpoint to offset label + ax.plot([labeled_pt[0], lx], [labeled_pt[1], ly], + color="#9b7b56", linewidth=1.0, alpha=0.7, + linestyle=(0, (2.0, 3.2)), zorder=3.8) + # Dot at the actual endpoint + ax.scatter([labeled_pt[0]], [labeled_pt[1]], s=110, + facecolors=lc, edgecolors="white", linewidths=1.8, zorder=4.0) + # Label at offset position + ax.text(lx, ly, line_labels[i], + fontsize=17, fontweight="bold", color=lc, + ha="center", va="center", zorder=5.0, + bbox=dict(boxstyle="round,pad=0.2", facecolor="#f3efe8", + edgecolor=lc, alpha=0.95, linewidth=1.2)) + + fig.savefig(out_path, dpi=100, bbox_inches="tight", pad_inches=0) + plt.close(fig) + + +# ── Main ─────────────────────────────────────────────────────────── + +def main(): + parser = argparse.ArgumentParser() + parser.add_argument("--output-root", type=Path, required=True) + parser.add_argument("--count", type=int, default=20) + parser.add_argument("--seed", type=int, default=42) + parser.add_argument("--width", type=int, default=1024) + parser.add_argument("--height", type=int, default=1024) + parser.add_argument("--min-lines", type=int, default=3) + parser.add_argument("--max-lines", type=int, default=5) + parser.add_argument("--min-crossings", type=int, default=3) + parser.add_argument("--max-crossings", type=int, default=25) + parser.add_argument("--difficulty", type=int, default=5, + help="Integer difficulty >=0; scales line and crossing counts.") + parser.add_argument("--workers", type=int, default=8) + args = parser.parse_args() + + d = max(0, int(args.difficulty)) + # num_lines grows as 3 + d//3 (no cap). num_crossings ∈ [3, 5 + 2d]. + # Fixed 45° minimum crossing angle (handled inside sample_instance). + args.min_lines = 3 + d // 3 + args.max_lines = 3 + d // 3 + args.min_crossings = 10 + args.max_crossings = 10 + 2 * d + + # Canvas scaling: area ∝ N_d / N_0 with N_d = 5 + 2d. + N_d = 5 + 2 * d + N_0 = 5 + s = math.sqrt(max(1.0, N_d / N_0)) + args.width = int(round(args.width * s)) + args.height = int(round(args.height * s)) + + out_root = args.output_root + img_dir = out_root / "images" + img_dir.mkdir(parents=True, exist_ok=True) + + sys.path.insert(0, str(Path(__file__).resolve().parents[3])) + from _sample_pool import parallel_sample_records # noqa: E402 + + def _attempt(rng): + return sample_instance( + rng, args.width, args.height, + min_lines=args.min_lines, max_lines=args.max_lines, + min_total_crossings=args.min_crossings, + max_total_crossings=args.max_crossings, + max_attempts=50, + ) + + records_raw = parallel_sample_records( + _attempt, count=args.count, workers=args.workers, + seed_base=args.seed, + ) + rng_render = random.Random(args.seed ^ 0xA5A5) + records = [] + for idx, record in enumerate(records_raw): + name = f"line_intersections_{idx:05d}.png" + ns = rng_render.randint(0, 10**9) + render_instance(img_dir / name, record, noise_seed=ns) + record.pop("polylines") + record["image"] = f"images/{name}" + records.append(record) + print(f" {len(records)}/{args.count} valid samples (workers={args.workers})") + + with (out_root / "annotations.jsonl").open("w") as fh: + for r in records: + fh.write(json.dumps(r) + "\n") + + data_json = { + "task": "line_intersections", + "category": "sequential_traversal", + "count": len(records), + "items": [ + {"image": r["image"], "question": r["question"], "answer": r["answer"]} + for r in records + ], + } + (out_root / "data.json").write_text(json.dumps(data_json, indent=2)) + print(f"Saved {len(records)} items to {out_root}") + + +if __name__ == "__main__": + main() diff --git a/code/sequential_traversal/line_intersections/data.json b/code/sequential_traversal/line_intersections/data.json new file mode 100644 index 0000000000000000000000000000000000000000..ed5915e1baa4e7406feaa3de2ffcb46ee29c3843 --- /dev/null +++ b/code/sequential_traversal/line_intersections/data.json @@ -0,0 +1,32 @@ +{ + "task": "line_intersections", + "category": "sequential_traversal", + "count": 5, + "items": [ + { + "image": "images/line_intersections_00000.png", + "question": "The image shows several curves or strands tangled together. Each curve has a single labeled endpoint on the perimeter, marked with a distinct letter. Starting from the endpoint labeled A, follow that curve through the interior tangle. At every crossing you pass through, record the label of the OTHER curve involved \u2014 when curve A crosses itself, record A. Report the ordered sequence of labels, separated by commas. Provide your final answer enclosed in ... tags.", + "answer": "C, C, D, C, D, D, C, C, C, B, C" + }, + { + "image": "images/line_intersections_00001.png", + "question": "The image shows several curves or strands tangled together. Each curve has a single labeled endpoint on the perimeter, marked with a distinct letter. Starting from the endpoint labeled A, follow that curve through the interior tangle. At every crossing you pass through, record the label of the OTHER curve involved \u2014 when curve A crosses itself, record A. Report the ordered sequence of labels, separated by commas. Provide your final answer enclosed in ... tags.", + "answer": "D, B, B, C, C, B, B, B, C" + }, + { + "image": "images/line_intersections_00002.png", + "question": "The image shows several curves or strands tangled together. Each curve has a single labeled endpoint on the perimeter, marked with a distinct letter. Starting from the endpoint labeled A, follow that curve through the interior tangle. At every crossing you pass through, record the label of the OTHER curve involved \u2014 when curve A crosses itself, record A. Report the ordered sequence of labels, separated by commas. Provide your final answer enclosed in ... tags.", + "answer": "D, C, C, B, B, D, B, D, C" + }, + { + "image": "images/line_intersections_00003.png", + "question": "The image shows several curves or strands tangled together. Each curve has a single labeled endpoint on the perimeter, marked with a distinct letter. Starting from the endpoint labeled A, follow that curve through the interior tangle. At every crossing you pass through, record the label of the OTHER curve involved \u2014 when curve A crosses itself, record A. Report the ordered sequence of labels, separated by commas. Provide your final answer enclosed in ... tags.", + "answer": "D, C, B, C, D, C, D, C, C" + }, + { + "image": "images/line_intersections_00004.png", + "question": "The image shows several curves or strands tangled together. Each curve has a single labeled endpoint on the perimeter, marked with a distinct letter. Starting from the endpoint labeled A, follow that curve through the interior tangle. At every crossing you pass through, record the label of the OTHER curve involved \u2014 when curve A crosses itself, record A. Report the ordered sequence of labels, separated by commas. Provide your final answer enclosed in ... tags.", + "answer": "C, C, B, B, C, B, C, B, C, D, B" + } + ] +} \ No newline at end of file diff --git a/code/sequential_traversal/maze/creation.md b/code/sequential_traversal/maze/creation.md new file mode 100644 index 0000000000000000000000000000000000000000..9777deb3877311f3e10df261c65192cd8b85ee2b --- /dev/null +++ b/code/sequential_traversal/maze/creation.md @@ -0,0 +1,286 @@ +# Maze Dataset Generation + +## Goal + +Generate an image-based puzzle dataset where: + +- A rectangular grid maze is rendered as a black-and-white image. +- The maze has exactly two connected openings on its border — the correct entry and exit. +- Additional decoy openings lead into isolated dead-end pockets with no path to any other opening. +- The task can only be solved by visually tracing the interior path from one opening to another. + +This dataset is intended to test visual maze navigation under clutter, decoy paths, and structural ambiguity. + +## Core Puzzle Definition + +Each sample contains: + +- A grid maze of `rows × cols` cells. +- Between `min_openings` and `max_openings` labeled border openings. +- Exactly one connected pair among all openings — the two that share an interior path. +- All other openings lead to isolated dead-end regions that cannot reach any exit. + +Recommended defaults: + +- `rows, cols ∈ [8, 20]` +- `num_openings ∈ [3, 6]` + +The main question for a sample can be one of: + +- Identify which two labeled openings are connected. +- Given one opening label, identify the opening it connects to. +- Verify whether a proposed pair of labels is connected. + +## Visual Structure + +### Grid Layout + +- Each cell is rendered as a square of fixed `cell_size` pixels. +- Walls are drawn as filled black rectangles of `wall_width` pixels. +- A uniform `margin` is applied outside the grid boundary. +- Additional label padding is added outside the margin to accommodate opening labels. + +### Border Openings + +- Each opening is a gap in the outer border wall at a specific column (top/bottom) or row (left/right). +- Openings are placed at least one cell from each corner (`corner_buffer = 1`). +- Each opening is labeled with a consecutive uppercase letter (A, B, C, …). +- Labels are rendered just outside the maze boundary, adjacent to their opening. + +### Interior Walls + +Walls are drawn using two arrays: + +- `vertical_walls[r][c]`: wall on the right side of cell `(r, c)`. +- `horizontal_walls[r][c]`: wall on the bottom of cell `(r, c)`. + +The outer border is always fully walled except at openings. + +## Difficulty Requirements + +The puzzle should be hard for both humans and models. + +### Required difficulty factors + +- Long interior path between the connected pair. +- Many dead-end branches throughout the maze. +- Decoy openings that visually resemble the true entry/exit. +- No color coding or directional hints. +- Decoy regions large enough to require exploration before rejection. + +### Important anti-shortcut rule + +The connected pair must not be predictable by position: + +- The two connected openings must not always be the first two labels. +- Opening label order is shuffled after placement. +- The connected pair should not always appear on opposite sides. + +Allowed variation: + +- Random grid size within the configured range. +- Variable path length depending on grid size. + +Not allowed: + +- Any visual cue distinguishing the connected openings from decoys. +- Labels ordered to reveal the answer by position or alphabetical proximity. + +## Path Design + +### Connected Pair + +The connected pair is selected to maximize interior path length: + +- A minimum path length of `(rows × cols) / 3` cells is enforced. +- All border cell pairs are evaluated; those meeting the threshold are candidates. +- One candidate is chosen at random. + +### Decoy Openings + +Each decoy opening leads to an isolated subtree: + +- The subtree is grown from the decoy cell using open maze passages. +- It is then walled off from the rest of the maze. +- Minimum decoy size: 3 cells. Maximum: `max(12, (rows × cols × 3) / 5)` cells. +- Decoy subtrees must not overlap the main path cells or each other. + +### Validation + +After placement, each sample is verified: + +- The connected pair must be reachable from each other. +- Each decoy opening must not reach either connected opening. +- Samples failing validation have their decoys dropped rather than discarded entirely. + +## Appearance Constraints + +### Background + +- Plain white background. +- No textures, gradients, or background elements. + +### Wall styling + +- All walls use the same solid black fill. +- Wall thickness is uniform across the entire image (`wall_width` pixels). +- No anti-aliasing or blur applied to walls. + +### Text styling + +- Opening labels use a readable sans-serif font (Arial or DejaVu Sans fallback). +- Font size is fixed at 16pt. +- Labels are rendered in black on the white margin area. +- No decorative fonts, colors, or size variation. + +## Data Generation Procedure + +### 1. Generate maze + +- Use recursive DFS (backtracking) starting from cell `(0, 0)`. +- All cells are visited; the result is a perfect maze (no loops, fully connected). + +### 2. Select connected pair + +- Enumerate all border cell pairs. +- Compute BFS path length between each pair. +- Collect pairs with path length ≥ `(rows × cols) / 3`. +- Randomly select one qualifying pair as the main connected pair. + +### 3. Place decoy openings + +- Enumerate remaining border cells not on the main path. +- For each candidate, grow an isolated subtree using existing maze passages. +- Wall off the subtree from the rest of the maze. +- Accept the decoy if the subtree meets the minimum size and does not overlap forbidden cells. +- Continue until the desired number of openings is reached or candidates are exhausted. + +### 4. Shuffle label order + +- Concatenate the connected pair and all accepted decoy openings. +- Randomly permute the full list to assign labels A, B, C, … in shuffled order. +- Record the indices of the connected pair in the shuffled order. + +### 5. Render image + +- Draw outer border walls with gaps at each opening position. +- Draw interior vertical and horizontal walls cell by cell. +- Draw opening labels in the margin outside each gap. +- Export final image as PNG. + +## Annotation Format + +Each image has machine-readable metadata. + +JSON fields per record: + +```json +{ + "image": "images/maze_00000.png", + "rows": 12, + "cols": 15, + "cell_size": 32, + "wall_width": 5, + "margin": 8, + "num_openings": 4, + "openings": [ + {"label": "A", "side": "top", "index": 3, "is_connected": false}, + {"label": "B", "side": "left", "index": 5, "is_connected": true}, + {"label": "C", "side": "bottom", "index": 9, "is_connected": true}, + {"label": "D", "side": "right", "index": 2, "is_connected": false} + ], + "connected_pair": ["B", "C"], + "answer": "B-C", + "path_length": 67, + "difficulty": "hard" +} +``` + +Optional extra annotations: + +- Full wall arrays (`vertical_walls`, `horizontal_walls`) +- Opening pixel coordinates +- BFS path cell list +- Decoy subtree cell lists + +## Difficulty Levels + +### Easy + +- Small grid (≤ 10 × 10) +- Fewer openings (≤ 3) +- Shorter paths +- Fewer dead-end branches + +### Medium + +- Moderate grid size (≤ 16 × 16) +- Up to 5 openings +- Moderate path complexity + +### Hard + +- Larger grid (> 16 × 16) +- Many openings (> 5) +- Long winding paths +- Dense dead-end regions near decoy openings + +Difficulty classification: + +- Easy: `rows × cols ≤ 100` and `num_openings ≤ 3` +- Medium: `rows × cols ≤ 256` and `num_openings ≤ 5` +- Hard: otherwise + +## Quality Checks + +Reject or degrade samples if: + +- The connected pair is not reachable from each other. +- A decoy opening can reach a connected opening. +- The path length is below the minimum threshold. +- Any label overlaps significantly with a wall or another label. +- The grid is too small to accommodate the requested number of openings. + +Recommended automated checks: + +- BFS reachability between connected pair +- BFS isolation check for each decoy +- Minimum path length enforcement +- Minimum endpoint separation (corner buffer) +- Label render bounds within image dimensions + +## Recommended Dataset Variants + +Create separate controlled variants: + +- Small grids (8–12) vs large grids (14–20) +- Few openings (3–4) vs many openings (5–6) +- Connected pair always on opposite sides vs any sides +- Fixed grid size vs variable grid size +- Thick walls vs thin walls + +## Output Organization + +Suggested structure: + +```text +maze/ + creation.py + creation.md + annotations.jsonl + images/ + maze_00000.png + maze_00001.png + ... +``` + +## Implementation Notes + +- Use recursive DFS for maze generation; set a random seed for reproducibility. +- Store both rendered images and structured annotations. +- Apply rejection logic only for hard validity failures (disconnected pair, reachable decoy). +- Use Pillow (`PIL`) for rendering; no external rendering dependencies required. + +## Summary + +This dataset should produce maze navigation puzzles where the only reliable strategy is to trace the interior path through the maze grid from one labeled opening to another, ignoring dead-end pockets connected to decoy openings. diff --git a/code/sequential_traversal/maze/creation.py b/code/sequential_traversal/maze/creation.py new file mode 100644 index 0000000000000000000000000000000000000000..e8f2eaa69cd9f662d70fe5bf8ec511a92716bc36 --- /dev/null +++ b/code/sequential_traversal/maze/creation.py @@ -0,0 +1,672 @@ +from __future__ import annotations + +import argparse +import json +import random +import string +from collections import deque +from dataclasses import dataclass +from pathlib import Path +from typing import Dict, List, Set, Tuple + +from PIL import Image, ImageDraw, ImageFont + + +# --------------------------------------------------------------------------- +# Data model +# --------------------------------------------------------------------------- + +@dataclass +class MazeLayout: + rows: int + cols: int + vertical_walls: List[List[bool]] + horizontal_walls: List[List[bool]] + # (side, index) for each opening in label order + openings: List[Tuple[str, int]] + # (side, index, row, col) in label order + opening_cells: List[Tuple[str, int, int, int]] + # Indices into openings[] that are the connected pair (always sorted) + connected_indices: Tuple[int, int] + path_length: int + + +# --------------------------------------------------------------------------- +# Maze generation (DFS / recursive backtracking) +# --------------------------------------------------------------------------- + +def _generate_maze( + rng: random.Random, + rows: int, + cols: int, +) -> Tuple[List[List[bool]], List[List[bool]]]: + """Return (vertical_walls, horizontal_walls). + + vertical_walls[r][c] = True → wall on the right side of cell (r, c) + horizontal_walls[r][c] = True → wall on the bottom of cell (r, c) + """ + visited = [[False] * cols for _ in range(rows)] + vertical_walls = [[True] * cols for _ in range(rows)] + horizontal_walls = [[True] * cols for _ in range(rows)] + + def dfs(r: int, c: int) -> None: + visited[r][c] = True + directions = [(-1, 0), (1, 0), (0, -1), (0, 1)] + rng.shuffle(directions) + for dr, dc in directions: + nr, nc = r + dr, c + dc + if 0 <= nr < rows and 0 <= nc < cols and not visited[nr][nc]: + if dr == -1: + horizontal_walls[nr][nc] = False + elif dr == 1: + horizontal_walls[r][c] = False + elif dc == -1: + vertical_walls[nr][nc] = False + elif dc == 1: + vertical_walls[r][c] = False + dfs(nr, nc) + + dfs(0, 0) + return vertical_walls, horizontal_walls + + +def _open_neighbors( + vertical_walls: List[List[bool]], + horizontal_walls: List[List[bool]], + rows: int, + cols: int, + r: int, + c: int, +) -> List[Tuple[int, int]]: + neighbors: List[Tuple[int, int]] = [] + if r > 0 and not horizontal_walls[r - 1][c]: + neighbors.append((r - 1, c)) + if r < rows - 1 and not horizontal_walls[r][c]: + neighbors.append((r + 1, c)) + if c > 0 and not vertical_walls[r][c - 1]: + neighbors.append((r, c - 1)) + if c < cols - 1 and not vertical_walls[r][c]: + neighbors.append((r, c + 1)) + return neighbors + + +def _reachable( + vertical_walls: List[List[bool]], + horizontal_walls: List[List[bool]], + rows: int, + cols: int, + start: Tuple[int, int], +) -> Set[Tuple[int, int]]: + queue: deque[Tuple[int, int]] = deque([start]) + seen: Set[Tuple[int, int]] = {start} + while queue: + r, c = queue.popleft() + for nr, nc in _open_neighbors(vertical_walls, horizontal_walls, rows, cols, r, c): + if (nr, nc) not in seen: + seen.add((nr, nc)) + queue.append((nr, nc)) + return seen + + +def _path_cells( + vertical_walls: List[List[bool]], + horizontal_walls: List[List[bool]], + rows: int, + cols: int, + start: Tuple[int, int], + goal: Tuple[int, int], +) -> Set[Tuple[int, int]]: + """BFS path between start and goal; returns the set of cells on the path.""" + queue: deque[Tuple[int, int]] = deque([start]) + prev: Dict[Tuple[int, int], Tuple[int, int] | None] = {start: None} + while queue: + r, c = queue.popleft() + if (r, c) == goal: + break + for nr, nc in _open_neighbors(vertical_walls, horizontal_walls, rows, cols, r, c): + if (nr, nc) not in prev: + prev[(nr, nc)] = (r, c) + queue.append((nr, nc)) + if goal not in prev: + return set() + cells: Set[Tuple[int, int]] = set() + cur: Tuple[int, int] | None = goal + while cur is not None: + cells.add(cur) + cur = prev[cur] + return cells + + +def _opening_candidates(rows: int, cols: int, corner_buffer: int = 1) -> List[Tuple[str, int, int, int]]: + """(side, border_index, row, col) for each candidate opening.""" + candidates: List[Tuple[str, int, int, int]] = [] + for c in range(corner_buffer, cols - corner_buffer): + candidates.append(("top", c, 0, c)) + candidates.append(("bottom", c, rows - 1, c)) + for r in range(corner_buffer, rows - corner_buffer): + candidates.append(("left", r, r, 0)) + candidates.append(("right", r, r, cols - 1)) + return candidates + + +def _looks_natural( + vertical_walls: List[List[bool]], + horizontal_walls: List[List[bool]], + rows: int, + cols: int, + r: int, + c: int, +) -> bool: + return len(_open_neighbors(vertical_walls, horizontal_walls, rows, cols, r, c)) >= 1 + + +def _isolate_decoy( + vertical_walls: List[List[bool]], + horizontal_walls: List[List[bool]], + rows: int, + cols: int, + rng: random.Random, + root: Tuple[int, int], + forbidden: Set[Tuple[int, int]], + min_size: int, + max_size: int, +) -> Set[Tuple[int, int]]: + """Grow a compact subtree from root that avoids forbidden cells, then wall + it off so it forms an isolated dead-end region.""" + if root in forbidden: + return set() + + queue: deque[Tuple[int, int]] = deque([root]) + subtree: Set[Tuple[int, int]] = {root} + max_depth = rows * cols + + while queue and len(subtree) < max_size: + r, c = queue.popleft() + neighbors = _open_neighbors(vertical_walls, horizontal_walls, rows, cols, r, c) + rng.shuffle(neighbors) + for nr, nc in neighbors: + if (nr, nc) in subtree or (nr, nc) in forbidden: + continue + subtree.add((nr, nc)) + queue.append((nr, nc)) + if len(subtree) >= max_size: + break + + if len(subtree) < min_size: + return set() + + # Wall off subtree from the rest of the maze + for r, c in list(subtree): + if r > 0 and (r - 1, c) not in subtree: + horizontal_walls[r - 1][c] = True + if r < rows - 1 and (r + 1, c) not in subtree: + horizontal_walls[r][c] = True + if c > 0 and (r, c - 1) not in subtree: + vertical_walls[r][c - 1] = True + if c < cols - 1 and (r, c + 1) not in subtree: + vertical_walls[r][c] = True + + return subtree + + +# --------------------------------------------------------------------------- +# Layout sampling +# --------------------------------------------------------------------------- + +def sample_layout( + rng: random.Random, + rows: int, + cols: int, + num_openings: int, + min_path_length_floor: int = 0, +) -> MazeLayout: + """Generate a perfect maze and place border openings. + + Exactly two openings are connected through the maze interior; all others + lead to isolated dead-end pockets (decoys). The opening order is shuffled + so that the connected pair does not always occupy the first two label slots. + """ + vertical_walls, horizontal_walls = _generate_maze(rng, rows, cols) + border_cells = _opening_candidates(rows, cols, corner_buffer=1) + k = max(2, min(num_openings, len(border_cells))) + + # Pick a connected pair with a long interior path + min_path_len = max(min_path_length_floor, (rows * cols) // 3) + candidates: List[Tuple] = [] + for i in range(len(border_cells)): + for j in range(i + 1, len(border_cells)): + a, b = border_cells[i], border_cells[j] + path = _path_cells( + vertical_walls, horizontal_walls, rows, cols, + (a[2], a[3]), (b[2], b[3]), + ) + if len(path) >= min_path_len: + candidates.append((a, b, path)) + + if candidates: + first, second, path = rng.choice(candidates) + else: + first, second = rng.sample(border_cells, 2) + path = _path_cells( + vertical_walls, horizontal_walls, rows, cols, + (first[2], first[3]), (second[2], second[3]), + ) + + main_pair = [first, second] + forbidden: Set[Tuple[int, int]] = set(path) + + # Build decoy openings + total = rows * cols + remaining = [ + cell for cell in border_cells + if cell not in main_pair + and (cell[2], cell[3]) not in path + and _looks_natural(vertical_walls, horizontal_walls, rows, cols, cell[2], cell[3]) + ] + rng.shuffle(remaining) + + extras: List[Tuple[str, int, int, int]] = [] + for side, idx, r, c in remaining: + if len(extras) >= k - 2: + break + subtree = _isolate_decoy( + vertical_walls, horizontal_walls, rows, cols, rng, + (r, c), forbidden, + min_size=3, + max_size=max(12, (total * 3) // 5), + ) + if not subtree: + continue + extras.append((side, idx, r, c)) + forbidden.update(subtree) + + # Validate: ensure main pair is connected and decoys cannot reach the exit + ra, ca = first[2], first[3] + rb, cb = second[2], second[3] + main_component = _reachable(vertical_walls, horizontal_walls, rows, cols, (ra, ca)) + + if (rb, cb) not in main_component: + extras = [] + else: + valid_extras = [] + for side, idx, r, c in extras: + comp = _reachable(vertical_walls, horizontal_walls, rows, cols, (r, c)) + if (rb, cb) not in comp: + valid_extras.append((side, idx, r, c)) + extras = valid_extras + + all_cells = main_pair + extras + + # Shuffle label order so connected pair is not always A-B + order = list(range(len(all_cells))) + rng.shuffle(order) + shuffled = [all_cells[i] for i in order] + + conn_a = order.index(0) + conn_b = order.index(1) + if conn_a > conn_b: + conn_a, conn_b = conn_b, conn_a + + return MazeLayout( + rows=rows, + cols=cols, + vertical_walls=vertical_walls, + horizontal_walls=horizontal_walls, + openings=[(side, idx) for (side, idx, _, _) in shuffled], + opening_cells=shuffled, + connected_indices=(conn_a, conn_b), + path_length=len(path), + ) + + +# --------------------------------------------------------------------------- +# Rendering +# --------------------------------------------------------------------------- + +def _carve_openings( + draw: ImageDraw.ImageDraw, + rows: int, + cols: int, + cell_size: int, + wall_width: int, + margin: int, + openings: List[Tuple[str, int]], +) -> None: + x0, y0 = margin, margin + x1, y1 = margin + cols * cell_size, margin + rows * cell_size + opening_set = set(openings) + + for c in range(cols): + if ("top", c) not in opening_set: + xs = margin + c * cell_size + draw.rectangle( + [xs - wall_width // 2, y0 - wall_width // 2, + xs + cell_size + (wall_width - 1) // 2, y0 + (wall_width - 1) // 2], + fill="black", + ) + if ("bottom", c) not in opening_set: + xs = margin + c * cell_size + draw.rectangle( + [xs - wall_width // 2, y1 - wall_width // 2, + xs + cell_size + (wall_width - 1) // 2, y1 + (wall_width - 1) // 2], + fill="black", + ) + + for r in range(rows): + if ("left", r) not in opening_set: + ys = margin + r * cell_size + draw.rectangle( + [x0 - wall_width // 2, ys - wall_width // 2, + x0 + (wall_width - 1) // 2, ys + cell_size + (wall_width - 1) // 2], + fill="black", + ) + if ("right", r) not in opening_set: + ys = margin + r * cell_size + draw.rectangle( + [x1 - wall_width // 2, ys - wall_width // 2, + x1 + (wall_width - 1) // 2, ys + cell_size + (wall_width - 1) // 2], + fill="black", + ) + + +def _label_openings( + draw: ImageDraw.ImageDraw, + rows: int, + cols: int, + cell_size: int, + wall_width: int, + margin: int, + openings: List[Tuple[str, int]], +) -> None: + if not openings: + return + + x0, y0 = margin, margin + x1, y1 = margin + cols * cell_size, margin + rows * cell_size + img_w = cols * cell_size + 2 * margin + img_h = rows * cell_size + 2 * margin + + try: + font = ImageFont.truetype("Arial.ttf", size=16) + except Exception: + try: + font = ImageFont.truetype("DejaVuSans.ttf", size=16) + except Exception: + font = ImageFont.load_default() + + def measure(text: str) -> Tuple[float, float]: + if hasattr(draw, "textbbox"): + l, t, r, b = draw.textbbox((0, 0), text, font=font) + return r - l, b - t + if hasattr(font, "getbbox"): + l, t, r, b = font.getbbox(text) + return r - l, b - t + return font.getsize(text) # type: ignore[attr-defined] + + offset = wall_width + 4 + + for i, (side, index) in enumerate(openings): + label = string.ascii_uppercase[i] + w, h = measure(label) + + if side == "top": + col = max(0, min(cols - 1, index)) + cx = margin + col * cell_size + cell_size / 2 + x = cx - w / 2 + y = y0 - offset - h + elif side == "bottom": + col = max(0, min(cols - 1, index)) + cx = margin + col * cell_size + cell_size / 2 + x = cx - w / 2 + y = y1 + offset + elif side == "left": + row = max(0, min(rows - 1, index)) + cy = margin + row * cell_size + cell_size / 2 + x = x0 - offset - w + y = cy - h / 2 + elif side == "right": + row = max(0, min(rows - 1, index)) + cy = margin + row * cell_size + cell_size / 2 + x = x1 + offset + y = cy - h / 2 + else: + continue + + x = max(0.0, min(float(img_w - w), x)) + y = max(0.0, min(float(img_h - h), y)) + draw.text((x, y), label, fill="black", font=font) + + +def render_sample( + out_path: Path, + layout: MazeLayout, + cell_size: int, + wall_width: int, + margin: int, +) -> None: + if wall_width % 2 == 0: + wall_width += 1 + + rows, cols = layout.rows, layout.cols + label_pad = max(16, wall_width * 2) + draw_margin = margin + label_pad + + width = cols * cell_size + 2 * draw_margin + height = rows * cell_size + 2 * draw_margin + + img = Image.new("RGB", (width, height), "white") + draw = ImageDraw.Draw(img) + + _carve_openings(draw, rows, cols, cell_size, wall_width, draw_margin, layout.openings) + _label_openings(draw, rows, cols, cell_size, wall_width, draw_margin, layout.openings) + + for r in range(rows): + for c in range(cols): + cx = draw_margin + c * cell_size + cy = draw_margin + r * cell_size + + if layout.vertical_walls[r][c] and c != cols - 1: + x = cx + cell_size + draw.rectangle( + [x - wall_width // 2, cy - wall_width // 2, + x + (wall_width - 1) // 2, cy + cell_size + (wall_width - 1) // 2], + fill="black", + ) + + if layout.horizontal_walls[r][c] and r != rows - 1: + y = cy + cell_size + draw.rectangle( + [cx - wall_width // 2, y - wall_width // 2, + cx + cell_size + (wall_width - 1) // 2, y + (wall_width - 1) // 2], + fill="black", + ) + + img.save(out_path) + + +# --------------------------------------------------------------------------- +# Annotation building +# --------------------------------------------------------------------------- + +def choose_difficulty(rows: int, cols: int, num_openings: int) -> str: + total = rows * cols + if total <= 100 and num_openings <= 3: + return "easy" + if total <= 256 and num_openings <= 5: + return "medium" + return "hard" + + +def build_record( + image_name: str, + layout: MazeLayout, + cell_size: int, + wall_width: int, + margin: int, +) -> Dict[str, object]: + labels = [string.ascii_uppercase[i] for i in range(len(layout.openings))] + answer_a = labels[layout.connected_indices[0]] + answer_b = labels[layout.connected_indices[1]] + + opening_details = [] + for i, (side, idx) in enumerate(layout.openings): + opening_details.append({ + "label": labels[i], + "side": side, + "index": idx, + "is_connected": i in layout.connected_indices, + }) + + label_list = ", ".join(labels) + question = ( + f"The image shows a maze with {len(labels)} labeled openings on its border: " + f"{label_list}. Exactly one pair of openings is connected by a path through " + f"the maze; all other pairs are blocked by walls. " + f"Identify the connected pair. " + f"Trace passages carefully from each opening — a path exists only if you " + f"can travel from one opening to another through corridors without crossing " + f"any wall. Report the connected pair in alphabetical order, formatted as " + f"X-Y (for example, A-C). " + f"Provide your final answer enclosed in ... tags." + ) + + return { + "image": image_name, + "rows": layout.rows, + "cols": layout.cols, + "cell_size": cell_size, + "wall_width": wall_width, + "margin": margin, + "num_openings": len(layout.openings), + "openings": opening_details, + "connected_pair": [answer_a, answer_b], + "question": question, + "answer": f"{answer_a}-{answer_b}", + "path_length": layout.path_length, + "difficulty": choose_difficulty(layout.rows, layout.cols, len(layout.openings)), + } + + +# --------------------------------------------------------------------------- +# Dataset generation +# --------------------------------------------------------------------------- + +def ensure_output_dir(root: Path) -> Tuple[Path, Path]: + root.mkdir(parents=True, exist_ok=True) + images_dir = root / "images" + images_dir.mkdir(exist_ok=True) + return root, images_dir + + +def generate_dataset( + rng: random.Random, + count: int, + output_dir: Path, + images_dir: Path, + min_rows: int, + max_rows: int, + min_cols: int, + max_cols: int, + cell_size: int, + wall_width: int, + margin: int, + min_openings: int, + max_openings: int, + min_path_length_floor: int = 0, +) -> None: + records: List[Dict[str, object]] = [] + + for idx in range(count): + rows = rng.randint(min_rows, max_rows) + cols = rng.randint(min_cols, max_cols) + num_openings = rng.randint(min_openings, max_openings) + + layout = sample_layout(rng, rows, cols, num_openings, + min_path_length_floor=min_path_length_floor) + + image_name = f"maze_{idx:05d}.png" + render_sample(images_dir / image_name, layout, cell_size, wall_width, margin) + records.append(build_record(f"images/{image_name}", layout, cell_size, wall_width, margin)) + + with (output_dir / "annotations.jsonl").open("w", encoding="utf-8") as fh: + for record in records: + fh.write(json.dumps(record) + "\n") + + data_json = { + "task": "maze", + "category": "sequential_traversal", + "count": len(records), + "items": [ + {"image": r["image"], "question": r["question"], "answer": r["answer"]} + for r in records + ], + } + (output_dir / "data.json").write_text(json.dumps(data_json, indent=2)) + + +# --------------------------------------------------------------------------- +# CLI +# --------------------------------------------------------------------------- + +def parse_args() -> argparse.Namespace: + parser = argparse.ArgumentParser(description="Generate a maze puzzle dataset.") + parser.add_argument("--output-root", type=Path, default=".", help="Dataset root directory.") + parser.add_argument("--count", type=int, default=36) + parser.add_argument("--min-rows", type=int, default=8) + parser.add_argument("--max-rows", type=int, default=20) + parser.add_argument("--min-cols", type=int, default=8) + parser.add_argument("--max-cols", type=int, default=20) + parser.add_argument("--cell-size", type=int, default=32) + parser.add_argument("--wall-width", type=int, default=5) + parser.add_argument("--margin", type=int, default=8) + parser.add_argument("--min-openings", type=int, default=3) + parser.add_argument("--max-openings", type=int, default=6) + parser.add_argument("--seed", type=int, default=42) + parser.add_argument("--difficulty", type=int, default=5, + help="Integer difficulty >=0; scales maze size, openings, path length.") + return parser.parse_args() + + +def main() -> None: + args = parse_args() + rng = random.Random(args.seed) + output_dir, images_dir = ensure_output_dir(args.output_root) + + d = max(0, int(args.difficulty)) + # Formulas (no canvas_scale — canvas self-scales via rows × cols × cell_size): + # rows = cols = 10 + 2·d + # num_openings = min(10, 3 + d) + # min_path_length = 8 + 3·d + # wall_thickness_px = max(2, 5 − 0.3·d) + # dead_end_count = 3·d (decoys) + args.min_rows = args.max_rows = 10 + 2 * d + args.min_cols = args.max_cols = 10 + 2 * d + num_openings_val = min(10, 3 + d) + args.min_openings = num_openings_val + args.max_openings = num_openings_val + min_path_length_floor = 8 + 3 * d + args.wall_width = max(2, int(round(5 - 0.3 * d))) + dead_end_count = 3 * d # noqa: F841 (reserved; decoys already grown from extras) + + generate_dataset( + rng=rng, + count=args.count, + output_dir=output_dir, + images_dir=images_dir, + min_rows=args.min_rows, + max_rows=args.max_rows, + min_cols=args.min_cols, + max_cols=args.max_cols, + cell_size=args.cell_size, + wall_width=args.wall_width, + margin=args.margin, + min_openings=args.min_openings, + max_openings=args.max_openings, + min_path_length_floor=min_path_length_floor, + ) + + print(f"Saved dataset to {args.output_root}") + + +if __name__ == "__main__": + main() diff --git a/code/sequential_traversal/maze/data.json b/code/sequential_traversal/maze/data.json new file mode 100644 index 0000000000000000000000000000000000000000..45a84fb096191a1df7a045bc7946171bbe8063ff --- /dev/null +++ b/code/sequential_traversal/maze/data.json @@ -0,0 +1,32 @@ +{ + "task": "maze", + "category": "sequential_traversal", + "count": 5, + "items": [ + { + "image": "images/maze_00000.png", + "question": "The image shows a maze with several labeled openings on its border. Exactly one pair of openings is connected by a path through the maze; all other pairs are blocked by walls. Identify the connected pair. A path exists only if you can travel from one opening to another through corridors without crossing any wall. Report the connected pair in alphabetical order, formatted as X-Y. Provide your final answer enclosed in ... tags.", + "answer": "E-F" + }, + { + "image": "images/maze_00001.png", + "question": "The image shows a maze with several labeled openings on its border. Exactly one pair of openings is connected by a path through the maze; all other pairs are blocked by walls. Identify the connected pair. A path exists only if you can travel from one opening to another through corridors without crossing any wall. Report the connected pair in alphabetical order, formatted as X-Y. Provide your final answer enclosed in ... tags.", + "answer": "C-E" + }, + { + "image": "images/maze_00002.png", + "question": "The image shows a maze with several labeled openings on its border. Exactly one pair of openings is connected by a path through the maze; all other pairs are blocked by walls. Identify the connected pair. A path exists only if you can travel from one opening to another through corridors without crossing any wall. Report the connected pair in alphabetical order, formatted as X-Y. Provide your final answer enclosed in ... tags.", + "answer": "A-G" + }, + { + "image": "images/maze_00003.png", + "question": "The image shows a maze with several labeled openings on its border. Exactly one pair of openings is connected by a path through the maze; all other pairs are blocked by walls. Identify the connected pair. A path exists only if you can travel from one opening to another through corridors without crossing any wall. Report the connected pair in alphabetical order, formatted as X-Y. Provide your final answer enclosed in ... tags.", + "answer": "C-G" + }, + { + "image": "images/maze_00004.png", + "question": "The image shows a maze with several labeled openings on its border. Exactly one pair of openings is connected by a path through the maze; all other pairs are blocked by walls. Identify the connected pair. A path exists only if you can travel from one opening to another through corridors without crossing any wall. Report the connected pair in alphabetical order, formatted as X-Y. Provide your final answer enclosed in ... tags.", + "answer": "B-C" + } + ] +} \ No newline at end of file diff --git a/code/sequential_traversal/maze/maze.py b/code/sequential_traversal/maze/maze.py new file mode 100644 index 0000000000000000000000000000000000000000..b8f6a9ecf449c7de9c329c924cfc370b35023c22 --- /dev/null +++ b/code/sequential_traversal/maze/maze.py @@ -0,0 +1,743 @@ +import argparse +import random +import string +from collections import deque +from PIL import Image, ImageDraw, ImageFont + + +def generate_maze(rows, cols): + """ + Generate a perfect maze using DFS / recursive backtracking. + Returns: + vertical_walls[r][c]: whether there is a wall on the right of cell (r, c) + horizontal_walls[r][c]: whether there is a wall below cell (r, c) + + In a perfect maze, there is exactly one simple path between any two cells, + so any chosen entrance and exit will have a unique solution path. + """ + visited = [[False] * cols for _ in range(rows)] + + # Initially, every cell has a right wall and a bottom wall + vertical_walls = [[True] * cols for _ in range(rows)] + horizontal_walls = [[True] * cols for _ in range(rows)] + + def dfs(r, c): + visited[r][c] = True + + directions = [(-1, 0), (1, 0), (0, -1), (0, 1)] + random.shuffle(directions) + + for dr, dc in directions: + nr, nc = r + dr, c + dc + + if 0 <= nr < rows and 0 <= nc < cols and not visited[nr][nc]: + # Remove the wall between the current cell and the neighboring cell + if dr == -1: # up + horizontal_walls[nr][nc] = False + elif dr == 1: # down + horizontal_walls[r][c] = False + elif dc == -1: # left + vertical_walls[nr][nc] = False + elif dc == 1: # right + vertical_walls[r][c] = False + + dfs(nr, nc) + + dfs(0, 0) + return vertical_walls, horizontal_walls + + +def _carve_openings(draw, rows, cols, cell_size, wall_width, margin, openings): + """ + Draw the outer border of the maze while leaving openings. + + openings is a list of (side, index) tuples where: + - side in {"top", "bottom", "left", "right"} + - index is the row (for left/right) or column (for top/bottom), 0-based + """ + x0, y0 = margin, margin + x1, y1 = margin + cols * cell_size, margin + rows * cell_size + openings_set = set(openings) + + # Top border: one rectangle per cell-width segment, skipping openings. + # Each segment extends half a wall-width beyond its nominal endpoints so + # that neighboring segments overlap slightly and avoid pixel gaps. + for c in range(cols): + if ("top", c) in openings_set: + continue + x_start = margin + c * cell_size + x_end = x_start + cell_size + y = y0 + draw.rectangle( + [ + x_start - wall_width // 2, + y - wall_width // 2, + x_end + (wall_width - 1) // 2, + y + (wall_width - 1) // 2, + ], + fill="black", + ) + + # Bottom border + for c in range(cols): + if ("bottom", c) in openings_set: + continue + x_start = margin + c * cell_size + x_end = x_start + cell_size + y = y1 + draw.rectangle( + [ + x_start - wall_width // 2, + y - wall_width // 2, + x_end + (wall_width - 1) // 2, + y + (wall_width - 1) // 2, + ], + fill="black", + ) + + # Left border + for r in range(rows): + if ("left", r) in openings_set: + continue + y_start = margin + r * cell_size + y_end = y_start + cell_size + x = x0 + draw.rectangle( + [ + x - wall_width // 2, + y_start - wall_width // 2, + x + (wall_width - 1) // 2, + y_end + (wall_width - 1) // 2, + ], + fill="black", + ) + + # Right border + for r in range(rows): + if ("right", r) in openings_set: + continue + y_start = margin + r * cell_size + y_end = y_start + cell_size + x = x1 + draw.rectangle( + [ + x - wall_width // 2, + y_start - wall_width // 2, + x + (wall_width - 1) // 2, + y_end + (wall_width - 1) // 2, + ], + fill="black", + ) + + +def _label_openings(draw, rows, cols, cell_size, wall_width, margin, openings): + """ + Draw letter labels (A, B, C, ...) just outside each opening on the border. + Labels are placed on the outside of the maze, on the side where the wall + was opened. + """ + if not openings: + return + + x0, y0 = margin, margin + x1, y1 = margin + cols * cell_size, margin + rows * cell_size + width = cols * cell_size + 2 * margin + height = rows * cell_size + 2 * margin + + # Try to use a more readable TrueType font; fall back to default bitmap font. + try: + font = ImageFont.truetype("Arial.ttf", size=16) + except Exception: + try: + font = ImageFont.truetype("DejaVuSans.ttf", size=16) + except Exception: + font = ImageFont.load_default() + + def _measure(text: str): + """ + Measure text width and height in pixels in a Pillow-version-tolerant way. + """ + # Newer Pillow versions: prefer textbbox for accurate metrics + if hasattr(draw, "textbbox"): + left, top, right, bottom = draw.textbbox((0, 0), text, font=font) + return right - left, bottom - top + + # Fallback to font methods + if hasattr(font, "getbbox"): + left, top, right, bottom = font.getbbox(text) + return right - left, bottom - top + + # Oldest fallback + return font.getsize(text) + offset = wall_width + 4 # distance from the outer wall to the label + + # Compute a geometric position for each opening so that labels can be + # ordered from left to right (and then top to bottom when x is equal). + opening_positions = [] + for side, index in openings: + if side == "left": + row = max(0, min(rows - 1, index)) + cx = x0 + cy = margin + row * cell_size + cell_size / 2 + elif side == "right": + row = max(0, min(rows - 1, index)) + cx = x1 + cy = margin + row * cell_size + cell_size / 2 + elif side == "top": + col = max(0, min(cols - 1, index)) + cx = margin + col * cell_size + cell_size / 2 + cy = y0 + elif side == "bottom": + col = max(0, min(cols - 1, index)) + cx = margin + col * cell_size + cell_size / 2 + cy = y1 + else: + continue + + opening_positions.append((cx, cy, side, index)) + + # Preserve the logical order of openings (e.g., main openings first, + # followed by decoys) instead of reordering by geometry. + for i, (_cx, _cy, side, index) in enumerate(opening_positions): + label = string.ascii_uppercase[i % len(string.ascii_uppercase)] + + if side == "left": + row = max(0, min(rows - 1, index)) + cy = margin + row * cell_size + cell_size / 2 + w, h = _measure(label) + x = x0 - offset - w + y = cy - h / 2 + elif side == "right": + row = max(0, min(rows - 1, index)) + cy = margin + row * cell_size + cell_size / 2 + w, h = _measure(label) + x = x1 + offset + y = cy - h / 2 + elif side == "top": + col = max(0, min(cols - 1, index)) + cx = margin + col * cell_size + cell_size / 2 + w, h = _measure(label) + x = cx - w / 2 + y = y0 - offset - h + elif side == "bottom": + col = max(0, min(cols - 1, index)) + cx = margin + col * cell_size + cell_size / 2 + w, h = _measure(label) + x = cx - w / 2 + y = y1 + offset + else: + continue + + # Clamp to stay within the image so labels are always visible, + # even when the margin is small. + x = max(0, min(width - w, x)) + y = max(0, min(height - h, y)) + + draw.text((x, y), label, fill="black", font=font) + + +def _find_path_cells(vertical_walls, horizontal_walls, rows, cols, start, goal): + """ + Return the set of (r, c) cells on the unique path between start and goal + in the maze defined by vertical_walls and horizontal_walls. + """ + (sr, sc) = start + (gr, gc) = goal + + queue = deque([(sr, sc)]) + prev = {(sr, sc): None} + + while queue: + r, c = queue.popleft() + if (r, c) == (gr, gc): + break + + # Explore neighbors based on wall configuration + # Up + if r > 0 and not horizontal_walls[r - 1][c]: + nr, nc = r - 1, c + if (nr, nc) not in prev: + prev[(nr, nc)] = (r, c) + queue.append((nr, nc)) + # Down + if r < rows - 1 and not horizontal_walls[r][c]: + nr, nc = r + 1, c + if (nr, nc) not in prev: + prev[(nr, nc)] = (r, c) + queue.append((nr, nc)) + # Left + if c > 0 and not vertical_walls[r][c - 1]: + nr, nc = r, c - 1 + if (nr, nc) not in prev: + prev[(nr, nc)] = (r, c) + queue.append((nr, nc)) + # Right + if c < cols - 1 and not vertical_walls[r][c]: + nr, nc = r, c + 1 + if (nr, nc) not in prev: + prev[(nr, nc)] = (r, c) + queue.append((nr, nc)) + + # Reconstruct path from goal back to start + if (gr, gc) not in prev: + # Should not happen in a perfect maze, but guard anyway + return set() + + path_cells = set() + cur = (gr, gc) + while cur is not None: + path_cells.add(cur) + cur = prev[cur] + + return path_cells + + +def _build_opening_candidates(rows, cols, corner_buffer: int = 1): + """ + Build a list of candidate border cells to use as openings. + corner_buffer controls how many cells near each corner are excluded so + that openings do not sit directly on the corners. + """ + candidates = [] + + # Top and bottom rows, skipping a few cells near the corners + for c in range(corner_buffer, cols - corner_buffer): + candidates.append(("top", c, 0, c)) + candidates.append(("bottom", c, rows - 1, c)) + + # Left and right columns, skipping corners (and respecting the buffer) + for r in range(corner_buffer, rows - corner_buffer): + candidates.append(("left", r, r, 0)) + candidates.append(("right", r, r, cols - 1)) + + return candidates + + +def _get_open_neighbors(vertical_walls, horizontal_walls, rows, cols, r, c): + """ + Return a list of neighbor cells (nr, nc) that are reachable from (r, c) + given the current wall configuration. + """ + neighbors = [] + + # Up + if r > 0 and not horizontal_walls[r - 1][c]: + neighbors.append((r - 1, c)) + + # Down + if r < rows - 1 and not horizontal_walls[r][c]: + neighbors.append((r + 1, c)) + + # Left + if c > 0 and not vertical_walls[r][c - 1]: + neighbors.append((r, c - 1)) + + # Right + if c < cols - 1 and not vertical_walls[r][c]: + neighbors.append((r, c + 1)) + + return neighbors + + +def _reachable_cells(vertical_walls, horizontal_walls, rows, cols, start): + """ + Return the set of all cells reachable from start under the current wall + configuration. + """ + queue = deque([start]) + seen = {start} + + while queue: + r, c = queue.popleft() + for nr, nc in _get_open_neighbors(vertical_walls, horizontal_walls, rows, cols, r, c): + if (nr, nc) not in seen: + seen.add((nr, nc)) + queue.append((nr, nc)) + + return seen + + +def _opening_looks_natural(vertical_walls, horizontal_walls, rows, cols, r, c): + """ + Heuristic check: an opening is considered natural if its interior cell + has at least one open neighbor (and optionally could be extended). + """ + neighbors = _get_open_neighbors(vertical_walls, horizontal_walls, rows, cols, r, c) + return len(neighbors) >= 1 + + +def _isolate_decoy_region( + vertical_walls, + horizontal_walls, + rows, + cols, + root_cell, + forbidden_cells, + min_size: int = 3, + max_size: int = 12, + max_depth: int = 10, +): + """ + Starting from the given root_cell, perform a BFS in the current maze + topology to get a connected subtree that does not contain any cells in + forbidden_cells. Then, by adding walls around this subtree, cut it off + from the main maze as an isolated "fake entrance" region. + + The resulting decoy region is still a valid maze-like structure composed + of cell edges/walls, but there is no path between it and the main maze. + """ + (sr, sc) = root_cell + if (sr, sc) in forbidden_cells: + return set() + + # BFS queue stores (row, col, depth_from_root). + queue = deque([(sr, sc, 0)]) + subtree = {(sr, sc)} + + # Grow a limited-size, limited-depth subtree around root_cell that does not + # intersect the forbidden_cells set. This keeps decoy regions compact and + # shallow rather than cutting out huge, full-featured submazes. + while queue and len(subtree) < max_size: + r, c, depth = queue.popleft() + + # Do not extend beyond the maximum depth from the opening. + if depth >= max_depth: + continue + + neighbors = _get_open_neighbors(vertical_walls, horizontal_walls, rows, cols, r, c) + random.shuffle(neighbors) + + for nr, nc in neighbors: + if (nr, nc) in subtree or (nr, nc) in forbidden_cells: + continue + + subtree.add((nr, nc)) + queue.append((nr, nc, depth + 1)) + + if len(subtree) >= max_size: + break + + # If the subtree is too small, we simply skip using it as a decoy. + if len(subtree) < min_size: + return set() + + # Add walls on all edges between this subtree and the rest of the maze + # so that it is completely disconnected from the main maze. + for r, c in list(subtree): + # Up neighbor + if r > 0: + nr, nc = r - 1, c + if (nr, nc) not in subtree: + horizontal_walls[r - 1][c] = True + + # Down neighbor + if r < rows - 1: + nr, nc = r + 1, c + if (nr, nc) not in subtree: + horizontal_walls[r][c] = True + + # Left neighbor + if c > 0: + nr, nc = r, c - 1 + if (nr, nc) not in subtree: + vertical_walls[r][c - 1] = True + + # Right neighbor + if c < cols - 1: + nr, nc = r, c + 1 + if (nr, nc) not in subtree: + vertical_walls[r][c] = True + + return subtree + + +def draw_maze( + rows, + cols, + cell_size=32, + wall_width=5, + margin=8, + filename="maze.png", + entrances=None, + num_openings=4, +): + """ + Draw a maze image. + + Parameters: + rows, cols: maze grid size + cell_size: size (in pixels) of each cell + wall_width: wall thickness (in pixels). If an even value is provided, + it will be rounded up to the next odd value to keep walls + symmetric around grid lines. + margin: outer margin (in pixels) + filename: output image filename + entrances: optional list of (side, index) tuples that define openings + in the outer border. Examples: + [("left", 0), ("right", rows - 1)] + [("top", 0), ("bottom", cols - 1)] + [("left", 0), ("left", rows - 1), ("right", rows // 2)] + If provided, all of these openings are carved visually, + but only two of them are connected to the interior; all + others are dead ends. + num_openings: if entrances is None, this many random openings will be + carved on the outer border (clamped so it is at least 2 + and at most the number of border cells). Exactly one + entrance and one exit will connect to the maze interior; + all others are dead ends (no path to an exit). + + The underlying maze is perfect (DFS backtracking) before we cut any + additional dead-end openings, so between the unique entrance and exit + there is exactly one simple path. Any other opening leads only to an + isolated pocket with no exit. + """ + # Ensure wall width is an odd number so rectangles are symmetric + if wall_width % 2 == 0: + wall_width += 1 + + vertical_walls, horizontal_walls = generate_maze(rows, cols) + + # Give extra canvas space so border labels have room outside the maze. + # This does not change the "inner" margin between maze and outer walls + # seen visually by the user; it only affects total image size. + label_pad = max(16, wall_width * 2) + draw_margin = margin + label_pad + + width = cols * cell_size + 2 * draw_margin + height = rows * cell_size + 2 * draw_margin + + img = Image.new("RGB", (width, height), "white") + draw = ImageDraw.Draw(img) + + # Decide entrance / exit and (optionally) additional decoy openings + if entrances is None: + # Build filtered list of border candidates that are not too close to corners. + border_cells = _build_opening_candidates(rows, cols, corner_buffer=1) + max_openings = len(border_cells) + k = max(2, min(num_openings, max_openings)) + + # Choose entrance/exit pair based on actual path length through the maze. + min_path_len = (rows * cols) // 3 + candidate_pairs = [] + + for i in range(len(border_cells)): + for j in range(i + 1, len(border_cells)): + a = border_cells[i] + b = border_cells[j] + path = _find_path_cells( + vertical_walls, + horizontal_walls, + rows, + cols, + (a[2], a[3]), + (b[2], b[3]), + ) + if len(path) >= min_path_len: + candidate_pairs.append((a, b, path)) + + if candidate_pairs: + first, second, path_cells = random.choice(candidate_pairs) + else: + # Fallback: pick any two border cells and use their path + first, second = random.sample(border_cells, 2) + path_cells = _find_path_cells( + vertical_walls, + horizontal_walls, + rows, + cols, + (first[2], first[3]), + (second[2], second[3]), + ) + + main_border_cells = [first, second] + (side_a, idx_a, ra, ca) = first + (side_b, idx_b, rb, cb) = second + + # Track cells already "used" (main path) so decoys never reuse them. + forbidden_cells = set(path_cells) + + # Only place decoys on border cells that are not on the main path and + # whose interior cell looks reasonably natural as an opening. + remaining_border = [ + cell for cell in border_cells + if cell not in main_border_cells + and (cell[2], cell[3]) not in path_cells + and _opening_looks_natural( + vertical_walls, horizontal_walls, rows, cols, cell[2], cell[3] + ) + ] + + extra_to_pick = max(0, min(k - 2, len(remaining_border))) + extra_border_cells = [] + + random.shuffle(remaining_border) + for side, idx, r, c in remaining_border: + if len(extra_border_cells) >= extra_to_pick: + break + + # Allow decoy regions to be fairly large and deep so they feel + # comparable in scale to the main maze, while still being isolated + # and reachable only from a single opening. + total_cells = rows * cols + decoy_min_size = 3 + # Up to roughly 60% of the maze cells for a single decoy region. + decoy_max_size = max(12, (total_cells * 3) // 5) + # Depth bound effectively lets the region explore as far as area allows. + decoy_max_depth = total_cells + + # From (r, c), grow a small decoy subtree that does not intersect + # the main path, and cut it off to form a fake region. + subtree = _isolate_decoy_region( + vertical_walls, + horizontal_walls, + rows, + cols, + (r, c), + forbidden_cells, + min_size=decoy_min_size, + max_size=decoy_max_size, + max_depth=decoy_max_depth, + ) + + if not subtree: + continue + + extra_border_cells.append((side, idx, r, c)) + # Add these cells to forbidden so later decoys do not overlap. + forbidden_cells.update(subtree) + + # All border cells where we actually create visual openings + all_border_cells = main_border_cells + extra_border_cells + entrances_to_draw = [(side, idx) for (side, idx, _r, _c) in all_border_cells] + + # Final validation: ensure main entrance/exit are connected, and decoys + # cannot reach the exit. + main_component = _reachable_cells( + vertical_walls, + horizontal_walls, + rows, + cols, + (ra, ca), + ) + if (rb, cb) not in main_component: + # In the unlikely event this fails, fall back to a simple, direct pair. + first, second = random.sample(border_cells, 2) + main_border_cells = [first, second] + (side_a, idx_a, ra, ca) = first + (side_b, idx_b, rb, cb) = second + path_cells = _find_path_cells( + vertical_walls, + horizontal_walls, + rows, + cols, + (ra, ca), + (rb, cb), + ) + entrances_to_draw = [(first[0], first[1]), (second[0], second[1])] + + else: + # Verify each decoy opening does not connect to the real exit. + valid = True + for side, idx, r, c in extra_border_cells: + decoy_component = _reachable_cells( + vertical_walls, + horizontal_walls, + rows, + cols, + (r, c), + ) + if (rb, cb) in decoy_component: + valid = False + break + + if not valid: + # If validation fails, degrade gracefully by dropping decoys. + entrances_to_draw = [ + (first[0], first[1]), + (second[0], second[1]), + ] + else: + # If explicit entrances are provided, treat all of them as real openings + entrances_to_draw = entrances + + # Finally, carve the visual openings in the outer border and label them + _carve_openings(draw, rows, cols, cell_size, wall_width, draw_margin, entrances_to_draw) + _label_openings(draw, rows, cols, cell_size, wall_width, draw_margin, entrances_to_draw) + + # Draw internal walls + for r in range(rows): + for c in range(cols): + cell_x = draw_margin + c * cell_size + cell_y = draw_margin + r * cell_size + + # Right wall (extend half a wall-width above and below so it + # slightly overlaps neighboring horizontal walls) + if vertical_walls[r][c] and c != cols - 1: + x = cell_x + cell_size + draw.rectangle( + [ + x - wall_width // 2, + cell_y - wall_width // 2, + x + (wall_width - 1) // 2, + cell_y + cell_size + (wall_width - 1) // 2, + ], + fill="black", + ) + + # Bottom wall (extend half a wall-width left and right so it + # slightly overlaps neighboring vertical walls) + if horizontal_walls[r][c] and r != rows - 1: + y = cell_y + cell_size + draw.rectangle( + [ + cell_x - wall_width // 2, + y - wall_width // 2, + cell_x + cell_size + (wall_width - 1) // 2, + y + (wall_width - 1) // 2, + ], + fill="black", + ) + + img.save(filename) + print(f"Maze has been saved to {filename}") + + +if __name__ == "__main__": + parser = argparse.ArgumentParser(description="Generate and save a random perfect maze image.") + parser.add_argument("--rows", type=int, default=10, help="Number of rows in the maze grid.") + parser.add_argument("--cols", type=int, default=10, help="Number of columns in the maze grid.") + parser.add_argument("--cell-size", type=int, default=32, help="Pixel size of each maze cell.") + parser.add_argument( + "--wall-width", + type=int, + default=5, + help="Wall thickness in pixels. Even values are rounded up to the next odd value.", + ) + parser.add_argument("--margin", type=int, default=8, help="Outer margin in pixels around the maze.") + parser.add_argument( + "--num-openings", + type=int, + default=4, + help=( + "Total number of border openings. Exactly two of them form the unique " + "solvable entrance/exit pair; the rest are dead-end decoys." + ), + ) + parser.add_argument( + "--output", + type=str, + default="random_maze.png", + help="Output image filename.", + ) + + args = parser.parse_args() + + draw_maze( + rows=args.rows, + cols=args.cols, + cell_size=args.cell_size, + wall_width=args.wall_width, + margin=args.margin, + filename=args.output, + num_openings=args.num_openings, + ) \ No newline at end of file diff --git a/code/sequential_traversal/traverse_ordering/creation.py b/code/sequential_traversal/traverse_ordering/creation.py new file mode 100644 index 0000000000000000000000000000000000000000..5d37ec0dd12817ddbe3f4f9a1088b377d82b2ea2 --- /dev/null +++ b/code/sequential_traversal/traverse_ordering/creation.py @@ -0,0 +1,951 @@ +"""Traverse Ordering (two-curve variant). + +Two smooth tangled curves cross each other multiple times (perimeter-anchored +turtle walks, same generator style as line_intersections). Labeled marks +(A, B, C, ...) are placed on BOTH curves at well-spaced arc-length positions, +avoiding crossings and close-strand regions. Exactly one curve additionally +has a green 'S' marker at one of its endpoints. The task: starting from S, +follow that curve to its other endpoint and list the labels of marks on the +S-curve in the order visited. Marks on the other curve are distractors. +""" +from __future__ import annotations + +import argparse +import json +import math +import os +import random +import string +import sys +from collections import defaultdict +from pathlib import Path +from typing import Dict, List, Tuple + +import matplotlib +matplotlib.use("Agg") +import matplotlib.pyplot as plt +import numpy as np +from tqdm import tqdm + + +LINE_COLOR = "#2f2f2f" +MARK_COLORS = [ + "#e63946", "#457b9d", "#2a9d8f", "#e9c46a", "#f4a261", + "#264653", "#6a4c93", "#1982c4", "#8ac926", "#ff595e", +] +S_COLOR = "#1f9d3f" + + +# ── Perimeter-anchored turtle curve (from line_intersections) ────── + +def _perimeter_point(s: float, width: int, height: int, margin: int + ) -> Tuple[np.ndarray, np.ndarray]: + side = int(s) % 4 + t = s - int(s) + if side == 0: + x = margin + (width - 2 * margin) * t + y = margin + inward = np.array([0.0, 1.0]) + elif side == 1: + x = width - margin + y = margin + (height - 2 * margin) * t + inward = np.array([-1.0, 0.0]) + elif side == 2: + x = width - margin - (width - 2 * margin) * t + y = height - margin + inward = np.array([0.0, -1.0]) + else: + x = margin + y = height - margin - (height - 2 * margin) * t + inward = np.array([1.0, 0.0]) + return np.array([x, y]), inward + + +def sample_perimeter_anchors( + rng: random.Random, + num_anchors: int, + width: int, + height: int, + margin: int, + min_gap: float | None = None, + jitter_deg: float = 10.0, + max_attempts: int = 300, +) -> List[Tuple[np.ndarray, np.ndarray]]: + if min_gap is None: + min_gap = (4.0 / num_anchors) * 0.5 + for _ in range(max_attempts): + positions = sorted(rng.uniform(0.0, 4.0) for _ in range(num_anchors)) + ok = True + for i in range(num_anchors): + gap = (positions[(i + 1) % num_anchors] - positions[i]) % 4.0 + if gap < min_gap: + ok = False + break + if ok: + anchors = [] + for s in positions: + pt, inward = _perimeter_point(s, width, height, margin) + ang = math.atan2(inward[1], inward[0]) + ang += math.radians(rng.uniform(-jitter_deg, jitter_deg)) + inward = np.array([math.cos(ang), math.sin(ang)]) + anchors.append((pt, inward)) + return anchors + return None + + +def _angle_diff(current: float, target: float) -> float: + d = target - current + return (d + math.pi) % (2 * math.pi) - math.pi + + +def build_anchored_curve( + rng: random.Random, + width: int, + height: int, + start_pos: np.ndarray, + start_inward: np.ndarray, + end_pos: np.ndarray, + end_inward: np.ndarray, + leader_length: float = 150.0, + step_size: float = 3.3, + max_turn: float = 0.028, + drift_rate: float = 0.007, + num_waypoints: int = 2, + interior_margin: int = 260, + max_steps: int = 1800, +) -> np.ndarray: + pts: List[np.ndarray] = [start_pos.copy()] + x, y = float(start_pos[0]), float(start_pos[1]) + theta = math.atan2(start_inward[1], start_inward[0]) + + raw_wps: List[np.ndarray] = [] + for _ in range(num_waypoints): + wx = rng.uniform(interior_margin, width - interior_margin) + wy = rng.uniform(interior_margin, height - interior_margin) + raw_wps.append(np.array([wx, wy])) + direction = end_pos - start_pos + dir_norm = direction / (np.linalg.norm(direction) + 1e-9) + raw_wps.sort(key=lambda p: float(np.dot(p - start_pos, dir_norm))) + + funnel = end_pos + end_inward * 400.0 + lookahead = end_pos + (-end_inward) * 240.0 + waypoints: List[np.ndarray] = raw_wps + [funnel, lookahead] + + exit_theta = math.atan2(-end_inward[1], -end_inward[0]) + entry_steps = max(2, int(leader_length / step_size)) + + wp_idx = 0 + drift = 0.0 + perp = np.array([-end_inward[1], end_inward[0]]) + + for step_count in range(max_steps): + sd = ((x - end_pos[0]) * end_inward[0] + + (y - end_pos[1]) * end_inward[1]) + lat = abs((x - end_pos[0]) * perp[0] + (y - end_pos[1]) * perp[1]) + + if sd <= 0.0: + break + + if step_count < entry_steps: + turn = 0.0 + elif sd < leader_length and lat < 40.0: + head_err = _angle_diff(theta, exit_theta) + turn = max(-max_turn, min(max_turn, head_err)) + else: + is_final = (wp_idx == len(waypoints) - 1) + if is_final: + wx, wy = waypoints[-1] + pull = 0.18 + else: + wx, wy = waypoints[wp_idx] + steer_dist = math.hypot(wx - x, wy - y) + if steer_dist < 140.0: + wp_idx += 1 + continue + pull = 0.09 + + drift += rng.gauss(0, drift_rate) + drift = max(-max_turn * 0.7, min(max_turn * 0.7, drift)) + turn = drift + rng.gauss(0, max_turn * 0.12) + target_ang = math.atan2(wy - y, wx - x) + steer = _angle_diff(theta, target_ang) + turn += steer * pull + turn = max(-max_turn, min(max_turn, turn)) + + theta += turn + x += step_size * math.cos(theta) + y += step_size * math.sin(theta) + pts.append(np.array([x, y])) + + return np.array(pts) + + +# ── Intersection detection ───────────────────────────────────────── + +def _segments_cross(p0, p1, q0, q1) -> bool: + eps = 1e-8 + o1 = (p1[0]-p0[0])*(q0[1]-p0[1]) - (p1[1]-p0[1])*(q0[0]-p0[0]) + o2 = (p1[0]-p0[0])*(q1[1]-p0[1]) - (p1[1]-p0[1])*(q1[0]-p0[0]) + o3 = (q1[0]-q0[0])*(p0[1]-q0[1]) - (q1[1]-q0[1])*(p0[0]-q0[0]) + o4 = (q1[0]-q0[0])*(p1[1]-q0[1]) - (q1[1]-q0[1])*(p1[0]-q0[0]) + return ((o1 > eps and o2 < -eps) or (o1 < -eps and o2 > eps)) and \ + ((o3 > eps and o4 < -eps) or (o3 < -eps and o4 > eps)) + + +def _crossing_angle(p0, p1, q0, q1) -> float: + d1 = p1 - p0 + d2 = q1 - q0 + cos_a = np.dot(d1, d2) / (np.linalg.norm(d1) * np.linalg.norm(d2) + 1e-12) + return math.degrees(math.acos(min(1.0, abs(cos_a)))) + + +def _find_crossings( + poly_a: np.ndarray, + poly_b: np.ndarray, + same_curve: bool = False, + min_seg_gap: int = 10, +): + na = len(poly_a) - 1 + nb = len(poly_b) - 1 + + a_min_x = np.minimum(poly_a[:-1, 0], poly_a[1:, 0]) + a_max_x = np.maximum(poly_a[:-1, 0], poly_a[1:, 0]) + a_min_y = np.minimum(poly_a[:-1, 1], poly_a[1:, 1]) + a_max_y = np.maximum(poly_a[:-1, 1], poly_a[1:, 1]) + b_min_x = np.minimum(poly_b[:-1, 0], poly_b[1:, 0]) + b_max_x = np.maximum(poly_b[:-1, 0], poly_b[1:, 0]) + b_min_y = np.minimum(poly_b[:-1, 1], poly_b[1:, 1]) + b_max_y = np.maximum(poly_b[:-1, 1], poly_b[1:, 1]) + + all_min_x = np.concatenate([a_min_x, b_min_x]) + all_max_x = np.concatenate([a_max_x, b_max_x]) + cell_size = max(np.median(all_max_x - all_min_x), 1.0) * 3 + + grid_a = defaultdict(list) + for i in range(na): + cx0 = int(a_min_x[i] / cell_size); cx1 = int(a_max_x[i] / cell_size) + cy0 = int(a_min_y[i] / cell_size); cy1 = int(a_max_y[i] / cell_size) + for gx in range(cx0, cx1 + 1): + for gy in range(cy0, cy1 + 1): + grid_a[(gx, gy)].append(i) + + grid_b = defaultdict(list) + for j in range(nb): + cx0 = int(b_min_x[j] / cell_size); cx1 = int(b_max_x[j] / cell_size) + cy0 = int(b_min_y[j] / cell_size); cy1 = int(b_max_y[j] / cell_size) + for gx in range(cx0, cx1 + 1): + for gy in range(cy0, cy1 + 1): + grid_b[(gx, gy)].append(j) + + checked = set() + worst_angle = 180.0 + details: List[Dict] = [] + for cell_key in grid_a: + if cell_key not in grid_b: + continue + for i in grid_a[cell_key]: + for j in grid_b[cell_key]: + if same_curve: + ii, jj = min(i, j), max(i, j) + if jj - ii < min_seg_gap: + continue + key = (ii, jj) + else: + key = (i, j) + if key in checked: + continue + checked.add(key) + if a_max_x[i] < b_min_x[j] or b_max_x[j] < a_min_x[i] or \ + a_max_y[i] < b_min_y[j] or b_max_y[j] < a_min_y[i]: + continue + if same_curve: + si, sj = key + p0, p1 = poly_a[si], poly_a[si + 1] + q0, q1 = poly_b[sj], poly_b[sj + 1] + else: + si, sj = i, j + p0, p1 = poly_a[si], poly_a[si + 1] + q0, q1 = poly_b[sj], poly_b[sj + 1] + if not _segments_cross(p0, p1, q0, q1): + continue + d1 = p1 - p0 + d2 = q1 - q0 + denom = d1[0] * d2[1] - d1[1] * d2[0] + if abs(denom) < 1e-12: + continue + ti = ((q0[0] - p0[0]) * d2[1] - (q0[1] - p0[1]) * d2[0]) / denom + tj = ((q0[0] - p0[0]) * d1[1] - (q0[1] - p0[1]) * d1[0]) / denom + px = float(p0[0] + ti * d1[0]) + py = float(p0[1] + ti * d1[1]) + angle = _crossing_angle(p0, p1, q0, q1) + worst_angle = min(worst_angle, angle) + details.append({ + "i": int(si), "j": int(sj), + "ti": float(ti), "tj": float(tj), + "px": px, "py": py, "angle": float(angle), + }) + return len(details), worst_angle, details + + +def _details_to_points(details: List[Dict]) -> np.ndarray: + if not details: + return np.zeros((0, 2)) + return np.array([[d["px"], d["py"]] for d in details]) + + +def _curves_too_close( + poly_a: np.ndarray, + poly_b: np.ndarray, + same_curve: bool = False, + min_dist: float = 7.0, + sample_step: int = 3, + self_index_gap: int = 50, + known_crossings: np.ndarray | None = None, + crossing_exclude_radius: float = 30.0, +) -> bool: + nb = len(poly_b) - 1 + b_pts = poly_b[:-1] + b_vecs = poly_b[1:] - poly_b[:-1] + b_lens_sq = np.maximum((b_vecs ** 2).sum(axis=1), 1e-12) + has_crossings = known_crossings is not None and len(known_crossings) > 0 + for idx in range(0, len(poly_a), sample_step): + px, py = poly_a[idx] + p = np.array([px, py]) + dp = p - b_pts + t = (dp * b_vecs).sum(axis=1) / b_lens_sq + t = np.clip(t, 0.0, 1.0) + proj = b_pts + t[:, None] * b_vecs + dists = np.sqrt(((p - proj) ** 2).sum(axis=1)) + if same_curve: + mask = np.abs(np.arange(nb) - idx) < self_index_gap + dists[mask] = 9999.0 + if dists.min() < min_dist: + if has_crossings: + cross_dists = np.sqrt(((p - known_crossings) ** 2).sum(axis=1)) + if cross_dists.min() < crossing_exclude_radius: + continue + return True + return False + + +# ── Mark placement ───────────────────────────────────────────────── + +def place_marks_on_curve( + rng: random.Random, + polyline: np.ndarray, + other_polyline: np.ndarray, + num_marks: int, + avoid_points: List[np.ndarray], + existing_marks: List[Tuple[np.ndarray, int]] | None = None, + min_arc_frac: float = 0.10, + min_pixel_to_crossing: float = 70.0, + min_pixel_to_other_strand: float = 36.0, + min_pixel_between_marks: float = 90.0, + endpoint_arc_margin_frac: float = 0.08, + max_iter: int = 1200, + extra_polylines: List[np.ndarray] | None = None, +) -> List[int]: + """Place `num_marks` indices along `polyline`, avoiding crossing points, + regions where the OTHER curve passes close, and existing marks on either + curve. Endpoints are reserved (for the S marker). + """ + n = len(polyline) + diffs = np.diff(polyline, axis=0) + seg_lens = np.sqrt((diffs ** 2).sum(axis=1)) + arc = np.concatenate([[0.0], np.cumsum(seg_lens)]) + total_len = arc[-1] + min_arc_dist = total_len * min_arc_frac + margin_arc = total_len * endpoint_arc_margin_frac + + # Precompute other curve's segments for fast distance queries. + op = other_polyline + o_pts = op[:-1] + o_vecs = op[1:] - op[:-1] + o_lens_sq = np.maximum((o_vecs ** 2).sum(axis=1), 1e-12) + + def dist_to_other(pt: np.ndarray) -> float: + dp = pt - o_pts + t = (dp * o_vecs).sum(axis=1) / o_lens_sq + t = np.clip(t, 0.0, 1.0) + proj = o_pts + t[:, None] * o_vecs + return float(np.sqrt(((pt - proj) ** 2).sum(axis=1)).min()) + + extra_bundles = [] + for ep in (extra_polylines or []): + e_pts = ep[:-1] + e_vecs = ep[1:] - ep[:-1] + e_lens_sq = np.maximum((e_vecs ** 2).sum(axis=1), 1e-12) + extra_bundles.append((e_pts, e_vecs, e_lens_sq)) + + def dist_to_any_extra(pt: np.ndarray) -> float: + if not extra_bundles: + return 1e9 + best = 1e9 + for e_pts, e_vecs, e_lens_sq in extra_bundles: + dp = pt - e_pts + t = (dp * e_vecs).sum(axis=1) / e_lens_sq + t = np.clip(t, 0.0, 1.0) + proj = e_pts + t[:, None] * e_vecs + d = float(np.sqrt(((pt - proj) ** 2).sum(axis=1)).min()) + if d < best: + best = d + return best + + def too_near_avoid(pt: np.ndarray) -> bool: + for cp in avoid_points: + if np.sqrt(((pt - cp) ** 2).sum()) < min_pixel_to_crossing: + return True + return False + + existing = existing_marks or [] + mark_indices: List[int] = [] + for _ in range(max_iter): + if len(mark_indices) == num_marks: + break + t = rng.uniform(margin_arc, total_len - margin_arc) + idx = int(np.searchsorted(arc, t)) + idx = max(1, min(n - 2, idx)) + pt = polyline[idx] + if too_near_avoid(pt): + continue + if dist_to_other(pt) < min_pixel_to_other_strand: + continue + if dist_to_any_extra(pt) < min_pixel_to_other_strand: + continue + if any(abs(arc[idx] - arc[m]) < min_arc_dist for m in mark_indices): + continue + if any(np.sqrt(((pt - polyline[m]) ** 2).sum()) < min_pixel_between_marks + for m in mark_indices): + continue + if any(np.sqrt(((pt - ep) ** 2).sum()) < min_pixel_between_marks + for ep, _ in existing): + continue + mark_indices.append(idx) + + mark_indices.sort() + return mark_indices + + +def label_anchor_for_mark( + polyline: np.ndarray, + other_polyline: np.ndarray, + mark_idx: int, + width: int, + height: int, + offset: float = 48.0, + tangent_span: int = 6, +) -> Tuple[float, float]: + n = len(polyline) + lo = max(0, mark_idx - tangent_span) + hi = min(n - 1, mark_idx + tangent_span) + tan = polyline[hi] - polyline[lo] + tnorm = math.hypot(float(tan[0]), float(tan[1])) + if tnorm < 1e-8: + ux, uy = 0.0, -1.0 + else: + ux = float(tan[0]) / tnorm + uy = float(tan[1]) / tnorm + perps = [(-uy, ux), (uy, -ux)] + pt = polyline[mark_idx] + best = None + best_score = -1e9 + for px, py in perps: + cx = float(pt[0]) + px * offset + cy = float(pt[1]) + py * offset + score = 0.0 + d_self = float(np.sqrt(((polyline - np.array([cx, cy])) ** 2).sum(axis=1)).min()) + d_other = float(np.sqrt(((other_polyline - np.array([cx, cy])) ** 2).sum(axis=1)).min()) + score += min(d_self, d_other) + if 30 < cx < width - 30 and 30 < cy < height - 30: + score += 200.0 + if score > best_score: + best_score = score + best = (cx, cy) + cx, cy = best + return ( + float(np.clip(cx, 34, width - 34)), + float(np.clip(cy, 34, height - 34)), + ) + + +# ── Instance sampling ────────────────────────────────────────────── + +def sample_instance( + rng: random.Random, + width: int, + height: int, + min_marks_total: int = 4, + max_marks_total: int = 7, + min_crossings: int = 7, + max_crossings: int = 25, + min_pixel_between_marks: float = 90.0, + s_endpoint_excl_px: float = 90.0, + perimeter_margin: int = 70, + interior_margin: int = 260, + max_attempts: int = 2500, + num_waypoints: int = 5, + s_marks_override: int | None = None, + o_marks_override: int | None = None, + num_extra_distractor_curves: int = 0, + extra_distractor_max_attempts: int = 80, +) -> Dict | None: + labels_pool = list(string.ascii_uppercase) + # reserve S + labels_pool = [l for l in labels_pool if l != "S"] + + for _ in range(max_attempts): + anchors = sample_perimeter_anchors( + rng, 4, width, height, margin=perimeter_margin, + ) + if anchors is None: + continue + + # Pair anchor i with anchor i+2 (opposite-side pairing). + pairs = [(anchors[0], anchors[2]), (anchors[1], anchors[3])] + rng.shuffle(pairs) + + polylines = [] + ok = True + for (sp, sd), (ep, ed) in pairs: + poly = build_anchored_curve( + rng, width, height, + start_pos=sp, start_inward=sd, + end_pos=ep, end_inward=ed, + leader_length=150.0, + step_size=rng.uniform(3.0, 3.8), + max_turn=rng.uniform(0.034, 0.044), + drift_rate=rng.uniform(0.010, 0.016), + num_waypoints=num_waypoints, + interior_margin=interior_margin, + max_steps=3200, + ) + if len(poly) < 50: + ok = False + break + polylines.append(poly) + if not ok: + continue + + # Crossings between the two curves (no self-crossings desired, but allow). + pair_count, pair_min_angle, pair_details = _find_crossings( + polylines[0], polylines[1] + ) + self_counts = [] + self_details_list = [] + self_min_angle = 180.0 + for p in polylines: + sc, sa, sd_ = _find_crossings(p, p, same_curve=True) + self_counts.append(sc) + self_details_list.append(sd_) + self_min_angle = min(self_min_angle, sa) + + total_cross = pair_count + sum(self_counts) + if pair_count < min_crossings or pair_count > max_crossings: + continue + # Keep self-crossings modest so traversal order stays clear. + if sum(self_counts) > 3: + continue + + min_angle = min(pair_min_angle, self_min_angle) + if min_angle < 30.0: + continue + + # Hygiene: no close strands outside crossing regions. + all_cross_pts = _details_to_points(pair_details) + # 1% of canvas short side keeps strands visibly separated at any + # resolution while staying tractable for the sampler. + min_dist_px = 0.01 * float(min(width, height)) + if _curves_too_close(polylines[0], polylines[1], + min_dist=min_dist_px, + known_crossings=all_cross_pts): + continue + too_close_self = False + for i, p in enumerate(polylines): + if _curves_too_close(p, p, same_curve=True, + min_dist=min_dist_px, + known_crossings=_details_to_points(self_details_list[i])): + too_close_self = True + break + if too_close_self: + continue + + # Extra unlabeled distractor curves (difficulty-driven). + extra_polylines: List[np.ndarray] = [] + extra_pair_details_all: List[Dict] = [] + extra_self_details_all: List[Dict] = [] + extra_ok = True + for _extra_idx in range(num_extra_distractor_curves): + placed = False + for _attempt in range(extra_distractor_max_attempts): + extra_anchors = sample_perimeter_anchors( + rng, 2, width, height, margin=perimeter_margin, + ) + if extra_anchors is None: + continue + (esp, esd), (eep, eed) = extra_anchors[0], extra_anchors[1] + epoly = build_anchored_curve( + rng, width, height, + start_pos=esp, start_inward=esd, + end_pos=eep, end_inward=eed, + leader_length=150.0, + step_size=rng.uniform(3.0, 3.8), + max_turn=rng.uniform(0.034, 0.044), + drift_rate=rng.uniform(0.010, 0.016), + num_waypoints=num_waypoints, + interior_margin=interior_margin, + max_steps=3200, + ) + if len(epoly) < 50: + continue + # Self-crossing hygiene for the extra curve. + esc, esa, esd_det = _find_crossings(epoly, epoly, same_curve=True) + if esc > 3 or esa < 30.0: + continue + if _curves_too_close(epoly, epoly, same_curve=True, + min_dist=min_dist_px, + known_crossings=_details_to_points(esd_det)): + continue + # Cross-curve hygiene vs all existing polylines. + all_existing = polylines + extra_polylines + bad = False + candidate_pair_details: List[Dict] = [] + for existing_poly in all_existing: + ec, ea, ed_det = _find_crossings(epoly, existing_poly) + if ea < 30.0: + bad = True + break + if _curves_too_close(epoly, existing_poly, + min_dist=min_dist_px, + known_crossings=_details_to_points(ed_det)): + bad = True + break + candidate_pair_details.extend(ed_det) + if bad: + continue + extra_polylines.append(epoly) + extra_pair_details_all.extend(candidate_pair_details) + extra_self_details_all.extend(esd_det) + placed = True + break + if not placed: + extra_ok = False + break + if not extra_ok: + continue + + # Collect points to avoid for mark placement: all crossings. + avoid_pts: List[np.ndarray] = [np.array([d["px"], d["py"]]) + for d in pair_details] + for sd_ in self_details_list: + avoid_pts.extend(np.array([d["px"], d["py"]]) for d in sd_) + for d_ in extra_pair_details_all: + avoid_pts.append(np.array([d_["px"], d_["py"]])) + for d_ in extra_self_details_all: + avoid_pts.append(np.array([d_["px"], d_["py"]])) + + # Pick S-curve and which endpoint is S. + s_curve = rng.randint(0, 1) + s_at_start = rng.random() < 0.5 + other_curve = 1 - s_curve + + # Decide mark counts. Default: 10 total, 5 on S-curve, 5 on the + # distractor curve. If the caller narrows the range we split roughly + # evenly, keeping at least 2 on each curve. + total_marks = rng.randint(min_marks_total, max_marks_total) + s_marks = total_marks // 2 + s_marks = max(2, min(s_marks, total_marks - 2)) + o_marks = total_marks - s_marks + + s_poly = polylines[s_curve] + o_poly = polylines[other_curve] + + s_indices = place_marks_on_curve( + rng, s_poly, o_poly, s_marks, avoid_points=avoid_pts, + min_pixel_between_marks=min_pixel_between_marks, + extra_polylines=extra_polylines, + ) + if len(s_indices) < s_marks: + continue + + # Existing marks (pixel-space) to avoid from other curve's perspective. + existing_for_other = [(s_poly[i], i) for i in s_indices] + o_indices = place_marks_on_curve( + rng, o_poly, s_poly, o_marks, avoid_points=avoid_pts, + existing_marks=existing_for_other, + min_pixel_between_marks=min_pixel_between_marks, + extra_polylines=extra_polylines, + ) + if len(o_indices) < o_marks: + continue + + # Also avoid marks too close to the S endpoint on s_poly. + s_endpoint_pt = s_poly[0] if s_at_start else s_poly[-1] + all_mark_pts = [s_poly[i] for i in s_indices] + [o_poly[i] for i in o_indices] + if any(np.sqrt(((s_endpoint_pt - p) ** 2).sum()) < s_endpoint_excl_px + for p in all_mark_pts): + continue + + # Assign labels. Pool of letters sized to total_marks, shuffled. + pool = labels_pool[:total_marks] + rng.shuffle(pool) + s_labels = pool[:s_marks] + o_labels = pool[s_marks:s_marks + o_marks] + + # Determine traversal order on the S-curve. + # s_indices is ascending in polyline order. If s_at_start, forward; + # else reverse. + order_pairs = list(zip(s_indices, s_labels)) + if not s_at_start: + order_pairs = order_pairs[::-1] + answer_sequence = [lab for _, lab in order_pairs] + answer = ", ".join(answer_sequence) + + # Build mark info. + marks_info: List[Dict] = [] + for idx, lab in zip(s_indices, s_labels): + lx, ly = label_anchor_for_mark(s_poly, o_poly, idx, width, height) + marks_info.append({ + "label": lab, + "curve": s_curve, + "on_s_curve": True, + "polyline_index": int(idx), + "x": round(float(s_poly[idx, 0]), 2), + "y": round(float(s_poly[idx, 1]), 2), + "label_x": round(lx, 2), + "label_y": round(ly, 2), + }) + for idx, lab in zip(o_indices, o_labels): + lx, ly = label_anchor_for_mark(o_poly, s_poly, idx, width, height) + marks_info.append({ + "label": lab, + "curve": other_curve, + "on_s_curve": False, + "polyline_index": int(idx), + "x": round(float(o_poly[idx, 0]), 2), + "y": round(float(o_poly[idx, 1]), 2), + "label_x": round(lx, 2), + "label_y": round(ly, 2), + }) + + # S marker info. + s_pt = s_poly[0] if s_at_start else s_poly[-1] + # Label offset for S: away from curve tangent. + s_tan_idx = 10 if s_at_start else len(s_poly) - 11 + s_idx_for_anchor = 0 if s_at_start else len(s_poly) - 1 + slx, sly = label_anchor_for_mark(s_poly, o_poly, s_idx_for_anchor, + width, height, offset=52.0, + tangent_span=10) + + all_labels_display = sorted(s_labels + o_labels) + question = ( + "The image shows two smooth curves that cross each other. " + f"Labeled points ({', '.join(all_labels_display)}) are marked on " + "both curves. A single green 'S' marks the start endpoint of one " + "of the curves. Starting from S, follow THAT curve (the one S " + "lies on) all the way to its other endpoint, and list the labels " + "of the marked points you visit in order, separated by commas " + "(for example: B, D, A). Ignore any labeled points that lie on " + "the OTHER curve. Provide your final answer enclosed in " + "... tags." + ) + + all_polylines = polylines + extra_polylines + return { + "width": width, + "height": height, + "polylines": all_polylines, + "num_extra_distractor_curves": int(len(extra_polylines)), + "s_curve": int(s_curve), + "s_at_start": bool(s_at_start), + "s_point": { + "x": round(float(s_pt[0]), 2), + "y": round(float(s_pt[1]), 2), + "label_x": round(float(slx), 2), + "label_y": round(float(sly), 2), + }, + "marks": marks_info, + "num_marks_total": total_marks, + "num_marks_on_s": s_marks, + "num_pair_crossings": int(pair_count), + "num_self_crossings": int(sum(self_counts)), + "question": question, + "answer": answer, + } + return None + + +# ── Rendering ────────────────────────────────────────────────────── + +def render_instance(out_path: Path, record: Dict, noise_seed: int) -> None: + width = int(record["width"]) + height = int(record["height"]) + polylines = record["polylines"] + marks = record["marks"] + s_point = record["s_point"] + + fig = plt.figure(figsize=(width / 100, height / 100), dpi=100) + ax = fig.add_axes([0, 0, 1, 1]) + ax.set_xlim(0, width) + ax.set_ylim(height, 0) + ax.axis("off") + ax.set_facecolor("#f3efe8") + + nrng = np.random.default_rng(noise_seed) + noise = nrng.normal(0.0, 1.0, size=(height, width)) + noise = (noise - noise.min()) / max(noise.max() - noise.min(), 1e-6) + ax.imshow(noise, cmap="Greys", alpha=0.06, extent=(0, width, height, 0), + interpolation="bilinear") + + for i, poly in enumerate(polylines): + ax.plot(poly[:, 0], poly[:, 1], + color=LINE_COLOR, linewidth=2.8, + solid_capstyle="round", solid_joinstyle="round", + zorder=2.0 + i * 0.05) + + # Subtle dots at all four endpoints. + for poly in polylines: + for ep in (poly[0], poly[-1]): + ax.scatter([ep[0]], [ep[1]], s=30, + facecolors="#f3efe8", edgecolors="#999", + linewidths=1.0, zorder=3.2) + + # Marks with offset labels. + for i, mark in enumerate(marks): + x, y = mark["x"], mark["y"] + lx = mark.get("label_x", x) + ly = mark.get("label_y", y - 22) + color = MARK_COLORS[i % len(MARK_COLORS)] + ax.plot([x, lx], [y, ly], + color="#9b7b56", linewidth=1.0, alpha=0.7, + linestyle=(0, (2.0, 3.2)), zorder=3.8) + ax.scatter([x], [y], s=120, facecolors=color, edgecolors="white", + linewidths=1.5, zorder=4.0) + ax.text(lx, ly, mark["label"], + fontsize=18, fontweight="bold", color=color, + ha="center", va="center", zorder=5, + bbox=dict(facecolor="#f3efe8", edgecolor=color, + boxstyle="round,pad=0.25", alpha=0.95, linewidth=1.2)) + + # S marker on the S-curve endpoint — green, larger. + sx, sy = s_point["x"], s_point["y"] + slx, sly = s_point["label_x"], s_point["label_y"] + ax.plot([sx, slx], [sy, sly], + color=S_COLOR, linewidth=1.2, alpha=0.8, + linestyle=(0, (2.0, 3.2)), zorder=4.2) + ax.scatter([sx], [sy], s=180, facecolors=S_COLOR, edgecolors="white", + linewidths=2.0, zorder=5.0) + ax.text(slx, sly, "S", + fontsize=20, fontweight="bold", color=S_COLOR, + ha="center", va="center", zorder=6, + bbox=dict(facecolor="#f3efe8", edgecolor=S_COLOR, + boxstyle="round,pad=0.3", alpha=0.95, linewidth=1.5)) + + fig.savefig(out_path, dpi=100, bbox_inches="tight", pad_inches=0) + plt.close(fig) + + +# ── Main ─────────────────────────────────────────────────────────── + +def _polyline_to_list(p: np.ndarray) -> List[List[float]]: + return [[round(float(x), 2), round(float(y), 2)] for x, y in p] + + +def main(): + parser = argparse.ArgumentParser() + parser.add_argument("--output-root", type=Path, required=True) + parser.add_argument("--count", type=int, default=6) + parser.add_argument("--seed", type=int, default=42) + parser.add_argument("--width", type=int, default=1024) + parser.add_argument("--height", type=int, default=1024) + parser.add_argument("--min-marks", type=int, default=10) + parser.add_argument("--max-marks", type=int, default=10) + parser.add_argument("--min-crossings", type=int, default=7) + parser.add_argument("--max-crossings", type=int, default=25) + parser.add_argument("--difficulty", type=int, default=5, + help="Integer difficulty >=0; scales crossings and marks.") + parser.add_argument("--workers", type=int, default=8) + args = parser.parse_args() + + d = max(0, int(args.difficulty)) + # Defaults (used whether difficulty override fires or not). + min_pixel_between_marks = 90.0 + s_endpoint_excl_px = 90.0 + num_waypoints = 5 + num_extra_distractor_curves = 0 + if d > 0: + args.min_crossings = 4 + d + args.max_crossings = 6 + 2 * d + s_marks = 4 + d // 2 + o_marks = 4 + d // 2 + args.min_marks = s_marks + o_marks + args.max_marks = s_marks + o_marks + num_waypoints = 5 + d + min_pixel_between_marks = float(max(55, 90 - 5 * d)) + s_endpoint_excl_px = float(max(50, 90 - 5 * d)) + num_extra_distractor_curves = d // 3 + + # Canvas scaling with difficulty (sqrt of mark-density growth). + N_d = 8 + 4 * d + N_0 = 8 + s_scale = math.sqrt(max(1.0, N_d / N_0)) + args.width = int(round(args.width * s_scale)) + args.height = int(round(args.height * s_scale)) + + out_root = args.output_root + img_dir = out_root / "images" + img_dir.mkdir(parents=True, exist_ok=True) + + rng = random.Random(args.seed) + records = [] + + sys.path.insert(0, str(Path(__file__).resolve().parents[3])) + from _sample_pool import parallel_sample_records # noqa: E402 + + def _attempt(rng): + return sample_instance( + rng, args.width, args.height, + min_marks_total=args.min_marks, + max_marks_total=args.max_marks, + min_crossings=args.min_crossings, + max_crossings=args.max_crossings, + min_pixel_between_marks=min_pixel_between_marks, + s_endpoint_excl_px=s_endpoint_excl_px, + num_waypoints=num_waypoints, + num_extra_distractor_curves=num_extra_distractor_curves, + max_attempts=50, + ) + + records_raw = parallel_sample_records( + _attempt, count=args.count, workers=args.workers, + seed_base=args.seed, + ) + rng_render = random.Random(args.seed ^ 0xA5A5) + for idx, record in enumerate(records_raw): + name = f"traverse_ordering_{idx:05d}.png" + ns = rng_render.randint(0, 10**9) + render_instance(img_dir / name, record, noise_seed=ns) + polylines = record.pop("polylines") + record["polylines"] = [_polyline_to_list(p) for p in polylines] + record["image"] = f"images/{name}" + records.append(record) + print(f" {len(records)}/{args.count} valid samples (workers={args.workers})") + + with (out_root / "annotations.jsonl").open("w") as fh: + for r in records: + fh.write(json.dumps(r) + "\n") + + data_json = { + "task": "traverse_ordering", + "category": "sequential_traversal", + "count": len(records), + "items": [ + {"image": r["image"], "question": r["question"], "answer": r["answer"]} + for r in records + ], + } + (out_root / "data.json").write_text(json.dumps(data_json, indent=2)) + print(f"Saved {len(records)} items to {out_root}") + + +if __name__ == "__main__": + main() diff --git a/code/sequential_traversal/traverse_ordering/data.json b/code/sequential_traversal/traverse_ordering/data.json new file mode 100644 index 0000000000000000000000000000000000000000..dab26777a167773f2f294099a46a15e4a5136be9 --- /dev/null +++ b/code/sequential_traversal/traverse_ordering/data.json @@ -0,0 +1,32 @@ +{ + "task": "traverse_ordering", + "category": "sequential_traversal", + "count": 5, + "items": [ + { + "image": "images/traverse_ordering_00000.png", + "question": "The image shows two curves or strands that cross each other. Several labeled points are marked on the two curves combined. A green 'S' marks the start endpoint of one of the curves. Starting from S, follow THAT curve (the one S lies on) all the way to its other endpoint, and list the labels of the marked points you visit in order, separated by commas. Ignore any labeled points that lie on the OTHER curve. Provide your final answer enclosed in ... tags.", + "answer": "L, G, H, F, C, K" + }, + { + "image": "images/traverse_ordering_00001.png", + "question": "The image shows two curves or strands that cross each other. Several labeled points are marked on the two curves combined. A green 'S' marks the start endpoint of one of the curves. Starting from S, follow THAT curve (the one S lies on) all the way to its other endpoint, and list the labels of the marked points you visit in order, separated by commas. Ignore any labeled points that lie on the OTHER curve. Provide your final answer enclosed in ... tags.", + "answer": "J, I, A, L, B, E" + }, + { + "image": "images/traverse_ordering_00002.png", + "question": "The image shows two curves or strands that cross each other. Several labeled points are marked on the two curves combined. A green 'S' marks the start endpoint of one of the curves. Starting from S, follow THAT curve (the one S lies on) all the way to its other endpoint, and list the labels of the marked points you visit in order, separated by commas. Ignore any labeled points that lie on the OTHER curve. Provide your final answer enclosed in ... tags.", + "answer": "I, A, E, F, C, H" + }, + { + "image": "images/traverse_ordering_00003.png", + "question": "The image shows two curves or strands that cross each other. Several labeled points are marked on the two curves combined. A green 'S' marks the start endpoint of one of the curves. Starting from S, follow THAT curve (the one S lies on) all the way to its other endpoint, and list the labels of the marked points you visit in order, separated by commas. Ignore any labeled points that lie on the OTHER curve. Provide your final answer enclosed in ... tags.", + "answer": "B, H, D, A, F, G" + }, + { + "image": "images/traverse_ordering_00004.png", + "question": "The image shows two curves or strands that cross each other. Several labeled points are marked on the two curves combined. A green 'S' marks the start endpoint of one of the curves. Starting from S, follow THAT curve (the one S lies on) all the way to its other endpoint, and list the labels of the marked points you visit in order, separated by commas. Ignore any labeled points that lie on the OTHER curve. Provide your final answer enclosed in ... tags.", + "answer": "D, F, A, C, B, L" + } + ] +} \ No newline at end of file diff --git a/code/visual_attribute_transfer/constellation_match_count/creation.py b/code/visual_attribute_transfer/constellation_match_count/creation.py new file mode 100644 index 0000000000000000000000000000000000000000..de08bbd5368b87eefb40e6db63e88eea01a1be73 --- /dev/null +++ b/code/visual_attribute_transfer/constellation_match_count/creation.py @@ -0,0 +1,560 @@ +"""Generate constellation_match_count samples. + +Each sample produces a SINGLE side-by-side image: Template on the left, +Field on the right, separated by a thin divider. + +The Template panel shows the reference constellation, and the Field shows +30-60 white "star" dots (with planted copies of the template among +distractor stars). + +The task: count how many translated copies of the template pattern occur +in the Field (same relative offsets within a tolerance, no rotation/reflection). + +The generator performs rejection sampling in TWO directions: + * it places the desired number of planted template instances, and + * it scatters additional non-template "distractor" stars while rejecting + any configuration that would accidentally form another template match. +After all dots are placed the final match-count is verified; a mismatch +against the intended count causes the sample to be regenerated. +""" + +from __future__ import annotations + +import argparse +import io +import json +import math +import random +from pathlib import Path +from typing import List, Tuple + +import matplotlib +matplotlib.use("Agg") +import matplotlib.pyplot as plt +import numpy as np +from PIL import Image +from tqdm import tqdm + + +QUESTION = ( + "This image has two panels separated by a thin vertical divider. " + "The left panel shows the Template: a small constellation of white dots. " + "The right panel shows the Field: a larger scene with many white stars. " + "The two panels are drawn at the SAME pixel scale and the dots are the " + "SAME pixel size. Count how many copies of the Template pattern appear " + "in the Field. A copy may be rotated by a small angle (up to about 20 " + "degrees in either direction) and individual dot positions may be jittered " + "slightly, but the overall pattern of relative dot positions must be " + "preserved. The patterns may be rotated by a small angle. No mirroring " + "or scaling. Report the count as a non-negative integer. " + "Provide your final answer enclosed in ... tags." +) + +# Per-copy rotation angle drawn from [-MAX_ANGLE_DEG, +MAX_ANGLE_DEG]. +# Chosen large enough to break translation-only Hausdorff matching (which +# succeeded at 100% on the unrotated version) but small enough that human +# viewers readily identify each copy as the same pattern. +MAX_ANGLE_DEG = 20.0 +# Candidate angles sampled in the matching verifier. +ANGLE_SEARCH_STEP = 2.0 + + +# Shared dot size (matplotlib scatter ``s`` is marker area in points^2). +# Same value used for both the Template panel and the Field so the model has +# a direct pixel-to-pixel correspondence. +DOT_SIZE = 130 + + +# --------------------------------------------------------------------------- +# Template generation +# --------------------------------------------------------------------------- + + +def random_template(rng: random.Random) -> np.ndarray: + """Build a template of exactly 5 dots with distinct relative offsets. + + Returned as an (n, 2) ndarray of offsets relative to the first dot, + so the first row is always (0, 0). + """ + n = 5 + # Draw points on a coarse integer grid so that offsets are well separated, + # then jitter slightly to keep the pattern visually non-lattice. + scale = rng.uniform(36.0, 54.0) # pixel scale of the template + while True: + grid_pts = set() + grid_pts.add((0, 0)) + while len(grid_pts) < n: + gx = rng.randint(-3, 3) + gy = rng.randint(-3, 3) + grid_pts.add((gx, gy)) + pts_grid = list(grid_pts) + pts = [] + for gx, gy in pts_grid: + jitter_x = rng.uniform(-4.0, 4.0) + jitter_y = rng.uniform(-4.0, 4.0) + pts.append((gx * scale + jitter_x, gy * scale + jitter_y)) + arr = np.asarray(pts, dtype=np.float64) + # Normalise so first point is origin + arr = arr - arr[0] + # Reject degenerate patterns: ensure min pairwise distance is comfortable, + # and that the pattern occupies a reasonable bounding box. + diffs = arr[:, None, :] - arr[None, :, :] + dists = np.linalg.norm(diffs, axis=-1) + np.fill_diagonal(dists, np.inf) + if dists.min() < 28.0: + continue + bbox = arr.max(axis=0) - arr.min(axis=0) + if bbox[0] < 40.0 or bbox[1] < 40.0: + continue + if bbox[0] > 260.0 or bbox[1] > 260.0: + continue + return arr + + +# --------------------------------------------------------------------------- +# Template matching +# --------------------------------------------------------------------------- + + +def _rot_matrix(theta_deg: float) -> np.ndarray: + t = np.deg2rad(theta_deg) + c, s = np.cos(t), np.sin(t) + return np.array([[c, -s], [s, c]]) + + +def count_template_matches(stars: np.ndarray, template: np.ndarray, + tol: float, + max_angle_deg: float = MAX_ANGLE_DEG, + angle_step_deg: float = ANGLE_SEARCH_STEP) -> int: + """Count distinct (translation, rotation) placements such that every + (R(theta) @ template[i] + T) has a star within `tol` pixels, where T is + a candidate translation and theta is a rotation angle in + [-max_angle_deg, +max_angle_deg]. We deduplicate by translation only: + two matches are the same if their anchor translations differ by less + than `tol`. + + stars: (S, 2) array, template: (n, 2) with template[0] = (0,0). + """ + if len(stars) == 0: + return 0 + n = template.shape[0] + tol_sq = tol * tol + angles = np.arange(-max_angle_deg, max_angle_deg + 1e-6, angle_step_deg) + + found_translations: List[np.ndarray] = [] + for anchor in stars: + anchor_matched = False + for theta in angles: + R = _rot_matrix(float(theta)) + rotated = template @ R.T # (n, 2), first row still (0, 0) + ok = True + for k in range(1, n): + target = anchor + rotated[k] + d2 = np.sum((stars - target) ** 2, axis=1) + if d2.min() > tol_sq: + ok = False + break + if ok: + anchor_matched = True + break + if not anchor_matched: + continue + # Deduplicate by anchor translation + is_new = True + for t in found_translations: + if np.sum((t - anchor) ** 2) < tol_sq: + is_new = False + break + if is_new: + found_translations.append(anchor.copy()) + return len(found_translations) + + +# --------------------------------------------------------------------------- +# Sample building +# --------------------------------------------------------------------------- + + +def build_sample(rng: random.Random, width: int, height: int, + target_matches: int, total_stars: int, + tol: float) -> Tuple[np.ndarray, np.ndarray, int]: + """Return (stars, template, realised_matches). Raises RuntimeError if + it could not construct the intended configuration after many tries.""" + + for attempt in range(30): + template = random_template(rng) + n_tmpl = template.shape[0] + + # Margins for anchor placement so full template stays inside + tmin = template.min(axis=0) + tmax = template.max(axis=0) + pad = 60.0 + anchor_xmin = pad - tmin[0] + anchor_xmax = width - pad - tmax[0] + anchor_ymin = pad - tmin[1] + anchor_ymax = height - pad - tmax[1] + if anchor_xmax - anchor_xmin < 100 or anchor_ymax - anchor_ymin < 100: + continue + + placed_stars: List[np.ndarray] = [] + # Separation so planted instances don't overlap each other excessively, + # and so the template anchors are themselves far enough apart to be + # counted as distinct translations. + min_anchor_sep = float(np.linalg.norm(tmax - tmin)) + 40.0 + + # 1. Plant template instances. Each planted copy carries a per-copy + # rotation in [-MAX_ANGLE_DEG, +MAX_ANGLE_DEG]. This breaks + # translation-only matching (Hausdorff + centroid-anchored offsets). + anchors: List[np.ndarray] = [] + planted_angles: List[float] = [] + for _ in range(target_matches): + ok = False + for _try in range(400): + ax = rng.uniform(anchor_xmin, anchor_xmax) + ay = rng.uniform(anchor_ymin, anchor_ymax) + theta = rng.uniform(-MAX_ANGLE_DEG, MAX_ANGLE_DEG) + candidate = np.array([ax, ay]) + too_close = False + for a in anchors: + if np.linalg.norm(candidate - a) < min_anchor_sep: + too_close = True + break + if too_close: + continue + R = _rot_matrix(theta) + rotated_template = template @ R.T + # Check rotated bounding box still inside the panel padding. + rtmin = rotated_template.min(axis=0) + rtmax = rotated_template.max(axis=0) + if (candidate[0] + rtmin[0] < pad or + candidate[0] + rtmax[0] > width - pad or + candidate[1] + rtmin[1] < pad or + candidate[1] + rtmax[1] > height - pad): + continue + # Try adding this instance's dots while ensuring each new dot + # is at least `tol * 2.5` from all existing dots. + new_dots = [candidate + rotated_template[k] for k in range(n_tmpl)] + clash = False + for nd in new_dots: + for ex in placed_stars: + if np.linalg.norm(nd - ex) < tol * 2.5: + clash = True + break + if clash: + break + if clash: + continue + anchors.append(candidate) + planted_angles.append(float(theta)) + placed_stars.extend(new_dots) + ok = True + break + if not ok: + break + if len(anchors) != target_matches: + continue + + # 2. Scatter distractor stars, rejection-sample to avoid new matches + distractors_needed = total_stars - len(placed_stars) + if distractors_needed < 0: + continue + fail = False + for _d in range(distractors_needed): + added = False + for _try in range(500): + dx = rng.uniform(pad, width - pad) + dy = rng.uniform(pad, height - pad) + cand = np.array([dx, dy]) + # Keep distractors from overlapping existing dots + too_close = False + for ex in placed_stars: + if np.linalg.norm(cand - ex) < 22.0: + too_close = True + break + if too_close: + continue + # Tentatively add and verify match count is unchanged + trial = np.asarray(placed_stars + [cand]) + count = count_template_matches(trial, template, tol) + if count != target_matches: + continue + placed_stars.append(cand) + added = True + break + if not added: + fail = True + break + if fail: + continue + + final_stars = np.asarray(placed_stars) + realised = count_template_matches(final_stars, template, tol) + if realised == target_matches: + return final_stars, template, realised + + raise RuntimeError("Failed to build sample after many attempts") + + +# --------------------------------------------------------------------------- +# Rendering - two separate images per sample +# --------------------------------------------------------------------------- + + +def _add_dust(ax, rng: random.Random, width: int, height: int, + n_dust: int = 800) -> None: + dust_rng = np.random.default_rng(rng.randint(0, 2**31 - 1)) + dx = dust_rng.uniform(0, width, size=n_dust) + dy = dust_rng.uniform(0, height, size=n_dust) + ds = dust_rng.uniform(0.3, 2.5, size=n_dust) + ax.scatter(dx, dy, s=ds, c="#1a2238", alpha=0.45, linewidths=0, zorder=1) + + +def render_field(width: int, height: int, + stars: np.ndarray, rng: random.Random) -> Image.Image: + """Render the field image: dark sky with scattered white stars, no + template overlay. Returns a PIL Image.""" + fig = plt.figure(figsize=(width / 100, height / 100), dpi=100, + facecolor="#0a1020") + ax = fig.add_axes([0, 0, 1, 1]) + ax.set_xlim(0, width) + ax.set_ylim(height, 0) + ax.axis("off") + ax.set_facecolor("#0a1020") + + _add_dust(ax, rng, width, height) + + # All stars at a single, larger pixel size so the model can compare + # dot-to-dot spacings directly between Template and Field. + ax.scatter(stars[:, 0], stars[:, 1], s=DOT_SIZE, c="white", alpha=1.0, + linewidths=0, zorder=3) + + buf = io.BytesIO() + fig.savefig(buf, format="png", dpi=100, bbox_inches="tight", pad_inches=0, + facecolor=fig.get_facecolor()) + plt.close(fig) + buf.seek(0) + return Image.open(buf).convert("RGB") + + +def render_template(template: np.ndarray, + rng: random.Random) -> Image.Image: + """Render the template panel at the SAME pixel scale as the field. + + The canvas is a tight bounding box around the template dots plus a + small margin, so pixel-distances between template dots equal pixel- + distances between the corresponding dots in the field (1:1 scale). + A short header strip labels the panel as TEMPLATE. + Returns a PIL Image. + """ + from matplotlib.patches import Rectangle + + tpl_min = template.min(axis=0) + tpl_max = template.max(axis=0) + span_x = float(tpl_max[0] - tpl_min[0]) + span_y = float(tpl_max[1] - tpl_min[1]) + + margin = 50.0 # empty space around the dots + header_h = 34.0 # top strip for the TEMPLATE label + min_width = 240.0 # keep the header legible for narrow patterns + + content_w = span_x + 2 * margin + content_h = span_y + 2 * margin + canvas_w = max(content_w, min_width) + canvas_h = header_h + content_h + + # Offset that places tpl_min at (margin_x, header_h + margin) with the + # dots horizontally centred inside the canvas. + margin_x = (canvas_w - span_x) / 2.0 + ox = margin_x - tpl_min[0] + oy = (header_h + margin) - tpl_min[1] + disp = template + np.array([ox, oy]) + + fig = plt.figure(figsize=(canvas_w / 100, canvas_h / 100), dpi=100, + facecolor="#070b14") + ax = fig.add_axes([0, 0, 1, 1]) + ax.set_xlim(0, canvas_w) + ax.set_ylim(canvas_h, 0) + ax.axis("off") + ax.set_facecolor("#070b14") + + _add_dust(ax, rng, int(canvas_w), int(canvas_h), + n_dust=max(40, int(canvas_w * canvas_h / 1800))) + + # Header strip + ax.add_patch(Rectangle((0, 0), canvas_w, header_h, + facecolor="#11182a", edgecolor="none", zorder=4)) + ax.plot([0, canvas_w], [header_h, header_h], + color="#2d3a5a", linewidth=1.0, zorder=5) + ax.text(canvas_w / 2, header_h / 2, "TEMPLATE", + color="#cfe0ff", fontsize=14, fontweight="bold", + ha="center", va="center", zorder=6, + family="DejaVu Sans") + + # Subtle border around the whole canvas + ax.add_patch(Rectangle((1.5, 1.5), canvas_w - 3, canvas_h - 3, + facecolor="none", edgecolor="#7aa6ff", + linewidth=2.0, zorder=6)) + + ax.scatter(disp[:, 0], disp[:, 1], s=DOT_SIZE, c="white", + linewidths=0, zorder=7) + + buf = io.BytesIO() + fig.savefig(buf, format="png", dpi=100, bbox_inches=None, pad_inches=0, + facecolor=fig.get_facecolor()) + plt.close(fig) + buf.seek(0) + return Image.open(buf).convert("RGB") + + +BG_COLOR = "#0a1020" +DIVIDER_WIDTH = 3 +DIVIDER_COLOR = (51, 51, 85) # #333355 + + +def render_combined(out_path: Path, template_img: Image.Image, + field_img: Image.Image) -> None: + """Concatenate template (left) and field (right) with a thin divider.""" + bg = tuple(int(BG_COLOR.lstrip("#")[i:i+2], 16) for i in (0, 2, 4)) + tw, th = template_img.size + fw, fh = field_img.size + combined_h = max(th, fh) + combined_w = tw + DIVIDER_WIDTH + fw + combined = Image.new("RGB", (combined_w, combined_h), bg) + # Vertically centre template + t_y = (combined_h - th) // 2 + combined.paste(template_img, (0, t_y)) + # Divider + for x in range(tw, tw + DIVIDER_WIDTH): + for y in range(combined_h): + combined.putpixel((x, y), DIVIDER_COLOR) + # Field + f_y = (combined_h - fh) // 2 + combined.paste(field_img, (tw + DIVIDER_WIDTH, f_y)) + combined.save(out_path) + + +# --------------------------------------------------------------------------- +# Main +# --------------------------------------------------------------------------- + + +def main() -> None: + parser = argparse.ArgumentParser() + parser.add_argument("--output-root", type=Path, required=True) + parser.add_argument("--count", type=int, default=30) + parser.add_argument("--seed", type=int, default=0) + parser.add_argument("--width", type=int, default=900) + parser.add_argument("--height", type=int, default=900) + parser.add_argument("--tolerance", type=float, default=10.0, + help="Position tolerance (pixels) for a template match") + parser.add_argument("--difficulty", type=int, default=5, + help="Integer difficulty >=0; scales copies, angle, field stars.") + args = parser.parse_args() + + d = max(0, int(args.difficulty)) + # Canvas scaling: N_d = 50 + 5*d, N_0 = 50 + N_d = 50 + 5 * d + N_0 = 50 + s = math.sqrt(max(1.0, N_d / N_0)) + args.width = int(round(args.width * s)) + args.height = int(round(args.height * s)) + + if d > 0: + _min_copies = 5 + _max_copies = 5 + d + # Override module-level MAX_ANGLE_DEG so both the builder and the + # verifier see the same scaled max angle. + global MAX_ANGLE_DEG + MAX_ANGLE_DEG = float(min(10, d)) + _field_stars_target = 50 + 5 * d + else: + _min_copies = 5 + _max_copies = 5 + _field_stars_target = 50 + + out_root: Path = args.output_root + img_dir = out_root / "images" + img_dir.mkdir(parents=True, exist_ok=True) + + ann_path = out_root / "annotations.jsonl" + master_rng = random.Random(args.seed) + + # Force evenly-spaced answers across [_min_copies, _max_copies]. + if args.count > 1: + plan = [ + int(round(_min_copies + i * (_max_copies - _min_copies) / (args.count - 1))) + for i in range(args.count) + ] + else: + plan = [_min_copies] + print(f"forced constellation match counts: {plan}") + + records = [] + with ann_path.open("w") as f: + for i in tqdm(range(args.count), desc="constellation_match_count"): + target = plan[i] + # Total star count: default 30-60, scaled a bit with target so high-match + # images stay legible. With difficulty override, use _field_stars_target. + if _field_stars_target is not None: + base_total = master_rng.randint(_field_stars_target, + _field_stars_target + 15) + total_stars = max(base_total, target * 5 + master_rng.randint(8, 18)) + else: + base_total = master_rng.randint(30, 55) + total_stars = max(base_total, target * 5 + master_rng.randint(8, 18)) + total_stars = min(total_stars, 60) + + # Retry loop in case build_sample can't meet constraints + built = False + for retry in range(12): + sub_seed = master_rng.randint(0, 2**31 - 1) + sub_rng = random.Random(sub_seed) + try: + stars, template, realised = build_sample( + sub_rng, args.width, args.height, + target_matches=target, + total_stars=total_stars, + tol=args.tolerance, + ) + except RuntimeError: + continue + built = True + break + if not built: + raise RuntimeError(f"sample {i} could not be generated") + + img_name = f"constellation_match_count_{i:05d}.png" + tpl_img = render_template(template, random.Random(sub_seed + 2)) + fld_img = render_field(args.width, args.height, + stars, random.Random(sub_seed + 1)) + render_combined(img_dir / img_name, tpl_img, fld_img) + + rec = { + "image": f"images/{img_name}", + "question": QUESTION, + "answer": realised, + "num_template_dots": int(template.shape[0]), + "total_stars": int(stars.shape[0]), + "num_matches": realised, + "metadata": { + "seed": sub_seed, + "tolerance": args.tolerance, + "template_offsets": template.tolist(), + }, + } + f.write(json.dumps(rec) + "\n") + f.flush() + records.append(rec) + + data_json = { + "task": "constellation_match_count", + "category": "visual_attribute_transfer", + "count": len(records), + "items": records, + } + (out_root / "data.json").write_text(json.dumps(data_json, indent=2)) + print(f"Saved to {out_root}") + + +if __name__ == "__main__": + main() diff --git a/code/visual_attribute_transfer/constellation_match_count/data.json b/code/visual_attribute_transfer/constellation_match_count/data.json new file mode 100644 index 0000000000000000000000000000000000000000..207ee5443c0854ac5f7864e01b9a7db0883e9ae7 --- /dev/null +++ b/code/visual_attribute_transfer/constellation_match_count/data.json @@ -0,0 +1,177 @@ +{ + "task": "constellation_match_count", + "category": "visual_attribute_transfer", + "count": 5, + "items": [ + { + "image": "images/constellation_match_count_00000.png", + "question": "This image has two panels separated by a thin vertical divider. The left panel shows the Template: a small constellation of marker points. The right panel shows the Field: a larger scene with many marker points. Both panels are drawn at the SAME pixel scale. Count how many copies of the Template pattern appear in the Field. A copy may be rotated by a small angle (up to about 20\u00b0) and individual point positions may be jittered slightly, but the overall pattern of relative positions must be preserved. No mirroring or scaling. Report the count as a non-negative integer. Provide your final answer enclosed in ... tags.", + "answer": 5, + "num_template_dots": 5, + "total_stars": 87, + "num_matches": 5, + "metadata": { + "seed": 173879092, + "tolerance": 10.0, + "template_offsets": [ + [ + 0.0, + 0.0 + ], + [ + -3.4616725711160994, + -46.954577705132806 + ], + [ + -93.61996391282639, + -185.43041481111788 + ], + [ + -136.61615800989028, + 43.50301029449731 + ], + [ + -46.03632956018431, + 43.912698774913856 + ] + ] + } + }, + { + "image": "images/constellation_match_count_00001.png", + "question": "This image has two panels separated by a thin vertical divider. The left panel shows the Template: a small constellation of marker points. The right panel shows the Field: a larger scene with many marker points. Both panels are drawn at the SAME pixel scale. Count how many copies of the Template pattern appear in the Field. A copy may be rotated by a small angle (up to about 20\u00b0) and individual point positions may be jittered slightly, but the overall pattern of relative positions must be preserved. No mirroring or scaling. Report the count as a non-negative integer. Provide your final answer enclosed in ... tags.", + "answer": 6, + "num_template_dots": 5, + "total_stars": 83, + "num_matches": 6, + "metadata": { + "seed": 2087043557, + "tolerance": 10.0, + "template_offsets": [ + [ + 0.0, + 0.0 + ], + [ + 41.471863900170575, + -3.5227577023131147 + ], + [ + 5.448720316866826, + 115.88056826599409 + ], + [ + 121.46995054662344, + 79.74046043527612 + ], + [ + 80.93273678646665, + 200.86023062793714 + ] + ] + } + }, + { + "image": "images/constellation_match_count_00002.png", + "question": "This image has two panels separated by a thin vertical divider. The left panel shows the Template: a small constellation of marker points. The right panel shows the Field: a larger scene with many marker points. Both panels are drawn at the SAME pixel scale. Count how many copies of the Template pattern appear in the Field. A copy may be rotated by a small angle (up to about 20\u00b0) and individual point positions may be jittered slightly, but the overall pattern of relative positions must be preserved. No mirroring or scaling. Report the count as a non-negative integer. Provide your final answer enclosed in ... tags.", + "answer": 8, + "num_template_dots": 5, + "total_stars": 87, + "num_matches": 8, + "metadata": { + "seed": 2046968324, + "tolerance": 10.0, + "template_offsets": [ + [ + 0.0, + 0.0 + ], + [ + 77.91193064942368, + 33.84202775872543 + ], + [ + 2.8003529401257765, + 34.648794353110745 + ], + [ + 72.87157848243598, + 111.78624537905125 + ], + [ + 111.2982677920412, + -75.27606286426611 + ] + ] + } + }, + { + "image": "images/constellation_match_count_00003.png", + "question": "This image has two panels separated by a thin vertical divider. The left panel shows the Template: a small constellation of marker points. The right panel shows the Field: a larger scene with many marker points. Both panels are drawn at the SAME pixel scale. Count how many copies of the Template pattern appear in the Field. A copy may be rotated by a small angle (up to about 20\u00b0) and individual point positions may be jittered slightly, but the overall pattern of relative positions must be preserved. No mirroring or scaling. Report the count as a non-negative integer. Provide your final answer enclosed in ... tags.", + "answer": 9, + "num_template_dots": 5, + "total_stars": 86, + "num_matches": 9, + "metadata": { + "seed": 938204377, + "tolerance": 10.0, + "template_offsets": [ + [ + 0.0, + 0.0 + ], + [ + 47.632343557594254, + -44.75983700245754 + ], + [ + -46.37957980821564, + 0.3673736111899366 + ], + [ + 96.88568769916276, + 146.00365752516385 + ], + [ + 5.919548083564312, + -45.64291592166866 + ] + ] + } + }, + { + "image": "images/constellation_match_count_00004.png", + "question": "This image has two panels separated by a thin vertical divider. The left panel shows the Template: a small constellation of marker points. The right panel shows the Field: a larger scene with many marker points. Both panels are drawn at the SAME pixel scale. Count how many copies of the Template pattern appear in the Field. A copy may be rotated by a small angle (up to about 20\u00b0) and individual point positions may be jittered slightly, but the overall pattern of relative positions must be preserved. No mirroring or scaling. Report the count as a non-negative integer. Provide your final answer enclosed in ... tags.", + "answer": 10, + "num_template_dots": 5, + "total_stars": 79, + "num_matches": 10, + "metadata": { + "seed": 600203567, + "tolerance": 10.0, + "template_offsets": [ + [ + 0.0, + 0.0 + ], + [ + -50.866585971769034, + -96.4278447997325 + ], + [ + -47.43200738991797, + 51.02936959014866 + ], + [ + 46.583067551880816, + -4.411768013683243 + ], + [ + -2.2544303107114274, + -98.25764451187348 + ] + ] + } + } + ] +} \ No newline at end of file diff --git a/code/visual_attribute_transfer/constellation_match_count/image.png b/code/visual_attribute_transfer/constellation_match_count/image.png new file mode 100644 index 0000000000000000000000000000000000000000..0fcbeccb67cdbb9014bf74e2cb0d6ed2dfcefd8d --- /dev/null +++ b/code/visual_attribute_transfer/constellation_match_count/image.png @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:9188c7500fbc299bc03e1910bb92ab807d9507357e2d40e9a6b656388a36d89d +size 100608 diff --git a/code/visual_attribute_transfer/contour_silhouette_count/creation.py b/code/visual_attribute_transfer/contour_silhouette_count/creation.py new file mode 100644 index 0000000000000000000000000000000000000000..33fe213aaff5a6899fb2d5dd1c728b0280911475 --- /dev/null +++ b/code/visual_attribute_transfer/contour_silhouette_count/creation.py @@ -0,0 +1,593 @@ +"""Generate contour_silhouette_count samples. + +Each sample produces a SINGLE side-by-side image: Template on the left, +Field on the right, separated by a thin divider. + +The Template panel shows a single smooth closed blob outline (white on dark), +and the Field shows 25-35 blob outlines scattered across a dark canvas. + +Some blobs in the field are exact translated copies of the template shape; +the rest are distractors (different Fourier-descriptor blobs). The task is +to count how many field contours match the template exactly (translation only). + +Blob shapes are generated via Fourier descriptors: + r(theta) = r0 + sum_{k=1}^{K} a_k * cos(k*theta + phi_k) +""" + +from __future__ import annotations + +import argparse +import io +import json +import math +import random +from pathlib import Path +from typing import List, Tuple + +import matplotlib +matplotlib.use("Agg") +import matplotlib.pyplot as plt +import matplotlib.patches as mpatches +from matplotlib.patches import Rectangle +from matplotlib.path import Path as MplPath +import numpy as np +from PIL import Image +from scipy.spatial.distance import directed_hausdorff +from tqdm import tqdm + + +QUESTION = ( + "This image has two panels separated by a thin vertical divider. " + "The left panel shows the Template: a single closed contour. " + "The right panel shows the Field: many closed contours scattered across the " + "canvas, all at the SAME pixel scale as the Template. Count the number of " + "Field contours that are exact copies of the Template shape (translation " + "only — same size, same orientation, no rotation or mirroring) and report " + "the integer count. " + "Provide your final answer enclosed in ... tags." +) + +STROKE_WIDTH = 2.5 +BG_COLOR = "#0a1020" +N_CONTOUR_PTS = 200 # number of points to sample on each blob boundary + + +# --------------------------------------------------------------------------- +# Fourier-descriptor blob generation +# --------------------------------------------------------------------------- + + +def fourier_blob(rng: random.Random, r0: float = 45.0, + K_range: Tuple[int, int] = (4, 6), + amp_scale: float = 0.18) -> np.ndarray: + """Generate a closed blob shape via Fourier descriptors. + + Returns an (N, 2) array of (x, y) points (centred near origin). + """ + K = rng.randint(*K_range) + amplitudes = [rng.uniform(0.05, amp_scale) * r0 for _ in range(K)] + phases = [rng.uniform(0, 2 * np.pi) for _ in range(K)] + + theta = np.linspace(0, 2 * np.pi, N_CONTOUR_PTS, endpoint=False) + r = np.full_like(theta, r0) + for k_idx in range(K): + k = k_idx + 1 + r = r + amplitudes[k_idx] * np.cos(k * theta + phases[k_idx]) + + # Ensure r stays positive + r = np.clip(r, 5.0, None) + + x = r * np.cos(theta) + y = r * np.sin(theta) + pts = np.column_stack([x, y]) + + # Centre on centroid + pts -= pts.mean(axis=0) + return pts + + +def blob_hausdorff(a: np.ndarray, b: np.ndarray) -> float: + """Symmetric Hausdorff distance between two blob point sets.""" + d1 = directed_hausdorff(a, b)[0] + d2 = directed_hausdorff(b, a)[0] + return max(d1, d2) + + +def blob_bounding_radius(pts: np.ndarray) -> float: + """Max distance from centroid (assumed ~origin) to any point.""" + return float(np.max(np.linalg.norm(pts, axis=1))) + + +# --------------------------------------------------------------------------- +# Sample building +# --------------------------------------------------------------------------- + + +DISTRACTOR_HAUSDORFF_MIN = 8.0 # module-level override target (difficulty-tunable) +TEMPLATE_AMP_SCALE = 0.18 # base Fourier amplitude scale for the template blob + + +def build_sample(rng: random.Random, width: int, height: int, + target_matches: int, total_blobs: int + ) -> Tuple[np.ndarray, List[np.ndarray], List[np.ndarray], + List[Tuple[np.ndarray, float]], + List[Tuple[np.ndarray, float]], int]: + """Build a sample. Returns (template_pts, match_placements, distractor_placements, + match_centres, distractor_centres, realised_matches). + + Each placement is the blob points translated to their canvas position. + match_placements: list of (N,2) arrays (copies of template at various offsets). + distractor_placements: list of (N,2) arrays (different blob shapes). + match_centres / distractor_centres: list of (centre_xy, bounding_radius). + """ + hausdorff_thresh = DISTRACTOR_HAUSDORFF_MIN # distractors must differ by at least this much + outline_gap = 10.0 # min visible gap between any two blob outlines + + for _attempt in range(60): + # Generate template blob + r0 = rng.uniform(40.0, 50.0) + template = fourier_blob(rng, r0=r0, K_range=(4, 6), amp_scale=TEMPLATE_AMP_SCALE) + t_radius = blob_bounding_radius(template) + pad = t_radius + 20.0 + + # Track (centre, bounding_radius) for non-overlap checks. + placed_items: List[Tuple[np.ndarray, float]] = [] + + def fits(cand: np.ndarray, cand_radius: float) -> bool: + for c, r in placed_items: + if np.linalg.norm(cand - c) < (cand_radius + r + outline_gap): + return False + return True + + # 1. Place match copies (same template, same radius). + match_placements: List[np.ndarray] = [] + match_centres: List[Tuple[np.ndarray, float]] = [] + ok = True + for _ in range(target_matches): + placed = False + for _try in range(800): + cx = rng.uniform(pad, width - pad) + cy = rng.uniform(pad, height - pad) + cand = np.array([cx, cy]) + if not fits(cand, t_radius): + continue + placed_items.append((cand, t_radius)) + match_placements.append(template + cand) + match_centres.append((cand, t_radius)) + placed = True + break + if not placed: + ok = False + break + if not ok: + continue + + # 2. Place distractors with per-blob bounding radii. + n_distractors = total_blobs - target_matches + distractor_placements: List[np.ndarray] = [] + distractor_centres: List[Tuple[np.ndarray, float]] = [] + fail = False + for _ in range(n_distractors): + placed = False + for _try in range(800): + d_r0 = rng.uniform(38.0, 52.0) + distractor = fourier_blob(rng, r0=d_r0, K_range=(4, 6), + amp_scale=0.22) + if blob_hausdorff(distractor, template) < hausdorff_thresh: + continue + + d_radius = blob_bounding_radius(distractor) + d_pad = d_radius + 20.0 + if d_pad >= width / 2 or d_pad >= height / 2: + continue + cx = rng.uniform(d_pad, width - d_pad) + cy = rng.uniform(d_pad, height - d_pad) + cand = np.array([cx, cy]) + if not fits(cand, d_radius): + continue + + placed_items.append((cand, d_radius)) + distractor_placements.append(distractor + cand) + distractor_centres.append((cand, d_radius)) + placed = True + break + if not placed: + fail = True + break + + if fail: + continue + + realised = target_matches # by construction, matches are exact copies + return (template, match_placements, distractor_placements, + match_centres, distractor_centres, realised) + + raise RuntimeError("Failed to build sample after many attempts") + + +# --------------------------------------------------------------------------- +# Rendering +# --------------------------------------------------------------------------- + + +def _draw_blob_outline(ax, pts: np.ndarray, stroke_width: float, + color: str = "white", alpha: float = 1.0, + zorder: int = 3) -> None: + """Draw a closed smooth polygon outline (anti-aliased).""" + # Close the loop + closed = np.vstack([pts, pts[0:1]]) + ax.plot(closed[:, 0], closed[:, 1], color=color, linewidth=stroke_width, + alpha=alpha, solid_capstyle="round", solid_joinstyle="round", + antialiased=True, zorder=zorder) + + +LABEL_BADGE_COLOR = "#f5d76e" # warm yellow filled circle behind label letter +LABEL_TEXT_COLOR = "#0a1020" # dark text on the badge (matches BG) +LABEL_BADGE_EDGE = "#ffffff" # thin white outline so badge stands out + + +def _pick_label_position(centre: np.ndarray, bbox_r: float, + own_idx: int, + all_contours: List[np.ndarray], + placed_label_centres: List[np.ndarray], + width: int, height: int, + badge_radius: float, + other_clearance: float = 10.0, + label_clearance: float = 24.0) -> np.ndarray: + """Multi-candidate / best-score label placement. + + Samples ~16 angular anchors around the contour's bounding circle, with + several radial offsets each. Filters by hard gates (panel margin, distance + to other contours, distance to placed labels), scores remaining candidates, + and returns the argmax. Falls back by relaxing other_clearance. + """ + margin = badge_radius + 4.0 + n_angles = 16 + angles = [2 * math.pi * k / n_angles for k in range(n_angles)] + radial_offsets = [8.0, 14.0, 22.0, 30.0] + + # Build candidate list (pos, offset) + candidates: List[Tuple[np.ndarray, float]] = [] + for ang in angles: + ux, uy = math.cos(ang), math.sin(ang) + for off in radial_offsets: + r = bbox_r + off + bx = centre[0] + ux * r + by = centre[1] + uy * r + candidates.append((np.array([bx, by]), off)) + + # Other-contour points: concat all contours except own + other_pts_list = [c for k, c in enumerate(all_contours) if k != own_idx] + if other_pts_list: + other_pts = np.vstack(other_pts_list) + else: + other_pts = None + + def gate_and_score(min_other: float) -> Tuple[np.ndarray, float] | None: + best = None + best_score = -1e18 + for pos, off in candidates: + bx, by = pos[0], pos[1] + # Panel margin + if not (margin <= bx <= width - margin and + margin <= by <= height - margin): + continue + # Distance to other contour points + if other_pts is not None and len(other_pts) > 0: + other_d = float(np.min(np.linalg.norm(other_pts - pos, axis=1))) + else: + other_d = 1e9 + if other_d < min_other: + continue + # Distance to placed label centres + if placed_label_centres: + lbl_arr = np.array(placed_label_centres) + label_d = float(np.min(np.linalg.norm(lbl_arr - pos, axis=1))) + else: + label_d = 1e9 + if label_d < label_clearance: + continue + score = other_d + 0.5 * label_d - 0.4 * off + if score > best_score: + best_score = score + best = pos + return (best, best_score) if best is not None else None + + for relax in (1.0, 0.5, 0.3): + res = gate_and_score(other_clearance * relax) + if res is not None: + return res[0] + + # Last-resort: pick any in-panel candidate maximising other_d + best = None + best_d = -1.0 + for pos, off in candidates: + bx, by = pos[0], pos[1] + if not (margin <= bx <= width - margin and + margin <= by <= height - margin): + continue + if other_pts is not None and len(other_pts) > 0: + other_d = float(np.min(np.linalg.norm(other_pts - pos, axis=1))) + else: + other_d = 1e9 + if other_d > best_d: + best_d = other_d + best = pos + if best is not None: + return best + # Absolute fallback: upper-right diagonal + ux, uy = 1.0 / math.sqrt(2), -1.0 / math.sqrt(2) + return np.array([centre[0] + ux * (bbox_r + badge_radius + 6.0), + centre[1] + uy * (bbox_r + badge_radius + 6.0)]) + + +def _index_to_letters(idx: int) -> str: + """0->A, 25->Z, 26->AA, 27->AB, ...""" + letters = "" + n = idx + while True: + letters = chr(ord("A") + (n % 26)) + letters + n = n // 26 - 1 + if n < 0: + break + return letters + + +def render_field(width: int, height: int, + match_placements: List[np.ndarray], + distractor_placements: List[np.ndarray], + labels: List[str], + centres: List[Tuple[np.ndarray, float]]) -> Image.Image: + """Render the field image: dark canvas with many blob outlines plus labels. + + `labels` and `centres` are aligned by index with the concatenation + `match_placements + distractor_placements`. + Returns PIL Image. + """ + fig = plt.figure(figsize=(width / 100, height / 100), dpi=100, + facecolor=BG_COLOR) + ax = fig.add_axes([0, 0, 1, 1]) + ax.set_xlim(0, width) + ax.set_ylim(height, 0) + ax.axis("off") + ax.set_facecolor(BG_COLOR) + + all_blobs = match_placements + distractor_placements + for blob in all_blobs: + _draw_blob_outline(ax, blob, STROKE_WIDTH) + + # Letter labels removed — task is now a count, no per-contour labels needed. + + buf = io.BytesIO() + fig.savefig(buf, format="png", dpi=100, bbox_inches="tight", pad_inches=0, + facecolor=fig.get_facecolor()) + plt.close(fig) + buf.seek(0) + return Image.open(buf).convert("RGB") + + +def render_template(template: np.ndarray) -> Image.Image: + """Render the template panel at the SAME pixel scale as the field. + + Tight bounding box + margin, with a TEMPLATE header strip. + Returns a PIL Image. + """ + tpl_min = template.min(axis=0) + tpl_max = template.max(axis=0) + span_x = float(tpl_max[0] - tpl_min[0]) + span_y = float(tpl_max[1] - tpl_min[1]) + + margin = 40.0 + header_h = 34.0 + min_width = 200.0 + + content_w = span_x + 2 * margin + content_h = span_y + 2 * margin + canvas_w = max(content_w, min_width) + canvas_h = header_h + content_h + + # Offset to centre blob in canvas + margin_x = (canvas_w - span_x) / 2.0 + ox = margin_x - tpl_min[0] + oy = (header_h + margin) - tpl_min[1] + disp = template + np.array([ox, oy]) + + fig = plt.figure(figsize=(canvas_w / 100, canvas_h / 100), dpi=100, + facecolor="#070b14") + ax = fig.add_axes([0, 0, 1, 1]) + ax.set_xlim(0, canvas_w) + ax.set_ylim(canvas_h, 0) + ax.axis("off") + ax.set_facecolor("#070b14") + + # Header strip + ax.add_patch(Rectangle((0, 0), canvas_w, header_h, + facecolor="#11182a", edgecolor="none", zorder=4)) + ax.plot([0, canvas_w], [header_h, header_h], + color="#2d3a5a", linewidth=1.0, zorder=5) + ax.text(canvas_w / 2, header_h / 2, "TEMPLATE", + color="#cfe0ff", fontsize=14, fontweight="bold", + ha="center", va="center", zorder=6, + family="DejaVu Sans") + + # Border + ax.add_patch(Rectangle((1.5, 1.5), canvas_w - 3, canvas_h - 3, + facecolor="none", edgecolor="#7aa6ff", + linewidth=2.0, zorder=6)) + + _draw_blob_outline(ax, disp, STROKE_WIDTH, zorder=7) + + buf = io.BytesIO() + fig.savefig(buf, format="png", dpi=100, bbox_inches=None, pad_inches=0, + facecolor=fig.get_facecolor()) + plt.close(fig) + buf.seek(0) + return Image.open(buf).convert("RGB") + + +DIVIDER_WIDTH = 3 +DIVIDER_COLOR = (51, 51, 85) # #333355 + + +def render_combined(out_path: Path, template_img: Image.Image, + field_img: Image.Image) -> None: + """Concatenate template (left) and field (right) with a thin divider, + then pad to a square canvas (BG colour) so downstream image-edit + models receive a 1:1 input.""" + bg = tuple(int(BG_COLOR.lstrip("#")[i:i+2], 16) for i in (0, 2, 4)) + tw, th = template_img.size + fw, fh = field_img.size + inner_h = max(th, fh) + inner_w = tw + DIVIDER_WIDTH + fw + side = max(inner_w, inner_h) + combined = Image.new("RGB", (side, side), bg) + x0 = (side - inner_w) // 2 + y0 = (side - inner_h) // 2 + combined.paste(template_img, (x0, y0 + (inner_h - th) // 2)) + for x in range(x0 + tw, x0 + tw + DIVIDER_WIDTH): + for y in range(y0, y0 + inner_h): + combined.putpixel((x, y), DIVIDER_COLOR) + combined.paste(field_img, (x0 + tw + DIVIDER_WIDTH, y0 + (inner_h - fh) // 2)) + combined.save(out_path) + + +# --------------------------------------------------------------------------- +# Main +# --------------------------------------------------------------------------- + + +def main() -> None: + parser = argparse.ArgumentParser() + parser.add_argument("--output-root", type=Path, required=True) + parser.add_argument("--count", type=int, default=30) + parser.add_argument("--seed", type=int, default=0) + parser.add_argument("--width", type=int, default=900) + parser.add_argument("--height", type=int, default=900) + parser.add_argument("--difficulty", type=int, default=5, + help="Integer difficulty >=0; scales match count, total contours, distractor similarity.") + args = parser.parse_args() + + d = max(0, int(args.difficulty)) + # Canvas scaling: N_d = 10 + 2*d, N_0 = 10 + N_d = 10 + 2 * d + N_0 = 10 + s = math.sqrt(max(1.0, N_d / N_0)) + args.width = int(round(args.width * s)) + args.height = int(round(args.height * s)) + + global DISTRACTOR_HAUSDORFF_MIN, TEMPLATE_AMP_SCALE + if d > 0: + _min_matches = 5 + _max_matches = 5 + 2 * d + _total_contours = 10 + 2 * d + DISTRACTOR_HAUSDORFF_MIN = float(max(6, 30 - 3 * d)) + TEMPLATE_AMP_SCALE = 0.18 * (1 + 0.1 * d) + else: + _min_matches = 3 + _max_matches = 5 + _total_contours = 10 + DISTRACTOR_HAUSDORFF_MIN = 30.0 + TEMPLATE_AMP_SCALE = 0.18 + + out_root: Path = args.output_root + img_dir = out_root / "images" + img_dir.mkdir(parents=True, exist_ok=True) + + ann_path = out_root / "annotations.jsonl" + master_rng = random.Random(args.seed) + + # Force evenly-spaced answers across [_min_matches, _max_matches]. + if args.count > 1: + plan = [int(round(_min_matches + i * (_max_matches - _min_matches) / (args.count - 1))) for i in range(args.count)] + else: + plan = [_min_matches] + print(f"forced contour silhouette match counts: {plan}") + + records = [] + with ann_path.open("w") as f: + for i in tqdm(range(args.count), desc="contour_silhouette_count"): + target = plan[i] + if _total_contours is not None: + total_blobs = max(_total_contours, target + 12) + else: + total_blobs = master_rng.randint(18, 26) + total_blobs = max(total_blobs, target + 12) + total_blobs = min(total_blobs, 26) + + built = False + for retry in range(20): + sub_seed = master_rng.randint(0, 2**31 - 1) + sub_rng = random.Random(sub_seed) + try: + (template, match_pl, distractor_pl, + match_centres, distractor_centres, + realised) = build_sample( + sub_rng, args.width, args.height, + target_matches=target, + total_blobs=total_blobs, + ) + except RuntimeError: + continue + built = True + break + if not built: + raise RuntimeError(f"sample {i} could not be generated") + + # Assign letter labels to all field contours in a randomised order so + # match labels are not always the first letters. We label every + # contour (matches + distractors) and record which labels are matches. + n_total = len(match_pl) + len(distractor_pl) + order = list(range(n_total)) + sub_rng.shuffle(order) + # `order[k]` = original index that should receive the k-th letter. + label_for_orig = [""] * n_total + for k, orig in enumerate(order): + label_for_orig[orig] = _index_to_letters(k) + + match_labels = label_for_orig[:len(match_pl)] + distractor_labels = label_for_orig[len(match_pl):] + all_labels = match_labels + distractor_labels + all_centres = match_centres + distractor_centres + + answer = str(realised) + + img_name = f"contour_silhouette_count_{i:05d}.png" + tpl_img = render_template(template) + fld_img = render_field(args.width, args.height, + match_pl, distractor_pl, + all_labels, all_centres) + render_combined(img_dir / img_name, tpl_img, fld_img) + + rec = { + "image": f"images/{img_name}", + "question": QUESTION, + "answer": answer, + "num_matches": realised, + "match_labels": sorted(match_labels), + "total_blobs": n_total, + "metadata": { + "seed": sub_seed, + "template_points": template.tolist(), + }, + } + f.write(json.dumps(rec) + "\n") + f.flush() + records.append(rec) + + data_json = { + "task": "contour_silhouette_count", + "category": "visual_attribute_transfer", + "count": len(records), + "items": records, + } + (out_root / "data.json").write_text(json.dumps(data_json, indent=2)) + print(f"Saved {len(records)} samples to {out_root}") + + # Print answer distribution + from collections import Counter + dist = Counter(r["num_matches"] for r in records) + print("Match-count distribution:", dict(sorted(dist.items()))) + + +if __name__ == "__main__": + main() diff --git a/code/visual_attribute_transfer/contour_silhouette_count/data.json b/code/visual_attribute_transfer/contour_silhouette_count/data.json new file mode 100644 index 0000000000000000000000000000000000000000..2d7f32c153a061ed43a4e4bc584beec34ff75371 --- /dev/null +++ b/code/visual_attribute_transfer/contour_silhouette_count/data.json @@ -0,0 +1,4127 @@ +{ + "task": "contour_silhouette_count", + "category": "visual_attribute_transfer", + "count": 5, + "items": [ + { + "image": "images/contour_silhouette_count_00000.png", + "question": "This image has two panels separated by a thin vertical divider. The left panel shows the Template: a single closed silhouette. The right panel shows the Field: many closed silhouettes scattered across the canvas, all at the SAME pixel scale as the Template. Count the number of Field silhouettes that are exact copies of the Template (translation only \u2014 same size, same orientation, no rotation or mirroring) and report the integer count. Provide your final answer enclosed in ... tags.", + "answer": "5", + "num_matches": 5, + "match_labels": [ + "B", + "H", + "L", + "M", + "R" + ], + "total_blobs": 20, + "metadata": { + "seed": 1654615998, + "template_points": [ + [ + 59.88288659048854, + -2.5776266602431703 + ], + [ + 61.059053348364095, + -0.6760765930405448 + ], + [ + 61.89677537242383, + 1.2819381701431545 + ], + [ + 62.40304707733168, + 3.269136921193678 + ], + [ + 62.591688736192246, + 5.259966682456739 + ], + [ + 62.48249190828891, + 7.23139776308075 + ], + [ + 62.10020217664202, + 9.163559455472562 + ], + [ + 61.473383588421676, + 11.040195556219723 + ], + [ + 60.633212588661934, + 12.848927987534973 + ], + [ + 59.61225060504461, + 14.581325726453352 + ], + [ + 58.44324375222302, + 16.23278511695815 + ], + [ + 57.1579954403894, + 17.802236048750277 + ], + [ + 55.78635314200893, + 19.291696063623622 + ], + [ + 54.355344418465165, + 20.705700865388526 + ], + [ + 52.8884898280052, + 22.050644686624672 + ], + [ + 51.4053118751392, + 23.334067297599415 + ], + [ + 49.92105010509636, + 24.563925998541816 + ], + [ + 48.4465832014306, + 25.747890667402817 + ], + [ + 46.988549919331966, + 26.892697876458296 + ], + [ + 45.54965227488268, + 28.00359635961049 + ], + [ + 44.12911697157756, + 29.083910900942225 + ], + [ + 42.723284891236645, + 30.13474528354358 + ], + [ + 41.326293856011844, + 31.154837599769493 + ], + [ + 39.93081695774208, + 32.14057333339159 + ], + [ + 38.52881764670731, + 33.08615355717403 + ], + [ + 37.11228348671518, + 33.98390772842901 + ], + [ + 35.673902947239185, + 34.824733280238156 + ], + [ + 34.207653666936274, + 35.59863783830324 + ], + [ + 32.7092760660557, + 36.2953547391317 + ], + [ + 31.176612727500416, + 36.904998822562135 + ], + [ + 29.609801280407797, + 37.41872738959214 + ], + [ + 28.01131624778007, + 37.82937084766865 + ], + [ + 26.385863088410062, + 38.131998922123884 + ], + [ + 24.740135103804246, + 38.324391325637336 + ], + [ + 23.082450643990473, + 38.40738630158765 + ], + [ + 21.422293819307566, + 38.38508627583122 + ], + [ + 19.769786446317347, + 38.26490668878817 + ], + [ + 18.135122024907915, + 38.05746161276282 + ], + [ + 16.52799403151529, + 37.776287633390524 + ], + [ + 14.957050667594032, + 37.43741531890784 + ], + [ + 13.429406448572962, + 37.05880504840394 + ], + [ + 11.950237758361656, + 36.659670671058365 + ], + [ + 10.522484900689369, + 36.25972010836202 + ], + [ + 9.146677485966961, + 35.87834632564768 + ], + [ + 7.820893486388201, + 35.533804884662125 + ], + [ + 6.540855294844877, + 35.24241541287904 + ], + [ + 5.300158978489034, + 35.01782373194027 + ], + [ + 4.090625974012855, + 34.87035910022665 + ], + [ + 2.9027600660538946, + 34.80651714406897 + ], + [ + 1.7262869324073156, + 34.828593752411344 + ], + [ + 0.550749098043001, + 34.934488729483235 + ], + [ + -0.6338739717561552, + 35.117690631509774 + ], + [ + -1.8365525404805274, + 35.367446289102254 + ], + [ + -3.064632005374065, + 35.66911039394901 + ], + [ + -4.323327519751736, + 36.00466257245505 + ], + [ + -5.6153153261396955, + 36.35337193752422 + ], + [ + -6.940443949356691, + 36.692582535784666 + ], + [ + -8.295582900896102, + 36.998587684835385 + ], + [ + -9.674620151797535, + 37.24755716459276 + ], + [ + -11.068612639357465, + 37.41647876637572 + ], + [ + -12.466086809953856, + 37.48407491991044 + ], + [ + -13.853479004783678, + 37.431656043522906 + ], + [ + -15.21569870289899, + 37.243874851656784 + ], + [ + -16.53679156313527, + 36.90934998791152 + ], + [ + -17.800674136446574, + 36.42113284443762 + ], + [ + -18.991908289491455, + 35.77699803401981 + ], + [ + -20.09648096864344, + 34.97954540611806 + ], + [ + -21.102554055422548, + 34.03610941498546 + ], + [ + -22.001149764100496, + 32.95847971000889 + ], + [ + -22.786739282629387, + 31.762444675383975 + ], + [ + -23.457706061677428, + 30.467176960038906 + ], + [ + -24.01666015074388, + 29.09448649837748 + ], + [ + -24.470586044901975, + 27.667971857583847 + ], + [ + -24.830813373386164, + 26.212104739809973 + ], + [ + -25.112807130008648, + 24.751284960725524 + ], + [ + -25.33578169246875, + 23.30890413026859 + ], + [ + -25.522150274141723, + 21.906455558127934 + ], + [ + -25.69682837780262, + 20.5627256469556 + ], + [ + -25.88641597891927, + 19.293098338970978 + ], + [ + -26.118288295734267, + 18.10899922552583 + ], + [ + -26.41962889087091, + 17.017499945082495 + ], + [ + -26.816441337560914, + 16.021096754170223 + ], + [ + -27.332576678425685, + 15.117669957214028 + ], + [ + -27.98881337773919, + 14.30062353759662 + ], + [ + -28.802024457326432, + 13.559197156582092 + ], + [ + -29.784463113452844, + 12.878935977328124 + ], + [ + -30.943193497096495, + 12.242297799578553 + ], + [ + -32.27968771274045, + 11.629371989645172 + ], + [ + -33.78960370092904, + 11.018680844865077 + ], + [ + -35.46275179487822, + 10.388031471742282 + ], + [ + -37.28325067402088, + 9.71538505297166 + ], + [ + -39.229866471820344, + 8.97971054012996 + ], + [ + -41.27652221441852, + 8.161791286057884 + ], + [ + -43.39295882990678, + 7.244955818351002 + ], + [ + -45.54552389980802, + 6.215707698349146 + ], + [ + -47.698060305608735, + 5.0642340134794175 + ], + [ + -49.812864084213466, + 3.7847772893715024 + ], + [ + -51.85167922248367, + 2.375861237424006 + ], + [ + -53.776696811490474, + 0.840366521895771 + ], + [ + -55.55152690916365, + -0.8145416088031812 + ], + [ + -57.142113537085365, + -2.577626660243189 + ], + [ + -58.51756632860964, + -4.433923255573747 + ], + [ + -59.65088627843028, + -6.365192518079535 + ], + [ + -60.51956761651535, + -8.350471248702243 + ], + [ + -61.10606282200521, + -10.366693387071457 + ], + [ + -61.398102977916984, + -12.389360925041602 + ], + [ + -61.38887082159823, + -14.393241081455109 + ], + [ + -61.07702875909, + -16.35306709341491 + ], + [ + -60.46660859740384, + -18.244221341381824 + ], + [ + -59.56677365175849, + -20.04338157925248 + ], + [ + -58.39146708659835, + -21.72911363825115 + ], + [ + -56.95896277208185, + -23.282396951642117 + ], + [ + -55.291336546461025, + -24.68707243658349 + ], + [ + -53.41387657593104, + -25.930205503789296 + ], + [ + -51.35445154266572, + -27.002360090915968 + ], + [ + -49.14285474871543, + -27.89778249726086 + ], + [ + -46.81014100600429, + -28.6144963269337 + ], + [ + -44.38797151895393, + -29.154311947136605 + ], + [ + -41.907979996379304, + -29.52275549276275 + ], + [ + -39.4011710965934, + -29.72892358724036 + ], + [ + -36.897360152189634, + -29.78527062392815 + ], + [ + -34.42466106359755, + -29.707335713184193 + ], + [ + -32.0090273980389, + -29.51341632258574 + ], + [ + -29.673850162272316, + -29.224195314863863 + ], + [ + -27.43961448361261, + -28.86232762397537 + ], + [ + -25.32361655322002, + -28.45199231150621 + ], + [ + -23.339741646708593, + -28.01841531568411 + ], + [ + -21.498303798995497, + -27.58736793388824 + ], + [ + -19.805947707255992, + -27.184646038096336 + ], + [ + -18.265613582482715, + -26.835535258427363 + ], + [ + -16.87656586810165, + -26.564267902001365 + ], + [ + -15.634486889446752, + -26.39347819183567 + ], + [ + -14.531636489019215, + -26.34366347227851 + ], + [ + -13.557078447792579, + -26.432660263991238 + ], + [ + -12.696973918103495, + -26.675145368463618 + ], + [ + -11.934941148151717, + -27.082173505908113 + ], + [ + -11.252479439511909, + -27.660764095043493 + ], + [ + -10.629453556650612, + -28.413550618014316 + ], + [ + -10.044632743495757, + -29.338506431575347 + ], + [ + -9.476276171103374, + -30.4287607722028 + ], + [ + -8.90275414635492, + -31.672517964159496 + ], + [ + -8.303191883159387, + -33.05309141003869 + ], + [ + -7.658120221992822, + -34.54906179150569 + ], + [ + -6.9501155385695235, + -36.13456604117504 + ], + [ + -6.1644093675959875, + -37.77972011338906 + ], + [ + -5.289447134951489, + -39.451174472311614 + ], + [ + -4.317374976348218, + -41.11279666005423 + ], + [ + -3.2444340317256515, + -42.726470470810504 + ], + [ + -2.0712429176275755, + -44.252996333576135 + ], + [ + -0.8029513294928126, + -45.65307271127734 + ], + [ + 0.5507490980429989, + -46.88833388450373 + ], + [ + 1.9757664999005844, + -47.92241563045697 + ], + [ + 3.4539080775066466, + -48.722017248197965 + ], + [ + 4.963336583076504, + -49.25792631362831 + ], + [ + 6.479177749927955, + -49.50597163277862 + ], + [ + 7.974266391866445, + -49.447870218337336 + ], + [ + 9.42001291345694, + -49.07193580996717 + ], + [ + 10.787366309569471, + -48.3736195056932 + ], + [ + 12.047844652101974, + -47.35585742186016 + ], + [ + 13.174599818659418, + -46.02920584510685 + ], + [ + 14.143480043349546, + -44.41175091586575 + ], + [ + 14.934051959665188, + -42.52878727061975 + ], + [ + 15.530543309631836, + -40.412268005968116 + ], + [ + 15.922668507025948, + -38.10003651287142 + ], + [ + 16.10630179884772, + -35.634858842700865 + ], + [ + 16.083966835907212, + -33.06328297687436 + ], + [ + 15.865116940922718, + -30.434358352819643 + ], + [ + 15.46618708640091, + -27.798254944144116 + ], + [ + 14.910406340060662, + -25.204825829235826 + ], + [ + 14.227368025642317, + -22.70216028370318 + ], + [ + 13.452363762718091, + -20.335175829587694 + ], + [ + 12.625496542532204, + -18.144297266445214 + ], + [ + 11.790596705088712, + -16.16426846740976 + ], + [ + 10.993972743448207, + -14.423138694434751 + ], + [ + 10.283035928700373, + -12.941459493910633 + ], + [ + 9.704843509484379, + -11.731721071684529 + ], + [ + 9.304609425712234, + -10.798048675120562 + ], + [ + 9.124233878842166, + -10.136170244338546 + ], + [ + 9.200903581600768, + -9.733656792818532 + ], + [ + 9.565813006353308, + -9.57042702472232 + ], + [ + 10.243053482466047, + -9.619497989899319 + ], + [ + 11.248711660083542, + -9.84795450983817 + ], + [ + 12.590211841029777, + -10.21810204900637 + ], + [ + 14.265928230164949, + -10.688760988128514 + ], + [ + 16.26508359885129, + -11.216655158175877 + ], + [ + 18.567940543276595, + -11.757844230002979 + ], + [ + 21.14628086710199, + -12.26914826333449 + ], + [ + 23.96415804220242, + -12.709513457122446 + ], + [ + 26.978897626484688, + -13.04127088331951 + ], + [ + 30.142311350646896, + -13.231244615892743 + ], + [ + 33.40208269896334, + -13.25167199470005 + ], + [ + 36.703275525246724, + -13.08090652605619 + ], + [ + 39.98991282226694, + -12.703882794021776 + ], + [ + 43.20657038359646, + -12.112332367292277 + ], + [ + 46.29992985971352, + -11.304749633673204 + ], + [ + 49.220237625286735, + -10.286116361512077 + ], + [ + 51.92261986264518, + -9.067403163549134 + ], + [ + 54.36821016203676, + -7.664874534197942 + ], + [ + 56.525053497532305, + -6.099231396291537 + ], + [ + 58.368759344462674, + -4.394630832344362 + ] + ] + } + }, + { + "image": "images/contour_silhouette_count_00001.png", + "question": "This image has two panels separated by a thin vertical divider. The left panel shows the Template: a single closed silhouette. The right panel shows the Field: many closed silhouettes scattered across the canvas, all at the SAME pixel scale as the Template. Count the number of Field silhouettes that are exact copies of the Template (translation only \u2014 same size, same orientation, no rotation or mirroring) and report the integer count. Provide your final answer enclosed in ... tags.", + "answer": "8", + "num_matches": 8, + "match_labels": [ + "C", + "E", + "F", + "G", + "H", + "M", + "Q", + "T" + ], + "total_blobs": 20, + "metadata": { + "seed": 1806341205, + "template_points": [ + [ + 48.45124596599337, + 1.3161463304618664 + ], + [ + 49.93759942799447, + 2.7548766253845485 + ], + [ + 51.18724928873264, + 4.2750728442749 + ], + [ + 52.15985647968932, + 5.853803371616253 + ], + [ + 52.821110994488755, + 7.4639215534016445 + ], + [ + 53.14395705066804, + 9.075003772281857 + ], + [ + 53.10961863254095, + 10.654460937314381 + ], + [ + 52.70838289962222, + 12.168787089642382 + ], + [ + 51.9401062723717, + 13.584901657057848 + ], + [ + 50.81441698697131, + 14.87153616956397 + ], + [ + 49.35059824325595, + 16.00061228419878 + ], + [ + 47.57714738551271, + 16.948556006828728 + ], + [ + 45.53101843004472, + 17.69749321090065 + ], + [ + 43.256567222944, + 18.23627402039449 + ], + [ + 40.80423009355578, + 18.561278334203617 + ], + [ + 38.22897758154273, + 18.676961612322163 + ], + [ + 35.58859419796207, + 18.59610881504821 + ], + [ + 32.94184281545347, + 18.339774789261636 + ], + [ + 30.346577813326057, + 17.936901054899668 + ], + [ + 27.857874252649246, + 17.423611417984606 + ], + [ + 25.526240939818425, + 16.842201632956403 + ], + [ + 23.39598317295203, + 16.23985093662667 + ], + [ + 21.50377628133603, + 15.667095151595955 + ], + [ + 19.877503901655988, + 15.176111696733784 + ], + [ + 18.535405530318123, + 14.818875772639952 + ], + [ + 17.485566591480346, + 14.645253796712867 + ], + [ + 16.72577149376681, + 14.701104510210897 + ], + [ + 16.24372641205197, + 15.026459828300608 + ], + [ + 16.017644369857816, + 15.653856320960088 + ], + [ + 16.017171184578423, + 16.60688418081416 + ], + [ + 16.204617545704828, + 17.899013755376853 + ], + [ + 16.536450476724927, + 19.53275041442528 + ], + [ + 16.96498718924049, + 21.499157016675834 + ], + [ + 17.440226309263473, + 23.777769959610875 + ], + [ + 17.911745988668823, + 26.336920249472236 + ], + [ + 18.33059575338461, + 29.134455783344734 + ], + [ + 18.651109212489285, + 32.11884569795172 + ], + [ + 18.832567965110268, + 35.23063282896505 + ], + [ + 18.840653077406696, + 38.404186645587636 + ], + [ + 18.64862912285875, + 41.56969704397828 + ], + [ + 18.23821663806897, + 44.65533960222231 + ], + [ + 17.600121499520128, + 47.58953573588924 + ], + [ + 16.734203652601366, + 50.30322695957719 + ], + [ + 15.649282245187418, + 52.73208135077107 + ], + [ + 14.362588925258983, + 54.81855239488717 + ], + [ + 12.898895241302547, + 56.5137156002689 + ], + [ + 11.28935314245935, + 57.778816415337474 + ], + [ + 9.570098966287897, + 58.58647374113747 + ], + [ + 7.780680549694777, + 58.92149628572509 + ], + [ + 5.9623738183724955, + 58.78128363477843 + ], + [ + 4.156459125229339, + 58.17579962748132 + ], + [ + 2.402528562165642, + 57.12712179424229 + ], + [ + 0.7368934325050724, + 55.66858658001116 + ], + [ + -0.8088438578621178, + 53.8435651990302 + ], + [ + -2.2089968312361385, + 51.70391863459751 + ], + [ + -3.444734162313229, + 49.30819196302153 + ], + [ + -4.504716193828755, + 46.719617381170885 + ], + [ + -5.385407046322061, + 44.00400169273629 + ], + [ + -6.091054632590596, + 41.22757731952651 + ], + [ + -6.633345090792434, + 38.45489603984955 + ], + [ + -7.0307519353899535, + 35.746841638634784 + ], + [ + -7.307612999411117, + 33.15883163763882 + ], + [ + -7.4929794588331395, + 30.739269537726944 + ], + [ + -7.619290414845179, + 28.5282979398426 + ], + [ + -7.7209332781286255, + 26.55689000263316 + ], + [ + -7.832754274701711, + 24.84630250190969 + ], + [ + -7.988584618446914, + 23.407898888142135 + ], + [ + -8.21984624028746, + 22.243335823347344 + ], + [ + -8.554296524014177, + 21.34509234334178 + ], + [ + -9.01496449190757, + 20.697307628814663 + ], + [ + -9.61932163928163, + 20.27688191510934 + ], + [ + -10.378719562322122, + 20.054785782061074 + ], + [ + -11.298114161857558, + 19.997516298737207 + ], + [ + -12.376083095357703, + 20.06863449688241 + ], + [ + -13.605129878430333, + 20.230317531661605 + ], + [ + -14.972255196616466, + 20.444860652307682 + ], + [ + -16.459764146617527, + 20.676068616615282 + ], + [ + -18.04626780335652, + 20.89048319186601 + ], + [ + -19.707829154639768, + 21.058402534997324 + ], + [ + -21.419197417227817, + 21.154659092087872 + ], + [ + -23.155071299449975, + 21.159134688903716 + ], + [ + -24.89133104206947, + 21.057004143131437 + ], + [ + -26.60618106493557, + 20.838711438547705 + ], + [ + -28.2811496641933, + 20.499694691867404 + ], + [ + -29.901899219332357, + 20.039887276433113 + ], + [ + -31.458809451464944, + 19.463032060866812 + ], + [ + -32.94730700406547, + 18.775853369360956 + ], + [ + -34.36792650426381, + 17.98713666118498 + ], + [ + -35.72610076741054, + 17.10676885451063 + ], + [ + -37.031690366632276, + 16.14479259255178 + ], + [ + -38.298274840188746, + 15.11052559415435 + ], + [ + -39.54223881651197, + 14.011791686753432 + ], + [ + -40.78169581359384, + 12.854303434195096 + ], + [ + -42.03530000075769, + 11.641227786726025 + ], + [ + -43.32100147076072, + 10.372956314576962 + ], + [ + -44.65480333547853, + 9.047090816835116 + ], + [ + -46.04957911799182, + 7.658643935969525 + ], + [ + -47.514006472669884, + 6.200443379454988 + ], + [ + -49.0516683426471, + 4.663717965293082 + ], + [ + -50.66036548936527, + 3.038834444451833 + ], + [ + -52.33167522761792, + 1.3161463304618481 + ], + [ + -54.05078058003975, + -0.5130898701650408 + ], + [ + -55.796582398984924, + -2.4557793278783198 + ], + [ + -57.542094804983414, + -4.516084159459128 + ], + [ + -59.255112092191084, + -6.694598047725981 + ], + [ + -60.899123580487554, + -8.98764572636971 + ], + [ + -62.434442251284366, + -11.386747482038533 + ], + [ + -63.81950384031778, + -13.87828159277374 + ], + [ + -65.01228575760798, + -16.44336885685929 + ], + [ + -65.97179005963312, + -19.05799347332881 + ], + [ + -66.65953191228418, + -21.693363965879268 + ], + [ + -67.0409746519436, + -24.316507072888495 + ], + [ + -67.08685466584879, + -26.891077038990968 + ], + [ + -66.7743437566557, + -29.37835300216489 + ], + [ + -66.08800321647541, + -31.73838860050351 + ], + [ + -65.02049221185287, + -33.9312708942192 + ], + [ + -63.57300289955255, + -35.91844050817554 + ], + [ + -61.75540552458838, + -37.66402176110545 + ], + [ + -59.58609813190781, + -39.13611057973017 + ], + [ + -57.09156697183977, + -40.307969222743786 + ], + [ + -54.30567472336254, + -41.15908018848189 + ], + [ + -51.26870385156222, + -41.67601698722094 + ], + [ + -48.02619135484588, + -41.85309647819153 + ], + [ + -44.6275985038276, + -41.69278588623204 + ], + [ + -41.12486466262849, + -41.205847052497134 + ], + [ + -37.570897734614796, + -40.411210529382295 + ], + [ + -34.01805509861253, + -39.33558237487005 + ], + [ + -30.51666810047483, + -38.01279650950533 + ], + [ + -27.113660329767463, + -36.48293486313299 + ], + [ + -23.85130521614485, + -34.79124588810454 + ], + [ + -20.766162171183073, + -32.98689903267002 + ], + [ + -17.888222885088698, + -31.1216181991815 + ], + [ + -15.240290814269713, + -29.248240877050865 + ], + [ + -12.83760774327217, + -27.419251440409774 + ], + [ + -10.687731961247962, + -25.685337017190154 + ], + [ + -8.790663439907053, + -24.09401243193762 + ], + [ + -7.139202794233483, + -22.688357135809206 + ], + [ + -5.719523068174008, + -21.50590196666386 + ], + [ + -4.511926783339589, + -20.577697287451006 + ], + [ + -3.4917554273794327, + -19.927586831076535 + ], + [ + -2.63041478107135, + -19.571703759844084 + ], + [ + -1.896477260285753, + -19.51819736362373 + ], + [ + -1.2568217820028114, + -19.767190804708846 + ], + [ + -0.677772487406691, + -20.310962681316443 + ], + [ + -0.12619984483979785, + -21.134338205941678 + ], + [ + 0.4294489646780115, + -22.215269715928383 + ], + [ + 1.0182191221903558, + -23.525581236319212 + ], + [ + 1.6658403627636145, + -25.03184802626912 + ], + [ + 2.3940603819927064, + -26.696379526684858 + ], + [ + 3.220131780043888, + -28.47827289496686 + ], + [ + 4.1564591252293415, + -30.334504312817632 + ], + [ + 5.210407237306676, + -32.22102638434413 + ], + [ + 6.384266762325124, + -34.09384206086236 + ], + [ + 7.675368691268394, + -35.91002845995573 + ], + [ + 9.076335775682232, + -37.6286874917966 + ], + [ + 10.575455880801442, + -39.21180415772569 + ], + [ + 12.157160210865804, + -40.624997538055815 + ], + [ + 13.802588020286258, + -41.838153643449296 + ], + [ + 15.490218828627494, + -42.82593329384 + ], + [ + 17.196553197050577, + -43.56815186663725 + ], + [ + 18.896823686751503, + -44.050031012496426 + ], + [ + 20.565718580762063, + -44.262325200894836 + ], + [ + 22.178102180236156, + -44.20132819661562 + ], + [ + 23.709716861322093, + -43.86876628686991 + ], + [ + 25.137853488739324, + -43.27158631566766 + ], + [ + 26.441978137486036, + -42.4216474036278 + ], + [ + 27.60430431056219, + -41.33532572495038 + ], + [ + 28.61030092288961, + -40.03304197855734 + ], + [ + 29.449127244304922, + -38.53872133106878 + ], + [ + 30.113986780889814, + -36.87919572400359 + ], + [ + 30.602392772801213, + -35.08355861203627 + ], + [ + 30.91633866740361, + -33.18248249881328 + ], + [ + 31.062367671926474, + -31.207510101606122 + ], + [ + 31.051536389161576, + -29.190330616490503 + ], + [ + 30.89926867969388, + -27.1620533514473 + ], + [ + 30.625097350916775, + -25.152491895225523 + ], + [ + 30.252293104036156, + -23.189472917073694 + ], + [ + 29.80738240720155, + -21.29818454620661 + ], + [ + 29.319558606437226, + -19.500579944055964 + ], + [ + 28.81999360205048, + -17.81485203327403 + ], + [ + 28.341060736259486, + -16.25499526281847 + ], + [ + 27.91548305213966, + -14.830469656766102 + ], + [ + 27.57542465669624, + -13.545981124328259 + ], + [ + 27.351546387401385, + -12.401390036953545 + ], + [ + 27.27205015831516, + -11.391757378162543 + ], + [ + 27.361739055654038, + -10.507534356731805 + ], + [ + 27.641122270889642, + -9.734897302187612 + ], + [ + 28.12559512172865, + -9.056225036507138 + ], + [ + 28.82472456057828, + -8.450710883849487 + ], + [ + 29.74166958346189, + -7.895096226119084 + ], + [ + 30.872763750717027, + -7.3645072529173605 + ], + [ + 32.207283586752865, + -6.833371529083299 + ], + [ + 33.727421969528166, + -6.276386462009571 + ], + [ + 35.40847984126795, + -5.6695079436230795 + ], + [ + 37.21928282054986, + -4.990924603574296 + ], + [ + 39.12282177952259, + -4.221981448723112 + ], + [ + 41.07710842687085, + -3.3480163476412255 + ], + [ + 43.03622870705945, + -2.3590739648587444 + ], + [ + 44.951568719327355, + -1.2504644152269802 + ], + [ + 46.77318022199566, + -0.023138084620409893 + ] + ] + } + }, + { + "image": "images/contour_silhouette_count_00002.png", + "question": "This image has two panels separated by a thin vertical divider. The left panel shows the Template: a single closed silhouette. The right panel shows the Field: many closed silhouettes scattered across the canvas, all at the SAME pixel scale as the Template. Count the number of Field silhouettes that are exact copies of the Template (translation only \u2014 same size, same orientation, no rotation or mirroring) and report the integer count. Provide your final answer enclosed in ... tags.", + "answer": "10", + "num_matches": 10, + "match_labels": [ + "C", + "D", + "G", + "I", + "J", + "L", + "M", + "N", + "Q", + "V" + ], + "total_blobs": 22, + "metadata": { + "seed": 173879092, + "template_points": [ + [ + 43.26594215961219, + -1.1508314185001145 + ], + [ + 44.7217786121908, + 0.1840500569756689 + ], + [ + 45.934566987178876, + 1.597873001219399 + ], + [ + 46.855118751048884, + 3.066051105945041 + ], + [ + 47.44422793768217, + 4.559138572971672 + ], + [ + 47.674339646594106, + 6.044443564303187 + ], + [ + 47.53073356250932, + 7.487853499031258 + ], + [ + 47.012170732079774, + 8.85578074399496 + ], + [ + 46.130975565868546, + 10.117130517652557 + ], + [ + 44.91254996893097, + 11.245190889059268 + ], + [ + 43.39434150108849, + 12.219347704412419 + ], + [ + 41.62431138979985, + 13.026534974760153 + ], + [ + 39.658969980687516, + 13.662343343920364 + ], + [ + 37.56106583456309, + 14.13172514852781 + ], + [ + 35.39702933650639, + 14.449253522981925 + ], + [ + 33.23428173124748, + 14.638914085717985 + ], + [ + 31.138525515328393, + 14.73342996240404 + ], + [ + 29.171131910748894, + 14.773143195496226 + ], + [ + 27.386735769665133, + 14.804496894462979 + ], + [ + 25.83113800587024, + 14.878181781476421 + ], + [ + 24.539601029589804, + 15.047027164633185 + ], + [ + 23.53560438791215, + 15.363729046246284 + ], + [ + 22.830106756474752, + 15.878516445988822 + ], + [ + 22.421337581383597, + 16.63686069155736 + ], + [ + 22.29511809877534, + 17.677331230622816 + ], + [ + 22.425688249868827, + 19.029695505952404 + ], + [ + 22.776994219436613, + 20.713349897254265 + ], + [ + 23.304371934326248, + 22.736154168385568 + ], + [ + 23.956545721048954, + 25.093723955743183 + ], + [ + 24.67784912975946, + 27.769215436918678 + ], + [ + 25.410567184705545, + 30.733614386602582 + ], + [ + 26.09729630191039, + 33.94651938631751 + ], + [ + 26.683219880993853, + 37.357387052239794 + ], + [ + 27.11820396120712, + 40.90718679702557 + ], + [ + 27.358627948340953, + 44.53039478441827 + ], + [ + 27.36887969137884, + 48.15724218389064 + ], + [ + 27.122461373706, + 51.71612224020438 + ], + [ + 26.60267191501773, + 55.13605450206573 + ], + [ + 25.802851905374453, + 58.349103055682846 + ], + [ + 24.726197524517186, + 61.292648811233164 + ], + [ + 23.385169461946134, + 63.911423599048625 + ], + [ + 21.800540629749186, + 66.15922564234363 + ], + [ + 20.000141636811374, + 68.00025128872615 + ], + [ + 18.017374896792195, + 69.40999594723593 + ], + [ + 15.889576371878322, + 70.37569711195665 + ], + [ + 13.656308002521996, + 70.89631319825506 + ], + [ + 11.357663739262888, + 70.98205268124987 + ], + [ + 9.032667883922906, + 70.65348773099323 + ], + [ + 6.717836471926093, + 69.94030426974147 + ], + [ + 4.445961175552011, + 68.87975532380645 + ], + [ + 2.2451613254163014, + 67.51489603972028 + ], + [ + 0.13823390141577319, + 65.89268629035294 + ], + [ + -1.8576854170129002, + 64.06205011654765 + ], + [ + -3.7311539425134295, + 62.07198024841546 + ], + [ + -5.476166357243979, + 59.969770753655624 + ], + [ + -7.091646288694745, + 57.79945179732298 + ], + [ + -8.580731858770516, + 55.60048808252907 + ], + [ + -9.949911334784783, + 53.406787446788414 + ], + [ + -11.20807101358051, + 51.24604910951906 + ], + [ + -12.36551985958566, + 49.13946308079284 + ], + [ + -13.4330541253158, + 47.10175416650478 + ], + [ + -14.421120334721, + 45.14154674997471 + ], + [ + -15.339126919078218, + 43.26201095123055 + ], + [ + -16.19494394491763, + 41.46173762448007 + ], + [ + -16.994617390690216, + 39.73577958083386 + ], + [ + -17.74231005112863, + 38.0767898835905 + ], + [ + -18.440466184940195, + 36.47618533860982 + ], + [ + -19.090182311387807, + 34.925264475776 + ], + [ + -19.69175292870677, + 33.416214271750114 + ], + [ + -20.245348138697263, + 31.94294828796912 + ], + [ + -20.751770885804316, + 30.501730303325132 + ], + [ + -21.213235290927287, + 29.09155126942314 + ], + [ + -21.63410475378071, + 27.714242751428905 + ], + [ + -22.02152930460588, + 26.37432610452243 + ], + [ + -22.385926105220765, + 25.078612604711847 + ], + [ + -22.741254835549025, + 23.835584742465226 + ], + [ + -23.10505057438687, + 22.654602091133466 + ], + [ + -23.498190144863695, + 21.544985867438974 + ], + [ + -23.944383058561407, + 20.515043928165642 + ], + [ + -24.46939436478462, + 19.571102077182598 + ], + [ + -25.100023034690242, + 18.716607955264365 + ], + [ + -25.86287510368105, + 17.951370413183874 + ], + [ + -26.782984802695676, + 17.270990286331198 + ], + [ + -27.882348540474887, + 16.66652824795923 + ], + [ + -29.17844517424297, + 16.12444244336254 + ], + [ + -30.682820990482533, + 15.626813571952834 + ], + [ + -32.399818850169666, + 15.151858775437198 + ], + [ + -34.32552786899404, + 14.674718969541631 + ], + [ + -36.44702284445046, + 14.168488015909938 + ], + [ + -38.741951657766144, + 13.60543724688427 + ], + [ + -41.178514517374076, + 12.958376145655516 + ], + [ + -43.715861799157715, + 12.202080162065297 + ], + [ + -46.304918154774214, + 11.314710284397464 + ], + [ + -48.88962039525441, + 10.27914649323483 + ], + [ + -51.40853637713497, + 9.084158805012414 + ], + [ + -53.796812712641, + 7.725345274199766 + ], + [ + -55.98838156139236, + 6.2057758591051995 + ], + [ + -57.918341935973785, + 4.536294060946755 + ], + [ + -59.52541964953709, + 2.7354441296036924 + ], + [ + -60.75440287752167, + 0.8290096467565922 + ], + [ + -61.55844773701337, + -1.150831418500135 + ], + [ + -61.90115053394438, + -3.1667104806921635 + ], + [ + -61.75829039224654, + -5.177587286401909 + ], + [ + -61.11915763309967, + -7.140523063800909 + ], + [ + -59.987399068077636, + -9.012632092120096 + ], + [ + -58.38133065579565, + -10.753124419784378 + ], + [ + -56.33368992146548, + -12.325344928086635 + ], + [ + -53.890824188767, + -13.698710824450817 + ], + [ + -51.11133496614194, + -14.850451208294736 + ], + [ + -48.06422266263919, + -15.767058612668887 + ], + [ + -44.82659808604136, + -16.445373187619406 + ], + [ + -41.48104686067777, + -16.89323503637101 + ], + [ + -38.11274906414352, + -17.129658518275424 + ], + [ + -34.80646824345283, + -17.184503291187134 + ], + [ + -31.643530944397725, + -17.097639537668343 + ], + [ + -28.698919604827932, + -16.91762816367481 + ], + [ + -26.03859799333722, + -16.69995968548549 + ], + [ + -23.717179435919576, + -16.504916943225464 + ], + [ + -21.776034223845944, + -16.395145664143552 + ], + [ + -20.241914423426138, + -16.43303231653521 + ], + [ + -19.126152601654592, + -16.677999865870397 + ], + [ + -18.424466696836138, + -17.18383837637828 + ], + [ + -18.117377479257925, + -17.99618852038303 + ], + [ + -18.171218916773647, + -19.150291828771262 + ], + [ + -18.539696457510473, + -20.66911205119555 + ], + [ + -19.165924906967856, + -22.561917650884293 + ], + [ + -19.98485726321084, + -24.82339682379762 + ], + [ + -20.92599949972674, + -27.43335429841409 + ], + [ + -21.916294590320142, + -30.357014504201352 + ], + [ + -22.88305258042099, + -33.54592958453905 + ], + [ + -23.756802512525972, + -36.939464341789105 + ], + [ + -24.4739465464739, + -40.46680473056019 + ], + [ + -24.979106460275613, + -44.049413121017686 + ], + [ + -25.227067412657014, + -47.60383331318532 + ], + [ + -25.18424270996441, + -51.044732135801674 + ], + [ + -24.829605472260624, + -54.28805316972289 + ], + [ + -24.15505751064899, + -57.254152240161304 + ], + [ + -23.165231281946948, + -59.870784125738226 + ], + [ + -21.87674629974074, + -62.075815479127385 + ], + [ + -20.316965679630847, + -63.81955002661925 + ], + [ + -18.5223204679828, + -65.06656824263135 + ], + [ + -16.53628804635037, + -65.79700417869108 + ], + [ + -14.407125374972697, + -66.00720606308383 + ], + [ + -12.18546749302106, + -65.70975361599798 + ], + [ + -9.92190611275383, + -64.93283257317239 + ], + [ + -7.66466215824652, + -63.718994448638554 + ], + [ + -5.457459789910894, + -62.123355862710525 + ], + [ + -3.337698156314867, + -60.21131563867349 + ], + [ + -1.3350013910475749, + -58.05588826367015 + ], + [ + 0.5297920030518009, + -55.73476830741447 + ], + [ + 2.2451613254163028, + -53.32725128795307 + ], + [ + 3.808942077573817, + -50.9111417907318 + ], + [ + 5.2278794976692335, + -48.55977916398585 + ], + [ + 6.516729715329303, + -46.33930487254033 + ], + [ + 7.696962813715034, + -44.30628389802331 + ], + [ + 8.795141074177987, + -42.505775978335045 + ], + [ + 9.841060958090456, + -40.96993175668801 + ], + [ + 10.865758233549391, + -39.71716501978371 + ], + [ + 11.89948157907138, + -38.75192624562779 + ], + [ + 12.96974072686976, + -38.065075843584545 + ], + [ + 14.09953072927966, + -37.63482897636439 + ], + [ + 15.305824477799998, + -37.428218907235745 + ], + [ + 16.59841164788605, + -37.40300353952519 + ], + [ + 17.97914447235492, + -37.50992120355892 + ], + [ + 19.441630032648685, + -37.69518761768171 + ], + [ + 20.971386111282918, + -37.90311691169279 + ], + [ + 22.546454173576716, + -38.07874602037925 + ], + [ + 24.138439884290044, + -38.170343743933515 + ], + [ + 25.713929841554073, + -38.13169318390383 + ], + [ + 27.236213983416665, + -37.9240487000515 + ], + [ + 28.66722732828218, + -37.517685366340686 + ], + [ + 29.969613121405597, + -36.892979304389 + ], + [ + 31.108802645432366, + -36.040980251102205 + ], + [ + 32.05500525456727, + -34.96346217217479 + ], + [ + 32.78500570455252, + -33.67246250263121 + ], + [ + 33.28367441921657, + -32.18934451230032 + ], + [ + 33.54510956092505, + -30.543439240494713 + ], + [ + 33.573347034675464, + -28.77034240478973 + ], + [ + 33.38259503706749, + -26.90995679772674 + ], + [ + 32.99697248937052, + -25.004381266046146 + ], + [ + 32.44975458622891, + -23.095752963584555 + ], + [ + 31.782152607880775, + -21.2241499657269 + ], + [ + 31.041677939899994, + -19.425656564102123 + ], + [ + 30.280160826707274, + -17.73068390504858 + ], + [ + 29.55151176229052, + -16.16262460557779 + ], + [ + 28.909326752353255, + -14.736902293053946 + ], + [ + 28.40444631491157, + -13.460456556608376 + ], + [ + 28.08258159057142, + -12.331681583219455 + ], + [ + 27.98211912026919, + -11.340813870727203 + ], + [ + 28.132208778812597, + -10.47074197996387 + ], + [ + 28.55122733683688, + -9.698190396013649 + ], + [ + 29.245693705131245, + -8.995211220867011 + ], + [ + 30.209691845203995, + -8.330902494153005 + ], + [ + 31.424834532816742, + -7.673261142638503 + ], + [ + 32.8607766908469, + -6.991072396655514 + ], + [ + 34.476261997952456, + -6.255736259563515 + ], + [ + 36.22066209048607, + -5.442935312546634 + ], + [ + 38.03594503949728, + -4.534056579198947 + ], + [ + 39.858989954087185, + -3.5172929308087433 + ], + [ + 41.624148458321216, + -2.3883659446564884 + ] + ] + } + }, + { + "image": "images/contour_silhouette_count_00003.png", + "question": "This image has two panels separated by a thin vertical divider. The left panel shows the Template: a single closed silhouette. The right panel shows the Field: many closed silhouettes scattered across the canvas, all at the SAME pixel scale as the Template. Count the number of Field silhouettes that are exact copies of the Template (translation only \u2014 same size, same orientation, no rotation or mirroring) and report the integer count. Provide your final answer enclosed in ... tags.", + "answer": "12", + "num_matches": 12, + "match_labels": [ + "A", + "C", + "E", + "H", + "I", + "J", + "N", + "O", + "P", + "U", + "W", + "X" + ], + "total_blobs": 24, + "metadata": { + "seed": 1112038970, + "template_points": [ + [ + 43.62552417336153, + -2.0987071309275165 + ], + [ + 43.2748751825859, + -0.6010273974338005 + ], + [ + 42.843846396244594, + 0.8724954754287668 + ], + [ + 42.34317338194335, + 2.318128472579693 + ], + [ + 41.78418689647445, + 3.7334469873984024 + ], + [ + 41.17855970944056, + 5.117387179125531 + ], + [ + 40.53804550737015, + 6.470262551840861 + ], + [ + 39.87421575997371, + 7.793744026902713 + ], + [ + 39.19820050332733, + 9.090803652527532 + ], + [ + 38.520438931344884, + 10.365622967556519 + ], + [ + 37.85044548793271, + 11.623467890806264 + ], + [ + 37.19659682190351, + 12.870532822521671 + ], + [ + 36.565944512322446, + 14.113757400991338 + ], + [ + 35.96405790364632, + 15.360620037318375 + ], + [ + 35.39490072065109, + 16.618912938384344 + ], + [ + 34.860744377952805, + 17.896503808059606 + ], + [ + 34.362120075172434, + 19.201089778038472 + ], + [ + 33.89781089534952, + 20.53994935340471 + ], + [ + 33.46488422113136, + 21.91969825823109 + ], + [ + 33.05876387131837, + 23.34605503036807 + ], + [ + 32.67334046052015, + 24.82362204245329 + ], + [ + 32.30111761768952, + 26.355687321629304 + ], + [ + 31.9333908851158, + 27.94405211014879 + ], + [ + 31.560455376812968, + 29.58888856257919 + ], + [ + 31.171837621210244, + 31.288631325024156 + ], + [ + 30.75654646265499, + 33.039906002422306 + ], + [ + 30.30333746207407, + 34.8374967084259 + ], + [ + 29.80098492910411, + 36.67435402715317 + ], + [ + 29.23855554305458, + 38.54164381706415 + ], + [ + 28.605677482076032, + 40.42883637493793 + ], + [ + 27.89279907955457, + 42.32383457335006 + ], + [ + 27.09143126153551, + 44.213138708901624 + ], + [ + 26.194368383261178, + 46.08204497081411 + ], + [ + 25.195882568067887, + 47.91487367931681 + ], + [ + 24.091887246521605, + 49.695222767881454 + ], + [ + 22.88006628387403, + 51.40624140818748 + ], + [ + 21.559965853607565, + 53.03091821477748 + ], + [ + 20.133047046137733, + 54.552378128094986 + ], + [ + 18.602698075416466, + 55.95418186752211 + ], + [ + 16.974205842081755, + 57.22062177463673 + ], + [ + 15.254687509345505, + 58.33700793253995 + ], + [ + 13.452983626469273, + 59.289938647942364 + ], + [ + 11.57951517444656, + 60.06754971381507 + ], + [ + 9.646107690387034, + 60.65973732389186 + ], + [ + 7.665786333522633, + 61.058350075446356 + ], + [ + 5.652546371012213, + 61.25734616032832 + ], + [ + 3.6211040723031713, + 61.252912590783794 + ], + [ + 1.5866333957214263, + 61.043544118823284 + ], + [ + -0.4355058780291352, + 60.630080367137985 + ], + [ + -2.4300427689682156, + 60.01570057611244 + ], + [ + -4.382066670694837, + 59.20587626514868 + ], + [ + -6.277290186416771, + 58.208282987070206 + ], + [ + -8.10229778859137, + 57.032673202022224 + ], + [ + -9.844775229583846, + 55.69071309313762 + ], + [ + -11.493715108116536, + 54.19578687274174 + ], + [ + -13.039594571990179, + 52.5627727692174 + ], + [ + -14.474521797348288, + 50.8077954271367 + ], + [ + -15.792348612519906, + 48.947959885594514 + ], + [ + -16.988747412657794, + 47.001072613214596 + ], + [ + -18.06125132140718, + 44.98535526725143 + ], + [ + -19.009257378543904, + 42.91915690574241 + ], + [ + -19.83399334864684, + 40.82067031594362 + ], + [ + -20.538449536507127, + 38.70765793244075 + ], + [ + -21.12727774201385, + 36.597192510372956 + ], + [ + -21.606660173807523, + 34.50541730184773 + ], + [ + -21.984151751795583, + 32.44732996803632 + ], + [ + -22.2684997503871, + 30.436593858977403 + ], + [ + -22.469445156012302, + 28.485379622976904 + ], + [ + -22.597510425650576, + 26.60423938432332 + ], + [ + -22.66377853190379, + 24.802014969537392 + ], + [ + -22.679668261641552, + 23.085780886836144 + ], + [ + -22.65671069929267, + 21.460821989378175 + ], + [ + -22.606331675179845, + 19.93064499833152 + ], + [ + -22.53964469934961, + 18.497022344313212 + ], + [ + -22.467258540176395, + 17.160066121591672 + ], + [ + -22.399103155012533, + 15.918329353390194 + ], + [ + -22.344277149791324, + 14.768931251594053 + ], + [ + -22.31091934999432, + 13.707702730883495 + ], + [ + -22.30610642237548, + 12.729348114133337 + ], + [ + -22.33577781188846, + 11.827618748599548 + ], + [ + -22.40468856854576, + 10.995494144021881 + ], + [ + -22.51638995173856, + 10.225366244646493 + ], + [ + -22.673237031855987, + 9.50922255493292 + ], + [ + -22.876421877134018, + 8.83882404835684 + ], + [ + -23.12603033269278, + 8.205874092788088 + ], + [ + -23.421119882357306, + 7.602175014693149 + ], + [ + -23.759815643958948, + 7.019769386187955 + ], + [ + -24.139421195146678, + 6.451063640384815 + ], + [ + -24.556540666749623, + 5.888932186871861 + ], + [ + -25.007208379386626, + 5.326800794940796 + ], + [ + -25.487022238699435, + 4.758708621237519 + ], + [ + -25.991277145051868, + 4.179348864645837 + ], + [ + -26.515094811963703, + 3.5840886185411542 + ], + [ + -27.05354661861351, + 2.9689690439039826 + ], + [ + -27.60176643779792, + 2.3306874920994756 + ], + [ + -28.155050772007463, + 1.6665636508029769 + ], + [ + -28.708943985153375, + 0.9744921597427574 + ], + [ + -29.259306922812332, + 0.252884435831088 + ], + [ + -29.802367755315252, + -0.4993973467010364 + ], + [ + -30.334754440469407, + -1.2831110605345564 + ], + [ + -30.853508770640754, + -2.098707130927525 + ], + [ + -31.356082526812177, + -2.9463997294794364 + ], + [ + -31.840316794963762, + -3.826233800863383 + ], + [ + -32.30440599337779, + -4.738145308552728 + ], + [ + -32.74684860011728, + -5.682012401830857 + ], + [ + -33.16638694632642, + -6.657695587905637 + ], + [ + -33.56193874332939, + -7.665065428215199 + ], + [ + -33.932523231984064, + -8.70401675407748 + ], + [ + -34.277184975858745, + -9.774468899724017 + ], + [ + -34.59491836240142, + -10.876351965826599 + ], + [ + -34.88459582768609, + -12.009579638952497 + ], + [ + -35.144902682338135, + -13.174009587171021 + ], + [ + -35.37428119307052, + -14.36939291500356 + ], + [ + -35.57088627240555, + -15.595314578671728 + ], + [ + -35.73255475724527, + -16.85112702302838 + ], + [ + -35.85678982550375, + -18.135879594094252 + ], + [ + -35.940761621155566, + -19.448246497075793 + ], + [ + -35.981324645206975, + -20.786456202500606 + ], + [ + -35.97505193760984, + -22.148225248339752 + ], + [ + -35.91828553791713, + -23.530699341746864 + ], + [ + -35.807202185588785, + -24.930404530829634 + ], + [ + -35.637892719120586, + -26.343210997622172 + ], + [ + -35.40645317077778, + -27.76431172345336 + ], + [ + -35.10908514385279, + -29.188217904750754 + ], + [ + -34.74220271385173, + -30.608772560574405 + ], + [ + -34.302542823993576, + -32.01918328422628 + ], + [ + -33.78727595706773, + -33.41207456302882 + ], + [ + -33.194113766061285, + -34.77955953688171 + ], + [ + -32.52141033870827, + -36.11333050236926 + ], + [ + -31.768253857454496, + -37.40476691030981 + ], + [ + -30.934545595001854, + -38.64505906603557 + ], + [ + -30.021063452849006, + -39.82534523831941 + ], + [ + -29.029507599943365, + -40.93685942889467 + ], + [ + -27.962526192319345, + -41.971086662971004 + ], + [ + -26.82371964200137, + -42.9199223435592 + ], + [ + -25.617622442286788, + -43.77583197849442 + ], + [ + -24.34966213312325, + -44.532007446430896 + ], + [ + -23.026095589838867, + -45.182515922110085 + ], + [ + -21.653923425413442, + -45.72243763477974 + ], + [ + -20.24078389487206, + -46.14798878710166 + ], + [ + -18.794828264367514, + -46.45662621300569 + ], + [ + -17.324580141692373, + -46.64713069694665 + ], + [ + -15.838781744777979, + -46.71966630668015 + ], + [ + -14.346230496896968, + -46.675813597488556 + ], + [ + -12.855609670085993, + -46.51857511619239 + ], + [ + -11.375317041947863, + -46.252352254928134 + ], + [ + -9.91329567786194, + -45.882893162735996 + ], + [ + -8.476870995498807, + -45.417212101572254 + ], + [ + -7.072598208713493, + -44.863481315799206 + ], + [ + -5.706124083336148, + -44.230897153593325 + ], + [ + -4.382066670694835, + -43.529522818237524 + ], + [ + -3.1039163211268335, + -42.770110720676925 + ], + [ + -1.8739608269984398, + -41.96390793672014 + ], + [ + -0.6932370129360792, + -41.12244872891768 + ], + [ + 0.4384895077408748, + -40.257338462214726 + ], + [ + 1.5227153430052525, + -39.38003351375239 + ], + [ + 2.5621607201614154, + -38.50162194279041 + ], + [ + 3.5607016034025225, + -37.63260974125951 + ], + [ + 4.523273575683867, + -36.78271742623022 + ], + [ + 5.455749228962639, + -35.96069156266392 + ], + [ + 6.364791474595681, + -35.17413552104591 + ], + [ + 7.257685805067637, + -34.429363385479085 + ], + [ + 8.142155093409762, + -33.73128044175679 + ], + [ + 9.026160993241234, + -33.08329310249801 + ], + [ + 9.917696387674344, + -32.48725048047929 + ], + [ + 10.824573618841352, + -31.943419116609057 + ], + [ + 11.754213403369285, + -31.450491621886123 + ], + [ + 12.713439397210653, + -31.00562922066798 + ], + [ + 13.708283313049506, + -30.604537403896558 + ], + [ + 14.743805315125506, + -30.24157313415507 + ], + [ + 15.823934122704657, + -29.909881307988712 + ], + [ + 16.951330850360364, + -29.601557492654635 + ], + [ + 18.12728010920941, + -29.307833331192906 + ], + [ + 19.35161129929074, + -29.019280466804155 + ], + [ + 20.622652352685602, + -28.726028388539433 + ], + [ + 21.93721745502001, + -28.417991256642317 + ], + [ + 23.290629496554892, + -28.085098536471126 + ], + [ + 24.67677720126049, + -27.71752416102243 + ], + [ + 26.08820607198355, + -27.305908957075133 + ], + [ + 27.516241491285406, + -26.841571209322225 + ], + [ + 28.95114154986385, + -26.31670049802126 + ], + [ + 30.382276456198234, + -25.724530323200657 + ], + [ + 31.798330729665956, + -25.05948551401258 + ], + [ + 33.187523810890845, + -24.317301004490535 + ], + [ + 34.53784425167361, + -23.495109223427768 + ], + [ + 35.83729228447167, + -22.591494080936474 + ], + [ + 37.074125327488495, + -21.606510320349717 + ], + [ + 38.23710086269329, + -20.54166782304145 + ], + [ + 39.31571113427845, + -19.399881286135837 + ], + [ + 40.300404254904294, + -18.185386519200257 + ], + [ + 41.18278657422668, + -16.903625406164334 + ], + [ + 41.95580155327594, + -15.561102333695471 + ], + [ + 42.61388089095425, + -14.165215578889063 + ], + [ + 43.153064254202995, + -12.724067760625427 + ], + [ + 43.571084657738346, + -11.246259975356963 + ], + [ + 43.867417306948646, + -9.740674646674227 + ], + [ + 44.04329054107557, + -8.21625240852121 + ], + [ + 44.10165837422217, + -6.681768506916146 + ], + [ + 44.047135009156584, + -5.145614239975258 + ], + [ + 43.88589257290602, + -3.61558885948655 + ] + ] + } + }, + { + "image": "images/contour_silhouette_count_00004.png", + "question": "This image has two panels separated by a thin vertical divider. The left panel shows the Template: a single closed silhouette. The right panel shows the Field: many closed silhouettes scattered across the canvas, all at the SAME pixel scale as the Template. Count the number of Field silhouettes that are exact copies of the Template (translation only \u2014 same size, same orientation, no rotation or mirroring) and report the integer count. Provide your final answer enclosed in ... tags.", + "answer": "15", + "num_matches": 15, + "match_labels": [ + "A", + "AA", + "B", + "C", + "F", + "G", + "I", + "K", + "M", + "N", + "R", + "T", + "U", + "Y", + "Z" + ], + "total_blobs": 27, + "metadata": { + "seed": 2087043557, + "template_points": [ + [ + 44.44277960010965, + -1.5135772298455548 + ], + [ + 44.97757930029059, + -0.2440484942287342 + ], + [ + 45.37233049140971, + 1.0528259592544305 + ], + [ + 45.61693668353232, + 2.3655062466698427 + ], + [ + 45.70405171643831, + 3.6815329821615146 + ], + [ + 45.62920720230568, + 4.987896132239603 + ], + [ + 45.39086727675366, + 6.271412885526659 + ], + [ + 44.99041173498287, + 7.519103762969489 + ], + [ + 44.432050499270076, + 8.718556718100722 + ], + [ + 43.722674057189735, + 9.858269778852979 + ], + [ + 42.87164598361344, + 10.927963820842383 + ], + [ + 41.89054488019117, + 11.918858290243424 + ], + [ + 40.79286401023389, + 12.823904061749165 + ], + [ + 39.593677561820016, + 13.63796907111185 + ], + [ + 38.30928283513834, + 14.357973849931984 + ], + [ + 36.95682772921209, + 14.982975562252019 + ], + [ + 35.553932715135865, + 15.514200551414596 + ], + [ + 34.11831605287599, + 15.95502671017379 + ], + [ + 32.66743036841311, + 16.310918152398415 + ], + [ + 31.218117894652263, + 16.58931566363479 + ], + [ + 29.78629073374007, + 16.79948722125561 + ], + [ + 28.38664146262484, + 16.952343492342532 + ], + [ + 27.032388320305536, + 17.060223636677524 + ], + [ + 25.73505812499355, + 17.136656969100002 + ], + [ + 24.504309009902755, + 17.19610608316108 + ], + [ + 23.347794070567975, + 17.25369692590135 + ], + [ + 22.271066111792468, + 17.324941066198065 + ], + [ + 21.277522889392785, + 17.42545504466127 + ], + [ + 20.368391574712216, + 17.570681261879525 + ], + [ + 19.542750635130066, + 17.775614384952686 + ], + [ + 18.797586921197553, + 18.05453675985827 + ], + [ + 18.127885473716304, + 18.4207658371472 + ], + [ + 17.52674939934268, + 18.886416175052428 + ], + [ + 16.985547093595716, + 19.462178196965635 + ], + [ + 16.494084094240034, + 20.157115563567142 + ], + [ + 16.04079690231743, + 20.97848278181372 + ], + [ + 15.612966188067784, + 21.931564515348754 + ], + [ + 15.196946880531282, + 23.019537979311302 + ], + [ + 14.778412700519826, + 24.24335978672414 + ], + [ + 14.342612717756158, + 25.601678648141004 + ], + [ + 13.874637479393797, + 27.090775391211285 + ], + [ + 13.359692159046055, + 28.704531839244613 + ], + [ + 12.783374008817873, + 30.434430142780638 + ], + [ + 12.131951163702304, + 32.26958417012093 + ], + [ + 11.392639556285292, + 34.196804507216804 + ], + [ + 10.553874364111309, + 36.20069847205164 + ], + [ + 9.605572051757857, + 38.2638062952421 + ], + [ + 8.539378708620122, + 40.36677424347513 + ], + [ + 7.348900049017654, + 42.488564958007956 + ], + [ + 6.0299081630428075, + 44.606704645906916 + ], + [ + 4.580519914821062, + 46.69756600331179 + ], + [ + 3.0013418089627173, + 48.73668488147073 + ], + [ + 1.29557621298949, + 50.69910774848937 + ], + [ + -0.5309159434453452, + 52.55976598028303 + ], + [ + -2.469600454830715, + 54.293871966616244 + ], + [ + -4.509306961815716, + 55.87733098065381 + ], + [ + -6.636320948315171, + 57.28716177484654 + ], + [ + -8.834532647824776, + 58.501917975787386 + ], + [ + -11.085643706784609, + 59.50210159956524 + ], + [ + -13.3694311074943, + 60.27055943907067 + ], + [ + -15.664066379165874, + 60.79285272408182 + ], + [ + -17.946486561464774, + 61.057590356873 + ], + [ + -20.19281177538708, + 61.05671620666854 + ], + [ + -22.378802652045966, + 60.78574142321933 + ], + [ + -24.480349325221344, + 60.24391351122132 + ], + [ + -26.473982264676064, + 59.43431499086814 + ], + [ + -28.33739397054308, + 58.36388584221145 + ], + [ + -30.049959518695655, + 57.04336556779326 + ], + [ + -31.593243192823838, + 55.487152574038824 + ], + [ + -32.95147800453453, + 53.713080621934246 + ], + [ + -34.112004823521445, + 51.74211427739272 + ], + [ + -35.06565814112402, + 49.597967539846415 + ], + [ + -35.80708618648496, + 47.30665207676962 + ], + [ + -36.33499420668933, + 44.89596367132335 + ], + [ + -36.65230119932973, + 42.39491752818542 + ], + [ + -36.76620222321951, + 39.83314490819503 + ], + [ + -36.688130572756386, + 37.24026510872519 + ], + [ + -36.43361653368368, + 34.64524801287697 + ], + [ + -36.02204208150547, + 32.07578324432357 + ], + [ + -35.47629366781861, + 29.55767234414188 + ], + [ + -34.82231808598208, + 27.11426030184367 + ], + [ + -34.08858923223166, + 24.765922209459088 + ], + [ + -33.30549629524533, + 22.52961976413299 + ], + [ + -32.50466642999047, + 20.418540835818153 + ], + [ + -31.718237217033117, + 18.441833372206773 + ], + [ + -30.97809609864316, + 16.604442577814584 + ], + [ + -30.315105448686214, + 14.907057636785257 + ], + [ + -29.75833291606079, + 13.34617132052444 + ], + [ + -29.334307136078326, + 11.914252713136129 + ], + [ + -29.066318800434498, + 10.600030089286413 + ], + [ + -28.973786400415978, + 9.388877785407146 + ], + [ + -29.071704713147636, + 8.263297813354576 + ], + [ + -29.370192308066347, + 7.203484072395116 + ], + [ + -29.87415204889425, + 6.187954413578732 + ], + [ + -30.583055810289753, + 5.1942335862044695 + ], + [ + -31.490861488485077, + 4.199568325449963 + ], + [ + -32.58606694538961, + 3.181654587168145 + ], + [ + -33.851901880735326, + 2.119356249455761 + ], + [ + -35.26665488006259, + 0.9933945133725608 + ], + [ + -36.804129146133384, + -0.21301223834528016 + ], + [ + -38.43421679809976, + -1.5135772298455685 + ], + [ + -40.1235782253502, + -2.9184601112266626 + ], + [ + -41.83640991540011, + -4.433882925049929 + ], + [ + -43.53528153322503, + -6.061859586182322 + ], + [ + -45.18202089743696, + -7.800048080557205 + ], + [ + -46.73862394686966, + -9.64173110831382 + ], + [ + -48.16816587309066, + -11.575927197511241 + ], + [ + -49.435689345246224, + -13.587630505184812 + ], + [ + -50.50904618879574, + -15.65817371684751 + ], + [ + -51.359669993887614, + -17.76570476900916 + ], + [ + -51.96325889652723, + -19.88576466508951 + ], + [ + -52.30035015037597, + -21.99195054067777 + ], + [ + -52.35677102424825, + -24.056645457687832 + ], + [ + -52.12395393834833, + -26.05179425442076 + ], + [ + -51.599107494574596, + -27.949703220519933 + ], + [ + -50.78523905458284, + -29.723840455331306 + ], + [ + -49.691028656975504, + -31.34961353889662 + ], + [ + -48.33055822023135, + -32.805101609684186 + ], + [ + -46.72290402782137, + -34.071720093934935 + ], + [ + -44.89160431595091, + -35.13479813875478 + ], + [ + -42.864017268408226, + -35.98405121499333 + ], + [ + -40.67058776284405, + -36.61393430771023 + ], + [ + -38.34404371736259, + -37.023864515861504 + ], + [ + -35.91854478050869, + -37.21830563856792 + ], + [ + -33.4288073349233, + -37.20671132142759 + ], + [ + -30.909230308676772, + -37.00332745334342 + ], + [ + -28.393046093524323, + -36.62685861854349 + ], + [ + -25.91151996291998, + -36.10000739575982 + ], + [ + -23.4932197931499, + -35.448899036231374 + ], + [ + -21.163375667821477, + -34.70240743087422 + ], + [ + -18.943346157916885, + -33.891401191947296 + ], + [ + -16.850204802604793, + -33.047931037176284 + ], + [ + -14.89645667041762, + -32.20438140269498 + ], + [ + -13.08989096800731, + -31.392610272486863 + ], + [ + -11.433571604112107, + -30.643101564126003 + ], + [ + -9.925963533318473, + -29.984154043100798 + ], + [ + -8.561188721657352, + -29.44112966247649 + ], + [ + -7.3294018143560455, + -29.03578247438533 + ], + [ + -6.217272158128261, + -28.78568688871514 + ], + [ + -5.20855583829024, + -28.703781135180424 + ], + [ + -4.284738922768757, + -28.79803840709318 + ], + [ + -3.4257312320234172, + -29.07127443176816 + ], + [ + -2.610588728462478, + -29.52109623729855 + ], + [ + -1.8182420730872337, + -30.139992789081827 + ], + [ + -1.0282090415403298, + -30.91556407579733 + ], + [ + -0.22126931561948915, + -31.830881256689914 + ], + [ + 0.6199183622286606, + -32.86496675862416 + ], + [ + 1.5102746166786836, + -33.993379842891365 + ], + [ + 2.461848801682191, + -35.18889024703358 + ], + [ + 3.4834716158198336, + -36.42222013027498 + ], + [ + 4.580519914821064, + -37.662832779672854 + ], + [ + 5.754799057256123, + -38.87974541588352 + ], + [ + 7.004544392950542, + -40.042343000101425 + ], + [ + 8.32453977593039, + -41.12117019374613 + ], + [ + 9.706347370700115, + -42.088679545145176 + ], + [ + 11.138639633580786, + -42.91991553758503 + ], + [ + 12.6076212925737, + -43.593116276231235 + ], + [ + 14.097526511161238, + -44.090217245781375 + ], + [ + 15.591174280743928, + -44.39724464962584 + ], + [ + 17.070563503606643, + -44.504589246028125 + ], + [ + 18.51748824568083, + -44.40715521986554 + ], + [ + 19.91415327875394, + -44.10438235692475 + ], + [ + 21.243770298077287, + -43.60014350702302 + ], + [ + 22.49111607666075, + -42.90252291960665 + ], + [ + 23.643035266037902, + -42.02348440365109 + ], + [ + 24.688872521457213, + -40.978441304031456 + ], + [ + 25.62082104814021, + -39.785742912113186 + ], + [ + 26.434177452038487, + -38.466094066430266 + ], + [ + 27.127495840565373, + -37.04192629360182 + ], + [ + 27.702637355907374, + -35.53673985154561 + ], + [ + 28.16471463154778, + -33.974436446819915 + ], + [ + 28.521933936758586, + -32.378662204877436 + ], + [ + 28.785340911977084, + -30.772179694241615 + ], + [ + 28.96847870409253, + -29.17628647909045 + ], + [ + 29.086969897587807, + -27.610295851821494 + ], + [ + 29.158035829676198, + -26.09109314470065 + ], + [ + 29.19996861332896, + -24.63277841651155 + ], + [ + 29.23157242526122, + -23.246403444370845 + ], + [ + 29.271591317150722, + -21.939807916996145 + ], + [ + 29.33814096563915, + -20.71755662129573 + ], + [ + 29.448161395500208, + -19.580976336749128 + ], + [ + 29.616906813198568, + -18.52828819606062 + ], + [ + 29.857487313247123, + -17.554828524281017 + ], + [ + 30.180475419968364, + -16.653348711489475 + ], + [ + 30.593588267490674, + -15.814382574534186 + ], + [ + 31.10145377604295, + -15.026667976496626 + ], + [ + 31.705466535171382, + -14.277608239171375 + ], + [ + 32.40373634120966, + -13.553758129157224 + ], + [ + 33.19112954560022, + -12.841318931448846 + ], + [ + 34.05940063958818, + -12.126627339229213 + ], + [ + 34.99740891238052, + -11.396623563282581 + ], + [ + 35.991412650367494, + -10.63928516344723 + ], + [ + 37.02543126172345, + -9.844014579729759 + ], + [ + 38.08166396994583, + -9.001970133518324 + ], + [ + 39.14095236547051, + -8.10633231289256 + ], + [ + 40.18327316663501, + -7.152499377646647 + ], + [ + 41.18824703598766, + -6.13820864338725 + ], + [ + 42.135649227001394, + -5.063582153383432 + ], + [ + 43.005908187474624, + -3.9310977471118447 + ], + [ + 43.78057899403311, + -2.7454887153839573 + ] + ] + } + } + ] +} \ No newline at end of file diff --git a/code/visual_attribute_transfer/spot_the_contour_diff/creation.py b/code/visual_attribute_transfer/spot_the_contour_diff/creation.py new file mode 100644 index 0000000000000000000000000000000000000000..5549fbc0302999134836bba8df7bb46f9f4548dc --- /dev/null +++ b/code/visual_attribute_transfer/spot_the_contour_diff/creation.py @@ -0,0 +1,600 @@ +"""Generate 'Spot the Contour Diff' visual benchmark dataset. + +Two panels are shown side by side. Each panel contains the same set of closed +contours (Fourier-descriptor blobs) at the same positions, with each contour +pair tagged by an uppercase letter label that appears at the same position in +both panels. Some contours in the right panel have a different shape from the +corresponding contour in the left panel. The task is to list the letter labels +of contour pairs whose shape differs between the two panels. + +Blob generation is reused from the sibling ``contour_silhouette_count`` task. +""" +from __future__ import annotations + +import argparse +import json +import math +import random +import sys +from pathlib import Path +from typing import Any, Dict, List, Tuple + +import matplotlib + +matplotlib.use("Agg") +import matplotlib.pyplot as plt +import matplotlib.patches as mpatches +import numpy as np +from tqdm import tqdm + +# Reuse blob generation and stroke helpers from contour_silhouette_count. +_SIBLING = Path(__file__).resolve().parent.parent / "contour_silhouette_count" +sys.path.insert(0, str(_SIBLING)) +from creation import ( # type: ignore # noqa: E402 + STROKE_WIDTH, + _draw_blob_outline, + blob_bounding_radius, + blob_hausdorff, + fourier_blob, +) + + +# --------------------------------------------------------------------------- +# Constants +# --------------------------------------------------------------------------- + +NUM_ITEMS = 15 +MIN_DIFFS = 4 +MAX_DIFFS = 7 + +PANEL_W = 900 +PANEL_H = 900 +PADDING = 18 # visible gap between blob outlines +EDGE_PADDING = 18 # gap between blob bounding circle and panel edge + +HAUSDORFF_THRESH = 5.0 # changed blob must differ from original by at least this +HAUSDORFF_MAX = 20.0 # …but not by more than this (keep diffs subtle) + +BG_COLOR = "#0a1020" +BORDER_COLOR = "#7aa6ff" +BORDER_WIDTH = 2.0 +HIGHLIGHT_COLOR = "#dc1e1e" + +LABEL_BADGE_COLOR = "#f5d76e" +LABEL_TEXT_COLOR = "#1a1a1a" +LABEL_BADGE_PAD = 8.0 # gap between bounding circle and badge centre offset + +MARGIN_PX = 60 +GAP_PX = 100 +LABEL_HEIGHT = 44 + +QUESTION = ( + "Two panels are shown side by side. Each panel contains the same set of " + "closed contours at the same positions. Some contours in the right " + "panel have a different shape from the corresponding contour in the " + "left panel. Count the number of contour pairs whose shape differs " + "between the two panels and report the integer count. " + "Provide your final answer enclosed in ... tags." +) + + +def _index_to_label(i: int) -> str: + """Map 0->A, 1->B, ..., 25->Z, 26->AA, 27->AB, ...""" + letters = [] + n = i + while True: + letters.append(chr(ord("A") + (n % 26))) + n = n // 26 - 1 + if n < 0: + break + return "".join(reversed(letters)) + + +# --------------------------------------------------------------------------- +# Blob generation / swapping +# --------------------------------------------------------------------------- + + +def _generate_blob(rng: random.Random) -> np.ndarray: + r0 = rng.uniform(38.0, 52.0) + return fourier_blob(rng, r0=r0, K_range=(4, 6), amp_scale=0.22) + + +def _polygon_area(poly: np.ndarray) -> float: + x = poly[:, 0] + y = poly[:, 1] + return 0.5 * abs(float(np.dot(x, np.roll(y, -1)) - np.dot(y, np.roll(x, -1)))) + + +# Minimum fractional difference in enclosed area between the original and the +# swapped contour. Together with HAUSDORFF_THRESH this guarantees that the +# swap differs both in OUTLINE shape and in ENCLOSED REGION. +MIN_AREA_RATIO_DIFF = 0.18 + + +def _sample_different_blob( + rng: random.Random, + old: np.ndarray, + old_radius: float, + max_tries: int = 800, +) -> np.ndarray: + """Sample a new blob with Hausdorff distance >= HAUSDORFF_THRESH from + ``old``, area differing by at least MIN_AREA_RATIO_DIFF, and bounding + radius no larger than ``old_radius``.""" + old_area = _polygon_area(old) + for _ in range(max_tries): + cand = _generate_blob(rng) + if blob_bounding_radius(cand) > old_radius: + continue + d = blob_hausdorff(cand, old) + if not (HAUSDORFF_THRESH <= d <= HAUSDORFF_MAX): + continue + cand_area = _polygon_area(cand) + if old_area > 0 and abs(cand_area - old_area) / old_area < MIN_AREA_RATIO_DIFF: + continue + return cand + raise RuntimeError("Could not sample a sufficiently different blob.") + + +# --------------------------------------------------------------------------- +# Layout +# --------------------------------------------------------------------------- + + +def _place_items( + rng: random.Random, + radii: List[float], + max_tries_per_item: int = 5000, +) -> List[Tuple[float, float]]: + order = sorted(range(len(radii)), key=lambda i: -radii[i]) + centers: List[Tuple[float, float] | None] = [None] * len(radii) + + for i in order: + r = radii[i] + placed = False + for _ in range(max_tries_per_item): + x = rng.uniform(r + EDGE_PADDING, PANEL_W - r - EDGE_PADDING) + y = rng.uniform(r + EDGE_PADDING, PANEL_H - r - EDGE_PADDING) + ok = True + for j, c in enumerate(centers): + if c is None or j == i: + continue + min_dist = radii[i] + radii[j] + PADDING + dx = x - c[0] + dy = y - c[1] + if dx * dx + dy * dy < min_dist * min_dist: + ok = False + break + if ok: + centers[i] = (x, y) + placed = True + break + if not placed: + raise RuntimeError(f"Could not place blob {i} (r={r:.1f}) without overlap.") + return [c for c in centers] # type: ignore[return-value] + + +# --------------------------------------------------------------------------- +# Sample construction +# --------------------------------------------------------------------------- + + +def build_sample( + rng: random.Random, + num_items: int, + num_diffs: int, +) -> Tuple[List[Dict[str, Any]], List[Dict[str, Any]]]: + left_blobs = [_generate_blob(rng) for _ in range(num_items)] + radii = [blob_bounding_radius(b) for b in left_blobs] + + right_blobs: List[np.ndarray] = [b.copy() for b in left_blobs] + changed_idx = rng.sample(range(num_items), num_diffs) + for i in changed_idx: + right_blobs[i] = _sample_different_blob(rng, left_blobs[i], radii[i]) + + centers = _place_items(rng, radii) + changed_set = set(changed_idx) + + items: List[Dict[str, Any]] = [] + for i in range(num_items): + x, y = centers[i] + items.append({ + "index": i, + "label": _index_to_label(i), + "x": x, + "y": y, + "bounding_radius": radii[i], + "left_pts": left_blobs[i], + "right_pts": right_blobs[i], + "changed": i in changed_set, + }) + + diffs: List[Dict[str, Any]] = [ + { + "index": i, + "label": items[i]["label"], + "x": items[i]["x"], + "y": items[i]["y"], + "bounding_radius": items[i]["bounding_radius"], + } + for i in sorted(changed_idx) + ] + return items, diffs + + +# --------------------------------------------------------------------------- +# Rendering +# --------------------------------------------------------------------------- + + +def _panel_origins() -> Tuple[Tuple[float, float], Tuple[float, float]]: + oy = MARGIN_PX + LABEL_HEIGHT + ox_left = MARGIN_PX + ox_right = MARGIN_PX + PANEL_W + GAP_PX + return (ox_left, oy), (ox_right, oy) + + +def _canvas_size() -> Tuple[int, int]: + w = MARGIN_PX + PANEL_W + GAP_PX + PANEL_W + MARGIN_PX + h = MARGIN_PX + LABEL_HEIGHT + PANEL_H + MARGIN_PX + return w, h + + +def _badge_offset( + x: float, + y: float, + radius: float, + badge_r: float, +) -> Tuple[float, float]: + """Legacy compass-based fallback (unused — kept for backward compatibility).""" + margin = radius + badge_r + LABEL_BADGE_PAD + candidates = [ + (1.0, -1.0), (1.0, 1.0), (-1.0, -1.0), (-1.0, 1.0), + (1.0, 0.0), (0.0, -1.0), (-1.0, 0.0), (0.0, 1.0), + ] + inset = badge_r + 2.0 + for sx, sy in candidates: + norm = math.hypot(sx, sy) or 1.0 + dx = sx / norm * margin + dy = sy / norm * margin + bx = x + dx + by = y + dy + if (inset <= bx <= PANEL_W - inset + and inset <= by <= PANEL_H - inset): + return dx, dy + return margin / math.sqrt(2), -margin / math.sqrt(2) + + +def _plan_label_positions( + items: List[Dict[str, Any]], + other_clearance: float = 14.0, + label_clearance: float = 24.0, + n_angles: int = 16, + radial_steps: Tuple[float, ...] = (0.0, 6.0, 14.0, 24.0), +) -> List[Tuple[float, float]]: + """Choose label-badge centres (in LEFT-panel local coords) using a + multi-candidate, best-score search. + + Returns one (px, py) per item. Identical coordinates are reused for the + right panel (just translated by that panel's origin) since both panels + have the same width/height and contour positions. + """ + n = len(items) + # Pre-compute badge radii and translated point sets (panel-local). + badge_rs: List[float] = [] + other_pts_list: List[np.ndarray] = [] + for it in items: + label = it["label"] + badge_rs.append(22.0 + 6.0 * (len(label) - 1)) + # Translate the LEFT contour's points into panel-local coords. + other_pts_list.append(it["left_pts"] + np.array([it["x"], it["y"]])) + + placements: List[Tuple[float, float]] = [] + placed_centres: List[Tuple[float, float]] = [] + + for i, it in enumerate(items): + x, y = it["x"], it["y"] + radius = it["bounding_radius"] + badge_r = badge_rs[i] + inset = badge_r + 2.0 + base_margin = radius + badge_r + LABEL_BADGE_PAD + + # Other contours' points (everything except self). + if n > 1: + other_pts = np.concatenate( + [other_pts_list[j] for j in range(n) if j != i], axis=0 + ) + else: + other_pts = np.zeros((0, 2)) + + def evaluate(oc: float, lc: float): + best = None + best_score = -1e18 + for a_idx in range(n_angles): + ang = 2.0 * math.pi * a_idx / n_angles + cos_a = math.cos(ang) + sin_a = math.sin(ang) + for off in radial_steps: + margin = base_margin + off + px = x + cos_a * margin + py = y + sin_a * margin + # Panel-bounds gate. + if not (inset <= px <= PANEL_W - inset + and inset <= py <= PANEL_H - inset): + continue + # Other-contour clearance. + if other_pts.shape[0] > 0: + dx = other_pts[:, 0] - px + dy = other_pts[:, 1] - py + other_d = float(np.sqrt(np.min(dx * dx + dy * dy))) + else: + other_d = 1e6 + if other_d < oc: + continue + # Label-clearance gate. + if placed_centres: + label_d = min( + math.hypot(px - lx, py - ly) + for lx, ly in placed_centres + ) + else: + label_d = 1e6 + if label_d < lc: + continue + score = other_d + 0.5 * label_d - 0.4 * off + if score > best_score: + best_score = score + best = (px, py) + return best + + chosen = evaluate(other_clearance, label_clearance) + # Soft fallback: progressively relax other_clearance. + relax = other_clearance + while chosen is None and relax > 1.0: + relax *= 0.6 + chosen = evaluate(relax, max(4.0, label_clearance * 0.6)) + if chosen is None: + # Last-resort: legacy compass. + dx, dy = _badge_offset(x, y, radius, badge_r) + chosen = (x + dx, y + dy) + + placements.append(chosen) + placed_centres.append(chosen) + + return placements + + +def _draw_panel( + ax: plt.Axes, + items: List[Dict[str, Any]], + side: str, + ox: float, + oy: float, + label_positions: List[Tuple[float, float]], +) -> None: + key = "left_pts" if side == "left" else "right_pts" + for it in items: + translated = it[key] + np.array([ox + it["x"], oy + it["y"]]) + _draw_blob_outline(ax, translated, STROKE_WIDTH, color="white", zorder=3) + + # Letter-label badges removed: the task is now a count, not a label-list, + # so explicit per-contour anchors are no longer needed (and would let a + # model match in text space rather than visually). + + +def _render( + out_path: Path, + items: List[Dict[str, Any]], + diffs: List[Dict[str, Any]] | None = None, +) -> None: + w, h = _canvas_size() + dpi = 100 + fig, ax = plt.subplots(1, 1, figsize=(w / dpi, h / dpi), dpi=dpi) + ax.set_xlim(0, w) + ax.set_ylim(h, 0) + ax.set_aspect("equal") + ax.axis("off") + fig.patch.set_facecolor(BG_COLOR) + ax.set_facecolor(BG_COLOR) + + (ox_left, oy), (ox_right, _) = _panel_origins() + + for ox in (ox_left, ox_right): + border = mpatches.Rectangle( + (ox, oy), PANEL_W, PANEL_H, + facecolor="none", edgecolor=BORDER_COLOR, + linewidth=BORDER_WIDTH, zorder=1, + ) + ax.add_patch(border) + + label_positions = _plan_label_positions(items) + + _draw_panel(ax, items, "left", ox_left, oy, label_positions) + _draw_panel(ax, items, "right", ox_right, oy, label_positions) + + ax.text( + ox_left + PANEL_W / 2, MARGIN_PX + LABEL_HEIGHT * 0.5, + "Left", ha="center", va="center", + fontsize=16, fontweight="bold", color="#cfe0ff", + ) + ax.text( + ox_right + PANEL_W / 2, MARGIN_PX + LABEL_HEIGHT * 0.5, + "Right", ha="center", va="center", + fontsize=16, fontweight="bold", color="#cfe0ff", + ) + + if diffs: + for diff in diffs: + hl_r = diff["bounding_radius"] + 12 + for ox in (ox_left, ox_right): + cx = ox + diff["x"] + cy = oy + diff["y"] + ring = mpatches.Circle( + (cx, cy), hl_r, + facecolor="none", edgecolor=HIGHLIGHT_COLOR, + linewidth=2.5, zorder=10, + ) + ax.add_patch(ring) + + fig.savefig(out_path, facecolor=BG_COLOR) + plt.close(fig) + + +def render_pair(out_path: Path, items: List[Dict[str, Any]]) -> None: + _render(out_path, items) + + +def render_answer( + out_path: Path, + items: List[Dict[str, Any]], + diffs: List[Dict[str, Any]], +) -> None: + _render(out_path, items, diffs) + + +# --------------------------------------------------------------------------- +# Annotation +# --------------------------------------------------------------------------- + + +def _answer_string(diffs: List[Dict[str, Any]]) -> str: + return str(len(diffs)) + + +def build_annotation( + image_name: str, + items: List[Dict[str, Any]], + diffs: List[Dict[str, Any]], +) -> Dict[str, Any]: + return { + "image": image_name, + "num_items": len(items), + "num_differences": len(diffs), + "differences": [ + { + "index": d["index"], + "label": d["label"], + "x": d["x"], + "y": d["y"], + "bounding_radius": d["bounding_radius"], + } + for d in diffs + ], + "question": QUESTION, + "answer": _answer_string(diffs), + } + + +# --------------------------------------------------------------------------- +# Dataset generation +# --------------------------------------------------------------------------- + + +def generate_dataset( + rng: random.Random, + count: int, + output_dir: Path, + num_items: int = NUM_ITEMS, + min_diffs: int = MIN_DIFFS, + max_diffs: int = MAX_DIFFS, +) -> None: + images_dir = output_dir / "images" + answers_dir = output_dir / "answers" + images_dir.mkdir(parents=True, exist_ok=True) + answers_dir.mkdir(parents=True, exist_ok=True) + + annotations: List[Dict[str, Any]] = [] + data_items: List[Dict[str, Any]] = [] + + # Force evenly-spaced num_diffs across [min_diffs, max_diffs]. + if count > 1: + forced = [int(round(min_diffs + i * (max_diffs - min_diffs) / (count - 1))) for i in range(count)] + else: + forced = [min_diffs] + print(f"forced contour diff counts: {forced}") + + for idx in tqdm(range(count), desc="Generating contour diff pairs"): + num_diffs = forced[idx] + for attempt in range(200): + try: + items, diffs = build_sample(rng, num_items, num_diffs) + break + except RuntimeError: + continue + else: + raise RuntimeError(f"Failed to build sample {idx} after many retries") + + image_name = f"contour_diff_{idx:05d}.png" + img_path = images_dir / image_name + ans_path = answers_dir / image_name + + render_pair(img_path, items) + render_answer(ans_path, items, diffs) + + rel_image = f"images/{image_name}" + annotations.append(build_annotation(rel_image, items, diffs)) + data_items.append({ + "image": rel_image, + "question": QUESTION, + "answer": _answer_string(diffs), + }) + + with (output_dir / "annotations.jsonl").open("w", encoding="utf-8") as fh: + for rec in annotations: + fh.write(json.dumps(rec) + "\n") + + data_json = { + "task": "spot_the_contour_diff", + "category": "visual_attribute_transfer", + "count": len(data_items), + "items": data_items, + } + with (output_dir / "data.json").open("w", encoding="utf-8") as fh: + json.dump(data_json, fh, indent=2) + fh.write("\n") + + +# --------------------------------------------------------------------------- +# CLI +# --------------------------------------------------------------------------- + + +def parse_args() -> argparse.Namespace: + p = argparse.ArgumentParser( + description="Generate 'Spot the Contour Diff' visual benchmark dataset." + ) + p.add_argument("--output-root", type=Path, default=".") + p.add_argument("--count", type=int, default=20) + p.add_argument("--seed", type=int, default=42) + p.add_argument("--difficulty", type=int, default=5, + help="Integer difficulty >=0; scales diff count and subtlety.") + return p.parse_args() + + +def main() -> None: + args = parse_args() + rng = random.Random(args.seed) + d = max(0, int(args.difficulty)) + + N_d = 10 + 2 * d + N_0 = 10 + s = math.sqrt(max(1.0, N_d / N_0)) + global PANEL_W, PANEL_H + PANEL_W = int(round(PANEL_W * s)) + PANEL_H = int(round(PANEL_H * s)) + + global HAUSDORFF_THRESH, HAUSDORFF_MAX + HAUSDORFF_THRESH = max(18.0, 22.0 - 0.4 * d) + HAUSDORFF_MAX = max(HAUSDORFF_THRESH + 12.0, 50.0 - 1.0 * d) + + total_contours = 10 + 2 * d + min_diffs = 5 + max_diffs = 5 + 2 * d + generate_dataset(rng, args.count, args.output_root, + num_items=total_contours, + min_diffs=min_diffs, max_diffs=max_diffs) + print(f"Saved {args.count} image pairs to {args.output_root}") + + +if __name__ == "__main__": + main() diff --git a/code/visual_attribute_transfer/spot_the_contour_diff/data.json b/code/visual_attribute_transfer/spot_the_contour_diff/data.json new file mode 100644 index 0000000000000000000000000000000000000000..8136c74955508249c362789538ebca2f208f111b --- /dev/null +++ b/code/visual_attribute_transfer/spot_the_contour_diff/data.json @@ -0,0 +1,32 @@ +{ + "task": "spot_the_contour_diff", + "category": "visual_attribute_transfer", + "count": 5, + "items": [ + { + "image": "images/contour_diff_00000.png", + "question": "Two panels are shown side by side. Each panel contains the same set of closed silhouettes at the same positions. Some silhouettes in the right panel have a different shape from the corresponding silhouette in the left panel. Count the number of pairs whose shape differs between the two panels and report the integer count. Provide your final answer enclosed in ... tags.", + "answer": "5" + }, + { + "image": "images/contour_diff_00001.png", + "question": "Two panels are shown side by side. Each panel contains the same set of closed silhouettes at the same positions. Some silhouettes in the right panel have a different shape from the corresponding silhouette in the left panel. Count the number of pairs whose shape differs between the two panels and report the integer count. Provide your final answer enclosed in ... tags.", + "answer": "8" + }, + { + "image": "images/contour_diff_00002.png", + "question": "Two panels are shown side by side. Each panel contains the same set of closed silhouettes at the same positions. Some silhouettes in the right panel have a different shape from the corresponding silhouette in the left panel. Count the number of pairs whose shape differs between the two panels and report the integer count. Provide your final answer enclosed in ... tags.", + "answer": "10" + }, + { + "image": "images/contour_diff_00003.png", + "question": "Two panels are shown side by side. Each panel contains the same set of closed silhouettes at the same positions. Some silhouettes in the right panel have a different shape from the corresponding silhouette in the left panel. Count the number of pairs whose shape differs between the two panels and report the integer count. Provide your final answer enclosed in ... tags.", + "answer": "12" + }, + { + "image": "images/contour_diff_00004.png", + "question": "Two panels are shown side by side. Each panel contains the same set of closed silhouettes at the same positions. Some silhouettes in the right panel have a different shape from the corresponding silhouette in the left panel. Count the number of pairs whose shape differs between the two panels and report the integer count. Provide your final answer enclosed in ... tags.", + "answer": "15" + } + ] +} \ No newline at end of file diff --git a/code/visual_attribute_transfer/spot_the_field_diff/apply_effects.py b/code/visual_attribute_transfer/spot_the_field_diff/apply_effects.py new file mode 100644 index 0000000000000000000000000000000000000000..ad8b56379a70ff034c84de15179b923329d13272 --- /dev/null +++ b/code/visual_attribute_transfer/spot_the_field_diff/apply_effects.py @@ -0,0 +1,542 @@ +#!/usr/bin/env python3 +"""Combined effects pipeline for spot_the_field_diff. + +Single-file version of the print-and-capture degradation chain used to +produce the bad-illumination + reversed-colors + sheet-surface variants: + + reverse_dark_background_colors -> composite_on_paper -> color_and_tone + -> perspective_capture -> pre_blur -> lighting_variation + -> scanner_grain -> post_blur -> jpeg_roundtrip -> sheet_surface + +Every effect can be toggled independently via EffectsConfig (or matching +CLI flags). Disabled stages are skipped, preserving the input image for +that stage. + +Usage: + python apply_effects.py # full default chain + python apply_effects.py --no-sheet-surface # skip surface composite + python apply_effects.py --no-perspective-capture --no-jpeg-roundtrip + python apply_effects.py --illumination-profile standard --overwrite +""" + +from __future__ import annotations + +import argparse +import json +import math +import random +from dataclasses import asdict, dataclass +from io import BytesIO +from pathlib import Path + +import numpy as np +from PIL import Image, ImageEnhance, ImageFilter + + +HERE = Path(__file__).resolve().parent +DEFAULT_IMAGES_DIR = HERE / "images" +DEFAULT_OUTPUT_DIR = HERE / "images_with_effects" +IMAGE_EXTENSIONS = {".png", ".jpg", ".jpeg", ".webp"} +OUTPUT_SUFFIX = "_captured.png" + + +# ===================================================================== +# Toggle configuration +# ===================================================================== + +@dataclass +class EffectsConfig: + """Per-effect on/off switches. + + Defaults reproduce the canonical "bad illumination + reversed colors + + sheet on surface" chain. + """ + + reverse_dark_background: bool = True + composite_on_paper: bool = True + color_and_tone: bool = True + perspective_capture: bool = True + pre_blur: bool = True + lighting_variation: bool = True + scanner_grain: bool = True + post_blur: bool = True + jpeg_roundtrip: bool = True + sheet_surface: bool = True # second JPEG roundtrip is also chained when on + + illumination_profile: str = "bad" # "bad" or "standard" + paper_luma_threshold: float = 50.0 + paper_coverage_threshold: float = 0.55 + + +# ===================================================================== +# Print/capture pipeline +# ===================================================================== + +def _luminance(rgb: np.ndarray) -> np.ndarray: + return (rgb[..., 0] * 0.2126) + (rgb[..., 1] * 0.7152) + (rgb[..., 2] * 0.0722) + + +def _estimate_border_rgb(image: Image.Image) -> np.ndarray: + rgba = np.asarray(image.convert("RGBA"), dtype=np.float32) + height, width, _ = rgba.shape + border = max(4, min(height, width) // 50) + border_pixels = np.concatenate([ + rgba[:border].reshape(-1, 4), rgba[-border:].reshape(-1, 4), + rgba[:, :border].reshape(-1, 4), rgba[:, -border:].reshape(-1, 4), + ], axis=0) + visible = border_pixels[border_pixels[:, 3] > 16] + if len(visible) == 0: + visible = border_pixels + return np.median(visible[:, :3], axis=0) + + +def _background_coverage(image: Image.Image, background_rgb: np.ndarray) -> float: + rgba = np.asarray(image.convert("RGBA"), dtype=np.float32) + visible = rgba[..., 3] > 16 + if not np.any(visible): + return 0.0 + distance = np.linalg.norm(rgba[..., :3] - background_rgb, axis=2) + return float(((distance <= 32.0) & visible).sum() / visible.sum()) + + +def reverse_dark_background_colors(image: Image.Image, luma_threshold: float, + coverage_threshold: float) -> tuple[Image.Image, bool]: + rgba = image.convert("RGBA") + background_rgb = _estimate_border_rgb(rgba) + if (float(_luminance(background_rgb)) > luma_threshold + or _background_coverage(rgba, background_rgb) < coverage_threshold): + return rgba, False + arr = np.asarray(rgba, dtype=np.float32) + rgb = arr[..., :3] + alpha = arr[..., 3:4] / 255.0 + distance = np.linalg.norm(rgb - background_rgb, axis=2, keepdims=True) + soft_foreground = np.clip((distance - 10.0) / 42.0, 0.0, 1.0) * alpha + new_background = np.array([246.0, 245.0, 236.0], dtype=np.float32) + source_luma = _luminance(rgb)[..., None] / 255.0 + foreground_scale = 0.10 + (1.0 - source_luma) * 0.18 + new_foreground = (rgb * foreground_scale) + (background_rgb * 0.18) + new_foreground = np.clip(new_foreground, 12.0, 88.0) + reversed_rgb = (new_background * (1.0 - soft_foreground)) + (new_foreground * soft_foreground) + reversed_rgba = np.concatenate([reversed_rgb, arr[..., 3:4]], axis=2) + return Image.fromarray(np.clip(reversed_rgba, 0, 255).astype(np.uint8), mode="RGBA"), True + + +def composite_on_paper(image: Image.Image, paper_rgb: tuple[int, int, int]) -> Image.Image: + rgba = image.convert("RGBA") + paper = Image.new("RGBA", rgba.size, paper_rgb + (255,)) + paper.alpha_composite(rgba) + return paper.convert("RGB") + + +def _perspective_coefficients(source_points, destination_points): + matrix, vector = [], [] + for (sx, sy), (dx, dy) in zip(source_points, destination_points): + matrix.append([dx, dy, 1, 0, 0, 0, -sx * dx, -sx * dy]) + matrix.append([0, 0, 0, dx, dy, 1, -sy * dx, -sy * dy]) + vector.extend([sx, sy]) + return np.linalg.solve(np.asarray(matrix, dtype=np.float64), np.asarray(vector)).tolist() + + +def apply_perspective_capture(image, rng, paper_rgb): + width, height = image.size + max_x = max(1.0, width * rng.uniform(0.008, 0.024)) + max_y = max(1.0, height * rng.uniform(0.012, 0.034)) + destination = [ + (rng.uniform(0, max_x), rng.uniform(0, max_y)), + (width - 1 - rng.uniform(0, max_x), rng.uniform(0, max_y)), + (width - 1 - rng.uniform(0, max_x), height - 1 - rng.uniform(0, max_y)), + (rng.uniform(0, max_x), height - 1 - rng.uniform(0, max_y)), + ] + source = [(0, 0), (width - 1, 0), (width - 1, height - 1), (0, height - 1)] + coeffs = _perspective_coefficients(source, destination) + return image.transform(image.size, Image.Transform.PERSPECTIVE, coeffs, + Image.Resampling.BICUBIC, fillcolor=paper_rgb) + + +def apply_color_and_tone(image, rng): + arr = np.asarray(image).astype(np.float32) / 255.0 + arr *= np.array([rng.uniform(0.97, 1.06), rng.uniform(0.96, 1.04), + rng.uniform(0.93, 1.05)], dtype=np.float32) + arr = np.clip(arr, 0.0, 1.0) ** rng.uniform(0.92, 1.10) + arr = np.clip(arr, 0.0, 1.0) + toned = Image.fromarray((arr * 255.0).astype(np.uint8), mode="RGB") + toned = ImageEnhance.Color(toned).enhance(rng.uniform(0.88, 1.07)) + toned = ImageEnhance.Contrast(toned).enhance(rng.uniform(0.90, 1.05)) + toned = ImageEnhance.Brightness(toned).enhance(rng.uniform(0.94, 1.05)) + return toned + + +def apply_lighting_variation(image, rng, illumination_profile): + width, height = image.size + y, x = np.mgrid[0:height, 0:width].astype(np.float32) + xn = (x / max(width - 1, 1)) - 0.5 + yn = (y / max(height - 1, 1)) - 0.5 + angle = rng.uniform(0.0, math.tau) + directional = math.cos(angle) * xn + math.sin(angle) * yn + directional *= (rng.uniform(0.28, 0.58) if illumination_profile == "bad" + else rng.uniform(0.10, 0.22)) + center_limit = 0.35 if illumination_profile == "bad" else 0.20 + cx_off, cy_off = rng.uniform(-center_limit, center_limit), rng.uniform(-center_limit, center_limit) + radius = np.sqrt((xn - cx_off) ** 2 + (yn - cy_off) ** 2) + vignette_strength = (rng.uniform(0.22, 0.44) if illumination_profile == "bad" + else rng.uniform(0.07, 0.18)) + vignette_power = 1.18 if illumination_profile == "bad" else 1.45 + vignette = -vignette_strength * np.clip(radius / 0.72, 0.0, 1.0) ** vignette_power + hw = (rng.uniform(0.08, 0.22) if illumination_profile == "bad" + else rng.uniform(0.18, 0.36)) + hh = (rng.uniform(0.08, 0.26) if illumination_profile == "bad" + else rng.uniform(0.18, 0.42)) + hotspot = np.exp(-(((xn - cx_off) ** 2) / hw + ((yn - cy_off) ** 2) / hh)) + hotspot *= (rng.uniform(0.08, 0.24) if illumination_profile == "bad" + else rng.uniform(0.03, 0.10)) + base_exposure = rng.uniform(0.68, 0.84) if illumination_profile == "bad" else 1.0 + illumination = base_exposure + directional + vignette + hotspot + if illumination_profile == "bad": + sa = rng.uniform(0.0, math.tau) + sax = math.cos(sa) * xn + math.sin(sa) * yn + cast = np.exp(-((sax - rng.uniform(-0.35, 0.35)) ** 2) / rng.uniform(0.010, 0.035)) + illumination -= cast * rng.uniform(0.10, 0.24) + sa2 = sa + rng.uniform(0.6, 1.4) + sax2 = math.cos(sa2) * xn + math.sin(sa2) * yn + cast2 = np.exp(-((sax2 - rng.uniform(-0.45, 0.45)) ** 2) / rng.uniform(0.020, 0.060)) + illumination -= cast2 * rng.uniform(0.04, 0.12) + lower, upper = (0.34, 1.42) if illumination_profile == "bad" else (0.72, 1.24) + illumination = np.clip(illumination, lower, upper) + arr = np.asarray(image).astype(np.float32) * illumination[..., None] + if illumination_profile == "bad": + cast_color = np.array(rng.choice([(1.08, 1.00, 0.86), (0.88, 0.96, 1.10), + (1.05, 0.94, 0.90)]), dtype=np.float32) + arr *= cast_color + arr += rng.uniform(-14.0, 2.0) + else: + arr += rng.uniform(-3.0, 4.0) + return Image.fromarray(np.clip(arr, 0, 255).astype(np.uint8), mode="RGB") + + +def apply_scanner_grain(image, rng, np_rng, illumination_profile): + arr = np.asarray(image).astype(np.float32) + height, width, _ = arr.shape + grain_sigma = (rng.uniform(7.0, 13.5) if illumination_profile == "bad" + else rng.uniform(3.2, 7.8)) + arr += np_rng.normal(0.0, grain_sigma, arr.shape).astype(np.float32) + row_sigma = (rng.uniform(2.4, 5.2) if illumination_profile == "bad" + else rng.uniform(1.2, 3.0)) + col_sigma = (rng.uniform(0.8, 2.2) if illumination_profile == "bad" + else rng.uniform(0.4, 1.2)) + arr += np_rng.normal(0.0, row_sigma, (height, 1, 1)).astype(np.float32) + arr += np_rng.normal(0.0, col_sigma, (1, width, 1)).astype(np.float32) + scanlines = np.sin(np.arange(height, dtype=np.float32)[:, None] / rng.uniform(3.5, 7.0)) + scan_strength = (rng.uniform(2.2, 5.0) if illumination_profile == "bad" + else rng.uniform(1.1, 2.8)) + arr += scanlines[..., None] * scan_strength + y, x = np.mgrid[0:height, 0:width].astype(np.float32) + diagonal = np.sin((x * rng.uniform(0.015, 0.028)) + (y * rng.uniform(0.006, 0.014))) + diag_strength = (rng.uniform(1.0, 3.0) if illumination_profile == "bad" + else rng.uniform(0.5, 1.8)) + arr += diagonal[..., None] * diag_strength + return Image.fromarray(np.clip(arr, 0, 255).astype(np.uint8), mode="RGB") + + +def jpeg_roundtrip(image, rng, illumination_profile): + quality_range = (70, 84) if illumination_profile == "bad" else (78, 88) + buffer = BytesIO() + image.save(buffer, format="JPEG", quality=rng.randint(*quality_range), + subsampling=rng.choice([1, 2]), optimize=False) + buffer.seek(0) + return Image.open(buffer).convert("RGB") + + +# ----- Sheet-on-surface composite (only used when sheet_surface=True) ----- + +def _apply_rgba_light_warp(image, rng): + arr = np.asarray(image.convert("RGBA"), dtype=np.float32) + height, width, _ = arr.shape + y, x = np.mgrid[0:height, 0:width].astype(np.float32) + xn, yn = x / max(width - 1, 1), y / max(height - 1, 1) + px, py = rng.uniform(0.0, math.tau), rng.uniform(0.0, math.tau) + dx = width * rng.uniform(0.0025, 0.0065) * np.sin((yn * math.tau * rng.uniform(1.1, 1.8)) + px) + dx += width * rng.uniform(0.0015, 0.0035) * np.sin(((xn + yn) * math.tau * rng.uniform(0.7, 1.3)) + py) + dy = height * rng.uniform(0.0020, 0.0050) * np.sin((xn * math.tau * rng.uniform(0.9, 1.5)) + py) + dy += height * rng.uniform(0.0010, 0.0025) * np.sin(((yn * 1.7 - xn * 0.4) * math.tau) + px) + src_x, src_y = x - dx, y - dy + valid = (src_x >= 0) & (src_x <= width - 1) & (src_y >= 0) & (src_y <= height - 1) + x0 = np.floor(np.clip(src_x, 0, width - 1)).astype(np.int32) + y0 = np.floor(np.clip(src_y, 0, height - 1)).astype(np.int32) + x1 = np.clip(x0 + 1, 0, width - 1) + y1 = np.clip(y0 + 1, 0, height - 1) + wx = (src_x - x0)[..., None] + wy = (src_y - y0)[..., None] + top = (arr[y0, x0] * (1.0 - wx)) + (arr[y0, x1] * wx) + bottom = (arr[y1, x0] * (1.0 - wx)) + (arr[y1, x1] * wx) + warped = (top * (1.0 - wy)) + (bottom * wy) + warped[..., 3] *= valid.astype(np.float32) + return Image.fromarray(np.clip(warped, 0, 255).astype(np.uint8), mode="RGBA") + + +def _apply_sheet_perspective(image, rng): + width, height = image.size + max_x = max(1.0, width * rng.uniform(0.010, 0.026)) + max_y = max(1.0, height * rng.uniform(0.010, 0.028)) + destination = [ + (rng.uniform(0, max_x), rng.uniform(0, max_y)), + (width - 1 - rng.uniform(0, max_x), rng.uniform(0, max_y)), + (width - 1 - rng.uniform(0, max_x), height - 1 - rng.uniform(0, max_y)), + (rng.uniform(0, max_x), height - 1 - rng.uniform(0, max_y)), + ] + source = [(0, 0), (width - 1, 0), (width - 1, height - 1), (0, height - 1)] + coeffs = _perspective_coefficients(source, destination) + return image.transform(image.size, Image.Transform.PERSPECTIVE, coeffs, + Image.Resampling.BICUBIC, fillcolor=(0, 0, 0, 0)) + + +def _make_surface_background(size, rng, np_rng): + width, height = size + base = np.array([rng.randint(166, 190), rng.randint(158, 181), + rng.randint(142, 166)], dtype=np.float32) + y, x = np.mgrid[0:height, 0:width].astype(np.float32) + xn = (x / max(width - 1, 1)) - 0.5 + yn = (y / max(height - 1, 1)) - 0.5 + angle = rng.uniform(0.0, math.tau) + directional = (math.cos(angle) * xn) + (math.sin(angle) * yn) + shade = 1.0 + directional * rng.uniform(0.10, 0.22) + shade -= np.clip(np.sqrt(xn**2 + yn**2) / 0.72, 0.0, 1.0) ** 1.35 * rng.uniform(0.06, 0.16) + low_noise = np_rng.normal(0.0, 1.0, (32, 32)).astype(np.float32) + noise_image = Image.fromarray(((low_noise - low_noise.min()) / + (np.ptp(low_noise) + 1e-6) * 255).astype(np.uint8)) + noise_image = noise_image.resize((width, height), Image.Resampling.BICUBIC) + low_noise = (np.asarray(noise_image).astype(np.float32) - 127.5) / 127.5 + fine_noise = np_rng.normal(0.0, 2.4, (height, width, 1)).astype(np.float32) + surface = base[None, None, :] * shade[..., None] + surface += low_noise[..., None] * rng.uniform(3.0, 7.5) + surface += fine_noise + return Image.fromarray(np.clip(surface, 0, 255).astype(np.uint8), mode="RGB") + + +def _add_paper_curl_shading(paper, rng, np_rng): + arr = np.asarray(paper.convert("RGBA"), dtype=np.float32) + height, width, _ = arr.shape + y, x = np.mgrid[0:height, 0:width].astype(np.float32) + xn, yn = x / max(width - 1, 1), y / max(height - 1, 1) + left_curl = np.exp(-xn / rng.uniform(0.035, 0.070)) + right_curl = np.exp(-(1.0 - xn) / rng.uniform(0.040, 0.080)) + top_roll = np.exp(-yn / rng.uniform(0.060, 0.110)) + bottom_roll = np.exp(-(1.0 - yn) / rng.uniform(0.060, 0.120)) + shade = 1.0 + shade += left_curl * rng.uniform(0.025, 0.060) + shade -= right_curl * rng.uniform(0.020, 0.055) + shade += top_roll * np.sin((xn * math.tau * rng.uniform(0.75, 1.35)) + rng.uniform(0, math.tau)) * 0.025 + shade -= bottom_roll * rng.uniform(0.020, 0.052) + shade += np.sin((yn * math.tau * rng.uniform(1.2, 2.1)) + rng.uniform(0, math.tau)) * rng.uniform(0.006, 0.014) + paper_fiber = np_rng.normal(0.0, 1.6, (height, width, 1)).astype(np.float32) + arr[..., :3] *= shade[..., None] + arr[..., :3] += paper_fiber + return Image.fromarray(np.clip(arr, 0, 255).astype(np.uint8), mode="RGBA") + + +def _apply_surface_lighting(image, rng): + arr = np.asarray(image.convert("RGB"), dtype=np.float32) + height, width, _ = arr.shape + y, x = np.mgrid[0:height, 0:width].astype(np.float32) + xn = (x / max(width - 1, 1)) - 0.5 + yn = (y / max(height - 1, 1)) - 0.5 + angle = rng.uniform(0.0, math.tau) + gradient = (math.cos(angle) * xn) + (math.sin(angle) * yn) + cx_off = rng.uniform(-0.20, 0.20) + cy_off = rng.uniform(-0.20, 0.20) + hotspot = np.exp(-(((xn - cx_off) ** 2) / 0.14 + ((yn - cy_off) ** 2) / 0.22)) + shade = 1.0 + gradient * rng.uniform(0.08, 0.17) + shade += hotspot * rng.uniform(0.025, 0.070) + shade -= np.clip(np.sqrt(xn**2 + yn**2) / 0.74, 0.0, 1.0) ** 1.6 * rng.uniform(0.05, 0.13) + arr *= np.clip(shade, 0.72, 1.24)[..., None] + return Image.fromarray(np.clip(arr, 0, 255).astype(np.uint8), mode="RGB") + + +def apply_sheet_surface_effect(image, rng, np_rng): + width, height = image.size + surface = _make_surface_background((width, height), rng, np_rng).convert("RGBA") + scale = rng.uniform(0.88, 0.94) + sheet_w = max(8, int(width * scale)) + sheet_h = max(8, int(height * scale)) + paper_rgb = (rng.randint(234, 246), rng.randint(231, 243), rng.randint(218, 235)) + paper = Image.new("RGBA", (sheet_w, sheet_h), paper_rgb + (255,)) + inset = max(3, int(min(sheet_w, sheet_h) * rng.uniform(0.010, 0.022))) + content_w = max(2, sheet_w - (inset * 2)) + content_h = max(2, sheet_h - (inset * 2)) + content = image.convert("RGB").resize((content_w, content_h), Image.Resampling.BICUBIC) + paper.alpha_composite(content.convert("RGBA"), (inset, inset)) + paper = _add_paper_curl_shading(paper, rng, np_rng) + sheet_layer = Image.new("RGBA", (width, height), (0, 0, 0, 0)) + left = (width - sheet_w) // 2 + rng.randint(-max(1, width // 90), max(1, width // 90)) + top = (height - sheet_h) // 2 + rng.randint(-max(1, height // 90), max(1, height // 90)) + sheet_layer.alpha_composite(paper, (left, top)) + sheet_layer = _apply_rgba_light_warp(sheet_layer, rng) + sheet_layer = _apply_sheet_perspective(sheet_layer, rng) + alpha = sheet_layer.getchannel("A") + shadow_offset = (rng.randint(max(1, width // 180), max(2, width // 70)), + rng.randint(max(1, height // 180), max(2, height // 70))) + shadow_alpha = alpha.filter(ImageFilter.GaussianBlur( + radius=max(2.0, min(width, height) * 0.018))) + shadow_alpha = ImageEnhance.Brightness(shadow_alpha).enhance(rng.uniform(0.26, 0.42)) + shifted_alpha = Image.new("L", (width, height), 0) + sx, sy = shadow_offset + shifted_alpha.paste(shadow_alpha.crop((0, 0, max(1, width - sx), max(1, height - sy))), + shadow_offset) + shadow = Image.new("RGBA", (width, height), (0, 0, 0, 0)) + shadow.putalpha(shifted_alpha) + composed = Image.alpha_composite(surface, shadow) + composed = Image.alpha_composite(composed, sheet_layer) + composed = _apply_surface_lighting(composed.convert("RGB"), rng) + composed = composed.filter(ImageFilter.GaussianBlur(radius=rng.uniform(0.06, 0.18))) + return composed + + +# ===================================================================== +# Top-level orchestrator +# ===================================================================== + +def process_image(image: Image.Image, seed: int, config: EffectsConfig) -> Image.Image: + """Run the configured chain of effects on a single image.""" + rng = random.Random(seed) + np_rng = np.random.default_rng(seed) + + if config.reverse_dark_background: + image, _ = reverse_dark_background_colors( + image, config.paper_luma_threshold, config.paper_coverage_threshold) + image = image.convert("RGB") if image.mode != "RGB" else image + + paper_rgb = (rng.randint(241, 249), rng.randint(238, 247), rng.randint(229, 241)) + + if config.composite_on_paper: + image = composite_on_paper(image, paper_rgb) + if config.color_and_tone: + image = apply_color_and_tone(image, rng) + if config.perspective_capture: + image = apply_perspective_capture(image, rng, paper_rgb) + if config.pre_blur: + radius = (rng.uniform(0.60, 1.35) if config.illumination_profile == "bad" + else rng.uniform(0.35, 0.95)) + image = image.filter(ImageFilter.GaussianBlur(radius=radius)) + if config.lighting_variation: + image = apply_lighting_variation(image, rng, config.illumination_profile) + if config.scanner_grain: + image = apply_scanner_grain(image, rng, np_rng, config.illumination_profile) + if config.post_blur: + image = image.filter(ImageFilter.GaussianBlur(radius=rng.uniform(0.08, 0.28))) + if config.jpeg_roundtrip: + image = jpeg_roundtrip(image, rng, config.illumination_profile) + if config.sheet_surface: + image = apply_sheet_surface_effect(image, rng, np_rng) + if config.jpeg_roundtrip: + image = jpeg_roundtrip(image, rng, config.illumination_profile) + return image + + +# ===================================================================== +# CLI driver +# ===================================================================== + +@dataclass(frozen=True) +class OutputRecord: + source: str + output: str + seed: int + width: int + height: int + config: dict + + +def _build_config_from_args(args: argparse.Namespace) -> EffectsConfig: + return EffectsConfig( + reverse_dark_background=args.reverse_dark_background, + composite_on_paper=args.composite_on_paper, + color_and_tone=args.color_and_tone, + perspective_capture=args.perspective_capture, + pre_blur=args.pre_blur, + lighting_variation=args.lighting_variation, + scanner_grain=args.scanner_grain, + post_blur=args.post_blur, + jpeg_roundtrip=args.jpeg_roundtrip, + sheet_surface=args.sheet_surface, + illumination_profile=args.illumination_profile, + ) + + +def parse_args() -> argparse.Namespace: + parser = argparse.ArgumentParser(description=__doc__) + parser.add_argument("--images-dir", type=Path, default=DEFAULT_IMAGES_DIR) + parser.add_argument("--output-dir", type=Path, default=DEFAULT_OUTPUT_DIR) + # First seed = 20260424 + 5 * 1009 = 20265469 (matches the original + # generated_bad_illumination_reversed_colors_sheet_surface manifest). + parser.add_argument("--seed", type=int, default=20265469) + parser.add_argument("--overwrite", action="store_true") + parser.add_argument("--manifest", type=Path, default=None) + parser.add_argument("--illumination-profile", + choices=("standard", "bad"), default="bad") + + defaults = EffectsConfig() + for name in ( + "reverse_dark_background", "composite_on_paper", "color_and_tone", + "perspective_capture", "pre_blur", "lighting_variation", + "scanner_grain", "post_blur", "jpeg_roundtrip", "sheet_surface", + ): + flag = name.replace("_", "-") + default = getattr(defaults, name) + parser.add_argument(f"--{flag}", dest=name, action="store_true", + default=default) + parser.add_argument(f"--no-{flag}", dest=name, action="store_false") + return parser.parse_args() + + +def main() -> None: + args = parse_args() + config = _build_config_from_args(args) + + images_dir = args.images_dir + if not images_dir.exists(): + raise SystemExit(f"Input directory does not exist: {images_dir}") + output_dir = args.output_dir + output_dir.mkdir(parents=True, exist_ok=True) + + sources = sorted( + path for path in images_dir.iterdir() + if path.is_file() and path.suffix.lower() in IMAGE_EXTENSIONS + and not path.stem.endswith("_captured") + ) + if not sources: + raise SystemExit(f"No source images found in {images_dir}") + + records: list[OutputRecord] = [] + for index, source_path in enumerate(sources): + seed = args.seed + index + output_path = output_dir / f"{source_path.stem}{OUTPUT_SUFFIX}" + if output_path.exists() and not args.overwrite: + with Image.open(output_path) as existing: + width, height = existing.size + else: + with Image.open(source_path) as image: + result = process_image(image, seed, config) + result.save(output_path, format="PNG", compress_level=4) + width, height = result.size + records.append(OutputRecord( + source=source_path.as_posix(), + output=output_path.as_posix(), + seed=seed, width=width, height=height, + config=asdict(config), + )) + print(f"wrote {output_path.name} (seed={seed})") + + manifest_path = args.manifest or (output_dir / "manifest_captured.json") + manifest = { + "effect": "spot_the_field_diff_pipeline", + "images_dir": images_dir.as_posix(), + "output_dir": output_dir.as_posix(), + "config": asdict(config), + "count": len(records), + "records": [asdict(r) for r in records], + } + manifest_path.write_text(json.dumps(manifest, indent=2) + "\n", encoding="utf-8") + print(f"Wrote manifest to {manifest_path}") + + +if __name__ == "__main__": + main() diff --git a/code/visual_attribute_transfer/spot_the_field_diff/creation.py b/code/visual_attribute_transfer/spot_the_field_diff/creation.py new file mode 100644 index 0000000000000000000000000000000000000000..5d67a6fe2510c511e9380159640b66ed1526a887 --- /dev/null +++ b/code/visual_attribute_transfer/spot_the_field_diff/creation.py @@ -0,0 +1,650 @@ +"""Generate 'Spot the Field Diff' visual benchmark dataset. + +Two 2D color fields are shown side by side. Each sample uses a randomly- +chosen colour mode: a scalar 2D multi-scale noise field rendered through +one of several colormaps (viridis / plasma / inferno / magma / cividis / +coolwarm), OR a full-RGB variant where three independent scalar fields are +stacked directly as red, green and blue channels. + +Each panel is divided into a 3×3 grid of 9 cells. Independently for each +cell, the right panel may or may not have a radially-faded perturbation +applied at the cell's centre. At the perturbation's rim the right panel +matches the left exactly (seamless), and at the centre its content is +blended toward a freshly-drawn, DC-matched field so only the texture +differs — there is no visible "dot" of uniform offset. + +Task: count how many of the 9 cells are identical between the two panels +(i.e. how many cells have NO perturbation). Answer is an integer 0..9. +""" +from __future__ import annotations + +import argparse +import json +import random +from pathlib import Path +from typing import Any, Dict, List, Tuple + +import matplotlib + +matplotlib.use("Agg") +import matplotlib.pyplot as plt +import matplotlib.patches as mpatches +import numpy as np +from tqdm import tqdm + + +# --------------------------------------------------------------------------- +# Constants +# --------------------------------------------------------------------------- + +GRID_N = 3 # 3×3 grid → 9 cells (difficulty-tunable) +CELL_W = 200 # pixels per cell (fixed; panel auto-scales) +CELL_H = 200 +PANEL_W = CELL_W * GRID_N # pixels per panel (square field) +PANEL_H = CELL_H * GRID_N + +# Radius of the (optional) circular perturbation centred inside each cell. +# Must stay within the cell so perturbations don't cross cell boundaries. +CIRCLE_RADIUS = 70 +assert CIRCLE_RADIUS < min(CELL_W, CELL_H) // 2 + +# Visibility thresholds for the blended-fresh texture inside a perturbed +# cell. Peak and mean |fresh − orig| (after DC-matching) must exceed these +# multiples of the field std; otherwise the fresh field is resampled. +MIN_BLEND_DIFF_MULT = 2.6 +MIN_BLEND_MEAN_DIFF_MULT = 1.0 + +# Frequency scale multiplier on generate_field sigmas (difficulty-tunable). +# Smaller value → finer-grained texture, harder task. +FIELD_FREQUENCY_SCALE = 1.0 + +# Rendering modes — one is picked at random per sample. +COLOR_MODES: List[str] = [ + "viridis", "plasma", "inferno", "magma", "cividis", "coolwarm", "rgb", +] +JITTER_AMPLITUDE = 0.012 + +BG_COLOR = "#ffffff" +BORDER_COLOR = "#888888" +BORDER_WIDTH = 1.5 +GRID_LINE_COLOR = "#ffffff" +GRID_LINE_WIDTH = 1.6 +GRID_LINE_ALPHA = 0.85 +LABEL_COLOR = "#333333" +HIGHLIGHT_COLOR = "#dc1e1e" + +MARGIN_PX = 70 +GAP_PX = 90 +LABEL_HEIGHT = 44 + +def _build_question() -> str: + n = GRID_N + total = n * n + return ( + f"Three panels are shown side by side. The leftmost panel is an INDEX " + f"key: a {n}×{n} grid whose cells are numbered 1 through {total} in " + f"row-major order (top-left = 1, increasing left-to-right then " + f"top-to-bottom). The middle and right panels are 2D color fields, " + f"each divided into the same {n}×{n} grid by thin white gridlines. " + f"Compare the middle (Field A) and right (Field B) panels cell by " + f"cell. List the numbers of all cells that DIFFER between Field A " + f"and Field B, in ascending order, separated by commas (for example: " + f"\"1, 4, {total}\"). If no cells differ, write \"none\". " + f"Provide your final answer enclosed in ... tags." + ) + + +QUESTION = ( + "Two panels are shown side by side. Each panel is a 2D color field, and " + "each panel is divided into a 3×3 grid of 9 cells by thin white " + "gridlines. Compare the left and right panels cell by cell. Count how " + "many of the 9 cells are DIFFERENT between the two panels. Report the " + "count as an integer between 0 and 9. " + "Provide your final answer enclosed in ... tags." +) + + +# --------------------------------------------------------------------------- +# 2D field generator (multi-scale smoothed Gaussian noise) +# --------------------------------------------------------------------------- + + +def _separable_gaussian_blur(arr: np.ndarray, sigma: float) -> np.ndarray: + """Apply a separable 2D Gaussian blur via two 1D convolutions.""" + max_radius = max(1, min(arr.shape) // 2 - 1) + radius = max(1, min(int(sigma * 4), max_radius)) + k = np.arange(-radius, radius + 1, dtype=np.float64) + w = np.exp(-(k * k) / (2.0 * sigma * sigma)) + kernel = w / w.sum() + out = np.apply_along_axis( + lambda row: np.convolve(row, kernel, mode="same"), axis=1, arr=arr + ) + out = np.apply_along_axis( + lambda col: np.convolve(col, kernel, mode="same"), axis=0, arr=out + ) + return out + + +def _smoothed_noise_2d( + np_rng: np.random.Generator, + shape: Tuple[int, int], + sigma: float, +) -> np.ndarray: + raw = np_rng.standard_normal(size=shape) + return _separable_gaussian_blur(raw, sigma) + + +def generate_field( + np_rng: np.random.Generator, + shape: Tuple[int, int], +) -> np.ndarray: + """Arbitrary non-periodic 2D random field built from multi-scale smoothed + Gaussian noise.""" + fs = max(0.1, float(FIELD_FREQUENCY_SCALE)) + coarse = _smoothed_noise_2d(np_rng, shape, sigma=np_rng.uniform(28.0, 44.0) * fs) + medium = _smoothed_noise_2d(np_rng, shape, sigma=np_rng.uniform(9.0, 16.0) * fs) + fine = _smoothed_noise_2d(np_rng, shape, sigma=np_rng.uniform(2.5, 4.5) * fs) + micro = _smoothed_noise_2d(np_rng, shape, sigma=np_rng.uniform(0.9, 1.4) * fs) + + def _norm(a: np.ndarray) -> np.ndarray: + s = float(np.std(a)) or 1.0 + return a / s + + f = ( + 1.8 * _norm(coarse) + + 0.75 * _norm(medium) + + 0.40 * _norm(fine) + + 0.20 * _norm(micro) + ) + f = f + np_rng.normal(0.0, 0.04, size=shape) + return f + + +# --------------------------------------------------------------------------- +# Cell geometry and perturbation selection +# --------------------------------------------------------------------------- + + +def _cell_center(row: int, col: int) -> Tuple[int, int]: + cx = col * CELL_W + CELL_W // 2 + cy = row * CELL_H + CELL_H // 2 + return cx, cy + + +MIN_PERTURBED = 0 +MAX_PERTURBED = GRID_N * GRID_N + + +def _sample_perturbed_cells(rng: random.Random) -> List[Tuple[int, int]]: + """Return list of (row, col) cells to perturb. Draws ``num_perturbed`` + uniformly from [MIN_PERTURBED..MAX_PERTURBED] then picks that many cells.""" + num_perturbed = rng.randint(MIN_PERTURBED, MAX_PERTURBED) + all_cells = [(r, c) for r in range(GRID_N) for c in range(GRID_N)] + rng.shuffle(all_cells) + return sorted(all_cells[:num_perturbed]) + + +# --------------------------------------------------------------------------- +# Radial-fade perturbation (circle-shaped, magnitude fades to zero at rim) +# --------------------------------------------------------------------------- + + +def _build_disk_mask(radius: int) -> np.ndarray: + """Smootherstep radial fade mask, exactly 0 at the rim and C²-smooth.""" + n = 2 * radius + 1 + yy, xx = np.mgrid[0:n, 0:n].astype(np.float64) + c = float(radius) + dist = np.sqrt((xx - c) ** 2 + (yy - c) ** 2) + t = np.clip(dist / float(radius), 0.0, 1.0) + mask = 1.0 - (6.0 * t ** 5 - 15.0 * t ** 4 + 10.0 * t ** 3) + return mask + + +def _apply_radial_perturbation( + np_rng: np.random.Generator, + left: np.ndarray, + right: np.ndarray, + cx: int, + cy: int, + radius: int, + field_std: float, + max_tries: int = 40, +) -> None: + r = radius + n = 2 * r + 1 + y_lo, y_hi = cy - r, cy + r + 1 + x_lo, x_hi = cx - r, cx + r + 1 + orig = left[y_lo:y_hi, x_lo:x_hi] + + mask = _build_disk_mask(r) + min_peak = MIN_BLEND_DIFF_MULT * field_std + min_mean = MIN_BLEND_MEAN_DIFF_MULT * field_std + inner = mask > 0.5 + + pad_canvas = n + 60 + pad_off = (pad_canvas - n) // 2 + + for _ in range(max_tries): + fresh_full = generate_field(np_rng, shape=(pad_canvas, pad_canvas)) + fresh = fresh_full[pad_off:pad_off + n, pad_off:pad_off + n].copy() + + # DC-match: remove bulk brightness offset so only texture differs. + fresh = fresh - (float(np.mean(fresh[inner])) - float(np.mean(orig[inner]))) + + abs_diff = np.abs(fresh[inner] - orig[inner]) + if float(np.max(abs_diff)) < min_peak: + continue + if float(np.mean(abs_diff)) < min_mean: + continue + + right[y_lo:y_hi, x_lo:x_hi] = (1.0 - mask) * orig + mask * fresh + return + + raise RuntimeError("Could not find a sufficiently different fresh field for this cell.") + + +def build_sample( + rng: random.Random, + np_rng: np.random.Generator, + mode: str, +) -> Tuple[np.ndarray, np.ndarray, List[Dict[str, Any]]]: + """Return ``(left, right, cell_records)``. One record per cell with + ``row``, ``col``, ``perturbed`` (bool), ``cx``, ``cy``, ``radius``.""" + n_channels = 3 if mode == "rgb" else 1 + + channels_left = [ + generate_field(np_rng, shape=(PANEL_H, PANEL_W)) for _ in range(n_channels) + ] + channels_right = [c.copy() for c in channels_left] + + perturbed = set(_sample_perturbed_cells(rng)) + + cell_records: List[Dict[str, Any]] = [] + for row in range(GRID_N): + for col in range(GRID_N): + cx, cy = _cell_center(row, col) + is_perturbed = (row, col) in perturbed + if is_perturbed: + for ch_left, ch_right in zip(channels_left, channels_right): + ch_std = float(np.std(ch_left)) or 1.0 + _apply_radial_perturbation( + np_rng, ch_left, ch_right, cx, cy, CIRCLE_RADIUS, ch_std, + ) + cell_records.append({ + "row": row, + "col": col, + "perturbed": is_perturbed, + "cx": int(cx), + "cy": int(cy), + "radius": int(CIRCLE_RADIUS), + }) + + if mode == "rgb": + left = np.stack(channels_left, axis=-1) + right = np.stack(channels_right, axis=-1) + else: + left = channels_left[0] + right = channels_right[0] + return left, right, cell_records + + +# --------------------------------------------------------------------------- +# Rendering +# --------------------------------------------------------------------------- + + +LEGEND_SCALE = 0.4 # legend is 40% the dimensions of a field panel + + +def _legend_dims() -> Tuple[int, int]: + return int(round(PANEL_W * LEGEND_SCALE)), int(round(PANEL_H * LEGEND_SCALE)) + + +def _panel_origins() -> Tuple[Tuple[float, float], + Tuple[float, float], + Tuple[float, float]]: + """(ox, oy) for legend (small, vertically centered), left field, right + field. Layout: [legend] [GAP] [Field A] [GAP] [Field B].""" + oy_field = MARGIN_PX + LABEL_HEIGHT + leg_w, leg_h = _legend_dims() + ox_legend = MARGIN_PX + oy_legend = oy_field + (PANEL_H - leg_h) / 2 + ox_left = ox_legend + leg_w + GAP_PX + ox_right = ox_left + PANEL_W + GAP_PX + return (ox_legend, oy_legend), (ox_left, oy_field), (ox_right, oy_field) + + +def _canvas_size() -> Tuple[int, int]: + leg_w, _ = _legend_dims() + w = MARGIN_PX + leg_w + GAP_PX + PANEL_W + GAP_PX + PANEL_W + MARGIN_PX + h = MARGIN_PX + LABEL_HEIGHT + PANEL_H + MARGIN_PX + return int(w), int(h) + + +def _draw_legend(ax: plt.Axes, ox: float, oy: float) -> None: + """Small index-key panel: light grey background, N×N gridlines, each + cell labeled with its 1-based row-major number in bold font.""" + leg_w, leg_h = _legend_dims() + cell_w = leg_w / GRID_N + cell_h = leg_h / GRID_N + legend_bg = mpatches.Rectangle( + (ox, oy), leg_w, leg_h, + facecolor="#e8e6e0", edgecolor=BORDER_COLOR, + linewidth=BORDER_WIDTH, zorder=2, + ) + ax.add_patch(legend_bg) + for i in range(1, GRID_N): + gx = ox + i * cell_w + ax.plot([gx, gx], [oy, oy + leg_h], + color="#888", linewidth=GRID_LINE_WIDTH, zorder=3) + gy = oy + i * cell_h + ax.plot([ox, ox + leg_w], [gy, gy], + color="#888", linewidth=GRID_LINE_WIDTH, zorder=3) + fontsize = max(10, int(min(cell_w, cell_h) * 0.36)) + for r in range(GRID_N): + for c in range(GRID_N): + n = r * GRID_N + c + 1 + cx = ox + c * cell_w + cell_w / 2 + cy = oy + r * cell_h + cell_h / 2 + ax.text(cx, cy, str(n), + ha="center", va="center", + fontsize=fontsize, fontweight="bold", + color="#222", zorder=4) + + +def _normalise(field: np.ndarray, v_lo: float, v_hi: float) -> np.ndarray: + if v_hi <= v_lo: + return np.zeros_like(field, dtype=np.float64) + return (field - v_lo) / (v_hi - v_lo) + + +def _field_to_rgba( + field: np.ndarray, + mode: str, + v_lo: Any, + v_hi: Any, + np_rng: np.random.Generator, +) -> np.ndarray: + if mode == "rgb": + norm = np.stack( + [_normalise(field[..., c], v_lo[c], v_hi[c]) for c in range(3)], + axis=-1, + ) + norm = norm + np_rng.uniform(-JITTER_AMPLITUDE, JITTER_AMPLITUDE, size=norm.shape) + norm = np.clip(norm, 0.0, 1.0) + alpha = np.ones(norm.shape[:2] + (1,), dtype=norm.dtype) + rgba = np.concatenate([norm, alpha], axis=-1) + else: + norm = _normalise(field, v_lo, v_hi) + norm = norm + np_rng.uniform(-JITTER_AMPLITUDE, JITTER_AMPLITUDE, size=norm.shape) + norm = np.clip(norm, 0.0, 1.0) + rgba = plt.get_cmap(mode)(norm) + + return (rgba * 255.0 + 0.5).astype(np.uint8) + + +def _draw_grid(ax: plt.Axes, ox: float, oy: float) -> None: + """Draw the 3×3 white gridlines on a panel (2 vertical + 2 horizontal).""" + for i in range(1, GRID_N): + gx = ox + i * CELL_W + ax.plot( + [gx, gx], [oy, oy + PANEL_H], + color=GRID_LINE_COLOR, linewidth=GRID_LINE_WIDTH, + alpha=GRID_LINE_ALPHA, zorder=4, + ) + gy = oy + i * CELL_H + ax.plot( + [ox, ox + PANEL_W], [gy, gy], + color=GRID_LINE_COLOR, linewidth=GRID_LINE_WIDTH, + alpha=GRID_LINE_ALPHA, zorder=4, + ) + + +def _render( + out_path: Path, + left: np.ndarray, + right: np.ndarray, + mode: str, + np_rng: np.random.Generator, + cell_records: List[Dict[str, Any]] | None = None, +) -> None: + w, h = _canvas_size() + dpi = 100 + fig, ax = plt.subplots(1, 1, figsize=(w / dpi, h / dpi), dpi=dpi) + ax.set_xlim(0, w) + ax.set_ylim(h, 0) + ax.set_aspect("auto") + ax.axis("off") + fig.patch.set_facecolor(BG_COLOR) + ax.set_facecolor(BG_COLOR) + + (ox_legend, oy_legend), (ox_left, oy), (ox_right, _) = _panel_origins() + _draw_legend(ax, ox_legend, oy_legend) + leg_w, _ = _legend_dims() + ax.text( + ox_legend + leg_w / 2, oy_legend - 12, + "Index", ha="center", va="bottom", + fontsize=13, fontweight="bold", color=LABEL_COLOR, + ) + + if mode == "rgb": + v_lo, v_hi = [], [] + for c in range(3): + both = np.concatenate([left[..., c].ravel(), right[..., c].ravel()]) + v_lo.append(float(np.quantile(both, 0.01))) + v_hi.append(float(np.quantile(both, 0.99))) + else: + both = np.concatenate([left.ravel(), right.ravel()]) + v_lo = float(np.quantile(both, 0.01)) + v_hi = float(np.quantile(both, 0.99)) + + left_rgba = _field_to_rgba(left, mode, v_lo, v_hi, np_rng) + right_rgba = _field_to_rgba(right, mode, v_lo, v_hi, np_rng) + + ax.imshow( + left_rgba, + extent=(ox_left, ox_left + PANEL_W, oy + PANEL_H, oy), + interpolation="nearest", + zorder=2, + ) + ax.imshow( + right_rgba, + extent=(ox_right, ox_right + PANEL_W, oy + PANEL_H, oy), + interpolation="nearest", + zorder=2, + ) + + for ox in (ox_left, ox_right): + border = mpatches.Rectangle( + (ox, oy), PANEL_W, PANEL_H, + facecolor="none", edgecolor=BORDER_COLOR, + linewidth=BORDER_WIDTH, zorder=3, + ) + ax.add_patch(border) + _draw_grid(ax, ox, oy) + + ax.text( + ox_left + PANEL_W / 2, MARGIN_PX + LABEL_HEIGHT * 0.5, + "Field A", ha="center", va="center", + fontsize=15, fontweight="bold", color=LABEL_COLOR, + ) + ax.text( + ox_right + PANEL_W / 2, MARGIN_PX + LABEL_HEIGHT * 0.5, + "Field B", ha="center", va="center", + fontsize=15, fontweight="bold", color=LABEL_COLOR, + ) + + if cell_records: + for rec in cell_records: + if not rec["perturbed"]: + continue + cx, cy, rr = rec["cx"], rec["cy"], rec["radius"] + for ox in (ox_left, ox_right): + ax.add_patch(mpatches.Circle( + (ox + cx, oy + cy), rr, + facecolor="none", edgecolor=HIGHLIGHT_COLOR, + linewidth=2.2, zorder=5, + )) + + fig.savefig(out_path, facecolor=BG_COLOR) + plt.close(fig) + + +def render_pair( + out_path: Path, + left: np.ndarray, + right: np.ndarray, + mode: str, + np_rng: np.random.Generator, +) -> None: + _render(out_path, left, right, mode, np_rng) + + +def render_answer( + out_path: Path, + left: np.ndarray, + right: np.ndarray, + cell_records: List[Dict[str, Any]], + mode: str, + np_rng: np.random.Generator, +) -> None: + _render(out_path, left, right, mode, np_rng, cell_records) + + +# --------------------------------------------------------------------------- +# Annotation +# --------------------------------------------------------------------------- + + +def build_annotation( + image_name: str, + cell_records: List[Dict[str, Any]], + mode: str, +) -> Dict[str, Any]: + num_identical = sum(1 for r in cell_records if not r["perturbed"]) + num_different = GRID_N * GRID_N - num_identical + diff_indices = sorted( + r["row"] * GRID_N + r["col"] + 1 + for r in cell_records if r["perturbed"] + ) + answer = ", ".join(str(i) for i in diff_indices) if diff_indices else "none" + return { + "image": image_name, + "color_mode": mode, + "grid": GRID_N, + "num_cells": GRID_N * GRID_N, + "num_identical": num_identical, + "num_different": num_different, + "diff_indices": diff_indices, + "cells": cell_records, + "question": _build_question(), + "answer": answer, + } + + +# --------------------------------------------------------------------------- +# Dataset generation +# --------------------------------------------------------------------------- + + +def generate_dataset( + rng: random.Random, + np_rng: np.random.Generator, + count: int, + output_dir: Path, +) -> None: + images_dir = output_dir / "images" + answers_dir = output_dir / "answers" + images_dir.mkdir(parents=True, exist_ok=True) + answers_dir.mkdir(parents=True, exist_ok=True) + + annotations: List[Dict[str, Any]] = [] + data_items: List[Dict[str, Any]] = [] + + for idx in tqdm(range(count), desc="Generating field diff pairs"): + mode = rng.choice(COLOR_MODES) + for _ in range(20): + try: + left, right, cell_records = build_sample(rng, np_rng, mode) + break + except RuntimeError: + continue + else: + raise RuntimeError(f"Failed to build sample {idx}") + + image_name = f"field_diff_{idx:05d}.png" + img_path = images_dir / image_name + ans_path = answers_dir / image_name + + render_pair(img_path, left, right, mode, np_rng) + render_answer(ans_path, left, right, cell_records, mode, np_rng) + + rel_image = f"images/{image_name}" + diff_indices = sorted( + r["row"] * GRID_N + r["col"] + 1 + for r in cell_records if r["perturbed"] + ) + answer = ", ".join(str(i) for i in diff_indices) if diff_indices else "none" + annotations.append(build_annotation(rel_image, cell_records, mode)) + data_items.append({ + "image": rel_image, + "question": _build_question(), + "answer": answer, + }) + + with (output_dir / "annotations.jsonl").open("w", encoding="utf-8") as fh: + for rec in annotations: + fh.write(json.dumps(rec) + "\n") + + data_json = { + "task": "spot_the_field_diff", + "category": "visual_attribute_transfer", + "count": len(data_items), + "items": data_items, + } + with (output_dir / "data.json").open("w", encoding="utf-8") as fh: + json.dump(data_json, fh, indent=2) + fh.write("\n") + + +# --------------------------------------------------------------------------- +# CLI +# --------------------------------------------------------------------------- + + +def parse_args() -> argparse.Namespace: + p = argparse.ArgumentParser( + description="Generate 'Spot the Field Diff' visual benchmark dataset." + ) + p.add_argument("--output-root", type=Path, default=".") + p.add_argument("--count", type=int, default=20) + p.add_argument("--seed", type=int, default=42) + p.add_argument("--difficulty", type=int, default=5, + help="Integer difficulty >=0; scales perturbed cells and blend subtlety.") + return p.parse_args() + + +def main() -> None: + args = parse_args() + rng = random.Random(args.seed) + np_rng = np.random.default_rng(args.seed) + d = max(0, int(args.difficulty)) + + global GRID_N, PANEL_W, PANEL_H, MIN_PERTURBED, MAX_PERTURBED + global MIN_BLEND_DIFF_MULT, MIN_BLEND_MEAN_DIFF_MULT, FIELD_FREQUENCY_SCALE + + GRID_N = 3 + d // 3 + PANEL_W = CELL_W * GRID_N + PANEL_H = CELL_H * GRID_N + MIN_PERTURBED = 1 + MAX_PERTURBED = GRID_N * GRID_N + MIN_BLEND_DIFF_MULT = 3.0 - 0.1 * d + MIN_BLEND_MEAN_DIFF_MULT = max(0.5, 1.0 - 0.05 * d) + FIELD_FREQUENCY_SCALE = max(0.4, 1.0 - 0.08 * d) + + generate_dataset(rng, np_rng, args.count, args.output_root) + print(f"Saved {args.count} field diff pairs to {args.output_root}") + + +if __name__ == "__main__": + main() diff --git a/code/visual_attribute_transfer/spot_the_field_diff/data.json b/code/visual_attribute_transfer/spot_the_field_diff/data.json new file mode 100644 index 0000000000000000000000000000000000000000..928b1db4c0af3d326c214824c5aad0bd3abda7ff --- /dev/null +++ b/code/visual_attribute_transfer/spot_the_field_diff/data.json @@ -0,0 +1,32 @@ +{ + "task": "spot_the_field_diff", + "category": "visual_attribute_transfer", + "count": 5, + "items": [ + { + "image": "images_with_effects/field_diff_00000_captured.png", + "question": "Three panels are shown side by side. The leftmost panel is an INDEX key: a grid of cells numbered in row-major order (top-left = 1, increasing left-to-right then top-to-bottom). The middle and right panels are 2D color fields, each divided into the same grid. Compare the middle (Field A) and right (Field B) panels cell by cell. List the numbers of all cells that DIFFER between Field A and Field B, in ascending order, separated by commas. If no cells differ, write \"none\". Provide your final answer enclosed in ... tags.", + "answer": "6, 7, 8, 11" + }, + { + "image": "images_with_effects/field_diff_00001_captured.png", + "question": "Three panels are shown side by side. The leftmost panel is an INDEX key: a grid of cells numbered in row-major order (top-left = 1, increasing left-to-right then top-to-bottom). The middle and right panels are 2D color fields, each divided into the same grid. Compare the middle (Field A) and right (Field B) panels cell by cell. List the numbers of all cells that DIFFER between Field A and Field B, in ascending order, separated by commas. If no cells differ, write \"none\". Provide your final answer enclosed in ... tags.", + "answer": "6" + }, + { + "image": "images_with_effects/field_diff_00002_captured.png", + "question": "Three panels are shown side by side. The leftmost panel is an INDEX key: a grid of cells numbered in row-major order (top-left = 1, increasing left-to-right then top-to-bottom). The middle and right panels are 2D color fields, each divided into the same grid. Compare the middle (Field A) and right (Field B) panels cell by cell. List the numbers of all cells that DIFFER between Field A and Field B, in ascending order, separated by commas. If no cells differ, write \"none\". Provide your final answer enclosed in ... tags.", + "answer": "3, 4, 8, 9, 10, 12, 13" + }, + { + "image": "images_with_effects/field_diff_00003_captured.png", + "question": "Three panels are shown side by side. The leftmost panel is an INDEX key: a grid of cells numbered in row-major order (top-left = 1, increasing left-to-right then top-to-bottom). The middle and right panels are 2D color fields, each divided into the same grid. Compare the middle (Field A) and right (Field B) panels cell by cell. List the numbers of all cells that DIFFER between Field A and Field B, in ascending order, separated by commas. If no cells differ, write \"none\". Provide your final answer enclosed in ... tags.", + "answer": "3, 5, 6, 7, 8, 9, 11, 13, 14, 15" + }, + { + "image": "images_with_effects/field_diff_00004_captured.png", + "question": "Three panels are shown side by side. The leftmost panel is an INDEX key: a grid of cells numbered in row-major order (top-left = 1, increasing left-to-right then top-to-bottom). The middle and right panels are 2D color fields, each divided into the same grid. Compare the middle (Field A) and right (Field B) panels cell by cell. List the numbers of all cells that DIFFER between Field A and Field B, in ascending order, separated by commas. If no cells differ, write \"none\". Provide your final answer enclosed in ... tags.", + "answer": "1, 2, 3, 5, 7, 8, 9, 10, 11, 12, 13, 14" + } + ] +} \ No newline at end of file diff --git a/code/visual_attribute_transfer/spot_the_signal_diff/apply_effects.py b/code/visual_attribute_transfer/spot_the_signal_diff/apply_effects.py new file mode 100644 index 0000000000000000000000000000000000000000..0a7a40a9539296086bc2907ed3b75680dff82add --- /dev/null +++ b/code/visual_attribute_transfer/spot_the_signal_diff/apply_effects.py @@ -0,0 +1,931 @@ +#!/usr/bin/env python3 +"""Combined effects pipeline for spot_the_signal_diff. + +Single-file version of: + - Stage 1: blue digital data-stream "enhance" effect + - Stage 2: print-and-capture degradation + - Optional: sheet-on-surface composite, JPEG roundtrip, etc. + +Every effect can be toggled independently via EffectsConfig (or matching +CLI flags). Disabled stages are simply skipped, preserving the input image +for that stage. + +Usage: + python apply_effects.py # enhance + bad-illumination capture + python apply_effects.py --no-enhance # capture only on existing originals + python apply_effects.py --no-perspective-capture --no-jpeg-roundtrip + python apply_effects.py --sheet-surface --overwrite +""" + +from __future__ import annotations + +import argparse +import json +import math +import random +import tempfile +from dataclasses import asdict, dataclass, field, replace +from io import BytesIO +from pathlib import Path +from typing import Iterable + +import numpy as np +from PIL import Image, ImageDraw, ImageEnhance, ImageFilter, ImageFont + + +HERE = Path(__file__).resolve().parent +DEFAULT_IMAGES_DIR = HERE / "images" +DEFAULT_OUTPUT_DIR = HERE / "images_with_effects" +IMAGE_EXTENSIONS = {".png", ".jpg", ".jpeg", ".webp"} +OUTPUT_SUFFIX = "_captured.png" + + +# ===================================================================== +# Toggle configuration +# ===================================================================== + +@dataclass +class EffectsConfig: + """Per-effect on/off switches. + + Setting any field to False makes that step a no-op (its input image is + passed through unchanged). Defaults reproduce the canonical + enhance + bad-illumination + JPEG capture chain. + """ + + # Stage 1 - "enhance" pass (blue data-stream visuals). + enhance: bool = True + + # Stage 2 - print/capture degradation, broken into substeps. + reverse_dark_background: bool = False # only meaningful for dark-bg inputs + composite_on_paper: bool = True + color_and_tone: bool = True + perspective_capture: bool = True + pre_blur: bool = True + lighting_variation: bool = True + scanner_grain: bool = True + post_blur: bool = True + jpeg_roundtrip: bool = True + sheet_surface: bool = False # off by default; second JPEG roundtrip if on + + # Tunables. + illumination_profile: str = "bad" # "bad" or "standard" + paper_luma_threshold: float = 50.0 + paper_coverage_threshold: float = 0.55 + + +# ===================================================================== +# Stage 1: enhance (blue data-stream) +# ===================================================================== + +@dataclass(frozen=True) +class PanelGeometry: + left_panel: tuple[int, int, int, int] + right_panel: tuple[int, int, int, int] + + +def load_font(size: int, bold: bool = False, mono: bool = False) -> ImageFont.ImageFont: + if mono: + candidates = [ + "/System/Library/Fonts/Menlo.ttc", + "/Library/Fonts/Menlo.ttc", + "/usr/share/fonts/truetype/dejavu/DejaVuSansMono.ttf", + ] + elif bold: + candidates = [ + "/System/Library/Fonts/Supplemental/Arial Bold.ttf", + "/Library/Fonts/Arial Bold.ttf", + "/usr/share/fonts/truetype/dejavu/DejaVuSans-Bold.ttf", + ] + else: + candidates = [ + "/System/Library/Fonts/Supplemental/Arial.ttf", + "/Library/Fonts/Arial.ttf", + "/usr/share/fonts/truetype/dejavu/DejaVuSans.ttf", + ] + for path in candidates: + try: + return ImageFont.truetype(path, size=size) + except OSError: + continue + return ImageFont.load_default() + + +def group_runs(indices: Iterable[int]) -> list[tuple[int, int]]: + runs: list[tuple[int, int]] = [] + start: int | None = None + previous: int | None = None + for idx in indices: + if start is None: + start = previous = idx + continue + if previous is not None and idx == previous + 1: + previous = idx + continue + runs.append((start, previous if previous is not None else start)) + start = previous = idx + if start is not None: + runs.append((start, previous if previous is not None else start)) + return runs + + +def detect_panel_geometry(image: Image.Image) -> PanelGeometry: + rgb = np.asarray(image.convert("RGB"), dtype=np.uint8) + non_white = np.any(rgb < 248, axis=2) + height, width = non_white.shape + x_counts = non_white.sum(axis=0) + y_counts = non_white.sum(axis=1) + x_runs = group_runs(np.flatnonzero(x_counts > height * 0.36).tolist()) + y_runs = group_runs(np.flatnonzero(y_counts > width * 0.30).tolist()) + if len(x_runs) < 22 or len(y_runs) < 2: + raise ValueError("Could not detect the two 10-segment plot panels.") + return PanelGeometry( + left_panel=(x_runs[0][0], y_runs[0][0], x_runs[10][1] + 1, y_runs[-1][1] + 1), + right_panel=(x_runs[11][0], y_runs[0][0], x_runs[21][1] + 1, y_runs[-1][1] + 1), + ) + + +def alpha_from_white(crop: Image.Image) -> np.ndarray: + rgb = np.asarray(crop.convert("RGB"), dtype=np.float32) + distance = np.linalg.norm(255.0 - rgb, axis=2) + return np.clip((distance - 7.0) * 8.0, 0.0, 255.0).astype(np.uint8) + + +def curve_mask_from_crop(crop: Image.Image, alpha: np.ndarray) -> np.ndarray: + rgb = np.asarray(crop.convert("RGB"), dtype=np.int16) + red, green, blue = rgb[..., 0], rgb[..., 1], rgb[..., 2] + luma = (red * 0.2126) + (green * 0.7152) + (blue * 0.0722) + curve = ( + (alpha > 16) + & (blue - red > 22) + & (green - red > 10) + & (blue > 80) + & (luma < 165) + ) + return (curve.astype(np.uint8) * 255) + + +def resize_mask(mask: np.ndarray, size: tuple[int, int], + resample: int = Image.Resampling.LANCZOS) -> np.ndarray: + return np.asarray(Image.fromarray(mask, mode="L").resize(size, resample), dtype=np.uint8) + + +def draw_binary_rain_columns( + overlay: Image.Image, + size: tuple[int, int], + rng: random.Random, + *, + column_count: int, + font_sizes: tuple[int, ...], + row_step_range: tuple[int, int], + skip_probability: float, + faint_colors: tuple[tuple[int, int, int, int], ...], + bright_colors: tuple[tuple[int, int, int, int], ...], + margin: int, + large_glow_count: int = 0, +) -> None: + width, height = size + fonts = [load_font(size=font_size, mono=True) for font_size in font_sizes] + draw = ImageDraw.Draw(overlay) + glow_layer = Image.new("RGBA", size, (0, 0, 0, 0)) + glow_draw = ImageDraw.Draw(glow_layer) + direct_digits: list[tuple[str, int, int, ImageFont.ImageFont, tuple[int, int, int, int]]] = [] + + if column_count <= 1: + x_positions = [width // 2] + else: + column_step = (width - margin * 2) / (column_count - 1) + x_positions = [ + int(margin + idx * column_step + rng.uniform(-column_step * 0.28, column_step * 0.28)) + for idx in range(column_count) + ] + + for x0 in x_positions: + font = rng.choice(fonts) + bbox = draw.textbbox((0, 0), "0", font=font) + text_h = bbox[3] - bbox[1] + row_step = rng.randint(*row_step_range) + y0 = rng.randint(-row_step, margin) + column_brightness = rng.choice(("faint", "faint", "medium", "medium", "bright")) + while y0 < height - margin: + if rng.random() >= skip_probability: + digit = rng.choice("01") + if column_brightness == "bright" or rng.random() < 0.10: + color = rng.choice(bright_colors) + if rng.random() < 0.20: + glow_draw.text((x0, y0), digit, font=font, fill=(120, 230, 255, 60)) + elif column_brightness == "medium": + color = rng.choice(tuple( + (r, g, b, min(255, int(a * 1.25))) for r, g, b, a in faint_colors + )) + else: + color = rng.choice(faint_colors) + direct_digits.append((digit, x0, y0, font, color)) + y0 += max(text_h + 8, row_step + rng.randint(-5, 6)) + + large_fonts = [load_font(size=size, mono=True) for size in (46, 54, 62)] + for _ in range(large_glow_count): + digit = rng.choice("01") + font = rng.choice(large_fonts) + bbox = glow_draw.textbbox((0, 0), digit, font=font) + text_w = bbox[2] - bbox[0] + text_h = bbox[3] - bbox[1] + x0 = rng.randint(margin, max(margin, width - margin - text_w)) + y0 = rng.randint(margin, max(margin, height - margin - text_h)) + glow_draw.text((x0, y0), digit, font=font, fill=(150, 235, 255, 95)) + direct_digits.append((digit, x0, y0, font, rng.choice(bright_colors))) + + overlay.alpha_composite(glow_layer.filter(ImageFilter.GaussianBlur(radius=5))) + for digit, x0, y0, font, color in direct_digits: + draw.text((x0, y0), digit, font=font, fill=color) + + +def make_canvas_background(size, rng, np_rng): + width, height = size + y, x = np.mgrid[0:height, 0:width].astype(np.float32) + xn, yn = x / max(width - 1, 1), y / max(height - 1, 1) + base = np.zeros((height, width, 3), dtype=np.float32) + base[..., 0] = 4 + 10 * yn + base[..., 1] = 20 + 34 * xn + 10 * yn + base[..., 2] = 58 + 82 * (1.0 - yn) + 18 * np.sin(xn * np.pi) + sweep = np.exp(-((yn - rng.uniform(0.25, 0.70)) ** 2) / rng.uniform(0.020, 0.060)) + base[..., 1] += sweep * 12 + base[..., 2] += sweep * 34 + base += np_rng.normal(0.0, 3.0, base.shape).astype(np.float32) + image = Image.fromarray(np.clip(base, 0, 255).astype(np.uint8), mode="RGB").convert("RGBA") + overlay = Image.new("RGBA", (width, height), (0, 0, 0, 0)) + draw = ImageDraw.Draw(overlay) + for _ in range(160): + x0 = rng.randint(0, width - 1) + y0 = rng.randint(0, height - 1) + length = rng.randint(18, 120) + alpha = rng.randint(10, 34) + draw.line( + [(x0, y0), (min(width, x0 + rng.randint(-8, 12)), min(height, y0 + length))], + fill=(80, 190, 255, alpha), + width=rng.choice([1, 1, 2]), + ) + draw_binary_rain_columns( + overlay, (width, height), rng, + column_count=60, font_sizes=(16, 18, 20), + row_step_range=(38, 60), skip_probability=0.55, + faint_colors=((74, 174, 255, 22), (145, 222, 255, 18), (30, 118, 220, 24)), + bright_colors=((110, 220, 255, 38), (220, 252, 255, 30)), + margin=16, large_glow_count=4, + ) + return Image.alpha_composite(image, overlay) + + +def make_panel_background(size, rng, np_rng): + width, height = size + y, x = np.mgrid[0:height, 0:width].astype(np.float32) + xn, yn = x / max(width - 1, 1), y / max(height - 1, 1) + angle = rng.uniform(0.0, 2.0 * np.pi) + dir_x, dir_y = np.cos(angle), np.sin(angle) + grad = (xn - 0.5) * dir_x + (yn - 0.5) * dir_y + grad = (grad - grad.min()) / max(grad.max() - grad.min(), 1e-6) + strength = rng.uniform(40.0, 70.0) + bg = np.zeros((height, width, 4), dtype=np.float32) + bg[..., 0] = 6 + 10 * yn + grad * (strength * 0.18) + bg[..., 1] = 30 + 28 * xn + grad * (strength * 0.55) + bg[..., 2] = 80 + 50 * (1.0 - yn) + grad * strength + bg[..., 3] = 255 + bg[..., :3] += np_rng.normal(0.0, 2.0, (height, width, 3)).astype(np.float32) + image = Image.fromarray(np.clip(bg, 0, 255).astype(np.uint8), mode="RGBA") + overlay = Image.new("RGBA", size, (0, 0, 0, 0)) + draw_binary_rain_columns( + overlay, size, rng, + column_count=64, font_sizes=(22, 26, 30), + row_step_range=(34, 50), skip_probability=0.30, + faint_colors=((60, 180, 255, 90), (120, 220, 255, 80), + (220, 250, 255, 70), (24, 110, 230, 95)), + bright_colors=((235, 252, 255, 230), (255, 255, 255, 245), + (210, 248, 255, 220)), + margin=22, large_glow_count=18, + ) + return Image.alpha_composite(image, overlay) + + +def draw_small_black_digits(overlay, size, rng, *, count): + width, height = size + draw = ImageDraw.Draw(overlay) + fonts = [load_font(size=s, mono=True) for s in (10, 11, 12, 13)] + for _ in range(count): + digit = rng.choice("01") + font = rng.choice(fonts) + x0 = rng.randint(6, max(7, width - 12)) + y0 = rng.randint(4, max(5, height - 14)) + alpha = rng.randint(160, 230) + draw.text((x0, y0), digit, font=font, fill=(0, 0, 0, alpha)) + + +def paste_alpha_tint(base, mask, xy, color, opacity=1.0): + alpha = np.clip(mask.astype(np.float32) * opacity, 0, 255).astype(np.uint8) + layer = Image.new("RGBA", (mask.shape[1], mask.shape[0]), color + (0,)) + layer.putalpha(Image.fromarray(alpha, mode="L")) + base.alpha_composite(layer, xy) + + +def draw_index_box_strip(canvas, center_xy, box_size=(54, 54), spacing=6, + label_font_size=16, num_font_size=18): + draw = ImageDraw.Draw(canvas) + cx, cy = center_xy + bw, bh = box_size + total_w = 10 * bw + 9 * spacing + start_x = cx - total_w // 2 + label_font = load_font(label_font_size, bold=True) + num_font = load_font(num_font_size, bold=True) + label_text = "Index" + lbbox = draw.textbbox((0, 0), label_text, font=label_font) + label_w = lbbox[2] - lbbox[0] + label_h = lbbox[3] - lbbox[1] + label_y = cy - bh // 2 - label_h - 12 + draw.text((cx - label_w / 2, label_y), label_text, font=label_font, + fill=(225, 248, 255, 235)) + box_top = cy - bh // 2 + for idx in range(10): + bx1 = start_x + idx * (bw + spacing) + bx2 = bx1 + bw + by1 = box_top + by2 = by1 + bh + draw.rounded_rectangle((bx1, by1, bx2, by2), radius=6, + fill=(238, 246, 255, 245), + outline=(120, 200, 255, 220), width=2) + text = str(idx + 1) + tbbox = draw.textbbox((0, 0), text, font=num_font) + tw, th = tbbox[2] - tbbox[0], tbbox[3] - tbbox[1] + draw.text((bx1 + (bw - tw) / 2, by1 + (bh - th) / 2 - 4), + text, font=num_font, fill=(18, 36, 78, 255)) + + +def add_panel(canvas, source, source_box, target_box, rng, np_rng): + x1, y1, x2, y2 = target_box + target_size = (x2 - x1, y2 - y1) + crop = source.crop(source_box).convert("RGB") + all_alpha = alpha_from_white(crop) + curve_alpha = curve_mask_from_crop(crop, all_alpha) + all_alpha = resize_mask(all_alpha, target_size) + curve_alpha = resize_mask(curve_alpha, target_size) + grid_alpha = np.clip(all_alpha.astype(np.int16) - (curve_alpha.astype(np.int16) // 2), + 0, 255).astype(np.uint8) + panel_bg = make_panel_background(target_size, rng, np_rng) + canvas.alpha_composite(panel_bg, (x1, y1)) + black_overlay = Image.new("RGBA", target_size, (0, 0, 0, 0)) + draw_small_black_digits(black_overlay, target_size, rng, count=320) + canvas.alpha_composite(black_overlay, (x1, y1)) + glow = Image.fromarray(curve_alpha, mode="L").filter(ImageFilter.GaussianBlur(radius=8)) + glow_strong = Image.fromarray(curve_alpha, mode="L").filter(ImageFilter.GaussianBlur(radius=3)) + glow_outer_radius = rng.randint(18, 32) + glow_outer = Image.fromarray(curve_alpha, mode="L").filter( + ImageFilter.GaussianBlur(radius=glow_outer_radius)) + outer_color = rng.choice([(90, 200, 255), (140, 230, 255), (60, 170, 255), + (180, 240, 255), (200, 160, 255), (110, 255, 230)]) + outer_strength = rng.uniform(0.55, 1.05) + glow_layer = Image.new("RGBA", target_size, (0, 0, 0, 0)) + outer = Image.new("RGBA", target_size, outer_color + (0,)) + outer.putalpha(Image.eval(glow_outer, lambda value: min(220, int(value * outer_strength)))) + cyan = Image.new("RGBA", target_size, (90, 220, 255, 0)) + cyan.putalpha(Image.eval(glow, lambda value: min(220, int(value * 0.85)))) + white = Image.new("RGBA", target_size, (235, 252, 255, 0)) + white.putalpha(Image.eval(glow_strong, lambda value: min(240, int(value * 0.65)))) + glow_layer = Image.alpha_composite(glow_layer, outer) + glow_layer = Image.alpha_composite(glow_layer, cyan) + glow_layer = Image.alpha_composite(glow_layer, white) + canvas.alpha_composite(glow_layer, (x1, y1)) + paste_alpha_tint(canvas, grid_alpha, (x1, y1), (150, 222, 255), opacity=0.85) + paste_alpha_tint(canvas, curve_alpha, (x1, y1), (245, 253, 255), opacity=1.0) + paste_alpha_tint(canvas, curve_alpha, (x1, y1), (255, 255, 255), opacity=1.0) + draw = ImageDraw.Draw(canvas) + panel_w, panel_h = x2 - x1, y2 - y1 + for idx in range(11): + gx = round(x1 + panel_w * idx / 10) + line_color = (155, 226, 255, 150 if idx in (0, 10) else 95) + draw.line([(gx, y1), (gx, y2)], fill=line_color, width=1) + draw.rectangle((x1, y1, x2, y2), outline=(188, 238, 255, 200), width=2) + + +def enhance_image(source: Image.Image, seed: int) -> Image.Image: + """Build the blue data-stream composite around the source's two panels.""" + rng = random.Random(seed) + np_rng = np.random.default_rng(seed) + source = source.convert("RGBA") + src_w, _ = source.size + geometry = detect_panel_geometry(source) + + canvas_w, canvas_h = src_w, 1240 + canvas = make_canvas_background((canvas_w, canvas_h), rng, np_rng) + index_area_w = 900 + panel_x1 = index_area_w + 60 + panel_x2 = canvas_w - 80 + panel_h = 430 + top_y1 = 90 + top_y2 = top_y1 + panel_h + bot_y1 = top_y2 + 160 + bot_y2 = bot_y1 + panel_h + left_target = (panel_x1, top_y1, panel_x2, top_y2) + right_target = (panel_x1, bot_y1, panel_x2, bot_y2) + title_font = load_font(40, bold=True) + draw = ImageDraw.Draw(canvas) + for label, target in (("Signal A", left_target), ("Signal B", right_target)): + x1, y1, x2, _ = target + bbox = draw.textbbox((0, 0), label, font=title_font) + draw.text(((x1 + x2) / 2 - (bbox[2] - bbox[0]) / 2, y1 - bbox[3] - 14), + label, font=title_font, fill=(225, 248, 255, 240)) + index_cx = 60 + index_area_w // 2 + index_cy = (top_y2 + bot_y1) // 2 + draw_index_box_strip(canvas, (index_cx, index_cy), box_size=(70, 70), + spacing=8, label_font_size=28, num_font_size=36) + add_panel(canvas, source, geometry.left_panel, left_target, rng, np_rng) + add_panel(canvas, source, geometry.right_panel, right_target, rng, np_rng) + return canvas.convert("RGB") + + +# ===================================================================== +# Stage 2: print/capture degradation +# ===================================================================== + +def _luminance(rgb: np.ndarray) -> np.ndarray: + return (rgb[..., 0] * 0.2126) + (rgb[..., 1] * 0.7152) + (rgb[..., 2] * 0.0722) + + +def _estimate_border_rgb(image: Image.Image) -> np.ndarray: + rgba = np.asarray(image.convert("RGBA"), dtype=np.float32) + height, width, _ = rgba.shape + border = max(4, min(height, width) // 50) + border_pixels = np.concatenate([ + rgba[:border].reshape(-1, 4), rgba[-border:].reshape(-1, 4), + rgba[:, :border].reshape(-1, 4), rgba[:, -border:].reshape(-1, 4), + ], axis=0) + visible = border_pixels[border_pixels[:, 3] > 16] + if len(visible) == 0: + visible = border_pixels + return np.median(visible[:, :3], axis=0) + + +def _background_coverage(image: Image.Image, background_rgb: np.ndarray) -> float: + rgba = np.asarray(image.convert("RGBA"), dtype=np.float32) + visible = rgba[..., 3] > 16 + if not np.any(visible): + return 0.0 + distance = np.linalg.norm(rgba[..., :3] - background_rgb, axis=2) + return float(((distance <= 32.0) & visible).sum() / visible.sum()) + + +def reverse_dark_background_colors(image: Image.Image, luma_threshold: float, + coverage_threshold: float) -> tuple[Image.Image, bool]: + rgba = image.convert("RGBA") + background_rgb = _estimate_border_rgb(rgba) + if (float(_luminance(background_rgb)) > luma_threshold + or _background_coverage(rgba, background_rgb) < coverage_threshold): + return rgba, False + arr = np.asarray(rgba, dtype=np.float32) + rgb = arr[..., :3] + alpha = arr[..., 3:4] / 255.0 + distance = np.linalg.norm(rgb - background_rgb, axis=2, keepdims=True) + soft_foreground = np.clip((distance - 10.0) / 42.0, 0.0, 1.0) * alpha + new_background = np.array([246.0, 245.0, 236.0], dtype=np.float32) + source_luma = _luminance(rgb)[..., None] / 255.0 + foreground_scale = 0.10 + (1.0 - source_luma) * 0.18 + new_foreground = (rgb * foreground_scale) + (background_rgb * 0.18) + new_foreground = np.clip(new_foreground, 12.0, 88.0) + reversed_rgb = (new_background * (1.0 - soft_foreground)) + (new_foreground * soft_foreground) + reversed_rgba = np.concatenate([reversed_rgb, arr[..., 3:4]], axis=2) + return Image.fromarray(np.clip(reversed_rgba, 0, 255).astype(np.uint8), mode="RGBA"), True + + +def composite_on_paper(image: Image.Image, paper_rgb: tuple[int, int, int]) -> Image.Image: + rgba = image.convert("RGBA") + paper = Image.new("RGBA", rgba.size, paper_rgb + (255,)) + paper.alpha_composite(rgba) + return paper.convert("RGB") + + +def _perspective_coefficients(source_points, destination_points): + matrix, vector = [], [] + for (sx, sy), (dx, dy) in zip(source_points, destination_points): + matrix.append([dx, dy, 1, 0, 0, 0, -sx * dx, -sx * dy]) + matrix.append([0, 0, 0, dx, dy, 1, -sy * dx, -sy * dy]) + vector.extend([sx, sy]) + return np.linalg.solve(np.asarray(matrix, dtype=np.float64), np.asarray(vector)).tolist() + + +def apply_perspective_capture(image, rng, paper_rgb): + width, height = image.size + max_x = max(1.0, width * rng.uniform(0.008, 0.024)) + max_y = max(1.0, height * rng.uniform(0.012, 0.034)) + destination = [ + (rng.uniform(0, max_x), rng.uniform(0, max_y)), + (width - 1 - rng.uniform(0, max_x), rng.uniform(0, max_y)), + (width - 1 - rng.uniform(0, max_x), height - 1 - rng.uniform(0, max_y)), + (rng.uniform(0, max_x), height - 1 - rng.uniform(0, max_y)), + ] + source = [(0, 0), (width - 1, 0), (width - 1, height - 1), (0, height - 1)] + coeffs = _perspective_coefficients(source, destination) + return image.transform(image.size, Image.Transform.PERSPECTIVE, coeffs, + Image.Resampling.BICUBIC, fillcolor=paper_rgb) + + +def apply_color_and_tone(image, rng): + arr = np.asarray(image).astype(np.float32) / 255.0 + arr *= np.array([rng.uniform(0.97, 1.06), rng.uniform(0.96, 1.04), + rng.uniform(0.93, 1.05)], dtype=np.float32) + arr = np.clip(arr, 0.0, 1.0) ** rng.uniform(0.92, 1.10) + arr = np.clip(arr, 0.0, 1.0) + toned = Image.fromarray((arr * 255.0).astype(np.uint8), mode="RGB") + toned = ImageEnhance.Color(toned).enhance(rng.uniform(0.88, 1.07)) + toned = ImageEnhance.Contrast(toned).enhance(rng.uniform(0.90, 1.05)) + toned = ImageEnhance.Brightness(toned).enhance(rng.uniform(0.94, 1.05)) + return toned + + +def apply_lighting_variation(image, rng, illumination_profile): + width, height = image.size + y, x = np.mgrid[0:height, 0:width].astype(np.float32) + xn = (x / max(width - 1, 1)) - 0.5 + yn = (y / max(height - 1, 1)) - 0.5 + angle = rng.uniform(0.0, math.tau) + directional = math.cos(angle) * xn + math.sin(angle) * yn + directional *= (rng.uniform(0.28, 0.58) if illumination_profile == "bad" + else rng.uniform(0.10, 0.22)) + center_limit = 0.35 if illumination_profile == "bad" else 0.20 + cx_off, cy_off = rng.uniform(-center_limit, center_limit), rng.uniform(-center_limit, center_limit) + radius = np.sqrt((xn - cx_off) ** 2 + (yn - cy_off) ** 2) + vignette_strength = (rng.uniform(0.22, 0.44) if illumination_profile == "bad" + else rng.uniform(0.07, 0.18)) + vignette_power = 1.18 if illumination_profile == "bad" else 1.45 + vignette = -vignette_strength * np.clip(radius / 0.72, 0.0, 1.0) ** vignette_power + hw = (rng.uniform(0.08, 0.22) if illumination_profile == "bad" + else rng.uniform(0.18, 0.36)) + hh = (rng.uniform(0.08, 0.26) if illumination_profile == "bad" + else rng.uniform(0.18, 0.42)) + hotspot = np.exp(-(((xn - cx_off) ** 2) / hw + ((yn - cy_off) ** 2) / hh)) + hotspot *= (rng.uniform(0.08, 0.24) if illumination_profile == "bad" + else rng.uniform(0.03, 0.10)) + base_exposure = rng.uniform(0.68, 0.84) if illumination_profile == "bad" else 1.0 + illumination = base_exposure + directional + vignette + hotspot + if illumination_profile == "bad": + sa = rng.uniform(0.0, math.tau) + sax = math.cos(sa) * xn + math.sin(sa) * yn + cast = np.exp(-((sax - rng.uniform(-0.35, 0.35)) ** 2) / rng.uniform(0.010, 0.035)) + illumination -= cast * rng.uniform(0.10, 0.24) + sa2 = sa + rng.uniform(0.6, 1.4) + sax2 = math.cos(sa2) * xn + math.sin(sa2) * yn + cast2 = np.exp(-((sax2 - rng.uniform(-0.45, 0.45)) ** 2) / rng.uniform(0.020, 0.060)) + illumination -= cast2 * rng.uniform(0.04, 0.12) + lower, upper = (0.34, 1.42) if illumination_profile == "bad" else (0.72, 1.24) + illumination = np.clip(illumination, lower, upper) + arr = np.asarray(image).astype(np.float32) * illumination[..., None] + if illumination_profile == "bad": + cast_color = np.array(rng.choice([(1.08, 1.00, 0.86), (0.88, 0.96, 1.10), + (1.05, 0.94, 0.90)]), dtype=np.float32) + arr *= cast_color + arr += rng.uniform(-14.0, 2.0) + else: + arr += rng.uniform(-3.0, 4.0) + return Image.fromarray(np.clip(arr, 0, 255).astype(np.uint8), mode="RGB") + + +def apply_scanner_grain(image, rng, np_rng, illumination_profile): + arr = np.asarray(image).astype(np.float32) + height, width, _ = arr.shape + grain_sigma = (rng.uniform(7.0, 13.5) if illumination_profile == "bad" + else rng.uniform(3.2, 7.8)) + arr += np_rng.normal(0.0, grain_sigma, arr.shape).astype(np.float32) + row_sigma = (rng.uniform(2.4, 5.2) if illumination_profile == "bad" + else rng.uniform(1.2, 3.0)) + col_sigma = (rng.uniform(0.8, 2.2) if illumination_profile == "bad" + else rng.uniform(0.4, 1.2)) + arr += np_rng.normal(0.0, row_sigma, (height, 1, 1)).astype(np.float32) + arr += np_rng.normal(0.0, col_sigma, (1, width, 1)).astype(np.float32) + scanlines = np.sin(np.arange(height, dtype=np.float32)[:, None] / rng.uniform(3.5, 7.0)) + scan_strength = (rng.uniform(2.2, 5.0) if illumination_profile == "bad" + else rng.uniform(1.1, 2.8)) + arr += scanlines[..., None] * scan_strength + y, x = np.mgrid[0:height, 0:width].astype(np.float32) + diagonal = np.sin((x * rng.uniform(0.015, 0.028)) + (y * rng.uniform(0.006, 0.014))) + diag_strength = (rng.uniform(1.0, 3.0) if illumination_profile == "bad" + else rng.uniform(0.5, 1.8)) + arr += diagonal[..., None] * diag_strength + return Image.fromarray(np.clip(arr, 0, 255).astype(np.uint8), mode="RGB") + + +def jpeg_roundtrip(image, rng, illumination_profile): + quality_range = (70, 84) if illumination_profile == "bad" else (78, 88) + buffer = BytesIO() + image.save(buffer, format="JPEG", quality=rng.randint(*quality_range), + subsampling=rng.choice([1, 2]), optimize=False) + buffer.seek(0) + return Image.open(buffer).convert("RGB") + + +# ----- Sheet-on-surface composite (only used when sheet_surface=True) ----- + +def _apply_rgba_light_warp(image, rng): + arr = np.asarray(image.convert("RGBA"), dtype=np.float32) + height, width, _ = arr.shape + y, x = np.mgrid[0:height, 0:width].astype(np.float32) + xn, yn = x / max(width - 1, 1), y / max(height - 1, 1) + px, py = rng.uniform(0.0, math.tau), rng.uniform(0.0, math.tau) + dx = width * rng.uniform(0.0025, 0.0065) * np.sin((yn * math.tau * rng.uniform(1.1, 1.8)) + px) + dx += width * rng.uniform(0.0015, 0.0035) * np.sin(((xn + yn) * math.tau * rng.uniform(0.7, 1.3)) + py) + dy = height * rng.uniform(0.0020, 0.0050) * np.sin((xn * math.tau * rng.uniform(0.9, 1.5)) + py) + dy += height * rng.uniform(0.0010, 0.0025) * np.sin(((yn * 1.7 - xn * 0.4) * math.tau) + px) + src_x, src_y = x - dx, y - dy + valid = (src_x >= 0) & (src_x <= width - 1) & (src_y >= 0) & (src_y <= height - 1) + x0 = np.floor(np.clip(src_x, 0, width - 1)).astype(np.int32) + y0 = np.floor(np.clip(src_y, 0, height - 1)).astype(np.int32) + x1 = np.clip(x0 + 1, 0, width - 1) + y1 = np.clip(y0 + 1, 0, height - 1) + wx = (src_x - x0)[..., None] + wy = (src_y - y0)[..., None] + top = (arr[y0, x0] * (1.0 - wx)) + (arr[y0, x1] * wx) + bottom = (arr[y1, x0] * (1.0 - wx)) + (arr[y1, x1] * wx) + warped = (top * (1.0 - wy)) + (bottom * wy) + warped[..., 3] *= valid.astype(np.float32) + return Image.fromarray(np.clip(warped, 0, 255).astype(np.uint8), mode="RGBA") + + +def _apply_sheet_perspective(image, rng): + width, height = image.size + max_x = max(1.0, width * rng.uniform(0.010, 0.026)) + max_y = max(1.0, height * rng.uniform(0.010, 0.028)) + destination = [ + (rng.uniform(0, max_x), rng.uniform(0, max_y)), + (width - 1 - rng.uniform(0, max_x), rng.uniform(0, max_y)), + (width - 1 - rng.uniform(0, max_x), height - 1 - rng.uniform(0, max_y)), + (rng.uniform(0, max_x), height - 1 - rng.uniform(0, max_y)), + ] + source = [(0, 0), (width - 1, 0), (width - 1, height - 1), (0, height - 1)] + coeffs = _perspective_coefficients(source, destination) + return image.transform(image.size, Image.Transform.PERSPECTIVE, coeffs, + Image.Resampling.BICUBIC, fillcolor=(0, 0, 0, 0)) + + +def _make_surface_background(size, rng, np_rng): + width, height = size + base = np.array([rng.randint(166, 190), rng.randint(158, 181), + rng.randint(142, 166)], dtype=np.float32) + y, x = np.mgrid[0:height, 0:width].astype(np.float32) + xn = (x / max(width - 1, 1)) - 0.5 + yn = (y / max(height - 1, 1)) - 0.5 + angle = rng.uniform(0.0, math.tau) + directional = (math.cos(angle) * xn) + (math.sin(angle) * yn) + shade = 1.0 + directional * rng.uniform(0.10, 0.22) + shade -= np.clip(np.sqrt(xn**2 + yn**2) / 0.72, 0.0, 1.0) ** 1.35 * rng.uniform(0.06, 0.16) + low_noise = np_rng.normal(0.0, 1.0, (32, 32)).astype(np.float32) + noise_image = Image.fromarray(((low_noise - low_noise.min()) / + (np.ptp(low_noise) + 1e-6) * 255).astype(np.uint8)) + noise_image = noise_image.resize((width, height), Image.Resampling.BICUBIC) + low_noise = (np.asarray(noise_image).astype(np.float32) - 127.5) / 127.5 + fine_noise = np_rng.normal(0.0, 2.4, (height, width, 1)).astype(np.float32) + surface = base[None, None, :] * shade[..., None] + surface += low_noise[..., None] * rng.uniform(3.0, 7.5) + surface += fine_noise + return Image.fromarray(np.clip(surface, 0, 255).astype(np.uint8), mode="RGB") + + +def _add_paper_curl_shading(paper, rng, np_rng): + arr = np.asarray(paper.convert("RGBA"), dtype=np.float32) + height, width, _ = arr.shape + y, x = np.mgrid[0:height, 0:width].astype(np.float32) + xn, yn = x / max(width - 1, 1), y / max(height - 1, 1) + left_curl = np.exp(-xn / rng.uniform(0.035, 0.070)) + right_curl = np.exp(-(1.0 - xn) / rng.uniform(0.040, 0.080)) + top_roll = np.exp(-yn / rng.uniform(0.060, 0.110)) + bottom_roll = np.exp(-(1.0 - yn) / rng.uniform(0.060, 0.120)) + shade = 1.0 + shade += left_curl * rng.uniform(0.025, 0.060) + shade -= right_curl * rng.uniform(0.020, 0.055) + shade += top_roll * np.sin((xn * math.tau * rng.uniform(0.75, 1.35)) + rng.uniform(0, math.tau)) * 0.025 + shade -= bottom_roll * rng.uniform(0.020, 0.052) + shade += np.sin((yn * math.tau * rng.uniform(1.2, 2.1)) + rng.uniform(0, math.tau)) * rng.uniform(0.006, 0.014) + paper_fiber = np_rng.normal(0.0, 1.6, (height, width, 1)).astype(np.float32) + arr[..., :3] *= shade[..., None] + arr[..., :3] += paper_fiber + return Image.fromarray(np.clip(arr, 0, 255).astype(np.uint8), mode="RGBA") + + +def _apply_surface_lighting(image, rng): + arr = np.asarray(image.convert("RGB"), dtype=np.float32) + height, width, _ = arr.shape + y, x = np.mgrid[0:height, 0:width].astype(np.float32) + xn = (x / max(width - 1, 1)) - 0.5 + yn = (y / max(height - 1, 1)) - 0.5 + angle = rng.uniform(0.0, math.tau) + gradient = (math.cos(angle) * xn) + (math.sin(angle) * yn) + cx_off = rng.uniform(-0.20, 0.20) + cy_off = rng.uniform(-0.20, 0.20) + hotspot = np.exp(-(((xn - cx_off) ** 2) / 0.14 + ((yn - cy_off) ** 2) / 0.22)) + shade = 1.0 + gradient * rng.uniform(0.08, 0.17) + shade += hotspot * rng.uniform(0.025, 0.070) + shade -= np.clip(np.sqrt(xn**2 + yn**2) / 0.74, 0.0, 1.0) ** 1.6 * rng.uniform(0.05, 0.13) + arr *= np.clip(shade, 0.72, 1.24)[..., None] + return Image.fromarray(np.clip(arr, 0, 255).astype(np.uint8), mode="RGB") + + +def apply_sheet_surface_effect(image, rng, np_rng): + width, height = image.size + surface = _make_surface_background((width, height), rng, np_rng).convert("RGBA") + scale = rng.uniform(0.88, 0.94) + sheet_w = max(8, int(width * scale)) + sheet_h = max(8, int(height * scale)) + paper_rgb = (rng.randint(234, 246), rng.randint(231, 243), rng.randint(218, 235)) + paper = Image.new("RGBA", (sheet_w, sheet_h), paper_rgb + (255,)) + inset = max(3, int(min(sheet_w, sheet_h) * rng.uniform(0.010, 0.022))) + content_w = max(2, sheet_w - (inset * 2)) + content_h = max(2, sheet_h - (inset * 2)) + content = image.convert("RGB").resize((content_w, content_h), Image.Resampling.BICUBIC) + paper.alpha_composite(content.convert("RGBA"), (inset, inset)) + paper = _add_paper_curl_shading(paper, rng, np_rng) + sheet_layer = Image.new("RGBA", (width, height), (0, 0, 0, 0)) + left = (width - sheet_w) // 2 + rng.randint(-max(1, width // 90), max(1, width // 90)) + top = (height - sheet_h) // 2 + rng.randint(-max(1, height // 90), max(1, height // 90)) + sheet_layer.alpha_composite(paper, (left, top)) + sheet_layer = _apply_rgba_light_warp(sheet_layer, rng) + sheet_layer = _apply_sheet_perspective(sheet_layer, rng) + alpha = sheet_layer.getchannel("A") + shadow_offset = (rng.randint(max(1, width // 180), max(2, width // 70)), + rng.randint(max(1, height // 180), max(2, height // 70))) + shadow_alpha = alpha.filter(ImageFilter.GaussianBlur( + radius=max(2.0, min(width, height) * 0.018))) + shadow_alpha = ImageEnhance.Brightness(shadow_alpha).enhance(rng.uniform(0.26, 0.42)) + shifted_alpha = Image.new("L", (width, height), 0) + sx, sy = shadow_offset + shifted_alpha.paste(shadow_alpha.crop((0, 0, max(1, width - sx), max(1, height - sy))), + shadow_offset) + shadow = Image.new("RGBA", (width, height), (0, 0, 0, 0)) + shadow.putalpha(shifted_alpha) + composed = Image.alpha_composite(surface, shadow) + composed = Image.alpha_composite(composed, sheet_layer) + composed = _apply_surface_lighting(composed.convert("RGB"), rng) + composed = composed.filter(ImageFilter.GaussianBlur(radius=rng.uniform(0.06, 0.18))) + return composed + + +# ===================================================================== +# Top-level orchestrator +# ===================================================================== + +def process_image(image: Image.Image, seed: int, config: EffectsConfig) -> Image.Image: + """Run the configured chain of effects on a single image.""" + rng = random.Random(seed) + np_rng = np.random.default_rng(seed) + + # ---- Stage 1: enhance -------------------------------------------------- + if config.enhance: + image = enhance_image(image, seed) + else: + image = image.convert("RGB") + + # ---- Stage 2: print/capture pipeline ----------------------------------- + if config.reverse_dark_background: + image, _ = reverse_dark_background_colors( + image, config.paper_luma_threshold, config.paper_coverage_threshold) + image = image.convert("RGB") if image.mode != "RGB" else image + + paper_rgb = (rng.randint(241, 249), rng.randint(238, 247), rng.randint(229, 241)) + + if config.composite_on_paper: + image = composite_on_paper(image, paper_rgb) + + if config.color_and_tone: + image = apply_color_and_tone(image, rng) + + if config.perspective_capture: + image = apply_perspective_capture(image, rng, paper_rgb) + + if config.pre_blur: + radius = (rng.uniform(0.60, 1.35) if config.illumination_profile == "bad" + else rng.uniform(0.35, 0.95)) + image = image.filter(ImageFilter.GaussianBlur(radius=radius)) + + if config.lighting_variation: + image = apply_lighting_variation(image, rng, config.illumination_profile) + + if config.scanner_grain: + image = apply_scanner_grain(image, rng, np_rng, config.illumination_profile) + + if config.post_blur: + image = image.filter(ImageFilter.GaussianBlur(radius=rng.uniform(0.08, 0.28))) + + if config.jpeg_roundtrip: + image = jpeg_roundtrip(image, rng, config.illumination_profile) + + if config.sheet_surface: + image = apply_sheet_surface_effect(image, rng, np_rng) + if config.jpeg_roundtrip: + image = jpeg_roundtrip(image, rng, config.illumination_profile) + + return image + + +# ===================================================================== +# CLI driver +# ===================================================================== + +@dataclass(frozen=True) +class OutputRecord: + source: str + output: str + seed: int + width: int + height: int + config: dict + + +def _build_config_from_args(args: argparse.Namespace) -> EffectsConfig: + return EffectsConfig( + enhance=args.enhance, + reverse_dark_background=args.reverse_dark_background, + composite_on_paper=args.composite_on_paper, + color_and_tone=args.color_and_tone, + perspective_capture=args.perspective_capture, + pre_blur=args.pre_blur, + lighting_variation=args.lighting_variation, + scanner_grain=args.scanner_grain, + post_blur=args.post_blur, + jpeg_roundtrip=args.jpeg_roundtrip, + sheet_surface=args.sheet_surface, + illumination_profile=args.illumination_profile, + ) + + +def parse_args() -> argparse.Namespace: + parser = argparse.ArgumentParser(description=__doc__) + parser.add_argument("--images-dir", type=Path, default=DEFAULT_IMAGES_DIR) + parser.add_argument("--output-dir", type=Path, default=DEFAULT_OUTPUT_DIR) + parser.add_argument("--seed", type=int, default=20260430) + parser.add_argument("--overwrite", action="store_true") + parser.add_argument("--manifest", type=Path, default=None) + parser.add_argument("--illumination-profile", + choices=("standard", "bad"), default="bad") + + # Toggles. Each effect gets a paired --foo / --no-foo flag. + defaults = EffectsConfig() + for name in ( + "enhance", "reverse_dark_background", "composite_on_paper", + "color_and_tone", "perspective_capture", "pre_blur", + "lighting_variation", "scanner_grain", "post_blur", + "jpeg_roundtrip", "sheet_surface", + ): + flag = name.replace("_", "-") + default = getattr(defaults, name) + parser.add_argument(f"--{flag}", dest=name, action="store_true", + default=default) + parser.add_argument(f"--no-{flag}", dest=name, action="store_false") + return parser.parse_args() + + +def main() -> None: + args = parse_args() + config = _build_config_from_args(args) + + images_dir = args.images_dir + if not images_dir.exists(): + raise SystemExit(f"Input directory does not exist: {images_dir}") + output_dir = args.output_dir + output_dir.mkdir(parents=True, exist_ok=True) + + sources = sorted( + path for path in images_dir.iterdir() + if path.is_file() and path.suffix.lower() in IMAGE_EXTENSIONS + and not path.stem.endswith("_captured") + ) + if not sources: + raise SystemExit(f"No source images found in {images_dir}") + + records: list[OutputRecord] = [] + for index, source_path in enumerate(sources): + seed = args.seed + index + output_path = output_dir / f"{source_path.stem}{OUTPUT_SUFFIX}" + if output_path.exists() and not args.overwrite: + with Image.open(output_path) as existing: + width, height = existing.size + else: + with Image.open(source_path) as image: + result = process_image(image, seed, config) + result.save(output_path, format="PNG", compress_level=4) + width, height = result.size + records.append(OutputRecord( + source=source_path.as_posix(), + output=output_path.as_posix(), + seed=seed, width=width, height=height, + config=asdict(config), + )) + print(f"wrote {output_path.name} (seed={seed})") + + manifest_path = args.manifest or (output_dir / "manifest_captured.json") + manifest = { + "effect": "spot_the_signal_diff_pipeline", + "images_dir": images_dir.as_posix(), + "output_dir": output_dir.as_posix(), + "config": asdict(config), + "count": len(records), + "records": [asdict(r) for r in records], + } + manifest_path.write_text(json.dumps(manifest, indent=2) + "\n", encoding="utf-8") + print(f"Wrote manifest to {manifest_path}") + + +if __name__ == "__main__": + main() diff --git a/code/visual_attribute_transfer/spot_the_signal_diff/creation.py b/code/visual_attribute_transfer/spot_the_signal_diff/creation.py new file mode 100644 index 0000000000000000000000000000000000000000..3be4a2b5c6e6801a0e2208706285f9eff29f4389 --- /dev/null +++ b/code/visual_attribute_transfer/spot_the_signal_diff/creation.py @@ -0,0 +1,576 @@ +"""Generate 'Spot the Signal Diff' visual benchmark dataset. + +Two 1D signals are shown side by side. Each signal's x-axis is divided +into 9 equal segments. Independently for each segment, the right signal +may or may not have a short smooth perturbation applied at the segment's +centre (raised-cosine blend, endpoints anchored so the splice is +seamless). Task: count how many of the 9 segments are identical between +left and right. Answer is an integer 0..9. +""" +from __future__ import annotations + +import argparse +import json +import random +from pathlib import Path +from typing import Any, Dict, List, Tuple + +import matplotlib + +matplotlib.use("Agg") +import matplotlib.pyplot as plt +import matplotlib.patches as mpatches +import numpy as np +from tqdm import tqdm + + +# --------------------------------------------------------------------------- +# Constants +# --------------------------------------------------------------------------- + +N_SAMPLES = 2000 +X_RANGE = (0.0, 2000.0) + +GRID_N = 9 # segments along x +assert N_SAMPLES % GRID_N == 0 or True # integer cell boundaries handled numerically + +# Each perturbation is a contiguous span of this many samples, centred in +# its segment. Must be well below the segment length so the transition +# band never straddles a segment boundary. +SEGMENT_WIDTH = N_SAMPLES // GRID_N # ≈ 222 samples per segment +PERTURB_WIDTH = 110 # ~half of one segment +assert PERTURB_WIDTH < SEGMENT_WIDTH - 2 * 15 + +# Raised-cosine splice blend at each edge of the perturbation. +TRANSITION_BAND = 15 +RESYNTH_PAD = 40 +MIN_DEVIATION_MULT = 0.4 +INTRA_SEGMENT_NOISE_SIGMA = 0.04 + +PANEL_W = 1500 +PANEL_H = 440 + +BG_COLOR = "#ffffff" +LINE_COLOR = "#1f4e79" +GRID_COLOR = "#c0c6d0" # segment-boundary grid (visible) +GRID_WIDTH = 1.2 +BORDER_COLOR = "#888888" +BORDER_WIDTH = 1.5 +LABEL_COLOR = "#333333" +HIGHLIGHT_COLOR = "#dc1e1e" + +MARGIN_PX = 70 +GAP_PX = 110 +LABEL_HEIGHT = 44 + +def _build_question() -> str: + n = GRID_N + return ( + f"Two 1D signals are shown side by side. Each signal's x-axis is " + f"divided into {n} equal segments by thin vertical gridlines, and " + f"each segment is numbered 1 through {n} above the gridlines (the " + f"two signals share the same numbering). Compare the left (Signal A) " + f"and right (Signal B) signals segment by segment. List the numbers " + f"of all segments that DIFFER between Signal A and Signal B, in " + f"ascending order, separated by commas (for example: \"2, 5, {n}\"). " + f"If no segments differ, write \"none\". " + f"Provide your final answer enclosed in ... tags." + ) + + +QUESTION = ( + "Two 1D signals are shown side by side. Each signal's x-axis is " + "divided into 9 equal segments by thin vertical gridlines. Compare " + "the left and right signals segment by segment. Count how many of " + "the 9 segments are DIFFERENT between the two signals. Report the " + "count as an integer between 0 and 9. " + "Provide your final answer enclosed in ... tags." +) + + +# --------------------------------------------------------------------------- +# Signal generator +# --------------------------------------------------------------------------- + + +def _gaussian_kernel(sigma: float) -> np.ndarray: + radius = max(1, int(sigma * 4)) + k = np.arange(-radius, radius + 1, dtype=np.float64) + w = np.exp(-(k * k) / (2.0 * sigma * sigma)) + return w / w.sum() + + +def _smoothed_noise(rng: random.Random, sigma: float) -> np.ndarray: + raw = np.array([rng.gauss(0.0, 1.0) for _ in range(N_SAMPLES)], dtype=np.float64) + kernel = _gaussian_kernel(sigma) + return np.convolve(raw, kernel, mode="same") + + +def generate_signal(rng: random.Random) -> np.ndarray: + coarse = _smoothed_noise(rng, sigma=rng.uniform(60.0, 110.0)) + medium = _smoothed_noise(rng, sigma=rng.uniform(18.0, 35.0)) + fine = _smoothed_noise(rng, sigma=rng.uniform(4.0, 8.0)) + micro = _smoothed_noise(rng, sigma=rng.uniform(1.5, 2.5)) + + def _norm(a: np.ndarray) -> np.ndarray: + s = float(np.std(a)) or 1.0 + return a / s + + y = ( + 1.8 * _norm(coarse) + + 0.75 * _norm(medium) + + 0.40 * _norm(fine) + + 0.20 * _norm(micro) + ) + y = y + np.random.normal(0.0, INTRA_SEGMENT_NOISE_SIGMA, size=y.shape) + return y + + +# --------------------------------------------------------------------------- +# Segment geometry and perturbation selection +# --------------------------------------------------------------------------- + + +def _segment_bounds(idx: int) -> Tuple[int, int]: + """Return (start, end) sample-index bounds of segment idx (end exclusive).""" + start = round(idx * N_SAMPLES / GRID_N) + end = round((idx + 1) * N_SAMPLES / GRID_N) + return int(start), int(end) + + +def _perturb_window(segment_idx: int) -> Tuple[int, int]: + """Return (start, end) sample indices of the perturbation window, + centred within segment ``segment_idx``.""" + s, e = _segment_bounds(segment_idx) + mid = (s + e) // 2 + half = PERTURB_WIDTH // 2 + p_start = mid - half + p_end = p_start + PERTURB_WIDTH + return p_start, p_end + + +MIN_PERTURBED = 0 +MAX_PERTURBED = GRID_N + + +def _sample_perturbed_segments(rng: random.Random) -> List[int]: + """Uniform num_perturbed ∈ {MIN_PERTURBED..MAX_PERTURBED}, then that many + segments chosen at random.""" + num_perturbed = rng.randint(MIN_PERTURBED, MAX_PERTURBED) + all_idx = list(range(GRID_N)) + rng.shuffle(all_idx) + return sorted(all_idx[:num_perturbed]) + + +# --------------------------------------------------------------------------- +# Perturbation (segment-resynthesis with edge-anchored raised-cosine splice) +# --------------------------------------------------------------------------- + + +def _resynthesise_segment( + rng: random.Random, + left: np.ndarray, + start: int, + end: int, + sig_std: float, + max_tries: int = 30, +) -> np.ndarray: + n = end - start + orig = left[start:end] + a = float(left[start]) + b = float(left[end - 1]) + min_dev = MIN_DEVIATION_MULT * sig_std + + for _ in range(max_tries): + cand_full = generate_signal(rng) + if len(cand_full) <= n + 2 * RESYNTH_PAD: + continue + cs = rng.randint(RESYNTH_PAD, len(cand_full) - n - RESYNTH_PAD) + cand = cand_full[cs:cs + n].copy() + + ca, cb = float(cand[0]), float(cand[-1]) + t = np.linspace(0.0, 1.0, n) + cand = cand - ((1.0 - t) * (ca - a) + t * (cb - b)) + + margin = min(TRANSITION_BAND, max(5, n // 10)) + interior_dev = float(np.max(np.abs(cand[margin:n - margin] - orig[margin:n - margin]))) + if interior_dev < min_dev: + continue + return cand + + raise RuntimeError("Could not synthesise a sufficiently different segment.") + + +def _splice_with_blend( + signal: np.ndarray, + start: int, + end: int, + replacement: np.ndarray, +) -> None: + n = end - start + band = min(TRANSITION_BAND, n // 2) + + w = np.ones(n) + if band > 0: + edge = 0.5 * (1.0 - np.cos(np.linspace(0.0, np.pi, band + 2)[1:-1])) + w[:band] = edge + w[-band:] = edge[::-1] + + signal[start:end] = (1.0 - w) * signal[start:end] + w * replacement + + +def build_sample( + rng: random.Random, +) -> Tuple[np.ndarray, np.ndarray, np.ndarray, List[Dict[str, Any]]]: + x = np.linspace(X_RANGE[0], X_RANGE[1], N_SAMPLES) + left = generate_signal(rng) + right = left.copy() + + sig_std = float(np.std(left)) or 1.0 + perturbed = set(_sample_perturbed_segments(rng)) + + segment_records: List[Dict[str, Any]] = [] + for idx in range(GRID_N): + seg_start, seg_end = _segment_bounds(idx) + is_perturbed = idx in perturbed + record: Dict[str, Any] = { + "segment": idx, + "seg_start_index": seg_start, + "seg_end_index": seg_end, + "seg_x_start": float(x[seg_start]), + "seg_x_end": float(x[min(seg_end, N_SAMPLES) - 1]), + "perturbed": is_perturbed, + } + if is_perturbed: + p_start, p_end = _perturb_window(idx) + replacement = _resynthesise_segment(rng, left, p_start, p_end, sig_std) + _splice_with_blend(right, p_start, p_end, replacement) + record.update({ + "perturb_start_index": int(p_start), + "perturb_end_index": int(p_end), + "perturb_x_start": float(x[p_start]), + "perturb_x_end": float(x[p_end - 1]), + }) + segment_records.append(record) + + return x, left, right, segment_records + + +# --------------------------------------------------------------------------- +# Rendering +# --------------------------------------------------------------------------- + + +LEGEND_W_SCALE = 0.40 # legend width = 40% of a signal panel +LEGEND_H = 80 # legend bar height (px) — independent of signal panel height + + +def _legend_dims() -> Tuple[int, int]: + return int(round(PANEL_W * LEGEND_W_SCALE)), int(LEGEND_H) + + +def _panel_origins() -> Tuple[Tuple[float, float], + Tuple[float, float], + Tuple[float, float]]: + """(ox, oy) for legend (small bar, vertically centered), Signal A, + Signal B. Layout: [legend] [GAP] [Signal A] [GAP] [Signal B].""" + oy_signal = MARGIN_PX + LABEL_HEIGHT + leg_w, leg_h = _legend_dims() + ox_legend = MARGIN_PX + oy_legend = oy_signal + (PANEL_H - leg_h) / 2 + ox_left = ox_legend + leg_w + GAP_PX + ox_right = ox_left + PANEL_W + GAP_PX + return (ox_legend, oy_legend), (ox_left, oy_signal), (ox_right, oy_signal) + + +def _canvas_size() -> Tuple[int, int]: + leg_w, _ = _legend_dims() + w = MARGIN_PX + leg_w + GAP_PX + PANEL_W + GAP_PX + PANEL_W + MARGIN_PX + h = MARGIN_PX + LABEL_HEIGHT + PANEL_H + MARGIN_PX + return int(w), int(h) + + +def _draw_legend(ax: plt.Axes, ox: float, oy: float) -> None: + """Small horizontal index bar: N cells partitioned by vertical + gridlines, each labeled with its 1-based segment number.""" + leg_w, leg_h = _legend_dims() + cell_w = leg_w / GRID_N + legend_bg = mpatches.Rectangle( + (ox, oy), leg_w, leg_h, + facecolor="#e8e6e0", edgecolor=BORDER_COLOR, + linewidth=BORDER_WIDTH, zorder=2, + ) + ax.add_patch(legend_bg) + for i in range(1, GRID_N): + gx = ox + i * cell_w + ax.plot([gx, gx], [oy, oy + leg_h], + color="#888", linewidth=GRID_WIDTH, zorder=3) + fontsize = max(10, int(min(cell_w, leg_h) * 0.40)) + for i in range(GRID_N): + cx = ox + i * cell_w + cell_w / 2 + cy = oy + leg_h / 2 + ax.text(cx, cy, str(i + 1), + ha="center", va="center", + fontsize=fontsize, fontweight="bold", + color="#222", zorder=4) + + +def _draw_panel( + ax: plt.Axes, + x: np.ndarray, + y: np.ndarray, + ox: float, + oy: float, + y_lo: float, + y_hi: float, +) -> None: + sx = PANEL_W / (X_RANGE[1] - X_RANGE[0]) + sy = PANEL_H / (y_hi - y_lo) if y_hi > y_lo else 1.0 + + px = ox + (x - X_RANGE[0]) * sx + py = oy + PANEL_H - (y - y_lo) * sy + ax.plot(px, py, color=LINE_COLOR, linewidth=1.8, zorder=3, antialiased=True) + + +def _draw_segment_grid(ax: plt.Axes, ox: float, oy: float) -> None: + """Draw GRID_N-1 internal vertical dividers, partitioning the panel + x-axis into ``GRID_N`` equal segments.""" + for i in range(1, GRID_N): + gx = ox + PANEL_W * i / GRID_N + ax.plot( + [gx, gx], [oy, oy + PANEL_H], + color=GRID_COLOR, linewidth=GRID_WIDTH, zorder=1, + ) + + +def _render( + out_path: Path, + x: np.ndarray, + left: np.ndarray, + right: np.ndarray, + segment_records: List[Dict[str, Any]] | None = None, +) -> None: + w, h = _canvas_size() + dpi = 100 + fig, ax = plt.subplots(1, 1, figsize=(w / dpi, h / dpi), dpi=dpi) + ax.set_xlim(0, w) + ax.set_ylim(h, 0) + ax.set_aspect("auto") + ax.axis("off") + fig.patch.set_facecolor(BG_COLOR) + ax.set_facecolor(BG_COLOR) + + (ox_legend, oy_legend), (ox_left, oy), (ox_right, _) = _panel_origins() + _draw_legend(ax, ox_legend, oy_legend) + leg_w, _ = _legend_dims() + ax.text( + ox_legend + leg_w / 2, oy_legend - 12, + "Index", ha="center", va="bottom", + fontsize=13, fontweight="bold", color=LABEL_COLOR, + ) + + y_all = np.concatenate([left, right]) + y_lo = float(y_all.min()) + y_hi = float(y_all.max()) + pad = (y_hi - y_lo) * 0.08 if y_hi > y_lo else 1.0 + y_lo -= pad + y_hi += pad + + for ox in (ox_left, ox_right): + border = mpatches.Rectangle( + (ox, oy), PANEL_W, PANEL_H, + facecolor="none", edgecolor=BORDER_COLOR, + linewidth=BORDER_WIDTH, zorder=2, + ) + ax.add_patch(border) + _draw_segment_grid(ax, ox, oy) + + _draw_panel(ax, x, left, ox_left, oy, y_lo, y_hi) + _draw_panel(ax, x, right, ox_right, oy, y_lo, y_hi) + + ax.text( + ox_left + PANEL_W / 2, MARGIN_PX + LABEL_HEIGHT * 0.3, + "Signal A", ha="center", va="center", + fontsize=15, fontweight="bold", color=LABEL_COLOR, + ) + ax.text( + ox_right + PANEL_W / 2, MARGIN_PX + LABEL_HEIGHT * 0.3, + "Signal B", ha="center", va="center", + fontsize=15, fontweight="bold", color=LABEL_COLOR, + ) + + + if segment_records: + sx = PANEL_W / (X_RANGE[1] - X_RANGE[0]) + for rec in segment_records: + if not rec.get("perturbed"): + continue + px_start = (rec["perturb_x_start"] - X_RANGE[0]) * sx + px_end = (rec["perturb_x_end"] - X_RANGE[0]) * sx + for ox in (ox_left, ox_right): + ax.add_patch(mpatches.Rectangle( + (ox + px_start, oy), + px_end - px_start, PANEL_H, + facecolor=HIGHLIGHT_COLOR, edgecolor="none", alpha=0.18, zorder=4, + )) + + fig.savefig(out_path, facecolor=BG_COLOR) + plt.close(fig) + + +def render_pair( + out_path: Path, + x: np.ndarray, + left: np.ndarray, + right: np.ndarray, +) -> None: + _render(out_path, x, left, right) + + +def render_answer( + out_path: Path, + x: np.ndarray, + left: np.ndarray, + right: np.ndarray, + segment_records: List[Dict[str, Any]], +) -> None: + _render(out_path, x, left, right, segment_records) + + +# --------------------------------------------------------------------------- +# Annotation +# --------------------------------------------------------------------------- + + +def build_annotation( + image_name: str, + segment_records: List[Dict[str, Any]], +) -> Dict[str, Any]: + num_identical = sum(1 for r in segment_records if not r["perturbed"]) + num_different = GRID_N - num_identical + diff_indices = sorted( + r["segment"] + 1 for r in segment_records if r.get("perturbed") + ) + answer = ", ".join(str(i) for i in diff_indices) if diff_indices else "none" + return { + "image": image_name, + "num_segments": GRID_N, + "num_identical": num_identical, + "num_different": num_different, + "diff_indices": diff_indices, + "segments": segment_records, + "question": _build_question(), + "answer": answer, + } + + +# --------------------------------------------------------------------------- +# Dataset generation +# --------------------------------------------------------------------------- + + +def generate_dataset( + rng: random.Random, + count: int, + output_dir: Path, +) -> None: + images_dir = output_dir / "images" + answers_dir = output_dir / "answers" + images_dir.mkdir(parents=True, exist_ok=True) + answers_dir.mkdir(parents=True, exist_ok=True) + + annotations: List[Dict[str, Any]] = [] + data_items: List[Dict[str, Any]] = [] + + np.random.seed(rng.randint(0, 2**31 - 1)) + + for idx in tqdm(range(count), desc="Generating signal diff pairs"): + for _ in range(20): + try: + x, left, right, segment_records = build_sample(rng) + break + except RuntimeError: + continue + else: + raise RuntimeError(f"Failed to build sample {idx}") + + image_name = f"signal_diff_{idx:05d}.png" + img_path = images_dir / image_name + ans_path = answers_dir / image_name + + render_pair(img_path, x, left, right) + render_answer(ans_path, x, left, right, segment_records) + + rel_image = f"images/{image_name}" + diff_indices = sorted( + r["segment"] + 1 for r in segment_records if r.get("perturbed") + ) + answer = ", ".join(str(i) for i in diff_indices) if diff_indices else "none" + annotations.append(build_annotation(rel_image, segment_records)) + data_items.append({ + "image": rel_image, + "question": _build_question(), + "answer": answer, + }) + + with (output_dir / "annotations.jsonl").open("w", encoding="utf-8") as fh: + for rec in annotations: + fh.write(json.dumps(rec) + "\n") + + data_json = { + "task": "spot_the_signal_diff", + "category": "visual_attribute_transfer", + "count": len(data_items), + "items": data_items, + } + with (output_dir / "data.json").open("w", encoding="utf-8") as fh: + json.dump(data_json, fh, indent=2) + fh.write("\n") + + +# --------------------------------------------------------------------------- +# CLI +# --------------------------------------------------------------------------- + + +def parse_args() -> argparse.Namespace: + p = argparse.ArgumentParser( + description="Generate 'Spot the Signal Diff' visual benchmark dataset." + ) + p.add_argument("--output-root", type=Path, default=".") + p.add_argument("--count", type=int, default=20) + p.add_argument("--seed", type=int, default=42) + p.add_argument("--difficulty", type=int, default=5, + help="Integer difficulty >=0; scales perturbed segments and perturbation amplitude.") + return p.parse_args() + + +def main() -> None: + import math + args = parse_args() + rng = random.Random(args.seed) + d = max(0, int(args.difficulty)) + + global GRID_N, SEGMENT_WIDTH, PERTURB_WIDTH + global MIN_PERTURBED, MAX_PERTURBED, MIN_DEVIATION_MULT + global INTRA_SEGMENT_NOISE_SIGMA, PANEL_W + + GRID_N = 5 + d + SEGMENT_WIDTH = N_SAMPLES // GRID_N + PERTURB_WIDTH = min(110, max(30, SEGMENT_WIDTH - 2 * TRANSITION_BAND - 10)) + MIN_PERTURBED = 1 + MAX_PERTURBED = GRID_N + scale = 0.5 / (1.0 + 0.1 * d) + MIN_DEVIATION_MULT = MIN_DEVIATION_MULT * scale + INTRA_SEGMENT_NOISE_SIGMA = 0.03 + 0.01 * d + + # Canvas scaling (X-axis only): panel width scales with sqrt((5+d)/5). + s = math.sqrt(max(1.0, (5 + d) / 5)) + PANEL_W = int(round(PANEL_W * s)) + + generate_dataset(rng, args.count, args.output_root) + print(f"Saved {args.count} signal diff pairs to {args.output_root}") + + +if __name__ == "__main__": + main() diff --git a/code/visual_attribute_transfer/spot_the_signal_diff/data.json b/code/visual_attribute_transfer/spot_the_signal_diff/data.json new file mode 100644 index 0000000000000000000000000000000000000000..2f2c7036d156425c631d5e3951ea92d753bcebe2 --- /dev/null +++ b/code/visual_attribute_transfer/spot_the_signal_diff/data.json @@ -0,0 +1,32 @@ +{ + "task": "spot_the_signal_diff", + "category": "visual_attribute_transfer", + "count": 5, + "items": [ + { + "image": "images_with_effects/signal_diff_00000_captured.png", + "question": "Two 1D signals are shown side by side. Each signal's x-axis is divided into equal segments by thin vertical gridlines, and each segment is numbered above the gridlines (the two signals share the same numbering). Compare the left (Signal A) and right (Signal B) signals segment by segment. List the numbers of all segments that DIFFER between Signal A and Signal B, in ascending order, separated by commas. If no segments differ, write \"none\". Provide your final answer enclosed in ... tags.", + "answer": "7, 8, 9" + }, + { + "image": "images_with_effects/signal_diff_00001_captured.png", + "question": "Two 1D signals are shown side by side. Each signal's x-axis is divided into equal segments by thin vertical gridlines, and each segment is numbered above the gridlines (the two signals share the same numbering). Compare the left (Signal A) and right (Signal B) signals segment by segment. List the numbers of all segments that DIFFER between Signal A and Signal B, in ascending order, separated by commas. If no segments differ, write \"none\". Provide your final answer enclosed in ... tags.", + "answer": "2, 5, 7, 8, 10" + }, + { + "image": "images_with_effects/signal_diff_00002_captured.png", + "question": "Two 1D signals are shown side by side. Each signal's x-axis is divided into equal segments by thin vertical gridlines, and each segment is numbered above the gridlines (the two signals share the same numbering). Compare the left (Signal A) and right (Signal B) signals segment by segment. List the numbers of all segments that DIFFER between Signal A and Signal B, in ascending order, separated by commas. If no segments differ, write \"none\". Provide your final answer enclosed in ... tags.", + "answer": "8" + }, + { + "image": "images_with_effects/signal_diff_00003_captured.png", + "question": "Two 1D signals are shown side by side. Each signal's x-axis is divided into equal segments by thin vertical gridlines, and each segment is numbered above the gridlines (the two signals share the same numbering). Compare the left (Signal A) and right (Signal B) signals segment by segment. List the numbers of all segments that DIFFER between Signal A and Signal B, in ascending order, separated by commas. If no segments differ, write \"none\". Provide your final answer enclosed in ... tags.", + "answer": "3, 4, 6, 9" + }, + { + "image": "images_with_effects/signal_diff_00004_captured.png", + "question": "Two 1D signals are shown side by side. Each signal's x-axis is divided into equal segments by thin vertical gridlines, and each segment is numbered above the gridlines (the two signals share the same numbering). Compare the left (Signal A) and right (Signal B) signals segment by segment. List the numbers of all segments that DIFFER between Signal A and Signal B, in ascending order, separated by commas. If no segments differ, write \"none\". Provide your final answer enclosed in ... tags.", + "answer": "2, 3, 5, 6, 10" + } + ] +} \ No newline at end of file diff --git a/code/visual_attribute_transfer/spot_the_stroke_diff/creation.py b/code/visual_attribute_transfer/spot_the_stroke_diff/creation.py new file mode 100644 index 0000000000000000000000000000000000000000..a42a069f28d328de1e24e7844f151b3b20c66d29 --- /dev/null +++ b/code/visual_attribute_transfer/spot_the_stroke_diff/creation.py @@ -0,0 +1,682 @@ +"""Generate 'Spot the Stroke Diff' visual benchmark dataset. + +Two panels are shown side by side. Each panel contains the same set of brush +strokes (B-spline curves) at the same positions, but some strokes in the +right panel have a different shape from the corresponding stroke in the left +panel. Each stroke pair is labelled with an uppercase letter (A, B, C, ...) +placed identically in both panels. The task is to list the letters of the +strokes that differ. + +Stroke generation is reused from the sibling ``stroke_gesture_count`` task. +""" +from __future__ import annotations + +import argparse +import json +import math +import os +import random +import sys +from pathlib import Path +from typing import Any, Dict, List, Tuple + +import matplotlib + +matplotlib.use("Agg") +import matplotlib.pyplot as plt +import matplotlib.patches as mpatches +import numpy as np +from tqdm import tqdm + +# Reuse stroke generation helpers from stroke_gesture_count. +_SIBLING = Path(__file__).resolve().parent.parent / "stroke_gesture_count" +sys.path.insert(0, str(_SIBLING)) +from creation import ( # type: ignore # noqa: E402 + STROKE_WIDTH, + _curve_bbox, + _hausdorff_distance, + _render_curve_on_ax, + generate_distractor_stroke, + generate_template_stroke, +) + + +# --------------------------------------------------------------------------- +# Constants +# --------------------------------------------------------------------------- + +NUM_ITEMS = 14 +MIN_DIFFS = 4 +MAX_DIFFS = 7 + +PANEL_W = 900 +PANEL_H = 900 +PADDING = 20 +EDGE_PADDING = 20 + +HAUSDORFF_THRESH = 8.0 +HAUSDORFF_MAX = 18.0 + +BG_COLOR = "#0a1020" +BORDER_COLOR = "#7aa6ff" +BORDER_WIDTH = 2.0 +HIGHLIGHT_COLOR = "#dc1e1e" +LABEL_COLOR = "#cfe0ff" +STROKE_LABEL_COLOR = "#ffd24a" +STROKE_LABEL_OFFSET = 14.0 +STROKE_LABEL_FONTSIZE = 14 + +MARGIN_PX = 60 +GAP_PX = 100 +LABEL_HEIGHT = 44 + +QUESTION = ( + "Two panels are shown side by side. Each panel contains the same number " + "of brush strokes at the same positions. Some strokes in the right panel " + "have a different shape from the corresponding stroke in the left panel. " + "Count the number of stroke pairs whose shape differs between the two " + "panels and report the integer count. " + "Provide your final answer enclosed in ... tags." +) + + +# --------------------------------------------------------------------------- +# Geometry helpers +# --------------------------------------------------------------------------- + + +def _curve_radius(curve: np.ndarray) -> float: + """Max distance from origin to any point on a centred curve.""" + return float(np.max(np.sqrt((curve * curve).sum(axis=1)))) + + +def _index_to_letters(idx: int) -> str: + """Convert 0-based index to uppercase spreadsheet-style label. + + 0 -> A, 25 -> Z, 26 -> AA, 27 -> AB, ... + """ + n = idx + chars: List[str] = [] + while True: + chars.append(chr(ord("A") + (n % 26))) + n = n // 26 - 1 + if n < 0: + break + return "".join(reversed(chars)) + + +def _compute_label_positions( + width: int, + height: int, + centers: List[Tuple[float, float]], + curves: List[np.ndarray], +) -> List[Tuple[float, float]]: + """For each stroke, generate many candidate label positions and pick + the one that (a) hugs its own stroke, (b) maximally clears every other + stroke, and (c) doesn't collide with previously-placed labels. + + Strategy: for ~12 anchor points sampled along each stroke, try both + perpendicular normals at offsets {12, 18, 24, 30}; plus the two + endpoint-tangent extensions. Filter by hard clearance gates, then + score by other-stroke clearance and label-label clearance. + + All inputs/outputs are in panel-local pixel coordinates (the panel + being treated as a (width x height) box). + """ + own_max = 22.0 + other_clearance = 16.0 + label_clearance = 22.0 + label_half_w = 9.0 + label_half_h = 9.0 + offsets = [12.0, 18.0, 24.0, 30.0] + n_anchors = 12 + + shifted_all = [crv + np.array(ctr) for crv, ctr in zip(curves, centers)] + + positions: List[Tuple[float, float]] = [] + placed_label_pts: List[np.ndarray] = [] + + for i, sh in enumerate(shifted_all): + L = len(sh) + if L < 4: + positions.append((float(sh[0, 0]), float(sh[0, 1]))) + placed_label_pts.append(np.array(positions[-1])) + continue + + anchor_idxs = sorted(set( + [0, L - 1] + + [int(round(t * (L - 1))) + for t in np.linspace(0.0, 1.0, n_anchors)] + )) + + candidates: List[Tuple[float, float, float]] = [] + for idx in anchor_idxs: + anchor = sh[idx] + if idx == 0: + tan = sh[min(L - 1, 5)] - sh[0] + elif idx == L - 1: + tan = sh[L - 1] - sh[max(0, L - 6)] + else: + lo = max(0, idx - 3) + hi = min(L - 1, idx + 3) + tan = sh[hi] - sh[lo] + n = float(np.linalg.norm(tan)) + if n < 1e-6: + continue + t_unit = tan / n + normal_l = np.array([-t_unit[1], t_unit[0]]) + normal_r = -normal_l + directions = [normal_l, normal_r] + if idx == 0: + directions.append(-t_unit) + if idx == L - 1: + directions.append(t_unit) + + for direction in directions: + for offset in offsets: + pos = anchor + direction * offset + px, py = float(pos[0]), float(pos[1]) + if (px - label_half_w < 4 or px + label_half_w > width - 4 or + py - label_half_h < 4 or py + label_half_h > height - 4): + continue + + own_d = float("inf") + other_d = float("inf") + for j, sh_j in enumerate(shifted_all): + diffs = sh_j - np.array([px, py]) + d = float(np.sqrt((diffs * diffs).sum(axis=1)).min()) + if j == i: + if d < own_d: + own_d = d + else: + if d < other_d: + other_d = d + + if own_d > own_max: + continue + if other_d < other_clearance: + continue + + label_d = float("inf") + for lp in placed_label_pts: + d = float(np.hypot(px - lp[0], py - lp[1])) + if d < label_d: + label_d = d + if label_d < label_clearance: + continue + + score = other_d + 0.5 * label_d - 0.4 * offset + candidates.append((score, px, py)) + + if candidates: + candidates.sort(reverse=True) + _, px, py = candidates[0] + best_pos = (px, py) + else: + best_pos = None + for relax in (0.7, 0.5, 0.3): + clr = other_clearance * relax + for idx in anchor_idxs: + anchor = sh[idx] + for direction_sign in (1, -1): + if idx == 0: + tan = sh[min(L - 1, 5)] - sh[0] + elif idx == L - 1: + tan = sh[L - 1] - sh[max(0, L - 6)] + else: + tan = sh[min(L - 1, idx + 3)] - sh[max(0, idx - 3)] + n = float(np.linalg.norm(tan)) or 1.0 + t_unit = tan / n + normal = np.array([-t_unit[1] * direction_sign, + t_unit[0] * direction_sign]) + for offset in offsets: + pos = anchor + normal * offset + px, py = float(pos[0]), float(pos[1]) + if (px - label_half_w < 4 or px + label_half_w > width - 4 or + py - label_half_h < 4 or py + label_half_h > height - 4): + continue + other_d = min( + float(np.sqrt(((sh_j - np.array([px, py])) ** 2).sum(axis=1)).min()) + for j, sh_j in enumerate(shifted_all) if j != i + ) + if other_d >= clr: + best_pos = (px, py) + break + if best_pos: + break + if best_pos: + break + if best_pos: + break + if best_pos is None: + ep = sh[-1] + best_pos = ( + float(np.clip(ep[0] + 16, 4 + label_half_w, width - 4 - label_half_w)), + float(np.clip(ep[1], 4 + label_half_h, height - 4 - label_half_h)), + ) + + positions.append(best_pos) + placed_label_pts.append(np.array(best_pos)) + + return positions + + +# --------------------------------------------------------------------------- +# Stroke sampling / swapping +# --------------------------------------------------------------------------- + + +def _generate_stroke(rng: random.Random) -> np.ndarray: + _cp, curve = generate_template_stroke(rng) + return curve + + +def _sample_different_stroke( + rng: random.Random, + old: np.ndarray, + old_radius: float, + max_tries: int = 200, +) -> np.ndarray: + """Sample a new stroke with Hausdorff distance >= HAUSDORFF_THRESH from + ``old`` and bounding radius not larger than ``old_radius``.""" + for _ in range(max_tries): + try: + _cp, cand = generate_distractor_stroke( + rng, old, min_hausdorff=HAUSDORFF_THRESH, + max_hausdorff=HAUSDORFF_MAX, + ) + except RuntimeError: + continue + if _curve_radius(cand) <= old_radius: + return cand + raise RuntimeError("Could not sample a sufficiently different stroke.") + + +# --------------------------------------------------------------------------- +# Layout +# --------------------------------------------------------------------------- + + +def _place_items( + rng: random.Random, + radii: List[float], + max_tries_per_item: int = 5000, +) -> List[Tuple[float, float]]: + order = sorted(range(len(radii)), key=lambda i: -radii[i]) + centers: List[Tuple[float, float] | None] = [None] * len(radii) + + for i in order: + r = radii[i] + placed = False + for _ in range(max_tries_per_item): + x = rng.uniform(r + EDGE_PADDING, PANEL_W - r - EDGE_PADDING) + y = rng.uniform(r + EDGE_PADDING, PANEL_H - r - EDGE_PADDING) + ok = True + for j, c in enumerate(centers): + if c is None or j == i: + continue + min_dist = radii[i] + radii[j] + PADDING + dx = x - c[0] + dy = y - c[1] + if dx * dx + dy * dy < min_dist * min_dist: + ok = False + break + if ok: + centers[i] = (x, y) + placed = True + break + if not placed: + raise RuntimeError(f"Could not place stroke {i} (r={r:.1f}) without overlap.") + return [c for c in centers] # type: ignore[return-value] + + +# --------------------------------------------------------------------------- +# Sample construction +# --------------------------------------------------------------------------- + + +def build_sample( + rng: random.Random, + num_items: int, + num_diffs: int, +) -> Tuple[List[Dict[str, Any]], List[Dict[str, Any]]]: + left_strokes = [_generate_stroke(rng) for _ in range(num_items)] + radii = [_curve_radius(s) for s in left_strokes] + + right_strokes: List[np.ndarray] = [s.copy() for s in left_strokes] + changed_idx = rng.sample(range(num_items), num_diffs) + for i in changed_idx: + right_strokes[i] = _sample_different_stroke(rng, left_strokes[i], radii[i]) + + centers = _place_items(rng, radii) + changed_set = set(changed_idx) + + # Compute label positions in panel-local coords using the LEFT stroke set + # as geometry reference. The same (px, py) is used in both panels. + label_positions = _compute_label_positions( + PANEL_W, PANEL_H, centers, left_strokes + ) + + items: List[Dict[str, Any]] = [] + for i in range(num_items): + x, y = centers[i] + lx, ly = label_positions[i] + items.append({ + "index": i, + "label": _index_to_letters(i), + "x": x, + "y": y, + "bounding_radius": radii[i], + "left_curve": left_strokes[i], + "right_curve": right_strokes[i], + "changed": i in changed_set, + "label_pos": (lx, ly), + }) + + diffs: List[Dict[str, Any]] = [ + { + "index": i, + "label": items[i]["label"], + "x": items[i]["x"], + "y": items[i]["y"], + "bounding_radius": items[i]["bounding_radius"], + } + for i in sorted(changed_idx) + ] + return items, diffs + + +# --------------------------------------------------------------------------- +# Rendering +# --------------------------------------------------------------------------- + + +def _panel_origins() -> Tuple[Tuple[float, float], Tuple[float, float]]: + oy = MARGIN_PX + LABEL_HEIGHT + ox_left = MARGIN_PX + ox_right = MARGIN_PX + PANEL_W + GAP_PX + return (ox_left, oy), (ox_right, oy) + + +def _canvas_size() -> Tuple[int, int]: + w = MARGIN_PX + PANEL_W + GAP_PX + PANEL_W + MARGIN_PX + h = MARGIN_PX + LABEL_HEIGHT + PANEL_H + MARGIN_PX + return w, h + + +def _draw_panel( + ax: plt.Axes, + items: List[Dict[str, Any]], + side: str, + ox: float, + oy: float, +) -> None: + key = "left_curve" if side == "left" else "right_curve" + for it in items: + center = np.array([ox + it["x"], oy + it["y"]]) + _render_curve_on_ax(ax, it[key], center, color="white", linewidth=STROKE_WIDTH) + # Letter labels removed: task is now a count, not a label list. + + +def _render( + out_path: Path, + items: List[Dict[str, Any]], + diffs: List[Dict[str, Any]] | None = None, +) -> None: + w, h = _canvas_size() + dpi = 100 + fig, ax = plt.subplots(1, 1, figsize=(w / dpi, h / dpi), dpi=dpi) + ax.set_xlim(0, w) + ax.set_ylim(h, 0) + ax.set_aspect("equal") + ax.axis("off") + fig.patch.set_facecolor(BG_COLOR) + ax.set_facecolor(BG_COLOR) + + (ox_left, oy), (ox_right, _) = _panel_origins() + + for ox in (ox_left, ox_right): + border = mpatches.Rectangle( + (ox, oy), PANEL_W, PANEL_H, + facecolor="none", edgecolor=BORDER_COLOR, + linewidth=BORDER_WIDTH, zorder=1, + ) + ax.add_patch(border) + + _draw_panel(ax, items, "left", ox_left, oy) + _draw_panel(ax, items, "right", ox_right, oy) + + ax.text( + ox_left + PANEL_W / 2, MARGIN_PX + LABEL_HEIGHT * 0.5, + "Left", ha="center", va="center", + fontsize=16, fontweight="bold", color=LABEL_COLOR, + ) + ax.text( + ox_right + PANEL_W / 2, MARGIN_PX + LABEL_HEIGHT * 0.5, + "Right", ha="center", va="center", + fontsize=16, fontweight="bold", color=LABEL_COLOR, + ) + + if diffs: + for diff in diffs: + hl_r = diff["bounding_radius"] + 14 + for ox in (ox_left, ox_right): + cx = ox + diff["x"] + cy = oy + diff["y"] + ring = mpatches.Circle( + (cx, cy), hl_r, + facecolor="none", edgecolor=HIGHLIGHT_COLOR, + linewidth=2.5, zorder=10, + ) + ax.add_patch(ring) + + fig.savefig(out_path, facecolor=BG_COLOR) + plt.close(fig) + + +def render_pair(out_path: Path, items: List[Dict[str, Any]]) -> None: + _render(out_path, items) + + +def render_answer( + out_path: Path, + items: List[Dict[str, Any]], + diffs: List[Dict[str, Any]], +) -> None: + _render(out_path, items, diffs) + + +# --------------------------------------------------------------------------- +# Annotation +# --------------------------------------------------------------------------- + + +def _answer_string(diffs: List[Dict[str, Any]]) -> str: + return str(len(diffs)) + + +def build_annotation( + image_name: str, + items: List[Dict[str, Any]], + diffs: List[Dict[str, Any]], +) -> Dict[str, Any]: + return { + "image": image_name, + "num_items": len(items), + "num_differences": len(diffs), + "differences": diffs, + "question": QUESTION, + "answer": _answer_string(diffs), + } + + +# --------------------------------------------------------------------------- +# Dataset generation +# --------------------------------------------------------------------------- + + +def generate_dataset( + rng: random.Random, + count: int, + output_dir: Path, + num_items: int = NUM_ITEMS, + min_diffs: int = MIN_DIFFS, + max_diffs: int = MAX_DIFFS, +) -> None: + images_dir = output_dir / "images" + answers_dir = output_dir / "answers" + images_dir.mkdir(parents=True, exist_ok=True) + answers_dir.mkdir(parents=True, exist_ok=True) + + annotations: List[Dict[str, Any]] = [] + data_items: List[Dict[str, Any]] = [] + + if count > 1: + forced = [int(round(min_diffs + i * (max_diffs - min_diffs) / (count - 1))) for i in range(count)] + else: + forced = [min_diffs] + print(f"forced stroke diff counts: {forced}") + + for idx in tqdm(range(count), desc="Generating stroke diff pairs"): + num_diffs = forced[idx] + for _ in range(30): + try: + items, diffs = build_sample(rng, num_items, num_diffs) + break + except RuntimeError: + continue + else: + raise RuntimeError(f"Failed to build sample {idx} after many retries") + + image_name = f"stroke_diff_{idx:05d}.png" + img_path = images_dir / image_name + ans_path = answers_dir / image_name + + render_pair(img_path, items) + render_answer(ans_path, items, diffs) + + rel_image = f"images/{image_name}" + annotations.append(build_annotation(rel_image, items, diffs)) + data_items.append({ + "image": rel_image, + "question": QUESTION, + "answer": _answer_string(diffs), + }) + + with (output_dir / "annotations.jsonl").open("w", encoding="utf-8") as fh: + for rec in annotations: + fh.write(json.dumps(rec) + "\n") + + data_json = { + "task": "spot_the_stroke_diff", + "category": "visual_attribute_transfer", + "count": len(data_items), + "items": data_items, + } + with (output_dir / "data.json").open("w", encoding="utf-8") as fh: + json.dump(data_json, fh, indent=2) + fh.write("\n") + + +# --------------------------------------------------------------------------- +# CLI +# --------------------------------------------------------------------------- + + +def parse_args() -> argparse.Namespace: + p = argparse.ArgumentParser( + description="Generate 'Spot the Stroke Diff' visual benchmark dataset." + ) + p.add_argument("--output-root", type=Path, default=".") + p.add_argument("--count", type=int, default=20) + p.add_argument("--seed", type=int, default=42) + p.add_argument("--difficulty", type=int, default=5, + help="Integer difficulty >=0; scales diff count and stroke subtlety.") + p.add_argument("--workers", type=int, default=8) + return p.parse_args() + + +def main() -> None: + args = parse_args() + rng = random.Random(args.seed) + d = max(0, int(args.difficulty)) + + N_d = 10 + 2 * d + N_0 = 10 + s = math.sqrt(max(1.0, N_d / N_0)) + global PANEL_W, PANEL_H + PANEL_W = int(round(PANEL_W * s)) + PANEL_H = int(round(PANEL_H * s)) + + global HAUSDORFF_THRESH, HAUSDORFF_MAX + HAUSDORFF_THRESH = max(8.0, 12.0 - 0.4 * d) + HAUSDORFF_MAX = max(HAUSDORFF_THRESH + 10.0, 40.0 - 1.0 * d) + + stroke_count = 10 + 2 * d + min_diffs = 5 + max_diffs = 5 + 2 * d + + sys.path.insert(0, str(Path(__file__).resolve().parents[3])) + from _sample_pool import parallel_sample_records # noqa: E402 + + # Force evenly-spaced num_diffs across [min_diffs, max_diffs]. + if args.count > 1: + forced_targets = [ + int(round(min_diffs + i * (max_diffs - min_diffs) / (args.count - 1))) + for i in range(args.count) + ] + else: + forced_targets = [min_diffs] + print(f"forced stroke_diff num_diffs: {forced_targets}") + + output_dir = args.output_root + images_dir = output_dir / "images" + answers_dir = output_dir / "answers" + images_dir.mkdir(parents=True, exist_ok=True) + answers_dir.mkdir(parents=True, exist_ok=True) + + raw = [] + for ti, tgt in enumerate(forced_targets): + def _attempt(rng, _tgt=tgt): + try: + items, diffs = build_sample(rng, stroke_count, _tgt) + if len(diffs) != _tgt: + return None + return (items, diffs) + except RuntimeError: + return None + sub = parallel_sample_records( + _attempt, count=1, workers=args.workers, + seed_base=args.seed + ti * 977, + ) + raw.extend(sub) + + annotations = [] + data_items = [] + for idx, (items, diffs) in enumerate(raw): + image_name = f"stroke_diff_{idx:05d}.png" + img_path = images_dir / image_name + ans_path = answers_dir / image_name + render_pair(img_path, items) + render_answer(ans_path, items, diffs) + rel_image = f"images/{image_name}" + annotations.append(build_annotation(rel_image, items, diffs)) + data_items.append({ + "image": rel_image, "question": QUESTION, + "answer": _answer_string(diffs), + }) + + with (output_dir / "annotations.jsonl").open("w", encoding="utf-8") as fh: + for rec in annotations: + fh.write(json.dumps(rec) + "\n") + (output_dir / "data.json").write_text(json.dumps({ + "task": "spot_the_stroke_diff", + "category": "visual_attribute_transfer", + "count": len(data_items), + "items": data_items, + }, indent=2)) + print(f"Saved {len(data_items)} image pairs to {output_dir} (workers={args.workers})") + + +if __name__ == "__main__": + main() diff --git a/code/visual_attribute_transfer/spot_the_stroke_diff/data.json b/code/visual_attribute_transfer/spot_the_stroke_diff/data.json new file mode 100644 index 0000000000000000000000000000000000000000..405b7f69f5934c0a91c5725136c1cb8544d5915d --- /dev/null +++ b/code/visual_attribute_transfer/spot_the_stroke_diff/data.json @@ -0,0 +1,32 @@ +{ + "task": "spot_the_stroke_diff", + "category": "visual_attribute_transfer", + "count": 5, + "items": [ + { + "image": "images/stroke_diff_00000.png", + "question": "Two panels are shown side by side. Each panel contains the same number of elongated pieces (strands, twigs, threads, etc.) at the same positions. Some pieces in the right panel have a different shape from the corresponding piece in the left panel. Count the number of pairs whose shape differs between the two panels and report the integer count. Provide your final answer enclosed in ... tags.", + "answer": "5" + }, + { + "image": "images/stroke_diff_00001.png", + "question": "Two panels are shown side by side. Each panel contains the same number of elongated pieces (strands, twigs, threads, etc.) at the same positions. Some pieces in the right panel have a different shape from the corresponding piece in the left panel. Count the number of pairs whose shape differs between the two panels and report the integer count. Provide your final answer enclosed in ... tags.", + "answer": "8" + }, + { + "image": "images/stroke_diff_00002.png", + "question": "Two panels are shown side by side. Each panel contains the same number of elongated pieces (strands, twigs, threads, etc.) at the same positions. Some pieces in the right panel have a different shape from the corresponding piece in the left panel. Count the number of pairs whose shape differs between the two panels and report the integer count. Provide your final answer enclosed in ... tags.", + "answer": "10" + }, + { + "image": "images/stroke_diff_00003.png", + "question": "Two panels are shown side by side. Each panel contains the same number of elongated pieces (strands, twigs, threads, etc.) at the same positions. Some pieces in the right panel have a different shape from the corresponding piece in the left panel. Count the number of pairs whose shape differs between the two panels and report the integer count. Provide your final answer enclosed in ... tags.", + "answer": "12" + }, + { + "image": "images/stroke_diff_00004.png", + "question": "Two panels are shown side by side. Each panel contains the same number of elongated pieces (strands, twigs, threads, etc.) at the same positions. Some pieces in the right panel have a different shape from the corresponding piece in the left panel. Count the number of pairs whose shape differs between the two panels and report the integer count. Provide your final answer enclosed in ... tags.", + "answer": "15" + } + ] +} \ No newline at end of file diff --git a/code/visual_attribute_transfer/stroke_gesture_count/creation.py b/code/visual_attribute_transfer/stroke_gesture_count/creation.py new file mode 100644 index 0000000000000000000000000000000000000000..2e888d81a25b0073678515ef2f3786c4a30dc124 --- /dev/null +++ b/code/visual_attribute_transfer/stroke_gesture_count/creation.py @@ -0,0 +1,783 @@ +"""Generate stroke_gesture_count samples. + +Each sample produces a SINGLE side-by-side image: Template on the left, +Field on the right, separated by a thin divider. + +The Template panel shows a single white brush stroke with a specific +curvature profile, and the Field (900x900 dark canvas) shows 25-35 +scattered brush strokes, some matching the template's shape exactly +(translated only) and the rest being distractors. + +The task: count how many strokes in the field match the template's shape. +""" + +from __future__ import annotations + +import argparse +import io +import json +import random +from pathlib import Path +from typing import List, Tuple, Optional + +import matplotlib +matplotlib.use("Agg") +import matplotlib.pyplot as plt +import matplotlib.patches as mpatches +from matplotlib.patches import Rectangle +import numpy as np +from PIL import Image +from scipy.interpolate import splprep, splev +from scipy.spatial.distance import directed_hausdorff +from tqdm import tqdm + + +QUESTION = ( + "This image has two panels separated by a thin vertical divider. " + "The left panel shows the Template: a single white brush stroke with a " + "specific curvature profile. The right panel shows the Field: many white " + "brush strokes scattered across the canvas, all at the same thickness. " + "Both panels use the SAME pixel scale and stroke width. Count the number " + "of strokes in the Field that have the exact same shape as the Template " + "(translation only — same curvature, same orientation, no rotation or " + "mirroring) and report the integer count. " + "Provide your final answer enclosed in ... tags." +) + +STROKE_WIDTH = 4.0 +BG_COLOR = "#0a1020" +HEADER_BG = "#11182a" +BORDER_COLOR = "#7aa6ff" +LABEL_COLOR = "#cfe0ff" +STROKE_COLOR = "white" +LABEL_TEXT_COLOR = "#ffd24a" # bright yellow/gold, distinct from white strokes +LABEL_FONT_SIZE = 13 + + +# --------------------------------------------------------------------------- +# Stroke generation via B-splines +# --------------------------------------------------------------------------- + +def _random_control_points(rng: random.Random, n_pts: int = 5, + target_length: float = 120.0) -> np.ndarray: + """Generate n_pts control points that create an interesting curve. + + Returns control points centered around the origin. + The resulting curve will have arc length approximately target_length. + """ + # Spread control points in a roughly sequential manner along x + # with random y offsets to create interesting curvature + scale = target_length / (n_pts - 1) + pts = [] + for i in range(n_pts): + x = i * scale * rng.uniform(0.6, 1.4) + y = rng.uniform(-scale * 1.2, scale * 1.2) + pts.append([x, y]) + pts = np.array(pts, dtype=np.float64) + # Center around origin + centroid = pts.mean(axis=0) + pts -= centroid + return pts + + +def _sample_bspline(control_points: np.ndarray, + n_samples: int = 300) -> np.ndarray: + """Dense-sample a B-spline through the given control points. + + Returns (n_samples, 2) array of points along the curve. + """ + k = min(3, len(control_points) - 1) + tck, u = splprep([control_points[:, 0], control_points[:, 1]], + s=0, k=k) + u_fine = np.linspace(0, 1, n_samples) + x, y = splev(u_fine, tck) + return np.column_stack([x, y]) + + +def _arc_length(curve: np.ndarray) -> float: + """Compute arc length of a polyline.""" + diffs = np.diff(curve, axis=0) + return float(np.sum(np.sqrt(np.sum(diffs**2, axis=1)))) + + +def _curve_bbox(curve: np.ndarray) -> Tuple[float, float, float, float]: + """Return (xmin, ymin, xmax, ymax) of a curve.""" + return (curve[:, 0].min(), curve[:, 1].min(), + curve[:, 0].max(), curve[:, 1].max()) + + +def _hausdorff_distance(c1: np.ndarray, c2: np.ndarray) -> float: + """Symmetric Hausdorff distance between two point sets.""" + d1 = directed_hausdorff(c1, c2)[0] + d2 = directed_hausdorff(c2, c1)[0] + return max(d1, d2) + + +def generate_template_stroke(rng: random.Random) -> Tuple[np.ndarray, np.ndarray]: + """Generate a template stroke. + + Returns (control_points, sampled_curve) where the curve is centered + around origin. + """ + for _ in range(100): + n_pts = rng.randint(4, 5) + lo = max(40.0, 120.0 - TEMPLATE_LENGTH_STD) + hi = 120.0 + TEMPLATE_LENGTH_STD + target_len = rng.uniform(lo, hi) + cp = _random_control_points(rng, n_pts, target_len) + curve = _sample_bspline(cp, 300) + + length = _arc_length(curve) + if length < max(40.0, lo - 10.0) or length > hi + 30.0: + continue + + # Check bounding box isn't too degenerate + bbox = _curve_bbox(curve) + w = bbox[2] - bbox[0] + h = bbox[3] - bbox[1] + if w < 20 or h < 15: + continue + if w > 250 or h > 250: + continue + + # Center the curve + centroid = curve.mean(axis=0) + curve -= centroid + cp -= centroid + return cp, curve + + raise RuntimeError("Could not generate template stroke") + + +DISTRACTOR_HAUSDORFF_MIN = 15.0 # module-level override target (difficulty-tunable) +TEMPLATE_LENGTH_STD = 35.0 # range for uniform target_length around 120 + + +def generate_distractor_stroke(rng: random.Random, + template_curve: np.ndarray, + min_hausdorff: float = 15.0, + max_hausdorff: float | None = None) -> Tuple[np.ndarray, np.ndarray]: + """Generate a distractor stroke that is similar in size but different shape. + + Returns (control_points, sampled_curve) centered around origin. + """ + template_length = _arc_length(template_curve) + + for _ in range(200): + n_pts = rng.randint(4, 5) + # Match approximate length + target_len = template_length * rng.uniform(0.8, 1.2) + cp = _random_control_points(rng, n_pts, target_len) + curve = _sample_bspline(cp, 300) + + length = _arc_length(curve) + if length < 60 or length > 200: + continue + + bbox = _curve_bbox(curve) + w = bbox[2] - bbox[0] + h = bbox[3] - bbox[1] + if w < 15 or h < 10: + continue + if w > 280 or h > 280: + continue + + # Center + centroid = curve.mean(axis=0) + curve -= centroid + cp -= centroid + + # Verify sufficiently different from template + hd = _hausdorff_distance(curve, template_curve) + if hd < min_hausdorff: + continue + if max_hausdorff is not None and hd > max_hausdorff: + continue + + return cp, curve + + raise RuntimeError("Could not generate distractor stroke") + + +# --------------------------------------------------------------------------- +# Placement +# --------------------------------------------------------------------------- + +def _curves_overlap(centers: List[np.ndarray], curves: List[np.ndarray], + new_center: np.ndarray, new_curve: np.ndarray, + min_dist: float) -> bool: + """Check if a new stroke (at new_center) overlaps any existing strokes. + + Uses bounding-box pre-filter then checks minimum point-to-point distance + between the actual curve samples for any pair whose bboxes are close. + """ + min_pixel_gap = 14.0 # minimum pixel gap between any two strokes + new_shifted = new_curve + new_center + new_bbox = _curve_bbox(new_shifted) + + for i, (c, crv) in enumerate(zip(centers, curves)): + existing_shifted = crv + c + existing_bbox = _curve_bbox(existing_shifted) + + # Quick bbox rejection with generous margin + margin = min_pixel_gap + 5.0 + if (new_bbox[2] + margin < existing_bbox[0] or + new_bbox[0] - margin > existing_bbox[2] or + new_bbox[3] + margin < existing_bbox[1] or + new_bbox[1] - margin > existing_bbox[3]): + continue # bboxes far apart, no overlap + + # Bboxes are close -- subsample curves and check min distance + # Use every 10th point for speed (still ~30 points per curve) + ns = new_shifted[::10] + es = existing_shifted[::10] + # Compute pairwise distances + diffs = ns[:, None, :] - es[None, :, :] + dists_sq = np.sum(diffs**2, axis=-1) + min_d = float(np.sqrt(dists_sq.min())) + if min_d < min_pixel_gap: + return True + return False + + +def build_sample(rng: random.Random, width: int, height: int, + target_matches: int, total_strokes: int + ) -> Tuple[np.ndarray, np.ndarray, List[np.ndarray], List[np.ndarray], int]: + """Build a sample. + + Returns: + template_cp: control points for template + template_curve: sampled template curve (centered at origin) + centers: list of (x, y) center positions for each placed stroke + curves: list of sampled curves (each centered at origin) for each stroke + realised_matches: verified match count + """ + for attempt in range(40): + try: + template_cp, template_curve = generate_template_stroke(rng) + except RuntimeError: + continue + + t_bbox = _curve_bbox(template_curve) + max_stroke_extent = max(abs(t_bbox[0]), abs(t_bbox[1]), + abs(t_bbox[2]), abs(t_bbox[3])) + pad = max_stroke_extent + 30.0 + min_center_dist = max_stroke_extent * 1.8 + 10.0 + + centers: List[np.ndarray] = [] + curves: List[np.ndarray] = [] + is_match: List[bool] = [] + + # 1. Place template matches + placed_matches = 0 + for _ in range(target_matches): + placed = False + for _try in range(500): + cx = rng.uniform(pad, width - pad) + cy = rng.uniform(pad, height - pad) + center = np.array([cx, cy]) + + if _curves_overlap(centers, curves, center, template_curve, + min_center_dist): + continue + + # Check the stroke stays in bounds + shifted = template_curve + center + if (shifted[:, 0].min() < 5 or shifted[:, 0].max() > width - 5 or + shifted[:, 1].min() < 5 or shifted[:, 1].max() > height - 5): + continue + + centers.append(center) + curves.append(template_curve.copy()) + is_match.append(True) + placed_matches += 1 + placed = True + break + if not placed: + break + + if placed_matches != target_matches: + continue + + # 2. Place distractors + distractors_needed = total_strokes - placed_matches + fail = False + for _ in range(distractors_needed): + placed = False + for _try in range(300): + try: + d_cp, d_curve = generate_distractor_stroke( + random.Random(rng.randint(0, 2**31 - 1)), + template_curve, + min_hausdorff=DISTRACTOR_HAUSDORFF_MIN) + except RuntimeError: + continue + + d_bbox = _curve_bbox(d_curve) + d_extent = max(abs(d_bbox[0]), abs(d_bbox[1]), + abs(d_bbox[2]), abs(d_bbox[3])) + d_pad = d_extent + 20.0 + + cx = rng.uniform(d_pad, width - d_pad) + cy = rng.uniform(d_pad, height - d_pad) + center = np.array([cx, cy]) + + if _curves_overlap(centers, curves, center, d_curve, + min_center_dist * 0.7): + continue + + shifted = d_curve + center + if (shifted[:, 0].min() < 5 or shifted[:, 0].max() > width - 5 or + shifted[:, 1].min() < 5 or shifted[:, 1].max() > height - 5): + continue + + centers.append(center) + curves.append(d_curve) + is_match.append(False) + placed = True + break + if not placed: + fail = True + break + + if fail: + continue + + # 3. Verify match count by checking Hausdorff distance + verified_matches = 0 + for i, crv in enumerate(curves): + hd = _hausdorff_distance(crv, template_curve) + if hd < 2.0: # essentially identical shape + verified_matches += 1 + + if verified_matches == target_matches: + return template_cp, template_curve, centers, curves, verified_matches + + raise RuntimeError("Failed to build sample after many attempts") + + +# --------------------------------------------------------------------------- +# Rendering +# --------------------------------------------------------------------------- + +def _render_curve_on_ax(ax, curve: np.ndarray, center: np.ndarray, + color: str = "white", linewidth: float = STROKE_WIDTH): + """Render a single stroke on a matplotlib axes.""" + shifted = curve + center + ax.plot(shifted[:, 0], shifted[:, 1], + color=color, linewidth=linewidth, + solid_capstyle="round", solid_joinstyle="round", + antialiased=True, zorder=3) + + +def _label_index_to_text(idx: int) -> str: + """Map 0->A, 1->B, ..., 25->Z, 26->AA, 27->AB, ...""" + s = "" + n = idx + while True: + s = chr(ord("A") + (n % 26)) + s + n = n // 26 - 1 + if n < 0: + break + return s + + +def _compute_label_positions(width: int, height: int, + centers: List[np.ndarray], + curves: List[np.ndarray] + ) -> List[Tuple[float, float]]: + """For each stroke, generate many candidate label positions and pick + the one that (a) hugs its own stroke, (b) maximally clears every other + stroke, and (c) doesn't collide with previously-placed labels. + + Strategy: for ~12 anchor points sampled along each stroke, try both + perpendicular normals at offsets {12, 18, 24, 30}; plus the two + endpoint-tangent extensions. Filter by hard clearance gates, then + score by other-stroke clearance and label-label clearance. + """ + own_max = 22.0 # label must be within own_max of its own stroke + other_clearance = 16.0 # must clear all OTHER strokes by ≥ this + label_clearance = 22.0 # must clear other labels by ≥ this (centre-centre) + label_half_w = 9.0 + label_half_h = 9.0 + offsets = [12.0, 18.0, 24.0, 30.0] + n_anchors = 12 + + shifted_all = [crv + ctr for crv, ctr in zip(curves, centers)] + + positions: List[Tuple[float, float]] = [] + placed_label_pts: List[np.ndarray] = [] + + for i, sh in enumerate(shifted_all): + L = len(sh) + if L < 4: + positions.append((float(sh[0, 0]), float(sh[0, 1]))) + placed_label_pts.append(np.array(positions[-1])) + continue + + # Anchor indices: ~n_anchors uniformly along the curve, plus the + # two endpoints explicitly. + anchor_idxs = sorted(set( + [0, L - 1] + + [int(round(t * (L - 1))) + for t in np.linspace(0.0, 1.0, n_anchors)] + )) + + candidates: List[Tuple[float, float, float]] = [] # (score, px, py) + for idx in anchor_idxs: + anchor = sh[idx] + # Tangent at anchor (use neighbours if available). + if idx == 0: + tan = sh[min(L - 1, 5)] - sh[0] + elif idx == L - 1: + tan = sh[L - 1] - sh[max(0, L - 6)] + else: + lo = max(0, idx - 3) + hi = min(L - 1, idx + 3) + tan = sh[hi] - sh[lo] + n = float(np.linalg.norm(tan)) + if n < 1e-6: + continue + t_unit = tan / n + normal_l = np.array([-t_unit[1], t_unit[0]]) + normal_r = -normal_l + directions = [normal_l, normal_r] + # At endpoints, also try extending outward along the tangent. + if idx == 0: + directions.append(-t_unit) + if idx == L - 1: + directions.append(t_unit) + + for direction in directions: + for offset in offsets: + pos = anchor + direction * offset + px, py = float(pos[0]), float(pos[1]) + if (px - label_half_w < 4 or px + label_half_w > width - 4 or + py - label_half_h < 4 or py + label_half_h > height - 4): + continue + + # Distance to own stroke and to every other stroke. + own_d = float("inf") + other_d = float("inf") + for j, sh_j in enumerate(shifted_all): + diffs = sh_j - np.array([px, py]) + d = float(np.sqrt((diffs * diffs).sum(axis=1)).min()) + if j == i: + if d < own_d: + own_d = d + else: + if d < other_d: + other_d = d + + # Hard gates. + if own_d > own_max: + continue + if other_d < other_clearance: + continue + + # Label-label clearance. + label_d = float("inf") + for lp in placed_label_pts: + d = float(np.hypot(px - lp[0], py - lp[1])) + if d < label_d: + label_d = d + if label_d < label_clearance: + continue + + # Maximise (other_d, label_d) and minimise offset. + score = other_d + 0.5 * label_d - 0.4 * offset + candidates.append((score, px, py)) + + if candidates: + candidates.sort(reverse=True) + _, px, py = candidates[0] + best_pos = (px, py) + else: + # Soft fallback: relax other_clearance progressively. + best_pos = None + for relax in (0.7, 0.5, 0.3): + clr = other_clearance * relax + for idx in anchor_idxs: + anchor = sh[idx] + for direction_sign in (1, -1): + if idx == 0: + tan = sh[min(L - 1, 5)] - sh[0] + elif idx == L - 1: + tan = sh[L - 1] - sh[max(0, L - 6)] + else: + tan = sh[min(L - 1, idx + 3)] - sh[max(0, idx - 3)] + n = float(np.linalg.norm(tan)) or 1.0 + t_unit = tan / n + normal = np.array([-t_unit[1] * direction_sign, + t_unit[0] * direction_sign]) + for offset in offsets: + pos = anchor + normal * offset + px, py = float(pos[0]), float(pos[1]) + if (px - label_half_w < 4 or px + label_half_w > width - 4 or + py - label_half_h < 4 or py + label_half_h > height - 4): + continue + other_d = min( + float(np.sqrt(((sh_j - np.array([px, py])) ** 2).sum(axis=1)).min()) + for j, sh_j in enumerate(shifted_all) if j != i + ) + if other_d >= clr: + best_pos = (px, py) + break + if best_pos: + break + if best_pos: + break + if best_pos: + break + if best_pos is None: + ep = sh[-1] + best_pos = (float(np.clip(ep[0] + 16, 4 + label_half_w, width - 4 - label_half_w)), + float(np.clip(ep[1], 4 + label_half_h, height - 4 - label_half_h))) + + positions.append(best_pos) + placed_label_pts.append(np.array(best_pos)) + + return positions + + +def render_field(width: int, height: int, + centers: List[np.ndarray], curves: List[np.ndarray], + labels: Optional[List[str]] = None, + label_positions: Optional[List[Tuple[float, float]]] = None) -> Image.Image: + """Render the field image. Returns PIL Image.""" + fig = plt.figure(figsize=(width / 100, height / 100), dpi=100, + facecolor=BG_COLOR) + ax = fig.add_axes([0, 0, 1, 1]) + ax.set_xlim(0, width) + ax.set_ylim(height, 0) + ax.axis("off") + ax.set_facecolor(BG_COLOR) + + for center, curve in zip(centers, curves): + _render_curve_on_ax(ax, curve, center) + + # Letter labels removed — task is now a count, no per-stroke labels needed. + + buf = io.BytesIO() + fig.savefig(buf, format="png", dpi=100, bbox_inches="tight", pad_inches=0, + facecolor=fig.get_facecolor()) + plt.close(fig) + buf.seek(0) + return Image.open(buf).convert("RGB") + + +def render_template(template_curve: np.ndarray) -> Image.Image: + """Render the template panel at the SAME pixel scale as the field. Returns PIL Image.""" + t_bbox = _curve_bbox(template_curve) + span_x = t_bbox[2] - t_bbox[0] + span_y = t_bbox[3] - t_bbox[1] + + margin = 40.0 + header_h = 34.0 + min_width = 200.0 + + content_w = span_x + 2 * margin + content_h = span_y + 2 * margin + canvas_w = max(content_w, min_width) + canvas_h = header_h + content_h + + # Offset to center the stroke in the content area + margin_x = (canvas_w - span_x) / 2.0 + ox = margin_x - t_bbox[0] + oy = (header_h + margin) - t_bbox[1] + center = np.array([ox, oy]) + + fig = plt.figure(figsize=(canvas_w / 100, canvas_h / 100), dpi=100, + facecolor="#070b14") + ax = fig.add_axes([0, 0, 1, 1]) + ax.set_xlim(0, canvas_w) + ax.set_ylim(canvas_h, 0) + ax.axis("off") + ax.set_facecolor("#070b14") + + # Header strip + ax.add_patch(Rectangle((0, 0), canvas_w, header_h, + facecolor=HEADER_BG, edgecolor="none", zorder=4)) + ax.plot([0, canvas_w], [header_h, header_h], + color="#2d3a5a", linewidth=1.0, zorder=5) + ax.text(canvas_w / 2, header_h / 2, "TEMPLATE", + color=LABEL_COLOR, fontsize=14, fontweight="bold", + ha="center", va="center", zorder=6, + family="DejaVu Sans") + + # Border + ax.add_patch(Rectangle((1.5, 1.5), canvas_w - 3, canvas_h - 3, + facecolor="none", edgecolor=BORDER_COLOR, + linewidth=2.0, zorder=6)) + + # Draw the template stroke + _render_curve_on_ax(ax, template_curve, center, linewidth=STROKE_WIDTH) + + buf = io.BytesIO() + fig.savefig(buf, format="png", dpi=100, bbox_inches=None, pad_inches=0, + facecolor=fig.get_facecolor()) + plt.close(fig) + buf.seek(0) + return Image.open(buf).convert("RGB") + + +DIVIDER_WIDTH = 3 +DIVIDER_COLOR = (51, 51, 85) # #333355 + + +def render_combined(out_path: Path, template_img: Image.Image, + field_img: Image.Image) -> None: + """Concatenate template (left) and field (right) with a thin divider, + then pad to a square canvas (BG colour) so downstream image-edit + models receive a 1:1 input.""" + bg = tuple(int(BG_COLOR.lstrip("#")[i:i+2], 16) for i in (0, 2, 4)) + tw, th = template_img.size + fw, fh = field_img.size + inner_h = max(th, fh) + inner_w = tw + DIVIDER_WIDTH + fw + side = max(inner_w, inner_h) + combined = Image.new("RGB", (side, side), bg) + x0 = (side - inner_w) // 2 + y0 = (side - inner_h) // 2 + combined.paste(template_img, (x0, y0 + (inner_h - th) // 2)) + for x in range(x0 + tw, x0 + tw + DIVIDER_WIDTH): + for y in range(y0, y0 + inner_h): + combined.putpixel((x, y), DIVIDER_COLOR) + combined.paste(field_img, (x0 + tw + DIVIDER_WIDTH, y0 + (inner_h - fh) // 2)) + combined.save(out_path) + + +# --------------------------------------------------------------------------- +# Main +# --------------------------------------------------------------------------- + +def main() -> None: + parser = argparse.ArgumentParser( + description="Generate stroke_gesture_count samples") + parser.add_argument("--output-root", type=Path, required=True) + parser.add_argument("--count", type=int, default=30) + parser.add_argument("--seed", type=int, default=0) + parser.add_argument("--width", type=int, default=900) + parser.add_argument("--height", type=int, default=900) + parser.add_argument("--difficulty", type=int, default=5, + help="Integer difficulty >=0; scales matches, total strokes, and distractor similarity.") + args = parser.parse_args() + + import math + d = max(0, int(args.difficulty)) + _min_matches = 5 + _max_matches = 5 + 2 * d + _total_strokes_lo = 25 + 3 * d + global DISTRACTOR_HAUSDORFF_MIN, TEMPLATE_LENGTH_STD + DISTRACTOR_HAUSDORFF_MIN = max(8.0, 18.0 - 1.0 * d) + TEMPLATE_LENGTH_STD = 30.0 + 5.0 * d + + # Canvas scaling: N_d = 25 + 3d, N_0 = 25. Scale both axes by sqrt(N_d/N_0) + # to preserve stroke density. + n_d = 25 + 3 * d + n_0 = 25 + s = math.sqrt(max(1.0, n_d / n_0)) + args.width = int(round(args.width * s)) + args.height = int(round(args.height * s)) + + out_root: Path = args.output_root + img_dir = out_root / "images" + img_dir.mkdir(parents=True, exist_ok=True) + + ann_path = out_root / "annotations.jsonl" + master_rng = random.Random(args.seed) + + # Force evenly-spaced answers across [_min_matches, _max_matches]. + if args.count > 1: + plan = [int(round(_min_matches + i * (_max_matches - _min_matches) / (args.count - 1))) for i in range(args.count)] + else: + plan = [_min_matches] + print(f"forced stroke_gesture match counts: {plan}") + + records = [] + with ann_path.open("w") as f: + for i in tqdm(range(args.count), desc="stroke_gesture_count"): + target = plan[i] + total_strokes = master_rng.randint(_total_strokes_lo, + _total_strokes_lo + 10) + total_strokes = max(total_strokes, target + 20) + + built = False + for retry in range(20): + sub_seed = master_rng.randint(0, 2**31 - 1) + sub_rng = random.Random(sub_seed) + try: + template_cp, template_curve, centers, curves, realised = \ + build_sample(sub_rng, args.width, args.height, + target_matches=target, + total_strokes=total_strokes) + except RuntimeError: + continue + built = True + break + if not built: + raise RuntimeError(f"sample {i} could not be generated") + + # Shuffle the stroke order so match labels are not always first. + order = list(range(len(centers))) + master_rng.shuffle(order) + centers = [centers[k] for k in order] + curves = [curves[k] for k in order] + + # Recompute match flags after shuffling. + is_match = [] + for crv in curves: + hd = _hausdorff_distance(crv, template_curve) + is_match.append(hd < 2.0) + + labels = [_label_index_to_text(k) for k in range(len(centers))] + label_positions = _compute_label_positions( + args.width, args.height, centers, curves) + + matching_labels = sorted( + [labels[k] for k, m in enumerate(is_match) if m], + key=lambda s: (len(s), s)) + answer_str = str(realised) + + img_name = f"stroke_gesture_count_{i:05d}.png" + tpl_img = render_template(template_curve) + fld_img = render_field(args.width, args.height, centers, curves, + labels=None, label_positions=None) + render_combined(img_dir / img_name, tpl_img, fld_img) + + rec = { + "image": f"images/{img_name}", + "question": QUESTION, + "answer": answer_str, + "num_matches": realised, + "matching_labels": matching_labels, + "total_strokes": len(centers), + "metadata": { + "seed": sub_seed, + "template_control_points": template_cp.tolist(), + "template_arc_length": float(_arc_length(template_curve)), + }, + } + f.write(json.dumps(rec) + "\n") + f.flush() + records.append(rec) + + data_json = { + "task": "stroke_gesture_count", + "category": "visual_attribute_transfer", + "count": len(records), + "items": records, + } + (out_root / "data.json").write_text(json.dumps(data_json, indent=2)) + + # Print summary + from collections import Counter + answers = [r["answer"] for r in records] + dist = Counter(answers) + print(f"\nSaved {len(records)} samples to {out_root}") + print(f"Images: {len(records)} (single combined per sample)") + print(f"Answer distribution: {dict(sorted(dist.items()))}") + + +if __name__ == "__main__": + main() diff --git a/code/visual_attribute_transfer/stroke_gesture_count/data.json b/code/visual_attribute_transfer/stroke_gesture_count/data.json new file mode 100644 index 0000000000000000000000000000000000000000..ca6509389296ecb203198f0f41bbf58de2931c00 --- /dev/null +++ b/code/visual_attribute_transfer/stroke_gesture_count/data.json @@ -0,0 +1,224 @@ +{ + "task": "stroke_gesture_count", + "category": "visual_attribute_transfer", + "count": 5, + "items": [ + { + "image": "images/stroke_gesture_count_00000.png", + "question": "This image has two panels separated by a thin vertical divider. The left panel shows the Template: a single elongated piece with a specific curvature profile. The right panel shows the Field: many elongated pieces of similar thickness scattered across the canvas. Both panels use the SAME pixel scale. Count the number of pieces in the Field that have the exact same shape as the Template (translation only \u2014 same curvature, same orientation, no rotation or mirroring) and report the integer count. Provide your final answer enclosed in ... tags.", + "answer": "5", + "num_matches": 5, + "matching_labels": [ + "A", + "D", + "V", + "X", + "AT" + ], + "total_strokes": 46, + "metadata": { + "seed": 1806341205, + "template_control_points": [ + [ + -66.55486597185785, + 16.085256376174193 + ], + [ + -24.278162140806785, + -27.323924822020587 + ], + [ + 48.17117983381231, + 31.903219680759342 + ], + [ + 28.52056660025756, + 23.69733723731391 + ] + ], + "template_arc_length": 198.07439477803888 + } + }, + { + "image": "images/stroke_gesture_count_00001.png", + "question": "This image has two panels separated by a thin vertical divider. The left panel shows the Template: a single elongated piece with a specific curvature profile. The right panel shows the Field: many elongated pieces of similar thickness scattered across the canvas. Both panels use the SAME pixel scale. Count the number of pieces in the Field that have the exact same shape as the Template (translation only \u2014 same curvature, same orientation, no rotation or mirroring) and report the integer count. Provide your final answer enclosed in ... tags.", + "answer": "8", + "num_matches": 8, + "matching_labels": [ + "A", + "B", + "C", + "H", + "U", + "Z", + "AI", + "AN" + ], + "total_strokes": 41, + "metadata": { + "seed": 1712934065, + "template_control_points": [ + [ + -77.86385218338921, + -12.604188800149451 + ], + [ + -49.535287993337995, + -5.891695262338526 + ], + [ + -19.63836101282533, + -9.422801727457234 + ], + [ + 62.12764718751154, + 62.16519116595446 + ] + ], + "template_arc_length": 188.75627164371588 + } + }, + { + "image": "images/stroke_gesture_count_00002.png", + "question": "This image has two panels separated by a thin vertical divider. The left panel shows the Template: a single elongated piece with a specific curvature profile. The right panel shows the Field: many elongated pieces of similar thickness scattered across the canvas. Both panels use the SAME pixel scale. Count the number of pieces in the Field that have the exact same shape as the Template (translation only \u2014 same curvature, same orientation, no rotation or mirroring) and report the integer count. Provide your final answer enclosed in ... tags.", + "answer": "10", + "num_matches": 10, + "matching_labels": [ + "F", + "H", + "O", + "X", + "Y", + "AI", + "AJ", + "AK", + "AL", + "AP" + ], + "total_strokes": 44, + "metadata": { + "seed": 789710164, + "template_control_points": [ + [ + -30.378174474215683, + -16.505150636502307 + ], + [ + -17.063197201255036, + 8.159457749210215 + ], + [ + 13.54887764758545, + 7.1511961398748625 + ], + [ + 18.40984663054958, + -12.664839905287089 + ], + [ + 25.424568834037096, + 7.1443621701249995 + ] + ], + "template_arc_length": 108.86223215044478 + } + }, + { + "image": "images/stroke_gesture_count_00003.png", + "question": "This image has two panels separated by a thin vertical divider. The left panel shows the Template: a single elongated piece with a specific curvature profile. The right panel shows the Field: many elongated pieces of similar thickness scattered across the canvas. Both panels use the SAME pixel scale. Count the number of pieces in the Field that have the exact same shape as the Template (translation only \u2014 same curvature, same orientation, no rotation or mirroring) and report the integer count. Provide your final answer enclosed in ... tags.", + "answer": "12", + "num_matches": 12, + "matching_labels": [ + "C", + "G", + "V", + "X", + "AC", + "AH", + "AI", + "AJ", + "AM", + "AO", + "AP", + "AQ" + ], + "total_strokes": 45, + "metadata": { + "seed": 732213632, + "template_control_points": [ + [ + -57.61848404506308, + 17.867423319566896 + ], + [ + -33.24807698174301, + 4.74203273755055 + ], + [ + 9.926269958552135, + 7.3665532829045315 + ], + [ + 26.987059367428763, + -14.583106342140283 + ], + [ + 8.879367310127217, + 32.192621233437194 + ] + ], + "template_arc_length": 168.08698660845386 + } + }, + { + "image": "images/stroke_gesture_count_00004.png", + "question": "This image has two panels separated by a thin vertical divider. The left panel shows the Template: a single elongated piece with a specific curvature profile. The right panel shows the Field: many elongated pieces of similar thickness scattered across the canvas. Both panels use the SAME pixel scale. Count the number of pieces in the Field that have the exact same shape as the Template (translation only \u2014 same curvature, same orientation, no rotation or mirroring) and report the integer count. Provide your final answer enclosed in ... tags.", + "answer": "15", + "num_matches": 15, + "matching_labels": [ + "A", + "E", + "O", + "P", + "Q", + "T", + "U", + "V", + "Y", + "AE", + "AF", + "AH", + "AN", + "AO", + "AP" + ], + "total_strokes": 46, + "metadata": { + "seed": 856226606, + "template_control_points": [ + [ + -55.33878173562434, + -35.28033609360301 + ], + [ + -17.276457950348778, + 6.408432265668017 + ], + [ + 24.527879198738688, + 26.69385912896616 + ], + [ + 35.50841016932807, + -1.7883523737386202 + ], + [ + 53.6277057470718, + -21.297061647888007 + ] + ], + "template_arc_length": 165.02906785597352 + } + } + ] +} \ No newline at end of file diff --git a/croissant.json b/croissant.json new file mode 100644 index 0000000000000000000000000000000000000000..9e802c70e7f11fc61e76be6c697e708d40e764a6 --- /dev/null +++ b/croissant.json @@ -0,0 +1,265 @@ +{ + "@context": { + "@language": "en", + "@vocab": "https://schema.org/", + "citeAs": "cr:citeAs", + "column": "cr:column", + "conformsTo": "dct:conformsTo", + "cr": "http://mlcommons.org/croissant/", + "rai": "http://mlcommons.org/croissant/RAI/", + "data": { + "@id": "cr:data", + "@type": "@json" + }, + "dataType": { + "@id": "cr:dataType", + "@type": "@vocab" + }, + "dct": "http://purl.org/dc/terms/", + "equivalentProperty": "cr:equivalentProperty", + "examples": { + "@id": "cr:examples", + "@type": "@json" + }, + "extract": "cr:extract", + "field": "cr:field", + "fileProperty": "cr:fileProperty", + "fileObject": "cr:fileObject", + "fileSet": "cr:fileSet", + "format": "cr:format", + "includes": "cr:includes", + "isLiveDataset": "cr:isLiveDataset", + "jsonPath": "cr:jsonPath", + "key": "cr:key", + "md5": "cr:md5", + "parentField": "cr:parentField", + "path": "cr:path", + "prov": "http://www.w3.org/ns/prov#", + "recordSet": "cr:recordSet", + "references": "cr:references", + "regex": "cr:regex", + "repeated": "cr:repeated", + "replace": "cr:replace", + "samplingRate": "cr:samplingRate", + "sc": "https://schema.org/", + "separator": "cr:separator", + "source": "cr:source", + "subField": "cr:subField", + "transform": "cr:transform" + }, + "@type": "sc:Dataset", + "name": "ActiveVision", + "description": "ActiveVision is a benchmark for evaluating active visual reasoning in multimodal large language models. It contains 85 instances across 17 tasks organised into three families — Distributed Scanning (5 tasks), Sequential Traversal (5 tasks), and Visual Attribute Transfer (7 tasks) — each requiring iterative inspection of an image to accumulate evidence and answer. All instances are programmatically generated; the released images are photorealistic re-renders of matplotlib structural drafts produced via OpenAI gpt-image-2.", + "conformsTo": "http://mlcommons.org/croissant/1.0", + "dct:conformsTo": "http://mlcommons.org/croissant/RAI/1.0", + "license": "https://creativecommons.org/licenses/by/4.0/", + "url": "https://huggingface.co/datasets/ActiveVision/active-vision", + "version": "0.4.0", + "datePublished": "2026-05-06", + "creator": { + "@type": "sc:Organization", + "name": "Anonymous (NeurIPS 2026 double-blind submission)" + }, + "publisher": { + "@type": "sc:Organization", + "name": "Anonymous (NeurIPS 2026 double-blind submission)" + }, + "keywords": [ + "multimodal", + "vision-language", + "benchmark", + "active perception", + "active vision", + "visual reasoning", + "synthetic", + "MLLM evaluation" + ], + "citeAs": "Anonymous. ActiveVision: An Exam for Active Observers. NeurIPS 2026 Datasets and Benchmarks Track (under review).", + + "rai:dataCollection": "All 85 instances are programmatically generated. For each task, a Python generator (creation.py, seedable, deterministic) lays out a structural specification — region partition, arrow geometry, maze graph, brush-stroke field, etc. — and computes the ground-truth answer in closed form from the same seed. The structural specification is rendered in two stages: (1) a plain matplotlib draft used as the geometric reference, and (2) a photorealistic re-render produced via a single OpenAI gpt-image-2 image-edit call using a per-task prompt that preserves silhouettes, positions, counts, and labels but changes only the surface material (e.g. stones on sand, hedge maze from above, starfield). The released benchmark contains the photorealistic re-renders only.", + "rai:dataCollectionType": [ + "Software Collection", + "Experiments" + ], + "rai:dataCollectionTimeframe": [ + "2026-04-30", + "2026-05-06" + ], + "rai:dataCollectionRawData": "Per-task generators are included under code///creation.py; the matplotlib structural drafts can be regenerated from seed at any difficulty. The per-task gpt-image-2 image-edit prompts are documented in code/gpt_image_prompts.json.", + "rai:dataPreprocessingProtocol": [ + "Stage 1 (geometry): matplotlib + numpy + scipy generate the structural draft and ground-truth answer deterministically from a fixed seed (creation.py --seed --difficulty 4).", + "Stage 2 (re-render): the matplotlib draft is sent to OpenAI gpt-image-2 via the image-edit endpoint, with a per-task prompt that keeps silhouettes, positions, counts, and labels intact while replacing surface material with a photorealistic style. No human-in-the-loop curation occurs between Stage 1 and Stage 2." + ], + "rai:dataAnnotationProtocol": "Ground-truth answers are computed programmatically inside creation.py from the same random seed used to render the image. Per-instance metadata (region adjacency, arrow chains, traversal paths, blob Hausdorff distances, etc.) is recorded in data/annotations/.jsonl alongside the rendered image. No human annotators were used.", + "rai:annotationsPerItem": "1 (deterministic programmatic ground truth, derived from the generation seed)", + "rai:machineAnnotationTools": [ + "Python 3.11", + "matplotlib", + "numpy", + "scipy", + "OpenCV (cv2)", + "Pillow (PIL)", + "OpenAI gpt-image-2 (image-edit endpoint)" + ], + "rai:dataReleaseMaintenancePlan": "v0.4 is a frozen released snapshot at difficulty 4 with 5 instances per task (85 total). The maintained artifact is the generation pipeline (creation.py + gpt_image_prompts.json), which supports regeneration of additional instances at arbitrary difficulty and held-out splits with unpublished seeds. Future versions will increase per-task instance counts. Versioning follows semantic versioning by Git tag (v0.4.0, v0.5.0, ...).", + "rai:dataUseCases": [ + "Testing", + "Validation" + ], + "rai:dataLimitations": [ + "Adversarial design: tasks are deliberately constructed to defeat one-shot symbolic extraction, off-the-shelf computer-vision pipelines (OCR/edge/blob), prior leakage, and gestalt heuristics. Performance on this benchmark is therefore not predictive of general-purpose vision-language ability — it specifically diagnoses active visual observation.", + "Small released set: 5 instances per task (85 total), intended for evaluation only and not for training. Larger evaluation splits and held-out seeds can be regenerated from the released pipeline.", + "Re-render dependency: photorealistic images depend on OpenAI gpt-image-2 (a closed-source service); exact reproduction of these specific PNGs requires API access. The underlying matplotlib drafts are fully reproducible from seed.", + "Stylistic coverage: the gpt-image-2 re-render templates currently span a Western-centric set of natural surfaces (beach, forest floor, kitchen counter, hedge maze, starfield, etc.); broader cultural-aesthetic coverage is left to future versions." + ], + "rai:dataBiases": [ + "Both image families (matplotlib drafts and gpt-image-2 re-renders) inherit aesthetic and cultural priors from their generators.", + "Question text is in English only.", + "No human subjects, no demographic data, no scenes depicting real persons, locations, or events." + ], + "rai:dataSocialImpact": "Intended for diagnostic evaluation of multimodal LLM perception. Negative impacts are limited because the dataset contains no humans, no real-world events, and no personal data. Positive impact: surfaces gaps in active visual reasoning that are hidden by current static-perception benchmarks, motivating architectural research toward iterative perception-driven reasoning.", + "rai:personalSensitiveInformation": "None. The dataset contains only synthetic images of geometric or natural-material patterns (regions, arrows, mazes, contours, brush strokes, starfields). No human subjects, no faces, no PII, no health, location, age, gender, ethnicity, or socio-economic data.", + "rai:hasSyntheticData": true, + "prov:wasGeneratedBy": "code///creation.py — deterministic seeded generators (Python 3.11). See code/scope.md for the full pipeline specification and the design rationale (the generation pipeline is the artifact, per the paper).", + + "distribution": [ + { + "@type": "cr:FileObject", + "@id": "manifest", + "name": "manifest.json", + "description": "Canonical index of all 85 benchmark instances with id, task, category, image path, question, ground-truth answer, and image SHA-256.", + "contentUrl": "data/manifest.json", + "encodingFormat": "application/json", + "sha256": "2caa9eb71ca1687ab2831058a0f64f81619df9ac726a79d519e4ff015ffbba4c" + }, + { + "@type": "cr:FileSet", + "@id": "images", + "name": "images", + "description": "85 photorealistic benchmark images (one PNG per instance), produced by re-rendering matplotlib structural drafts through OpenAI gpt-image-2.", + "encodingFormat": "image/png", + "includes": "data/images/*.png" + }, + { + "@type": "cr:FileSet", + "@id": "annotations", + "name": "annotations", + "description": "Per-instance verification metadata in JSON Lines format, one file per task. Records include region adjacency, arrow chains, traversal paths, Hausdorff distances, etc., derived from the same generator seed as the image.", + "encodingFormat": "application/jsonlines", + "includes": "data/annotations/*.jsonl" + }, + { + "@type": "cr:FileSet", + "@id": "code", + "name": "code", + "description": "Generation pipeline: per-task creation.py generators, per-task creation.md specs, gpt-image-2 re-render prompts (gpt_image_prompts.json), and the overall design specification (scope.md). The pipeline is the artifact: any number of fresh instances at any difficulty can be regenerated from this code.", + "encodingFormat": "text/x-python", + "includes": "code/**/*" + } + ], + + "recordSet": [ + { + "@type": "cr:RecordSet", + "@id": "instances", + "name": "instances", + "description": "One record per benchmark instance. 85 records total, balanced across 17 tasks (5 each) in 3 categories (distributed_scanning: 25, sequential_traversal: 25, visual_attribute_transfer: 35). Records are joined by id: each PNG in the images FileSet is matched to a row of manifest.json by stripping the 'images/' prefix and '.png' extension from the file path and matching the resulting basename to manifest.json's id field.", + "key": {"@id": "instances/id"}, + "field": [ + { + "@type": "cr:Field", + "@id": "instances/id", + "name": "id", + "description": "Unique instance identifier formed from the task initials and a zero-based index (e.g. CR-0 for counting_regions instance 0). This field is the join key between the images FileSet (filename without extension) and manifest.json's id column.", + "dataType": "sc:Text", + "source": { + "fileSet": {"@id": "images"}, + "extract": {"fileProperty": "filename"}, + "transform": {"regex": "([^/]+)\\.png"} + }, + "references": { + "fileObject": {"@id": "manifest"}, + "extract": {"jsonPath": "$[*].id"} + } + }, + { + "@type": "cr:Field", + "@id": "instances/task", + "name": "task", + "description": "Task name. One of: attribute_group_counting, bounded_faces_counting, counting_connected_components, counting_regions, tangled_loops, arrow_chain, color_zone_sequence, line_intersections, maze, traverse_ordering, constellation_match_count, contour_silhouette_count, spot_the_contour_diff, spot_the_field_diff, spot_the_signal_diff, spot_the_stroke_diff, stroke_gesture_count.", + "dataType": "sc:Text", + "source": { + "fileObject": {"@id": "manifest"}, + "extract": {"jsonPath": "$[*].task"} + } + }, + { + "@type": "cr:Field", + "@id": "instances/category", + "name": "category", + "description": "Task family. One of: distributed_scanning, sequential_traversal, visual_attribute_transfer.", + "dataType": "sc:Text", + "source": { + "fileObject": {"@id": "manifest"}, + "extract": {"jsonPath": "$[*].category"} + } + }, + { + "@type": "cr:Field", + "@id": "instances/image_path", + "name": "image_path", + "description": "Relative path to the photorealistic image file inside the dataset distribution.", + "dataType": "sc:Text", + "source": { + "fileObject": {"@id": "manifest"}, + "extract": {"jsonPath": "$[*].image"} + } + }, + { + "@type": "cr:Field", + "@id": "instances/image_sha256", + "name": "image_sha256", + "description": "SHA-256 hex digest of the photorealistic image PNG bytes.", + "dataType": "sc:Text", + "source": { + "fileObject": {"@id": "manifest"}, + "extract": {"jsonPath": "$[*].image_sha256"} + } + }, + { + "@type": "cr:Field", + "@id": "instances/image", + "name": "image", + "description": "The photorealistic benchmark image as a binary PNG, sourced from the images FileSet. Joined to the rest of the record by instances/id (which is the regex-stripped filename).", + "dataType": "sc:ImageObject", + "source": { + "fileSet": {"@id": "images"}, + "extract": {"fileProperty": "content"} + } + }, + { + "@type": "cr:Field", + "@id": "instances/question", + "name": "question", + "description": "The natural-language task question presented to the model. Includes the answer-format instruction (final answer enclosed in ... tags).", + "dataType": "sc:Text", + "source": { + "fileObject": {"@id": "manifest"}, + "extract": {"jsonPath": "$[*].question"} + } + }, + { + "@type": "cr:Field", + "@id": "instances/answer", + "name": "answer", + "description": "Ground-truth answer. Type varies by task: integer count (e.g. 10), single letter (e.g. 'H'), letter pair (e.g. 'E-F'), or comma-separated letter sequence (e.g. 'L, G, H, F, C, K').", + "dataType": "sc:Text", + "source": { + "fileObject": {"@id": "manifest"}, + "extract": {"jsonPath": "$[*].answer"} + } + } + ] + } + ] +} diff --git a/data/annotations/arrow_chain.jsonl b/data/annotations/arrow_chain.jsonl new file mode 100644 index 0000000000000000000000000000000000000000..b4870bef917dce8f88b7b4126d35376bdbbe0ba1 --- /dev/null +++ b/data/annotations/arrow_chain.jsonl @@ -0,0 +1,5 @@ +{"width": 980, "height": 980, "num_hops": 15, "num_decoys": 40, "arrow_radius": 26.0, "terminus_radius": 26.0, "chain": [{"x": 522.8166440225942, "y": 468.55575983151533, "dir": 2.3766789376444217}, {"x": 371.17673155274855, "y": 614.1070236197154, "dir": 2.4574613459994947}, {"x": 290.2483575671503, "y": 680.1055066150918, "dir": -2.999009666044033}, {"x": 124.23431877957675, "y": 656.2730054867259, "dir": -1.40503686622824}, {"x": 158.2704036273674, "y": 452.8228686575884, "dir": 0.10304676659950365}, {"x": 314.1070713661408, "y": 468.9384155161839, "dir": -0.6967975426966935}, {"x": 412.13176551897425, "y": 386.9085431288311, "dir": 0.03261660068534492}, {"x": 516.8734122577162, "y": 390.32607158576013, "dir": 0.5603239282618767}, {"x": 701.2028576418405, "y": 505.9745278458641, "dir": 0.7296116392746714}, {"x": 856.0429476070258, "y": 644.4353825869066, "dir": 2.26669383656258}, {"x": 708.19698412317, "y": 821.4329327736771, "dir": -2.334336716513142}, {"x": 603.1028191996147, "y": 711.6410899783302, "dir": -1.4892995262950441}, {"x": 615.9809227214164, "y": 553.9713427591638, "dir": -1.708728295096792}, {"x": 584.5758681029935, "y": 327.7319932983463, "dir": -0.7590645281630918}, {"x": 728.0510193376806, "y": 191.62106231304924, "dir": 1.4613699914388887}], "decoys": [{"x": 543.645593999667, "y": 826.245524709013, "dir": 3.2779318104771207}, {"x": 485.68643179071, "y": 866.017015856981, "dir": 2.152849086306564}, {"x": 284.34684482387684, "y": 770.4325376400496, "dir": 3.254400379285789}, {"x": 313.09621448968625, "y": 551.8539854701578, "dir": 1.521187555996856}, {"x": 255.01961715542316, "y": 405.61122820226007, "dir": 4.02082109640801}, {"x": 455.4362985531568, "y": 644.0860932021125, "dir": 5.157192096829649}, {"x": 177.70012345919454, "y": 193.24712841363834, "dir": 1.893193201251688}, {"x": 390.0597486472437, "y": 297.0293946146676, "dir": 1.9030235883205728}, {"x": 294.6471309816154, "y": 229.85211718180824, "dir": 4.150274018231234}, {"x": 811.679006670908, "y": 267.2520743939792, "dir": 0.6560023198937338}, {"x": 846.5051991728827, "y": 481.15357977570204, "dir": 4.994975338164344}, {"x": 820.5907984560195, "y": 352.6952062448964, "dir": 4.318113580678957}, {"x": 511.0929861342116, "y": 757.8892291169661, "dir": 2.089449160426087}, {"x": 633.6288292990592, "y": 136.6891376377117, "dir": 1.2067463224609292}, {"x": 441.19337211350484, "y": 212.65220460995914, "dir": 1.6805230755383842}, {"x": 147.2571790959945, "y": 720.4833954423252, "dir": 5.517294845280439}, {"x": 694.7885899497977, "y": 361.4799544053519, "dir": 3.2294689849553464}, {"x": 146.9748207402772, "y": 790.9842912558544, "dir": 1.9866344280999568}, {"x": 796.8531424597512, "y": 155.4362081471095, "dir": 1.3292768028414028}, {"x": 200.66723952667994, "y": 522.5715085419404, "dir": 6.055434089177372}, {"x": 406.6835627749514, "y": 475.4850566942857, "dir": 5.2337770178550524}, {"x": 465.8065004814811, "y": 114.76493560508467, "dir": 2.587388700782586}, {"x": 246.26993659359738, "y": 852.8357350756188, "dir": 2.104189440563249}, {"x": 532.6217341306675, "y": 543.6725563348756, "dir": 1.651666595830063}, {"x": 621.7344627877186, "y": 832.9458812943433, "dir": 4.781667159337984}, {"x": 153.47292074522892, "y": 326.4134822680111, "dir": 5.972169153724803}, {"x": 862.3271052326628, "y": 182.54292037471413, "dir": 3.880006573315657}, {"x": 413.75168547856214, "y": 840.3496718198975, "dir": 0.23065888317264008}, {"x": 594.1838367786079, "y": 248.10650028858137, "dir": 3.841754290054504}, {"x": 546.2198946368295, "y": 111.1776609655214, "dir": 1.2063160800051633}, {"x": 303.1467661536599, "y": 326.3513854272194, "dir": 2.936337854595572}, {"x": 503.4705271675161, "y": 690.0177661546326, "dir": 1.887648517750681}, {"x": 116.5043310111564, "y": 161.25144067962353, "dir": 5.1500660113582875}, {"x": 864.4504596729672, "y": 838.3271247920865, "dir": 0.094046331273313}, {"x": 865.5151914938739, "y": 416.6421909460452, "dir": 0.4801646612265416}, {"x": 841.4522189914219, "y": 551.3402784097956, "dir": 4.744860716531376}, {"x": 351.6209018230319, "y": 180.17723285932126, "dir": 3.1270056139321905}, {"x": 365.30370969443356, "y": 718.263134182725, "dir": 1.1222477846794905}, {"x": 716.3831245546523, "y": 721.9454567469835, "dir": 5.213031947376286}, {"x": 679.4368743973457, "y": 429.23001790807655, "dir": 0.48031185850606045}], "termini": [{"x": 433.12656190546136, "y": 705.5597466248084, "label": "B"}, {"x": 528.83423792415, "y": 189.80069111196923, "label": "D"}, {"x": 483.9719383384686, "y": 302.26068298649835, "label": "G"}, {"x": 336.42524127748976, "y": 830.527838861177, "label": "C"}, {"x": 228.5523840459877, "y": 588.481290987329, "label": "A"}, {"x": 199.88862976814772, "y": 278.3950507840076, "label": "E"}, {"x": 206.8445841208963, "y": 128.8511680348361, "label": "F"}, {"x": 799.620218452795, "y": 843.0484461030989, "label": "H"}, {"x": 389.5925878979758, "y": 119.62151126048093, "label": "J"}, {"x": 667.8191800175238, "y": 613.7581018345487, "label": "I"}], "chosen_terminus_label": "H", "question": "The image shows many small pieces scattered across the canvas, each clearly oriented in a specific direction, plus several larger labeled terminus markers (each marker bears a distinct letter). The piece highlighted by a green ring is the starting point. From that piece, cast an infinitely thin ray (a mathematical half-line with zero width) from its centre along the direction it points; the next step is the FIRST other marker this ray enters. A marker only counts if the ray actually crosses its boundary \u2014 grazing nearby without entering does not count. Continue until the ray enters a labeled terminus marker and report its letter. Answer with a single letter. Provide your final answer enclosed in ... tags.", "answer": "H", "image": "images/arrow_chain_00000.png", "pool": "foot", "seed": 11} +{"width": 980, "height": 980, "num_hops": 17, "num_decoys": 40, "arrow_radius": 26.0, "terminus_radius": 26.0, "chain": [{"x": 397.79073771738564, "y": 762.9526544002506, "dir": 2.802894299864976}, {"x": 262.32281794937455, "y": 810.6743489973887, "dir": -2.1483294801194783}, {"x": 125.83465086604672, "y": 601.2237205133654, "dir": -1.0765256723788827}, {"x": 228.80681573551783, "y": 410.1404604730601, "dir": -0.08614014519527873}, {"x": 509.3915693633964, "y": 385.91089058040785, "dir": 0.5500198218055511}, {"x": 634.5130789560832, "y": 462.6269528593749, "dir": -0.22261860238305126}, {"x": 858.5156911558039, "y": 411.91935024927085, "dir": -1.8033448220829627}, {"x": 798.4959008004579, "y": 158.4929127145753, "dir": -2.9330693289130285}, {"x": 569.4124822564598, "y": 110.01905177572377, "dir": 2.87883880494848}, {"x": 326.36667353890556, "y": 175.39167658160687, "dir": 1.9157098577541418}, {"x": 273.32483231761455, "y": 323.02748416051367, "dir": 0.9939492811414715}, {"x": 398.25756814741527, "y": 515.0333725800097, "dir": 0.5926932045451289}, {"x": 619.5480935184958, "y": 664.0644161914238, "dir": -0.1915458529352948}, {"x": 796.2133341028447, "y": 629.804901822723, "dir": -1.6595034109755966}, {"x": 783.8550003066588, "y": 490.85431854302806, "dir": -2.2624511471973148}, {"x": 661.2539445838687, "y": 342.80721777701984, "dir": -2.314564491295491}, {"x": 556.2822005981258, "y": 228.71037679440295, "dir": 0.36561811505618386}], "decoys": [{"x": 521.5396995156178, "y": 286.5137403750974, "dir": 5.3648668915108795}, {"x": 643.1304447871271, "y": 802.6018046210465, "dir": 1.4932429957977427}, {"x": 126.19907193162479, "y": 812.6921696659925, "dir": 0.278978072620929}, {"x": 468.5874250979106, "y": 835.0372584897593, "dir": 1.8781032076356619}, {"x": 126.7975934292976, "y": 734.1380320536915, "dir": 1.6043286898502958}, {"x": 422.1654772470703, "y": 624.9848558765648, "dir": 4.996542780279238}, {"x": 202.0402357038314, "y": 570.3593829126612, "dir": 3.884953227692113}, {"x": 152.99857526265495, "y": 230.9466342582282, "dir": 4.333925781972885}, {"x": 158.73221654569866, "y": 353.94332504258284, "dir": 3.0530860146050696}, {"x": 429.4033030025542, "y": 239.3515754328865, "dir": 3.2764872756564607}, {"x": 132.64025589198047, "y": 165.20123215397592, "dir": 2.71162612271128}, {"x": 123.80603262417594, "y": 463.3935313373759, "dir": 4.309627490771437}, {"x": 583.7638786533349, "y": 727.0892024960785, "dir": 6.06935987081226}, {"x": 858.5060481058798, "y": 501.28455823036995, "dir": 3.1520126697512962}, {"x": 466.06742399145946, "y": 768.8209879032753, "dir": 2.5840860772962255}, {"x": 265.0903016378163, "y": 634.5700477812502, "dir": 3.0709189014718556}, {"x": 533.3023955346945, "y": 798.3250804438595, "dir": 3.1766853374554005}, {"x": 255.97459225797078, "y": 121.63912935305575, "dir": 2.8975346501430006}, {"x": 861.1873697652888, "y": 672.8755948888349, "dir": 3.6394367456790535}, {"x": 215.24916856575845, "y": 274.5823375404442, "dir": 2.525087159090158}, {"x": 723.5325634777379, "y": 589.2799060274178, "dir": 5.4144490987255285}, {"x": 490.425128279181, "y": 659.1941797193157, "dir": 5.986966534268388}, {"x": 573.0209562111315, "y": 556.5323783302905, "dir": 2.3693884456955376}, {"x": 456.7586568456098, "y": 299.855953070914, "dir": 5.8902713839559055}, {"x": 643.7939296834872, "y": 192.54041341279643, "dir": 1.2631930235916013}, {"x": 325.00547703950735, "y": 862.7412377544239, "dir": 1.9746986541642515}, {"x": 276.50261329923177, "y": 716.5908371459642, "dir": 3.7525758347064504}, {"x": 770.6937891004518, "y": 323.5493884518935, "dir": 4.5482248702555035}, {"x": 467.2222579720818, "y": 441.2638800487097, "dir": 5.537534694757201}, {"x": 401.99548056934856, "y": 849.1852337319458, "dir": 0.32250238492662076}, {"x": 658.8857719381645, "y": 732.4344757209519, "dir": 3.798228884935414}, {"x": 183.07018328932543, "y": 859.2659124626244, "dir": 5.752990565498215}, {"x": 344.6945983890613, "y": 332.07530977801895, "dir": 1.0294766976805292}, {"x": 233.1909795324372, "y": 208.3468438912492, "dir": 5.896720919940844}, {"x": 118.69570757607903, "y": 290.7299147247707, "dir": 3.7329933758941047}, {"x": 721.261779280448, "y": 213.57493184560315, "dir": 4.389495254657625}, {"x": 814.2788756229821, "y": 841.2591951880152, "dir": 0.43842407534854333}, {"x": 359.54392837665165, "y": 260.2126528551686, "dir": 1.6741107696342665}, {"x": 701.8612978057332, "y": 513.9311058807202, "dir": 4.741908315490108}, {"x": 865.8309811747395, "y": 574.651655617112, "dir": 4.44412296216683}], "termini": [{"x": 516.0886501437409, "y": 494.1889931221436, "label": "C"}, {"x": 776.6083129895404, "y": 726.0632501896825, "label": "G"}, {"x": 644.8095934588521, "y": 553.69674334215, "label": "H"}, {"x": 573.9667079557166, "y": 865.3247187666303, "label": "A"}, {"x": 729.3643594807411, "y": 846.5550891889065, "label": "J"}, {"x": 714.9101565109657, "y": 289.43801176792994, "label": "B"}, {"x": 264.8576839847272, "y": 464.2621441761273, "label": "F"}, {"x": 333.91202129989216, "y": 558.9697785219946, "label": "I"}, {"x": 341.60584035514245, "y": 688.1349342762894, "label": "E"}, {"x": 867.5375873018136, "y": 794.3450367219646, "label": "D"}], "chosen_terminus_label": "B", "question": "The image shows many small pieces scattered across the canvas, each clearly oriented in a specific direction, plus several larger labeled terminus markers (each marker bears a distinct letter). The piece highlighted by a green ring is the starting point. From that piece, cast an infinitely thin ray (a mathematical half-line with zero width) from its centre along the direction it points; the next step is the FIRST other marker this ray enters. A marker only counts if the ray actually crosses its boundary \u2014 grazing nearby without entering does not count. Continue until the ray enters a labeled terminus marker and report its letter. Answer with a single letter. Provide your final answer enclosed in ... tags.", "answer": "B", "image": "images/arrow_chain_00001.png", "pool": "fish", "seed": 22} +{"width": 980, "height": 980, "num_hops": 15, "num_decoys": 40, "arrow_radius": 26.0, "terminus_radius": 26.0, "chain": [{"x": 795.502253250225, "y": 778.9202833406912, "dir": 2.398810607216431}, {"x": 700.0614056061438, "y": 866.5544578572068, "dir": -2.522686978900401}, {"x": 578.7524430668544, "y": 780.1511509307875, "dir": -2.630922332459646}, {"x": 364.35757243181877, "y": 660.0387618750135, "dir": -1.8012231864739188}, {"x": 331.0602866726949, "y": 518.1027670819976, "dir": -2.4819066260843012}, {"x": 173.41157328275784, "y": 395.8301183394833, "dir": -1.0881561823621162}, {"x": 267.32958821629876, "y": 216.58740995410923, "dir": -0.41564770795168743}, {"x": 394.16464021714114, "y": 160.60718913368646, "dir": 0.4220106185613959}, {"x": 519.3360056547649, "y": 216.80741825489088, "dir": 0.013778532498284984}, {"x": 678.5864685418894, "y": 219.00179480087428, "dir": -0.8076292564051883}, {"x": 780.5704410350583, "y": 112.37951312677123, "dir": 0.8571052357966605}, {"x": 862.2358470712909, "y": 206.68494479282347, "dir": 1.5096317197147775}, {"x": 868.5283211120085, "y": 309.43431961695165, "dir": 2.0491395032706614}, {"x": 767.6375124691225, "y": 504.013857305555, "dir": 2.1641432360090134}, {"x": 692.7364262235099, "y": 615.0748641346869, "dir": -2.4661749138116407}], "decoys": [{"x": 661.6669892380737, "y": 341.19350529565486, "dir": 3.460083701624788}, {"x": 858.7103280324985, "y": 676.8979360753456, "dir": 4.501249801135706}, {"x": 745.2261325603625, "y": 233.42178221938752, "dir": 6.15617712261346}, {"x": 205.7465237835803, "y": 704.8502147680467, "dir": 1.9001228977199534}, {"x": 274.01738371029774, "y": 599.1968011239287, "dir": 0.6086558467575854}, {"x": 302.7438792070509, "y": 849.5841054226336, "dir": 4.154107291559996}, {"x": 686.4500663692536, "y": 723.3793572254672, "dir": 0.7715513503646851}, {"x": 766.2341001667834, "y": 718.8014516905013, "dir": 2.2877382443265355}, {"x": 259.53778695743347, "y": 757.673297059997, "dir": 3.4330848907314953}, {"x": 394.3385201118451, "y": 462.1611330776633, "dir": 1.3993427487409025}, {"x": 466.9073009313624, "y": 116.73085057192168, "dir": 1.8297459547305246}, {"x": 292.87553076796405, "y": 401.24169207835683, "dir": 0.8217493183750004}, {"x": 151.58122840380508, "y": 589.6271473460697, "dir": 3.438003996436236}, {"x": 393.56063483241564, "y": 821.8570330846328, "dir": 0.49840809132435665}, {"x": 139.27605154234317, "y": 119.61544120402603, "dir": 3.436316711066717}, {"x": 297.56329061746584, "y": 110.36918501936054, "dir": 2.813083067875707}, {"x": 811.5341159187494, "y": 258.27111537618396, "dir": 4.850933038496986}, {"x": 648.8033150670736, "y": 479.01115829267695, "dir": 5.1932337440663385}, {"x": 403.8021543074502, "y": 314.78230575068346, "dir": 3.0380254147166177}, {"x": 767.3252645130426, "y": 635.265457181938, "dir": 1.8064774486768582}, {"x": 540.1456058342778, "y": 403.5410033983287, "dir": 1.664376551866589}, {"x": 711.2674889965278, "y": 784.6786672269283, "dir": 1.9648279646837343}, {"x": 492.48293235067996, "y": 594.9857543870121, "dir": 1.0011435893107088}, {"x": 614.2215953734863, "y": 415.36064994886124, "dir": 0.8588784150202151}, {"x": 495.7803235991656, "y": 523.9513641945011, "dir": 5.0515772903085905}, {"x": 844.4220060231156, "y": 544.8877327061847, "dir": 5.227674840712144}, {"x": 628.417528674183, "y": 278.23517692914004, "dir": 2.0700216168017542}, {"x": 555.0505400153381, "y": 849.7048868668592, "dir": 5.719482365810369}, {"x": 369.93748411069674, "y": 240.37845104940658, "dir": 3.611771698003974}, {"x": 755.0210087737422, "y": 386.51411973320376, "dir": 6.099456605074564}, {"x": 492.9482967776989, "y": 308.08690469845, "dir": 1.761537920460199}, {"x": 459.87740921477246, "y": 250.12000525700643, "dir": 0.6684616933674143}, {"x": 723.7445039425608, "y": 298.73551047660067, "dir": 5.126724084294496}, {"x": 201.15768799832364, "y": 148.08584011044084, "dir": 2.03100378030813}, {"x": 828.9465098112757, "y": 842.3461276697958, "dir": 0.05259049577573621}, {"x": 111.06434207946992, "y": 694.3483678530237, "dir": 3.4479222404712506}, {"x": 387.1153933728516, "y": 746.8253072873266, "dir": 0.5318658710768733}, {"x": 198.45425594919948, "y": 481.3599523069398, "dir": 2.272309158355307}, {"x": 208.9331388361337, "y": 635.1340096280032, "dir": 0.6494573189203047}, {"x": 544.5739933097758, "y": 153.81132261915474, "dir": 1.366734059350855}], "termini": [{"x": 186.54919056875409, "y": 824.7206579637241, "label": "G"}, {"x": 395.5936090767498, "y": 566.7057147355624, "label": "A"}, {"x": 319.6816202723985, "y": 316.2167235631728, "label": "B"}, {"x": 623.9748536684913, "y": 677.1661592989233, "label": "D"}, {"x": 565.2712504828885, "y": 327.94182240368275, "label": "H"}, {"x": 686.4158902833519, "y": 136.44426419780913, "label": "F"}, {"x": 455.8175554121954, "y": 773.3180319094993, "label": "E"}, {"x": 497.66137593181946, "y": 668.3221506507596, "label": "C"}, {"x": 162.64788347451213, "y": 217.00762477226738, "label": "J"}, {"x": 867.2382434216795, "y": 770.8270785894656, "label": "I"}], "chosen_terminus_label": "B", "question": "The image shows many small pieces scattered across the canvas, each clearly oriented in a specific direction, plus several larger labeled terminus markers (each marker bears a distinct letter). The piece highlighted by a green ring is the starting point. From that piece, cast an infinitely thin ray (a mathematical half-line with zero width) from its centre along the direction it points; the next step is the FIRST other marker this ray enters. A marker only counts if the ray actually crosses its boundary \u2014 grazing nearby without entering does not count. Continue until the ray enters a labeled terminus marker and report its letter. Answer with a single letter. Provide your final answer enclosed in ... tags.", "answer": "B", "image": "images/arrow_chain_00002.png", "pool": "bird", "seed": 7} +{"width": 980, "height": 980, "num_hops": 15, "num_decoys": 40, "arrow_radius": 26.0, "terminus_radius": 26.0, "chain": [{"x": 737.3690898521855, "y": 529.1332848436077, "dir": 2.5167134196039402}, {"x": 534.5383280468989, "y": 675.435281167612, "dir": -2.7826004954587775}, {"x": 360.06084715345725, "y": 609.9621431234386, "dir": 3.1060921351990434}, {"x": 178.09247179823976, "y": 616.4248299537207, "dir": -1.7154414281969372}, {"x": 134.19769894701244, "y": 315.0788497720259, "dir": -0.8073197150190421}, {"x": 322.9479213633942, "y": 117.86637049810732, "dir": 0.036575559965712996}, {"x": 477.1641077913031, "y": 123.5094304720344, "dir": 0.888429533753667}, {"x": 565.6735612860556, "y": 232.43296099787858, "dir": 1.3786779298649878}, {"x": 600.8695054924506, "y": 413.3727053529212, "dir": 1.2505097571260824}, {"x": 692.7780267205956, "y": 690.4497322818678, "dir": 2.210011124566973}, {"x": 573.2122842849328, "y": 851.302513656968, "dir": -3.1344280030209566}, {"x": 397.12335260935987, "y": 850.0408764048528, "dir": -2.8704862819506944}, {"x": 279.06896051976526, "y": 817.2277035229449, "dir": -2.9144194231012426}, {"x": 126.83040336324203, "y": 782.0356929029681, "dir": -1.6376678487119452}, {"x": 110.71334494771551, "y": 541.3797799013391, "dir": 0.04502692120967415}], "decoys": [{"x": 711.1870309808542, "y": 763.691724151459, "dir": 2.93951987721132}, {"x": 670.8756667571372, "y": 169.62565720769283, "dir": 0.3414782582349503}, {"x": 703.0193627130717, "y": 417.0181915182342, "dir": 4.318117351707054}, {"x": 483.4316819519908, "y": 251.55504866785705, "dir": 5.581059269456021}, {"x": 747.6111168403933, "y": 841.4127985509376, "dir": 4.436586613839662}, {"x": 468.2569331624087, "y": 486.3314165785041, "dir": 0.9334031978797598}, {"x": 631.0709059848327, "y": 273.2748453104376, "dir": 2.428848118276887}, {"x": 566.6565828990142, "y": 146.22587557683082, "dir": 5.510463835439656}, {"x": 225.08618201479948, "y": 403.8723194419472, "dir": 0.31311754973235423}, {"x": 773.8877083766865, "y": 651.8343507947365, "dir": 1.1474344286675284}, {"x": 412.87141782018773, "y": 211.884841738109, "dir": 5.949674299122375}, {"x": 731.9082914131494, "y": 225.59555706443322, "dir": 2.5630878145268112}, {"x": 334.8800628873162, "y": 372.1786207566293, "dir": 5.704994279266484}, {"x": 479.5746922274152, "y": 398.07367199696785, "dir": 2.6829583007502134}, {"x": 544.7778792298993, "y": 494.2368878606493, "dir": 3.0106921157019984}, {"x": 118.36239590272845, "y": 848.6739021079451, "dir": 3.399329446465389}, {"x": 805.6696497130051, "y": 230.53124326805522, "dir": 3.8528463961221586}, {"x": 491.59940177919395, "y": 778.794299615087, "dir": 3.574595787107879}, {"x": 272.12153293031395, "y": 469.93409656166614, "dir": 0.22736629526944638}, {"x": 848.9954607615463, "y": 433.66985060039633, "dir": 0.12418306223583256}, {"x": 861.2557842034514, "y": 166.67420648577124, "dir": 5.86491164060848}, {"x": 842.9636252333884, "y": 529.8205455253126, "dir": 1.034910381435374}, {"x": 334.928351749659, "y": 690.1165938464516, "dir": 0.018967469930310364}, {"x": 730.4664867349732, "y": 118.19902063077126, "dir": 3.949947813261379}, {"x": 139.5503020151445, "y": 199.99642294875355, "dir": 0.9787290608113832}, {"x": 803.3213425170146, "y": 321.80231406807366, "dir": 0.12262802527759864}, {"x": 778.3048108815631, "y": 583.2891492375943, "dir": 1.0446748685093752}, {"x": 224.23382674059593, "y": 680.0773330173521, "dir": 0.15109501838310116}, {"x": 807.5355128608247, "y": 812.4556610490455, "dir": 5.737352842434323}, {"x": 578.3682190616303, "y": 743.4518275565575, "dir": 4.534445369269421}, {"x": 178.1418798215857, "y": 739.0586036324718, "dir": 3.567289803827924}, {"x": 314.8009717671171, "y": 210.30003615019, "dir": 3.175878173721713}, {"x": 360.17522474485725, "y": 781.6047938436928, "dir": 5.895233826606801}, {"x": 659.3743265623323, "y": 358.620758528462, "dir": 0.8262807197765834}, {"x": 416.3895291514079, "y": 435.19126792436055, "dir": 4.794654325637786}, {"x": 793.300821839503, "y": 722.3461657601518, "dir": 2.9671134880948573}, {"x": 515.5216249541559, "y": 334.15664924372913, "dir": 5.873917594682393}, {"x": 639.854641648566, "y": 859.3976318045627, "dir": 5.4327336896128}, {"x": 426.9273788601888, "y": 764.7549093271746, "dir": 0.5971027267538108}, {"x": 355.05311808272637, "y": 469.5286463133294, "dir": 0.692775744136298}], "termini": [{"x": 799.4984941469597, "y": 385.900454275569, "label": "B"}, {"x": 844.7015550684324, "y": 659.6942103181184, "label": "F"}, {"x": 868.8804632095198, "y": 780.3196027821309, "label": "D"}, {"x": 712.2215583482856, "y": 300.27262121451986, "label": "H"}, {"x": 404.052394566256, "y": 318.1064558793443, "label": "E"}, {"x": 539.5222976158584, "y": 560.700785880364, "label": "J"}, {"x": 257.61510691389014, "y": 324.1187592987771, "label": "I"}, {"x": 264.9275731552914, "y": 738.5251799204831, "label": "G"}, {"x": 421.9448675941253, "y": 699.1678329772026, "label": "A"}, {"x": 182.37309569780007, "y": 137.57740776721496, "label": "C"}], "chosen_terminus_label": "J", "question": "The image shows many small pieces scattered across the canvas, each clearly oriented in a specific direction, plus several larger labeled terminus markers (each marker bears a distinct letter). The piece highlighted by a green ring is the starting point. From that piece, cast an infinitely thin ray (a mathematical half-line with zero width) from its centre along the direction it points; the next step is the FIRST other marker this ray enters. A marker only counts if the ray actually crosses its boundary \u2014 grazing nearby without entering does not count. Continue until the ray enters a labeled terminus marker and report its letter. Answer with a single letter. Provide your final answer enclosed in ... tags.", "answer": "J", "image": "images/arrow_chain_00003.png", "pool": "key", "seed": 333} +{"width": 980, "height": 980, "num_hops": 16, "num_decoys": 40, "arrow_radius": 26.0, "terminus_radius": 26.0, "chain": [{"x": 550.8307964133126, "y": 638.4303801508297, "dir": -2.649153077456235}, {"x": 409.43973618908154, "y": 562.5704173573359, "dir": -2.114758246891904}, {"x": 298.86144090983163, "y": 379.7443903624645, "dir": -2.9915814099819853}, {"x": 172.66578470555464, "y": 360.6703310369094, "dir": -1.9129777874010294}, {"x": 133.66273538502134, "y": 251.17062998281958, "dir": -0.8946487799458436}, {"x": 240.78549297284385, "y": 117.65262142908628, "dir": 0.303801626573327}, {"x": 381.73343348131556, "y": 161.84072737285516, "dir": 0.7795718237969341}, {"x": 551.1839454659322, "y": 329.3281025886129, "dir": 0.31752404347264124}, {"x": 747.9908489824886, "y": 394.0074690546027, "dir": 0.20400562173467837}, {"x": 855.5470526125823, "y": 416.25909064502514, "dir": 1.638555956464199}, {"x": 841.6825919796559, "y": 620.5582472145908, "dir": 1.8729106151253903}, {"x": 800.5043709427478, "y": 752.6860939813856, "dir": 2.913503337131491}, {"x": 657.7683934570733, "y": 785.8192291466321, "dir": 2.9550366927961664}, {"x": 453.4912956408012, "y": 824.3766873937564, "dir": -2.7068918874416106}, {"x": 226.92985583435606, "y": 719.179033640515, "dir": -2.089226228995188}, {"x": 133.26457429919324, "y": 554.9919193234311, "dir": -1.7594202350693484}], "decoys": [{"x": 246.4777690581185, "y": 458.28729377500264, "dir": 2.8911073080727965}, {"x": 453.60970304195, "y": 736.7746753971566, "dir": 3.0984630725643196}, {"x": 671.6460855807336, "y": 193.38211208481414, "dir": 6.066002273893879}, {"x": 786.6975263542338, "y": 830.1449074066035, "dir": 3.096102996224052}, {"x": 155.35864265261023, "y": 123.47018658569947, "dir": 2.5188469679551138}, {"x": 360.34891607747363, "y": 269.01655446443215, "dir": 4.200525450788941}, {"x": 267.26991216245176, "y": 619.0079534065986, "dir": 1.943666559171882}, {"x": 618.0811265374402, "y": 728.4684497568985, "dir": 1.854188690289386}, {"x": 494.8325064158384, "y": 369.00563001672464, "dir": 6.283025705238679}, {"x": 480.3431545531592, "y": 534.9822720633122, "dir": 0.11830803625412971}, {"x": 293.3852856913511, "y": 285.50663258888943, "dir": 3.5435851137379384}, {"x": 754.4313667591965, "y": 475.4593282950824, "dir": 5.418528726722539}, {"x": 572.989753565097, "y": 497.83608146273775, "dir": 2.3897697491022964}, {"x": 120.48754103381356, "y": 658.9842609198953, "dir": 6.22487589374013}, {"x": 638.2189259187894, "y": 627.5031646765567, "dir": 4.717029596921798}, {"x": 822.0629486076514, "y": 144.79260835951914, "dir": 1.9122101765197863}, {"x": 728.588294344425, "y": 539.1822777765199, "dir": 1.3643820217441427}, {"x": 651.5624946326989, "y": 124.04192269648125, "dir": 5.512897674691212}, {"x": 292.0666971860491, "y": 195.6096344446473, "dir": 1.640841629008991}, {"x": 419.2627065182959, "y": 312.9495843961614, "dir": 0.1585326688548397}, {"x": 509.86482996901316, "y": 462.3071599904431, "dir": 3.889833444140799}, {"x": 788.2033337890646, "y": 581.2066622937255, "dir": 2.818093347721551}, {"x": 663.338967437736, "y": 862.3329116941972, "dir": 2.03899634219319}, {"x": 187.71588190810667, "y": 803.1003187609285, "dir": 4.726830096563127}, {"x": 300.8160734668634, "y": 684.7906416614964, "dir": 4.116530876206542}, {"x": 817.9847125096604, "y": 334.80996088244035, "dir": 1.8125891676819366}, {"x": 380.6461409338971, "y": 847.1571547185596, "dir": 5.554621297624121}, {"x": 613.8506970094941, "y": 410.9643030037579, "dir": 1.3075588758301326}, {"x": 387.6115270174128, "y": 697.036900574959, "dir": 5.246651167685891}, {"x": 327.0059874186244, "y": 578.5052289310877, "dir": 3.854818745760786}, {"x": 644.9571897568, "y": 558.0026014017265, "dir": 3.8414435271660854}, {"x": 578.6273307814156, "y": 568.2465807156445, "dir": 3.6811292918973417}, {"x": 605.6846222966685, "y": 262.09034478470625, "dir": 2.202760994519849}, {"x": 122.18981036062168, "y": 183.25951621591855, "dir": 0.9953841976951685}, {"x": 430.0593681159357, "y": 469.43751908263744, "dir": 5.132797538626687}, {"x": 865.3460312400038, "y": 244.70833570067327, "dir": 5.9374541183865785}, {"x": 217.99118134893092, "y": 563.3195779716046, "dir": 2.231911574429307}, {"x": 585.5374031913915, "y": 856.251188595919, "dir": 3.9227451150653643}, {"x": 716.2863891518393, "y": 642.2761411366388, "dir": 2.7335968320720316}, {"x": 533.4183370828533, "y": 133.0531284118201, "dir": 1.383481082745418}], "termini": [{"x": 400.3045550450442, "y": 387.8487367553969, "label": "I"}, {"x": 112.2458979258971, "y": 444.88494007544784, "label": "C"}, {"x": 449.3631109124521, "y": 120.06777621260349, "label": "J"}, {"x": 729.8291811150108, "y": 707.12400417989, "label": "A"}, {"x": 271.55236574416665, "y": 828.487593657753, "label": "F"}, {"x": 117.04761324232058, "y": 823.8003970250555, "label": "E"}, {"x": 798.0407614646055, "y": 215.78726851744153, "label": "G"}, {"x": 543.5869221395192, "y": 202.702712136932, "label": "B"}, {"x": 224.55458894883645, "y": 252.56600773070605, "label": "H"}, {"x": 699.5890305288738, "y": 300.3577501634841, "label": "D"}], "chosen_terminus_label": "C", "question": "The image shows many small pieces scattered across the canvas, each clearly oriented in a specific direction, plus several larger labeled terminus markers (each marker bears a distinct letter). The piece highlighted by a green ring is the starting point. From that piece, cast an infinitely thin ray (a mathematical half-line with zero width) from its centre along the direction it points; the next step is the FIRST other marker this ray enters. A marker only counts if the ray actually crosses its boundary \u2014 grazing nearby without entering does not count. Continue until the ray enters a labeled terminus marker and report its letter. Answer with a single letter. Provide your final answer enclosed in ... tags.", "answer": "C", "image": "images/arrow_chain_00004.png"} diff --git a/data/annotations/attribute_group_counting.jsonl b/data/annotations/attribute_group_counting.jsonl new file mode 100644 index 0000000000000000000000000000000000000000..dd789ed8c508a6782f8fb1d2b68558c299738361 --- /dev/null +++ b/data/annotations/attribute_group_counting.jsonl @@ -0,0 +1,5 @@ +{"width": 1273, "height": 1273, "num_groups": 3, "num_elements": 12, "answer": 3, "elements": [{"shape_id": 0, "group": 0, "x": 230.05, "y": 45.86}, {"shape_id": 0, "group": 0, "x": 1001.26, "y": 326.71}, {"shape_id": 0, "group": 0, "x": 785.48, "y": 1042.74}, {"shape_id": 0, "group": 0, "x": 1175.71, "y": 1138.77}, {"shape_id": 0, "group": 0, "x": 645.63, "y": 168.52}, {"shape_id": 1, "group": 1, "x": 934.79, "y": 955.86}, {"shape_id": 1, "group": 1, "x": 645.61, "y": 278.51}, {"shape_id": 1, "group": 1, "x": 486.58, "y": 234.08}, {"shape_id": 1, "group": 1, "x": 473.58, "y": 1213.55}, {"shape_id": 2, "group": 2, "x": 446.47, "y": 390.97}, {"shape_id": 2, "group": 2, "x": 1134.09, "y": 754.15}, {"shape_id": 2, "group": 2, "x": 1197.57, "y": 947.97}], "groups": [{"id": 0, "shape_id": 0, "size": 5, "shape_vertices": [[-0.4565, -0.6274], [-0.3859, -0.6413], [-0.3186, -0.6518], [-0.2548, -0.6608], [-0.1941, -0.6696], [-0.1357, -0.6794], [-0.0786, -0.6909], [-0.0215, -0.7041], [0.0366, -0.7188], [0.0967, -0.734], [0.1593, -0.7484], [0.2246, -0.7604], [0.2922, -0.7683], [0.3613, -0.7706], [0.4306, -0.7656], [0.4987, -0.7523], [0.5638, -0.73], [0.6242, -0.6985], [0.6782, -0.6582], [0.7246, -0.6099], [0.7624, -0.5547], [0.7912, -0.4943], [0.8111, -0.4302], [0.8226, -0.364], [0.8267, -0.2971], [0.8249, -0.2307], [0.8184, -0.1656], [0.8088, -0.1022], [0.7975, -0.0404], [0.7855, 0.0201], [0.7735, 0.0799], [0.7617, 0.14], [0.7498, 0.2009], [0.7372, 0.2634], [0.7227, 0.3279], [0.7051, 0.3943], [0.683, 0.462], [0.655, 0.5304], [0.6199, 0.5979], [0.5771, 0.6628], [0.5263, 0.7233], [0.4678, 0.7774], [0.4023, 0.8231], [0.3311, 0.8588], [0.256, 0.8833], [0.1789, 0.8959], [0.1019, 0.8965], [0.027, 0.8855], [-0.044, 0.8642], [-0.1099, 0.8342], [-0.1697, 0.7974], [-0.2233, 0.7561], [-0.271, 0.7126], [-0.3136, 0.6688], [-0.3524, 0.6265], [-0.3892, 0.587], [-0.4255, 0.5509], [-0.4631, 0.5183], [-0.5034, 0.4885], [-0.5473, 0.4607], [-0.5952, 0.4331], [-0.6469, 0.4042], [-0.7016, 0.3721], [-0.7576, 0.3352], [-0.8131, 0.2922], [-0.8656, 0.2421], [-0.9126, 0.1847], [-0.9516, 0.1203], [-0.9803, 0.0497], [-0.9969, -0.0255], [-1.0, -0.1033], [-0.9891, -0.1818], [-0.9642, -0.2584], [-0.9262, -0.331], [-0.8765, -0.3977], [-0.8172, -0.4569], [-0.7504, -0.5077], [-0.6787, -0.5496], [-0.6045, -0.5829], [-0.5298, -0.6084]]}, {"id": 1, "shape_id": 1, "size": 4, "shape_vertices": [[-0.0847, 0.7912], [-0.1508, 0.8359], [-0.2142, 0.8374], [-0.2574, 0.7699], [-0.2689, 0.6444], [-0.255, 0.5038], [-0.2391, 0.3965], [-0.2468, 0.3477], [-0.288, 0.3469], [-0.3494, 0.3606], [-0.4052, 0.3579], [-0.4393, 0.3304], [-0.4615, 0.2925], [-0.503, 0.2644], [-0.593, 0.252], [-0.7322, 0.2417], [-0.8829, 0.2112], [-0.9869, 0.1493], [-1.0, 0.065], [-0.9211, -0.0191], [-0.7956, -0.0852], [-0.6907, -0.1347], [-0.6569, -0.1877], [-0.7016, -0.2669], [-0.7882, -0.379], [-0.86, -0.5064], [-0.8717, -0.616], [-0.8086, -0.6773], [-0.6871, -0.6773], [-0.5395, -0.6243], [-0.3968, -0.5398], [-0.2782, -0.4476], [-0.1909, -0.3672], [-0.1335, -0.3122], [-0.0991, -0.2902], [-0.0784, -0.3003], [-0.0608, -0.3305], [-0.0395, -0.3602], [-0.0138, -0.3709], [0.0126, -0.3601], [0.0373, -0.3487], [0.0671, -0.3719], [0.1169, -0.4571], [0.2013, -0.6019], [0.3206, -0.7681], [0.4552, -0.8991], [0.5732, -0.9507], [0.6499, -0.9157], [0.6853, -0.8254], [0.7049, -0.7276], [0.7423, -0.6557], [0.8124, -0.611], [0.8969, -0.5685], [0.9522, -0.5005], [0.9381, -0.3987], [0.8467, -0.2795], [0.7114, -0.1702], [0.5879, -0.089], [0.5205, -0.0338], [0.5152, 0.0107], [0.5392, 0.0577], [0.5456, 0.1064], [0.5057, 0.1445], [0.4269, 0.1624], [0.3452, 0.166], [0.2995, 0.1763], [0.307, 0.217], [0.3554, 0.2977], [0.4142, 0.4084], [0.455, 0.5266], [0.4649, 0.6324], [0.4463, 0.7179], [0.4081, 0.7846], [0.3564, 0.8338], [0.2937, 0.8596], [0.2228, 0.8536], [0.1504, 0.8172], [0.0844, 0.7688], [0.0274, 0.7383], [-0.0261, 0.7474]]}, {"id": 2, "shape_id": 2, "size": 3, "shape_vertices": [[0.3343, 0.5266], [0.2649, 0.4729], [0.2155, 0.4402], [0.1904, 0.4515], [0.1836, 0.5159], [0.1819, 0.6246], [0.1707, 0.7528], [0.1403, 0.8675], [0.0894, 0.9374], [0.0252, 0.943], [-0.0399, 0.8829], [-0.0941, 0.7736], [-0.1314, 0.6442], [-0.1549, 0.5266], [-0.1761, 0.446], [-0.2104, 0.4133], [-0.2715, 0.423], [-0.3654, 0.4566], [-0.487, 0.4896], [-0.6212, 0.4998], [-0.747, 0.4742], [-0.8441, 0.4115], [-0.8986, 0.321], [-0.9076, 0.2175], [-0.8787, 0.1158], [-0.8266, 0.0259], [-0.7679, -0.0489], [-0.7151, -0.1106], [-0.6732, -0.1636], [-0.6399, -0.2111], [-0.6083, -0.2532], [-0.5715, -0.2876], [-0.5267, -0.3119], [-0.478, -0.327], [-0.4346, -0.3391], [-0.4076, -0.3596], [-0.4043, -0.4012], [-0.4251, -0.473], [-0.461, -0.5748], [-0.4966, -0.6946], [-0.5142, -0.81], [-0.5003, -0.8932], [-0.4502, -0.9198], [-0.3699, -0.8772], [-0.2741, -0.7701], [-0.1808, -0.621], [-0.1055, -0.4651], [-0.0551, -0.3404], [-0.0264, -0.2769], [-0.0077, -0.288], [0.0166, -0.3666], [0.0593, -0.4878], [0.1258, -0.6167], [0.2116, -0.719], [0.3044, -0.7711], [0.3895, -0.7652], [0.4556, -0.7098], [0.4997, -0.6245], [0.5287, -0.5316], [0.557, -0.4482], [0.6, -0.3808], [0.6676, -0.3255], [0.7587, -0.271], [0.8598, -0.2061], [0.9485, -0.125], [1.0, -0.0313], [0.996, 0.0634], [0.9315, 0.1441], [0.8172, 0.1986], [0.6776, 0.2235], [0.5431, 0.2261], [0.4417, 0.2223], [0.3898, 0.2308], [0.3883, 0.2656], [0.4228, 0.3299], [0.4696, 0.4143], [0.5034, 0.4995], [0.5063, 0.5634], [0.4726, 0.5892], [0.4098, 0.5732]]}], "radius": 36, "fourier_perturbation_amplitude": 0.1, "min_pairwise_silhouette_distance": 0.11, "blob_shapes": [[[-0.4565, -0.6274], [-0.3859, -0.6413], [-0.3186, -0.6518], [-0.2548, -0.6608], [-0.1941, -0.6696], [-0.1357, -0.6794], [-0.0786, -0.6909], [-0.0215, -0.7041], [0.0366, -0.7188], [0.0967, -0.734], [0.1593, -0.7484], [0.2246, -0.7604], [0.2922, -0.7683], [0.3613, -0.7706], [0.4306, -0.7656], [0.4987, -0.7523], [0.5638, -0.73], [0.6242, -0.6985], [0.6782, -0.6582], [0.7246, -0.6099], [0.7624, -0.5547], [0.7912, -0.4943], [0.8111, -0.4302], [0.8226, -0.364], [0.8267, -0.2971], [0.8249, -0.2307], [0.8184, -0.1656], [0.8088, -0.1022], [0.7975, -0.0404], [0.7855, 0.0201], [0.7735, 0.0799], [0.7617, 0.14], [0.7498, 0.2009], [0.7372, 0.2634], [0.7227, 0.3279], [0.7051, 0.3943], [0.683, 0.462], [0.655, 0.5304], [0.6199, 0.5979], [0.5771, 0.6628], [0.5263, 0.7233], [0.4678, 0.7774], [0.4023, 0.8231], [0.3311, 0.8588], [0.256, 0.8833], [0.1789, 0.8959], [0.1019, 0.8965], [0.027, 0.8855], [-0.044, 0.8642], [-0.1099, 0.8342], [-0.1697, 0.7974], [-0.2233, 0.7561], [-0.271, 0.7126], [-0.3136, 0.6688], [-0.3524, 0.6265], [-0.3892, 0.587], [-0.4255, 0.5509], [-0.4631, 0.5183], [-0.5034, 0.4885], [-0.5473, 0.4607], [-0.5952, 0.4331], [-0.6469, 0.4042], [-0.7016, 0.3721], [-0.7576, 0.3352], [-0.8131, 0.2922], [-0.8656, 0.2421], [-0.9126, 0.1847], [-0.9516, 0.1203], [-0.9803, 0.0497], [-0.9969, -0.0255], [-1.0, -0.1033], [-0.9891, -0.1818], [-0.9642, -0.2584], [-0.9262, -0.331], [-0.8765, -0.3977], [-0.8172, -0.4569], [-0.7504, -0.5077], [-0.6787, -0.5496], [-0.6045, -0.5829], [-0.5298, -0.6084]], [[-0.0847, 0.7912], [-0.1508, 0.8359], [-0.2142, 0.8374], [-0.2574, 0.7699], [-0.2689, 0.6444], [-0.255, 0.5038], [-0.2391, 0.3965], [-0.2468, 0.3477], [-0.288, 0.3469], [-0.3494, 0.3606], [-0.4052, 0.3579], [-0.4393, 0.3304], [-0.4615, 0.2925], [-0.503, 0.2644], [-0.593, 0.252], [-0.7322, 0.2417], [-0.8829, 0.2112], [-0.9869, 0.1493], [-1.0, 0.065], [-0.9211, -0.0191], [-0.7956, -0.0852], [-0.6907, -0.1347], [-0.6569, -0.1877], [-0.7016, -0.2669], [-0.7882, -0.379], [-0.86, -0.5064], [-0.8717, -0.616], [-0.8086, -0.6773], [-0.6871, -0.6773], [-0.5395, -0.6243], [-0.3968, -0.5398], [-0.2782, -0.4476], [-0.1909, -0.3672], [-0.1335, -0.3122], [-0.0991, -0.2902], [-0.0784, -0.3003], [-0.0608, -0.3305], [-0.0395, -0.3602], [-0.0138, -0.3709], [0.0126, -0.3601], [0.0373, -0.3487], [0.0671, -0.3719], [0.1169, -0.4571], [0.2013, -0.6019], [0.3206, -0.7681], [0.4552, -0.8991], [0.5732, -0.9507], [0.6499, -0.9157], [0.6853, -0.8254], [0.7049, -0.7276], [0.7423, -0.6557], [0.8124, -0.611], [0.8969, -0.5685], [0.9522, -0.5005], [0.9381, -0.3987], [0.8467, -0.2795], [0.7114, -0.1702], [0.5879, -0.089], [0.5205, -0.0338], [0.5152, 0.0107], [0.5392, 0.0577], [0.5456, 0.1064], [0.5057, 0.1445], [0.4269, 0.1624], [0.3452, 0.166], [0.2995, 0.1763], [0.307, 0.217], [0.3554, 0.2977], [0.4142, 0.4084], [0.455, 0.5266], [0.4649, 0.6324], [0.4463, 0.7179], [0.4081, 0.7846], [0.3564, 0.8338], [0.2937, 0.8596], [0.2228, 0.8536], [0.1504, 0.8172], [0.0844, 0.7688], [0.0274, 0.7383], [-0.0261, 0.7474]], [[0.3343, 0.5266], [0.2649, 0.4729], [0.2155, 0.4402], [0.1904, 0.4515], [0.1836, 0.5159], [0.1819, 0.6246], [0.1707, 0.7528], [0.1403, 0.8675], [0.0894, 0.9374], [0.0252, 0.943], [-0.0399, 0.8829], [-0.0941, 0.7736], [-0.1314, 0.6442], [-0.1549, 0.5266], [-0.1761, 0.446], [-0.2104, 0.4133], [-0.2715, 0.423], [-0.3654, 0.4566], [-0.487, 0.4896], [-0.6212, 0.4998], [-0.747, 0.4742], [-0.8441, 0.4115], [-0.8986, 0.321], [-0.9076, 0.2175], [-0.8787, 0.1158], [-0.8266, 0.0259], [-0.7679, -0.0489], [-0.7151, -0.1106], [-0.6732, -0.1636], [-0.6399, -0.2111], [-0.6083, -0.2532], [-0.5715, -0.2876], [-0.5267, -0.3119], [-0.478, -0.327], [-0.4346, -0.3391], [-0.4076, -0.3596], [-0.4043, -0.4012], [-0.4251, -0.473], [-0.461, -0.5748], [-0.4966, -0.6946], [-0.5142, -0.81], [-0.5003, -0.8932], [-0.4502, -0.9198], [-0.3699, -0.8772], [-0.2741, -0.7701], [-0.1808, -0.621], [-0.1055, -0.4651], [-0.0551, -0.3404], [-0.0264, -0.2769], [-0.0077, -0.288], [0.0166, -0.3666], [0.0593, -0.4878], [0.1258, -0.6167], [0.2116, -0.719], [0.3044, -0.7711], [0.3895, -0.7652], [0.4556, -0.7098], [0.4997, -0.6245], [0.5287, -0.5316], [0.557, -0.4482], [0.6, -0.3808], [0.6676, -0.3255], [0.7587, -0.271], [0.8598, -0.2061], [0.9485, -0.125], [1.0, -0.0313], [0.996, 0.0634], [0.9315, 0.1441], [0.8172, 0.1986], [0.6776, 0.2235], [0.5431, 0.2261], [0.4417, 0.2223], [0.3898, 0.2308], [0.3883, 0.2656], [0.4228, 0.3299], [0.4696, 0.4143], [0.5034, 0.4995], [0.5063, 0.5634], [0.4726, 0.5892], [0.4098, 0.5732]]], "image": "images/attribute_group_counting_00000.png", "question": "How many distinct shapes are in the image? The image contains many small pieces scattered across the canvas. Two pieces belong to the same type if and only if they have exactly the same silhouette. Pieces may differ in surface texture or colour tone \u2014 only the silhouette outline matters. Count the total number of distinct silhouette types and report the count as a positive integer. Provide your final answer enclosed in ... tags."} +{"width": 1273, "height": 1273, "num_groups": 5, "num_elements": 9, "answer": 5, "elements": [{"shape_id": 0, "group": 0, "x": 987.67, "y": 312.61}, {"shape_id": 1, "group": 1, "x": 170.89, "y": 558.7}, {"shape_id": 2, "group": 2, "x": 70.14, "y": 271.63}, {"shape_id": 2, "group": 2, "x": 432.3, "y": 1069.72}, {"shape_id": 2, "group": 2, "x": 589.3, "y": 404.87}, {"shape_id": 2, "group": 2, "x": 1158.75, "y": 179.14}, {"shape_id": 3, "group": 3, "x": 1191.63, "y": 373.88}, {"shape_id": 4, "group": 4, "x": 1208.59, "y": 679.56}, {"shape_id": 4, "group": 4, "x": 804.72, "y": 517.22}], "groups": [{"id": 0, "shape_id": 0, "size": 1, "shape_vertices": [[-0.2126, -0.381], [-0.1782, -0.3967], [-0.1498, -0.4296], [-0.1221, -0.4793], [-0.0893, -0.54], [-0.0479, -0.6025], [0.0028, -0.6578], [0.0609, -0.7004], [0.1239, -0.7303], [0.1906, -0.7525], [0.2625, -0.7746], [0.3435, -0.8027], [0.4373, -0.8387], [0.5453, -0.8777], [0.6633, -0.9095], [0.7811, -0.9208], [0.8834, -0.9], [0.9538, -0.8412], [0.9789, -0.7461], [0.9533, -0.6247], [0.8815, -0.492], [0.7776, -0.3642], [0.662, -0.2544], [0.556, -0.1692], [0.4767, -0.1083], [0.4321, -0.0657], [0.4205, -0.0327], [0.4319, -0.0014], [0.452, 0.0326], [0.4674, 0.0698], [0.469, 0.1078], [0.455, 0.143], [0.4294, 0.1738], [0.4, 0.2011], [0.3743, 0.2287], [0.3569, 0.2613], [0.3476, 0.3028], [0.3418, 0.3534], [0.3331, 0.4098], [0.316, 0.4659], [0.2875, 0.5151], [0.2485, 0.5531], [0.2023, 0.5799], [0.153, 0.6006], [0.103, 0.6228], [0.052, 0.6545], [-0.003, 0.6998], [-0.0658, 0.7565], [-0.1384, 0.8163], [-0.2193, 0.8659], [-0.3023, 0.892], [-0.3787, 0.8849], [-0.4395, 0.8428], [-0.4798, 0.7723], [-0.501, 0.6869], [-0.5114, 0.6029], [-0.5247, 0.5346], [-0.5548, 0.4893], [-0.6113, 0.4659], [-0.6947, 0.4553], [-0.7954, 0.4439], [-0.8945, 0.419], [-0.9694, 0.3725], [-1.0, 0.3043], [-0.975, 0.2216], [-0.8957, 0.1362], [-0.776, 0.0603], [-0.6388, 0.0021], [-0.5099, -0.0368], [-0.4107, -0.0614], [-0.353, -0.0811], [-0.3364, -0.1058], [-0.35, -0.1417], [-0.3768, -0.1895], [-0.3992, -0.2438], [-0.4043, -0.296], [-0.3874, -0.3374], [-0.3512, -0.3631], [-0.304, -0.3739], [-0.2554, -0.3766]]}, {"id": 1, "shape_id": 1, "size": 1, "shape_vertices": [[0.3748, 0.975], [0.3225, 1.0], [0.2619, 0.9978], [0.1961, 0.9642], [0.1302, 0.901], [0.0695, 0.8161], [0.0178, 0.7221], [-0.0241, 0.6334], [-0.0583, 0.563], [-0.09, 0.5198], [-0.1258, 0.5065], [-0.1715, 0.519], [-0.2313, 0.5479], [-0.3055, 0.5806], [-0.391, 0.6043], [-0.4814, 0.6087], [-0.569, 0.588], [-0.6462, 0.5416], [-0.7075, 0.4734], [-0.7507, 0.3899], [-0.7772, 0.2988], [-0.7912, 0.2062], [-0.7985, 0.1161], [-0.8047, 0.0293], [-0.8134, -0.0553], [-0.8256, -0.1397], [-0.8395, -0.2256], [-0.8509, -0.3129], [-0.8547, -0.3996], [-0.8467, -0.4821], [-0.8243, -0.556], [-0.7875, -0.6179], [-0.739, -0.6661], [-0.683, -0.7016], [-0.6244, -0.7273], [-0.5674, -0.7478], [-0.5144, -0.7674], [-0.466, -0.7891], [-0.4207, -0.8134], [-0.3764, -0.8383], [-0.3305, -0.8597], [-0.2816, -0.8731], [-0.2295, -0.8746], [-0.1754, -0.8628], [-0.1213, -0.8392], [-0.0689, -0.8083], [-0.0191, -0.776], [0.0284, -0.7484], [0.0756, -0.7301], [0.1251, -0.7221], [0.1794, -0.7223], [0.2396, -0.725], [0.3052, -0.7231], [0.3733, -0.7094], [0.4391, -0.6787], [0.4976, -0.6292], [0.5446, -0.5628], [0.5779, -0.4844], [0.5985, -0.4005], [0.6105, -0.3171], [0.6199, -0.2383], [0.6337, -0.1652], [0.6573, -0.0955], [0.6936, -0.0252], [0.7413, 0.0504], [0.7952, 0.1346], [0.8474, 0.2277], [0.8888, 0.3269], [0.9114, 0.4261], [0.9099, 0.5181], [0.8837, 0.5961], [0.8361, 0.656], [0.7739, 0.6976], [0.7057, 0.7249], [0.6394, 0.7449], [0.5809, 0.7656], [0.5322, 0.7939], [0.4919, 0.833], [0.4557, 0.8811], [0.4183, 0.9318]]}, {"id": 2, "shape_id": 2, "size": 4, "shape_vertices": [[-0.0095, -0.7447], [0.043, -0.666], [0.0836, -0.5867], [0.1275, -0.5733], [0.1975, -0.6472], [0.3021, -0.7704], [0.4217, -0.8701], [0.518, -0.8859], [0.5596, -0.8057], [0.5464, -0.6684], [0.5103, -0.5329], [0.4918, -0.4387], [0.5084, -0.3862], [0.5403, -0.3467], [0.5454, -0.2918], [0.495, -0.2162], [0.4007, -0.1382], [0.3135, -0.0809], [0.2912, -0.0508], [0.3588, -0.0333], [0.4892, -0.0062], [0.6195, 0.0417], [0.6907, 0.1024], [0.6822, 0.1576], [0.6185, 0.1959], [0.545, 0.2217], [0.494, 0.2483], [0.4664, 0.2828], [0.4402, 0.317], [0.3963, 0.3358], [0.3384, 0.3359], [0.2913, 0.3384], [0.2793, 0.3811], [0.3045, 0.4916], [0.3417, 0.6617], [0.3552, 0.8422], [0.3228, 0.9688], [0.2493, 1.0], [0.159, 0.943], [0.0764, 0.8476], [0.0098, 0.7715], [-0.0479, 0.743], [-0.1064, 0.7461], [-0.1643, 0.7388], [-0.2103, 0.689], [-0.2355, 0.6005], [-0.2475, 0.5106], [-0.2696, 0.4611], [-0.324, 0.4664], [-0.4105, 0.5022], [-0.4995, 0.5216], [-0.5467, 0.4877], [-0.5226, 0.397], [-0.4347, 0.279], [-0.3271, 0.175], [-0.2563, 0.1119], [-0.2595, 0.0895], [-0.3367, 0.0869], [-0.4558, 0.0795], [-0.5755, 0.0535], [-0.6673, 0.0085], [-0.7236, -0.0487], [-0.7507, -0.1113], [-0.7562, -0.1747], [-0.7431, -0.2354], [-0.7131, -0.2901], [-0.6723, -0.338], [-0.6312, -0.3827], [-0.5971, -0.4301], [-0.5659, -0.4796], [-0.5227, -0.5188], [-0.4535, -0.5268], [-0.3588, -0.4895], [-0.2586, -0.4175], [-0.1805, -0.3494], [-0.1411, -0.3346], [-0.1338, -0.4015], [-0.1331, -0.5341], [-0.1137, -0.6743], [-0.0681, -0.7555]]}, {"id": 3, "shape_id": 3, "size": 1, "shape_vertices": [[-0.1073, 0.5307], [-0.1078, 0.4091], [-0.1084, 0.3317], [-0.1304, 0.3313], [-0.1947, 0.4191], [-0.3115, 0.5753], [-0.4679, 0.7474], [-0.6246, 0.8668], [-0.7287, 0.8796], [-0.7404, 0.7758], [-0.6574, 0.5946], [-0.5198, 0.4022], [-0.3905, 0.2547], [-0.3222, 0.1732], [-0.3349, 0.1434], [-0.4165, 0.1342], [-0.5404, 0.1181], [-0.6826, 0.0788], [-0.8238, 0.0097], [-0.9405, -0.088], [-1.0, -0.2022], [-0.9725, -0.307], [-0.8527, -0.3715], [-0.6719, -0.3793], [-0.4867, -0.3432], [-0.3479, -0.2995], [-0.2729, -0.283], [-0.2428, -0.3012], [-0.224, -0.3318], [-0.1962, -0.348], [-0.1625, -0.3484], [-0.1379, -0.3631], [-0.1288, -0.4273], [-0.1247, -0.5441], [-0.1087, -0.6722], [-0.0735, -0.7538], [-0.0274, -0.7618], [0.0175, -0.7243], [0.0585, -0.7016], [0.1044, -0.7331], [0.1615, -0.7985], [0.2192, -0.8323], [0.2549, -0.7799], [0.256, -0.6506], [0.2407, -0.5181], [0.2528, -0.4668], [0.3283, -0.5245], [0.4603, -0.6389], [0.5931, -0.7159], [0.6584, -0.6898], [0.6275, -0.5676], [0.5362, -0.4148], [0.4602, -0.3001], [0.4574, -0.2459], [0.524, -0.2244], [0.5996, -0.1932], [0.6196, -0.1354], [0.5697, -0.0657], [0.4985, -0.0059], [0.4775, 0.0447], [0.5414, 0.1095], [0.658, 0.2077], [0.7511, 0.3272], [0.7571, 0.4274], [0.6689, 0.4716], [0.536, 0.4615], [0.4235, 0.4393], [0.3659, 0.4539], [0.3506, 0.5193], [0.3391, 0.6017], [0.3016, 0.6467], [0.2376, 0.6255], [0.1685, 0.5591], [0.1155, 0.5038], [0.0821, 0.508], [0.0563, 0.5775], [0.0242, 0.6726], [-0.0178, 0.7367], [-0.0609, 0.7309], [-0.093, 0.6526]]}, {"id": 4, "shape_id": 4, "size": 2, "shape_vertices": [[0.2412, -0.7082], [0.3472, -0.7526], [0.4479, -0.7552], [0.5248, -0.7099], [0.565, -0.6247], [0.5673, -0.5181], [0.5443, -0.4122], [0.5178, -0.3246], [0.5114, -0.2633], [0.541, -0.2255], [0.6091, -0.2003], [0.7032, -0.1749], [0.8009, -0.1398], [0.8781, -0.0921], [0.9179, -0.0353], [0.9159, 0.0234], [0.8808, 0.0777], [0.8295, 0.1247], [0.7803, 0.1658], [0.7458, 0.2056], [0.7286, 0.2481], [0.7223, 0.2948], [0.7154, 0.3429], [0.6965, 0.3871], [0.6599, 0.4214], [0.6063, 0.4428], [0.5427, 0.4524], [0.4783, 0.4555], [0.421, 0.4596], [0.3746, 0.4722], [0.3382, 0.4979], [0.3076, 0.5378], [0.277, 0.5901], [0.2409, 0.6511], [0.1954, 0.7164], [0.1376, 0.7817], [0.0665, 0.8419], [-0.0177, 0.8909], [-0.1121, 0.9212], [-0.2109, 0.9249], [-0.3051, 0.8959], [-0.3842, 0.8327], [-0.4388, 0.7399], [-0.4643, 0.6281], [-0.4629, 0.5119], [-0.4439, 0.4054], [-0.4214, 0.3191], [-0.4094, 0.2567], [-0.4175, 0.215], [-0.4475, 0.1865], [-0.4934, 0.1622], [-0.545, 0.1356], [-0.5928, 0.1035], [-0.6323, 0.0663], [-0.6661, 0.0256], [-0.7023, -0.018], [-0.7494, -0.0661], [-0.8108, -0.1219], [-0.8815, -0.1873], [-0.948, -0.2613], [-0.9927, -0.338], [-1.0, -0.4081], [-0.9634, -0.4618], [-0.8879, -0.4934], [-0.7891, -0.504], [-0.6872, -0.5019], [-0.5997, -0.4999], [-0.5354, -0.5098], [-0.4922, -0.5373], [-0.4591, -0.5788], [-0.4229, -0.6225], [-0.3735, -0.6531], [-0.3091, -0.6585], [-0.2354, -0.636], [-0.162, -0.5941], [-0.0969, -0.5505], [-0.0415, -0.5252], [0.0106, -0.5325], [0.07, -0.575], [0.1461, -0.6408]]}], "radius": 36, "fourier_perturbation_amplitude": 0.1, "min_pairwise_silhouette_distance": 0.11, "blob_shapes": [[[-0.2126, -0.381], [-0.1782, -0.3967], [-0.1498, -0.4296], [-0.1221, -0.4793], [-0.0893, -0.54], [-0.0479, -0.6025], [0.0028, -0.6578], [0.0609, -0.7004], [0.1239, -0.7303], [0.1906, -0.7525], [0.2625, -0.7746], [0.3435, -0.8027], [0.4373, -0.8387], [0.5453, -0.8777], [0.6633, -0.9095], [0.7811, -0.9208], [0.8834, -0.9], [0.9538, -0.8412], [0.9789, -0.7461], [0.9533, -0.6247], [0.8815, -0.492], [0.7776, -0.3642], [0.662, -0.2544], [0.556, -0.1692], [0.4767, -0.1083], [0.4321, -0.0657], [0.4205, -0.0327], [0.4319, -0.0014], [0.452, 0.0326], [0.4674, 0.0698], [0.469, 0.1078], [0.455, 0.143], [0.4294, 0.1738], [0.4, 0.2011], [0.3743, 0.2287], [0.3569, 0.2613], [0.3476, 0.3028], [0.3418, 0.3534], [0.3331, 0.4098], [0.316, 0.4659], [0.2875, 0.5151], [0.2485, 0.5531], [0.2023, 0.5799], [0.153, 0.6006], [0.103, 0.6228], [0.052, 0.6545], [-0.003, 0.6998], [-0.0658, 0.7565], [-0.1384, 0.8163], [-0.2193, 0.8659], [-0.3023, 0.892], [-0.3787, 0.8849], [-0.4395, 0.8428], [-0.4798, 0.7723], [-0.501, 0.6869], [-0.5114, 0.6029], [-0.5247, 0.5346], [-0.5548, 0.4893], [-0.6113, 0.4659], [-0.6947, 0.4553], [-0.7954, 0.4439], [-0.8945, 0.419], [-0.9694, 0.3725], [-1.0, 0.3043], [-0.975, 0.2216], [-0.8957, 0.1362], [-0.776, 0.0603], [-0.6388, 0.0021], [-0.5099, -0.0368], [-0.4107, -0.0614], [-0.353, -0.0811], [-0.3364, -0.1058], [-0.35, -0.1417], [-0.3768, -0.1895], [-0.3992, -0.2438], [-0.4043, -0.296], [-0.3874, -0.3374], [-0.3512, -0.3631], [-0.304, -0.3739], [-0.2554, -0.3766]], [[0.3748, 0.975], [0.3225, 1.0], [0.2619, 0.9978], [0.1961, 0.9642], [0.1302, 0.901], [0.0695, 0.8161], [0.0178, 0.7221], [-0.0241, 0.6334], [-0.0583, 0.563], [-0.09, 0.5198], [-0.1258, 0.5065], [-0.1715, 0.519], [-0.2313, 0.5479], [-0.3055, 0.5806], [-0.391, 0.6043], [-0.4814, 0.6087], [-0.569, 0.588], [-0.6462, 0.5416], [-0.7075, 0.4734], [-0.7507, 0.3899], [-0.7772, 0.2988], [-0.7912, 0.2062], [-0.7985, 0.1161], [-0.8047, 0.0293], [-0.8134, -0.0553], [-0.8256, -0.1397], [-0.8395, -0.2256], [-0.8509, -0.3129], [-0.8547, -0.3996], [-0.8467, -0.4821], [-0.8243, -0.556], [-0.7875, -0.6179], [-0.739, -0.6661], [-0.683, -0.7016], [-0.6244, -0.7273], [-0.5674, -0.7478], [-0.5144, -0.7674], [-0.466, -0.7891], [-0.4207, -0.8134], [-0.3764, -0.8383], [-0.3305, -0.8597], [-0.2816, -0.8731], [-0.2295, -0.8746], [-0.1754, -0.8628], [-0.1213, -0.8392], [-0.0689, -0.8083], [-0.0191, -0.776], [0.0284, -0.7484], [0.0756, -0.7301], [0.1251, -0.7221], [0.1794, -0.7223], [0.2396, -0.725], [0.3052, -0.7231], [0.3733, -0.7094], [0.4391, -0.6787], [0.4976, -0.6292], [0.5446, -0.5628], [0.5779, -0.4844], [0.5985, -0.4005], [0.6105, -0.3171], [0.6199, -0.2383], [0.6337, -0.1652], [0.6573, -0.0955], [0.6936, -0.0252], [0.7413, 0.0504], [0.7952, 0.1346], [0.8474, 0.2277], [0.8888, 0.3269], [0.9114, 0.4261], [0.9099, 0.5181], [0.8837, 0.5961], [0.8361, 0.656], [0.7739, 0.6976], [0.7057, 0.7249], [0.6394, 0.7449], [0.5809, 0.7656], [0.5322, 0.7939], [0.4919, 0.833], [0.4557, 0.8811], [0.4183, 0.9318]], [[-0.0095, -0.7447], [0.043, -0.666], [0.0836, -0.5867], [0.1275, -0.5733], [0.1975, -0.6472], [0.3021, -0.7704], [0.4217, -0.8701], [0.518, -0.8859], [0.5596, -0.8057], [0.5464, -0.6684], [0.5103, -0.5329], [0.4918, -0.4387], [0.5084, -0.3862], [0.5403, -0.3467], [0.5454, -0.2918], [0.495, -0.2162], [0.4007, -0.1382], [0.3135, -0.0809], [0.2912, -0.0508], [0.3588, -0.0333], [0.4892, -0.0062], [0.6195, 0.0417], [0.6907, 0.1024], [0.6822, 0.1576], [0.6185, 0.1959], [0.545, 0.2217], [0.494, 0.2483], [0.4664, 0.2828], [0.4402, 0.317], [0.3963, 0.3358], [0.3384, 0.3359], [0.2913, 0.3384], [0.2793, 0.3811], [0.3045, 0.4916], [0.3417, 0.6617], [0.3552, 0.8422], [0.3228, 0.9688], [0.2493, 1.0], [0.159, 0.943], [0.0764, 0.8476], [0.0098, 0.7715], [-0.0479, 0.743], [-0.1064, 0.7461], [-0.1643, 0.7388], [-0.2103, 0.689], [-0.2355, 0.6005], [-0.2475, 0.5106], [-0.2696, 0.4611], [-0.324, 0.4664], [-0.4105, 0.5022], [-0.4995, 0.5216], [-0.5467, 0.4877], [-0.5226, 0.397], [-0.4347, 0.279], [-0.3271, 0.175], [-0.2563, 0.1119], [-0.2595, 0.0895], [-0.3367, 0.0869], [-0.4558, 0.0795], [-0.5755, 0.0535], [-0.6673, 0.0085], [-0.7236, -0.0487], [-0.7507, -0.1113], [-0.7562, -0.1747], [-0.7431, -0.2354], [-0.7131, -0.2901], [-0.6723, -0.338], [-0.6312, -0.3827], [-0.5971, -0.4301], [-0.5659, -0.4796], [-0.5227, -0.5188], [-0.4535, -0.5268], [-0.3588, -0.4895], [-0.2586, -0.4175], [-0.1805, -0.3494], [-0.1411, -0.3346], [-0.1338, -0.4015], [-0.1331, -0.5341], [-0.1137, -0.6743], [-0.0681, -0.7555]], [[-0.1073, 0.5307], [-0.1078, 0.4091], [-0.1084, 0.3317], [-0.1304, 0.3313], [-0.1947, 0.4191], [-0.3115, 0.5753], [-0.4679, 0.7474], [-0.6246, 0.8668], [-0.7287, 0.8796], [-0.7404, 0.7758], [-0.6574, 0.5946], [-0.5198, 0.4022], [-0.3905, 0.2547], [-0.3222, 0.1732], [-0.3349, 0.1434], [-0.4165, 0.1342], [-0.5404, 0.1181], [-0.6826, 0.0788], [-0.8238, 0.0097], [-0.9405, -0.088], [-1.0, -0.2022], [-0.9725, -0.307], [-0.8527, -0.3715], [-0.6719, -0.3793], [-0.4867, -0.3432], [-0.3479, -0.2995], [-0.2729, -0.283], [-0.2428, -0.3012], [-0.224, -0.3318], [-0.1962, -0.348], [-0.1625, -0.3484], [-0.1379, -0.3631], [-0.1288, -0.4273], [-0.1247, -0.5441], [-0.1087, -0.6722], [-0.0735, -0.7538], [-0.0274, -0.7618], [0.0175, -0.7243], [0.0585, -0.7016], [0.1044, -0.7331], [0.1615, -0.7985], [0.2192, -0.8323], [0.2549, -0.7799], [0.256, -0.6506], [0.2407, -0.5181], [0.2528, -0.4668], [0.3283, -0.5245], [0.4603, -0.6389], [0.5931, -0.7159], [0.6584, -0.6898], [0.6275, -0.5676], [0.5362, -0.4148], [0.4602, -0.3001], [0.4574, -0.2459], [0.524, -0.2244], [0.5996, -0.1932], [0.6196, -0.1354], [0.5697, -0.0657], [0.4985, -0.0059], [0.4775, 0.0447], [0.5414, 0.1095], [0.658, 0.2077], [0.7511, 0.3272], [0.7571, 0.4274], [0.6689, 0.4716], [0.536, 0.4615], [0.4235, 0.4393], [0.3659, 0.4539], [0.3506, 0.5193], [0.3391, 0.6017], [0.3016, 0.6467], [0.2376, 0.6255], [0.1685, 0.5591], [0.1155, 0.5038], [0.0821, 0.508], [0.0563, 0.5775], [0.0242, 0.6726], [-0.0178, 0.7367], [-0.0609, 0.7309], [-0.093, 0.6526]], [[0.2412, -0.7082], [0.3472, -0.7526], [0.4479, -0.7552], [0.5248, -0.7099], [0.565, -0.6247], [0.5673, -0.5181], [0.5443, -0.4122], [0.5178, -0.3246], [0.5114, -0.2633], [0.541, -0.2255], [0.6091, -0.2003], [0.7032, -0.1749], [0.8009, -0.1398], [0.8781, -0.0921], [0.9179, -0.0353], [0.9159, 0.0234], [0.8808, 0.0777], [0.8295, 0.1247], [0.7803, 0.1658], [0.7458, 0.2056], [0.7286, 0.2481], [0.7223, 0.2948], [0.7154, 0.3429], [0.6965, 0.3871], [0.6599, 0.4214], [0.6063, 0.4428], [0.5427, 0.4524], [0.4783, 0.4555], [0.421, 0.4596], [0.3746, 0.4722], [0.3382, 0.4979], [0.3076, 0.5378], [0.277, 0.5901], [0.2409, 0.6511], [0.1954, 0.7164], [0.1376, 0.7817], [0.0665, 0.8419], [-0.0177, 0.8909], [-0.1121, 0.9212], [-0.2109, 0.9249], [-0.3051, 0.8959], [-0.3842, 0.8327], [-0.4388, 0.7399], [-0.4643, 0.6281], [-0.4629, 0.5119], [-0.4439, 0.4054], [-0.4214, 0.3191], [-0.4094, 0.2567], [-0.4175, 0.215], [-0.4475, 0.1865], [-0.4934, 0.1622], [-0.545, 0.1356], [-0.5928, 0.1035], [-0.6323, 0.0663], [-0.6661, 0.0256], [-0.7023, -0.018], [-0.7494, -0.0661], [-0.8108, -0.1219], [-0.8815, -0.1873], [-0.948, -0.2613], [-0.9927, -0.338], [-1.0, -0.4081], [-0.9634, -0.4618], [-0.8879, -0.4934], [-0.7891, -0.504], [-0.6872, -0.5019], [-0.5997, -0.4999], [-0.5354, -0.5098], [-0.4922, -0.5373], [-0.4591, -0.5788], [-0.4229, -0.6225], [-0.3735, -0.6531], [-0.3091, -0.6585], [-0.2354, -0.636], [-0.162, -0.5941], [-0.0969, -0.5505], [-0.0415, -0.5252], [0.0106, -0.5325], [0.07, -0.575], [0.1461, -0.6408]]], "image": "images/attribute_group_counting_00001.png", "question": "How many distinct shapes are in the image? The image contains many small pieces scattered across the canvas. Two pieces belong to the same type if and only if they have exactly the same silhouette. Pieces may differ in surface texture or colour tone \u2014 only the silhouette outline matters. Count the total number of distinct silhouette types and report the count as a positive integer. Provide your final answer enclosed in ... tags."} +{"width": 1273, "height": 1273, "num_groups": 6, "num_elements": 21, "answer": 6, "elements": [{"shape_id": 0, "group": 0, "x": 696.95, "y": 698.63}, {"shape_id": 0, "group": 0, "x": 725.91, "y": 1172.48}, {"shape_id": 0, "group": 0, "x": 257.79, "y": 737.96}, {"shape_id": 0, "group": 0, "x": 1061.84, "y": 434.14}, {"shape_id": 0, "group": 0, "x": 866.78, "y": 384.69}, {"shape_id": 1, "group": 1, "x": 472.16, "y": 941.41}, {"shape_id": 1, "group": 1, "x": 1230.49, "y": 841.34}, {"shape_id": 1, "group": 1, "x": 415.96, "y": 426.37}, {"shape_id": 1, "group": 1, "x": 785.61, "y": 945.63}, {"shape_id": 2, "group": 2, "x": 796.76, "y": 626.66}, {"shape_id": 2, "group": 2, "x": 696.06, "y": 582.79}, {"shape_id": 2, "group": 2, "x": 438.3, "y": 838.79}, {"shape_id": 2, "group": 2, "x": 533.35, "y": 265.04}, {"shape_id": 2, "group": 2, "x": 1165.84, "y": 1009.33}, {"shape_id": 3, "group": 3, "x": 326.52, "y": 230.31}, {"shape_id": 3, "group": 3, "x": 816.64, "y": 1131.21}, {"shape_id": 3, "group": 3, "x": 872.22, "y": 186.38}, {"shape_id": 3, "group": 3, "x": 1195.55, "y": 522.56}, {"shape_id": 3, "group": 3, "x": 941.2, "y": 601.43}, {"shape_id": 4, "group": 4, "x": 150.49, "y": 455.73}, {"shape_id": 5, "group": 5, "x": 362.45, "y": 89.84}], "groups": [{"id": 0, "shape_id": 0, "size": 5, "shape_vertices": [[0.4193, 0.4173], [0.3936, 0.46], [0.3679, 0.5069], [0.3406, 0.5583], [0.3102, 0.6137], [0.2751, 0.6723], [0.2337, 0.7323], [0.1851, 0.7919], [0.1287, 0.8486], [0.0645, 0.8997], [-0.0066, 0.9426], [-0.0835, 0.9749], [-0.1642, 0.9946], [-0.2464, 1.0], [-0.3275, 0.9903], [-0.4047, 0.9655], [-0.4755, 0.926], [-0.5375, 0.8734], [-0.5889, 0.8096], [-0.6285, 0.7372], [-0.6557, 0.6588], [-0.6708, 0.5774], [-0.6748, 0.4956], [-0.6691, 0.4161], [-0.6559, 0.3407], [-0.6374, 0.2707], [-0.6163, 0.207], [-0.595, 0.1496], [-0.5755, 0.0978], [-0.5595, 0.0506], [-0.5482, 0.0064], [-0.5421, -0.0363], [-0.541, -0.0794], [-0.544, -0.1244], [-0.5499, -0.1726], [-0.5569, -0.2248], [-0.563, -0.2813], [-0.5662, -0.3418], [-0.5646, -0.4056], [-0.5564, -0.4715], [-0.5403, -0.5378], [-0.5157, -0.6027], [-0.4821, -0.6644], [-0.4399, -0.7209], [-0.3896, -0.7708], [-0.3325, -0.8127], [-0.2699, -0.8458], [-0.2032, -0.8696], [-0.1341, -0.8841], [-0.0638, -0.8897], [0.0062, -0.8871], [0.0751, -0.8772], [0.1422, -0.861], [0.2069, -0.8396], [0.2691, -0.8137], [0.3287, -0.7842], [0.3859, -0.7515], [0.4406, -0.7159], [0.4928, -0.6775], [0.5423, -0.6361], [0.5888, -0.5916], [0.6317, -0.5437], [0.6704, -0.4924], [0.7038, -0.4377], [0.7312, -0.3798], [0.7518, -0.3193], [0.7647, -0.2569], [0.7696, -0.1935], [0.7663, -0.1302], [0.7551, -0.0683], [0.7366, -0.0086], [0.7117, 0.0477], [0.6817, 0.1001], [0.648, 0.1482], [0.6123, 0.1922], [0.576, 0.2325], [0.5405, 0.27], [0.5067, 0.3059], [0.4752, 0.3414], [0.4462, 0.3781]]}, {"id": 1, "shape_id": 1, "size": 4, "shape_vertices": [[0.3581, -0.0923], [0.3504, -0.0576], [0.359, -0.0265], [0.3865, 0.0058], [0.4302, 0.0442], [0.4821, 0.0922], [0.5308, 0.1492], [0.5646, 0.211], [0.5749, 0.2703], [0.5588, 0.3199], [0.5206, 0.3551], [0.4699, 0.3766], [0.4189, 0.3912], [0.3785, 0.4101], [0.3549, 0.4456], [0.3473, 0.5068], [0.3489, 0.5956], [0.3491, 0.7048], [0.337, 0.819], [0.3056, 0.9182], [0.2535, 0.9829], [0.1856, 1.0], [0.111, 0.9661], [0.0399, 0.8888], [-0.0197, 0.7846], [-0.0649, 0.6743], [-0.0978, 0.5773], [-0.1243, 0.5064], [-0.1517, 0.4658], [-0.1855, 0.4507], [-0.2272, 0.4501], [-0.2744, 0.4514], [-0.3215, 0.4442], [-0.3627, 0.4233], [-0.3944, 0.3892], [-0.4169, 0.3467], [-0.4345, 0.3019], [-0.4535, 0.2595], [-0.48, 0.2208], [-0.5162, 0.1837], [-0.5598, 0.1444], [-0.6034, 0.0992], [-0.6368, 0.047], [-0.6502, -0.0097], [-0.6377, -0.0656], [-0.5999, -0.1147], [-0.5433, -0.1527], [-0.4797, -0.1792], [-0.4219, -0.1984], [-0.3805, -0.2178], [-0.3607, -0.246], [-0.3612, -0.2895], [-0.3747, -0.35], [-0.391, -0.4236], [-0.3997, -0.5019], [-0.3937, -0.5745], [-0.3705, -0.6324], [-0.3322, -0.6707], [-0.284, -0.6902], [-0.2319, -0.6969], [-0.1804, -0.6994], [-0.131, -0.7059], [-0.0829, -0.7214], [-0.0335, -0.7459], [0.0195, -0.7743], [0.0769, -0.7986], [0.1373, -0.8106], [0.1973, -0.804], [0.253, -0.777], [0.3011, -0.7317], [0.3399, -0.6734], [0.3697, -0.6083], [0.392, -0.5417], [0.4085, -0.4768], [0.4197, -0.4142], [0.4252, -0.3536], [0.4234, -0.2942], [0.4136, -0.2366], [0.3965, -0.1824], [0.3759, -0.1338]]}, {"id": 2, "shape_id": 2, "size": 5, "shape_vertices": [[-0.61, 0.0979], [-0.6772, 0.0581], [-0.7813, 0.0093], [-0.8926, -0.0553], [-0.9747, -0.1335], [-1.0, -0.2139], [-0.9596, -0.282], [-0.8654, -0.3274], [-0.743, -0.348], [-0.6204, -0.3513], [-0.5182, -0.3494], [-0.4447, -0.3538], [-0.397, -0.3711], [-0.3662, -0.4021], [-0.3426, -0.4436], [-0.319, -0.4914], [-0.2909, -0.5411], [-0.2559, -0.5884], [-0.2126, -0.6273], [-0.1612, -0.6502], [-0.1043, -0.65], [-0.0471, -0.6244], [0.0049, -0.5791], [0.0488, -0.5287], [0.087, -0.4922], [0.1279, -0.4858], [0.1818, -0.515], [0.2555, -0.5712], [0.3472, -0.6341], [0.4455, -0.6791], [0.5326, -0.6872], [0.5922, -0.6521], [0.6159, -0.581], [0.6067, -0.4905], [0.5772, -0.3984], [0.5426, -0.3172], [0.5142, -0.2511], [0.4954, -0.198], [0.4839, -0.153], [0.4766, -0.1129], [0.4757, -0.0763], [0.4896, -0.042], [0.5292, -0.0063], [0.5997, 0.0372], [0.6937, 0.095], [0.7899, 0.1689], [0.8592, 0.2525], [0.8763, 0.3315], [0.8311, 0.3893], [0.734, 0.4156], [0.6123, 0.4128], [0.4986, 0.3966], [0.4175, 0.3902], [0.376, 0.4127], [0.363, 0.47], [0.3572, 0.5503], [0.3387, 0.6301], [0.298, 0.6853], [0.2385, 0.7036], [0.1713, 0.6908], [0.1071, 0.6672], [0.0495, 0.6565], [-0.0057, 0.6723], [-0.0654, 0.7092], [-0.1317, 0.7449], [-0.1978, 0.7508], [-0.2496, 0.7069], [-0.274, 0.6127], [-0.2681, 0.4895], [-0.2434, 0.3711], [-0.2237, 0.2886], [-0.2338, 0.2574], [-0.2873, 0.271], [-0.3785, 0.306], [-0.4839, 0.334], [-0.5724, 0.3346], [-0.6211, 0.3034], [-0.6261, 0.2502], [-0.6054, 0.1915], [-0.5903, 0.1399]]}, {"id": 3, "shape_id": 3, "size": 5, "shape_vertices": [[0.7589, 0.4291], [0.583, 0.3767], [0.4297, 0.3154], [0.3469, 0.2882], [0.3403, 0.3196], [0.3749, 0.3989], [0.404, 0.489], [0.4, 0.5549], [0.3659, 0.5882], [0.3217, 0.6095], [0.2823, 0.6465], [0.2461, 0.7079], [0.2006, 0.7729], [0.138, 0.8063], [0.064, 0.7861], [-0.0078, 0.7213], [-0.0692, 0.6476], [-0.1256, 0.6029], [-0.1911, 0.603], [-0.275, 0.6322], [-0.3707, 0.6556], [-0.4568, 0.6412], [-0.5094, 0.578], [-0.5182, 0.4798], [-0.4956, 0.3752], [-0.473, 0.2916], [-0.4871, 0.2415], [-0.5605, 0.2187], [-0.6877, 0.2034], [-0.8326, 0.1748], [-0.9416, 0.1229], [-0.9699, 0.0543], [-0.9069, -0.0135], [-0.7845, -0.0653], [-0.6621, -0.0994], [-0.5914, -0.1278], [-0.5864, -0.1655], [-0.6163, -0.2154], [-0.6311, -0.264], [-0.5993, -0.2936], [-0.5321, -0.3009], [-0.4751, -0.307], [-0.4722, -0.3466], [-0.5312, -0.4412], [-0.6145, -0.5772], [-0.6652, -0.7078], [-0.6451, -0.7808], [-0.5586, -0.7748], [-0.4452, -0.7157], [-0.3487, -0.6607], [-0.2885, -0.6606], [-0.2526, -0.7264], [-0.2137, -0.8236], [-0.1534, -0.8963], [-0.0735, -0.9025], [0.0091, -0.8359], [0.0773, -0.7234], [0.126, -0.6049], [0.162, -0.5113], [0.1976, -0.4542], [0.2425, -0.4288], [0.3003, -0.4216], [0.3672, -0.4167], [0.4323, -0.4003], [0.4814, -0.3645], [0.5043, -0.3108], [0.5048, -0.2503], [0.5056, -0.1973], [0.5381, -0.1592], [0.623, -0.1308], [0.751, -0.098], [0.8835, -0.0495], [0.9741, 0.0145], [1.0, 0.0833], [0.9773, 0.1467], [0.9474, 0.2048], [0.943, 0.2662], [0.96, 0.3355], [0.9591, 0.4012], [0.8959, 0.4389]]}, {"id": 4, "shape_id": 4, "size": 1, "shape_vertices": [[-0.507, 0.0291], [-0.4907, -0.0279], [-0.458, -0.0783], [-0.4192, -0.1201], [-0.3872, -0.1567], [-0.372, -0.1963], [-0.3778, -0.2481], [-0.4003, -0.318], [-0.4289, -0.4046], [-0.4503, -0.4987], [-0.4537, -0.5858], [-0.4342, -0.6518], [-0.3943, -0.6886], [-0.3423, -0.6986], [-0.2883, -0.6938], [-0.2403, -0.6917], [-0.2012, -0.7081], [-0.1683, -0.7504], [-0.136, -0.8136], [-0.0985, -0.8815], [-0.0535, -0.9325], [-0.0032, -0.9478], [0.0469, -0.9184], [0.0904, -0.8491], [0.1238, -0.7567], [0.1483, -0.6637], [0.1692, -0.5901], [0.1939, -0.5464], [0.2276, -0.5304], [0.2709, -0.5292], [0.3182, -0.5255], [0.3602, -0.5048], [0.3881, -0.4608], [0.3973, -0.3978], [0.3916, -0.3269], [0.382, -0.2614], [0.3841, -0.2095], [0.4122, -0.1712], [0.473, -0.1381], [0.5624, -0.0974], [0.665, -0.0382], [0.7578, 0.0431], [0.8169, 0.1397], [0.8242, 0.2361], [0.7733, 0.3131], [0.6711, 0.3541], [0.5361, 0.3521], [0.3928, 0.3121], [0.2659, 0.2509], [0.1737, 0.1924], [0.1248, 0.1611], [0.117, 0.1757], [0.1395, 0.2436], [0.1764, 0.3601], [0.2116, 0.5091], [0.232, 0.6677], [0.2307, 0.8119], [0.2067, 0.9214], [0.1646, 0.9846], [0.1118, 1.0], [0.056, 0.9758], [0.0031, 0.9271], [-0.0445, 0.8715], [-0.0877, 0.8242], [-0.1299, 0.7939], [-0.1745, 0.7812], [-0.2232, 0.7786], [-0.2747, 0.7741], [-0.3243, 0.7555], [-0.3659, 0.7148], [-0.3941, 0.6509], [-0.4068, 0.57], [-0.4065, 0.4827], [-0.4, 0.4004], [-0.3961, 0.3307], [-0.4022, 0.2752], [-0.4213, 0.2297], [-0.4503, 0.187], [-0.481, 0.1404], [-0.5028, 0.087]]}, {"id": 5, "shape_id": 5, "size": 1, "shape_vertices": [[0.5191, 0.8895], [0.4662, 0.9061], [0.3918, 0.8734], [0.3105, 0.8064], [0.2374, 0.7354], [0.1807, 0.692], [0.1385, 0.6943], [0.1011, 0.7396], [0.0583, 0.8076], [0.0044, 0.8713], [-0.0599, 0.909], [-0.1291, 0.9106], [-0.197, 0.8781], [-0.259, 0.8207], [-0.3137, 0.75], [-0.3633, 0.6771], [-0.413, 0.6114], [-0.4717, 0.5589], [-0.5487, 0.52], [-0.6493, 0.4877], [-0.7682, 0.4483], [-0.8864, 0.3876], [-0.9735, 0.2989], [-1.0, 0.1889], [-0.9515, 0.0762], [-0.8402, -0.0184], [-0.7043, -0.0835], [-0.5937, -0.1257], [-0.5479, -0.1656], [-0.5768, -0.226], [-0.6563, -0.3158], [-0.74, -0.423], [-0.7824, -0.5197], [-0.7611, -0.5787], [-0.6856, -0.5906], [-0.5892, -0.571], [-0.5087, -0.5523], [-0.4653, -0.5649], [-0.456, -0.619], [-0.4597, -0.6996], [-0.4527, -0.7758], [-0.4223, -0.8208], [-0.3716, -0.8283], [-0.3135, -0.8142], [-0.2597, -0.8046], [-0.2132, -0.8163], [-0.1683, -0.8442], [-0.1181, -0.8638], [-0.0613, -0.8483], [-0.004, -0.7883], [0.0462, -0.7016], [0.0887, -0.626], [0.1342, -0.5983], [0.1998, -0.6332], [0.2983, -0.7131], [0.4276, -0.7971], [0.5681, -0.8409], [0.6912, -0.819], [0.7737, -0.7333], [0.8095, -0.6081], [0.8118, -0.4738], [0.8034, -0.3513], [0.8022, -0.2463], [0.8109, -0.1532], [0.8172, -0.0654], [0.8041, 0.0176], [0.7629, 0.0905], [0.7005, 0.1483], [0.6368, 0.1925], [0.5924, 0.2321], [0.5773, 0.2778], [0.5855, 0.3347], [0.6005, 0.3989], [0.606, 0.4608], [0.5958, 0.5133], [0.5762, 0.5584], [0.5589, 0.6069], [0.5517, 0.6698], [0.5514, 0.7486], [0.5453, 0.8298]]}], "radius": 36, "fourier_perturbation_amplitude": 0.1, "min_pairwise_silhouette_distance": 0.11, "blob_shapes": [[[0.4193, 0.4173], [0.3936, 0.46], [0.3679, 0.5069], [0.3406, 0.5583], [0.3102, 0.6137], [0.2751, 0.6723], [0.2337, 0.7323], [0.1851, 0.7919], [0.1287, 0.8486], [0.0645, 0.8997], [-0.0066, 0.9426], [-0.0835, 0.9749], [-0.1642, 0.9946], [-0.2464, 1.0], [-0.3275, 0.9903], [-0.4047, 0.9655], [-0.4755, 0.926], [-0.5375, 0.8734], [-0.5889, 0.8096], [-0.6285, 0.7372], [-0.6557, 0.6588], [-0.6708, 0.5774], [-0.6748, 0.4956], [-0.6691, 0.4161], [-0.6559, 0.3407], [-0.6374, 0.2707], [-0.6163, 0.207], [-0.595, 0.1496], [-0.5755, 0.0978], [-0.5595, 0.0506], [-0.5482, 0.0064], [-0.5421, -0.0363], [-0.541, -0.0794], [-0.544, -0.1244], [-0.5499, -0.1726], [-0.5569, -0.2248], [-0.563, -0.2813], [-0.5662, -0.3418], [-0.5646, -0.4056], [-0.5564, -0.4715], [-0.5403, -0.5378], [-0.5157, -0.6027], [-0.4821, -0.6644], [-0.4399, -0.7209], [-0.3896, -0.7708], [-0.3325, -0.8127], [-0.2699, -0.8458], [-0.2032, -0.8696], [-0.1341, -0.8841], [-0.0638, -0.8897], [0.0062, -0.8871], [0.0751, -0.8772], [0.1422, -0.861], [0.2069, -0.8396], [0.2691, -0.8137], [0.3287, -0.7842], [0.3859, -0.7515], [0.4406, -0.7159], [0.4928, -0.6775], [0.5423, -0.6361], [0.5888, -0.5916], [0.6317, -0.5437], [0.6704, -0.4924], [0.7038, -0.4377], [0.7312, -0.3798], [0.7518, -0.3193], [0.7647, -0.2569], [0.7696, -0.1935], [0.7663, -0.1302], [0.7551, -0.0683], [0.7366, -0.0086], [0.7117, 0.0477], [0.6817, 0.1001], [0.648, 0.1482], [0.6123, 0.1922], [0.576, 0.2325], [0.5405, 0.27], [0.5067, 0.3059], [0.4752, 0.3414], [0.4462, 0.3781]], [[0.3581, -0.0923], [0.3504, -0.0576], [0.359, -0.0265], [0.3865, 0.0058], [0.4302, 0.0442], [0.4821, 0.0922], [0.5308, 0.1492], [0.5646, 0.211], [0.5749, 0.2703], [0.5588, 0.3199], [0.5206, 0.3551], [0.4699, 0.3766], [0.4189, 0.3912], [0.3785, 0.4101], [0.3549, 0.4456], [0.3473, 0.5068], [0.3489, 0.5956], [0.3491, 0.7048], [0.337, 0.819], [0.3056, 0.9182], [0.2535, 0.9829], [0.1856, 1.0], [0.111, 0.9661], [0.0399, 0.8888], [-0.0197, 0.7846], [-0.0649, 0.6743], [-0.0978, 0.5773], [-0.1243, 0.5064], [-0.1517, 0.4658], [-0.1855, 0.4507], [-0.2272, 0.4501], [-0.2744, 0.4514], [-0.3215, 0.4442], [-0.3627, 0.4233], [-0.3944, 0.3892], [-0.4169, 0.3467], [-0.4345, 0.3019], [-0.4535, 0.2595], [-0.48, 0.2208], [-0.5162, 0.1837], [-0.5598, 0.1444], [-0.6034, 0.0992], [-0.6368, 0.047], [-0.6502, -0.0097], [-0.6377, -0.0656], [-0.5999, -0.1147], [-0.5433, -0.1527], [-0.4797, -0.1792], [-0.4219, -0.1984], [-0.3805, -0.2178], [-0.3607, -0.246], [-0.3612, -0.2895], [-0.3747, -0.35], [-0.391, -0.4236], [-0.3997, -0.5019], [-0.3937, -0.5745], [-0.3705, -0.6324], [-0.3322, -0.6707], [-0.284, -0.6902], [-0.2319, -0.6969], [-0.1804, -0.6994], [-0.131, -0.7059], [-0.0829, -0.7214], [-0.0335, -0.7459], [0.0195, -0.7743], [0.0769, -0.7986], [0.1373, -0.8106], [0.1973, -0.804], [0.253, -0.777], [0.3011, -0.7317], [0.3399, -0.6734], [0.3697, -0.6083], [0.392, -0.5417], [0.4085, -0.4768], [0.4197, -0.4142], [0.4252, -0.3536], [0.4234, -0.2942], [0.4136, -0.2366], [0.3965, -0.1824], [0.3759, -0.1338]], [[-0.61, 0.0979], [-0.6772, 0.0581], [-0.7813, 0.0093], [-0.8926, -0.0553], [-0.9747, -0.1335], [-1.0, -0.2139], [-0.9596, -0.282], [-0.8654, -0.3274], [-0.743, -0.348], [-0.6204, -0.3513], [-0.5182, -0.3494], [-0.4447, -0.3538], [-0.397, -0.3711], [-0.3662, -0.4021], [-0.3426, -0.4436], [-0.319, -0.4914], [-0.2909, -0.5411], [-0.2559, -0.5884], [-0.2126, -0.6273], [-0.1612, -0.6502], [-0.1043, -0.65], [-0.0471, -0.6244], [0.0049, -0.5791], [0.0488, -0.5287], [0.087, -0.4922], [0.1279, -0.4858], [0.1818, -0.515], [0.2555, -0.5712], [0.3472, -0.6341], [0.4455, -0.6791], [0.5326, -0.6872], [0.5922, -0.6521], [0.6159, -0.581], [0.6067, -0.4905], [0.5772, -0.3984], [0.5426, -0.3172], [0.5142, -0.2511], [0.4954, -0.198], [0.4839, -0.153], [0.4766, -0.1129], [0.4757, -0.0763], [0.4896, -0.042], [0.5292, -0.0063], [0.5997, 0.0372], [0.6937, 0.095], [0.7899, 0.1689], [0.8592, 0.2525], [0.8763, 0.3315], [0.8311, 0.3893], [0.734, 0.4156], [0.6123, 0.4128], [0.4986, 0.3966], [0.4175, 0.3902], [0.376, 0.4127], [0.363, 0.47], [0.3572, 0.5503], [0.3387, 0.6301], [0.298, 0.6853], [0.2385, 0.7036], [0.1713, 0.6908], [0.1071, 0.6672], [0.0495, 0.6565], [-0.0057, 0.6723], [-0.0654, 0.7092], [-0.1317, 0.7449], [-0.1978, 0.7508], [-0.2496, 0.7069], [-0.274, 0.6127], [-0.2681, 0.4895], [-0.2434, 0.3711], [-0.2237, 0.2886], [-0.2338, 0.2574], [-0.2873, 0.271], [-0.3785, 0.306], [-0.4839, 0.334], [-0.5724, 0.3346], [-0.6211, 0.3034], [-0.6261, 0.2502], [-0.6054, 0.1915], [-0.5903, 0.1399]], [[0.7589, 0.4291], [0.583, 0.3767], [0.4297, 0.3154], [0.3469, 0.2882], [0.3403, 0.3196], [0.3749, 0.3989], [0.404, 0.489], [0.4, 0.5549], [0.3659, 0.5882], [0.3217, 0.6095], [0.2823, 0.6465], [0.2461, 0.7079], [0.2006, 0.7729], [0.138, 0.8063], [0.064, 0.7861], [-0.0078, 0.7213], [-0.0692, 0.6476], [-0.1256, 0.6029], [-0.1911, 0.603], [-0.275, 0.6322], [-0.3707, 0.6556], [-0.4568, 0.6412], [-0.5094, 0.578], [-0.5182, 0.4798], [-0.4956, 0.3752], [-0.473, 0.2916], [-0.4871, 0.2415], [-0.5605, 0.2187], [-0.6877, 0.2034], [-0.8326, 0.1748], [-0.9416, 0.1229], [-0.9699, 0.0543], [-0.9069, -0.0135], [-0.7845, -0.0653], [-0.6621, -0.0994], [-0.5914, -0.1278], [-0.5864, -0.1655], [-0.6163, -0.2154], [-0.6311, -0.264], [-0.5993, -0.2936], [-0.5321, -0.3009], [-0.4751, -0.307], [-0.4722, -0.3466], [-0.5312, -0.4412], [-0.6145, -0.5772], [-0.6652, -0.7078], [-0.6451, -0.7808], [-0.5586, -0.7748], [-0.4452, -0.7157], [-0.3487, -0.6607], [-0.2885, -0.6606], [-0.2526, -0.7264], [-0.2137, -0.8236], [-0.1534, -0.8963], [-0.0735, -0.9025], [0.0091, -0.8359], [0.0773, -0.7234], [0.126, -0.6049], [0.162, -0.5113], [0.1976, -0.4542], [0.2425, -0.4288], [0.3003, -0.4216], [0.3672, -0.4167], [0.4323, -0.4003], [0.4814, -0.3645], [0.5043, -0.3108], [0.5048, -0.2503], [0.5056, -0.1973], [0.5381, -0.1592], [0.623, -0.1308], [0.751, -0.098], [0.8835, -0.0495], [0.9741, 0.0145], [1.0, 0.0833], [0.9773, 0.1467], [0.9474, 0.2048], [0.943, 0.2662], [0.96, 0.3355], [0.9591, 0.4012], [0.8959, 0.4389]], [[-0.507, 0.0291], [-0.4907, -0.0279], [-0.458, -0.0783], [-0.4192, -0.1201], [-0.3872, -0.1567], [-0.372, -0.1963], [-0.3778, -0.2481], [-0.4003, -0.318], [-0.4289, -0.4046], [-0.4503, -0.4987], [-0.4537, -0.5858], [-0.4342, -0.6518], [-0.3943, -0.6886], [-0.3423, -0.6986], [-0.2883, -0.6938], [-0.2403, -0.6917], [-0.2012, -0.7081], [-0.1683, -0.7504], [-0.136, -0.8136], [-0.0985, -0.8815], [-0.0535, -0.9325], [-0.0032, -0.9478], [0.0469, -0.9184], [0.0904, -0.8491], [0.1238, -0.7567], [0.1483, -0.6637], [0.1692, -0.5901], [0.1939, -0.5464], [0.2276, -0.5304], [0.2709, -0.5292], [0.3182, -0.5255], [0.3602, -0.5048], [0.3881, -0.4608], [0.3973, -0.3978], [0.3916, -0.3269], [0.382, -0.2614], [0.3841, -0.2095], [0.4122, -0.1712], [0.473, -0.1381], [0.5624, -0.0974], [0.665, -0.0382], [0.7578, 0.0431], [0.8169, 0.1397], [0.8242, 0.2361], [0.7733, 0.3131], [0.6711, 0.3541], [0.5361, 0.3521], [0.3928, 0.3121], [0.2659, 0.2509], [0.1737, 0.1924], [0.1248, 0.1611], [0.117, 0.1757], [0.1395, 0.2436], [0.1764, 0.3601], [0.2116, 0.5091], [0.232, 0.6677], [0.2307, 0.8119], [0.2067, 0.9214], [0.1646, 0.9846], [0.1118, 1.0], [0.056, 0.9758], [0.0031, 0.9271], [-0.0445, 0.8715], [-0.0877, 0.8242], [-0.1299, 0.7939], [-0.1745, 0.7812], [-0.2232, 0.7786], [-0.2747, 0.7741], [-0.3243, 0.7555], [-0.3659, 0.7148], [-0.3941, 0.6509], [-0.4068, 0.57], [-0.4065, 0.4827], [-0.4, 0.4004], [-0.3961, 0.3307], [-0.4022, 0.2752], [-0.4213, 0.2297], [-0.4503, 0.187], [-0.481, 0.1404], [-0.5028, 0.087]], [[0.5191, 0.8895], [0.4662, 0.9061], [0.3918, 0.8734], [0.3105, 0.8064], [0.2374, 0.7354], [0.1807, 0.692], [0.1385, 0.6943], [0.1011, 0.7396], [0.0583, 0.8076], [0.0044, 0.8713], [-0.0599, 0.909], [-0.1291, 0.9106], [-0.197, 0.8781], [-0.259, 0.8207], [-0.3137, 0.75], [-0.3633, 0.6771], [-0.413, 0.6114], [-0.4717, 0.5589], [-0.5487, 0.52], [-0.6493, 0.4877], [-0.7682, 0.4483], [-0.8864, 0.3876], [-0.9735, 0.2989], [-1.0, 0.1889], [-0.9515, 0.0762], [-0.8402, -0.0184], [-0.7043, -0.0835], [-0.5937, -0.1257], [-0.5479, -0.1656], [-0.5768, -0.226], [-0.6563, -0.3158], [-0.74, -0.423], [-0.7824, -0.5197], [-0.7611, -0.5787], [-0.6856, -0.5906], [-0.5892, -0.571], [-0.5087, -0.5523], [-0.4653, -0.5649], [-0.456, -0.619], [-0.4597, -0.6996], [-0.4527, -0.7758], [-0.4223, -0.8208], [-0.3716, -0.8283], [-0.3135, -0.8142], [-0.2597, -0.8046], [-0.2132, -0.8163], [-0.1683, -0.8442], [-0.1181, -0.8638], [-0.0613, -0.8483], [-0.004, -0.7883], [0.0462, -0.7016], [0.0887, -0.626], [0.1342, -0.5983], [0.1998, -0.6332], [0.2983, -0.7131], [0.4276, -0.7971], [0.5681, -0.8409], [0.6912, -0.819], [0.7737, -0.7333], [0.8095, -0.6081], [0.8118, -0.4738], [0.8034, -0.3513], [0.8022, -0.2463], [0.8109, -0.1532], [0.8172, -0.0654], [0.8041, 0.0176], [0.7629, 0.0905], [0.7005, 0.1483], [0.6368, 0.1925], [0.5924, 0.2321], [0.5773, 0.2778], [0.5855, 0.3347], [0.6005, 0.3989], [0.606, 0.4608], [0.5958, 0.5133], [0.5762, 0.5584], [0.5589, 0.6069], [0.5517, 0.6698], [0.5514, 0.7486], [0.5453, 0.8298]]], "image": "images/attribute_group_counting_00002.png", "question": "How many distinct shapes are in the image? The image contains many small pieces scattered across the canvas. Two pieces belong to the same type if and only if they have exactly the same silhouette. Pieces may differ in surface texture or colour tone \u2014 only the silhouette outline matters. Count the total number of distinct silhouette types and report the count as a positive integer. Provide your final answer enclosed in ... tags."} +{"width": 1273, "height": 1273, "num_groups": 8, "num_elements": 36, "answer": 8, "elements": [{"shape_id": 0, "group": 0, "x": 797.37, "y": 784.56}, {"shape_id": 1, "group": 1, "x": 567.79, "y": 402.05}, {"shape_id": 1, "group": 1, "x": 400.39, "y": 293.16}, {"shape_id": 1, "group": 1, "x": 252.63, "y": 960.92}, {"shape_id": 1, "group": 1, "x": 410.92, "y": 445.68}, {"shape_id": 1, "group": 1, "x": 620.61, "y": 835.22}, {"shape_id": 2, "group": 2, "x": 494.07, "y": 878.67}, {"shape_id": 2, "group": 2, "x": 773.24, "y": 644.9}, {"shape_id": 2, "group": 2, "x": 648.96, "y": 177.58}, {"shape_id": 2, "group": 2, "x": 205.17, "y": 345.81}, {"shape_id": 2, "group": 2, "x": 930.2, "y": 250.9}, {"shape_id": 3, "group": 3, "x": 629.51, "y": 939.04}, {"shape_id": 3, "group": 3, "x": 636.81, "y": 1032.86}, {"shape_id": 3, "group": 3, "x": 146.79, "y": 1046.09}, {"shape_id": 3, "group": 3, "x": 322.92, "y": 216.12}, {"shape_id": 3, "group": 3, "x": 115.57, "y": 165.06}, {"shape_id": 4, "group": 4, "x": 331.3, "y": 742.17}, {"shape_id": 4, "group": 4, "x": 96.0, "y": 511.97}, {"shape_id": 4, "group": 4, "x": 64.28, "y": 1137.26}, {"shape_id": 4, "group": 4, "x": 538.64, "y": 178.94}, {"shape_id": 4, "group": 4, "x": 723.77, "y": 885.59}, {"shape_id": 5, "group": 5, "x": 1000.57, "y": 127.7}, {"shape_id": 5, "group": 5, "x": 1015.73, "y": 1011.08}, {"shape_id": 5, "group": 5, "x": 817.03, "y": 424.3}, {"shape_id": 5, "group": 5, "x": 1111.06, "y": 450.35}, {"shape_id": 5, "group": 5, "x": 664.99, "y": 512.57}, {"shape_id": 6, "group": 6, "x": 945.49, "y": 575.75}, {"shape_id": 6, "group": 6, "x": 1031.28, "y": 1135.61}, {"shape_id": 6, "group": 6, "x": 221.49, "y": 123.19}, {"shape_id": 6, "group": 6, "x": 66.8, "y": 428.15}, {"shape_id": 6, "group": 6, "x": 648.33, "y": 1119.14}, {"shape_id": 7, "group": 7, "x": 994.77, "y": 776.95}, {"shape_id": 7, "group": 7, "x": 754.6, "y": 51.14}, {"shape_id": 7, "group": 7, "x": 1067.05, "y": 726.58}, {"shape_id": 7, "group": 7, "x": 592.16, "y": 701.08}, {"shape_id": 7, "group": 7, "x": 432.14, "y": 51.19}], "groups": [{"id": 0, "shape_id": 0, "size": 1, "shape_vertices": [[-0.55, 0.5349], [-0.6102, 0.4881], [-0.6644, 0.4337], [-0.7103, 0.3728], [-0.7461, 0.3066], [-0.7708, 0.2373], [-0.784, 0.167], [-0.7867, 0.0978], [-0.7803, 0.0314], [-0.7673, -0.0311], [-0.7506, -0.0894], [-0.7331, -0.1442], [-0.7175, -0.1966], [-0.7058, -0.2486], [-0.6992, -0.3022], [-0.6976, -0.3593], [-0.6999, -0.4215], [-0.7039, -0.4891], [-0.7068, -0.562], [-0.7053, -0.6384], [-0.6962, -0.7158], [-0.6768, -0.7909], [-0.6453, -0.8595], [-0.6007, -0.9179], [-0.5437, -0.9624], [-0.4758, -0.9903], [-0.3996, -1.0], [-0.3185, -0.9912], [-0.2359, -0.9652], [-0.1553, -0.9243], [-0.0794, -0.8721], [-0.0103, -0.8127], [0.0511, -0.7503], [0.1051, -0.6888], [0.153, -0.6315], [0.1966, -0.5807], [0.2384, -0.5372], [0.281, -0.501], [0.3263, -0.4707], [0.376, -0.4442], [0.4306, -0.4188], [0.49, -0.3919], [0.5527, -0.3608], [0.6169, -0.3238], [0.6801, -0.2795], [0.7395, -0.2277], [0.7927, -0.1689], [0.8374, -0.1041], [0.8721, -0.0351], [0.8961, 0.0363], [0.9094, 0.1084], [0.9125, 0.1795], [0.9065, 0.2484], [0.8929, 0.3145], [0.8729, 0.3773], [0.8479, 0.4368], [0.8187, 0.493], [0.7859, 0.5461], [0.7497, 0.596], [0.7099, 0.6425], [0.6662, 0.685], [0.6185, 0.7227], [0.5665, 0.7547], [0.5106, 0.7801], [0.451, 0.7984], [0.3887, 0.809], [0.3246, 0.8122], [0.2598, 0.8086], [0.1953, 0.799], [0.1319, 0.785], [0.07, 0.768], [0.0095, 0.7494], [-0.0498, 0.7304], [-0.1086, 0.7117], [-0.168, 0.6934], [-0.2285, 0.6749], [-0.2908, 0.6553], [-0.355, 0.633], [-0.4204, 0.6065], [-0.486, 0.5742]]}, {"id": 1, "shape_id": 1, "size": 5, "shape_vertices": [[0.6145, -0.1204], [0.6407, -0.0643], [0.6439, -0.0046], [0.6094, 0.0519], [0.5353, 0.095], [0.4345, 0.1177], [0.331, 0.1215], [0.251, 0.1172], [0.2131, 0.1219], [0.2208, 0.1511], [0.2619, 0.2112], [0.3133, 0.2948], [0.3503, 0.3825], [0.3559, 0.4502], [0.3268, 0.4798], [0.2729, 0.4676], [0.2121, 0.4281], [0.1612, 0.3894], [0.1294, 0.383], [0.1143, 0.4307], [0.105, 0.5357], [0.0871, 0.6796], [0.0508, 0.8285], [-0.0052, 0.9447], [-0.0731, 1.0], [-0.1402, 0.985], [-0.1951, 0.9113], [-0.2333, 0.8056], [-0.2585, 0.6986], [-0.2809, 0.6144], [-0.3111, 0.5625], [-0.3547, 0.5372], [-0.4092, 0.5227], [-0.4646, 0.5016], [-0.5079, 0.4626], [-0.5296, 0.4039], [-0.5276, 0.3327], [-0.5092, 0.2599], [-0.4872, 0.1945], [-0.4745, 0.14], [-0.4784, 0.0938], [-0.497, 0.0499], [-0.5202, 0.0037], [-0.5343, -0.0455], [-0.5282, -0.0937], [-0.4983, -0.135], [-0.4501, -0.1652], [-0.3959, -0.1848], [-0.35, -0.2002], [-0.3228, -0.2209], [-0.3169, -0.2556], [-0.3275, -0.3081], [-0.3441, -0.3757], [-0.3556, -0.4499], [-0.3542, -0.5201], [-0.3372, -0.5777], [-0.3066, -0.6188], [-0.2668, -0.6445], [-0.2224, -0.6583], [-0.1761, -0.6635], [-0.1294, -0.6604], [-0.0829, -0.6469], [-0.0381, -0.6206], [0.0032, -0.5827], [0.0394, -0.5395], [0.0715, -0.5028], [0.104, -0.4856], [0.1439, -0.4971], [0.1988, -0.5372], [0.2722, -0.5954], [0.361, -0.6529], [0.4547, -0.6887], [0.5382, -0.6875], [0.597, -0.6447], [0.6234, -0.5677], [0.6198, -0.4727], [0.5986, -0.3774], [0.5773, -0.2946], [0.571, -0.228], [0.5856, -0.1728]]}, {"id": 2, "shape_id": 2, "size": 5, "shape_vertices": [[-0.6257, 0.4803], [-0.7177, 0.4251], [-0.7386, 0.327], [-0.7001, 0.2184], [-0.6424, 0.1256], [-0.6081, 0.0546], [-0.6175, -0.0048], [-0.6626, -0.0657], [-0.7197, -0.1338], [-0.7701, -0.2075], [-0.8114, -0.2848], [-0.8525, -0.3677], [-0.8979, -0.4595], [-0.9358, -0.5551], [-0.9407, -0.6365], [-0.8886, -0.6782], [-0.7752, -0.6623], [-0.6231, -0.593], [-0.4736, -0.5005], [-0.3668, -0.4302], [-0.3236, -0.4215], [-0.3379, -0.4904], [-0.3823, -0.6213], [-0.4216, -0.7733], [-0.4276, -0.8954], [-0.3887, -0.9444], [-0.3114, -0.8992], [-0.216, -0.7688], [-0.1269, -0.5907], [-0.0617, -0.4203], [-0.0237, -0.3114], [-0.0005, -0.2945], [0.0283, -0.3641], [0.079, -0.4814], [0.154, -0.5941], [0.2424, -0.6621], [0.3302, -0.6758], [0.4133, -0.6536], [0.5024, -0.6235], [0.6137, -0.6006], [0.7505, -0.5761], [0.89, -0.5272], [0.9875, -0.4372], [1.0, -0.312], [0.9151, -0.1789], [0.7656, -0.0687], [0.6177, 0.0048], [0.5366, 0.0532], [0.5514, 0.1025], [0.6401, 0.1725], [0.7448, 0.2614], [0.8065, 0.3479], [0.7967, 0.4077], [0.7268, 0.4311], [0.6335, 0.4286], [0.5527, 0.4218], [0.5005, 0.4276], [0.4735, 0.4505], [0.4608, 0.487], [0.4566, 0.5355], [0.4612, 0.6008], [0.4727, 0.686], [0.4793, 0.7789], [0.4618, 0.8472], [0.4071, 0.8524], [0.32, 0.7776], [0.2247, 0.6489], [0.1492, 0.531], [0.1063, 0.495], [0.0844, 0.5745], [0.0564, 0.7395], [0.0016, 0.9085], [-0.0772, 0.9931], [-0.1555, 0.9481], [-0.2064, 0.7961], [-0.2238, 0.6113], [-0.2312, 0.473], [-0.2666, 0.4216], [-0.3554, 0.4411], [-0.4887, 0.4783]]}, {"id": 3, "shape_id": 3, "size": 5, "shape_vertices": [[0.5811, 0.7744], [0.4775, 0.7241], [0.4151, 0.7216], [0.3676, 0.7411], [0.3079, 0.7324], [0.2288, 0.6586], [0.145, 0.5257], [0.0786, 0.3837], [0.0394, 0.2953], [0.0181, 0.2957], [-0.005, 0.3686], [-0.0417, 0.457], [-0.0869, 0.5011], [-0.1247, 0.4767], [-0.1461, 0.4083], [-0.1623, 0.3499], [-0.2021, 0.346], [-0.2909, 0.4027], [-0.4295, 0.4851], [-0.5869, 0.5417], [-0.7155, 0.5369], [-0.7782, 0.4686], [-0.769, 0.3632], [-0.7127, 0.2538], [-0.6452, 0.1614], [-0.5891, 0.0897], [-0.5448, 0.0328], [-0.5011, -0.0138], [-0.4555, -0.0511], [-0.4263, -0.083], [-0.4449, -0.1231], [-0.531, -0.1905], [-0.6723, -0.2971], [-0.8218, -0.4335], [-0.9212, -0.568], [-0.9331, -0.663], [-0.8621, -0.6992], [-0.7505, -0.6907], [-0.6512, -0.6779], [-0.5964, -0.7018], [-0.5817, -0.7753], [-0.5745, -0.871], [-0.5383, -0.9357], [-0.457, -0.9213], [-0.3421, -0.8137], [-0.2229, -0.6416], [-0.1271, -0.461], [-0.0668, -0.3264], [-0.0356, -0.2667], [-0.017, -0.2785], [0.0045, -0.3358], [0.0373, -0.4084], [0.0825, -0.4753], [0.1382, -0.5281], [0.2023, -0.5655], [0.2724, -0.587], [0.3441, -0.5893], [0.4103, -0.5682], [0.4624, -0.5222], [0.4936, -0.4555], [0.5029, -0.3773], [0.4965, -0.299], [0.4871, -0.2301], [0.49, -0.1745], [0.5166, -0.1292], [0.5675, -0.0864], [0.6287, -0.0378], [0.6752, 0.0186], [0.6834, 0.0766], [0.6458, 0.1258], [0.5818, 0.1609], [0.5324, 0.191], [0.5411, 0.2391], [0.6273, 0.3309], [0.7705, 0.475], [0.9156, 0.6506], [1.0, 0.8111], [0.9858, 0.9073], [0.8798, 0.9159], [0.7265, 0.855]]}, {"id": 4, "shape_id": 4, "size": 5, "shape_vertices": [[-0.3664, -0.2965], [-0.244, -0.2316], [-0.2394, -0.2661], [-0.3334, -0.4353], [-0.4417, -0.6818], [-0.4774, -0.8822], [-0.4104, -0.926], [-0.2796, -0.7941], [-0.1533, -0.5765], [-0.0758, -0.4136], [-0.0415, -0.403], [-0.013, -0.5371], [0.0389, -0.7154], [0.1092, -0.8182], [0.1686, -0.7868], [0.1945, -0.6532], [0.1947, -0.5054], [0.2006, -0.4196], [0.2378, -0.4114], [0.3002, -0.4366], [0.3512, -0.434], [0.3521, -0.3715], [0.2937, -0.265], [0.2078, -0.1599], [0.1478, -0.0963], [0.1552, -0.0846], [0.2349, -0.1052], [0.3547, -0.1265], [0.4662, -0.1261], [0.5305, -0.0997], [0.5342, -0.0575], [0.4903, -0.0141], [0.4276, 0.0214], [0.3789, 0.049], [0.3696, 0.0778], [0.4108, 0.1209], [0.4947, 0.189], [0.595, 0.2829], [0.6749, 0.3888], [0.7007, 0.481], [0.6574, 0.5321], [0.5567, 0.5284], [0.4334, 0.4818], [0.3296, 0.4303], [0.2744, 0.4236], [0.2694, 0.4979], [0.2887, 0.6513], [0.2943, 0.8361], [0.2593, 0.9754], [0.1832, 1.0], [0.0915, 0.8882], [0.0165, 0.6833], [-0.0258, 0.4744], [-0.0465, 0.3485], [-0.0728, 0.3397], [-0.1223, 0.4108], [-0.1847, 0.4796], [-0.2268, 0.4745], [-0.2206, 0.3817], [-0.173, 0.2517], [-0.1297, 0.1602], [-0.1458, 0.1538], [-0.2429, 0.2191], [-0.3867, 0.2975], [-0.5064, 0.33], [-0.5445, 0.2969], [-0.4988, 0.2233], [-0.4237, 0.1511], [-0.3866, 0.1046], [-0.4155, 0.0781], [-0.4801, 0.0517], [-0.5218, 0.015], [-0.5077, -0.0254], [-0.4622, -0.0598], [-0.4463, -0.0939], [-0.5024, -0.1478], [-0.6099, -0.233], [-0.6924, -0.3292], [-0.6744, -0.3885], [-0.5435, -0.3731]]}, {"id": 5, "shape_id": 5, "size": 5, "shape_vertices": [[-0.4166, 0.1309], [-0.442, 0.1099], [-0.4917, 0.0908], [-0.5607, 0.0682], [-0.6408, 0.0376], [-0.7223, -0.0038], [-0.7949, -0.0561], [-0.8495, -0.1176], [-0.8789, -0.1845], [-0.8793, -0.2516], [-0.8504, -0.3135], [-0.7953, -0.3655], [-0.7199, -0.4041], [-0.6321, -0.4284], [-0.5399, -0.4395], [-0.4512, -0.4411], [-0.3714, -0.4386], [-0.3037, -0.4385], [-0.2481, -0.4471], [-0.2019, -0.4697], [-0.16, -0.5094], [-0.1166, -0.5668], [-0.0655, -0.639], [-0.0021, -0.7207], [0.0757, -0.8042], [0.1674, -0.8803], [0.2691, -0.9399], [0.3742, -0.9751], [0.4741, -0.9802], [0.5597, -0.9525], [0.6225, -0.8929], [0.656, -0.8057], [0.6572, -0.6982], [0.6267, -0.5797], [0.5696, -0.4599], [0.4942, -0.3483], [0.4117, -0.2526], [0.3347, -0.1775], [0.2752, -0.1249], [0.2433, -0.093], [0.2457, -0.0772], [0.285, -0.0709], [0.3587, -0.0663], [0.4601, -0.056], [0.5789, -0.0339], [0.7024, 0.0037], [0.8172, 0.0577], [0.9109, 0.1262], [0.9739, 0.2044], [1.0, 0.2861], [0.9873, 0.364], [0.9384, 0.4312], [0.8593, 0.4824], [0.7591, 0.5145], [0.6478, 0.5273], [0.5354, 0.5235], [0.4303, 0.5081], [0.3381, 0.4881], [0.2614, 0.4709], [0.1992, 0.4635], [0.148, 0.4712], [0.1022, 0.4968], [0.0553, 0.5402], [0.0018, 0.5983], [-0.0626, 0.665], [-0.1394, 0.7329], [-0.2271, 0.7933], [-0.3217, 0.8382], [-0.4165, 0.8611], [-0.504, 0.8577], [-0.5765, 0.827], [-0.6277, 0.7709], [-0.6534, 0.6942], [-0.6528, 0.6038], [-0.6286, 0.5076], [-0.5865, 0.4134], [-0.5348, 0.3281], [-0.4831, 0.2563], [-0.4411, 0.2002], [-0.417, 0.1594]]}, {"id": 6, "shape_id": 6, "size": 5, "shape_vertices": [[0.4995, -0.1937], [0.4751, -0.1327], [0.4426, -0.0785], [0.4075, -0.0327], [0.3778, 0.0052], [0.3623, 0.0383], [0.3687, 0.0726], [0.4014, 0.1158], [0.4598, 0.1753], [0.5376, 0.2564], [0.6235, 0.3596], [0.7034, 0.4797], [0.7628, 0.6064], [0.7903, 0.7254], [0.7795, 0.8216], [0.731, 0.8826], [0.6517, 0.9018], [0.5529, 0.8803], [0.4484, 0.8273], [0.3505, 0.758], [0.268, 0.6912], [0.204, 0.6444], [0.1563, 0.6304], [0.1183, 0.6541], [0.0815, 0.7119], [0.0381, 0.7919], [-0.017, 0.8775], [-0.0842, 0.9502], [-0.1598, 0.9944], [-0.2367, 1.0], [-0.3067, 0.9643], [-0.3623, 0.8917], [-0.3991, 0.7922], [-0.4164, 0.6788], [-0.417, 0.5641], [-0.4063, 0.458], [-0.3906, 0.3664], [-0.3753, 0.2911], [-0.3639, 0.2302], [-0.3574, 0.1803], [-0.3549, 0.1376], [-0.3549, 0.0991], [-0.3563, 0.0632], [-0.36, 0.0289], [-0.3687, -0.0051], [-0.3868, -0.0409], [-0.4187, -0.0824], [-0.4673, -0.1348], [-0.5324, -0.203], [-0.6095, -0.2907], [-0.6902, -0.398], [-0.7632, -0.5206], [-0.8168, -0.6493], [-0.8407, -0.7717], [-0.8291, -0.8738], [-0.7814, -0.9434], [-0.7029, -0.9728], [-0.6032, -0.9604], [-0.4944, -0.9121], [-0.3883, -0.8397], [-0.2943, -0.759], [-0.2172, -0.6861], [-0.1573, -0.6344], [-0.1107, -0.6118], [-0.071, -0.6194], [-0.0313, -0.6517], [0.0135, -0.6984], [0.0662, -0.747], [0.1262, -0.7856], [0.1906, -0.8054], [0.255, -0.8019], [0.3151, -0.7756], [0.3678, -0.73], [0.4117, -0.6711], [0.4469, -0.6045], [0.4744, -0.5348], [0.4953, -0.4647], [0.5095, -0.3952], [0.516, -0.3265], [0.5132, -0.259]]}, {"id": 7, "shape_id": 7, "size": 5, "shape_vertices": [[0.6668, 0.0256], [0.7332, 0.0871], [0.779, 0.1564], [0.7936, 0.2264], [0.7801, 0.2914], [0.7488, 0.3499], [0.7089, 0.4026], [0.6625, 0.4491], [0.6056, 0.4845], [0.5334, 0.5005], [0.4467, 0.4906], [0.3545, 0.4566], [0.2711, 0.4121], [0.2094, 0.3798], [0.1741, 0.3835], [0.1588, 0.4373], [0.1496, 0.5388], [0.1311, 0.6692], [0.0934, 0.8004], [0.0353, 0.9066], [-0.0374, 0.9732], [-0.1161, 1.0], [-0.1948, 0.9975], [-0.271, 0.9786], [-0.3446, 0.9512], [-0.4144, 0.9153], [-0.4764, 0.8663], [-0.525, 0.8002], [-0.5564, 0.7189], [-0.5722, 0.6306], [-0.5796, 0.5459], [-0.5884, 0.4727], [-0.6049, 0.4121], [-0.6279, 0.3587], [-0.6489, 0.3053], [-0.6576, 0.2478], [-0.6496, 0.1874], [-0.632, 0.1289], [-0.6215, 0.0758], [-0.6364, 0.0264], [-0.6851, -0.0263], [-0.7575, -0.09], [-0.8256, -0.1657], [-0.8545, -0.2437], [-0.8183, -0.3057], [-0.7139, -0.3336], [-0.5659, -0.3214], [-0.4171, -0.2827], [-0.3107, -0.2486], [-0.2721, -0.2554], [-0.2977, -0.3269], [-0.358, -0.4612], [-0.4128, -0.6274], [-0.4291, -0.7783], [-0.3951, -0.8705], [-0.3216, -0.8853], [-0.2326, -0.8376], [-0.1507, -0.7691], [-0.0849, -0.7278], [-0.029, -0.7448], [0.0314, -0.8184], [0.1063, -0.9153], [0.1927, -0.9868], [0.2749, -0.9926], [0.3335, -0.9204], [0.3583, -0.7913], [0.3569, -0.649], [0.3537, -0.5391], [0.3776, -0.4878], [0.4469, -0.4925], [0.5577, -0.5253], [0.6832, -0.5489], [0.7848, -0.5346], [0.8306, -0.4745], [0.8099, -0.381], [0.7386, -0.2783], [0.6519, -0.188], [0.5878, -0.1198], [0.571, -0.0696], [0.6035, -0.0251]]}], "radius": 36, "fourier_perturbation_amplitude": 0.1, "min_pairwise_silhouette_distance": 0.11, "blob_shapes": [[[-0.55, 0.5349], [-0.6102, 0.4881], [-0.6644, 0.4337], [-0.7103, 0.3728], [-0.7461, 0.3066], [-0.7708, 0.2373], [-0.784, 0.167], [-0.7867, 0.0978], [-0.7803, 0.0314], [-0.7673, -0.0311], [-0.7506, -0.0894], [-0.7331, -0.1442], [-0.7175, -0.1966], [-0.7058, -0.2486], [-0.6992, -0.3022], [-0.6976, -0.3593], [-0.6999, -0.4215], [-0.7039, -0.4891], [-0.7068, -0.562], [-0.7053, -0.6384], [-0.6962, -0.7158], [-0.6768, -0.7909], [-0.6453, -0.8595], [-0.6007, -0.9179], [-0.5437, -0.9624], [-0.4758, -0.9903], [-0.3996, -1.0], [-0.3185, -0.9912], [-0.2359, -0.9652], [-0.1553, -0.9243], [-0.0794, -0.8721], [-0.0103, -0.8127], [0.0511, -0.7503], [0.1051, -0.6888], [0.153, -0.6315], [0.1966, -0.5807], [0.2384, -0.5372], [0.281, -0.501], [0.3263, -0.4707], [0.376, -0.4442], [0.4306, -0.4188], [0.49, -0.3919], [0.5527, -0.3608], [0.6169, -0.3238], [0.6801, -0.2795], [0.7395, -0.2277], [0.7927, -0.1689], [0.8374, -0.1041], [0.8721, -0.0351], [0.8961, 0.0363], [0.9094, 0.1084], [0.9125, 0.1795], [0.9065, 0.2484], [0.8929, 0.3145], [0.8729, 0.3773], [0.8479, 0.4368], [0.8187, 0.493], [0.7859, 0.5461], [0.7497, 0.596], [0.7099, 0.6425], [0.6662, 0.685], [0.6185, 0.7227], [0.5665, 0.7547], [0.5106, 0.7801], [0.451, 0.7984], [0.3887, 0.809], [0.3246, 0.8122], [0.2598, 0.8086], [0.1953, 0.799], [0.1319, 0.785], [0.07, 0.768], [0.0095, 0.7494], [-0.0498, 0.7304], [-0.1086, 0.7117], [-0.168, 0.6934], [-0.2285, 0.6749], [-0.2908, 0.6553], [-0.355, 0.633], [-0.4204, 0.6065], [-0.486, 0.5742]], [[0.6145, -0.1204], [0.6407, -0.0643], [0.6439, -0.0046], [0.6094, 0.0519], [0.5353, 0.095], [0.4345, 0.1177], [0.331, 0.1215], [0.251, 0.1172], [0.2131, 0.1219], [0.2208, 0.1511], [0.2619, 0.2112], [0.3133, 0.2948], [0.3503, 0.3825], [0.3559, 0.4502], [0.3268, 0.4798], [0.2729, 0.4676], [0.2121, 0.4281], [0.1612, 0.3894], [0.1294, 0.383], [0.1143, 0.4307], [0.105, 0.5357], [0.0871, 0.6796], [0.0508, 0.8285], [-0.0052, 0.9447], [-0.0731, 1.0], [-0.1402, 0.985], [-0.1951, 0.9113], [-0.2333, 0.8056], [-0.2585, 0.6986], [-0.2809, 0.6144], [-0.3111, 0.5625], [-0.3547, 0.5372], [-0.4092, 0.5227], [-0.4646, 0.5016], [-0.5079, 0.4626], [-0.5296, 0.4039], [-0.5276, 0.3327], [-0.5092, 0.2599], [-0.4872, 0.1945], [-0.4745, 0.14], [-0.4784, 0.0938], [-0.497, 0.0499], [-0.5202, 0.0037], [-0.5343, -0.0455], [-0.5282, -0.0937], [-0.4983, -0.135], [-0.4501, -0.1652], [-0.3959, -0.1848], [-0.35, -0.2002], [-0.3228, -0.2209], [-0.3169, -0.2556], [-0.3275, -0.3081], [-0.3441, -0.3757], [-0.3556, -0.4499], [-0.3542, -0.5201], [-0.3372, -0.5777], [-0.3066, -0.6188], [-0.2668, -0.6445], [-0.2224, -0.6583], [-0.1761, -0.6635], [-0.1294, -0.6604], [-0.0829, -0.6469], [-0.0381, -0.6206], [0.0032, -0.5827], [0.0394, -0.5395], [0.0715, -0.5028], [0.104, -0.4856], [0.1439, -0.4971], [0.1988, -0.5372], [0.2722, -0.5954], [0.361, -0.6529], [0.4547, -0.6887], [0.5382, -0.6875], [0.597, -0.6447], [0.6234, -0.5677], [0.6198, -0.4727], [0.5986, -0.3774], [0.5773, -0.2946], [0.571, -0.228], [0.5856, -0.1728]], [[-0.6257, 0.4803], [-0.7177, 0.4251], [-0.7386, 0.327], [-0.7001, 0.2184], [-0.6424, 0.1256], [-0.6081, 0.0546], [-0.6175, -0.0048], [-0.6626, -0.0657], [-0.7197, -0.1338], [-0.7701, -0.2075], [-0.8114, -0.2848], [-0.8525, -0.3677], [-0.8979, -0.4595], [-0.9358, -0.5551], [-0.9407, -0.6365], [-0.8886, -0.6782], [-0.7752, -0.6623], [-0.6231, -0.593], [-0.4736, -0.5005], [-0.3668, -0.4302], [-0.3236, -0.4215], [-0.3379, -0.4904], [-0.3823, -0.6213], [-0.4216, -0.7733], [-0.4276, -0.8954], [-0.3887, -0.9444], [-0.3114, -0.8992], [-0.216, -0.7688], [-0.1269, -0.5907], [-0.0617, -0.4203], [-0.0237, -0.3114], [-0.0005, -0.2945], [0.0283, -0.3641], [0.079, -0.4814], [0.154, -0.5941], [0.2424, -0.6621], [0.3302, -0.6758], [0.4133, -0.6536], [0.5024, -0.6235], [0.6137, -0.6006], [0.7505, -0.5761], [0.89, -0.5272], [0.9875, -0.4372], [1.0, -0.312], [0.9151, -0.1789], [0.7656, -0.0687], [0.6177, 0.0048], [0.5366, 0.0532], [0.5514, 0.1025], [0.6401, 0.1725], [0.7448, 0.2614], [0.8065, 0.3479], [0.7967, 0.4077], [0.7268, 0.4311], [0.6335, 0.4286], [0.5527, 0.4218], [0.5005, 0.4276], [0.4735, 0.4505], [0.4608, 0.487], [0.4566, 0.5355], [0.4612, 0.6008], [0.4727, 0.686], [0.4793, 0.7789], [0.4618, 0.8472], [0.4071, 0.8524], [0.32, 0.7776], [0.2247, 0.6489], [0.1492, 0.531], [0.1063, 0.495], [0.0844, 0.5745], [0.0564, 0.7395], [0.0016, 0.9085], [-0.0772, 0.9931], [-0.1555, 0.9481], [-0.2064, 0.7961], [-0.2238, 0.6113], [-0.2312, 0.473], [-0.2666, 0.4216], [-0.3554, 0.4411], [-0.4887, 0.4783]], [[0.5811, 0.7744], [0.4775, 0.7241], [0.4151, 0.7216], [0.3676, 0.7411], [0.3079, 0.7324], [0.2288, 0.6586], [0.145, 0.5257], [0.0786, 0.3837], [0.0394, 0.2953], [0.0181, 0.2957], [-0.005, 0.3686], [-0.0417, 0.457], [-0.0869, 0.5011], [-0.1247, 0.4767], [-0.1461, 0.4083], [-0.1623, 0.3499], [-0.2021, 0.346], [-0.2909, 0.4027], [-0.4295, 0.4851], [-0.5869, 0.5417], [-0.7155, 0.5369], [-0.7782, 0.4686], [-0.769, 0.3632], [-0.7127, 0.2538], [-0.6452, 0.1614], [-0.5891, 0.0897], [-0.5448, 0.0328], [-0.5011, -0.0138], [-0.4555, -0.0511], [-0.4263, -0.083], [-0.4449, -0.1231], [-0.531, -0.1905], [-0.6723, -0.2971], [-0.8218, -0.4335], [-0.9212, -0.568], [-0.9331, -0.663], [-0.8621, -0.6992], [-0.7505, -0.6907], [-0.6512, -0.6779], [-0.5964, -0.7018], [-0.5817, -0.7753], [-0.5745, -0.871], [-0.5383, -0.9357], [-0.457, -0.9213], [-0.3421, -0.8137], [-0.2229, -0.6416], [-0.1271, -0.461], [-0.0668, -0.3264], [-0.0356, -0.2667], [-0.017, -0.2785], [0.0045, -0.3358], [0.0373, -0.4084], [0.0825, -0.4753], [0.1382, -0.5281], [0.2023, -0.5655], [0.2724, -0.587], [0.3441, -0.5893], [0.4103, -0.5682], [0.4624, -0.5222], [0.4936, -0.4555], [0.5029, -0.3773], [0.4965, -0.299], [0.4871, -0.2301], [0.49, -0.1745], [0.5166, -0.1292], [0.5675, -0.0864], [0.6287, -0.0378], [0.6752, 0.0186], [0.6834, 0.0766], [0.6458, 0.1258], [0.5818, 0.1609], [0.5324, 0.191], [0.5411, 0.2391], [0.6273, 0.3309], [0.7705, 0.475], [0.9156, 0.6506], [1.0, 0.8111], [0.9858, 0.9073], [0.8798, 0.9159], [0.7265, 0.855]], [[-0.3664, -0.2965], [-0.244, -0.2316], [-0.2394, -0.2661], [-0.3334, -0.4353], [-0.4417, -0.6818], [-0.4774, -0.8822], [-0.4104, -0.926], [-0.2796, -0.7941], [-0.1533, -0.5765], [-0.0758, -0.4136], [-0.0415, -0.403], [-0.013, -0.5371], [0.0389, -0.7154], [0.1092, -0.8182], [0.1686, -0.7868], [0.1945, -0.6532], [0.1947, -0.5054], [0.2006, -0.4196], [0.2378, -0.4114], [0.3002, -0.4366], [0.3512, -0.434], [0.3521, -0.3715], [0.2937, -0.265], [0.2078, -0.1599], [0.1478, -0.0963], [0.1552, -0.0846], [0.2349, -0.1052], [0.3547, -0.1265], [0.4662, -0.1261], [0.5305, -0.0997], [0.5342, -0.0575], [0.4903, -0.0141], [0.4276, 0.0214], [0.3789, 0.049], [0.3696, 0.0778], [0.4108, 0.1209], [0.4947, 0.189], [0.595, 0.2829], [0.6749, 0.3888], [0.7007, 0.481], [0.6574, 0.5321], [0.5567, 0.5284], [0.4334, 0.4818], [0.3296, 0.4303], [0.2744, 0.4236], [0.2694, 0.4979], [0.2887, 0.6513], [0.2943, 0.8361], [0.2593, 0.9754], [0.1832, 1.0], [0.0915, 0.8882], [0.0165, 0.6833], [-0.0258, 0.4744], [-0.0465, 0.3485], [-0.0728, 0.3397], [-0.1223, 0.4108], [-0.1847, 0.4796], [-0.2268, 0.4745], [-0.2206, 0.3817], [-0.173, 0.2517], [-0.1297, 0.1602], [-0.1458, 0.1538], [-0.2429, 0.2191], [-0.3867, 0.2975], [-0.5064, 0.33], [-0.5445, 0.2969], [-0.4988, 0.2233], [-0.4237, 0.1511], [-0.3866, 0.1046], [-0.4155, 0.0781], [-0.4801, 0.0517], [-0.5218, 0.015], [-0.5077, -0.0254], [-0.4622, -0.0598], [-0.4463, -0.0939], [-0.5024, -0.1478], [-0.6099, -0.233], [-0.6924, -0.3292], [-0.6744, -0.3885], [-0.5435, -0.3731]], [[-0.4166, 0.1309], [-0.442, 0.1099], [-0.4917, 0.0908], [-0.5607, 0.0682], [-0.6408, 0.0376], [-0.7223, -0.0038], [-0.7949, -0.0561], [-0.8495, -0.1176], [-0.8789, -0.1845], [-0.8793, -0.2516], [-0.8504, -0.3135], [-0.7953, -0.3655], [-0.7199, -0.4041], [-0.6321, -0.4284], [-0.5399, -0.4395], [-0.4512, -0.4411], [-0.3714, -0.4386], [-0.3037, -0.4385], [-0.2481, -0.4471], [-0.2019, -0.4697], [-0.16, -0.5094], [-0.1166, -0.5668], [-0.0655, -0.639], [-0.0021, -0.7207], [0.0757, -0.8042], [0.1674, -0.8803], [0.2691, -0.9399], [0.3742, -0.9751], [0.4741, -0.9802], [0.5597, -0.9525], [0.6225, -0.8929], [0.656, -0.8057], [0.6572, -0.6982], [0.6267, -0.5797], [0.5696, -0.4599], [0.4942, -0.3483], [0.4117, -0.2526], [0.3347, -0.1775], [0.2752, -0.1249], [0.2433, -0.093], [0.2457, -0.0772], [0.285, -0.0709], [0.3587, -0.0663], [0.4601, -0.056], [0.5789, -0.0339], [0.7024, 0.0037], [0.8172, 0.0577], [0.9109, 0.1262], [0.9739, 0.2044], [1.0, 0.2861], [0.9873, 0.364], [0.9384, 0.4312], [0.8593, 0.4824], [0.7591, 0.5145], [0.6478, 0.5273], [0.5354, 0.5235], [0.4303, 0.5081], [0.3381, 0.4881], [0.2614, 0.4709], [0.1992, 0.4635], [0.148, 0.4712], [0.1022, 0.4968], [0.0553, 0.5402], [0.0018, 0.5983], [-0.0626, 0.665], [-0.1394, 0.7329], [-0.2271, 0.7933], [-0.3217, 0.8382], [-0.4165, 0.8611], [-0.504, 0.8577], [-0.5765, 0.827], [-0.6277, 0.7709], [-0.6534, 0.6942], [-0.6528, 0.6038], [-0.6286, 0.5076], [-0.5865, 0.4134], [-0.5348, 0.3281], [-0.4831, 0.2563], [-0.4411, 0.2002], [-0.417, 0.1594]], [[0.4995, -0.1937], [0.4751, -0.1327], [0.4426, -0.0785], [0.4075, -0.0327], [0.3778, 0.0052], [0.3623, 0.0383], [0.3687, 0.0726], [0.4014, 0.1158], [0.4598, 0.1753], [0.5376, 0.2564], [0.6235, 0.3596], [0.7034, 0.4797], [0.7628, 0.6064], [0.7903, 0.7254], [0.7795, 0.8216], [0.731, 0.8826], [0.6517, 0.9018], [0.5529, 0.8803], [0.4484, 0.8273], [0.3505, 0.758], [0.268, 0.6912], [0.204, 0.6444], [0.1563, 0.6304], [0.1183, 0.6541], [0.0815, 0.7119], [0.0381, 0.7919], [-0.017, 0.8775], [-0.0842, 0.9502], [-0.1598, 0.9944], [-0.2367, 1.0], [-0.3067, 0.9643], [-0.3623, 0.8917], [-0.3991, 0.7922], [-0.4164, 0.6788], [-0.417, 0.5641], [-0.4063, 0.458], [-0.3906, 0.3664], [-0.3753, 0.2911], [-0.3639, 0.2302], [-0.3574, 0.1803], [-0.3549, 0.1376], [-0.3549, 0.0991], [-0.3563, 0.0632], [-0.36, 0.0289], [-0.3687, -0.0051], [-0.3868, -0.0409], [-0.4187, -0.0824], [-0.4673, -0.1348], [-0.5324, -0.203], [-0.6095, -0.2907], [-0.6902, -0.398], [-0.7632, -0.5206], [-0.8168, -0.6493], [-0.8407, -0.7717], [-0.8291, -0.8738], [-0.7814, -0.9434], [-0.7029, -0.9728], [-0.6032, -0.9604], [-0.4944, -0.9121], [-0.3883, -0.8397], [-0.2943, -0.759], [-0.2172, -0.6861], [-0.1573, -0.6344], [-0.1107, -0.6118], [-0.071, -0.6194], [-0.0313, -0.6517], [0.0135, -0.6984], [0.0662, -0.747], [0.1262, -0.7856], [0.1906, -0.8054], [0.255, -0.8019], [0.3151, -0.7756], [0.3678, -0.73], [0.4117, -0.6711], [0.4469, -0.6045], [0.4744, -0.5348], [0.4953, -0.4647], [0.5095, -0.3952], [0.516, -0.3265], [0.5132, -0.259]], [[0.6668, 0.0256], [0.7332, 0.0871], [0.779, 0.1564], [0.7936, 0.2264], [0.7801, 0.2914], [0.7488, 0.3499], [0.7089, 0.4026], [0.6625, 0.4491], [0.6056, 0.4845], [0.5334, 0.5005], [0.4467, 0.4906], [0.3545, 0.4566], [0.2711, 0.4121], [0.2094, 0.3798], [0.1741, 0.3835], [0.1588, 0.4373], [0.1496, 0.5388], [0.1311, 0.6692], [0.0934, 0.8004], [0.0353, 0.9066], [-0.0374, 0.9732], [-0.1161, 1.0], [-0.1948, 0.9975], [-0.271, 0.9786], [-0.3446, 0.9512], [-0.4144, 0.9153], [-0.4764, 0.8663], [-0.525, 0.8002], [-0.5564, 0.7189], [-0.5722, 0.6306], [-0.5796, 0.5459], [-0.5884, 0.4727], [-0.6049, 0.4121], [-0.6279, 0.3587], [-0.6489, 0.3053], [-0.6576, 0.2478], [-0.6496, 0.1874], [-0.632, 0.1289], [-0.6215, 0.0758], [-0.6364, 0.0264], [-0.6851, -0.0263], [-0.7575, -0.09], [-0.8256, -0.1657], [-0.8545, -0.2437], [-0.8183, -0.3057], [-0.7139, -0.3336], [-0.5659, -0.3214], [-0.4171, -0.2827], [-0.3107, -0.2486], [-0.2721, -0.2554], [-0.2977, -0.3269], [-0.358, -0.4612], [-0.4128, -0.6274], [-0.4291, -0.7783], [-0.3951, -0.8705], [-0.3216, -0.8853], [-0.2326, -0.8376], [-0.1507, -0.7691], [-0.0849, -0.7278], [-0.029, -0.7448], [0.0314, -0.8184], [0.1063, -0.9153], [0.1927, -0.9868], [0.2749, -0.9926], [0.3335, -0.9204], [0.3583, -0.7913], [0.3569, -0.649], [0.3537, -0.5391], [0.3776, -0.4878], [0.4469, -0.4925], [0.5577, -0.5253], [0.6832, -0.5489], [0.7848, -0.5346], [0.8306, -0.4745], [0.8099, -0.381], [0.7386, -0.2783], [0.6519, -0.188], [0.5878, -0.1198], [0.571, -0.0696], [0.6035, -0.0251]]], "image": "images/attribute_group_counting_00003.png", "question": "How many distinct shapes are in the image? The image contains many small pieces scattered across the canvas. Two pieces belong to the same type if and only if they have exactly the same silhouette. Pieces may differ in surface texture or colour tone \u2014 only the silhouette outline matters. Count the total number of distinct silhouette types and report the count as a positive integer. Provide your final answer enclosed in ... tags."} +{"width": 1273, "height": 1273, "num_groups": 10, "num_elements": 34, "answer": 10, "elements": [{"shape_id": 0, "group": 0, "x": 1152.41, "y": 1127.26}, {"shape_id": 0, "group": 0, "x": 639.13, "y": 871.75}, {"shape_id": 0, "group": 0, "x": 981.9, "y": 861.59}, {"shape_id": 0, "group": 0, "x": 865.13, "y": 715.91}, {"shape_id": 0, "group": 0, "x": 1026.57, "y": 137.85}, {"shape_id": 1, "group": 1, "x": 122.81, "y": 221.57}, {"shape_id": 1, "group": 1, "x": 453.23, "y": 158.15}, {"shape_id": 1, "group": 1, "x": 1204.04, "y": 1199.9}, {"shape_id": 1, "group": 1, "x": 1018.05, "y": 1146.43}, {"shape_id": 2, "group": 2, "x": 490.45, "y": 607.97}, {"shape_id": 2, "group": 2, "x": 255.15, "y": 1133.56}, {"shape_id": 2, "group": 2, "x": 61.89, "y": 900.69}, {"shape_id": 2, "group": 2, "x": 715.8, "y": 1051.99}, {"shape_id": 2, "group": 2, "x": 825.89, "y": 60.71}, {"shape_id": 3, "group": 3, "x": 761.55, "y": 771.89}, {"shape_id": 3, "group": 3, "x": 750.18, "y": 1194.05}, {"shape_id": 3, "group": 3, "x": 397.84, "y": 712.55}, {"shape_id": 3, "group": 3, "x": 553.95, "y": 1159.32}, {"shape_id": 3, "group": 3, "x": 113.88, "y": 800.26}, {"shape_id": 4, "group": 4, "x": 364.33, "y": 1140.83}, {"shape_id": 4, "group": 4, "x": 79.67, "y": 523.88}, {"shape_id": 4, "group": 4, "x": 140.31, "y": 1047.1}, {"shape_id": 4, "group": 4, "x": 674.8, "y": 531.08}, {"shape_id": 5, "group": 5, "x": 135.95, "y": 1182.61}, {"shape_id": 6, "group": 6, "x": 463.43, "y": 987.18}, {"shape_id": 7, "group": 7, "x": 318.02, "y": 132.39}, {"shape_id": 7, "group": 7, "x": 778.92, "y": 342.13}, {"shape_id": 8, "group": 8, "x": 678.99, "y": 382.87}, {"shape_id": 8, "group": 8, "x": 871.01, "y": 398.67}, {"shape_id": 9, "group": 9, "x": 587.75, "y": 331.17}, {"shape_id": 9, "group": 9, "x": 60.62, "y": 288.31}, {"shape_id": 9, "group": 9, "x": 1221.1, "y": 814.72}, {"shape_id": 9, "group": 9, "x": 302.06, "y": 584.2}, {"shape_id": 9, "group": 9, "x": 799.09, "y": 877.68}], "groups": [{"id": 0, "shape_id": 0, "size": 5, "shape_vertices": [[0.232, 0.6081], [0.1641, 0.5979], [0.103, 0.5933], [0.0462, 0.5979], [-0.0098, 0.6138], [-0.0688, 0.6407], [-0.134, 0.6761], [-0.2069, 0.7151], [-0.2873, 0.7516], [-0.372, 0.7785], [-0.4563, 0.7897], [-0.5336, 0.7805], [-0.597, 0.7489], [-0.6408, 0.6957], [-0.6609, 0.6244], [-0.6567, 0.541], [-0.6308, 0.4528], [-0.5891, 0.367], [-0.5403, 0.2899], [-0.4942, 0.2256], [-0.4606, 0.1757], [-0.4476, 0.1387], [-0.4602, 0.1108], [-0.4996, 0.0866], [-0.563, 0.06], [-0.644, 0.0256], [-0.7334, -0.0207], [-0.8208, -0.0804], [-0.8963, -0.1529], [-0.9513, -0.2352], [-0.9802, -0.3227], [-0.9805, -0.4102], [-0.953, -0.4924], [-0.9012, -0.5655], [-0.8305, -0.6266], [-0.7469, -0.6752], [-0.6562, -0.7119], [-0.5634, -0.7386], [-0.4717, -0.7577], [-0.383, -0.7711], [-0.2976, -0.7804], [-0.2156, -0.7856], [-0.1365, -0.7861], [-0.0603, -0.7804], [0.0122, -0.7667], [0.0799, -0.7437], [0.1408, -0.7108], [0.1935, -0.6687], [0.2367, -0.6194], [0.2705, -0.566], [0.2959, -0.5122], [0.3157, -0.4618], [0.3332, -0.4179], [0.3525, -0.3827], [0.3775, -0.3567], [0.4113, -0.3388], [0.4554, -0.3269], [0.5099, -0.3176], [0.5732, -0.3075], [0.6424, -0.2933], [0.7135, -0.2721], [0.7824, -0.2425], [0.8454, -0.2036], [0.8994, -0.156], [0.9425, -0.1005], [0.9737, -0.0386], [0.9928, 0.028], [1.0, 0.098], [0.9957, 0.1699], [0.9802, 0.2424], [0.9534, 0.3139], [0.9151, 0.3828], [0.8652, 0.447], [0.8038, 0.5043], [0.7319, 0.5523], [0.6515, 0.589], [0.5654, 0.6133], [0.4769, 0.6253], [0.3899, 0.6262], [0.3075, 0.6191]]}, {"id": 1, "shape_id": 1, "size": 4, "shape_vertices": [[0.063, -0.7631], [0.1078, -0.7534], [0.1508, -0.736], [0.1905, -0.7082], [0.2255, -0.6707], [0.2556, -0.627], [0.2823, -0.5821], [0.3081, -0.5408], [0.3362, -0.506], [0.3689, -0.4776], [0.4066, -0.4527], [0.4476, -0.4269], [0.4878, -0.3953], [0.5218, -0.3544], [0.5445, -0.3036], [0.5528, -0.2449], [0.5467, -0.1823], [0.5299, -0.1206], [0.5091, -0.0631], [0.4927, -0.0103], [0.4885, 0.0403], [0.5014, 0.0942], [0.5322, 0.1577], [0.5767, 0.2361], [0.6263, 0.3315], [0.6702, 0.4409], [0.6974, 0.5565], [0.6994, 0.6668], [0.6725, 0.7591], [0.6179, 0.8226], [0.542, 0.8512], [0.4541, 0.8454], [0.3645, 0.8127], [0.2821, 0.7656], [0.2121, 0.7195], [0.1556, 0.6886], [0.1097, 0.6823], [0.0691, 0.7038], [0.0279, 0.7484], [-0.0183, 0.8055], [-0.0711, 0.861], [-0.1289, 0.9011], [-0.1875, 0.915], [-0.2415, 0.898], [-0.2863, 0.8517], [-0.3194, 0.7835], [-0.3415, 0.7041], [-0.3558, 0.6246], [-0.3677, 0.5534], [-0.3824, 0.495], [-0.4033, 0.4491], [-0.4311, 0.4112], [-0.463, 0.3752], [-0.4936, 0.3353], [-0.5163, 0.2879], [-0.5258, 0.2329], [-0.5194, 0.1733], [-0.4988, 0.1136], [-0.4695, 0.0582], [-0.4401, 0.0092], [-0.4201, -0.0347], [-0.4174, -0.0784], [-0.4362, -0.1293], [-0.4758, -0.1948], [-0.5299, -0.2804], [-0.5885, -0.3871], [-0.6398, -0.5105], [-0.6727, -0.6413], [-0.6792, -0.7667], [-0.6562, -0.8736], [-0.6056, -0.9511], [-0.5334, -0.9932], [-0.4486, -1.0], [-0.3602, -0.9775], [-0.2759, -0.9359], [-0.2005, -0.8869], [-0.1353, -0.8413], [-0.0791, -0.806], [-0.0292, -0.7833], [0.0175, -0.7709]]}, {"id": 2, "shape_id": 2, "size": 5, "shape_vertices": [[0.9365, 0.0309], [0.9948, 0.1022], [1.0, 0.1737], [0.9543, 0.2356], [0.8707, 0.2814], [0.7687, 0.3104], [0.6687, 0.328], [0.5863, 0.3429], [0.5288, 0.3646], [0.4941, 0.3991], [0.4734, 0.4469], [0.4544, 0.5022], [0.4266, 0.5551], [0.384, 0.5946], [0.3268, 0.6125], [0.26, 0.6057], [0.191, 0.5776], [0.1269, 0.5363], [0.0715, 0.4919], [0.0253, 0.4534], [-0.0141, 0.4259], [-0.0502, 0.4099], [-0.0858, 0.4017], [-0.1218, 0.396], [-0.1576, 0.3885], [-0.1924, 0.3776], [-0.2269, 0.365], [-0.264, 0.355], [-0.3092, 0.3517], [-0.3686, 0.357], [-0.4466, 0.3691], [-0.5427, 0.3821], [-0.6506, 0.3879], [-0.7581, 0.3787], [-0.85, 0.3494], [-0.9113, 0.2998], [-0.9319, 0.2345], [-0.9093, 0.1617], [-0.8495, 0.0904], [-0.7656, 0.0279], [-0.6739, -0.0222], [-0.5899, -0.0606], [-0.5241, -0.0911], [-0.4806, -0.1186], [-0.4567, -0.1476], [-0.4458, -0.18], [-0.4401, -0.2158], [-0.4335, -0.2535], [-0.4238, -0.2922], [-0.4118, -0.3326], [-0.4001, -0.3777], [-0.3904, -0.4315], [-0.3822, -0.4973], [-0.3713, -0.575], [-0.3519, -0.6596], [-0.3182, -0.7413], [-0.2671, -0.8075], [-0.2, -0.8454], [-0.1231, -0.8466], [-0.0453, -0.8095], [0.0244, -0.7404], [0.0799, -0.6522], [0.1198, -0.5608], [0.1478, -0.4808], [0.1711, -0.4217], [0.1965, -0.3855], [0.2283, -0.3673], [0.2657, -0.3573], [0.3035, -0.3452], [0.334, -0.3235], [0.3511, -0.2901], [0.3531, -0.2486], [0.3455, -0.206], [0.3399, -0.1698], [0.351, -0.1443], [0.3918, -0.1289], [0.4688, -0.118], [0.5786, -0.1029], [0.7075, -0.0753], [0.8345, -0.0304]]}, {"id": 3, "shape_id": 3, "size": 5, "shape_vertices": [[-0.5289, 0.7133], [-0.5946, 0.668], [-0.6303, 0.5926], [-0.6265, 0.4926], [-0.5837, 0.3815], [-0.5148, 0.2763], [-0.4416, 0.1907], [-0.3893, 0.1308], [-0.3787, 0.0936], [-0.4199, 0.0687], [-0.5093, 0.0426], [-0.6311, 0.004], [-0.762, -0.053], [-0.8779, -0.127], [-0.96, -0.2114], [-0.999, -0.2966], [-0.9951, -0.3742], [-0.9563, -0.4385], [-0.8946, -0.4883], [-0.8223, -0.5258], [-0.749, -0.5554], [-0.6805, -0.5815], [-0.6188, -0.6078], [-0.5632, -0.6361], [-0.5112, -0.6663], [-0.4596, -0.6964], [-0.4056, -0.7226], [-0.3471, -0.7405], [-0.284, -0.7453], [-0.2177, -0.7336], [-0.1512, -0.705], [-0.0884, -0.6628], [-0.0322, -0.6145], [0.0169, -0.5702], [0.0613, -0.5404], [0.1067, -0.5323], [0.1599, -0.5476], [0.2266, -0.5814], [0.3086, -0.6231], [0.4033, -0.6595], [0.5031, -0.6784], [0.5981, -0.672], [0.6793, -0.6387], [0.7409, -0.5826], [0.7823, -0.5113], [0.8071, -0.4333], [0.8216, -0.3548], [0.8311, -0.2793], [0.8384, -0.2073], [0.8429, -0.1378], [0.8416, -0.0703], [0.8311, -0.0052], [0.8102, 0.0563], [0.781, 0.113], [0.7476, 0.1646], [0.7149, 0.2123], [0.6857, 0.2578], [0.6595, 0.3024], [0.6326, 0.3453], [0.6001, 0.3837], [0.5587, 0.4143], [0.5089, 0.4349], [0.4553, 0.4473], [0.4056, 0.4581], [0.3666, 0.4778], [0.3412, 0.517], [0.3268, 0.5823], [0.3153, 0.6726], [0.2962, 0.7774], [0.2609, 0.8792], [0.2056, 0.9584], [0.1334, 1.0], [0.0523, 0.9982], [-0.0283, 0.9588], [-0.1019, 0.8975], [-0.1672, 0.8335], [-0.2287, 0.783], [-0.2936, 0.7535], [-0.3671, 0.7411], [-0.4483, 0.7332]]}, {"id": 4, "shape_id": 4, "size": 4, "shape_vertices": [[-0.5976, -0.2911], [-0.621, -0.3633], [-0.6346, -0.4394], [-0.6156, -0.4998], [-0.5548, -0.5257], [-0.4649, -0.5136], [-0.3728, -0.4816], [-0.302, -0.4595], [-0.2589, -0.4695], [-0.2322, -0.5117], [-0.2046, -0.5647], [-0.1659, -0.6018], [-0.1172, -0.6096], [-0.0661, -0.5963], [-0.0179, -0.5835], [0.0292, -0.5887], [0.08, -0.6127], [0.1368, -0.6402], [0.196, -0.6538], [0.2528, -0.6474], [0.307, -0.6302], [0.3654, -0.6176], [0.4364, -0.6175], [0.5206, -0.6231], [0.6062, -0.6169], [0.6738, -0.5836], [0.7078, -0.52], [0.7061, -0.4366], [0.6807, -0.3496], [0.6495, -0.271], [0.6248, -0.2047], [0.6103, -0.1483], [0.6048, -0.0979], [0.6097, -0.0506], [0.6305, -0.0035], [0.6687, 0.0478], [0.7128, 0.1063], [0.7355, 0.168], [0.7061, 0.2191], [0.6107, 0.2419], [0.466, 0.227], [0.3158, 0.1848], [0.2086, 0.1444], [0.1706, 0.1385], [0.1925, 0.1824], [0.2376, 0.2625], [0.2672, 0.3452], [0.2631, 0.4003], [0.2341, 0.4246], [0.2021, 0.4454], [0.1819, 0.5019], [0.1694, 0.6146], [0.1473, 0.7667], [0.1009, 0.9102], [0.0305, 0.995], [-0.0495, 1.0], [-0.1233, 0.9444], [-0.1862, 0.8716], [-0.2453, 0.8182], [-0.3088, 0.7908], [-0.3734, 0.7666], [-0.4236, 0.7159], [-0.4443, 0.6287], [-0.4379, 0.5242], [-0.4291, 0.4367], [-0.4509, 0.3905], [-0.5195, 0.3817], [-0.6168, 0.3814], [-0.696, 0.3574], [-0.7093, 0.296], [-0.6397, 0.2096], [-0.5153, 0.1252], [-0.3938, 0.0637], [-0.3293, 0.0273], [-0.3413, 0.0019], [-0.4084, -0.0292], [-0.4874, -0.0727], [-0.5437, -0.1242], [-0.5698, -0.1768], [-0.5814, -0.2303]]}, {"id": 5, "shape_id": 5, "size": 1, "shape_vertices": [[-0.6153, 0.1302], [-0.6012, 0.0547], [-0.5678, -0.0144], [-0.5395, -0.0749], [-0.5329, -0.1339], [-0.5474, -0.199], [-0.5663, -0.2704], [-0.5672, -0.3371], [-0.5355, -0.3833], [-0.4724, -0.3986], [-0.3933, -0.3857], [-0.3194, -0.3605], [-0.2675, -0.3458], [-0.2446, -0.3612], [-0.2467, -0.4162], [-0.262, -0.5067], [-0.2751, -0.6139], [-0.2723, -0.7089], [-0.2467, -0.7611], [-0.2009, -0.7528], [-0.1463, -0.6914], [-0.0963, -0.6114], [-0.0581, -0.5588], [-0.0284, -0.5657], [0.0027, -0.6277], [0.0424, -0.7037], [0.0877, -0.7392], [0.1262, -0.6998], [0.1461, -0.5934], [0.1473, -0.4633], [0.1432, -0.3604], [0.1517, -0.3117], [0.1824, -0.3095], [0.2309, -0.3251], [0.2857, -0.3335], [0.3416, -0.3285], [0.4075, -0.3183], [0.4983, -0.309], [0.6178, -0.2927], [0.7447, -0.252], [0.8372, -0.1771], [0.8561, -0.0778], [0.7905, 0.0201], [0.6686, 0.0929], [0.543, 0.1364], [0.46, 0.1672], [0.4345, 0.2075], [0.4444, 0.2641], [0.449, 0.3214], [0.4164, 0.3513], [0.3409, 0.3343], [0.2432, 0.2745], [0.1546, 0.1998], [0.0997, 0.1472], [0.0868, 0.1465], [0.1087, 0.2103], [0.1496, 0.3339], [0.1918, 0.4993], [0.2202, 0.6796], [0.2247, 0.842], [0.2021, 0.9554], [0.1575, 1.0], [0.1016, 0.9772], [0.0458, 0.9103], [-0.0035, 0.8341], [-0.0468, 0.7765], [-0.0884, 0.7451], [-0.1308, 0.7257], [-0.1708, 0.6937], [-0.2006, 0.6308], [-0.2131, 0.5363], [-0.208, 0.4274], [-0.1941, 0.3293], [-0.187, 0.2632], [-0.2029, 0.2369], [-0.2521, 0.2423], [-0.333, 0.2601], [-0.4319, 0.2678], [-0.5257, 0.249], [-0.5913, 0.2001]]}, {"id": 6, "shape_id": 6, "size": 1, "shape_vertices": [[-0.7855, -0.5247], [-0.7902, -0.6335], [-0.7715, -0.7369], [-0.7229, -0.8206], [-0.6444, -0.8715], [-0.5432, -0.8819], [-0.4315, -0.8528], [-0.3229, -0.7948], [-0.2282, -0.7258], [-0.1523, -0.6666], [-0.0933, -0.6352], [-0.0439, -0.6417], [0.0055, -0.6856], [0.0631, -0.7562], [0.1326, -0.8354], [0.2123, -0.9034], [0.2953, -0.9433], [0.3727, -0.946], [0.4364, -0.9109], [0.4815, -0.845], [0.5078, -0.7603], [0.5191, -0.6691], [0.5212, -0.5813], [0.5198, -0.5023], [0.5185, -0.4335], [0.5186, -0.3732], [0.52, -0.3191], [0.5223, -0.2695], [0.5274, -0.2236], [0.5393, -0.1815], [0.5638, -0.1423], [0.6063, -0.1031], [0.669, -0.0593], [0.7486, -0.0053], [0.836, 0.0631], [0.917, 0.1471], [0.9761, 0.2427], [1.0, 0.3416], [0.9818, 0.433], [0.9227, 0.5065], [0.832, 0.5558], [0.7243, 0.5807], [0.6154, 0.5878], [0.5182, 0.5882], [0.4395, 0.5944], [0.3791, 0.6156], [0.3315, 0.6552], [0.2883, 0.7096], [0.2418, 0.769], [0.1877, 0.8216], [0.1259, 0.8571], [0.0596, 0.8704], [-0.0069, 0.8633], [-0.0704, 0.8433], [-0.1304, 0.8211], [-0.1894, 0.8058], [-0.2511, 0.8023], [-0.3185, 0.8085], [-0.3912, 0.8166], [-0.4647, 0.8156], [-0.5312, 0.7952], [-0.5817, 0.7496], [-0.6094, 0.6796], [-0.6126, 0.592], [-0.5955, 0.4979], [-0.568, 0.4088], [-0.5425, 0.333], [-0.5304, 0.2736], [-0.5385, 0.2284], [-0.567, 0.1908], [-0.6095, 0.1538], [-0.6561, 0.1115], [-0.6962, 0.0617], [-0.7231, 0.0051], [-0.7351, -0.0555], [-0.7367, -0.1181], [-0.7353, -0.1828], [-0.7385, -0.2523], [-0.7502, -0.3309], [-0.7685, -0.4218]]}, {"id": 7, "shape_id": 7, "size": 2, "shape_vertices": [[0.5211, -0.4433], [0.5742, -0.4081], [0.6763, -0.3973], [0.7969, -0.3799], [0.8893, -0.334], [0.9209, -0.2594], [0.8938, -0.1727], [0.8439, -0.0918], [0.8159, -0.022], [0.8315, 0.0442], [0.873, 0.1157], [0.893, 0.1897], [0.8446, 0.248], [0.714, 0.2694], [0.5325, 0.2474], [0.3617, 0.2015], [0.2594, 0.1702], [0.2493, 0.1905], [0.3099, 0.2739], [0.3904, 0.3979], [0.4399, 0.517], [0.4327, 0.5884], [0.3758, 0.5954], [0.2972, 0.555], [0.2252, 0.5046], [0.1734, 0.4789], [0.1378, 0.49], [0.1063, 0.5247], [0.0699, 0.5577], [0.0277, 0.5713], [-0.0165, 0.5675], [-0.0613, 0.5661], [-0.112, 0.5891], [-0.1779, 0.6449], [-0.2642, 0.72], [-0.3653, 0.7853], [-0.4645, 0.8111], [-0.5417, 0.7822], [-0.5836, 0.7041], [-0.5903, 0.5981], [-0.5748, 0.4891], [-0.5546, 0.3942], [-0.5426, 0.3188], [-0.5433, 0.259], [-0.5547, 0.2083], [-0.5755, 0.1621], [-0.61, 0.1178], [-0.6663, 0.0724], [-0.7488, 0.0202], [-0.8489, -0.0451], [-0.943, -0.125], [-1.0, -0.2124], [-0.996, -0.2925], [-0.9281, -0.3502], [-0.8176, -0.3799], [-0.7002, -0.39], [-0.607, -0.3983], [-0.5486, -0.4192], [-0.5113, -0.452], [-0.4685, -0.4775], [-0.3995, -0.4696], [-0.3044, -0.4139], [-0.204, -0.3232], [-0.1268, -0.2367], [-0.0898, -0.2013], [-0.0886, -0.2448], [-0.1005, -0.3574], [-0.1001, -0.4942], [-0.0752, -0.6], [-0.031, -0.6402], [0.018, -0.6207], [0.063, -0.5819], [0.1087, -0.5716], [0.1693, -0.6138], [0.2543, -0.6929], [0.3555, -0.7643], [0.4484, -0.7828], [0.5065, -0.7312], [0.5219, -0.6297], [0.5142, -0.521]]}, {"id": 8, "shape_id": 8, "size": 2, "shape_vertices": [[0.0355, -0.9133], [0.1049, -0.9255], [0.1707, -0.9026], [0.2257, -0.8442], [0.2621, -0.7509], [0.2733, -0.6273], [0.2577, -0.4876], [0.2256, -0.358], [0.1999, -0.2691], [0.2086, -0.2396], [0.2688, -0.264], [0.3719, -0.3118], [0.4822, -0.3429], [0.5535, -0.3302], [0.5566, -0.2737], [0.5004, -0.1972], [0.4315, -0.1303], [0.4093, -0.0876], [0.4712, -0.0605], [0.6095, -0.0272], [0.7744, 0.0301], [0.9023, 0.1108], [0.9498, 0.1979], [0.9134, 0.271], [0.8238, 0.3205], [0.7208, 0.3509], [0.629, 0.3725], [0.5497, 0.3888], [0.4708, 0.3932], [0.3854, 0.3777], [0.3023, 0.347], [0.2413, 0.325], [0.2171, 0.3453], [0.227, 0.431], [0.2497, 0.5765], [0.2583, 0.7464], [0.2357, 0.8925], [0.1818, 0.9793], [0.1097, 1.0], [0.0343, 0.9742], [-0.0361, 0.9299], [-0.1002, 0.8845], [-0.1585, 0.8382], [-0.209, 0.7818], [-0.2485, 0.7119], [-0.2783, 0.6389], [-0.3076, 0.5819], [-0.3492, 0.5541], [-0.4091, 0.5506], [-0.4782, 0.5493], [-0.5331, 0.5236], [-0.5475, 0.459], [-0.508, 0.3613], [-0.4232, 0.2525], [-0.3204, 0.1576], [-0.2334, 0.092], [-0.1877, 0.0567], [-0.1941, 0.0415], [-0.251, 0.0322], [-0.35, 0.0156], [-0.4795, -0.0186], [-0.6236, -0.0766], [-0.7589, -0.1581], [-0.8576, -0.2544], [-0.8962, -0.3486], [-0.8677, -0.4225], [-0.7863, -0.4656], [-0.6822, -0.4825], [-0.5868, -0.4901], [-0.5175, -0.5072], [-0.4727, -0.5425], [-0.4376, -0.5894], [-0.3973, -0.6318], [-0.346, -0.6571], [-0.2885, -0.6662], [-0.233, -0.6733], [-0.1838, -0.6962], [-0.1379, -0.743], [-0.0885, -0.8068], [-0.0306, -0.8698]]}, {"id": 9, "shape_id": 9, "size": 5, "shape_vertices": [[-0.4803, -0.2801], [-0.4791, -0.3524], [-0.4663, -0.4251], [-0.4318, -0.484], [-0.3824, -0.5264], [-0.3321, -0.5646], [-0.2892, -0.6162], [-0.2504, -0.6856], [-0.2053, -0.755], [-0.1481, -0.792], [-0.0837, -0.7716], [-0.0242, -0.6968], [0.0213, -0.5996], [0.0539, -0.521], [0.0818, -0.4821], [0.1105, -0.4696], [0.1348, -0.4475], [0.143, -0.3887], [0.1319, -0.3021], [0.1189, -0.2341], [0.1396, -0.2393], [0.2266, -0.3413], [0.3831, -0.5099], [0.5718, -0.6741], [0.7297, -0.7624], [0.8054, -0.7442], [0.7911, -0.6436], [0.7291, -0.5182], [0.684, -0.4197], [0.6988, -0.3639], [0.765, -0.3296], [0.8283, -0.2836], [0.826, -0.2103], [0.7325, -0.1218], [0.579, -0.0441], [0.4354, 0.0073], [0.3666, 0.0418], [0.3916, 0.085], [0.4741, 0.1554], [0.5479, 0.2459], [0.5588, 0.3259], [0.4947, 0.3639], [0.3861, 0.352], [0.2811, 0.315], [0.2138, 0.2942], [0.1884, 0.3203], [0.185, 0.3942], [0.179, 0.4902], [0.157, 0.5774], [0.1203, 0.6429], [0.0757, 0.6981], [0.0265, 0.7648], [-0.0303, 0.8521], [-0.0975, 0.9427], [-0.1697, 1.0], [-0.2331, 0.9903], [-0.2725, 0.9048], [-0.2821, 0.7667], [-0.2702, 0.619], [-0.2552, 0.5025], [-0.2554, 0.438], [-0.2798, 0.4215], [-0.3251, 0.4326], [-0.3793, 0.4472], [-0.4285, 0.4477], [-0.4615, 0.4265], [-0.4733, 0.3851], [-0.4668, 0.3317], [-0.4531, 0.2781], [-0.4504, 0.2346], [-0.4765, 0.2053], [-0.539, 0.1846], [-0.6273, 0.1597], [-0.7133, 0.1186], [-0.7633, 0.0581], [-0.7563, -0.0127], [-0.6964, -0.0794], [-0.6111, -0.1326], [-0.5356, -0.1755], [-0.4924, -0.221]]}], "radius": 36, "fourier_perturbation_amplitude": 0.1, "min_pairwise_silhouette_distance": 0.11, "blob_shapes": [[[0.232, 0.6081], [0.1641, 0.5979], [0.103, 0.5933], [0.0462, 0.5979], [-0.0098, 0.6138], [-0.0688, 0.6407], [-0.134, 0.6761], [-0.2069, 0.7151], [-0.2873, 0.7516], [-0.372, 0.7785], [-0.4563, 0.7897], [-0.5336, 0.7805], [-0.597, 0.7489], [-0.6408, 0.6957], [-0.6609, 0.6244], [-0.6567, 0.541], [-0.6308, 0.4528], [-0.5891, 0.367], [-0.5403, 0.2899], [-0.4942, 0.2256], [-0.4606, 0.1757], [-0.4476, 0.1387], [-0.4602, 0.1108], [-0.4996, 0.0866], [-0.563, 0.06], [-0.644, 0.0256], [-0.7334, -0.0207], [-0.8208, -0.0804], [-0.8963, -0.1529], [-0.9513, -0.2352], [-0.9802, -0.3227], [-0.9805, -0.4102], [-0.953, -0.4924], [-0.9012, -0.5655], [-0.8305, -0.6266], [-0.7469, -0.6752], [-0.6562, -0.7119], [-0.5634, -0.7386], [-0.4717, -0.7577], [-0.383, -0.7711], [-0.2976, -0.7804], [-0.2156, -0.7856], [-0.1365, -0.7861], [-0.0603, -0.7804], [0.0122, -0.7667], [0.0799, -0.7437], [0.1408, -0.7108], [0.1935, -0.6687], [0.2367, -0.6194], [0.2705, -0.566], [0.2959, -0.5122], [0.3157, -0.4618], [0.3332, -0.4179], [0.3525, -0.3827], [0.3775, -0.3567], [0.4113, -0.3388], [0.4554, -0.3269], [0.5099, -0.3176], [0.5732, -0.3075], [0.6424, -0.2933], [0.7135, -0.2721], [0.7824, -0.2425], [0.8454, -0.2036], [0.8994, -0.156], [0.9425, -0.1005], [0.9737, -0.0386], [0.9928, 0.028], [1.0, 0.098], [0.9957, 0.1699], [0.9802, 0.2424], [0.9534, 0.3139], [0.9151, 0.3828], [0.8652, 0.447], [0.8038, 0.5043], [0.7319, 0.5523], [0.6515, 0.589], [0.5654, 0.6133], [0.4769, 0.6253], [0.3899, 0.6262], [0.3075, 0.6191]], [[0.063, -0.7631], [0.1078, -0.7534], [0.1508, -0.736], [0.1905, -0.7082], [0.2255, -0.6707], [0.2556, -0.627], [0.2823, -0.5821], [0.3081, -0.5408], [0.3362, -0.506], [0.3689, -0.4776], [0.4066, -0.4527], [0.4476, -0.4269], [0.4878, -0.3953], [0.5218, -0.3544], [0.5445, -0.3036], [0.5528, -0.2449], [0.5467, -0.1823], [0.5299, -0.1206], [0.5091, -0.0631], [0.4927, -0.0103], [0.4885, 0.0403], [0.5014, 0.0942], [0.5322, 0.1577], [0.5767, 0.2361], [0.6263, 0.3315], [0.6702, 0.4409], [0.6974, 0.5565], [0.6994, 0.6668], [0.6725, 0.7591], [0.6179, 0.8226], [0.542, 0.8512], [0.4541, 0.8454], [0.3645, 0.8127], [0.2821, 0.7656], [0.2121, 0.7195], [0.1556, 0.6886], [0.1097, 0.6823], [0.0691, 0.7038], [0.0279, 0.7484], [-0.0183, 0.8055], [-0.0711, 0.861], [-0.1289, 0.9011], [-0.1875, 0.915], [-0.2415, 0.898], [-0.2863, 0.8517], [-0.3194, 0.7835], [-0.3415, 0.7041], [-0.3558, 0.6246], [-0.3677, 0.5534], [-0.3824, 0.495], [-0.4033, 0.4491], [-0.4311, 0.4112], [-0.463, 0.3752], [-0.4936, 0.3353], [-0.5163, 0.2879], [-0.5258, 0.2329], [-0.5194, 0.1733], [-0.4988, 0.1136], [-0.4695, 0.0582], [-0.4401, 0.0092], [-0.4201, -0.0347], [-0.4174, -0.0784], [-0.4362, -0.1293], [-0.4758, -0.1948], [-0.5299, -0.2804], [-0.5885, -0.3871], [-0.6398, -0.5105], [-0.6727, -0.6413], [-0.6792, -0.7667], [-0.6562, -0.8736], [-0.6056, -0.9511], [-0.5334, -0.9932], [-0.4486, -1.0], [-0.3602, -0.9775], [-0.2759, -0.9359], [-0.2005, -0.8869], [-0.1353, -0.8413], [-0.0791, -0.806], [-0.0292, -0.7833], [0.0175, -0.7709]], [[0.9365, 0.0309], [0.9948, 0.1022], [1.0, 0.1737], [0.9543, 0.2356], [0.8707, 0.2814], [0.7687, 0.3104], [0.6687, 0.328], [0.5863, 0.3429], [0.5288, 0.3646], [0.4941, 0.3991], [0.4734, 0.4469], [0.4544, 0.5022], [0.4266, 0.5551], [0.384, 0.5946], [0.3268, 0.6125], [0.26, 0.6057], [0.191, 0.5776], [0.1269, 0.5363], [0.0715, 0.4919], [0.0253, 0.4534], [-0.0141, 0.4259], [-0.0502, 0.4099], [-0.0858, 0.4017], [-0.1218, 0.396], [-0.1576, 0.3885], [-0.1924, 0.3776], [-0.2269, 0.365], [-0.264, 0.355], [-0.3092, 0.3517], [-0.3686, 0.357], [-0.4466, 0.3691], [-0.5427, 0.3821], [-0.6506, 0.3879], [-0.7581, 0.3787], [-0.85, 0.3494], [-0.9113, 0.2998], [-0.9319, 0.2345], [-0.9093, 0.1617], [-0.8495, 0.0904], [-0.7656, 0.0279], [-0.6739, -0.0222], [-0.5899, -0.0606], [-0.5241, -0.0911], [-0.4806, -0.1186], [-0.4567, -0.1476], [-0.4458, -0.18], [-0.4401, -0.2158], [-0.4335, -0.2535], [-0.4238, -0.2922], [-0.4118, -0.3326], [-0.4001, -0.3777], [-0.3904, -0.4315], [-0.3822, -0.4973], [-0.3713, -0.575], [-0.3519, -0.6596], [-0.3182, -0.7413], [-0.2671, -0.8075], [-0.2, -0.8454], [-0.1231, -0.8466], [-0.0453, -0.8095], [0.0244, -0.7404], [0.0799, -0.6522], [0.1198, -0.5608], [0.1478, -0.4808], [0.1711, -0.4217], [0.1965, -0.3855], [0.2283, -0.3673], [0.2657, -0.3573], [0.3035, -0.3452], [0.334, -0.3235], [0.3511, -0.2901], [0.3531, -0.2486], [0.3455, -0.206], [0.3399, -0.1698], [0.351, -0.1443], [0.3918, -0.1289], [0.4688, -0.118], [0.5786, -0.1029], [0.7075, -0.0753], [0.8345, -0.0304]], [[-0.5289, 0.7133], [-0.5946, 0.668], [-0.6303, 0.5926], [-0.6265, 0.4926], [-0.5837, 0.3815], [-0.5148, 0.2763], [-0.4416, 0.1907], [-0.3893, 0.1308], [-0.3787, 0.0936], [-0.4199, 0.0687], [-0.5093, 0.0426], [-0.6311, 0.004], [-0.762, -0.053], [-0.8779, -0.127], [-0.96, -0.2114], [-0.999, -0.2966], [-0.9951, -0.3742], [-0.9563, -0.4385], [-0.8946, -0.4883], [-0.8223, -0.5258], [-0.749, -0.5554], [-0.6805, -0.5815], [-0.6188, -0.6078], [-0.5632, -0.6361], [-0.5112, -0.6663], [-0.4596, -0.6964], [-0.4056, -0.7226], [-0.3471, -0.7405], [-0.284, -0.7453], [-0.2177, -0.7336], [-0.1512, -0.705], [-0.0884, -0.6628], [-0.0322, -0.6145], [0.0169, -0.5702], [0.0613, -0.5404], [0.1067, -0.5323], [0.1599, -0.5476], [0.2266, -0.5814], [0.3086, -0.6231], [0.4033, -0.6595], [0.5031, -0.6784], [0.5981, -0.672], [0.6793, -0.6387], [0.7409, -0.5826], [0.7823, -0.5113], [0.8071, -0.4333], [0.8216, -0.3548], [0.8311, -0.2793], [0.8384, -0.2073], [0.8429, -0.1378], [0.8416, -0.0703], [0.8311, -0.0052], [0.8102, 0.0563], [0.781, 0.113], [0.7476, 0.1646], [0.7149, 0.2123], [0.6857, 0.2578], [0.6595, 0.3024], [0.6326, 0.3453], [0.6001, 0.3837], [0.5587, 0.4143], [0.5089, 0.4349], [0.4553, 0.4473], [0.4056, 0.4581], [0.3666, 0.4778], [0.3412, 0.517], [0.3268, 0.5823], [0.3153, 0.6726], [0.2962, 0.7774], [0.2609, 0.8792], [0.2056, 0.9584], [0.1334, 1.0], [0.0523, 0.9982], [-0.0283, 0.9588], [-0.1019, 0.8975], [-0.1672, 0.8335], [-0.2287, 0.783], [-0.2936, 0.7535], [-0.3671, 0.7411], [-0.4483, 0.7332]], [[-0.5976, -0.2911], [-0.621, -0.3633], [-0.6346, -0.4394], [-0.6156, -0.4998], [-0.5548, -0.5257], [-0.4649, -0.5136], [-0.3728, -0.4816], [-0.302, -0.4595], [-0.2589, -0.4695], [-0.2322, -0.5117], [-0.2046, -0.5647], [-0.1659, -0.6018], [-0.1172, -0.6096], [-0.0661, -0.5963], [-0.0179, -0.5835], [0.0292, -0.5887], [0.08, -0.6127], [0.1368, -0.6402], [0.196, -0.6538], [0.2528, -0.6474], [0.307, -0.6302], [0.3654, -0.6176], [0.4364, -0.6175], [0.5206, -0.6231], [0.6062, -0.6169], [0.6738, -0.5836], [0.7078, -0.52], [0.7061, -0.4366], [0.6807, -0.3496], [0.6495, -0.271], [0.6248, -0.2047], [0.6103, -0.1483], [0.6048, -0.0979], [0.6097, -0.0506], [0.6305, -0.0035], [0.6687, 0.0478], [0.7128, 0.1063], [0.7355, 0.168], [0.7061, 0.2191], [0.6107, 0.2419], [0.466, 0.227], [0.3158, 0.1848], [0.2086, 0.1444], [0.1706, 0.1385], [0.1925, 0.1824], [0.2376, 0.2625], [0.2672, 0.3452], [0.2631, 0.4003], [0.2341, 0.4246], [0.2021, 0.4454], [0.1819, 0.5019], [0.1694, 0.6146], [0.1473, 0.7667], [0.1009, 0.9102], [0.0305, 0.995], [-0.0495, 1.0], [-0.1233, 0.9444], [-0.1862, 0.8716], [-0.2453, 0.8182], [-0.3088, 0.7908], [-0.3734, 0.7666], [-0.4236, 0.7159], [-0.4443, 0.6287], [-0.4379, 0.5242], [-0.4291, 0.4367], [-0.4509, 0.3905], [-0.5195, 0.3817], [-0.6168, 0.3814], [-0.696, 0.3574], [-0.7093, 0.296], [-0.6397, 0.2096], [-0.5153, 0.1252], [-0.3938, 0.0637], [-0.3293, 0.0273], [-0.3413, 0.0019], [-0.4084, -0.0292], [-0.4874, -0.0727], [-0.5437, -0.1242], [-0.5698, -0.1768], [-0.5814, -0.2303]], [[-0.6153, 0.1302], [-0.6012, 0.0547], [-0.5678, -0.0144], [-0.5395, -0.0749], [-0.5329, -0.1339], [-0.5474, -0.199], [-0.5663, -0.2704], [-0.5672, -0.3371], [-0.5355, -0.3833], [-0.4724, -0.3986], [-0.3933, -0.3857], [-0.3194, -0.3605], [-0.2675, -0.3458], [-0.2446, -0.3612], [-0.2467, -0.4162], [-0.262, -0.5067], [-0.2751, -0.6139], [-0.2723, -0.7089], [-0.2467, -0.7611], [-0.2009, -0.7528], [-0.1463, -0.6914], [-0.0963, -0.6114], [-0.0581, -0.5588], [-0.0284, -0.5657], [0.0027, -0.6277], [0.0424, -0.7037], [0.0877, -0.7392], [0.1262, -0.6998], [0.1461, -0.5934], [0.1473, -0.4633], [0.1432, -0.3604], [0.1517, -0.3117], [0.1824, -0.3095], [0.2309, -0.3251], [0.2857, -0.3335], [0.3416, -0.3285], [0.4075, -0.3183], [0.4983, -0.309], [0.6178, -0.2927], [0.7447, -0.252], [0.8372, -0.1771], [0.8561, -0.0778], [0.7905, 0.0201], [0.6686, 0.0929], [0.543, 0.1364], [0.46, 0.1672], [0.4345, 0.2075], [0.4444, 0.2641], [0.449, 0.3214], [0.4164, 0.3513], [0.3409, 0.3343], [0.2432, 0.2745], [0.1546, 0.1998], [0.0997, 0.1472], [0.0868, 0.1465], [0.1087, 0.2103], [0.1496, 0.3339], [0.1918, 0.4993], [0.2202, 0.6796], [0.2247, 0.842], [0.2021, 0.9554], [0.1575, 1.0], [0.1016, 0.9772], [0.0458, 0.9103], [-0.0035, 0.8341], [-0.0468, 0.7765], [-0.0884, 0.7451], [-0.1308, 0.7257], [-0.1708, 0.6937], [-0.2006, 0.6308], [-0.2131, 0.5363], [-0.208, 0.4274], [-0.1941, 0.3293], [-0.187, 0.2632], [-0.2029, 0.2369], [-0.2521, 0.2423], [-0.333, 0.2601], [-0.4319, 0.2678], [-0.5257, 0.249], [-0.5913, 0.2001]], [[-0.7855, -0.5247], [-0.7902, -0.6335], [-0.7715, -0.7369], [-0.7229, -0.8206], [-0.6444, -0.8715], [-0.5432, -0.8819], [-0.4315, -0.8528], [-0.3229, -0.7948], [-0.2282, -0.7258], [-0.1523, -0.6666], [-0.0933, -0.6352], [-0.0439, -0.6417], [0.0055, -0.6856], [0.0631, -0.7562], [0.1326, -0.8354], [0.2123, -0.9034], [0.2953, -0.9433], [0.3727, -0.946], [0.4364, -0.9109], [0.4815, -0.845], [0.5078, -0.7603], [0.5191, -0.6691], [0.5212, -0.5813], [0.5198, -0.5023], [0.5185, -0.4335], [0.5186, -0.3732], [0.52, -0.3191], [0.5223, -0.2695], [0.5274, -0.2236], [0.5393, -0.1815], [0.5638, -0.1423], [0.6063, -0.1031], [0.669, -0.0593], [0.7486, -0.0053], [0.836, 0.0631], [0.917, 0.1471], [0.9761, 0.2427], [1.0, 0.3416], [0.9818, 0.433], [0.9227, 0.5065], [0.832, 0.5558], [0.7243, 0.5807], [0.6154, 0.5878], [0.5182, 0.5882], [0.4395, 0.5944], [0.3791, 0.6156], [0.3315, 0.6552], [0.2883, 0.7096], [0.2418, 0.769], [0.1877, 0.8216], [0.1259, 0.8571], [0.0596, 0.8704], [-0.0069, 0.8633], [-0.0704, 0.8433], [-0.1304, 0.8211], [-0.1894, 0.8058], [-0.2511, 0.8023], [-0.3185, 0.8085], [-0.3912, 0.8166], [-0.4647, 0.8156], [-0.5312, 0.7952], [-0.5817, 0.7496], [-0.6094, 0.6796], [-0.6126, 0.592], [-0.5955, 0.4979], [-0.568, 0.4088], [-0.5425, 0.333], [-0.5304, 0.2736], [-0.5385, 0.2284], [-0.567, 0.1908], [-0.6095, 0.1538], [-0.6561, 0.1115], [-0.6962, 0.0617], [-0.7231, 0.0051], [-0.7351, -0.0555], [-0.7367, -0.1181], [-0.7353, -0.1828], [-0.7385, -0.2523], [-0.7502, -0.3309], [-0.7685, -0.4218]], [[0.5211, -0.4433], [0.5742, -0.4081], [0.6763, -0.3973], [0.7969, -0.3799], [0.8893, -0.334], [0.9209, -0.2594], [0.8938, -0.1727], [0.8439, -0.0918], [0.8159, -0.022], [0.8315, 0.0442], [0.873, 0.1157], [0.893, 0.1897], [0.8446, 0.248], [0.714, 0.2694], [0.5325, 0.2474], [0.3617, 0.2015], [0.2594, 0.1702], [0.2493, 0.1905], [0.3099, 0.2739], [0.3904, 0.3979], [0.4399, 0.517], [0.4327, 0.5884], [0.3758, 0.5954], [0.2972, 0.555], [0.2252, 0.5046], [0.1734, 0.4789], [0.1378, 0.49], [0.1063, 0.5247], [0.0699, 0.5577], [0.0277, 0.5713], [-0.0165, 0.5675], [-0.0613, 0.5661], [-0.112, 0.5891], [-0.1779, 0.6449], [-0.2642, 0.72], [-0.3653, 0.7853], [-0.4645, 0.8111], [-0.5417, 0.7822], [-0.5836, 0.7041], [-0.5903, 0.5981], [-0.5748, 0.4891], [-0.5546, 0.3942], [-0.5426, 0.3188], [-0.5433, 0.259], [-0.5547, 0.2083], [-0.5755, 0.1621], [-0.61, 0.1178], [-0.6663, 0.0724], [-0.7488, 0.0202], [-0.8489, -0.0451], [-0.943, -0.125], [-1.0, -0.2124], [-0.996, -0.2925], [-0.9281, -0.3502], [-0.8176, -0.3799], [-0.7002, -0.39], [-0.607, -0.3983], [-0.5486, -0.4192], [-0.5113, -0.452], [-0.4685, -0.4775], [-0.3995, -0.4696], [-0.3044, -0.4139], [-0.204, -0.3232], [-0.1268, -0.2367], [-0.0898, -0.2013], [-0.0886, -0.2448], [-0.1005, -0.3574], [-0.1001, -0.4942], [-0.0752, -0.6], [-0.031, -0.6402], [0.018, -0.6207], [0.063, -0.5819], [0.1087, -0.5716], [0.1693, -0.6138], [0.2543, -0.6929], [0.3555, -0.7643], [0.4484, -0.7828], [0.5065, -0.7312], [0.5219, -0.6297], [0.5142, -0.521]], [[0.0355, -0.9133], [0.1049, -0.9255], [0.1707, -0.9026], [0.2257, -0.8442], [0.2621, -0.7509], [0.2733, -0.6273], [0.2577, -0.4876], [0.2256, -0.358], [0.1999, -0.2691], [0.2086, -0.2396], [0.2688, -0.264], [0.3719, -0.3118], [0.4822, -0.3429], [0.5535, -0.3302], [0.5566, -0.2737], [0.5004, -0.1972], [0.4315, -0.1303], [0.4093, -0.0876], [0.4712, -0.0605], [0.6095, -0.0272], [0.7744, 0.0301], [0.9023, 0.1108], [0.9498, 0.1979], [0.9134, 0.271], [0.8238, 0.3205], [0.7208, 0.3509], [0.629, 0.3725], [0.5497, 0.3888], [0.4708, 0.3932], [0.3854, 0.3777], [0.3023, 0.347], [0.2413, 0.325], [0.2171, 0.3453], [0.227, 0.431], [0.2497, 0.5765], [0.2583, 0.7464], [0.2357, 0.8925], [0.1818, 0.9793], [0.1097, 1.0], [0.0343, 0.9742], [-0.0361, 0.9299], [-0.1002, 0.8845], [-0.1585, 0.8382], [-0.209, 0.7818], [-0.2485, 0.7119], [-0.2783, 0.6389], [-0.3076, 0.5819], [-0.3492, 0.5541], [-0.4091, 0.5506], [-0.4782, 0.5493], [-0.5331, 0.5236], [-0.5475, 0.459], [-0.508, 0.3613], [-0.4232, 0.2525], [-0.3204, 0.1576], [-0.2334, 0.092], [-0.1877, 0.0567], [-0.1941, 0.0415], [-0.251, 0.0322], [-0.35, 0.0156], [-0.4795, -0.0186], [-0.6236, -0.0766], [-0.7589, -0.1581], [-0.8576, -0.2544], [-0.8962, -0.3486], [-0.8677, -0.4225], [-0.7863, -0.4656], [-0.6822, -0.4825], [-0.5868, -0.4901], [-0.5175, -0.5072], [-0.4727, -0.5425], [-0.4376, -0.5894], [-0.3973, -0.6318], [-0.346, -0.6571], [-0.2885, -0.6662], [-0.233, -0.6733], [-0.1838, -0.6962], [-0.1379, -0.743], [-0.0885, -0.8068], [-0.0306, -0.8698]], [[-0.4803, -0.2801], [-0.4791, -0.3524], [-0.4663, -0.4251], [-0.4318, -0.484], [-0.3824, -0.5264], [-0.3321, -0.5646], [-0.2892, -0.6162], [-0.2504, -0.6856], [-0.2053, -0.755], [-0.1481, -0.792], [-0.0837, -0.7716], [-0.0242, -0.6968], [0.0213, -0.5996], [0.0539, -0.521], [0.0818, -0.4821], [0.1105, -0.4696], [0.1348, -0.4475], [0.143, -0.3887], [0.1319, -0.3021], [0.1189, -0.2341], [0.1396, -0.2393], [0.2266, -0.3413], [0.3831, -0.5099], [0.5718, -0.6741], [0.7297, -0.7624], [0.8054, -0.7442], [0.7911, -0.6436], [0.7291, -0.5182], [0.684, -0.4197], [0.6988, -0.3639], [0.765, -0.3296], [0.8283, -0.2836], [0.826, -0.2103], [0.7325, -0.1218], [0.579, -0.0441], [0.4354, 0.0073], [0.3666, 0.0418], [0.3916, 0.085], [0.4741, 0.1554], [0.5479, 0.2459], [0.5588, 0.3259], [0.4947, 0.3639], [0.3861, 0.352], [0.2811, 0.315], [0.2138, 0.2942], [0.1884, 0.3203], [0.185, 0.3942], [0.179, 0.4902], [0.157, 0.5774], [0.1203, 0.6429], [0.0757, 0.6981], [0.0265, 0.7648], [-0.0303, 0.8521], [-0.0975, 0.9427], [-0.1697, 1.0], [-0.2331, 0.9903], [-0.2725, 0.9048], [-0.2821, 0.7667], [-0.2702, 0.619], [-0.2552, 0.5025], [-0.2554, 0.438], [-0.2798, 0.4215], [-0.3251, 0.4326], [-0.3793, 0.4472], [-0.4285, 0.4477], [-0.4615, 0.4265], [-0.4733, 0.3851], [-0.4668, 0.3317], [-0.4531, 0.2781], [-0.4504, 0.2346], [-0.4765, 0.2053], [-0.539, 0.1846], [-0.6273, 0.1597], [-0.7133, 0.1186], [-0.7633, 0.0581], [-0.7563, -0.0127], [-0.6964, -0.0794], [-0.6111, -0.1326], [-0.5356, -0.1755], [-0.4924, -0.221]]], "image": "images/attribute_group_counting_00004.png", "question": "How many distinct shapes are in the image? The image contains many small pieces scattered across the canvas. Two pieces belong to the same type if and only if they have exactly the same silhouette. Pieces may differ in surface texture or colour tone \u2014 only the silhouette outline matters. Count the total number of distinct silhouette types and report the count as a positive integer. Provide your final answer enclosed in ... tags."} diff --git a/data/annotations/bounded_faces_counting.jsonl b/data/annotations/bounded_faces_counting.jsonl new file mode 100644 index 0000000000000000000000000000000000000000..bed7fef83eb5170234cb6b5636373d8aa8559d08 --- /dev/null +++ b/data/annotations/bounded_faces_counting.jsonl @@ -0,0 +1,5 @@ +{"image": "images/bounded_faces_counting_00000.png", "question": "How many distinct enclosed regions (bounded faces) are visible in this image? An enclosed region is a maximal area that is fully surrounded by drawn lines or strands on every side, with no opening to the outside background. The unbounded outside area does not count. Each enclosed region should be counted exactly once. Provide your final answer enclosed in ... tags.", "answer": 5, "metadata": {"bounded_faces": 5, "num_dots": 58, "num_edges": 62, "num_components": 1, "seed": 478163327}} +{"image": "images/bounded_faces_counting_00001.png", "question": "How many distinct enclosed regions (bounded faces) are visible in this image? An enclosed region is a maximal area that is fully surrounded by drawn lines or strands on every side, with no opening to the outside background. The unbounded outside area does not count. Each enclosed region should be counted exactly once. Provide your final answer enclosed in ... tags.", "answer": 8, "metadata": {"bounded_faces": 8, "num_dots": 69, "num_edges": 76, "num_components": 1, "seed": 798112150}} +{"image": "images/bounded_faces_counting_00002.png", "question": "How many distinct enclosed regions (bounded faces) are visible in this image? An enclosed region is a maximal area that is fully surrounded by drawn lines or strands on every side, with no opening to the outside background. The unbounded outside area does not count. Each enclosed region should be counted exactly once. Provide your final answer enclosed in ... tags.", "answer": 10, "metadata": {"bounded_faces": 10, "num_dots": 68, "num_edges": 77, "num_components": 1, "seed": 110287971}} +{"image": "images/bounded_faces_counting_00003.png", "question": "How many distinct enclosed regions (bounded faces) are visible in this image? An enclosed region is a maximal area that is fully surrounded by drawn lines or strands on every side, with no opening to the outside background. The unbounded outside area does not count. Each enclosed region should be counted exactly once. Provide your final answer enclosed in ... tags.", "answer": 12, "metadata": {"bounded_faces": 12, "num_dots": 60, "num_edges": 70, "num_components": 2, "seed": 1411178445}} +{"image": "images/bounded_faces_counting_00004.png", "question": "How many distinct enclosed regions (bounded faces) are visible in this image? An enclosed region is a maximal area that is fully surrounded by drawn lines or strands on every side, with no opening to the outside background. The unbounded outside area does not count. Each enclosed region should be counted exactly once. Provide your final answer enclosed in ... tags.", "answer": 15, "metadata": {"bounded_faces": 15, "num_dots": 61, "num_edges": 75, "num_components": 1, "seed": 963124353}} diff --git a/data/annotations/color_zone_sequence.jsonl b/data/annotations/color_zone_sequence.jsonl new file mode 100644 index 0000000000000000000000000000000000000000..050638eb256afa94f9b87b67049bf252f390f011 --- /dev/null +++ b/data/annotations/color_zone_sequence.jsonl @@ -0,0 +1,5 @@ +{"width": 1586, "height": 1586, "num_zones": 10, "region_labels": ["A", "B", "C", "D", "E", "F", "G", "H", "I", "J"], "region_colors": ["#e8b08a", "#4bb8ad", "#ec9863", "#6fa3dc", "#edd351", "#8cc571", "#7aa06a", "#c2a055", "#b49278", "#e2c28e"], "label_positions": [[614.0, 564.0], [1356.0, 745.0], [270.0, 336.0], [712.0, 1152.0], [955.0, 971.0], [1237.0, 1281.0], [1315.0, 299.0], [149.0, 1350.0], [1006.0, 144.0], [205.0, 916.0]], "sequence_length": 11, "sequence": ["H", "D", "F", "B", "I", "A", "C", "J", "A", "J", "H"], "visit_ids": [7, 3, 5, 1, 8, 0, 2, 9, 0, 9, 7], "question": "The image shows several distinct regions, each carrying a single-letter label and distinguished by its own fill (colour, texture, or material). A single smooth curve starts at the position marked 'Start' and travels through several regions, potentially re-entering earlier ones. Follow the curve from start to end and list the region labels it visits in order, separated by commas. Provide your final answer enclosed in ... tags.", "answer": "H, D, F, B, I, A, C, J, A, J, H", "image": "images/color_zone_00000.png"} +{"width": 1586, "height": 1586, "num_zones": 10, "region_labels": ["A", "B", "C", "D", "E", "F", "G", "H", "I", "J"], "region_colors": ["#8cc571", "#edd351", "#df7fba", "#7aa06a", "#b49278", "#e07a70", "#e8b08a", "#ec9863", "#e2c28e", "#4bb8ad"], "label_positions": [[433.0, 610.0], [983.0, 1378.0], [748.0, 618.0], [1421.0, 930.0], [974.0, 1026.0], [689.0, 221.0], [1308.0, 277.0], [509.0, 1308.0], [194.0, 1036.0], [166.0, 253.0]], "sequence_length": 11, "sequence": ["B", "D", "C", "G", "C", "E", "I", "A", "J", "F", "C"], "visit_ids": [1, 3, 2, 6, 2, 4, 8, 0, 9, 5, 2], "question": "The image shows several distinct regions, each carrying a single-letter label and distinguished by its own fill (colour, texture, or material). A single smooth curve starts at the position marked 'Start' and travels through several regions, potentially re-entering earlier ones. Follow the curve from start to end and list the region labels it visits in order, separated by commas. Provide your final answer enclosed in ... tags.", "answer": "B, D, C, G, C, E, I, A, J, F, C", "image": "images/color_zone_00001.png"} +{"width": 1586, "height": 1586, "num_zones": 10, "region_labels": ["A", "B", "C", "D", "E", "F", "G", "H", "I", "J"], "region_colors": ["#7aa06a", "#7ec4c1", "#e07a70", "#8cc571", "#d9a5e0", "#e2c28e", "#c2a055", "#6fa3dc", "#9d86d4", "#e8b08a"], "label_positions": [[207.0, 1376.0], [723.0, 825.0], [171.0, 171.0], [179.0, 1023.0], [1427.0, 614.0], [1087.0, 319.0], [1401.0, 1401.0], [1079.0, 1197.0], [382.0, 497.0], [655.0, 1296.0]], "sequence_length": 11, "sequence": ["F", "E", "F", "I", "C", "D", "A", "J", "H", "G", "E"], "visit_ids": [5, 4, 5, 8, 2, 3, 0, 9, 7, 6, 4], "question": "The image shows several distinct regions, each carrying a single-letter label and distinguished by its own fill (colour, texture, or material). A single smooth curve starts at the position marked 'Start' and travels through several regions, potentially re-entering earlier ones. Follow the curve from start to end and list the region labels it visits in order, separated by commas. Provide your final answer enclosed in ... tags.", "answer": "F, E, F, I, C, D, A, J, H, G, E", "image": "images/color_zone_00002.png"} +{"width": 1586, "height": 1586, "num_zones": 10, "region_labels": ["A", "B", "C", "D", "E", "F", "G", "H", "I", "J"], "region_colors": ["#c9d36b", "#e07a70", "#d9a5e0", "#4bb8ad", "#e2c28e", "#b49278", "#7ec4c1", "#7aa06a", "#e8b08a", "#8cc571"], "label_positions": [[1427.0, 1169.0], [1459.0, 126.0], [871.0, 1129.0], [875.0, 190.0], [249.0, 748.0], [197.0, 190.0], [1221.0, 389.0], [528.0, 1228.0], [1347.0, 1438.0], [178.0, 1259.0]], "sequence_length": 12, "sequence": ["G", "B", "D", "G", "A", "I", "C", "I", "A", "G", "D", "F"], "visit_ids": [6, 1, 3, 6, 0, 8, 2, 8, 0, 6, 3, 5], "question": "The image shows several distinct regions, each carrying a single-letter label and distinguished by its own fill (colour, texture, or material). A single smooth curve starts at the position marked 'Start' and travels through several regions, potentially re-entering earlier ones. Follow the curve from start to end and list the region labels it visits in order, separated by commas. Provide your final answer enclosed in ... tags.", "answer": "G, B, D, G, A, I, C, I, A, G, D, F", "image": "images/color_zone_00003.png"} +{"width": 1586, "height": 1586, "num_zones": 10, "region_labels": ["A", "B", "C", "D", "E", "F", "G", "H", "I", "J"], "region_colors": ["#4bb8ad", "#c9d36b", "#8cc571", "#e8b08a", "#d9a5e0", "#9d86d4", "#df7fba", "#e07a70", "#ec9863", "#e2c28e"], "label_positions": [[958.0, 572.0], [159.0, 1278.0], [770.0, 1032.0], [157.0, 439.0], [753.0, 236.0], [1358.0, 227.0], [1386.0, 912.0], [585.0, 1398.0], [1148.0, 1401.0], [180.0, 745.0]], "sequence_length": 13, "sequence": ["C", "A", "G", "A", "F", "E", "D", "J", "B", "C", "H", "I", "G"], "visit_ids": [2, 0, 6, 0, 5, 4, 3, 9, 1, 2, 7, 8, 6], "question": "The image shows several distinct regions, each carrying a single-letter label and distinguished by its own fill (colour, texture, or material). A single smooth curve starts at the position marked 'Start' and travels through several regions, potentially re-entering earlier ones. Follow the curve from start to end and list the region labels it visits in order, separated by commas. Provide your final answer enclosed in ... tags.", "answer": "C, A, G, A, F, E, D, J, B, C, H, I, G", "image": "images/color_zone_00004.png"} diff --git a/data/annotations/constellation_match_count.jsonl b/data/annotations/constellation_match_count.jsonl new file mode 100644 index 0000000000000000000000000000000000000000..c10877b4538e99bdae2973986608c8e1f6fe77a0 --- /dev/null +++ b/data/annotations/constellation_match_count.jsonl @@ -0,0 +1,5 @@ +{"image": "images/constellation_match_count_00000.png", "question": "This image has two panels separated by a thin vertical divider. The left panel shows the Template: a small constellation of marker points. The right panel shows the Field: a larger scene with many marker points. Both panels are drawn at the SAME pixel scale. Count how many copies of the Template pattern appear in the Field. A copy may be rotated by a small angle (up to about 20\u00b0) and individual point positions may be jittered slightly, but the overall pattern of relative positions must be preserved. No mirroring or scaling. Report the count as a non-negative integer. Provide your final answer enclosed in ... tags.", "answer": 5, "num_template_dots": 5, "total_stars": 87, "num_matches": 5, "metadata": {"seed": 173879092, "tolerance": 10.0, "template_offsets": [[0.0, 0.0], [-3.4616725711160994, -46.954577705132806], [-93.61996391282639, -185.43041481111788], [-136.61615800989028, 43.50301029449731], [-46.03632956018431, 43.912698774913856]]}} +{"image": "images/constellation_match_count_00001.png", "question": "This image has two panels separated by a thin vertical divider. The left panel shows the Template: a small constellation of marker points. The right panel shows the Field: a larger scene with many marker points. Both panels are drawn at the SAME pixel scale. Count how many copies of the Template pattern appear in the Field. A copy may be rotated by a small angle (up to about 20\u00b0) and individual point positions may be jittered slightly, but the overall pattern of relative positions must be preserved. No mirroring or scaling. Report the count as a non-negative integer. Provide your final answer enclosed in ... tags.", "answer": 6, "num_template_dots": 5, "total_stars": 83, "num_matches": 6, "metadata": {"seed": 2087043557, "tolerance": 10.0, "template_offsets": [[0.0, 0.0], [41.471863900170575, -3.5227577023131147], [5.448720316866826, 115.88056826599409], [121.46995054662344, 79.74046043527612], [80.93273678646665, 200.86023062793714]]}} +{"image": "images/constellation_match_count_00002.png", "question": "This image has two panels separated by a thin vertical divider. The left panel shows the Template: a small constellation of marker points. The right panel shows the Field: a larger scene with many marker points. Both panels are drawn at the SAME pixel scale. Count how many copies of the Template pattern appear in the Field. A copy may be rotated by a small angle (up to about 20\u00b0) and individual point positions may be jittered slightly, but the overall pattern of relative positions must be preserved. No mirroring or scaling. Report the count as a non-negative integer. Provide your final answer enclosed in ... tags.", "answer": 8, "num_template_dots": 5, "total_stars": 87, "num_matches": 8, "metadata": {"seed": 2046968324, "tolerance": 10.0, "template_offsets": [[0.0, 0.0], [77.91193064942368, 33.84202775872543], [2.8003529401257765, 34.648794353110745], [72.87157848243598, 111.78624537905125], [111.2982677920412, -75.27606286426611]]}} +{"image": "images/constellation_match_count_00003.png", "question": "This image has two panels separated by a thin vertical divider. The left panel shows the Template: a small constellation of marker points. The right panel shows the Field: a larger scene with many marker points. Both panels are drawn at the SAME pixel scale. Count how many copies of the Template pattern appear in the Field. A copy may be rotated by a small angle (up to about 20\u00b0) and individual point positions may be jittered slightly, but the overall pattern of relative positions must be preserved. No mirroring or scaling. Report the count as a non-negative integer. Provide your final answer enclosed in ... tags.", "answer": 9, "num_template_dots": 5, "total_stars": 86, "num_matches": 9, "metadata": {"seed": 938204377, "tolerance": 10.0, "template_offsets": [[0.0, 0.0], [47.632343557594254, -44.75983700245754], [-46.37957980821564, 0.3673736111899366], [96.88568769916276, 146.00365752516385], [5.919548083564312, -45.64291592166866]]}} +{"image": "images/constellation_match_count_00004.png", "question": "This image has two panels separated by a thin vertical divider. The left panel shows the Template: a small constellation of marker points. The right panel shows the Field: a larger scene with many marker points. Both panels are drawn at the SAME pixel scale. Count how many copies of the Template pattern appear in the Field. A copy may be rotated by a small angle (up to about 20\u00b0) and individual point positions may be jittered slightly, but the overall pattern of relative positions must be preserved. No mirroring or scaling. Report the count as a non-negative integer. Provide your final answer enclosed in ... tags.", "answer": 10, "num_template_dots": 5, "total_stars": 79, "num_matches": 10, "metadata": {"seed": 600203567, "tolerance": 10.0, "template_offsets": [[0.0, 0.0], [-50.866585971769034, -96.4278447997325], [-47.43200738991797, 51.02936959014866], [46.583067551880816, -4.411768013683243], [-2.2544303107114274, -98.25764451187348]]}} diff --git a/data/annotations/contour_silhouette_count.jsonl b/data/annotations/contour_silhouette_count.jsonl new file mode 100644 index 0000000000000000000000000000000000000000..f05623df27ef22bf014bb6c6b5be5170607252f2 --- /dev/null +++ b/data/annotations/contour_silhouette_count.jsonl @@ -0,0 +1,5 @@ +{"image": "images/contour_silhouette_count_00000.png", "question": "This image has two panels separated by a thin vertical divider. The left panel shows the Template: a single closed silhouette. The right panel shows the Field: many closed silhouettes scattered across the canvas, all at the SAME pixel scale as the Template. Count the number of Field silhouettes that are exact copies of the Template (translation only \u2014 same size, same orientation, no rotation or mirroring) and report the integer count. Provide your final answer enclosed in ... tags.", "answer": "5", "num_matches": 5, "match_labels": ["B", "H", "L", "M", "R"], "total_blobs": 20, "metadata": {"seed": 1654615998, "template_points": [[59.88288659048854, -2.5776266602431703], [61.059053348364095, -0.6760765930405448], [61.89677537242383, 1.2819381701431545], [62.40304707733168, 3.269136921193678], [62.591688736192246, 5.259966682456739], [62.48249190828891, 7.23139776308075], [62.10020217664202, 9.163559455472562], [61.473383588421676, 11.040195556219723], [60.633212588661934, 12.848927987534973], [59.61225060504461, 14.581325726453352], [58.44324375222302, 16.23278511695815], [57.1579954403894, 17.802236048750277], [55.78635314200893, 19.291696063623622], [54.355344418465165, 20.705700865388526], [52.8884898280052, 22.050644686624672], [51.4053118751392, 23.334067297599415], [49.92105010509636, 24.563925998541816], [48.4465832014306, 25.747890667402817], [46.988549919331966, 26.892697876458296], [45.54965227488268, 28.00359635961049], [44.12911697157756, 29.083910900942225], [42.723284891236645, 30.13474528354358], [41.326293856011844, 31.154837599769493], [39.93081695774208, 32.14057333339159], [38.52881764670731, 33.08615355717403], [37.11228348671518, 33.98390772842901], [35.673902947239185, 34.824733280238156], [34.207653666936274, 35.59863783830324], [32.7092760660557, 36.2953547391317], [31.176612727500416, 36.904998822562135], [29.609801280407797, 37.41872738959214], [28.01131624778007, 37.82937084766865], [26.385863088410062, 38.131998922123884], [24.740135103804246, 38.324391325637336], [23.082450643990473, 38.40738630158765], [21.422293819307566, 38.38508627583122], [19.769786446317347, 38.26490668878817], [18.135122024907915, 38.05746161276282], [16.52799403151529, 37.776287633390524], [14.957050667594032, 37.43741531890784], [13.429406448572962, 37.05880504840394], [11.950237758361656, 36.659670671058365], [10.522484900689369, 36.25972010836202], [9.146677485966961, 35.87834632564768], [7.820893486388201, 35.533804884662125], [6.540855294844877, 35.24241541287904], [5.300158978489034, 35.01782373194027], [4.090625974012855, 34.87035910022665], [2.9027600660538946, 34.80651714406897], [1.7262869324073156, 34.828593752411344], [0.550749098043001, 34.934488729483235], [-0.6338739717561552, 35.117690631509774], [-1.8365525404805274, 35.367446289102254], [-3.064632005374065, 35.66911039394901], [-4.323327519751736, 36.00466257245505], [-5.6153153261396955, 36.35337193752422], [-6.940443949356691, 36.692582535784666], [-8.295582900896102, 36.998587684835385], [-9.674620151797535, 37.24755716459276], [-11.068612639357465, 37.41647876637572], [-12.466086809953856, 37.48407491991044], [-13.853479004783678, 37.431656043522906], [-15.21569870289899, 37.243874851656784], [-16.53679156313527, 36.90934998791152], [-17.800674136446574, 36.42113284443762], [-18.991908289491455, 35.77699803401981], [-20.09648096864344, 34.97954540611806], [-21.102554055422548, 34.03610941498546], [-22.001149764100496, 32.95847971000889], [-22.786739282629387, 31.762444675383975], [-23.457706061677428, 30.467176960038906], [-24.01666015074388, 29.09448649837748], [-24.470586044901975, 27.667971857583847], [-24.830813373386164, 26.212104739809973], [-25.112807130008648, 24.751284960725524], [-25.33578169246875, 23.30890413026859], [-25.522150274141723, 21.906455558127934], [-25.69682837780262, 20.5627256469556], [-25.88641597891927, 19.293098338970978], [-26.118288295734267, 18.10899922552583], [-26.41962889087091, 17.017499945082495], [-26.816441337560914, 16.021096754170223], [-27.332576678425685, 15.117669957214028], [-27.98881337773919, 14.30062353759662], [-28.802024457326432, 13.559197156582092], [-29.784463113452844, 12.878935977328124], [-30.943193497096495, 12.242297799578553], [-32.27968771274045, 11.629371989645172], [-33.78960370092904, 11.018680844865077], [-35.46275179487822, 10.388031471742282], [-37.28325067402088, 9.71538505297166], [-39.229866471820344, 8.97971054012996], [-41.27652221441852, 8.161791286057884], [-43.39295882990678, 7.244955818351002], [-45.54552389980802, 6.215707698349146], [-47.698060305608735, 5.0642340134794175], [-49.812864084213466, 3.7847772893715024], [-51.85167922248367, 2.375861237424006], [-53.776696811490474, 0.840366521895771], [-55.55152690916365, -0.8145416088031812], [-57.142113537085365, -2.577626660243189], [-58.51756632860964, -4.433923255573747], [-59.65088627843028, -6.365192518079535], [-60.51956761651535, -8.350471248702243], [-61.10606282200521, -10.366693387071457], [-61.398102977916984, -12.389360925041602], [-61.38887082159823, -14.393241081455109], [-61.07702875909, -16.35306709341491], [-60.46660859740384, -18.244221341381824], [-59.56677365175849, -20.04338157925248], [-58.39146708659835, -21.72911363825115], [-56.95896277208185, -23.282396951642117], [-55.291336546461025, -24.68707243658349], [-53.41387657593104, -25.930205503789296], [-51.35445154266572, -27.002360090915968], [-49.14285474871543, -27.89778249726086], [-46.81014100600429, -28.6144963269337], [-44.38797151895393, -29.154311947136605], [-41.907979996379304, -29.52275549276275], [-39.4011710965934, -29.72892358724036], [-36.897360152189634, -29.78527062392815], [-34.42466106359755, -29.707335713184193], [-32.0090273980389, -29.51341632258574], [-29.673850162272316, -29.224195314863863], [-27.43961448361261, -28.86232762397537], [-25.32361655322002, -28.45199231150621], [-23.339741646708593, -28.01841531568411], [-21.498303798995497, -27.58736793388824], [-19.805947707255992, -27.184646038096336], [-18.265613582482715, -26.835535258427363], [-16.87656586810165, -26.564267902001365], [-15.634486889446752, -26.39347819183567], [-14.531636489019215, -26.34366347227851], [-13.557078447792579, -26.432660263991238], [-12.696973918103495, -26.675145368463618], [-11.934941148151717, -27.082173505908113], [-11.252479439511909, -27.660764095043493], [-10.629453556650612, -28.413550618014316], [-10.044632743495757, -29.338506431575347], [-9.476276171103374, -30.4287607722028], [-8.90275414635492, -31.672517964159496], [-8.303191883159387, -33.05309141003869], [-7.658120221992822, -34.54906179150569], [-6.9501155385695235, -36.13456604117504], [-6.1644093675959875, -37.77972011338906], [-5.289447134951489, -39.451174472311614], [-4.317374976348218, -41.11279666005423], [-3.2444340317256515, -42.726470470810504], [-2.0712429176275755, -44.252996333576135], [-0.8029513294928126, -45.65307271127734], [0.5507490980429989, -46.88833388450373], [1.9757664999005844, -47.92241563045697], [3.4539080775066466, -48.722017248197965], [4.963336583076504, -49.25792631362831], [6.479177749927955, -49.50597163277862], [7.974266391866445, -49.447870218337336], [9.42001291345694, -49.07193580996717], [10.787366309569471, -48.3736195056932], [12.047844652101974, -47.35585742186016], [13.174599818659418, -46.02920584510685], [14.143480043349546, -44.41175091586575], [14.934051959665188, -42.52878727061975], [15.530543309631836, -40.412268005968116], [15.922668507025948, -38.10003651287142], [16.10630179884772, -35.634858842700865], [16.083966835907212, -33.06328297687436], [15.865116940922718, -30.434358352819643], [15.46618708640091, -27.798254944144116], [14.910406340060662, -25.204825829235826], [14.227368025642317, -22.70216028370318], [13.452363762718091, -20.335175829587694], [12.625496542532204, -18.144297266445214], [11.790596705088712, -16.16426846740976], [10.993972743448207, -14.423138694434751], [10.283035928700373, -12.941459493910633], [9.704843509484379, -11.731721071684529], [9.304609425712234, -10.798048675120562], [9.124233878842166, -10.136170244338546], [9.200903581600768, -9.733656792818532], [9.565813006353308, -9.57042702472232], [10.243053482466047, -9.619497989899319], [11.248711660083542, -9.84795450983817], [12.590211841029777, -10.21810204900637], [14.265928230164949, -10.688760988128514], [16.26508359885129, -11.216655158175877], [18.567940543276595, -11.757844230002979], [21.14628086710199, -12.26914826333449], [23.96415804220242, -12.709513457122446], [26.978897626484688, -13.04127088331951], [30.142311350646896, -13.231244615892743], [33.40208269896334, -13.25167199470005], [36.703275525246724, -13.08090652605619], [39.98991282226694, -12.703882794021776], [43.20657038359646, -12.112332367292277], [46.29992985971352, -11.304749633673204], [49.220237625286735, -10.286116361512077], [51.92261986264518, -9.067403163549134], [54.36821016203676, -7.664874534197942], [56.525053497532305, -6.099231396291537], [58.368759344462674, -4.394630832344362]]}} +{"image": "images/contour_silhouette_count_00001.png", "question": "This image has two panels separated by a thin vertical divider. The left panel shows the Template: a single closed silhouette. The right panel shows the Field: many closed silhouettes scattered across the canvas, all at the SAME pixel scale as the Template. Count the number of Field silhouettes that are exact copies of the Template (translation only \u2014 same size, same orientation, no rotation or mirroring) and report the integer count. Provide your final answer enclosed in ... tags.", "answer": "8", "num_matches": 8, "match_labels": ["C", "E", "F", "G", "H", "M", "Q", "T"], "total_blobs": 20, "metadata": {"seed": 1806341205, "template_points": [[48.45124596599337, 1.3161463304618664], [49.93759942799447, 2.7548766253845485], [51.18724928873264, 4.2750728442749], [52.15985647968932, 5.853803371616253], [52.821110994488755, 7.4639215534016445], [53.14395705066804, 9.075003772281857], [53.10961863254095, 10.654460937314381], [52.70838289962222, 12.168787089642382], [51.9401062723717, 13.584901657057848], [50.81441698697131, 14.87153616956397], [49.35059824325595, 16.00061228419878], [47.57714738551271, 16.948556006828728], [45.53101843004472, 17.69749321090065], [43.256567222944, 18.23627402039449], [40.80423009355578, 18.561278334203617], [38.22897758154273, 18.676961612322163], [35.58859419796207, 18.59610881504821], [32.94184281545347, 18.339774789261636], [30.346577813326057, 17.936901054899668], [27.857874252649246, 17.423611417984606], [25.526240939818425, 16.842201632956403], [23.39598317295203, 16.23985093662667], [21.50377628133603, 15.667095151595955], [19.877503901655988, 15.176111696733784], [18.535405530318123, 14.818875772639952], [17.485566591480346, 14.645253796712867], [16.72577149376681, 14.701104510210897], [16.24372641205197, 15.026459828300608], [16.017644369857816, 15.653856320960088], [16.017171184578423, 16.60688418081416], [16.204617545704828, 17.899013755376853], [16.536450476724927, 19.53275041442528], [16.96498718924049, 21.499157016675834], [17.440226309263473, 23.777769959610875], [17.911745988668823, 26.336920249472236], [18.33059575338461, 29.134455783344734], [18.651109212489285, 32.11884569795172], [18.832567965110268, 35.23063282896505], [18.840653077406696, 38.404186645587636], [18.64862912285875, 41.56969704397828], [18.23821663806897, 44.65533960222231], [17.600121499520128, 47.58953573588924], [16.734203652601366, 50.30322695957719], [15.649282245187418, 52.73208135077107], [14.362588925258983, 54.81855239488717], [12.898895241302547, 56.5137156002689], [11.28935314245935, 57.778816415337474], [9.570098966287897, 58.58647374113747], [7.780680549694777, 58.92149628572509], [5.9623738183724955, 58.78128363477843], [4.156459125229339, 58.17579962748132], [2.402528562165642, 57.12712179424229], [0.7368934325050724, 55.66858658001116], [-0.8088438578621178, 53.8435651990302], [-2.2089968312361385, 51.70391863459751], [-3.444734162313229, 49.30819196302153], [-4.504716193828755, 46.719617381170885], [-5.385407046322061, 44.00400169273629], [-6.091054632590596, 41.22757731952651], [-6.633345090792434, 38.45489603984955], [-7.0307519353899535, 35.746841638634784], [-7.307612999411117, 33.15883163763882], [-7.4929794588331395, 30.739269537726944], [-7.619290414845179, 28.5282979398426], [-7.7209332781286255, 26.55689000263316], [-7.832754274701711, 24.84630250190969], [-7.988584618446914, 23.407898888142135], [-8.21984624028746, 22.243335823347344], [-8.554296524014177, 21.34509234334178], [-9.01496449190757, 20.697307628814663], [-9.61932163928163, 20.27688191510934], [-10.378719562322122, 20.054785782061074], [-11.298114161857558, 19.997516298737207], [-12.376083095357703, 20.06863449688241], [-13.605129878430333, 20.230317531661605], [-14.972255196616466, 20.444860652307682], [-16.459764146617527, 20.676068616615282], [-18.04626780335652, 20.89048319186601], [-19.707829154639768, 21.058402534997324], [-21.419197417227817, 21.154659092087872], [-23.155071299449975, 21.159134688903716], [-24.89133104206947, 21.057004143131437], [-26.60618106493557, 20.838711438547705], [-28.2811496641933, 20.499694691867404], [-29.901899219332357, 20.039887276433113], [-31.458809451464944, 19.463032060866812], [-32.94730700406547, 18.775853369360956], [-34.36792650426381, 17.98713666118498], [-35.72610076741054, 17.10676885451063], [-37.031690366632276, 16.14479259255178], [-38.298274840188746, 15.11052559415435], [-39.54223881651197, 14.011791686753432], [-40.78169581359384, 12.854303434195096], [-42.03530000075769, 11.641227786726025], [-43.32100147076072, 10.372956314576962], [-44.65480333547853, 9.047090816835116], [-46.04957911799182, 7.658643935969525], [-47.514006472669884, 6.200443379454988], [-49.0516683426471, 4.663717965293082], [-50.66036548936527, 3.038834444451833], [-52.33167522761792, 1.3161463304618481], [-54.05078058003975, -0.5130898701650408], [-55.796582398984924, -2.4557793278783198], [-57.542094804983414, -4.516084159459128], [-59.255112092191084, -6.694598047725981], [-60.899123580487554, -8.98764572636971], [-62.434442251284366, -11.386747482038533], [-63.81950384031778, -13.87828159277374], [-65.01228575760798, -16.44336885685929], [-65.97179005963312, -19.05799347332881], [-66.65953191228418, -21.693363965879268], [-67.0409746519436, -24.316507072888495], [-67.08685466584879, -26.891077038990968], [-66.7743437566557, -29.37835300216489], [-66.08800321647541, -31.73838860050351], [-65.02049221185287, -33.9312708942192], [-63.57300289955255, -35.91844050817554], [-61.75540552458838, -37.66402176110545], [-59.58609813190781, -39.13611057973017], [-57.09156697183977, -40.307969222743786], [-54.30567472336254, -41.15908018848189], [-51.26870385156222, -41.67601698722094], [-48.02619135484588, -41.85309647819153], [-44.6275985038276, -41.69278588623204], [-41.12486466262849, -41.205847052497134], [-37.570897734614796, -40.411210529382295], [-34.01805509861253, -39.33558237487005], [-30.51666810047483, -38.01279650950533], [-27.113660329767463, -36.48293486313299], [-23.85130521614485, -34.79124588810454], [-20.766162171183073, -32.98689903267002], [-17.888222885088698, -31.1216181991815], [-15.240290814269713, -29.248240877050865], [-12.83760774327217, -27.419251440409774], [-10.687731961247962, -25.685337017190154], [-8.790663439907053, -24.09401243193762], [-7.139202794233483, -22.688357135809206], [-5.719523068174008, -21.50590196666386], [-4.511926783339589, -20.577697287451006], [-3.4917554273794327, -19.927586831076535], [-2.63041478107135, -19.571703759844084], [-1.896477260285753, -19.51819736362373], [-1.2568217820028114, -19.767190804708846], [-0.677772487406691, -20.310962681316443], [-0.12619984483979785, -21.134338205941678], [0.4294489646780115, -22.215269715928383], [1.0182191221903558, -23.525581236319212], [1.6658403627636145, -25.03184802626912], [2.3940603819927064, -26.696379526684858], [3.220131780043888, -28.47827289496686], [4.1564591252293415, -30.334504312817632], [5.210407237306676, -32.22102638434413], [6.384266762325124, -34.09384206086236], [7.675368691268394, -35.91002845995573], [9.076335775682232, -37.6286874917966], [10.575455880801442, -39.21180415772569], [12.157160210865804, -40.624997538055815], [13.802588020286258, -41.838153643449296], [15.490218828627494, -42.82593329384], [17.196553197050577, -43.56815186663725], [18.896823686751503, -44.050031012496426], [20.565718580762063, -44.262325200894836], [22.178102180236156, -44.20132819661562], [23.709716861322093, -43.86876628686991], [25.137853488739324, -43.27158631566766], [26.441978137486036, -42.4216474036278], [27.60430431056219, -41.33532572495038], [28.61030092288961, -40.03304197855734], [29.449127244304922, -38.53872133106878], [30.113986780889814, -36.87919572400359], [30.602392772801213, -35.08355861203627], [30.91633866740361, -33.18248249881328], [31.062367671926474, -31.207510101606122], [31.051536389161576, -29.190330616490503], [30.89926867969388, -27.1620533514473], [30.625097350916775, -25.152491895225523], [30.252293104036156, -23.189472917073694], [29.80738240720155, -21.29818454620661], [29.319558606437226, -19.500579944055964], [28.81999360205048, -17.81485203327403], [28.341060736259486, -16.25499526281847], [27.91548305213966, -14.830469656766102], [27.57542465669624, -13.545981124328259], [27.351546387401385, -12.401390036953545], [27.27205015831516, -11.391757378162543], [27.361739055654038, -10.507534356731805], [27.641122270889642, -9.734897302187612], [28.12559512172865, -9.056225036507138], [28.82472456057828, -8.450710883849487], [29.74166958346189, -7.895096226119084], [30.872763750717027, -7.3645072529173605], [32.207283586752865, -6.833371529083299], [33.727421969528166, -6.276386462009571], [35.40847984126795, -5.6695079436230795], [37.21928282054986, -4.990924603574296], [39.12282177952259, -4.221981448723112], [41.07710842687085, -3.3480163476412255], [43.03622870705945, -2.3590739648587444], [44.951568719327355, -1.2504644152269802], [46.77318022199566, -0.023138084620409893]]}} +{"image": "images/contour_silhouette_count_00002.png", "question": "This image has two panels separated by a thin vertical divider. The left panel shows the Template: a single closed silhouette. The right panel shows the Field: many closed silhouettes scattered across the canvas, all at the SAME pixel scale as the Template. Count the number of Field silhouettes that are exact copies of the Template (translation only \u2014 same size, same orientation, no rotation or mirroring) and report the integer count. Provide your final answer enclosed in ... tags.", "answer": "10", "num_matches": 10, "match_labels": ["C", "D", "G", "I", "J", "L", "M", "N", "Q", "V"], "total_blobs": 22, "metadata": {"seed": 173879092, "template_points": [[43.26594215961219, -1.1508314185001145], [44.7217786121908, 0.1840500569756689], [45.934566987178876, 1.597873001219399], [46.855118751048884, 3.066051105945041], [47.44422793768217, 4.559138572971672], [47.674339646594106, 6.044443564303187], [47.53073356250932, 7.487853499031258], [47.012170732079774, 8.85578074399496], [46.130975565868546, 10.117130517652557], [44.91254996893097, 11.245190889059268], [43.39434150108849, 12.219347704412419], [41.62431138979985, 13.026534974760153], [39.658969980687516, 13.662343343920364], [37.56106583456309, 14.13172514852781], [35.39702933650639, 14.449253522981925], [33.23428173124748, 14.638914085717985], [31.138525515328393, 14.73342996240404], [29.171131910748894, 14.773143195496226], [27.386735769665133, 14.804496894462979], [25.83113800587024, 14.878181781476421], [24.539601029589804, 15.047027164633185], [23.53560438791215, 15.363729046246284], [22.830106756474752, 15.878516445988822], [22.421337581383597, 16.63686069155736], [22.29511809877534, 17.677331230622816], [22.425688249868827, 19.029695505952404], [22.776994219436613, 20.713349897254265], [23.304371934326248, 22.736154168385568], [23.956545721048954, 25.093723955743183], [24.67784912975946, 27.769215436918678], [25.410567184705545, 30.733614386602582], [26.09729630191039, 33.94651938631751], [26.683219880993853, 37.357387052239794], [27.11820396120712, 40.90718679702557], [27.358627948340953, 44.53039478441827], [27.36887969137884, 48.15724218389064], [27.122461373706, 51.71612224020438], [26.60267191501773, 55.13605450206573], [25.802851905374453, 58.349103055682846], [24.726197524517186, 61.292648811233164], [23.385169461946134, 63.911423599048625], [21.800540629749186, 66.15922564234363], [20.000141636811374, 68.00025128872615], [18.017374896792195, 69.40999594723593], [15.889576371878322, 70.37569711195665], [13.656308002521996, 70.89631319825506], [11.357663739262888, 70.98205268124987], [9.032667883922906, 70.65348773099323], [6.717836471926093, 69.94030426974147], [4.445961175552011, 68.87975532380645], [2.2451613254163014, 67.51489603972028], [0.13823390141577319, 65.89268629035294], [-1.8576854170129002, 64.06205011654765], [-3.7311539425134295, 62.07198024841546], [-5.476166357243979, 59.969770753655624], [-7.091646288694745, 57.79945179732298], [-8.580731858770516, 55.60048808252907], [-9.949911334784783, 53.406787446788414], [-11.20807101358051, 51.24604910951906], [-12.36551985958566, 49.13946308079284], [-13.4330541253158, 47.10175416650478], [-14.421120334721, 45.14154674997471], [-15.339126919078218, 43.26201095123055], [-16.19494394491763, 41.46173762448007], [-16.994617390690216, 39.73577958083386], [-17.74231005112863, 38.0767898835905], [-18.440466184940195, 36.47618533860982], [-19.090182311387807, 34.925264475776], [-19.69175292870677, 33.416214271750114], [-20.245348138697263, 31.94294828796912], [-20.751770885804316, 30.501730303325132], [-21.213235290927287, 29.09155126942314], [-21.63410475378071, 27.714242751428905], [-22.02152930460588, 26.37432610452243], [-22.385926105220765, 25.078612604711847], [-22.741254835549025, 23.835584742465226], [-23.10505057438687, 22.654602091133466], [-23.498190144863695, 21.544985867438974], [-23.944383058561407, 20.515043928165642], [-24.46939436478462, 19.571102077182598], [-25.100023034690242, 18.716607955264365], [-25.86287510368105, 17.951370413183874], [-26.782984802695676, 17.270990286331198], [-27.882348540474887, 16.66652824795923], [-29.17844517424297, 16.12444244336254], [-30.682820990482533, 15.626813571952834], [-32.399818850169666, 15.151858775437198], [-34.32552786899404, 14.674718969541631], [-36.44702284445046, 14.168488015909938], [-38.741951657766144, 13.60543724688427], [-41.178514517374076, 12.958376145655516], [-43.715861799157715, 12.202080162065297], [-46.304918154774214, 11.314710284397464], [-48.88962039525441, 10.27914649323483], [-51.40853637713497, 9.084158805012414], [-53.796812712641, 7.725345274199766], [-55.98838156139236, 6.2057758591051995], [-57.918341935973785, 4.536294060946755], [-59.52541964953709, 2.7354441296036924], [-60.75440287752167, 0.8290096467565922], [-61.55844773701337, -1.150831418500135], [-61.90115053394438, -3.1667104806921635], [-61.75829039224654, -5.177587286401909], [-61.11915763309967, -7.140523063800909], [-59.987399068077636, -9.012632092120096], [-58.38133065579565, -10.753124419784378], [-56.33368992146548, -12.325344928086635], [-53.890824188767, -13.698710824450817], [-51.11133496614194, -14.850451208294736], [-48.06422266263919, -15.767058612668887], [-44.82659808604136, -16.445373187619406], [-41.48104686067777, -16.89323503637101], [-38.11274906414352, -17.129658518275424], [-34.80646824345283, -17.184503291187134], [-31.643530944397725, -17.097639537668343], [-28.698919604827932, -16.91762816367481], [-26.03859799333722, -16.69995968548549], [-23.717179435919576, -16.504916943225464], [-21.776034223845944, -16.395145664143552], [-20.241914423426138, -16.43303231653521], [-19.126152601654592, -16.677999865870397], [-18.424466696836138, -17.18383837637828], [-18.117377479257925, -17.99618852038303], [-18.171218916773647, -19.150291828771262], [-18.539696457510473, -20.66911205119555], [-19.165924906967856, -22.561917650884293], [-19.98485726321084, -24.82339682379762], [-20.92599949972674, -27.43335429841409], [-21.916294590320142, -30.357014504201352], [-22.88305258042099, -33.54592958453905], [-23.756802512525972, -36.939464341789105], [-24.4739465464739, -40.46680473056019], [-24.979106460275613, -44.049413121017686], [-25.227067412657014, -47.60383331318532], [-25.18424270996441, -51.044732135801674], [-24.829605472260624, -54.28805316972289], [-24.15505751064899, -57.254152240161304], [-23.165231281946948, -59.870784125738226], [-21.87674629974074, -62.075815479127385], [-20.316965679630847, -63.81955002661925], [-18.5223204679828, -65.06656824263135], [-16.53628804635037, -65.79700417869108], [-14.407125374972697, -66.00720606308383], [-12.18546749302106, -65.70975361599798], [-9.92190611275383, -64.93283257317239], [-7.66466215824652, -63.718994448638554], [-5.457459789910894, -62.123355862710525], [-3.337698156314867, -60.21131563867349], [-1.3350013910475749, -58.05588826367015], [0.5297920030518009, -55.73476830741447], [2.2451613254163028, -53.32725128795307], [3.808942077573817, -50.9111417907318], [5.2278794976692335, -48.55977916398585], [6.516729715329303, -46.33930487254033], [7.696962813715034, -44.30628389802331], [8.795141074177987, -42.505775978335045], [9.841060958090456, -40.96993175668801], [10.865758233549391, -39.71716501978371], [11.89948157907138, -38.75192624562779], [12.96974072686976, -38.065075843584545], [14.09953072927966, -37.63482897636439], [15.305824477799998, -37.428218907235745], [16.59841164788605, -37.40300353952519], [17.97914447235492, -37.50992120355892], [19.441630032648685, -37.69518761768171], [20.971386111282918, -37.90311691169279], [22.546454173576716, -38.07874602037925], [24.138439884290044, -38.170343743933515], [25.713929841554073, -38.13169318390383], [27.236213983416665, -37.9240487000515], [28.66722732828218, -37.517685366340686], [29.969613121405597, -36.892979304389], [31.108802645432366, -36.040980251102205], [32.05500525456727, -34.96346217217479], [32.78500570455252, -33.67246250263121], [33.28367441921657, -32.18934451230032], [33.54510956092505, -30.543439240494713], [33.573347034675464, -28.77034240478973], [33.38259503706749, -26.90995679772674], [32.99697248937052, -25.004381266046146], [32.44975458622891, -23.095752963584555], [31.782152607880775, -21.2241499657269], [31.041677939899994, -19.425656564102123], [30.280160826707274, -17.73068390504858], [29.55151176229052, -16.16262460557779], [28.909326752353255, -14.736902293053946], [28.40444631491157, -13.460456556608376], [28.08258159057142, -12.331681583219455], [27.98211912026919, -11.340813870727203], [28.132208778812597, -10.47074197996387], [28.55122733683688, -9.698190396013649], [29.245693705131245, -8.995211220867011], [30.209691845203995, -8.330902494153005], [31.424834532816742, -7.673261142638503], [32.8607766908469, -6.991072396655514], [34.476261997952456, -6.255736259563515], [36.22066209048607, -5.442935312546634], [38.03594503949728, -4.534056579198947], [39.858989954087185, -3.5172929308087433], [41.624148458321216, -2.3883659446564884]]}} +{"image": "images/contour_silhouette_count_00003.png", "question": "This image has two panels separated by a thin vertical divider. The left panel shows the Template: a single closed silhouette. The right panel shows the Field: many closed silhouettes scattered across the canvas, all at the SAME pixel scale as the Template. Count the number of Field silhouettes that are exact copies of the Template (translation only \u2014 same size, same orientation, no rotation or mirroring) and report the integer count. Provide your final answer enclosed in ... tags.", "answer": "12", "num_matches": 12, "match_labels": ["A", "C", "E", "H", "I", "J", "N", "O", "P", "U", "W", "X"], "total_blobs": 24, "metadata": {"seed": 1112038970, "template_points": [[43.62552417336153, -2.0987071309275165], [43.2748751825859, -0.6010273974338005], [42.843846396244594, 0.8724954754287668], [42.34317338194335, 2.318128472579693], [41.78418689647445, 3.7334469873984024], [41.17855970944056, 5.117387179125531], [40.53804550737015, 6.470262551840861], [39.87421575997371, 7.793744026902713], [39.19820050332733, 9.090803652527532], [38.520438931344884, 10.365622967556519], [37.85044548793271, 11.623467890806264], [37.19659682190351, 12.870532822521671], [36.565944512322446, 14.113757400991338], [35.96405790364632, 15.360620037318375], [35.39490072065109, 16.618912938384344], [34.860744377952805, 17.896503808059606], [34.362120075172434, 19.201089778038472], [33.89781089534952, 20.53994935340471], [33.46488422113136, 21.91969825823109], [33.05876387131837, 23.34605503036807], [32.67334046052015, 24.82362204245329], [32.30111761768952, 26.355687321629304], [31.9333908851158, 27.94405211014879], [31.560455376812968, 29.58888856257919], [31.171837621210244, 31.288631325024156], [30.75654646265499, 33.039906002422306], [30.30333746207407, 34.8374967084259], [29.80098492910411, 36.67435402715317], [29.23855554305458, 38.54164381706415], [28.605677482076032, 40.42883637493793], [27.89279907955457, 42.32383457335006], [27.09143126153551, 44.213138708901624], [26.194368383261178, 46.08204497081411], [25.195882568067887, 47.91487367931681], [24.091887246521605, 49.695222767881454], [22.88006628387403, 51.40624140818748], [21.559965853607565, 53.03091821477748], [20.133047046137733, 54.552378128094986], [18.602698075416466, 55.95418186752211], [16.974205842081755, 57.22062177463673], [15.254687509345505, 58.33700793253995], [13.452983626469273, 59.289938647942364], [11.57951517444656, 60.06754971381507], [9.646107690387034, 60.65973732389186], [7.665786333522633, 61.058350075446356], [5.652546371012213, 61.25734616032832], [3.6211040723031713, 61.252912590783794], [1.5866333957214263, 61.043544118823284], [-0.4355058780291352, 60.630080367137985], [-2.4300427689682156, 60.01570057611244], [-4.382066670694837, 59.20587626514868], [-6.277290186416771, 58.208282987070206], [-8.10229778859137, 57.032673202022224], [-9.844775229583846, 55.69071309313762], [-11.493715108116536, 54.19578687274174], [-13.039594571990179, 52.5627727692174], [-14.474521797348288, 50.8077954271367], [-15.792348612519906, 48.947959885594514], [-16.988747412657794, 47.001072613214596], [-18.06125132140718, 44.98535526725143], [-19.009257378543904, 42.91915690574241], [-19.83399334864684, 40.82067031594362], [-20.538449536507127, 38.70765793244075], [-21.12727774201385, 36.597192510372956], [-21.606660173807523, 34.50541730184773], [-21.984151751795583, 32.44732996803632], [-22.2684997503871, 30.436593858977403], [-22.469445156012302, 28.485379622976904], [-22.597510425650576, 26.60423938432332], [-22.66377853190379, 24.802014969537392], [-22.679668261641552, 23.085780886836144], [-22.65671069929267, 21.460821989378175], [-22.606331675179845, 19.93064499833152], [-22.53964469934961, 18.497022344313212], [-22.467258540176395, 17.160066121591672], [-22.399103155012533, 15.918329353390194], [-22.344277149791324, 14.768931251594053], [-22.31091934999432, 13.707702730883495], [-22.30610642237548, 12.729348114133337], [-22.33577781188846, 11.827618748599548], [-22.40468856854576, 10.995494144021881], [-22.51638995173856, 10.225366244646493], [-22.673237031855987, 9.50922255493292], [-22.876421877134018, 8.83882404835684], [-23.12603033269278, 8.205874092788088], [-23.421119882357306, 7.602175014693149], [-23.759815643958948, 7.019769386187955], [-24.139421195146678, 6.451063640384815], [-24.556540666749623, 5.888932186871861], [-25.007208379386626, 5.326800794940796], [-25.487022238699435, 4.758708621237519], [-25.991277145051868, 4.179348864645837], [-26.515094811963703, 3.5840886185411542], [-27.05354661861351, 2.9689690439039826], [-27.60176643779792, 2.3306874920994756], [-28.155050772007463, 1.6665636508029769], [-28.708943985153375, 0.9744921597427574], [-29.259306922812332, 0.252884435831088], [-29.802367755315252, -0.4993973467010364], [-30.334754440469407, -1.2831110605345564], [-30.853508770640754, -2.098707130927525], [-31.356082526812177, -2.9463997294794364], [-31.840316794963762, -3.826233800863383], [-32.30440599337779, -4.738145308552728], [-32.74684860011728, -5.682012401830857], [-33.16638694632642, -6.657695587905637], [-33.56193874332939, -7.665065428215199], [-33.932523231984064, -8.70401675407748], [-34.277184975858745, -9.774468899724017], [-34.59491836240142, -10.876351965826599], [-34.88459582768609, -12.009579638952497], [-35.144902682338135, -13.174009587171021], [-35.37428119307052, -14.36939291500356], [-35.57088627240555, -15.595314578671728], [-35.73255475724527, -16.85112702302838], [-35.85678982550375, -18.135879594094252], [-35.940761621155566, -19.448246497075793], [-35.981324645206975, -20.786456202500606], [-35.97505193760984, -22.148225248339752], [-35.91828553791713, -23.530699341746864], [-35.807202185588785, -24.930404530829634], [-35.637892719120586, -26.343210997622172], [-35.40645317077778, -27.76431172345336], [-35.10908514385279, -29.188217904750754], [-34.74220271385173, -30.608772560574405], [-34.302542823993576, -32.01918328422628], [-33.78727595706773, -33.41207456302882], [-33.194113766061285, -34.77955953688171], [-32.52141033870827, -36.11333050236926], [-31.768253857454496, -37.40476691030981], [-30.934545595001854, -38.64505906603557], [-30.021063452849006, -39.82534523831941], [-29.029507599943365, -40.93685942889467], [-27.962526192319345, -41.971086662971004], [-26.82371964200137, -42.9199223435592], [-25.617622442286788, -43.77583197849442], [-24.34966213312325, -44.532007446430896], [-23.026095589838867, -45.182515922110085], [-21.653923425413442, -45.72243763477974], [-20.24078389487206, -46.14798878710166], [-18.794828264367514, -46.45662621300569], [-17.324580141692373, -46.64713069694665], [-15.838781744777979, -46.71966630668015], [-14.346230496896968, -46.675813597488556], [-12.855609670085993, -46.51857511619239], [-11.375317041947863, -46.252352254928134], [-9.91329567786194, -45.882893162735996], [-8.476870995498807, -45.417212101572254], [-7.072598208713493, -44.863481315799206], [-5.706124083336148, -44.230897153593325], [-4.382066670694835, -43.529522818237524], [-3.1039163211268335, -42.770110720676925], [-1.8739608269984398, -41.96390793672014], [-0.6932370129360792, -41.12244872891768], [0.4384895077408748, -40.257338462214726], [1.5227153430052525, -39.38003351375239], [2.5621607201614154, -38.50162194279041], [3.5607016034025225, -37.63260974125951], [4.523273575683867, -36.78271742623022], [5.455749228962639, -35.96069156266392], [6.364791474595681, -35.17413552104591], [7.257685805067637, -34.429363385479085], [8.142155093409762, -33.73128044175679], [9.026160993241234, -33.08329310249801], [9.917696387674344, -32.48725048047929], [10.824573618841352, -31.943419116609057], [11.754213403369285, -31.450491621886123], [12.713439397210653, -31.00562922066798], [13.708283313049506, -30.604537403896558], [14.743805315125506, -30.24157313415507], [15.823934122704657, -29.909881307988712], [16.951330850360364, -29.601557492654635], [18.12728010920941, -29.307833331192906], [19.35161129929074, -29.019280466804155], [20.622652352685602, -28.726028388539433], [21.93721745502001, -28.417991256642317], [23.290629496554892, -28.085098536471126], [24.67677720126049, -27.71752416102243], [26.08820607198355, -27.305908957075133], [27.516241491285406, -26.841571209322225], [28.95114154986385, -26.31670049802126], [30.382276456198234, -25.724530323200657], [31.798330729665956, -25.05948551401258], [33.187523810890845, -24.317301004490535], [34.53784425167361, -23.495109223427768], [35.83729228447167, -22.591494080936474], [37.074125327488495, -21.606510320349717], [38.23710086269329, -20.54166782304145], [39.31571113427845, -19.399881286135837], [40.300404254904294, -18.185386519200257], [41.18278657422668, -16.903625406164334], [41.95580155327594, -15.561102333695471], [42.61388089095425, -14.165215578889063], [43.153064254202995, -12.724067760625427], [43.571084657738346, -11.246259975356963], [43.867417306948646, -9.740674646674227], [44.04329054107557, -8.21625240852121], [44.10165837422217, -6.681768506916146], [44.047135009156584, -5.145614239975258], [43.88589257290602, -3.61558885948655]]}} +{"image": "images/contour_silhouette_count_00004.png", "question": "This image has two panels separated by a thin vertical divider. The left panel shows the Template: a single closed silhouette. The right panel shows the Field: many closed silhouettes scattered across the canvas, all at the SAME pixel scale as the Template. Count the number of Field silhouettes that are exact copies of the Template (translation only \u2014 same size, same orientation, no rotation or mirroring) and report the integer count. Provide your final answer enclosed in ... tags.", "answer": "15", "num_matches": 15, "match_labels": ["A", "AA", "B", "C", "F", "G", "I", "K", "M", "N", "R", "T", "U", "Y", "Z"], "total_blobs": 27, "metadata": {"seed": 2087043557, "template_points": [[44.44277960010965, -1.5135772298455548], [44.97757930029059, -0.2440484942287342], [45.37233049140971, 1.0528259592544305], [45.61693668353232, 2.3655062466698427], [45.70405171643831, 3.6815329821615146], [45.62920720230568, 4.987896132239603], [45.39086727675366, 6.271412885526659], [44.99041173498287, 7.519103762969489], [44.432050499270076, 8.718556718100722], [43.722674057189735, 9.858269778852979], [42.87164598361344, 10.927963820842383], [41.89054488019117, 11.918858290243424], [40.79286401023389, 12.823904061749165], [39.593677561820016, 13.63796907111185], [38.30928283513834, 14.357973849931984], [36.95682772921209, 14.982975562252019], [35.553932715135865, 15.514200551414596], [34.11831605287599, 15.95502671017379], [32.66743036841311, 16.310918152398415], [31.218117894652263, 16.58931566363479], [29.78629073374007, 16.79948722125561], [28.38664146262484, 16.952343492342532], [27.032388320305536, 17.060223636677524], [25.73505812499355, 17.136656969100002], [24.504309009902755, 17.19610608316108], [23.347794070567975, 17.25369692590135], [22.271066111792468, 17.324941066198065], [21.277522889392785, 17.42545504466127], [20.368391574712216, 17.570681261879525], [19.542750635130066, 17.775614384952686], [18.797586921197553, 18.05453675985827], [18.127885473716304, 18.4207658371472], [17.52674939934268, 18.886416175052428], [16.985547093595716, 19.462178196965635], [16.494084094240034, 20.157115563567142], [16.04079690231743, 20.97848278181372], [15.612966188067784, 21.931564515348754], [15.196946880531282, 23.019537979311302], [14.778412700519826, 24.24335978672414], [14.342612717756158, 25.601678648141004], [13.874637479393797, 27.090775391211285], [13.359692159046055, 28.704531839244613], [12.783374008817873, 30.434430142780638], [12.131951163702304, 32.26958417012093], [11.392639556285292, 34.196804507216804], [10.553874364111309, 36.20069847205164], [9.605572051757857, 38.2638062952421], [8.539378708620122, 40.36677424347513], [7.348900049017654, 42.488564958007956], [6.0299081630428075, 44.606704645906916], [4.580519914821062, 46.69756600331179], [3.0013418089627173, 48.73668488147073], [1.29557621298949, 50.69910774848937], [-0.5309159434453452, 52.55976598028303], [-2.469600454830715, 54.293871966616244], [-4.509306961815716, 55.87733098065381], [-6.636320948315171, 57.28716177484654], [-8.834532647824776, 58.501917975787386], [-11.085643706784609, 59.50210159956524], [-13.3694311074943, 60.27055943907067], [-15.664066379165874, 60.79285272408182], [-17.946486561464774, 61.057590356873], [-20.19281177538708, 61.05671620666854], [-22.378802652045966, 60.78574142321933], [-24.480349325221344, 60.24391351122132], [-26.473982264676064, 59.43431499086814], [-28.33739397054308, 58.36388584221145], [-30.049959518695655, 57.04336556779326], [-31.593243192823838, 55.487152574038824], [-32.95147800453453, 53.713080621934246], [-34.112004823521445, 51.74211427739272], [-35.06565814112402, 49.597967539846415], [-35.80708618648496, 47.30665207676962], [-36.33499420668933, 44.89596367132335], [-36.65230119932973, 42.39491752818542], [-36.76620222321951, 39.83314490819503], [-36.688130572756386, 37.24026510872519], [-36.43361653368368, 34.64524801287697], [-36.02204208150547, 32.07578324432357], [-35.47629366781861, 29.55767234414188], [-34.82231808598208, 27.11426030184367], [-34.08858923223166, 24.765922209459088], [-33.30549629524533, 22.52961976413299], [-32.50466642999047, 20.418540835818153], [-31.718237217033117, 18.441833372206773], [-30.97809609864316, 16.604442577814584], [-30.315105448686214, 14.907057636785257], [-29.75833291606079, 13.34617132052444], [-29.334307136078326, 11.914252713136129], [-29.066318800434498, 10.600030089286413], [-28.973786400415978, 9.388877785407146], [-29.071704713147636, 8.263297813354576], [-29.370192308066347, 7.203484072395116], [-29.87415204889425, 6.187954413578732], [-30.583055810289753, 5.1942335862044695], [-31.490861488485077, 4.199568325449963], [-32.58606694538961, 3.181654587168145], [-33.851901880735326, 2.119356249455761], [-35.26665488006259, 0.9933945133725608], [-36.804129146133384, -0.21301223834528016], [-38.43421679809976, -1.5135772298455685], [-40.1235782253502, -2.9184601112266626], [-41.83640991540011, -4.433882925049929], [-43.53528153322503, -6.061859586182322], [-45.18202089743696, -7.800048080557205], [-46.73862394686966, -9.64173110831382], [-48.16816587309066, -11.575927197511241], [-49.435689345246224, -13.587630505184812], [-50.50904618879574, -15.65817371684751], [-51.359669993887614, -17.76570476900916], [-51.96325889652723, -19.88576466508951], [-52.30035015037597, -21.99195054067777], [-52.35677102424825, -24.056645457687832], [-52.12395393834833, -26.05179425442076], [-51.599107494574596, -27.949703220519933], [-50.78523905458284, -29.723840455331306], [-49.691028656975504, -31.34961353889662], [-48.33055822023135, -32.805101609684186], [-46.72290402782137, -34.071720093934935], [-44.89160431595091, -35.13479813875478], [-42.864017268408226, -35.98405121499333], [-40.67058776284405, -36.61393430771023], [-38.34404371736259, -37.023864515861504], [-35.91854478050869, -37.21830563856792], [-33.4288073349233, -37.20671132142759], [-30.909230308676772, -37.00332745334342], [-28.393046093524323, -36.62685861854349], [-25.91151996291998, -36.10000739575982], [-23.4932197931499, -35.448899036231374], [-21.163375667821477, -34.70240743087422], [-18.943346157916885, -33.891401191947296], [-16.850204802604793, -33.047931037176284], [-14.89645667041762, -32.20438140269498], [-13.08989096800731, -31.392610272486863], [-11.433571604112107, -30.643101564126003], [-9.925963533318473, -29.984154043100798], [-8.561188721657352, -29.44112966247649], [-7.3294018143560455, -29.03578247438533], [-6.217272158128261, -28.78568688871514], [-5.20855583829024, -28.703781135180424], [-4.284738922768757, -28.79803840709318], [-3.4257312320234172, -29.07127443176816], [-2.610588728462478, -29.52109623729855], [-1.8182420730872337, -30.139992789081827], [-1.0282090415403298, -30.91556407579733], [-0.22126931561948915, -31.830881256689914], [0.6199183622286606, -32.86496675862416], [1.5102746166786836, -33.993379842891365], [2.461848801682191, -35.18889024703358], [3.4834716158198336, -36.42222013027498], [4.580519914821064, -37.662832779672854], [5.754799057256123, -38.87974541588352], [7.004544392950542, -40.042343000101425], [8.32453977593039, -41.12117019374613], [9.706347370700115, -42.088679545145176], [11.138639633580786, -42.91991553758503], [12.6076212925737, -43.593116276231235], [14.097526511161238, -44.090217245781375], [15.591174280743928, -44.39724464962584], [17.070563503606643, -44.504589246028125], [18.51748824568083, -44.40715521986554], [19.91415327875394, -44.10438235692475], [21.243770298077287, -43.60014350702302], [22.49111607666075, -42.90252291960665], [23.643035266037902, -42.02348440365109], [24.688872521457213, -40.978441304031456], [25.62082104814021, -39.785742912113186], [26.434177452038487, -38.466094066430266], [27.127495840565373, -37.04192629360182], [27.702637355907374, -35.53673985154561], [28.16471463154778, -33.974436446819915], [28.521933936758586, -32.378662204877436], [28.785340911977084, -30.772179694241615], [28.96847870409253, -29.17628647909045], [29.086969897587807, -27.610295851821494], [29.158035829676198, -26.09109314470065], [29.19996861332896, -24.63277841651155], [29.23157242526122, -23.246403444370845], [29.271591317150722, -21.939807916996145], [29.33814096563915, -20.71755662129573], [29.448161395500208, -19.580976336749128], [29.616906813198568, -18.52828819606062], [29.857487313247123, -17.554828524281017], [30.180475419968364, -16.653348711489475], [30.593588267490674, -15.814382574534186], [31.10145377604295, -15.026667976496626], [31.705466535171382, -14.277608239171375], [32.40373634120966, -13.553758129157224], [33.19112954560022, -12.841318931448846], [34.05940063958818, -12.126627339229213], [34.99740891238052, -11.396623563282581], [35.991412650367494, -10.63928516344723], [37.02543126172345, -9.844014579729759], [38.08166396994583, -9.001970133518324], [39.14095236547051, -8.10633231289256], [40.18327316663501, -7.152499377646647], [41.18824703598766, -6.13820864338725], [42.135649227001394, -5.063582153383432], [43.005908187474624, -3.9310977471118447], [43.78057899403311, -2.7454887153839573]]}} diff --git a/data/annotations/counting_connected_components.jsonl b/data/annotations/counting_connected_components.jsonl new file mode 100644 index 0000000000000000000000000000000000000000..4d48489a44173050b1662afcaaec40919c7602cb --- /dev/null +++ b/data/annotations/counting_connected_components.jsonl @@ -0,0 +1,5 @@ +{"width": 1916, "height": 1916, "grid_rows": 100, "grid_cols": 100, "square_left": 191.0, "square_top": 191.0, "square_size": 1534, "num_components": 10, "num_dots": 85, "question": "How many connected components are there in the image? A connected component is a maximal group of anchor points such that any two are linked by a path of one or more strands (directly or through other anchors in the same group). Anchors not linked by any chain belong to different components, even if they appear visually close. Count every connected component and report the total. Provide your final answer enclosed in ... tags.", "answer": 10, "dots": [[75, 13], [88, 11], [71, 31], [51, 73], [30, 25], [73, 55], [34, 42], [25, 52], [47, 57], [32, 75], [36, 6], [9, 7], [72, 18], [67, 10], [25, 41], [43, 8], [34, 85], [93, 36], [9, 39], [20, 53], [88, 56], [15, 14], [14, 69], [58, 7], [6, 74], [82, 24], [17, 80], [55, 29], [82, 44], [43, 40], [63, 30], [17, 38], [88, 78], [65, 53], [42, 59], [16, 33], [57, 76], [40, 66], [41, 94], [73, 69], [36, 53], [7, 68], [55, 52], [78, 8], [55, 34], [66, 77], [18, 87], [44, 47], [23, 85], [76, 26], [31, 67], [80, 36], [15, 23], [56, 83], [65, 71], [24, 9], [46, 93], [77, 81], [68, 45], [88, 35], [46, 64], [18, 59], [94, 14], [53, 21], [6, 25], [26, 30], [61, 19], [9, 59], [68, 26], [85, 72], [42, 20], [60, 42], [83, 52], [77, 88], [42, 71], [92, 31], [71, 82], [47, 22], [26, 72], [5, 44], [30, 93], [17, 73], [63, 47], [84, 64], [13, 49]], "components": [[0, 1, 2, 13, 23, 27, 30, 43, 62, 63, 66, 68, 70, 77], [3, 8, 9, 29, 34, 36, 37, 40, 45, 47, 50, 53, 54, 60, 74, 78], [4, 6, 11, 14, 18, 21, 31, 35, 52, 55, 64, 65, 79, 84], [5, 12, 20, 25, 28, 33, 42, 44, 49, 51, 58, 71, 72, 82], [7, 16, 19, 22, 24, 26, 41, 46, 48, 61, 67, 80, 81], [10, 15], [17, 59, 75], [32, 39, 69, 83], [38, 56], [57, 73, 76]], "edges": [[22, 41], [66, 68], [44, 71], [36, 53], [9, 74], [7, 19], [59, 75], [33, 82], [27, 77], [38, 56], [50, 78], [3, 60], [26, 48], [18, 84], [14, 31], [63, 66], [26, 81], [58, 71], [27, 30], [28, 72], [27, 63], [0, 13], [69, 83], [29, 47], [34, 37], [0, 43], [3, 36], [22, 61], [1, 43], [28, 51], [21, 55], [24, 41], [58, 82], [13, 66], [31, 35], [8, 60], [37, 50], [41, 67], [2, 30], [13, 23], [18, 31], [8, 47], [48, 80], [79, 84], [5, 72], [8, 34], [34, 40], [36, 54], [73, 76], [1, 62], [36, 45], [12, 49], [49, 51], [35, 52], [33, 42], [37, 74], [17, 59], [5, 33], [16, 80], [70, 77], [22, 81], [19, 61], [21, 52], [32, 69], [25, 49], [20, 72], [57, 76], [6, 14], [46, 48], [4, 65], [52, 64], [35, 65], [10, 15], [11, 21], [39, 83], [40, 50], [24, 81], [2, 68], [23, 63], [18, 35], [42, 82], [18, 79], [42, 71], [23, 43], [18, 64], [6, 65], [20, 28], [53, 74], [3, 53]], "dot_radius": 1.5, "image": "images/counting_connected_components_00000.png"} +{"width": 1916, "height": 1916, "grid_rows": 100, "grid_cols": 100, "square_left": 191.0, "square_top": 191.0, "square_size": 1534, "num_components": 12, "num_dots": 95, "question": "How many connected components are there in the image? A connected component is a maximal group of anchor points such that any two are linked by a path of one or more strands (directly or through other anchors in the same group). Anchors not linked by any chain belong to different components, even if they appear visually close. Count every connected component and report the total. Provide your final answer enclosed in ... tags.", "answer": 12, "dots": [[65, 12], [86, 94], [74, 92], [20, 25], [81, 28], [35, 56], [17, 32], [77, 19], [67, 64], [37, 51], [81, 87], [80, 71], [43, 91], [18, 94], [31, 36], [68, 57], [6, 47], [93, 69], [6, 67], [79, 13], [10, 85], [56, 87], [51, 20], [8, 59], [39, 10], [63, 72], [31, 73], [92, 87], [43, 61], [72, 39], [70, 30], [6, 34], [37, 71], [49, 64], [32, 60], [91, 5], [42, 38], [84, 21], [70, 12], [83, 6], [29, 20], [89, 54], [37, 36], [15, 59], [68, 71], [46, 84], [10, 75], [78, 45], [28, 85], [60, 5], [48, 24], [84, 11], [20, 51], [25, 31], [21, 82], [84, 56], [53, 30], [33, 7], [56, 50], [38, 5], [61, 81], [87, 80], [28, 92], [64, 31], [25, 8], [16, 76], [46, 5], [7, 42], [15, 90], [73, 22], [94, 9], [12, 43], [77, 35], [12, 49], [43, 72], [67, 81], [24, 39], [92, 46], [25, 67], [23, 91], [22, 60], [40, 21], [80, 50], [17, 69], [51, 51], [87, 64], [92, 58], [89, 15], [78, 59], [84, 47], [41, 79], [19, 46], [88, 37], [50, 81], [94, 31]], "components": [[0, 4, 7, 19, 35, 37, 38, 39, 49, 51, 69, 70, 72, 87], [1, 2, 10, 27, 61], [3, 6, 14, 16, 31, 36, 42, 53, 67, 71, 73, 76], [5, 9, 28, 33, 34], [8, 12, 15, 21, 25, 26, 32, 44, 45, 60, 74, 75, 90, 93], [11, 17, 29, 30, 41, 47, 55, 63, 77, 82, 85, 86, 88, 89], [13, 20, 68, 79], [18, 23, 43, 46, 48, 52, 54, 62, 65, 78, 80, 83, 91], [22, 50, 56, 81], [24, 40, 57, 59, 64, 66], [58, 84], [92, 94]], "edges": [[50, 81], [55, 82], [0, 38], [41, 77], [59, 66], [43, 80], [78, 80], [21, 60], [39, 51], [7, 37], [29, 30], [3, 6], [70, 87], [35, 70], [19, 38], [35, 39], [5, 28], [48, 62], [2, 10], [8, 15], [82, 89], [46, 65], [41, 55], [57, 59], [13, 68], [26, 32], [85, 86], [18, 23], [18, 46], [41, 86], [25, 44], [7, 19], [54, 65], [90, 93], [19, 39], [4, 69], [82, 88], [14, 42], [43, 52], [6, 53], [4, 7], [45, 90], [32, 74], [24, 59], [5, 34], [58, 84], [28, 33], [78, 83], [6, 76], [68, 79], [29, 47], [92, 94], [71, 73], [74, 90], [47, 82], [10, 61], [20, 68], [60, 75], [1, 10], [14, 76], [48, 54], [52, 91], [12, 45], [23, 43], [4, 72], [21, 93], [25, 75], [22, 50], [57, 64], [17, 85], [16, 73], [27, 61], [30, 63], [67, 71], [31, 67], [8, 25], [36, 42], [5, 9], [0, 49], [11, 85], [50, 56], [6, 31], [40, 64], [17, 86], [55, 86], [41, 89], [40, 57], [3, 53], [21, 45], [31, 71], [22, 56], [10, 27], [4, 37], [55, 88], [6, 71], [14, 53], [24, 40], [55, 89]], "dot_radius": 1.5, "image": "images/counting_connected_components_00001.png"} +{"width": 1916, "height": 1916, "grid_rows": 100, "grid_cols": 100, "square_left": 191.0, "square_top": 191.0, "square_size": 1534, "num_components": 15, "num_dots": 108, "question": "How many connected components are there in the image? A connected component is a maximal group of anchor points such that any two are linked by a path of one or more strands (directly or through other anchors in the same group). Anchors not linked by any chain belong to different components, even if they appear visually close. Count every connected component and report the total. Provide your final answer enclosed in ... tags.", "answer": 15, "dots": [[44, 11], [39, 28], [87, 17], [16, 58], [31, 32], [40, 89], [12, 20], [60, 56], [36, 32], [17, 34], [81, 31], [33, 65], [29, 56], [14, 68], [78, 74], [42, 38], [86, 73], [79, 82], [69, 30], [52, 74], [67, 44], [88, 39], [72, 48], [30, 20], [22, 39], [52, 84], [18, 92], [58, 42], [76, 86], [91, 83], [64, 86], [8, 41], [60, 94], [40, 80], [11, 77], [64, 26], [55, 22], [19, 54], [35, 13], [48, 64], [16, 50], [33, 49], [46, 87], [53, 37], [85, 7], [84, 92], [41, 52], [22, 77], [85, 60], [25, 44], [67, 9], [15, 44], [83, 26], [62, 78], [17, 11], [25, 53], [83, 68], [23, 69], [54, 55], [77, 9], [39, 57], [35, 85], [37, 39], [68, 63], [54, 10], [57, 67], [7, 16], [27, 91], [7, 23], [6, 83], [69, 25], [87, 51], [94, 32], [32, 44], [41, 62], [76, 20], [34, 54], [47, 77], [47, 59], [84, 78], [92, 60], [62, 39], [78, 43], [48, 25], [71, 74], [74, 33], [89, 77], [24, 26], [51, 49], [73, 63], [79, 65], [90, 94], [57, 74], [28, 69], [32, 77], [35, 90], [44, 5], [30, 87], [11, 54], [10, 11], [5, 68], [68, 82], [43, 16], [59, 19], [24, 34], [88, 87], [28, 8], [40, 47]], "components": [[0, 38, 64, 96, 102, 106], [1, 10, 18, 35, 36, 50, 52, 70, 75, 83, 85, 103], [2, 21, 72], [3, 9, 12, 24, 31, 37, 40, 41, 49, 51, 55, 98, 104], [4, 7, 8, 15, 23, 46, 58, 60, 62, 73, 76, 78, 87, 107], [5, 19, 25, 26, 33, 42, 61, 65, 67, 77, 94, 95, 97], [6, 54, 66, 68, 99], [11, 13, 34, 39, 47, 57, 69, 74, 93, 100], [14, 16, 29, 30, 32, 48, 56, 71, 79, 80, 84, 86, 90, 101], [17, 28], [20, 22, 27, 43, 81, 82, 88], [44, 59], [45, 91, 105], [53, 92], [63, 89]], "edges": [[73, 107], [19, 65], [35, 70], [25, 77], [16, 56], [20, 81], [12, 55], [61, 95], [11, 74], [14, 79], [43, 81], [0, 102], [33, 61], [67, 97], [3, 37], [46, 60], [41, 55], [20, 22], [39, 74], [24, 51], [79, 86], [19, 77], [9, 104], [22, 82], [58, 78], [67, 95], [25, 42], [29, 79], [37, 55], [31, 51], [45, 91], [37, 40], [48, 90], [4, 62], [62, 107], [5, 42], [33, 77], [60, 76], [57, 93], [18, 85], [0, 38], [8, 15], [6, 66], [48, 56], [70, 75], [27, 43], [15, 107], [10, 85], [46, 107], [6, 68], [30, 101], [54, 99], [44, 59], [26, 67], [33, 94], [14, 56], [17, 28], [63, 89], [24, 104], [38, 106], [4, 87], [36, 103], [41, 49], [14, 84], [13, 34], [53, 92], [60, 78], [1, 83], [23, 87], [91, 105], [48, 80], [18, 70], [84, 101], [48, 71], [0, 96], [30, 32], [24, 49], [52, 75], [47, 57], [7, 58], [36, 83], [66, 99], [13, 100], [13, 57], [3, 98], [34, 69], [35, 103], [21, 72], [11, 93], [50, 103], [0, 64], [43, 88], [2, 72], [38, 102], [20, 82], [56, 90], [40, 49], [19, 25], [8, 23], [27, 81], [3, 12], [96, 106], [5, 95], [40, 51], [3, 40], [62, 73], [50, 75], [6, 99], [70, 103], [49, 51], [11, 39]], "dot_radius": 1.5, "image": "images/counting_connected_components_00002.png"} +{"width": 1916, "height": 1916, "grid_rows": 100, "grid_cols": 100, "square_left": 191.0, "square_top": 191.0, "square_size": 1534, "num_components": 18, "num_dots": 95, "question": "How many connected components are there in the image? A connected component is a maximal group of anchor points such that any two are linked by a path of one or more strands (directly or through other anchors in the same group). Anchors not linked by any chain belong to different components, even if they appear visually close. Count every connected component and report the total. Provide your final answer enclosed in ... tags.", "answer": 18, "dots": [[20, 43], [87, 20], [29, 7], [49, 59], [13, 17], [62, 93], [22, 35], [84, 16], [40, 54], [75, 88], [94, 11], [17, 52], [89, 68], [38, 77], [92, 37], [19, 64], [59, 78], [57, 51], [41, 5], [62, 61], [53, 13], [48, 44], [31, 34], [6, 29], [79, 26], [30, 22], [92, 16], [49, 27], [63, 11], [13, 86], [6, 79], [87, 50], [6, 11], [93, 62], [80, 19], [25, 24], [51, 7], [92, 44], [7, 47], [40, 43], [22, 79], [60, 25], [51, 72], [74, 47], [68, 61], [69, 27], [58, 69], [77, 67], [21, 17], [87, 7], [33, 83], [48, 77], [81, 11], [12, 43], [92, 53], [42, 62], [87, 27], [34, 10], [20, 94], [73, 9], [29, 63], [60, 17], [73, 20], [84, 84], [57, 39], [54, 90], [83, 91], [20, 30], [42, 71], [37, 20], [6, 34], [66, 71], [7, 69], [16, 68], [83, 35], [25, 66], [67, 56], [21, 71], [27, 50], [64, 83], [10, 93], [87, 58], [37, 48], [6, 60], [71, 35], [35, 89], [36, 61], [32, 51], [49, 22], [75, 75], [18, 24], [50, 87], [16, 10], [46, 67], [9, 5]], "components": [[0, 6, 11, 22, 35, 48, 67, 78, 90], [1, 10, 24, 26, 34, 41, 45, 52, 62, 84], [2, 18, 20, 27, 28, 36, 57, 59, 61, 88], [3, 8, 21, 39, 55, 82, 86, 87, 93], [4, 32, 92, 94], [5, 65, 79, 91], [7, 49], [9, 63, 66], [12, 14, 31, 33, 37, 54, 56, 74, 81], [13, 16, 19, 42, 44, 46, 51, 68, 71, 76], [15, 30, 40, 60, 72, 73, 75, 77, 83], [17, 43, 64], [23, 70], [25, 69], [29, 58, 80], [38, 53], [47, 89], [50, 85]], "edges": [[41, 45], [24, 45], [18, 57], [19, 44], [20, 61], [75, 77], [33, 81], [42, 68], [44, 71], [11, 78], [48, 90], [45, 84], [10, 26], [29, 80], [38, 53], [56, 74], [1, 26], [14, 37], [63, 66], [92, 94], [24, 34], [20, 36], [6, 67], [4, 92], [50, 85], [0, 6], [28, 59], [15, 73], [55, 86], [1, 34], [67, 90], [31, 81], [5, 65], [34, 52], [34, 62], [20, 28], [4, 32], [12, 33], [5, 79], [40, 77], [35, 67], [0, 78], [13, 51], [15, 60], [51, 68], [2, 57], [31, 37], [82, 87], [8, 82], [9, 63], [46, 71], [7, 49], [16, 46], [6, 22], [19, 76], [14, 74], [3, 55], [42, 46], [47, 89], [55, 93], [25, 69], [72, 83], [60, 75], [18, 36], [20, 88], [72, 73], [8, 55], [39, 82], [37, 54], [23, 70], [30, 72], [21, 39], [27, 88], [58, 80], [65, 91], [43, 64], [17, 64], [44, 76], [3, 8], [32, 92], [73, 83], [19, 71], [6, 35], [61, 88], [33, 54], [65, 79], [0, 22], [15, 75], [16, 42], [40, 73], [24, 84], [1, 24]], "dot_radius": 1.5, "image": "images/counting_connected_components_00003.png"} +{"width": 1916, "height": 1916, "grid_rows": 100, "grid_cols": 100, "square_left": 191.0, "square_top": 191.0, "square_size": 1534, "num_components": 20, "num_dots": 108, "question": "How many connected components are there in the image? A connected component is a maximal group of anchor points such that any two are linked by a path of one or more strands (directly or through other anchors in the same group). Anchors not linked by any chain belong to different components, even if they appear visually close. Count every connected component and report the total. Provide your final answer enclosed in ... tags.", "answer": 20, "dots": [[42, 38], [42, 67], [10, 15], [66, 41], [52, 56], [28, 61], [52, 67], [78, 52], [92, 80], [32, 10], [57, 83], [42, 88], [70, 26], [37, 43], [59, 89], [42, 9], [18, 49], [75, 89], [76, 15], [8, 85], [85, 35], [35, 6], [89, 41], [19, 33], [52, 28], [81, 58], [32, 83], [89, 15], [83, 46], [20, 23], [5, 75], [25, 73], [33, 58], [79, 29], [72, 64], [71, 47], [8, 31], [77, 9], [25, 51], [72, 53], [74, 41], [20, 64], [45, 49], [5, 25], [17, 80], [68, 33], [37, 76], [90, 27], [66, 55], [25, 25], [18, 91], [31, 78], [9, 54], [85, 77], [81, 87], [52, 42], [38, 63], [66, 46], [5, 81], [45, 62], [10, 59], [26, 13], [55, 62], [12, 64], [28, 5], [67, 66], [30, 54], [84, 64], [51, 80], [17, 10], [10, 72], [61, 63], [66, 15], [47, 35], [89, 67], [58, 24], [70, 70], [20, 76], [47, 14], [90, 34], [57, 72], [93, 73], [76, 57], [70, 10], [88, 20], [14, 87], [37, 38], [26, 44], [47, 29], [87, 93], [32, 88], [83, 70], [52, 49], [57, 56], [34, 18], [22, 37], [94, 59], [65, 28], [45, 5], [93, 91], [27, 86], [37, 87], [52, 34], [47, 87], [53, 18], [20, 56], [94, 51], [89, 53]], "components": [[0, 4, 13, 42, 55, 73, 86, 88, 92, 102], [1, 6, 56, 59, 62, 71, 80, 93], [2, 69], [3, 12, 15, 24, 45, 75, 78, 97, 98, 104], [5, 16, 32, 38, 41, 63, 66, 70, 87, 105], [7, 22, 25, 28, 35, 39, 40, 48, 57, 107], [8, 53, 67, 74, 81, 91], [9, 21, 61, 64, 94], [10, 11, 14, 68, 103], [17, 54, 89, 99], [18, 37, 72, 83], [19, 30, 31, 44, 50, 58, 77, 85], [20, 27, 33, 47, 79, 84], [23, 95], [26, 46, 51, 90, 100, 101], [29, 49], [34, 65, 76, 82], [36, 43], [52, 60], [96, 106]], "edges": [[7, 39], [4, 92], [96, 106], [26, 90], [89, 99], [9, 94], [26, 100], [71, 93], [18, 83], [48, 57], [62, 71], [55, 102], [36, 43], [20, 47], [28, 107], [34, 82], [46, 51], [55, 92], [15, 98], [38, 105], [88, 102], [41, 63], [24, 75], [9, 61], [35, 40], [34, 76], [53, 81], [44, 77], [39, 48], [74, 81], [54, 89], [10, 68], [73, 88], [44, 85], [78, 104], [20, 33], [5, 105], [25, 107], [78, 98], [32, 66], [22, 28], [16, 38], [7, 25], [42, 92], [0, 13], [6, 59], [6, 80], [65, 76], [63, 70], [51, 100], [18, 37], [72, 83], [67, 91], [3, 45], [75, 104], [61, 64], [13, 86], [12, 97], [5, 66], [21, 64], [12, 45], [2, 69], [47, 79], [35, 57], [1, 59], [5, 41], [11, 103], [6, 62], [68, 103], [38, 87], [56, 59], [0, 73], [50, 85], [31, 77], [75, 97], [30, 58], [53, 91], [47, 84], [19, 85], [52, 60], [27, 84], [10, 14], [17, 54], [29, 49], [90, 101], [19, 58], [8, 53], [23, 95], [0, 86], [18, 72], [42, 55], [35, 39], [37, 83], [4, 42], [0, 88], [9, 64], [10, 103], [8, 81], [55, 73], [41, 70], [33, 47], [19, 44], [7, 40], [90, 100], [14, 103], [39, 57]], "dot_radius": 1.5, "image": "images/counting_connected_components_00004.png"} diff --git a/data/annotations/counting_regions.jsonl b/data/annotations/counting_regions.jsonl new file mode 100644 index 0000000000000000000000000000000000000000..d2189ea5d7c85e0533c46ca423ad2b359866957f --- /dev/null +++ b/data/annotations/counting_regions.jsonl @@ -0,0 +1,5 @@ +{"width": 1254, "height": 1254, "grid_rows": 200, "grid_cols": 200, "square_left": 125.0, "square_top": 125.0, "square_size": 1004.0, "num_regions": 10, "question": "How many separated regions are inside the image? A region is a maximal area filled with one continuous fill. Boundaries between regions may appear as continuous lines OR as a chain of small individual markers \u2014 discrete dots, footprints, dashes, or short segments aligned along a curve \u2014 that you must visually link together to recognize as a single boundary. Two locations belong to the same region if you can travel between them without crossing such a boundary (continuous or chained). Count every distinct region and report the total as a positive integer. Provide your final answer enclosed in ... tags.", "answer": 10, "difficulty": "medium", "region_seed_cells": [[199, 74], [21, 4], [108, 97], [33, 185], [49, 68], [113, 3], [181, 127], [131, 160], [175, 3], [3, 113]], "region_cell_counts": [74251, 80287, 99834, 140057, 110665, 106165, 110475, 125225, 74962, 86095], "region_adjacency": [[0, 2], [0, 5], [0, 6], [0, 8], [1, 4], [1, 5], [1, 9], [2, 3], [2, 4], [2, 5], [2, 6], [2, 7], [3, 4], [3, 7], [3, 9], [4, 5], [4, 9], [5, 8], [6, 7]], "image": "images/counting_regions_00000.png"} +{"width": 1254, "height": 1254, "grid_rows": 200, "grid_cols": 200, "square_left": 125.0, "square_top": 125.0, "square_size": 1004.0, "num_regions": 12, "question": "How many separated regions are inside the image? A region is a maximal area filled with one continuous fill. Boundaries between regions may appear as continuous lines OR as a chain of small individual markers \u2014 discrete dots, footprints, dashes, or short segments aligned along a curve \u2014 that you must visually link together to recognize as a single boundary. Two locations belong to the same region if you can travel between them without crossing such a boundary (continuous or chained). Count every distinct region and report the total as a positive integer. Provide your final answer enclosed in ... tags.", "answer": 12, "difficulty": "hard", "region_seed_cells": [[118, 113], [82, 50], [175, 46], [13, 58], [53, 170], [196, 105], [45, 117], [46, 0], [178, 192], [110, 192], [144, 5], [6, 196]], "region_cell_counts": [107644, 90017, 70137, 60871, 74055, 98974, 110114, 72037, 91826, 87191, 78872, 66278], "region_adjacency": [[0, 1], [0, 4], [0, 5], [0, 6], [0, 8], [0, 9], [0, 10], [1, 3], [1, 6], [1, 7], [1, 10], [2, 5], [2, 10], [3, 6], [3, 7], [4, 6], [4, 9], [4, 11], [5, 8], [5, 10], [6, 11], [7, 10], [8, 9]], "image": "images/counting_regions_00001.png"} +{"width": 1254, "height": 1254, "grid_rows": 200, "grid_cols": 200, "square_left": 125.0, "square_top": 125.0, "square_size": 1004.0, "num_regions": 15, "question": "How many separated regions are inside the image? A region is a maximal area filled with one continuous fill. Boundaries between regions may appear as continuous lines OR as a chain of small individual markers \u2014 discrete dots, footprints, dashes, or short segments aligned along a curve \u2014 that you must visually link together to recognize as a single boundary. Two locations belong to the same region if you can travel between them without crossing such a boundary (continuous or chained). Count every distinct region and report the total as a positive integer. Provide your final answer enclosed in ... tags.", "answer": 15, "difficulty": "hard", "region_seed_cells": [[46, 134], [106, 26], [10, 80], [135, 80], [27, 198], [57, 69], [184, 194], [15, 18], [92, 129], [152, 6], [100, 176], [163, 119], [199, 27], [199, 74], [47, 154]], "region_cell_counts": [59115, 63143, 63169, 70499, 55611, 74488, 64501, 87475, 72985, 62554, 98562, 87154, 47120, 50242, 51398], "region_adjacency": [[0, 2], [0, 4], [0, 5], [0, 8], [0, 14], [1, 5], [1, 7], [1, 9], [2, 5], [2, 7], [3, 5], [3, 8], [3, 9], [3, 10], [3, 11], [3, 12], [3, 13], [4, 14], [5, 7], [5, 8], [5, 9], [6, 10], [6, 11], [8, 10], [8, 14], [9, 12], [10, 11], [10, 14], [11, 13], [12, 13]], "image": "images/counting_regions_00002.png"} +{"width": 1254, "height": 1254, "grid_rows": 200, "grid_cols": 200, "square_left": 125.0, "square_top": 125.0, "square_size": 1004.0, "num_regions": 18, "question": "How many separated regions are inside the image? A region is a maximal area filled with one continuous fill. Boundaries between regions may appear as continuous lines OR as a chain of small individual markers \u2014 discrete dots, footprints, dashes, or short segments aligned along a curve \u2014 that you must visually link together to recognize as a single boundary. Two locations belong to the same region if you can travel between them without crossing such a boundary (continuous or chained). Count every distinct region and report the total as a positive integer. Provide your final answer enclosed in ... tags.", "answer": 18, "difficulty": "hard", "region_seed_cells": [[129, 183], [199, 19], [165, 99], [85, 94], [196, 133], [43, 184], [96, 147], [0, 35], [156, 40], [0, 111], [53, 45], [104, 9], [1, 154], [32, 2], [175, 179], [53, 141], [112, 59], [142, 141]], "region_cell_counts": [49542, 49619, 59811, 61799, 41901, 56407, 58257, 50999, 72110, 69684, 47373, 71734, 44532, 43735, 48309, 51149, 74918, 56137], "region_adjacency": [[0, 6], [0, 14], [0, 17], [1, 2], [1, 8], [2, 4], [2, 8], [2, 16], [2, 17], [3, 6], [3, 9], [3, 10], [3, 15], [3, 16], [3, 17], [4, 14], [4, 17], [5, 6], [5, 12], [5, 15], [6, 15], [6, 17], [7, 9], [7, 10], [7, 13], [8, 11], [8, 16], [9, 10], [9, 12], [9, 15], [10, 11], [10, 13], [10, 16], [11, 13], [11, 16], [12, 15], [14, 17], [16, 17]], "image": "images/counting_regions_00003.png"} +{"width": 1254, "height": 1254, "grid_rows": 200, "grid_cols": 200, "square_left": 125.0, "square_top": 125.0, "square_size": 1004.0, "num_regions": 20, "question": "How many separated regions are inside the image? A region is a maximal area filled with one continuous fill. Boundaries between regions may appear as continuous lines OR as a chain of small individual markers \u2014 discrete dots, footprints, dashes, or short segments aligned along a curve \u2014 that you must visually link together to recognize as a single boundary. Two locations belong to the same region if you can travel between them without crossing such a boundary (continuous or chained). Count every distinct region and report the total as a positive integer. Provide your final answer enclosed in ... tags.", "answer": 20, "difficulty": "hard", "region_seed_cells": [[96, 38], [161, 47], [39, 84], [45, 148], [124, 168], [2, 130], [96, 134], [182, 120], [29, 31], [31, 193], [167, 180], [155, 1], [197, 11], [74, 183], [134, 112], [196, 70], [118, 74], [64, 6], [78, 97], [2, 62]], "region_cell_counts": [68165, 53951, 53663, 47695, 72486, 39787, 54164, 57307, 50313, 49792, 63259, 37212, 31919, 48205, 58095, 35318, 48515, 37963, 52838, 47369], "region_adjacency": [[0, 1], [0, 11], [0, 16], [0, 17], [0, 18], [0, 19], [1, 11], [1, 12], [1, 14], [1, 15], [1, 16], [2, 3], [2, 5], [2, 18], [2, 19], [3, 5], [3, 6], [3, 9], [3, 13], [3, 18], [4, 6], [4, 7], [4, 10], [4, 13], [4, 14], [5, 9], [5, 19], [6, 13], [6, 14], [6, 18], [7, 10], [7, 14], [7, 15], [8, 17], [8, 19], [9, 13], [11, 12], [12, 15], [14, 15], [14, 16], [14, 18], [16, 18], [17, 19], [18, 19]], "image": "images/counting_regions_00004.png"} diff --git a/data/annotations/line_intersections.jsonl b/data/annotations/line_intersections.jsonl new file mode 100644 index 0000000000000000000000000000000000000000..e4ede467e2d8e5bbfafa2dfe8bf6ae04622eb383 --- /dev/null +++ b/data/annotations/line_intersections.jsonl @@ -0,0 +1,5 @@ +{"width": 1774, "height": 1774, "num_lines": 4, "query_line": 0, "query_label": "A", "line_labels": ["A", "B", "C", "D"], "total_crossings": 14, "sequence_length": 11, "label_starts": [true, true, true, true], "question": "The image shows several curves or strands tangled together. Each curve has a single labeled endpoint on the perimeter, marked with a distinct letter. Starting from the endpoint labeled A, follow that curve through the interior tangle. At every crossing you pass through, record the label of the OTHER curve involved \u2014 when curve A crosses itself, record A. Report the ordered sequence of labels, separated by commas. Provide your final answer enclosed in ... tags.", "answer": "C, C, D, C, D, D, C, C, C, B, C", "image": "images/line_intersections_00000.png"} +{"width": 1774, "height": 1774, "num_lines": 4, "query_line": 0, "query_label": "A", "line_labels": ["A", "B", "C", "D"], "total_crossings": 12, "sequence_length": 9, "label_starts": [true, false, true, true], "question": "The image shows several curves or strands tangled together. Each curve has a single labeled endpoint on the perimeter, marked with a distinct letter. Starting from the endpoint labeled A, follow that curve through the interior tangle. At every crossing you pass through, record the label of the OTHER curve involved \u2014 when curve A crosses itself, record A. Report the ordered sequence of labels, separated by commas. Provide your final answer enclosed in ... tags.", "answer": "D, B, B, C, C, B, B, B, C", "image": "images/line_intersections_00001.png"} +{"width": 1774, "height": 1774, "num_lines": 4, "query_line": 0, "query_label": "A", "line_labels": ["A", "B", "C", "D"], "total_crossings": 19, "sequence_length": 9, "label_starts": [false, true, false, true], "question": "The image shows several curves or strands tangled together. Each curve has a single labeled endpoint on the perimeter, marked with a distinct letter. Starting from the endpoint labeled A, follow that curve through the interior tangle. At every crossing you pass through, record the label of the OTHER curve involved \u2014 when curve A crosses itself, record A. Report the ordered sequence of labels, separated by commas. Provide your final answer enclosed in ... tags.", "answer": "D, C, C, B, B, D, B, D, C", "image": "images/line_intersections_00002.png"} +{"width": 1774, "height": 1774, "num_lines": 4, "query_line": 0, "query_label": "A", "line_labels": ["A", "B", "C", "D"], "total_crossings": 14, "sequence_length": 9, "label_starts": [true, true, true, false], "question": "The image shows several curves or strands tangled together. Each curve has a single labeled endpoint on the perimeter, marked with a distinct letter. Starting from the endpoint labeled A, follow that curve through the interior tangle. At every crossing you pass through, record the label of the OTHER curve involved \u2014 when curve A crosses itself, record A. Report the ordered sequence of labels, separated by commas. Provide your final answer enclosed in ... tags.", "answer": "D, C, B, C, D, C, D, C, C", "image": "images/line_intersections_00003.png"} +{"width": 1774, "height": 1774, "num_lines": 4, "query_line": 0, "query_label": "A", "line_labels": ["A", "B", "C", "D"], "total_crossings": 20, "sequence_length": 11, "label_starts": [true, false, true, false], "question": "The image shows several curves or strands tangled together. Each curve has a single labeled endpoint on the perimeter, marked with a distinct letter. Starting from the endpoint labeled A, follow that curve through the interior tangle. At every crossing you pass through, record the label of the OTHER curve involved \u2014 when curve A crosses itself, record A. Report the ordered sequence of labels, separated by commas. Provide your final answer enclosed in ... tags.", "answer": "C, C, B, B, C, B, C, B, C, D, B", "image": "images/line_intersections_00004.png"} diff --git a/data/annotations/maze.jsonl b/data/annotations/maze.jsonl new file mode 100644 index 0000000000000000000000000000000000000000..ab70f5ef7539f4a49baf854d6d7a405b2161b343 --- /dev/null +++ b/data/annotations/maze.jsonl @@ -0,0 +1,5 @@ +{"image": "images/maze_00000.png", "rows": 20, "cols": 20, "cell_size": 32, "wall_width": 4, "margin": 8, "num_openings": 7, "openings": [{"label": "A", "side": "right", "index": 2, "is_connected": false}, {"label": "B", "side": "left", "index": 10, "is_connected": false}, {"label": "C", "side": "bottom", "index": 15, "is_connected": false}, {"label": "D", "side": "left", "index": 2, "is_connected": false}, {"label": "E", "side": "left", "index": 15, "is_connected": true}, {"label": "F", "side": "top", "index": 2, "is_connected": true}, {"label": "G", "side": "bottom", "index": 2, "is_connected": false}], "connected_pair": ["E", "F"], "question": "The image shows a maze with several labeled openings on its border. Exactly one pair of openings is connected by a path through the maze; all other pairs are blocked by walls. Identify the connected pair. A path exists only if you can travel from one opening to another through corridors without crossing any wall. Report the connected pair in alphabetical order, formatted as X-Y. Provide your final answer enclosed in ... tags.", "answer": "E-F", "path_length": 178, "difficulty": "hard"} +{"image": "images/maze_00001.png", "rows": 20, "cols": 20, "cell_size": 32, "wall_width": 4, "margin": 8, "num_openings": 5, "openings": [{"label": "A", "side": "top", "index": 17, "is_connected": false}, {"label": "B", "side": "bottom", "index": 15, "is_connected": false}, {"label": "C", "side": "bottom", "index": 8, "is_connected": true}, {"label": "D", "side": "left", "index": 3, "is_connected": false}, {"label": "E", "side": "top", "index": 11, "is_connected": true}], "connected_pair": ["C", "E"], "question": "The image shows a maze with several labeled openings on its border. Exactly one pair of openings is connected by a path through the maze; all other pairs are blocked by walls. Identify the connected pair. A path exists only if you can travel from one opening to another through corridors without crossing any wall. Report the connected pair in alphabetical order, formatted as X-Y. Provide your final answer enclosed in ... tags.", "answer": "C-E", "path_length": 155, "difficulty": "hard"} +{"image": "images/maze_00002.png", "rows": 20, "cols": 20, "cell_size": 32, "wall_width": 4, "margin": 8, "num_openings": 7, "openings": [{"label": "A", "side": "bottom", "index": 16, "is_connected": true}, {"label": "B", "side": "right", "index": 17, "is_connected": false}, {"label": "C", "side": "left", "index": 7, "is_connected": false}, {"label": "D", "side": "top", "index": 17, "is_connected": false}, {"label": "E", "side": "bottom", "index": 14, "is_connected": false}, {"label": "F", "side": "top", "index": 2, "is_connected": false}, {"label": "G", "side": "top", "index": 3, "is_connected": true}], "connected_pair": ["A", "G"], "question": "The image shows a maze with several labeled openings on its border. Exactly one pair of openings is connected by a path through the maze; all other pairs are blocked by walls. Identify the connected pair. A path exists only if you can travel from one opening to another through corridors without crossing any wall. Report the connected pair in alphabetical order, formatted as X-Y. Provide your final answer enclosed in ... tags.", "answer": "A-G", "path_length": 161, "difficulty": "hard"} +{"image": "images/maze_00003.png", "rows": 20, "cols": 20, "cell_size": 32, "wall_width": 4, "margin": 8, "num_openings": 7, "openings": [{"label": "A", "side": "right", "index": 13, "is_connected": false}, {"label": "B", "side": "top", "index": 10, "is_connected": false}, {"label": "C", "side": "bottom", "index": 12, "is_connected": true}, {"label": "D", "side": "top", "index": 17, "is_connected": false}, {"label": "E", "side": "left", "index": 15, "is_connected": false}, {"label": "F", "side": "bottom", "index": 16, "is_connected": false}, {"label": "G", "side": "right", "index": 5, "is_connected": true}], "connected_pair": ["C", "G"], "question": "The image shows a maze with several labeled openings on its border. Exactly one pair of openings is connected by a path through the maze; all other pairs are blocked by walls. Identify the connected pair. A path exists only if you can travel from one opening to another through corridors without crossing any wall. Report the connected pair in alphabetical order, formatted as X-Y. Provide your final answer enclosed in ... tags.", "answer": "C-G", "path_length": 168, "difficulty": "hard"} +{"image": "images/maze_00004.png", "rows": 20, "cols": 20, "cell_size": 32, "wall_width": 4, "margin": 8, "num_openings": 6, "openings": [{"label": "A", "side": "right", "index": 7, "is_connected": false}, {"label": "B", "side": "top", "index": 8, "is_connected": true}, {"label": "C", "side": "right", "index": 17, "is_connected": true}, {"label": "D", "side": "left", "index": 9, "is_connected": false}, {"label": "E", "side": "right", "index": 16, "is_connected": false}, {"label": "F", "side": "top", "index": 3, "is_connected": false}], "connected_pair": ["B", "C"], "question": "The image shows a maze with several labeled openings on its border. Exactly one pair of openings is connected by a path through the maze; all other pairs are blocked by walls. Identify the connected pair. A path exists only if you can travel from one opening to another through corridors without crossing any wall. Report the connected pair in alphabetical order, formatted as X-Y. Provide your final answer enclosed in ... tags.", "answer": "B-C", "path_length": 135, "difficulty": "hard"} diff --git a/data/annotations/spot_the_contour_diff.jsonl b/data/annotations/spot_the_contour_diff.jsonl new file mode 100644 index 0000000000000000000000000000000000000000..b3d2792b11080820128119d212b63354fcb12a4b --- /dev/null +++ b/data/annotations/spot_the_contour_diff.jsonl @@ -0,0 +1,5 @@ +{"image": "images/contour_diff_00000.png", "num_items": 20, "num_differences": 5, "differences": [{"index": 2, "label": "C", "x": 105.32573202012559, "y": 1015.1429020205117, "bounding_radius": 69.6322400297027}, {"index": 8, "label": "I", "x": 1021.3285409244434, "y": 681.0411023666037, "bounding_radius": 67.81243133940893}, {"index": 9, "label": "J", "x": 755.4128884536265, "y": 481.1836575273767, "bounding_radius": 62.65951785269289}, {"index": 11, "label": "L", "x": 133.66690795121798, "y": 343.47806410497, "bounding_radius": 66.26014306374935}, {"index": 12, "label": "M", "x": 500.61575544439995, "y": 729.2346445428684, "bounding_radius": 57.7377851032875}], "question": "Two panels are shown side by side. Each panel contains the same set of closed silhouettes at the same positions. Some silhouettes in the right panel have a different shape from the corresponding silhouette in the left panel. Count the number of pairs whose shape differs between the two panels and report the integer count. Provide your final answer enclosed in ... tags.", "answer": "5"} +{"image": "images/contour_diff_00001.png", "num_items": 20, "num_differences": 8, "differences": [{"index": 0, "label": "A", "x": 1186.9892657398393, "y": 562.7530050172959, "bounding_radius": 61.048879264238664}, {"index": 4, "label": "E", "x": 188.68311461315207, "y": 438.89484268952026, "bounding_radius": 77.11218476572941}, {"index": 12, "label": "M", "x": 844.5881594335667, "y": 269.891938355782, "bounding_radius": 62.233157458497814}, {"index": 13, "label": "N", "x": 910.6131936019698, "y": 900.3116262031238, "bounding_radius": 64.53102174750144}, {"index": 14, "label": "O", "x": 774.0268527194689, "y": 115.2904975667536, "bounding_radius": 75.45195980575058}, {"index": 17, "label": "R", "x": 786.8169874454511, "y": 593.2930886387344, "bounding_radius": 73.65852507694208}, {"index": 18, "label": "S", "x": 1048.1860068183692, "y": 1120.262952094394, "bounding_radius": 72.78451015228984}, {"index": 19, "label": "T", "x": 1137.0391399416392, "y": 731.3649046481294, "bounding_radius": 61.93122378660632}], "question": "Two panels are shown side by side. Each panel contains the same set of closed silhouettes at the same positions. Some silhouettes in the right panel have a different shape from the corresponding silhouette in the left panel. Count the number of pairs whose shape differs between the two panels and report the integer count. Provide your final answer enclosed in ... tags.", "answer": "8"} +{"image": "images/contour_diff_00002.png", "num_items": 20, "num_differences": 10, "differences": [{"index": 0, "label": "A", "x": 672.3931321437459, "y": 628.9321752205313, "bounding_radius": 66.05976754044607}, {"index": 2, "label": "C", "x": 179.74365010938254, "y": 355.94085501771167, "bounding_radius": 56.67003345178697}, {"index": 5, "label": "F", "x": 160.32409886070656, "y": 838.1142919201296, "bounding_radius": 61.43364804574752}, {"index": 7, "label": "H", "x": 479.06178020903894, "y": 504.40152011590254, "bounding_radius": 68.4494262091954}, {"index": 10, "label": "K", "x": 330.1067033558254, "y": 722.1621874747385, "bounding_radius": 52.51111579167739}, {"index": 12, "label": "M", "x": 321.7571476379459, "y": 185.79549127588098, "bounding_radius": 59.512745971639134}, {"index": 13, "label": "N", "x": 969.7263422300136, "y": 1073.107727521923, "bounding_radius": 55.08906100069768}, {"index": 14, "label": "O", "x": 1129.1895648883592, "y": 345.5774312895564, "bounding_radius": 56.417679104359024}, {"index": 15, "label": "P", "x": 970.6129834778252, "y": 731.0457283135199, "bounding_radius": 70.04226748937728}, {"index": 17, "label": "R", "x": 609.5039412141149, "y": 385.6229858189672, "bounding_radius": 55.78982709605432}], "question": "Two panels are shown side by side. Each panel contains the same set of closed silhouettes at the same positions. Some silhouettes in the right panel have a different shape from the corresponding silhouette in the left panel. Count the number of pairs whose shape differs between the two panels and report the integer count. Provide your final answer enclosed in ... tags.", "answer": "10"} +{"image": "images/contour_diff_00003.png", "num_items": 20, "num_differences": 12, "differences": [{"index": 0, "label": "A", "x": 461.63144384159295, "y": 1110.3557278798817, "bounding_radius": 74.915999930236}, {"index": 2, "label": "C", "x": 120.39770208610801, "y": 725.5110399555472, "bounding_radius": 55.47608886744446}, {"index": 3, "label": "D", "x": 1126.2972587160536, "y": 1075.1905653763763, "bounding_radius": 62.135444258029466}, {"index": 5, "label": "F", "x": 203.82946486288114, "y": 86.48406256103708, "bounding_radius": 59.87721462216948}, {"index": 6, "label": "G", "x": 109.69785477531073, "y": 926.2729065792174, "bounding_radius": 67.25492197338679}, {"index": 9, "label": "J", "x": 358.20525244832675, "y": 756.4904504014221, "bounding_radius": 71.0439800800543}, {"index": 10, "label": "K", "x": 830.7883348735198, "y": 530.9012296694201, "bounding_radius": 61.647252479018036}, {"index": 13, "label": "N", "x": 184.2351434616582, "y": 542.0720761081284, "bounding_radius": 70.97896797489346}, {"index": 15, "label": "P", "x": 1113.021363182762, "y": 692.6274805794917, "bounding_radius": 55.639503328216556}, {"index": 17, "label": "R", "x": 1031.1279693293513, "y": 584.13775095841, "bounding_radius": 53.76490040016805}, {"index": 18, "label": "S", "x": 988.4364995699609, "y": 211.95241891190358, "bounding_radius": 73.8855261438851}, {"index": 19, "label": "T", "x": 943.6788678642199, "y": 1167.7738935173923, "bounding_radius": 55.078697046837206}], "question": "Two panels are shown side by side. Each panel contains the same set of closed silhouettes at the same positions. Some silhouettes in the right panel have a different shape from the corresponding silhouette in the left panel. Count the number of pairs whose shape differs between the two panels and report the integer count. Provide your final answer enclosed in ... tags.", "answer": "12"} +{"image": "images/contour_diff_00004.png", "num_items": 20, "num_differences": 15, "differences": [{"index": 0, "label": "A", "x": 434.7976585108059, "y": 647.0350002153475, "bounding_radius": 68.26965736059778}, {"index": 1, "label": "B", "x": 1149.1781823484541, "y": 1024.006217314532, "bounding_radius": 53.37783876391526}, {"index": 4, "label": "E", "x": 430.43863914351675, "y": 1174.0977171700363, "bounding_radius": 68.47364787165289}, {"index": 5, "label": "F", "x": 939.9039109366325, "y": 314.7843796798786, "bounding_radius": 54.12674961922259}, {"index": 7, "label": "H", "x": 743.1213937413829, "y": 154.67125714228408, "bounding_radius": 80.26868543064342}, {"index": 8, "label": "I", "x": 926.8442269714009, "y": 735.6913747275887, "bounding_radius": 64.15662084929374}, {"index": 9, "label": "J", "x": 1025.758614893776, "y": 946.9654708958001, "bounding_radius": 72.10743181800954}, {"index": 11, "label": "L", "x": 1128.7873169685972, "y": 1186.093131037898, "bounding_radius": 54.946894542020786}, {"index": 12, "label": "M", "x": 96.92558256893706, "y": 1196.5540777398635, "bounding_radius": 55.39410525844867}, {"index": 13, "label": "N", "x": 799.7526929866489, "y": 594.9129894121272, "bounding_radius": 77.86752536711263}, {"index": 14, "label": "O", "x": 762.42379053262, "y": 380.3591769436079, "bounding_radius": 53.59877286007314}, {"index": 16, "label": "Q", "x": 323.6101888981932, "y": 891.6907557166516, "bounding_radius": 60.48930123407488}, {"index": 17, "label": "R", "x": 589.6754257233171, "y": 504.2514810925725, "bounding_radius": 56.10111811686236}, {"index": 18, "label": "S", "x": 1045.9241686637658, "y": 640.5725717457696, "bounding_radius": 58.85203945844605}, {"index": 19, "label": "T", "x": 185.60221146069017, "y": 585.1868118543637, "bounding_radius": 68.44497131398936}], "question": "Two panels are shown side by side. Each panel contains the same set of closed silhouettes at the same positions. Some silhouettes in the right panel have a different shape from the corresponding silhouette in the left panel. Count the number of pairs whose shape differs between the two panels and report the integer count. Provide your final answer enclosed in ... tags.", "answer": "15"} diff --git a/data/annotations/spot_the_field_diff.jsonl b/data/annotations/spot_the_field_diff.jsonl new file mode 100644 index 0000000000000000000000000000000000000000..dc4147511f081ee402cd573bbbe0ade88ba72bab --- /dev/null +++ b/data/annotations/spot_the_field_diff.jsonl @@ -0,0 +1,5 @@ +{"image": "images/field_diff_00000.png", "color_mode": "coolwarm", "grid": 4, "num_cells": 16, "num_identical": 12, "num_different": 4, "diff_indices": [6, 7, 8, 11], "cells": [{"row": 0, "col": 0, "perturbed": false, "cx": 100, "cy": 100, "radius": 70}, {"row": 0, "col": 1, "perturbed": false, "cx": 300, "cy": 100, "radius": 70}, {"row": 0, "col": 2, "perturbed": false, "cx": 500, "cy": 100, "radius": 70}, {"row": 0, "col": 3, "perturbed": false, "cx": 700, "cy": 100, "radius": 70}, {"row": 1, "col": 0, "perturbed": false, "cx": 100, "cy": 300, "radius": 70}, {"row": 1, "col": 1, "perturbed": true, "cx": 300, "cy": 300, "radius": 70}, {"row": 1, "col": 2, "perturbed": true, "cx": 500, "cy": 300, "radius": 70}, {"row": 1, "col": 3, "perturbed": true, "cx": 700, "cy": 300, "radius": 70}, {"row": 2, "col": 0, "perturbed": false, "cx": 100, "cy": 500, "radius": 70}, {"row": 2, "col": 1, "perturbed": false, "cx": 300, "cy": 500, "radius": 70}, {"row": 2, "col": 2, "perturbed": true, "cx": 500, "cy": 500, "radius": 70}, {"row": 2, "col": 3, "perturbed": false, "cx": 700, "cy": 500, "radius": 70}, {"row": 3, "col": 0, "perturbed": false, "cx": 100, "cy": 700, "radius": 70}, {"row": 3, "col": 1, "perturbed": false, "cx": 300, "cy": 700, "radius": 70}, {"row": 3, "col": 2, "perturbed": false, "cx": 500, "cy": 700, "radius": 70}, {"row": 3, "col": 3, "perturbed": false, "cx": 700, "cy": 700, "radius": 70}], "question": "Three panels are shown side by side. The leftmost panel is an INDEX key: a grid of cells numbered in row-major order (top-left = 1, increasing left-to-right then top-to-bottom). The middle and right panels are 2D color fields, each divided into the same grid. Compare the middle (Field A) and right (Field B) panels cell by cell. List the numbers of all cells that DIFFER between Field A and Field B, in ascending order, separated by commas. If no cells differ, write \"none\". Provide your final answer enclosed in ... tags.", "answer": "6, 7, 8, 11"} +{"image": "images/field_diff_00001.png", "color_mode": "plasma", "grid": 4, "num_cells": 16, "num_identical": 15, "num_different": 1, "diff_indices": [6], "cells": [{"row": 0, "col": 0, "perturbed": false, "cx": 100, "cy": 100, "radius": 70}, {"row": 0, "col": 1, "perturbed": false, "cx": 300, "cy": 100, "radius": 70}, {"row": 0, "col": 2, "perturbed": false, "cx": 500, "cy": 100, "radius": 70}, {"row": 0, "col": 3, "perturbed": false, "cx": 700, "cy": 100, "radius": 70}, {"row": 1, "col": 0, "perturbed": false, "cx": 100, "cy": 300, "radius": 70}, {"row": 1, "col": 1, "perturbed": true, "cx": 300, "cy": 300, "radius": 70}, {"row": 1, "col": 2, "perturbed": false, "cx": 500, "cy": 300, "radius": 70}, {"row": 1, "col": 3, "perturbed": false, "cx": 700, "cy": 300, "radius": 70}, {"row": 2, "col": 0, "perturbed": false, "cx": 100, "cy": 500, "radius": 70}, {"row": 2, "col": 1, "perturbed": false, "cx": 300, "cy": 500, "radius": 70}, {"row": 2, "col": 2, "perturbed": false, "cx": 500, "cy": 500, "radius": 70}, {"row": 2, "col": 3, "perturbed": false, "cx": 700, "cy": 500, "radius": 70}, {"row": 3, "col": 0, "perturbed": false, "cx": 100, "cy": 700, "radius": 70}, {"row": 3, "col": 1, "perturbed": false, "cx": 300, "cy": 700, "radius": 70}, {"row": 3, "col": 2, "perturbed": false, "cx": 500, "cy": 700, "radius": 70}, {"row": 3, "col": 3, "perturbed": false, "cx": 700, "cy": 700, "radius": 70}], "question": "Three panels are shown side by side. The leftmost panel is an INDEX key: a grid of cells numbered in row-major order (top-left = 1, increasing left-to-right then top-to-bottom). The middle and right panels are 2D color fields, each divided into the same grid. Compare the middle (Field A) and right (Field B) panels cell by cell. List the numbers of all cells that DIFFER between Field A and Field B, in ascending order, separated by commas. If no cells differ, write \"none\". Provide your final answer enclosed in ... tags.", "answer": "6"} +{"image": "images/field_diff_00002.png", "color_mode": "plasma", "grid": 4, "num_cells": 16, "num_identical": 9, "num_different": 7, "diff_indices": [3, 4, 8, 9, 10, 12, 13], "cells": [{"row": 0, "col": 0, "perturbed": false, "cx": 100, "cy": 100, "radius": 70}, {"row": 0, "col": 1, "perturbed": false, "cx": 300, "cy": 100, "radius": 70}, {"row": 0, "col": 2, "perturbed": true, "cx": 500, "cy": 100, "radius": 70}, {"row": 0, "col": 3, "perturbed": true, "cx": 700, "cy": 100, "radius": 70}, {"row": 1, "col": 0, "perturbed": false, "cx": 100, "cy": 300, "radius": 70}, {"row": 1, "col": 1, "perturbed": false, "cx": 300, "cy": 300, "radius": 70}, {"row": 1, "col": 2, "perturbed": false, "cx": 500, "cy": 300, "radius": 70}, {"row": 1, "col": 3, "perturbed": true, "cx": 700, "cy": 300, "radius": 70}, {"row": 2, "col": 0, "perturbed": true, "cx": 100, "cy": 500, "radius": 70}, {"row": 2, "col": 1, "perturbed": true, "cx": 300, "cy": 500, "radius": 70}, {"row": 2, "col": 2, "perturbed": false, "cx": 500, "cy": 500, "radius": 70}, {"row": 2, "col": 3, "perturbed": true, "cx": 700, "cy": 500, "radius": 70}, {"row": 3, "col": 0, "perturbed": true, "cx": 100, "cy": 700, "radius": 70}, {"row": 3, "col": 1, "perturbed": false, "cx": 300, "cy": 700, "radius": 70}, {"row": 3, "col": 2, "perturbed": false, "cx": 500, "cy": 700, "radius": 70}, {"row": 3, "col": 3, "perturbed": false, "cx": 700, "cy": 700, "radius": 70}], "question": "Three panels are shown side by side. The leftmost panel is an INDEX key: a grid of cells numbered in row-major order (top-left = 1, increasing left-to-right then top-to-bottom). The middle and right panels are 2D color fields, each divided into the same grid. Compare the middle (Field A) and right (Field B) panels cell by cell. List the numbers of all cells that DIFFER between Field A and Field B, in ascending order, separated by commas. If no cells differ, write \"none\". Provide your final answer enclosed in ... tags.", "answer": "3, 4, 8, 9, 10, 12, 13"} +{"image": "images/field_diff_00003.png", "color_mode": "cividis", "grid": 4, "num_cells": 16, "num_identical": 6, "num_different": 10, "diff_indices": [3, 5, 6, 7, 8, 9, 11, 13, 14, 15], "cells": [{"row": 0, "col": 0, "perturbed": false, "cx": 100, "cy": 100, "radius": 70}, {"row": 0, "col": 1, "perturbed": false, "cx": 300, "cy": 100, "radius": 70}, {"row": 0, "col": 2, "perturbed": true, "cx": 500, "cy": 100, "radius": 70}, {"row": 0, "col": 3, "perturbed": false, "cx": 700, "cy": 100, "radius": 70}, {"row": 1, "col": 0, "perturbed": true, "cx": 100, "cy": 300, "radius": 70}, {"row": 1, "col": 1, "perturbed": true, "cx": 300, "cy": 300, "radius": 70}, {"row": 1, "col": 2, "perturbed": true, "cx": 500, "cy": 300, "radius": 70}, {"row": 1, "col": 3, "perturbed": true, "cx": 700, "cy": 300, "radius": 70}, {"row": 2, "col": 0, "perturbed": true, "cx": 100, "cy": 500, "radius": 70}, {"row": 2, "col": 1, "perturbed": false, "cx": 300, "cy": 500, "radius": 70}, {"row": 2, "col": 2, "perturbed": true, "cx": 500, "cy": 500, "radius": 70}, {"row": 2, "col": 3, "perturbed": false, "cx": 700, "cy": 500, "radius": 70}, {"row": 3, "col": 0, "perturbed": true, "cx": 100, "cy": 700, "radius": 70}, {"row": 3, "col": 1, "perturbed": true, "cx": 300, "cy": 700, "radius": 70}, {"row": 3, "col": 2, "perturbed": true, "cx": 500, "cy": 700, "radius": 70}, {"row": 3, "col": 3, "perturbed": false, "cx": 700, "cy": 700, "radius": 70}], "question": "Three panels are shown side by side. The leftmost panel is an INDEX key: a grid of cells numbered in row-major order (top-left = 1, increasing left-to-right then top-to-bottom). The middle and right panels are 2D color fields, each divided into the same grid. Compare the middle (Field A) and right (Field B) panels cell by cell. List the numbers of all cells that DIFFER between Field A and Field B, in ascending order, separated by commas. If no cells differ, write \"none\". Provide your final answer enclosed in ... tags.", "answer": "3, 5, 6, 7, 8, 9, 11, 13, 14, 15"} +{"image": "images/field_diff_00004.png", "color_mode": "coolwarm", "grid": 4, "num_cells": 16, "num_identical": 4, "num_different": 12, "diff_indices": [1, 2, 3, 5, 7, 8, 9, 10, 11, 12, 13, 14], "cells": [{"row": 0, "col": 0, "perturbed": true, "cx": 100, "cy": 100, "radius": 70}, {"row": 0, "col": 1, "perturbed": true, "cx": 300, "cy": 100, "radius": 70}, {"row": 0, "col": 2, "perturbed": true, "cx": 500, "cy": 100, "radius": 70}, {"row": 0, "col": 3, "perturbed": false, "cx": 700, "cy": 100, "radius": 70}, {"row": 1, "col": 0, "perturbed": true, "cx": 100, "cy": 300, "radius": 70}, {"row": 1, "col": 1, "perturbed": false, "cx": 300, "cy": 300, "radius": 70}, {"row": 1, "col": 2, "perturbed": true, "cx": 500, "cy": 300, "radius": 70}, {"row": 1, "col": 3, "perturbed": true, "cx": 700, "cy": 300, "radius": 70}, {"row": 2, "col": 0, "perturbed": true, "cx": 100, "cy": 500, "radius": 70}, {"row": 2, "col": 1, "perturbed": true, "cx": 300, "cy": 500, "radius": 70}, {"row": 2, "col": 2, "perturbed": true, "cx": 500, "cy": 500, "radius": 70}, {"row": 2, "col": 3, "perturbed": true, "cx": 700, "cy": 500, "radius": 70}, {"row": 3, "col": 0, "perturbed": true, "cx": 100, "cy": 700, "radius": 70}, {"row": 3, "col": 1, "perturbed": true, "cx": 300, "cy": 700, "radius": 70}, {"row": 3, "col": 2, "perturbed": false, "cx": 500, "cy": 700, "radius": 70}, {"row": 3, "col": 3, "perturbed": false, "cx": 700, "cy": 700, "radius": 70}], "question": "Three panels are shown side by side. The leftmost panel is an INDEX key: a grid of cells numbered in row-major order (top-left = 1, increasing left-to-right then top-to-bottom). The middle and right panels are 2D color fields, each divided into the same grid. Compare the middle (Field A) and right (Field B) panels cell by cell. List the numbers of all cells that DIFFER between Field A and Field B, in ascending order, separated by commas. If no cells differ, write \"none\". Provide your final answer enclosed in ... tags.", "answer": "1, 2, 3, 5, 7, 8, 9, 10, 11, 12, 13, 14"} diff --git a/data/annotations/spot_the_signal_diff.jsonl b/data/annotations/spot_the_signal_diff.jsonl new file mode 100644 index 0000000000000000000000000000000000000000..ce4cd78ec7637b6e18d1842ab2e60a27850aa24b --- /dev/null +++ b/data/annotations/spot_the_signal_diff.jsonl @@ -0,0 +1,5 @@ +{"image": "images/signal_diff_00000.png", "num_segments": 10, "num_identical": 7, "num_different": 3, "diff_indices": [7, 8, 9], "segments": [{"segment": 0, "seg_start_index": 0, "seg_end_index": 200, "seg_x_start": 0.0, "seg_x_end": 199.09954977488746, "perturbed": false}, {"segment": 1, "seg_start_index": 200, "seg_end_index": 400, "seg_x_start": 200.10005002501254, "seg_x_end": 399.1995997999, "perturbed": false}, {"segment": 2, "seg_start_index": 400, "seg_end_index": 600, "seg_x_start": 400.20010005002507, "seg_x_end": 599.2996498249125, "perturbed": false}, {"segment": 3, "seg_start_index": 600, "seg_end_index": 800, "seg_x_start": 600.3001500750375, "seg_x_end": 799.399699849925, "perturbed": false}, {"segment": 4, "seg_start_index": 800, "seg_end_index": 1000, "seg_x_start": 800.4002001000501, "seg_x_end": 999.4997498749375, "perturbed": false}, {"segment": 5, "seg_start_index": 1000, "seg_end_index": 1200, "seg_x_start": 1000.5002501250626, "seg_x_end": 1199.59979989995, "perturbed": false}, {"segment": 6, "seg_start_index": 1200, "seg_end_index": 1400, "seg_x_start": 1200.600300150075, "seg_x_end": 1399.6998499249626, "perturbed": true, "perturb_start_index": 1245, "perturb_end_index": 1355, "perturb_x_start": 1245.622811405703, "perturb_x_end": 1354.6773386693349}, {"segment": 7, "seg_start_index": 1400, "seg_end_index": 1600, "seg_x_start": 1400.7003501750876, "seg_x_end": 1599.799899949975, "perturbed": true, "perturb_start_index": 1445, "perturb_end_index": 1555, "perturb_x_start": 1445.7228614307155, "perturb_x_end": 1554.7773886943473}, {"segment": 8, "seg_start_index": 1600, "seg_end_index": 1800, "seg_x_start": 1600.8004002001003, "seg_x_end": 1799.8999499749875, "perturbed": true, "perturb_start_index": 1645, "perturb_end_index": 1755, "perturb_x_start": 1645.822911455728, "perturb_x_end": 1754.8774387193598}, {"segment": 9, "seg_start_index": 1800, "seg_end_index": 2000, "seg_x_start": 1800.9004502251128, "seg_x_end": 2000.0, "perturbed": false}], "question": "Two 1D signals are shown side by side. Each signal's x-axis is divided into equal segments by thin vertical gridlines, and each segment is numbered above the gridlines (the two signals share the same numbering). Compare the left (Signal A) and right (Signal B) signals segment by segment. List the numbers of all segments that DIFFER between Signal A and Signal B, in ascending order, separated by commas. If no segments differ, write \"none\". Provide your final answer enclosed in ... tags.", "answer": "7, 8, 9"} +{"image": "images/signal_diff_00001.png", "num_segments": 10, "num_identical": 5, "num_different": 5, "diff_indices": [2, 5, 7, 8, 10], "segments": [{"segment": 0, "seg_start_index": 0, "seg_end_index": 200, "seg_x_start": 0.0, "seg_x_end": 199.09954977488746, "perturbed": false}, {"segment": 1, "seg_start_index": 200, "seg_end_index": 400, "seg_x_start": 200.10005002501254, "seg_x_end": 399.1995997999, "perturbed": true, "perturb_start_index": 245, "perturb_end_index": 355, "perturb_x_start": 245.12256128064033, "perturb_x_end": 354.1770885442722}, {"segment": 2, "seg_start_index": 400, "seg_end_index": 600, "seg_x_start": 400.20010005002507, "seg_x_end": 599.2996498249125, "perturbed": false}, {"segment": 3, "seg_start_index": 600, "seg_end_index": 800, "seg_x_start": 600.3001500750375, "seg_x_end": 799.399699849925, "perturbed": false}, {"segment": 4, "seg_start_index": 800, "seg_end_index": 1000, "seg_x_start": 800.4002001000501, "seg_x_end": 999.4997498749375, "perturbed": true, "perturb_start_index": 845, "perturb_end_index": 955, "perturb_x_start": 845.4227113556778, "perturb_x_end": 954.4772386193097}, {"segment": 5, "seg_start_index": 1000, "seg_end_index": 1200, "seg_x_start": 1000.5002501250626, "seg_x_end": 1199.59979989995, "perturbed": false}, {"segment": 6, "seg_start_index": 1200, "seg_end_index": 1400, "seg_x_start": 1200.600300150075, "seg_x_end": 1399.6998499249626, "perturbed": true, "perturb_start_index": 1245, "perturb_end_index": 1355, "perturb_x_start": 1245.622811405703, "perturb_x_end": 1354.6773386693349}, {"segment": 7, "seg_start_index": 1400, "seg_end_index": 1600, "seg_x_start": 1400.7003501750876, "seg_x_end": 1599.799899949975, "perturbed": true, "perturb_start_index": 1445, "perturb_end_index": 1555, "perturb_x_start": 1445.7228614307155, "perturb_x_end": 1554.7773886943473}, {"segment": 8, "seg_start_index": 1600, "seg_end_index": 1800, "seg_x_start": 1600.8004002001003, "seg_x_end": 1799.8999499749875, "perturbed": false}, {"segment": 9, "seg_start_index": 1800, "seg_end_index": 2000, "seg_x_start": 1800.9004502251128, "seg_x_end": 2000.0, "perturbed": true, "perturb_start_index": 1845, "perturb_end_index": 1955, "perturb_x_start": 1845.9229614807405, "perturb_x_end": 1954.9774887443723}], "question": "Two 1D signals are shown side by side. Each signal's x-axis is divided into equal segments by thin vertical gridlines, and each segment is numbered above the gridlines (the two signals share the same numbering). Compare the left (Signal A) and right (Signal B) signals segment by segment. List the numbers of all segments that DIFFER between Signal A and Signal B, in ascending order, separated by commas. If no segments differ, write \"none\". Provide your final answer enclosed in ... tags.", "answer": "2, 5, 7, 8, 10"} +{"image": "images/signal_diff_00002.png", "num_segments": 10, "num_identical": 9, "num_different": 1, "diff_indices": [8], "segments": [{"segment": 0, "seg_start_index": 0, "seg_end_index": 200, "seg_x_start": 0.0, "seg_x_end": 199.09954977488746, "perturbed": false}, {"segment": 1, "seg_start_index": 200, "seg_end_index": 400, "seg_x_start": 200.10005002501254, "seg_x_end": 399.1995997999, "perturbed": false}, {"segment": 2, "seg_start_index": 400, "seg_end_index": 600, "seg_x_start": 400.20010005002507, "seg_x_end": 599.2996498249125, "perturbed": false}, {"segment": 3, "seg_start_index": 600, "seg_end_index": 800, "seg_x_start": 600.3001500750375, "seg_x_end": 799.399699849925, "perturbed": false}, {"segment": 4, "seg_start_index": 800, "seg_end_index": 1000, "seg_x_start": 800.4002001000501, "seg_x_end": 999.4997498749375, "perturbed": false}, {"segment": 5, "seg_start_index": 1000, "seg_end_index": 1200, "seg_x_start": 1000.5002501250626, "seg_x_end": 1199.59979989995, "perturbed": false}, {"segment": 6, "seg_start_index": 1200, "seg_end_index": 1400, "seg_x_start": 1200.600300150075, "seg_x_end": 1399.6998499249626, "perturbed": false}, {"segment": 7, "seg_start_index": 1400, "seg_end_index": 1600, "seg_x_start": 1400.7003501750876, "seg_x_end": 1599.799899949975, "perturbed": true, "perturb_start_index": 1445, "perturb_end_index": 1555, "perturb_x_start": 1445.7228614307155, "perturb_x_end": 1554.7773886943473}, {"segment": 8, "seg_start_index": 1600, "seg_end_index": 1800, "seg_x_start": 1600.8004002001003, "seg_x_end": 1799.8999499749875, "perturbed": false}, {"segment": 9, "seg_start_index": 1800, "seg_end_index": 2000, "seg_x_start": 1800.9004502251128, "seg_x_end": 2000.0, "perturbed": false}], "question": "Two 1D signals are shown side by side. Each signal's x-axis is divided into equal segments by thin vertical gridlines, and each segment is numbered above the gridlines (the two signals share the same numbering). Compare the left (Signal A) and right (Signal B) signals segment by segment. List the numbers of all segments that DIFFER between Signal A and Signal B, in ascending order, separated by commas. If no segments differ, write \"none\". Provide your final answer enclosed in ... tags.", "answer": "8"} +{"image": "images/signal_diff_00003.png", "num_segments": 10, "num_identical": 6, "num_different": 4, "diff_indices": [3, 4, 6, 9], "segments": [{"segment": 0, "seg_start_index": 0, "seg_end_index": 200, "seg_x_start": 0.0, "seg_x_end": 199.09954977488746, "perturbed": false}, {"segment": 1, "seg_start_index": 200, "seg_end_index": 400, "seg_x_start": 200.10005002501254, "seg_x_end": 399.1995997999, "perturbed": false}, {"segment": 2, "seg_start_index": 400, "seg_end_index": 600, "seg_x_start": 400.20010005002507, "seg_x_end": 599.2996498249125, "perturbed": true, "perturb_start_index": 445, "perturb_end_index": 555, "perturb_x_start": 445.22261130565283, "perturb_x_end": 554.2771385692847}, {"segment": 3, "seg_start_index": 600, "seg_end_index": 800, "seg_x_start": 600.3001500750375, "seg_x_end": 799.399699849925, "perturbed": true, "perturb_start_index": 645, "perturb_end_index": 755, "perturb_x_start": 645.3226613306654, "perturb_x_end": 754.3771885942972}, {"segment": 4, "seg_start_index": 800, "seg_end_index": 1000, "seg_x_start": 800.4002001000501, "seg_x_end": 999.4997498749375, "perturbed": false}, {"segment": 5, "seg_start_index": 1000, "seg_end_index": 1200, "seg_x_start": 1000.5002501250626, "seg_x_end": 1199.59979989995, "perturbed": true, "perturb_start_index": 1045, "perturb_end_index": 1155, "perturb_x_start": 1045.5227613806903, "perturb_x_end": 1154.5772886443222}, {"segment": 6, "seg_start_index": 1200, "seg_end_index": 1400, "seg_x_start": 1200.600300150075, "seg_x_end": 1399.6998499249626, "perturbed": false}, {"segment": 7, "seg_start_index": 1400, "seg_end_index": 1600, "seg_x_start": 1400.7003501750876, "seg_x_end": 1599.799899949975, "perturbed": false}, {"segment": 8, "seg_start_index": 1600, "seg_end_index": 1800, "seg_x_start": 1600.8004002001003, "seg_x_end": 1799.8999499749875, "perturbed": true, "perturb_start_index": 1645, "perturb_end_index": 1755, "perturb_x_start": 1645.822911455728, "perturb_x_end": 1754.8774387193598}, {"segment": 9, "seg_start_index": 1800, "seg_end_index": 2000, "seg_x_start": 1800.9004502251128, "seg_x_end": 2000.0, "perturbed": false}], "question": "Two 1D signals are shown side by side. Each signal's x-axis is divided into equal segments by thin vertical gridlines, and each segment is numbered above the gridlines (the two signals share the same numbering). Compare the left (Signal A) and right (Signal B) signals segment by segment. List the numbers of all segments that DIFFER between Signal A and Signal B, in ascending order, separated by commas. If no segments differ, write \"none\". Provide your final answer enclosed in ... tags.", "answer": "3, 4, 6, 9"} +{"image": "images/signal_diff_00004.png", "num_segments": 10, "num_identical": 5, "num_different": 5, "diff_indices": [2, 3, 5, 6, 10], "segments": [{"segment": 0, "seg_start_index": 0, "seg_end_index": 200, "seg_x_start": 0.0, "seg_x_end": 199.09954977488746, "perturbed": false}, {"segment": 1, "seg_start_index": 200, "seg_end_index": 400, "seg_x_start": 200.10005002501254, "seg_x_end": 399.1995997999, "perturbed": true, "perturb_start_index": 245, "perturb_end_index": 355, "perturb_x_start": 245.12256128064033, "perturb_x_end": 354.1770885442722}, {"segment": 2, "seg_start_index": 400, "seg_end_index": 600, "seg_x_start": 400.20010005002507, "seg_x_end": 599.2996498249125, "perturbed": true, "perturb_start_index": 445, "perturb_end_index": 555, "perturb_x_start": 445.22261130565283, "perturb_x_end": 554.2771385692847}, {"segment": 3, "seg_start_index": 600, "seg_end_index": 800, "seg_x_start": 600.3001500750375, "seg_x_end": 799.399699849925, "perturbed": false}, {"segment": 4, "seg_start_index": 800, "seg_end_index": 1000, "seg_x_start": 800.4002001000501, "seg_x_end": 999.4997498749375, "perturbed": true, "perturb_start_index": 845, "perturb_end_index": 955, "perturb_x_start": 845.4227113556778, "perturb_x_end": 954.4772386193097}, {"segment": 5, "seg_start_index": 1000, "seg_end_index": 1200, "seg_x_start": 1000.5002501250626, "seg_x_end": 1199.59979989995, "perturbed": true, "perturb_start_index": 1045, "perturb_end_index": 1155, "perturb_x_start": 1045.5227613806903, "perturb_x_end": 1154.5772886443222}, {"segment": 6, "seg_start_index": 1200, "seg_end_index": 1400, "seg_x_start": 1200.600300150075, "seg_x_end": 1399.6998499249626, "perturbed": false}, {"segment": 7, "seg_start_index": 1400, "seg_end_index": 1600, "seg_x_start": 1400.7003501750876, "seg_x_end": 1599.799899949975, "perturbed": false}, {"segment": 8, "seg_start_index": 1600, "seg_end_index": 1800, "seg_x_start": 1600.8004002001003, "seg_x_end": 1799.8999499749875, "perturbed": false}, {"segment": 9, "seg_start_index": 1800, "seg_end_index": 2000, "seg_x_start": 1800.9004502251128, "seg_x_end": 2000.0, "perturbed": true, "perturb_start_index": 1845, "perturb_end_index": 1955, "perturb_x_start": 1845.9229614807405, "perturb_x_end": 1954.9774887443723}], "question": "Two 1D signals are shown side by side. Each signal's x-axis is divided into equal segments by thin vertical gridlines, and each segment is numbered above the gridlines (the two signals share the same numbering). Compare the left (Signal A) and right (Signal B) signals segment by segment. List the numbers of all segments that DIFFER between Signal A and Signal B, in ascending order, separated by commas. If no segments differ, write \"none\". Provide your final answer enclosed in ... tags.", "answer": "2, 3, 5, 6, 10"} diff --git a/data/annotations/spot_the_stroke_diff.jsonl b/data/annotations/spot_the_stroke_diff.jsonl new file mode 100644 index 0000000000000000000000000000000000000000..f752fa394bc969b5aca4ae59c49273ff3521f494 --- /dev/null +++ b/data/annotations/spot_the_stroke_diff.jsonl @@ -0,0 +1,5 @@ +{"image": "images/stroke_diff_00000.png", "num_items": 20, "num_differences": 5, "differences": [{"index": 0, "label": "A", "x": 1183.272085511193, "y": 569.7778843142105, "bounding_radius": 52.786627836522484}, {"index": 8, "label": "I", "x": 1001.3330632321204, "y": 686.9539152538724, "bounding_radius": 60.8630185389318}, {"index": 10, "label": "K", "x": 584.0635458779055, "y": 1142.8835740822976, "bounding_radius": 69.19654770255644}, {"index": 14, "label": "O", "x": 266.9088179210339, "y": 1030.6567995291518, "bounding_radius": 68.14976911729399}, {"index": 15, "label": "P", "x": 1137.5044245193822, "y": 228.74805438199496, "bounding_radius": 58.4530305679033}], "question": "Two panels are shown side by side. Each panel contains the same number of elongated pieces (strands, twigs, threads, etc.) at the same positions. Some pieces in the right panel have a different shape from the corresponding piece in the left panel. Count the number of pairs whose shape differs between the two panels and report the integer count. Provide your final answer enclosed in ... tags.", "answer": "5"} +{"image": "images/stroke_diff_00001.png", "num_items": 20, "num_differences": 8, "differences": [{"index": 0, "label": "A", "x": 1115.286433909286, "y": 1175.1404270759583, "bounding_radius": 40.507843102258605}, {"index": 1, "label": "B", "x": 594.4315205885941, "y": 496.5680967206435, "bounding_radius": 73.49958652783556}, {"index": 4, "label": "E", "x": 456.2057963049053, "y": 586.6863211442076, "bounding_radius": 63.73820004097872}, {"index": 5, "label": "F", "x": 654.9414309785338, "y": 119.11166278560889, "bounding_radius": 53.66431804944811}, {"index": 7, "label": "H", "x": 952.5389019608119, "y": 473.70884792979757, "bounding_radius": 66.0639271398448}, {"index": 8, "label": "I", "x": 629.9062618552024, "y": 739.2607654998784, "bounding_radius": 62.18620342209474}, {"index": 14, "label": "O", "x": 457.92496128072503, "y": 103.71739643243247, "bounding_radius": 52.730109537649156}, {"index": 16, "label": "Q", "x": 776.2463223350809, "y": 656.1347293141475, "bounding_radius": 69.60983070647463}], "question": "Two panels are shown side by side. Each panel contains the same number of elongated pieces (strands, twigs, threads, etc.) at the same positions. Some pieces in the right panel have a different shape from the corresponding piece in the left panel. Count the number of pairs whose shape differs between the two panels and report the integer count. Provide your final answer enclosed in ... tags.", "answer": "8"} +{"image": "images/stroke_diff_00002.png", "num_items": 20, "num_differences": 10, "differences": [{"index": 1, "label": "B", "x": 696.0406328049361, "y": 850.5457019872772, "bounding_radius": 65.19177599198267}, {"index": 2, "label": "C", "x": 850.7740473852028, "y": 867.9248096619145, "bounding_radius": 52.118256699260755}, {"index": 9, "label": "J", "x": 142.97002870583503, "y": 439.24204864610283, "bounding_radius": 48.50416603963236}, {"index": 11, "label": "L", "x": 920.9435854297501, "y": 1166.623515699951, "bounding_radius": 75.34023401342182}, {"index": 12, "label": "M", "x": 1096.7015997475107, "y": 928.5384927907618, "bounding_radius": 56.33061909666607}, {"index": 13, "label": "N", "x": 832.4025849675247, "y": 88.45677741763801, "bounding_radius": 56.51663207349122}, {"index": 15, "label": "P", "x": 740.8017475516116, "y": 607.4995878112061, "bounding_radius": 50.33756013078856}, {"index": 16, "label": "Q", "x": 766.0673288902601, "y": 236.46798810042975, "bounding_radius": 73.06736299559509}, {"index": 17, "label": "R", "x": 1169.1073450564497, "y": 1047.6909336311583, "bounding_radius": 55.870295480847204}, {"index": 18, "label": "S", "x": 557.4946901128425, "y": 322.09099021370344, "bounding_radius": 83.11137983154067}], "question": "Two panels are shown side by side. Each panel contains the same number of elongated pieces (strands, twigs, threads, etc.) at the same positions. Some pieces in the right panel have a different shape from the corresponding piece in the left panel. Count the number of pairs whose shape differs between the two panels and report the integer count. Provide your final answer enclosed in ... tags.", "answer": "10"} +{"image": "images/stroke_diff_00003.png", "num_items": 20, "num_differences": 12, "differences": [{"index": 1, "label": "B", "x": 565.3628318062126, "y": 470.77510003431337, "bounding_radius": 62.73359404793353}, {"index": 3, "label": "D", "x": 349.73201009555873, "y": 1190.492404322952, "bounding_radius": 58.483392467314744}, {"index": 4, "label": "E", "x": 1123.9536853568704, "y": 1122.5513348995182, "bounding_radius": 64.9971254205356}, {"index": 5, "label": "F", "x": 954.5241638962042, "y": 930.7226195104262, "bounding_radius": 54.17288897799781}, {"index": 9, "label": "J", "x": 534.9270830019888, "y": 114.70003883756132, "bounding_radius": 72.89384833373377}, {"index": 10, "label": "K", "x": 210.96139466226336, "y": 779.3990125533917, "bounding_radius": 84.50218960788088}, {"index": 11, "label": "L", "x": 843.6578185033873, "y": 981.5980671943058, "bounding_radius": 41.49196060321283}, {"index": 12, "label": "M", "x": 270.8212842468754, "y": 1057.339460692566, "bounding_radius": 59.76520945735085}, {"index": 13, "label": "N", "x": 783.1206563423287, "y": 516.3687210109275, "bounding_radius": 57.897352354344456}, {"index": 14, "label": "O", "x": 1037.3834627291649, "y": 501.8120311286603, "bounding_radius": 79.69653626711077}, {"index": 15, "label": "P", "x": 1131.9157872080248, "y": 332.9286246579703, "bounding_radius": 64.26400210205598}, {"index": 16, "label": "Q", "x": 1103.501814589747, "y": 901.8619694753024, "bounding_radius": 54.936592601575484}], "question": "Two panels are shown side by side. Each panel contains the same number of elongated pieces (strands, twigs, threads, etc.) at the same positions. Some pieces in the right panel have a different shape from the corresponding piece in the left panel. Count the number of pairs whose shape differs between the two panels and report the integer count. Provide your final answer enclosed in ... tags.", "answer": "12"} +{"image": "images/stroke_diff_00004.png", "num_items": 20, "num_differences": 15, "differences": [{"index": 0, "label": "A", "x": 364.229207254163, "y": 889.3758711897652, "bounding_radius": 51.67997849740728}, {"index": 1, "label": "B", "x": 189.68093723991004, "y": 1168.8713714976263, "bounding_radius": 49.14980520281939}, {"index": 3, "label": "D", "x": 935.6871119602484, "y": 1001.1703322576201, "bounding_radius": 70.5527445623121}, {"index": 4, "label": "E", "x": 549.9840675335242, "y": 690.424760279989, "bounding_radius": 60.363392692760634}, {"index": 5, "label": "F", "x": 144.94361986836378, "y": 148.56228749203535, "bounding_radius": 64.10234842359652}, {"index": 6, "label": "G", "x": 279.39180144282864, "y": 1066.459024436862, "bounding_radius": 39.29773083625673}, {"index": 7, "label": "H", "x": 577.7548658637684, "y": 1126.293678869781, "bounding_radius": 71.34509444956434}, {"index": 8, "label": "I", "x": 668.0941466779606, "y": 864.4679187446649, "bounding_radius": 65.39159638098933}, {"index": 9, "label": "J", "x": 369.0159756623604, "y": 399.2701124912714, "bounding_radius": 82.5492376303858}, {"index": 12, "label": "M", "x": 1099.086695283693, "y": 1191.556953934657, "bounding_radius": 53.79585711012711}, {"index": 13, "label": "N", "x": 182.5748456893859, "y": 830.3720880328077, "bounding_radius": 59.401180874869475}, {"index": 14, "label": "O", "x": 637.718403989465, "y": 239.83595985576508, "bounding_radius": 54.41162025757867}, {"index": 17, "label": "R", "x": 1158.1619210290435, "y": 167.0115828335808, "bounding_radius": 55.117083680557975}, {"index": 18, "label": "S", "x": 1187.1088498000825, "y": 949.9514933146202, "bounding_radius": 48.8775771145461}, {"index": 19, "label": "T", "x": 715.6744341364719, "y": 701.8997683771095, "bounding_radius": 59.06291828365624}], "question": "Two panels are shown side by side. Each panel contains the same number of elongated pieces (strands, twigs, threads, etc.) at the same positions. Some pieces in the right panel have a different shape from the corresponding piece in the left panel. Count the number of pairs whose shape differs between the two panels and report the integer count. Provide your final answer enclosed in ... tags.", "answer": "15"} diff --git a/data/annotations/stroke_gesture_count.jsonl b/data/annotations/stroke_gesture_count.jsonl new file mode 100644 index 0000000000000000000000000000000000000000..cbb06877f8dc05d0d34a46f01b25f91ccf3ad576 --- /dev/null +++ b/data/annotations/stroke_gesture_count.jsonl @@ -0,0 +1,5 @@ +{"image": "images/stroke_gesture_count_00000.png", "question": "This image has two panels separated by a thin vertical divider. The left panel shows the Template: a single elongated piece with a specific curvature profile. The right panel shows the Field: many elongated pieces of similar thickness scattered across the canvas. Both panels use the SAME pixel scale. Count the number of pieces in the Field that have the exact same shape as the Template (translation only \u2014 same curvature, same orientation, no rotation or mirroring) and report the integer count. Provide your final answer enclosed in ... tags.", "answer": "5", "num_matches": 5, "matching_labels": ["A", "D", "V", "X", "AT"], "total_strokes": 46, "metadata": {"seed": 1806341205, "template_control_points": [[-66.55486597185785, 16.085256376174193], [-24.278162140806785, -27.323924822020587], [48.17117983381231, 31.903219680759342], [28.52056660025756, 23.69733723731391]], "template_arc_length": 198.07439477803888}} +{"image": "images/stroke_gesture_count_00001.png", "question": "This image has two panels separated by a thin vertical divider. The left panel shows the Template: a single elongated piece with a specific curvature profile. The right panel shows the Field: many elongated pieces of similar thickness scattered across the canvas. Both panels use the SAME pixel scale. Count the number of pieces in the Field that have the exact same shape as the Template (translation only \u2014 same curvature, same orientation, no rotation or mirroring) and report the integer count. Provide your final answer enclosed in ... tags.", "answer": "8", "num_matches": 8, "matching_labels": ["A", "B", "C", "H", "U", "Z", "AI", "AN"], "total_strokes": 41, "metadata": {"seed": 1712934065, "template_control_points": [[-77.86385218338921, -12.604188800149451], [-49.535287993337995, -5.891695262338526], [-19.63836101282533, -9.422801727457234], [62.12764718751154, 62.16519116595446]], "template_arc_length": 188.75627164371588}} +{"image": "images/stroke_gesture_count_00002.png", "question": "This image has two panels separated by a thin vertical divider. The left panel shows the Template: a single elongated piece with a specific curvature profile. The right panel shows the Field: many elongated pieces of similar thickness scattered across the canvas. Both panels use the SAME pixel scale. Count the number of pieces in the Field that have the exact same shape as the Template (translation only \u2014 same curvature, same orientation, no rotation or mirroring) and report the integer count. Provide your final answer enclosed in ... tags.", "answer": "10", "num_matches": 10, "matching_labels": ["F", "H", "O", "X", "Y", "AI", "AJ", "AK", "AL", "AP"], "total_strokes": 44, "metadata": {"seed": 789710164, "template_control_points": [[-30.378174474215683, -16.505150636502307], [-17.063197201255036, 8.159457749210215], [13.54887764758545, 7.1511961398748625], [18.40984663054958, -12.664839905287089], [25.424568834037096, 7.1443621701249995]], "template_arc_length": 108.86223215044478}} +{"image": "images/stroke_gesture_count_00003.png", "question": "This image has two panels separated by a thin vertical divider. The left panel shows the Template: a single elongated piece with a specific curvature profile. The right panel shows the Field: many elongated pieces of similar thickness scattered across the canvas. Both panels use the SAME pixel scale. Count the number of pieces in the Field that have the exact same shape as the Template (translation only \u2014 same curvature, same orientation, no rotation or mirroring) and report the integer count. Provide your final answer enclosed in ... tags.", "answer": "12", "num_matches": 12, "matching_labels": ["C", "G", "V", "X", "AC", "AH", "AI", "AJ", "AM", "AO", "AP", "AQ"], "total_strokes": 45, "metadata": {"seed": 732213632, "template_control_points": [[-57.61848404506308, 17.867423319566896], [-33.24807698174301, 4.74203273755055], [9.926269958552135, 7.3665532829045315], [26.987059367428763, -14.583106342140283], [8.879367310127217, 32.192621233437194]], "template_arc_length": 168.08698660845386}} +{"image": "images/stroke_gesture_count_00004.png", "question": "This image has two panels separated by a thin vertical divider. The left panel shows the Template: a single elongated piece with a specific curvature profile. The right panel shows the Field: many elongated pieces of similar thickness scattered across the canvas. Both panels use the SAME pixel scale. Count the number of pieces in the Field that have the exact same shape as the Template (translation only \u2014 same curvature, same orientation, no rotation or mirroring) and report the integer count. Provide your final answer enclosed in ... tags.", "answer": "15", "num_matches": 15, "matching_labels": ["A", "E", "O", "P", "Q", "T", "U", "V", "Y", "AE", "AF", "AH", "AN", "AO", "AP"], "total_strokes": 46, "metadata": {"seed": 856226606, "template_control_points": [[-55.33878173562434, -35.28033609360301], [-17.276457950348778, 6.408432265668017], [24.527879198738688, 26.69385912896616], [35.50841016932807, -1.7883523737386202], [53.6277057470718, -21.297061647888007]], "template_arc_length": 165.02906785597352}} diff --git a/data/annotations/tangled_loops.jsonl b/data/annotations/tangled_loops.jsonl new file mode 100644 index 0000000000000000000000000000000000000000..e9375e6417fd707f1352eceb892c20cf4075c7bb --- /dev/null +++ b/data/annotations/tangled_loops.jsonl @@ -0,0 +1,5 @@ +{"width": 1448, "height": 1448, "num_loops": 10, "inter_loop_crossings": 28, "question": "How many distinct closed loops are tangled together in this image? Each loop is a continuous closed curve (rope, cord, cable, etc.) with no loose endpoints anywhere. Loops may cross other loops or themselves freely. Count the total number of distinct closed loops and report the count as a positive integer. Provide your final answer enclosed in ... tags.", "answer": 10, "image": "images/tangled_loops_00000.png"} +{"width": 1448, "height": 1448, "num_loops": 12, "inter_loop_crossings": 36, "question": "How many distinct closed loops are tangled together in this image? Each loop is a continuous closed curve (rope, cord, cable, etc.) with no loose endpoints anywhere. Loops may cross other loops or themselves freely. Count the total number of distinct closed loops and report the count as a positive integer. Provide your final answer enclosed in ... tags.", "answer": 12, "image": "images/tangled_loops_00001.png"} +{"width": 1448, "height": 1448, "num_loops": 15, "inter_loop_crossings": 54, "question": "How many distinct closed loops are tangled together in this image? Each loop is a continuous closed curve (rope, cord, cable, etc.) with no loose endpoints anywhere. Loops may cross other loops or themselves freely. Count the total number of distinct closed loops and report the count as a positive integer. Provide your final answer enclosed in ... tags.", "answer": 15, "image": "images/tangled_loops_00002.png"} +{"width": 1448, "height": 1448, "num_loops": 18, "inter_loop_crossings": 64, "question": "How many distinct closed loops are tangled together in this image? Each loop is a continuous closed curve (rope, cord, cable, etc.) with no loose endpoints anywhere. Loops may cross other loops or themselves freely. Count the total number of distinct closed loops and report the count as a positive integer. Provide your final answer enclosed in ... tags.", "answer": 18, "image": "images/tangled_loops_00003.png"} +{"width": 1448, "height": 1448, "num_loops": 20, "inter_loop_crossings": 92, "question": "How many distinct closed loops are tangled together in this image? Each loop is a continuous closed curve (rope, cord, cable, etc.) with no loose endpoints anywhere. Loops may cross other loops or themselves freely. Count the total number of distinct closed loops and report the count as a positive integer. Provide your final answer enclosed in ... tags.", "answer": 20, "image": "images/tangled_loops_00004.png"} diff --git a/data/annotations/traverse_ordering.jsonl b/data/annotations/traverse_ordering.jsonl new file mode 100644 index 0000000000000000000000000000000000000000..390b16320626c403ff9a0f88f95438b480318164 --- /dev/null +++ b/data/annotations/traverse_ordering.jsonl @@ -0,0 +1,5 @@ +{"width": 1916, "height": 1916, "num_extra_distractor_curves": 1, "s_curve": 1, "s_at_start": false, "s_point": {"x": 68.96, "y": 1314.21, "label_x": 73.87, "label_y": 1262.44}, "marks": [{"label": "K", "curve": 1, "on_s_curve": true, "polyline_index": 246, "x": 1611.31, "y": 804.2, "label_x": 1653.65, "label_y": 826.81}, {"label": "C", "curve": 1, "on_s_curve": true, "polyline_index": 451, "x": 1104.9, "y": 643.38, "label_x": 1058.61, "label_y": 630.68}, {"label": "F", "curve": 1, "on_s_curve": true, "polyline_index": 633, "x": 1374.36, "y": 1117.2, "label_x": 1414.95, "label_y": 1091.58}, {"label": "H", "curve": 1, "on_s_curve": true, "polyline_index": 917, "x": 1260.45, "y": 1340.31, "label_x": 1296.47, "label_y": 1308.59}, {"label": "G", "curve": 1, "on_s_curve": true, "polyline_index": 1173, "x": 657.5, "y": 868.21, "label_x": 621.44, "label_y": 899.89}, {"label": "L", "curve": 1, "on_s_curve": true, "polyline_index": 1356, "x": 257.07, "y": 713.26, "label_x": 209.94, "label_y": 722.37}, {"label": "A", "curve": 0, "on_s_curve": false, "polyline_index": 220, "x": 514.05, "y": 551.68, "label_x": 506.74, "label_y": 599.12}, {"label": "J", "curve": 0, "on_s_curve": false, "polyline_index": 391, "x": 926.29, "y": 814.27, "label_x": 883.52, "label_y": 792.48}, {"label": "D", "curve": 0, "on_s_curve": false, "polyline_index": 532, "x": 1073.61, "y": 966.48, "label_x": 1104.9, "label_y": 1002.89}, {"label": "I", "curve": 0, "on_s_curve": false, "polyline_index": 754, "x": 1635.33, "y": 586.44, "label_x": 1668.44, "label_y": 551.69}, {"label": "B", "curve": 0, "on_s_curve": false, "polyline_index": 973, "x": 1198.44, "y": 1075.53, "label_x": 1165.27, "label_y": 1040.82}, {"label": "E", "curve": 0, "on_s_curve": false, "polyline_index": 1130, "x": 1111.3, "y": 1453.29, "label_x": 1109.18, "label_y": 1501.25}], "num_marks_total": 12, "num_marks_on_s": 6, "num_pair_crossings": 9, "num_self_crossings": 0, "question": "The image shows two curves or strands that cross each other. Several labeled points are marked on the two curves combined. A green 'S' marks the start endpoint of one of the curves. Starting from S, follow THAT curve (the one S lies on) all the way to its other endpoint, and list the labels of the marked points you visit in order, separated by commas. Ignore any labeled points that lie on the OTHER curve. Provide your final answer enclosed in ... tags.", "answer": "L, G, H, F, C, K", "polylines": [[[158.46, 70.0], [158.32, 73.22], [158.19, 76.44], [158.05, 79.67], [157.92, 82.89], [157.78, 86.11], [157.64, 89.33], [157.51, 92.55], [157.37, 95.78], [157.24, 99.0], [157.1, 102.22], [156.97, 105.44], [156.83, 108.66], [156.7, 111.89], [156.56, 115.11], [156.43, 118.33], [156.29, 121.55], [156.16, 124.77], [156.02, 128.0], [155.89, 131.22], [155.75, 134.44], [155.62, 137.66], [155.48, 140.89], [155.35, 144.11], [155.21, 147.33], [155.07, 150.55], [154.94, 153.77], [154.8, 157.0], [154.67, 160.22], [154.53, 163.44], [154.4, 166.66], [154.26, 169.88], [154.13, 173.11], [153.99, 176.33], [153.86, 179.55], [153.72, 182.77], [153.59, 185.99], [153.45, 189.22], [153.32, 192.44], [153.18, 195.66], [153.05, 198.88], [152.91, 202.1], [152.78, 205.33], [152.64, 208.55], [152.5, 211.77], [152.37, 214.99], [152.23, 218.21], [152.23, 221.44], [152.35, 224.66], [152.6, 227.88], [152.99, 231.08], [153.5, 234.26], [154.13, 237.42], [154.9, 240.56], [155.79, 243.66], [156.8, 246.72], [157.93, 249.74], [159.19, 252.71], [160.56, 255.63], [162.05, 258.49], [163.65, 261.29], [165.37, 264.02], [167.19, 266.68], [169.12, 269.26], [171.15, 271.77], [173.27, 274.19], [175.5, 276.53], [177.81, 278.77], [180.22, 280.92], [182.7, 282.98], [185.27, 284.93], [187.89, 286.81], [190.57, 288.6], [193.32, 290.28], [196.14, 291.85], [199.02, 293.3], [201.93, 294.69], [204.86, 296.03], [207.81, 297.34], [210.77, 298.63], [213.76, 299.83], [216.77, 301.0], [219.79, 302.12], [222.84, 303.18], [225.89, 304.22], [228.96, 305.2], [232.04, 306.15], [235.15, 307.03], [238.24, 307.92], [241.36, 308.77], [244.48, 309.56], [247.62, 310.3], [250.77, 310.98], [253.94, 311.57], [257.13, 312.08], [260.33, 312.49], [263.53, 312.87], [266.74, 313.13], [269.97, 313.28], [273.19, 313.32], [276.41, 313.26], [279.64, 313.11], [282.85, 312.9], [286.07, 312.62], [289.29, 312.48], [292.51, 312.46], [295.74, 312.57], [298.95, 312.81], [302.16, 313.18], [305.34, 313.68], [308.5, 314.31], [311.64, 315.06], [314.74, 315.94], [317.81, 316.94], [320.83, 318.07], [323.81, 319.31], [326.73, 320.68], [329.59, 322.16], [332.4, 323.75], [335.14, 325.45], [337.8, 327.27], [340.39, 329.19], [342.91, 331.21], [345.34, 333.33], [347.68, 335.54], [349.93, 337.85], [352.09, 340.25], [354.15, 342.73], [356.11, 345.29], [357.96, 347.93], [359.71, 350.64], [361.34, 353.42], [362.94, 356.22], [364.51, 359.04], [366.03, 361.88], [367.5, 364.76], [368.87, 367.68], [370.15, 370.63], [371.38, 373.62], [372.55, 376.62], [373.71, 379.63], [374.84, 382.65], [375.93, 385.69], [377.01, 388.72], [378.09, 391.76], [379.15, 394.81], [380.17, 397.87], [381.16, 400.94], [382.08, 404.03], [382.91, 407.14], [383.62, 410.29], [384.23, 413.46], [384.71, 416.64], [385.06, 419.85], [385.31, 423.07], [385.47, 426.29], [385.57, 429.51], [385.63, 432.73], [385.64, 435.96], [385.61, 439.18], [385.54, 442.41], [385.46, 445.63], [385.39, 448.86], [385.38, 452.08], [385.4, 455.31], [385.44, 458.53], [385.55, 461.75], [385.67, 464.98], [385.92, 468.19], [386.3, 471.39], [386.81, 474.58], [387.45, 477.74], [388.21, 480.87], [389.1, 483.97], [390.11, 487.04], [391.24, 490.06], [392.49, 493.03], [393.87, 495.95], [395.35, 498.81], [396.95, 501.61], [398.67, 504.34], [400.49, 507.0], [402.41, 509.59], [404.44, 512.09], [406.57, 514.52], [408.79, 516.85], [411.11, 519.1], [413.51, 521.25], [416.0, 523.3], [418.57, 525.25], [421.21, 527.1], [423.93, 528.84], [426.71, 530.47], [429.56, 531.98], [432.46, 533.38], [435.42, 534.67], [438.42, 535.85], [441.43, 537.01], [444.47, 538.07], [447.56, 539.02], [450.66, 539.9], [453.77, 540.73], [456.91, 541.49], [460.05, 542.2], [463.21, 542.85], [466.37, 543.48], [469.54, 544.11], [472.71, 544.68], [475.89, 545.21], [479.07, 545.76], [482.24, 546.34], [485.41, 546.93], [488.59, 547.5], [491.77, 548.05], [494.94, 548.61], [498.12, 549.13], [501.31, 549.64], [504.49, 550.18], [507.67, 550.69], [510.86, 551.15], [514.05, 551.68], [517.23, 552.2], [520.41, 552.73], [523.59, 553.23], [526.79, 553.68], [529.99, 554.1], [533.19, 554.5], [536.39, 554.88], [539.59, 555.26], [542.79, 555.65], [545.99, 556.07], [549.18, 556.51], [552.38, 556.95], [555.57, 557.4], [558.74, 557.98], [561.9, 558.65], [565.05, 559.36], [568.19, 560.05], [571.34, 560.76], [574.49, 561.44], [577.64, 562.13], [580.79, 562.83], [583.94, 563.52], [587.09, 564.19], [590.26, 564.81], [593.43, 565.4], [596.6, 566.0], [599.77, 566.61], [602.91, 567.34], [606.02, 568.19], [609.1, 569.15], [612.15, 570.18], [615.17, 571.33], [618.16, 572.52], [621.14, 573.75], [624.13, 574.96], [627.14, 576.13], [630.13, 577.33], [633.14, 578.48], [636.15, 579.64], [639.14, 580.85], [642.1, 582.13], [645.05, 583.44], [647.98, 584.78], [650.94, 586.06], [653.93, 587.28], [656.92, 588.48], [659.95, 589.58], [662.99, 590.67], [666.05, 591.68], [669.13, 592.63], [672.21, 593.6], [675.3, 594.5], [678.42, 595.34], [681.54, 596.14], [684.66, 596.96], [687.78, 597.77], [690.91, 598.56], [694.04, 599.34], [697.15, 600.19], [700.24, 601.11], [703.29, 602.16], [706.31, 603.29], [709.3, 604.5], [712.24, 605.83], [715.14, 607.23], [718.03, 608.66], [720.91, 610.12], [723.81, 611.53], [726.7, 612.95], [729.58, 614.41], [732.45, 615.88], [735.33, 617.34], [738.22, 618.75], [741.1, 620.21], [743.96, 621.7], [746.8, 623.23], [749.63, 624.78], [752.4, 626.42], [755.19, 628.05], [757.94, 629.73], [760.67, 631.44], [763.39, 633.18], [766.11, 634.91], [768.83, 636.64], [771.57, 638.34], [774.3, 640.06], [777.01, 641.81], [779.74, 643.53], [782.47, 645.24], [785.23, 646.91], [788.0, 648.57], [790.82, 650.12], [793.68, 651.61], [796.56, 653.07], [799.44, 654.52], [802.38, 655.85], [805.34, 657.11], [808.33, 658.33], [811.34, 659.49], [814.37, 660.58], [817.43, 661.61], [820.51, 662.56], [823.62, 663.44], [826.72, 664.3], [829.83, 665.15], [832.94, 666.02], [836.03, 666.94], [839.09, 667.96], [842.11, 669.09], [845.11, 670.28], [848.09, 671.51], [851.06, 672.76], [854.01, 674.06], [856.97, 675.35], [859.93, 676.63], [862.89, 677.9], [865.84, 679.22], [868.76, 680.58], [871.68, 681.95], [874.61, 683.29], [877.55, 684.61], [880.5, 685.92], [883.43, 687.27], [886.36, 688.61], [889.3, 689.94], [892.21, 691.33], [895.11, 692.75], [897.98, 694.21], [900.8, 695.78], [903.55, 697.46], [906.24, 699.25], [908.84, 701.14], [911.38, 703.14], [913.82, 705.24], [916.18, 707.44], [918.46, 709.73], [920.63, 712.1], [922.72, 714.57], [924.7, 717.11], [926.57, 719.74], [928.34, 722.43], [930.0, 725.2], [931.55, 728.03], [932.98, 730.92], [934.3, 733.86], [935.49, 736.85], [936.57, 739.89], [937.52, 742.98], [938.35, 746.09], [939.05, 749.24], [939.63, 752.41], [940.08, 755.61], [940.4, 758.82], [940.59, 762.03], [940.65, 765.26], [940.58, 768.48], [940.38, 771.7], [940.05, 774.91], [939.6, 778.1], [939.01, 781.27], [938.3, 784.42], [937.47, 787.53], [936.51, 790.61], [935.44, 793.66], [934.31, 796.68], [933.11, 799.67], [931.85, 802.64], [930.54, 805.59], [929.17, 808.51], [927.75, 811.4], [926.29, 814.27], [924.79, 817.13], [923.28, 819.98], [921.72, 822.8], [920.1, 825.59], [918.46, 828.37], [916.76, 831.11], [915.03, 833.83], [913.29, 836.54], [911.52, 839.24], [909.7, 841.9], [907.87, 844.56], [906.03, 847.21], [904.15, 849.83], [902.24, 852.42], [900.26, 854.97], [898.2, 857.45], [896.07, 859.88], [893.88, 862.24], [891.63, 864.56], [889.38, 866.87], [887.1, 869.15], [884.84, 871.45], [882.58, 873.74], [880.26, 875.98], [877.88, 878.16], [875.5, 880.34], [873.12, 882.51], [870.69, 884.63], [868.22, 886.71], [865.84, 888.88], [863.54, 891.14], [861.34, 893.5], [859.23, 895.94], [857.22, 898.46], [855.32, 901.07], [853.52, 903.74], [851.83, 906.49], [850.26, 909.3], [848.79, 912.18], [847.45, 915.11], [846.22, 918.09], [845.11, 921.12], [844.13, 924.19], [843.26, 927.3], [842.53, 930.44], [841.92, 933.6], [841.44, 936.79], [841.09, 940.0], [840.86, 943.22], [840.77, 946.44], [840.8, 949.66], [840.97, 952.88], [841.26, 956.1], [841.68, 959.29], [842.23, 962.47], [842.91, 965.62], [843.71, 968.75], [844.64, 971.84], [845.69, 974.88], [846.87, 977.89], [848.16, 980.84], [849.57, 983.74], [851.09, 986.58], [852.73, 989.36], [854.48, 992.07], [856.33, 994.71], [858.29, 997.27], [860.36, 999.75], [862.51, 1002.15], [864.77, 1004.45], [867.11, 1006.67], [869.54, 1008.79], [872.06, 1010.81], [874.65, 1012.72], [877.32, 1014.53], [880.06, 1016.24], [882.86, 1017.83], [885.73, 1019.31], [888.65, 1020.67], [891.63, 1021.91], [894.65, 1023.03], [897.72, 1024.03], [900.82, 1024.91], [903.96, 1025.66], [907.12, 1026.28], [910.31, 1026.78], [913.51, 1027.16], [916.73, 1027.41], [919.95, 1027.58], [923.17, 1027.62], [926.39, 1027.54], [929.61, 1027.33], [932.82, 1027.01], [936.02, 1026.56], [939.2, 1026.05], [942.37, 1025.47], [945.53, 1024.82], [948.68, 1024.14], [951.83, 1023.46], [954.96, 1022.67], [958.07, 1021.83], [961.18, 1020.96], [964.27, 1020.04], [967.35, 1019.09], [970.4, 1018.04], [973.45, 1016.98], [976.49, 1015.92], [979.53, 1014.84], [982.57, 1013.76], [985.61, 1012.69], [988.64, 1011.58], [991.66, 1010.43], [994.67, 1009.29], [997.69, 1008.15], [1000.7, 1006.99], [1003.7, 1005.82], [1006.7, 1004.64], [1009.71, 1003.46], [1012.7, 1002.28], [1015.68, 1001.05], [1018.63, 999.73], [1021.58, 998.43], [1024.52, 997.11], [1027.44, 995.73], [1030.34, 994.32], [1033.21, 992.86], [1036.09, 991.4], [1038.94, 989.89], [1041.78, 988.37], [1044.6, 986.8], [1047.41, 985.22], [1050.19, 983.59], [1052.92, 981.86], [1055.6, 980.07], [1058.24, 978.22], [1060.86, 976.34], [1063.46, 974.44], [1066.08, 972.56], [1068.66, 970.61], [1071.16, 968.58], [1073.61, 966.48], [1076.06, 964.39], [1078.46, 962.23], [1080.8, 960.01], [1083.08, 957.73], [1085.32, 955.41], [1087.52, 953.06], [1089.68, 950.66], [1091.86, 948.28], [1094.04, 945.91], [1096.22, 943.53], [1098.39, 941.14], [1100.53, 938.74], [1102.63, 936.29], [1104.7, 933.82], [1106.77, 931.34], [1108.8, 928.84], [1110.82, 926.32], [1112.87, 923.83], [1114.94, 921.36], [1117.07, 918.94], [1119.27, 916.58], [1121.56, 914.3], [1123.93, 912.12], [1126.32, 909.96], [1128.7, 907.78], [1131.02, 905.54], [1133.32, 903.28], [1135.62, 901.02], [1137.89, 898.73], [1140.14, 896.41], [1142.36, 894.08], [1144.55, 891.71], [1146.7, 889.31], [1148.81, 886.87], [1150.88, 884.4], [1152.91, 881.89], [1154.92, 879.37], [1156.9, 876.82], [1158.9, 874.29], [1161.0, 871.84], [1163.17, 869.46], [1165.42, 867.15], [1167.76, 864.93], [1170.19, 862.81], [1172.69, 860.77], [1175.2, 858.75], [1177.75, 856.78], [1180.32, 854.83], [1182.89, 852.88], [1185.49, 850.97], [1188.08, 849.05], [1190.65, 847.1], [1193.2, 845.12], [1195.71, 843.11], [1198.19, 841.04], [1200.62, 838.92], [1203.01, 836.76], [1205.38, 834.57], [1207.72, 832.35], [1210.06, 830.13], [1212.47, 827.99], [1214.9, 825.87], [1217.33, 823.75], [1219.78, 821.65], [1222.26, 819.59], [1224.77, 817.56], [1227.29, 815.55], [1229.83, 813.57], [1232.39, 811.61], [1235.01, 809.73], [1237.65, 807.87], [1240.25, 805.97], [1242.84, 804.04], [1245.37, 802.04], [1247.86, 799.99], [1250.3, 797.88], [1252.71, 795.74], [1255.11, 793.59], [1257.49, 791.41], [1259.87, 789.24], [1262.27, 787.08], [1264.68, 784.94], [1267.09, 782.8], [1269.54, 780.7], [1271.97, 778.57], [1274.34, 776.4], [1276.66, 774.15], [1278.94, 771.87], [1281.19, 769.56], [1283.43, 767.24], [1285.64, 764.89], [1287.83, 762.53], [1290.0, 760.14], [1292.16, 757.74], [1294.29, 755.32], [1296.39, 752.88], [1298.48, 750.42], [1300.61, 748.0], [1302.79, 745.62], [1304.98, 743.25], [1307.15, 740.87], [1309.34, 738.5], [1311.59, 736.19], [1313.86, 733.9], [1316.19, 731.67], [1318.51, 729.43], [1320.79, 727.16], [1323.07, 724.87], [1325.33, 722.57], [1327.63, 720.31], [1329.97, 718.08], [1332.38, 715.94], [1334.84, 713.86], [1337.35, 711.84], [1339.91, 709.87], [1342.46, 707.9], [1345.04, 705.97], [1347.66, 704.09], [1350.32, 702.26], [1352.98, 700.44], [1355.66, 698.65], [1358.33, 696.84], [1361.02, 695.06], [1363.7, 693.26], [1366.35, 691.43], [1368.98, 689.56], [1371.58, 687.66], [1374.19, 685.76], [1376.8, 683.86], [1379.46, 682.04], [1382.17, 680.29], [1384.93, 678.62], [1387.71, 676.98], [1390.49, 675.35], [1393.25, 673.69], [1395.99, 671.98], [1398.71, 670.25], [1401.4, 668.48], [1404.06, 666.65], [1406.67, 664.76], [1409.24, 662.81], [1411.78, 660.82], [1414.29, 658.8], [1416.76, 656.73], [1419.21, 654.63], [1421.67, 652.54], [1424.15, 650.48], [1426.63, 648.42], [1429.14, 646.4], [1431.63, 644.35], [1434.09, 642.26], [1436.5, 640.12], [1438.87, 637.93], [1441.22, 635.73], [1443.58, 633.53], [1445.96, 631.35], [1448.38, 629.22], [1450.87, 627.17], [1453.38, 625.15], [1455.98, 623.23], [1458.63, 621.39], [1461.32, 619.62], [1464.06, 617.92], [1466.81, 616.23], [1469.57, 614.56], [1472.34, 612.92], [1475.13, 611.3], [1477.93, 609.69], [1480.74, 608.11], [1483.52, 606.48], [1486.32, 604.87], [1489.11, 603.27], [1491.92, 601.67], [1494.75, 600.14], [1497.62, 598.66], [1500.51, 597.23], [1503.42, 595.85], [1506.33, 594.46], [1509.24, 593.06], [1512.14, 591.66], [1515.02, 590.2], [1517.85, 588.66], [1520.62, 587.0], [1523.32, 585.24], [1526.02, 583.48], [1528.73, 581.74], [1531.44, 579.98], [1534.17, 578.27], [1536.94, 576.62], [1539.76, 575.05], [1542.58, 573.49], [1545.47, 572.05], [1548.41, 570.72], [1551.4, 569.52], [1554.44, 568.43], [1557.52, 567.47], [1560.63, 566.63], [1563.78, 565.92], [1566.95, 565.34], [1570.14, 564.88], [1573.35, 564.55], [1576.57, 564.35], [1579.79, 564.28], [1583.01, 564.34], [1586.23, 564.53], [1589.44, 564.85], [1592.64, 565.29], [1595.81, 565.87], [1598.96, 566.57], [1602.07, 567.39], [1605.16, 568.34], [1608.2, 569.42], [1611.19, 570.61], [1614.14, 571.93], [1617.03, 573.36], [1619.86, 574.91], [1622.62, 576.56], [1625.32, 578.33], [1627.94, 580.21], [1630.49, 582.19], [1632.95, 584.27], [1635.33, 586.44], [1637.62, 588.71], [1639.82, 591.07], [1641.92, 593.52], [1643.92, 596.05], [1645.82, 598.66], [1647.61, 601.34], [1649.29, 604.09], [1650.86, 606.91], [1652.32, 609.79], [1653.66, 612.72], [1654.88, 615.7], [1655.98, 618.74], [1656.96, 621.81], [1657.81, 624.92], [1658.54, 628.06], [1659.14, 631.23], [1659.61, 634.42], [1659.96, 637.62], [1660.17, 640.84], [1660.26, 644.07], [1660.22, 647.29], [1660.04, 650.51], [1659.74, 653.72], [1659.31, 656.92], [1658.75, 660.09], [1658.07, 663.25], [1657.26, 666.37], [1656.32, 669.45], [1655.26, 672.5], [1654.08, 675.5], [1652.78, 678.45], [1651.37, 681.35], [1649.85, 684.19], [1648.22, 686.97], [1646.47, 689.69], [1644.62, 692.33], [1642.67, 694.89], [1640.61, 697.38], [1638.51, 699.82], [1636.3, 702.18], [1634.02, 704.45], [1631.68, 706.67], [1629.26, 708.8], [1626.78, 710.86], [1624.24, 712.85], [1621.7, 714.84], [1619.16, 716.82], [1616.55, 718.72], [1613.91, 720.57], [1611.27, 722.42], [1608.59, 724.22], [1605.89, 725.99], [1603.16, 727.71], [1600.41, 729.39], [1597.65, 731.05], [1594.88, 732.7], [1592.07, 734.29], [1589.26, 735.86], [1586.45, 737.45], [1583.67, 739.09], [1580.94, 740.8], [1578.24, 742.57], [1575.53, 744.31], [1572.84, 746.09], [1570.17, 747.9], [1567.55, 749.78], [1565.0, 751.75], [1562.5, 753.79], [1560.06, 755.9], [1557.69, 758.09], [1555.34, 760.3], [1553.08, 762.59], [1550.87, 764.94], [1548.71, 767.33], [1546.55, 769.73], [1544.42, 772.15], [1542.26, 774.55], [1540.04, 776.89], [1537.77, 779.17], [1535.44, 781.41], [1533.08, 783.61], [1530.69, 785.78], [1528.3, 787.94], [1525.92, 790.11], [1523.55, 792.3], [1521.23, 794.54], [1518.92, 796.79], [1516.58, 799.0], [1514.2, 801.18], [1511.78, 803.31], [1509.32, 805.4], [1506.81, 807.42], [1504.3, 809.45], [1501.76, 811.43], [1499.24, 813.45], [1496.7, 815.44], [1494.14, 817.4], [1491.59, 819.37], [1489.04, 821.35], [1486.53, 823.38], [1484.07, 825.46], [1481.68, 827.63], [1479.3, 829.8], [1476.88, 831.93], [1474.39, 833.98], [1471.85, 835.97], [1469.29, 837.93], [1466.71, 839.86], [1464.11, 841.77], [1461.49, 843.65], [1458.86, 845.52], [1456.22, 847.37], [1453.56, 849.2], [1450.93, 851.06], [1448.32, 852.95], [1445.75, 854.9], [1443.22, 856.9], [1440.72, 858.94], [1438.24, 861.0], [1435.78, 863.09], [1433.3, 865.15], [1430.77, 867.15], [1428.21, 869.11], [1425.61, 871.02], [1422.98, 872.88], [1420.36, 874.76], [1417.78, 876.69], [1415.21, 878.64], [1412.71, 880.68], [1410.27, 882.79], [1407.9, 884.98], [1405.58, 887.21], [1403.3, 889.5], [1401.02, 891.78], [1398.68, 894.0], [1396.32, 896.2], [1393.95, 898.38], [1391.55, 900.54], [1389.11, 902.65], [1386.65, 904.73], [1384.17, 906.79], [1381.69, 908.85], [1379.19, 910.89], [1376.69, 912.92], [1374.21, 914.99], [1371.77, 917.1], [1369.42, 919.3], [1367.13, 921.58], [1364.87, 923.88], [1362.61, 926.18], [1360.41, 928.54], [1358.26, 930.94], [1356.17, 933.39], [1354.09, 935.86], [1351.93, 938.26], [1349.69, 940.57], [1347.36, 942.8], [1345.02, 945.02], [1342.69, 947.25], [1340.36, 949.48], [1337.99, 951.67], [1335.62, 953.86], [1333.31, 956.11], [1330.96, 958.32], [1328.61, 960.53], [1326.26, 962.73], [1323.9, 964.93], [1321.53, 967.12], [1319.19, 969.34], [1316.87, 971.58], [1314.56, 973.83], [1312.28, 976.11], [1310.05, 978.44], [1307.85, 980.8], [1305.69, 983.19], [1303.47, 985.53], [1301.21, 987.83], [1298.93, 990.11], [1296.65, 992.39], [1294.31, 994.61], [1291.97, 996.83], [1289.61, 999.03], [1287.21, 1001.18], [1284.74, 1003.26], [1282.23, 1005.28], [1279.67, 1007.25], [1277.09, 1009.17], [1274.47, 1011.06], [1271.82, 1012.89], [1269.15, 1014.7], [1266.45, 1016.46], [1263.74, 1018.22], [1261.06, 1020.01], [1258.41, 1021.84], [1255.77, 1023.7], [1253.12, 1025.54], [1250.5, 1027.41], [1247.91, 1029.33], [1245.38, 1031.33], [1242.87, 1033.36], [1240.38, 1035.41], [1237.95, 1037.53], [1235.6, 1039.74], [1233.29, 1041.99], [1230.98, 1044.24], [1228.65, 1046.47], [1226.34, 1048.72], [1224.04, 1050.98], [1221.78, 1053.28], [1219.54, 1055.6], [1217.27, 1057.89], [1214.95, 1060.13], [1212.63, 1062.37], [1210.29, 1064.59], [1207.92, 1066.78], [1205.55, 1068.97], [1203.19, 1071.16], [1200.83, 1073.36], [1198.44, 1075.53], [1196.04, 1077.68], [1193.67, 1079.86], [1191.39, 1082.14], [1189.14, 1084.45], [1186.91, 1086.79], [1184.66, 1089.1], [1182.4, 1091.4], [1180.12, 1093.68], [1177.81, 1095.93], [1175.51, 1098.19], [1173.23, 1100.47], [1170.98, 1102.78], [1168.8, 1105.16], [1166.7, 1107.6], [1164.66, 1110.1], [1162.64, 1112.61], [1160.6, 1115.11], [1158.49, 1117.55], [1156.36, 1119.98], [1154.18, 1122.35], [1151.98, 1124.71], [1149.81, 1127.09], [1147.63, 1129.47], [1145.46, 1131.86], [1143.27, 1134.23], [1141.07, 1136.58], [1138.83, 1138.9], [1136.57, 1141.2], [1134.29, 1143.48], [1131.99, 1145.74], [1129.65, 1147.96], [1127.32, 1150.19], [1124.99, 1152.42], [1122.66, 1154.65], [1120.35, 1156.9], [1118.07, 1159.18], [1115.83, 1161.5], [1113.59, 1163.82], [1111.4, 1166.19], [1109.28, 1168.62], [1107.26, 1171.13], [1105.32, 1173.71], [1103.47, 1176.35], [1101.72, 1179.06], [1100.06, 1181.82], [1098.41, 1184.6], [1096.74, 1187.36], [1095.08, 1190.12], [1093.4, 1192.87], [1091.75, 1195.64], [1090.14, 1198.44], [1088.59, 1201.27], [1087.1, 1204.12], [1085.64, 1207.0], [1084.19, 1209.88], [1082.74, 1212.76], [1081.3, 1215.65], [1079.91, 1218.56], [1078.53, 1221.47], [1077.16, 1224.39], [1075.83, 1227.33], [1074.51, 1230.27], [1073.21, 1233.22], [1071.91, 1236.17], [1070.62, 1239.13], [1069.35, 1242.09], [1068.09, 1245.06], [1066.78, 1248.01], [1065.45, 1250.95], [1064.18, 1253.91], [1062.93, 1256.88], [1061.65, 1259.84], [1060.39, 1262.81], [1059.14, 1265.78], [1057.9, 1268.76], [1056.64, 1271.73], [1055.4, 1274.71], [1054.12, 1277.66], [1052.78, 1280.6], [1051.35, 1283.49], [1049.87, 1286.35], [1048.35, 1289.2], [1046.8, 1292.03], [1045.17, 1294.81], [1043.48, 1297.56], [1041.7, 1300.24], [1039.83, 1302.87], [1037.87, 1305.43], [1035.82, 1307.92], [1033.73, 1310.38], [1031.6, 1312.8], [1029.48, 1315.23], [1027.33, 1317.64], [1025.28, 1320.13], [1023.33, 1322.69], [1021.49, 1325.34], [1019.75, 1328.06], [1018.12, 1330.84], [1016.61, 1333.69], [1015.21, 1336.6], [1013.93, 1339.56], [1012.77, 1342.56], [1011.73, 1345.62], [1010.82, 1348.71], [1010.03, 1351.84], [1009.36, 1354.99], [1008.83, 1358.17], [1008.42, 1361.37], [1008.14, 1364.58], [1007.99, 1367.81], [1007.97, 1371.03], [1008.07, 1374.25], [1008.31, 1377.47], [1008.68, 1380.67], [1009.17, 1383.86], [1009.79, 1387.02], [1010.54, 1390.16], [1011.42, 1393.27], [1012.41, 1396.33], [1013.53, 1399.36], [1014.77, 1402.33], [1016.13, 1405.26], [1017.61, 1408.13], [1019.2, 1410.93], [1020.9, 1413.67], [1022.71, 1416.34], [1024.62, 1418.94], [1026.64, 1421.45], [1028.75, 1423.89], [1030.97, 1426.23], [1033.27, 1428.49], [1035.67, 1430.65], [1038.14, 1432.71], [1040.7, 1434.68], [1043.34, 1436.54], [1046.04, 1438.29], [1048.82, 1439.93], [1051.65, 1441.47], [1054.55, 1442.88], [1057.5, 1444.18], [1060.51, 1445.36], [1063.55, 1446.42], [1066.64, 1447.35], [1069.75, 1448.2], [1072.88, 1448.97], [1076.04, 1449.61], [1079.22, 1450.15], [1082.41, 1450.63], [1085.6, 1451.12], [1088.79, 1451.57], [1091.99, 1451.95], [1095.21, 1452.24], [1098.42, 1452.49], [1101.64, 1452.71], [1104.86, 1452.91], [1108.08, 1453.12], [1111.3, 1453.29], [1114.52, 1453.44], [1117.74, 1453.55], [1120.96, 1453.65], [1124.19, 1453.68], [1127.41, 1453.67], [1130.64, 1453.65], [1133.86, 1453.65], [1137.09, 1453.66], [1140.31, 1453.69], [1143.54, 1453.74], [1146.76, 1453.83], [1149.98, 1453.95], [1153.2, 1454.13], [1156.42, 1454.29], [1159.64, 1454.51], [1162.86, 1454.77], [1166.07, 1455.07], [1169.28, 1455.37], [1172.48, 1455.71], [1175.69, 1456.02], [1178.9, 1456.33], [1182.12, 1456.63], [1185.33, 1456.93], [1188.54, 1457.21], [1191.76, 1457.43], [1194.98, 1457.6], [1198.2, 1457.77], [1201.42, 1457.88], [1204.64, 1457.98], [1207.87, 1458.09], [1211.09, 1458.2], [1214.31, 1458.28], [1217.54, 1458.3], [1220.76, 1458.33], [1223.99, 1458.34], [1227.21, 1458.33], [1230.44, 1458.3], [1233.66, 1458.19], [1236.88, 1457.98], [1240.09, 1457.64], [1243.28, 1457.21], [1246.46, 1456.66], [1249.63, 1456.06], [1252.78, 1455.39], [1255.93, 1454.67], [1259.06, 1453.9], [1262.17, 1453.07], [1265.28, 1452.19], [1268.37, 1451.28], [1271.45, 1450.32], [1274.52, 1449.33], [1277.58, 1448.32], [1280.62, 1447.25], [1283.65, 1446.15], [1286.67, 1445.01], [1289.68, 1443.84], [1292.66, 1442.62], [1295.68, 1441.48], [1298.7, 1440.36], [1301.77, 1439.35], [1304.85, 1438.42], [1307.96, 1437.57], [1311.1, 1436.84], [1314.27, 1436.22], [1317.44, 1435.61], [1320.62, 1435.08], [1323.81, 1434.65], [1327.02, 1434.35], [1330.24, 1434.19], [1333.47, 1434.15], [1336.69, 1434.24], [1339.91, 1434.45], [1343.12, 1434.8], [1346.31, 1435.28], [1349.47, 1435.88], [1352.61, 1436.61], [1355.72, 1437.47], [1358.8, 1438.45], [1361.83, 1439.55], [1364.81, 1440.78], [1367.74, 1442.12], [1370.62, 1443.58], [1373.43, 1445.15], [1376.18, 1446.84], [1378.86, 1448.63], [1381.47, 1450.53], [1384.0, 1452.54], [1386.44, 1454.64], [1388.8, 1456.84], [1391.07, 1459.13], [1393.24, 1461.51], [1395.32, 1463.98], [1397.29, 1466.53], [1399.17, 1469.15], [1400.93, 1471.85], [1402.59, 1474.62], [1404.13, 1477.45], [1405.56, 1480.34], [1406.87, 1483.29], [1408.06, 1486.28], [1409.13, 1489.33], [1410.08, 1492.41], [1410.91, 1495.53], [1411.6, 1498.68], [1412.17, 1501.85], [1412.62, 1505.04], [1412.93, 1508.25], [1413.12, 1511.47], [1413.17, 1514.7], [1413.1, 1517.92], [1412.89, 1521.14], [1412.57, 1524.35], [1412.2, 1527.55], [1411.79, 1530.75], [1411.36, 1533.95], [1410.9, 1537.14], [1410.37, 1540.32], [1409.75, 1543.48], [1409.06, 1546.63], [1408.32, 1549.77], [1407.52, 1552.9], [1406.72, 1556.02], [1405.89, 1559.14], [1405.08, 1562.26], [1404.29, 1565.39], [1403.53, 1568.52], [1402.75, 1571.65], [1401.95, 1574.77], [1401.12, 1577.89], [1400.26, 1581.0], [1399.33, 1584.09], [1398.36, 1587.16], [1397.34, 1590.22], [1396.31, 1593.28], [1395.3, 1596.34], [1394.21, 1599.38], [1393.12, 1602.41], [1391.98, 1605.43], [1390.83, 1608.44], [1389.72, 1611.47], [1388.68, 1614.52], [1387.68, 1617.59], [1386.71, 1620.66], [1385.77, 1623.75], [1384.83, 1626.83], [1383.9, 1629.92], [1382.98, 1633.01], [1382.06, 1636.1], [1381.13, 1639.19], [1380.24, 1642.29], [1379.41, 1645.4], [1378.53, 1648.51], [1377.58, 1651.59], [1376.61, 1654.66], [1375.63, 1657.74], [1374.64, 1660.81], [1373.63, 1663.87], [1372.58, 1666.92], [1371.5, 1669.96], [1370.43, 1673.0], [1369.36, 1676.04], [1368.26, 1679.07], [1367.15, 1682.1], [1366.04, 1685.13], [1364.94, 1688.16], [1363.84, 1691.19], [1362.8, 1694.24], [1361.78, 1697.3], [1360.76, 1700.36], [1359.74, 1703.42], [1358.74, 1706.49], [1357.74, 1709.55], [1356.76, 1712.62], [1355.72, 1715.68], [1354.62, 1718.71], [1353.53, 1721.74], [1352.42, 1724.77], [1351.23, 1727.77], [1350.04, 1730.77], [1348.77, 1733.73], [1347.48, 1736.69], [1346.14, 1739.62], [1344.75, 1742.53], [1343.31, 1745.42], [1341.83, 1748.28], [1340.3, 1751.12], [1338.76, 1753.95], [1337.21, 1756.78], [1335.64, 1759.6], [1334.08, 1762.42], [1332.51, 1765.24], [1330.92, 1768.04], [1329.34, 1770.85], [1327.79, 1773.68], [1326.24, 1776.51], [1324.68, 1779.33], [1323.14, 1782.16], [1321.64, 1785.02], [1320.18, 1787.9], [1318.74, 1790.78], [1317.29, 1793.66], [1315.86, 1796.55], [1314.44, 1799.45], [1313.05, 1802.36], [1311.61, 1805.24], [1310.2, 1808.14], [1308.86, 1811.08], [1307.58, 1814.04], [1306.35, 1817.02], [1305.15, 1820.01], [1303.91, 1822.99], [1302.61, 1825.94], [1301.28, 1828.88], [1299.94, 1831.81], [1298.57, 1834.73], [1297.18, 1837.64], [1295.83, 1840.57], [1294.48, 1843.5], [1293.13, 1846.42]], [[1578.44, 70.0], [1578.42, 73.07], [1578.4, 76.14], [1578.38, 79.21], [1578.36, 82.28], [1578.34, 85.35], [1578.31, 88.42], [1578.29, 91.49], [1578.27, 94.56], [1578.25, 97.63], [1578.23, 100.7], [1578.21, 103.77], [1578.19, 106.83], [1578.16, 109.9], [1578.14, 112.97], [1578.12, 116.04], [1578.1, 119.11], [1578.08, 122.18], [1578.06, 125.25], [1578.03, 128.32], [1578.01, 131.39], [1577.99, 134.46], [1577.97, 137.53], [1577.95, 140.6], [1577.93, 143.67], [1577.91, 146.74], [1577.88, 149.81], [1577.86, 152.88], [1577.84, 155.95], [1577.82, 159.02], [1577.8, 162.09], [1577.78, 165.16], [1577.75, 168.23], [1577.73, 171.3], [1577.71, 174.37], [1577.69, 177.44], [1577.67, 180.5], [1577.65, 183.57], [1577.63, 186.64], [1577.6, 189.71], [1577.58, 192.78], [1577.56, 195.85], [1577.54, 198.92], [1577.52, 201.99], [1577.5, 205.06], [1577.47, 208.13], [1577.45, 211.2], [1577.43, 214.27], [1577.41, 217.34], [1577.28, 220.41], [1577.05, 223.47], [1576.72, 226.52], [1576.34, 229.57], [1575.88, 232.6], [1575.39, 235.63], [1574.89, 238.66], [1574.35, 241.68], [1573.73, 244.69], [1573.06, 247.68], [1572.29, 250.66], [1571.46, 253.61], [1570.55, 256.54], [1569.63, 259.47], [1568.66, 262.38], [1567.75, 265.31], [1566.81, 268.24], [1565.88, 271.16], [1565.06, 274.12], [1564.36, 277.11], [1563.77, 280.12], [1563.28, 283.15], [1562.88, 286.2], [1562.54, 289.25], [1562.22, 292.3], [1561.94, 295.36], [1561.67, 298.41], [1561.49, 301.48], [1561.34, 304.54], [1561.23, 307.61], [1561.08, 310.68], [1560.97, 313.75], [1560.93, 316.81], [1560.92, 319.88], [1560.9, 322.95], [1560.92, 326.02], [1560.94, 329.09], [1561.01, 332.16], [1561.09, 335.23], [1561.18, 338.3], [1561.22, 341.37], [1561.24, 344.44], [1561.27, 347.51], [1561.3, 350.58], [1561.4, 353.64], [1561.54, 356.71], [1561.71, 359.78], [1561.89, 362.84], [1562.06, 365.91], [1562.22, 368.97], [1562.35, 372.04], [1562.49, 375.1], [1562.65, 378.17], [1562.84, 381.23], [1563.0, 384.3], [1563.13, 387.37], [1563.23, 390.43], [1563.32, 393.5], [1563.32, 396.57], [1563.28, 399.64], [1563.2, 402.71], [1563.1, 405.78], [1563.06, 408.85], [1563.04, 411.92], [1563.02, 414.99], [1562.93, 418.06], [1562.84, 421.12], [1562.77, 424.19], [1562.71, 427.26], [1562.59, 430.33], [1562.34, 433.39], [1561.96, 436.43], [1561.45, 439.46], [1560.85, 442.47], [1560.11, 445.45], [1559.26, 448.4], [1558.32, 451.32], [1557.33, 454.23], [1556.33, 457.13], [1555.25, 460.0], [1554.12, 462.86], [1552.96, 465.7], [1551.83, 468.56], [1550.73, 471.42], [1549.67, 474.3], [1548.63, 477.19], [1547.56, 480.07], [1546.62, 482.99], [1545.76, 485.94], [1544.97, 488.9], [1544.27, 491.89], [1543.7, 494.91], [1543.27, 497.95], [1542.95, 501.0], [1542.77, 504.06], [1542.73, 507.13], [1542.78, 510.2], [1542.97, 513.27], [1543.24, 516.32], [1543.55, 519.38], [1543.91, 522.43], [1544.39, 525.46], [1544.99, 528.47], [1545.68, 531.46], [1546.48, 534.42], [1547.35, 537.37], [1548.31, 540.28], [1549.34, 543.17], [1550.44, 546.04], [1551.59, 548.89], [1552.77, 551.72], [1554.01, 554.53], [1555.27, 557.33], [1556.5, 560.14], [1557.76, 562.94], [1559.07, 565.72], [1560.37, 568.49], [1561.69, 571.27], [1562.98, 574.06], [1564.25, 576.85], [1565.5, 579.65], [1566.73, 582.46], [1567.94, 585.28], [1569.15, 588.11], [1570.35, 590.93], [1571.52, 593.77], [1572.63, 596.63], [1573.71, 599.5], [1574.76, 602.39], [1575.82, 605.27], [1576.91, 608.14], [1578.06, 610.99], [1579.2, 613.83], [1580.34, 616.68], [1581.53, 619.52], [1582.75, 622.33], [1584.0, 625.14], [1585.27, 627.93], [1586.54, 630.73], [1587.82, 633.52], [1589.09, 636.31], [1590.38, 639.1], [1591.68, 641.88], [1592.96, 644.67], [1594.17, 647.49], [1595.39, 650.31], [1596.58, 653.14], [1597.79, 655.95], [1599.0, 658.78], [1600.22, 661.59], [1601.45, 664.4], [1602.71, 667.21], [1603.96, 670.01], [1605.21, 672.81], [1606.43, 675.63], [1607.66, 678.44], [1608.93, 681.23], [1610.16, 684.05], [1611.3, 686.9], [1612.33, 689.79], [1613.24, 692.72], [1614.05, 695.68], [1614.78, 698.66], [1615.52, 701.64], [1616.22, 704.63], [1616.89, 707.63], [1617.47, 710.64], [1617.94, 713.68], [1618.4, 716.71], [1618.87, 719.74], [1619.29, 722.79], [1619.66, 725.83], [1620.01, 728.88], [1620.28, 731.94], [1620.47, 735.0], [1620.59, 738.07], [1620.68, 741.14], [1620.72, 744.21], [1620.72, 747.28], [1620.68, 750.35], [1620.59, 753.42], [1620.45, 756.48], [1620.3, 759.55], [1620.13, 762.61], [1619.99, 765.68], [1619.81, 768.74], [1619.65, 771.81], [1619.54, 774.88], [1619.3, 777.94], [1618.93, 780.98], [1618.42, 784.01], [1617.78, 787.02], [1617.02, 789.99], [1616.12, 792.92], [1615.1, 795.82], [1613.96, 798.67], [1612.69, 801.46], [1611.31, 804.2], [1609.8, 806.88], [1608.18, 809.49], [1606.46, 812.02], [1604.62, 814.48], [1602.67, 816.86], [1600.63, 819.15], [1598.49, 821.35], [1596.26, 823.45], [1593.93, 825.46], [1591.53, 827.37], [1589.04, 829.16], [1586.47, 830.85], [1583.84, 832.43], [1581.14, 833.89], [1578.38, 835.23], [1575.56, 836.45], [1572.7, 837.55], [1569.79, 838.53], [1566.84, 839.37], [1563.85, 840.09], [1560.84, 840.68], [1557.8, 841.14], [1554.75, 841.46], [1551.69, 841.65], [1548.62, 841.72], [1545.55, 841.64], [1542.49, 841.44], [1539.44, 841.1], [1536.4, 840.63], [1533.39, 840.03], [1530.4, 839.36], [1527.41, 838.66], [1524.42, 837.96], [1521.44, 837.24], [1518.48, 836.41], [1515.56, 835.45], [1512.67, 834.43], [1509.77, 833.41], [1506.91, 832.3], [1504.06, 831.17], [1501.21, 830.03], [1498.37, 828.86], [1495.57, 827.6], [1492.79, 826.3], [1490.02, 824.98], [1487.26, 823.63], [1484.55, 822.19], [1481.91, 820.63], [1479.32, 818.98], [1476.76, 817.29], [1474.18, 815.62], [1471.67, 813.85], [1469.2, 812.02], [1466.79, 810.12], [1464.39, 808.21], [1462.0, 806.28], [1459.59, 804.39], [1457.17, 802.49], [1454.76, 800.6], [1452.35, 798.69], [1449.99, 796.74], [1447.68, 794.72], [1445.39, 792.66], [1443.1, 790.62], [1440.79, 788.6], [1438.45, 786.61], [1436.06, 784.68], [1433.64, 782.8], [1431.19, 780.95], [1428.69, 779.17], [1426.12, 777.49], [1423.54, 775.83], [1420.93, 774.21], [1418.31, 772.6], [1415.65, 771.07], [1413.0, 769.54], [1410.33, 768.01], [1407.69, 766.46], [1405.1, 764.8], [1402.53, 763.13], [1399.97, 761.43], [1397.4, 759.75], [1394.81, 758.11], [1392.19, 756.5], [1389.62, 754.83], [1387.06, 753.14], [1384.52, 751.4], [1382.01, 749.64], [1379.53, 747.83], [1377.12, 745.94], [1374.73, 744.0], [1372.34, 742.07], [1369.91, 740.2], [1367.45, 738.36], [1364.95, 736.58], [1362.42, 734.85], [1359.86, 733.16], [1357.29, 731.48], [1354.73, 729.78], [1352.15, 728.12], [1349.61, 726.4], [1347.1, 724.63], [1344.62, 722.82], [1342.18, 720.96], [1339.76, 719.07], [1337.38, 717.13], [1334.97, 715.23], [1332.52, 713.38], [1330.04, 711.57], [1327.58, 709.74], [1325.13, 707.88], [1322.71, 706.0], [1320.32, 704.07], [1317.95, 702.12], [1315.58, 700.17], [1313.21, 698.21], [1310.84, 696.27], [1308.44, 694.35], [1306.0, 692.48], [1303.55, 690.64], [1301.08, 688.82], [1298.63, 686.97], [1296.23, 685.05], [1293.92, 683.03], [1291.66, 680.96], [1289.42, 678.85], [1287.2, 676.73], [1285.0, 674.59], [1282.82, 672.43], [1280.63, 670.28], [1278.39, 668.18], [1276.12, 666.12], [1273.82, 664.08], [1271.5, 662.08], [1269.15, 660.09], [1266.83, 658.09], [1264.57, 656.01], [1262.41, 653.83], [1260.33, 651.58], [1258.25, 649.32], [1256.19, 647.04], [1254.15, 644.74], [1252.18, 642.39], [1250.26, 639.99], [1248.36, 637.58], [1246.46, 635.17], [1244.52, 632.8], [1242.51, 630.48], [1240.42, 628.22], [1238.3, 626.01], [1236.11, 623.85], [1233.89, 621.74], [1231.65, 619.63], [1229.45, 617.49], [1227.25, 615.35], [1225.03, 613.24], [1222.79, 611.13], [1220.55, 609.04], [1218.21, 607.05], [1215.8, 605.15], [1213.3, 603.37], [1210.73, 601.69], [1208.09, 600.13], [1205.38, 598.68], [1202.61, 597.36], [1199.79, 596.15], [1196.92, 595.06], [1194.0, 594.11], [1191.05, 593.27], [1188.06, 592.57], [1185.04, 592.0], [1182.0, 591.56], [1178.95, 591.25], [1175.88, 591.07], [1172.82, 591.03], [1169.75, 591.12], [1166.69, 591.34], [1163.64, 591.69], [1160.61, 592.18], [1157.6, 592.79], [1154.62, 593.54], [1151.68, 594.41], [1148.78, 595.41], [1145.92, 596.54], [1143.11, 597.78], [1140.37, 599.15], [1137.68, 600.64], [1135.06, 602.24], [1132.51, 603.95], [1130.04, 605.77], [1127.65, 607.7], [1125.35, 609.72], [1123.13, 611.85], [1121.01, 614.07], [1118.99, 616.38], [1117.07, 618.77], [1115.25, 621.25], [1113.55, 623.8], [1111.95, 626.42], [1110.48, 629.11], [1109.11, 631.87], [1107.87, 634.67], [1106.76, 637.53], [1105.76, 640.44], [1104.9, 643.38], [1104.16, 646.36], [1103.55, 649.37], [1103.07, 652.4], [1102.69, 655.45], [1102.43, 658.51], [1102.31, 661.57], [1102.32, 664.64], [1102.46, 667.71], [1102.7, 670.77], [1103.02, 673.82], [1103.4, 676.87], [1103.8, 679.91], [1104.24, 682.95], [1104.72, 685.98], [1105.19, 689.02], [1105.67, 692.05], [1106.16, 695.08], [1106.67, 698.11], [1107.17, 701.13], [1107.68, 704.16], [1108.21, 707.18], [1108.76, 710.2], [1109.33, 713.22], [1109.93, 716.23], [1110.54, 719.24], [1111.17, 722.24], [1111.84, 725.24], [1112.5, 728.24], [1113.23, 731.22], [1114.02, 734.19], [1114.82, 737.15], [1115.58, 740.12], [1116.35, 743.09], [1117.08, 746.08], [1117.81, 749.06], [1118.58, 752.03], [1119.35, 755.0], [1120.16, 757.96], [1121.04, 760.9], [1122.0, 763.82], [1123.03, 766.71], [1124.17, 769.56], [1125.44, 772.36], [1126.78, 775.11], [1128.15, 777.86], [1129.56, 780.59], [1131.06, 783.27], [1132.61, 785.92], [1134.25, 788.51], [1135.95, 791.07], [1137.7, 793.59], [1139.49, 796.08], [1141.32, 798.55], [1143.17, 801.0], [1145.04, 803.43], [1146.92, 805.86], [1148.85, 808.24], [1150.83, 810.59], [1152.86, 812.9], [1154.89, 815.2], [1156.92, 817.5], [1158.92, 819.83], [1160.95, 822.13], [1162.94, 824.47], [1164.94, 826.8], [1166.93, 829.14], [1168.82, 831.55], [1170.71, 833.98], [1172.6, 836.39], [1174.48, 838.82], [1176.25, 841.32], [1177.94, 843.89], [1179.55, 846.5], [1181.19, 849.09], [1182.9, 851.65], [1184.66, 854.16], [1186.51, 856.61], [1188.39, 859.04], [1190.31, 861.43], [1192.27, 863.8], [1194.22, 866.17], [1196.2, 868.51], [1198.16, 870.87], [1200.15, 873.21], [1202.19, 875.5], [1204.25, 877.78], [1206.34, 880.03], [1208.45, 882.25], [1210.55, 884.5], [1212.62, 886.76], [1214.65, 889.06], [1216.64, 891.41], [1218.52, 893.83], [1220.4, 896.26], [1222.27, 898.69], [1224.11, 901.14], [1226.0, 903.57], [1227.89, 905.98], [1229.83, 908.36], [1231.81, 910.71], [1233.81, 913.04], [1235.77, 915.4], [1237.75, 917.75], [1239.75, 920.08], [1241.71, 922.44], [1243.57, 924.88], [1245.41, 927.34], [1247.26, 929.79], [1249.07, 932.27], [1250.93, 934.71], [1252.79, 937.15], [1254.64, 939.6], [1256.55, 942.01], [1258.49, 944.39], [1260.4, 946.78], [1262.28, 949.21], [1264.12, 951.67], [1265.94, 954.14], [1267.69, 956.67], [1269.38, 959.23], [1271.06, 961.79], [1272.77, 964.34], [1274.44, 966.92], [1276.05, 969.54], [1277.6, 972.18], [1279.14, 974.84], [1280.67, 977.5], [1282.2, 980.16], [1283.79, 982.78], [1285.46, 985.36], [1287.16, 987.92], [1288.91, 990.44], [1290.73, 992.91], [1292.62, 995.33], [1294.54, 997.72], [1296.45, 1000.13], [1298.31, 1002.57], [1300.12, 1005.05], [1301.91, 1007.54], [1303.74, 1010.01], [1305.57, 1012.48], [1307.42, 1014.92], [1309.31, 1017.34], [1311.24, 1019.73], [1313.14, 1022.14], [1315.03, 1024.56], [1316.89, 1027.0], [1318.7, 1029.48], [1320.45, 1032.0], [1322.09, 1034.6], [1323.67, 1037.23], [1325.18, 1039.9], [1326.59, 1042.63], [1327.9, 1045.4], [1329.13, 1048.21], [1330.34, 1051.04], [1331.5, 1053.88], [1332.67, 1056.72], [1333.92, 1059.52], [1335.26, 1062.28], [1336.65, 1065.02], [1338.13, 1067.71], [1339.72, 1070.33], [1341.39, 1072.91], [1343.12, 1075.44], [1344.86, 1077.98], [1346.55, 1080.54], [1348.31, 1083.05], [1350.11, 1085.54], [1351.92, 1088.02], [1353.77, 1090.47], [1355.66, 1092.89], [1357.58, 1095.28], [1359.54, 1097.65], [1361.54, 1099.97], [1363.52, 1102.32], [1365.43, 1104.72], [1367.29, 1107.17], [1369.1, 1109.64], [1370.88, 1112.14], [1372.63, 1114.66], [1374.36, 1117.2], [1376.02, 1119.78], [1377.57, 1122.43], [1379.03, 1125.13], [1380.45, 1127.86], [1381.8, 1130.61], [1383.13, 1133.38], [1384.44, 1136.15], [1385.7, 1138.95], [1386.9, 1141.78], [1388.08, 1144.61], [1389.3, 1147.43], [1390.61, 1150.2], [1392.02, 1152.93], [1393.47, 1155.64], [1394.95, 1158.33], [1396.47, 1161.0], [1398.09, 1163.6], [1399.76, 1166.18], [1401.49, 1168.71], [1403.29, 1171.2], [1405.15, 1173.64], [1407.05, 1176.05], [1408.98, 1178.44], [1410.92, 1180.81], [1412.87, 1183.18], [1414.85, 1185.54], [1416.87, 1187.85], [1418.91, 1190.14], [1420.95, 1192.43], [1423.02, 1194.7], [1425.08, 1196.98], [1427.1, 1199.29], [1429.09, 1201.62], [1431.05, 1203.99], [1433.01, 1206.35], [1434.95, 1208.73], [1436.88, 1211.11], [1438.72, 1213.57], [1440.55, 1216.04], [1442.43, 1218.46], [1444.34, 1220.86], [1446.28, 1223.25], [1448.19, 1225.65], [1450.08, 1228.07], [1451.95, 1230.5], [1453.84, 1232.92], [1455.77, 1235.31], [1457.73, 1237.67], [1459.71, 1240.01], [1461.72, 1242.33], [1463.72, 1244.66], [1465.65, 1247.05], [1467.48, 1249.52], [1469.25, 1252.03], [1470.93, 1254.59], [1472.57, 1257.19], [1474.21, 1259.78], [1475.93, 1262.33], [1477.67, 1264.86], [1479.39, 1267.4], [1481.08, 1269.96], [1482.71, 1272.56], [1484.25, 1275.22], [1485.7, 1277.92], [1487.05, 1280.68], [1488.32, 1283.47], [1489.52, 1286.3], [1490.66, 1289.15], [1491.81, 1291.99], [1492.98, 1294.83], [1494.13, 1297.68], [1495.36, 1300.49], [1496.67, 1303.27], [1498.04, 1306.01], [1499.5, 1308.71], [1500.99, 1311.4], [1502.49, 1314.08], [1503.95, 1316.78], [1505.35, 1319.51], [1506.72, 1322.25], [1508.03, 1325.03], [1509.27, 1327.84], [1510.5, 1330.65], [1511.61, 1333.51], [1512.65, 1336.4], [1513.62, 1339.31], [1514.57, 1342.23], [1515.57, 1345.14], [1516.66, 1348.01], [1517.81, 1350.85], [1518.95, 1353.7], [1520.04, 1356.57], [1521.11, 1359.45], [1522.16, 1362.33], [1523.15, 1365.24], [1524.14, 1368.14], [1525.09, 1371.06], [1526.0, 1374.0], [1526.88, 1376.94], [1527.71, 1379.89], [1528.53, 1382.85], [1529.46, 1385.78], [1530.46, 1388.68], [1531.55, 1391.55], [1532.76, 1394.37], [1534.02, 1397.16], [1535.37, 1399.93], [1536.75, 1402.66], [1538.13, 1405.41], [1539.52, 1408.15], [1540.91, 1410.88], [1542.32, 1413.61], [1543.69, 1416.35], [1544.98, 1419.14], [1546.21, 1421.95], [1547.37, 1424.79], [1548.49, 1427.65], [1549.57, 1430.53], [1550.6, 1433.42], [1551.69, 1436.29], [1552.81, 1439.15], [1553.97, 1441.99], [1555.15, 1444.82], [1556.42, 1447.61], [1557.77, 1450.37], [1559.22, 1453.08], [1560.78, 1455.72], [1562.22, 1458.43], [1563.54, 1461.2], [1564.75, 1464.03], [1565.82, 1466.9], [1566.78, 1469.82], [1567.6, 1472.78], [1568.3, 1475.76], [1568.87, 1478.78], [1569.3, 1481.82], [1569.6, 1484.87], [1569.78, 1487.94], [1569.81, 1491.01], [1569.72, 1494.08], [1569.49, 1497.14], [1569.13, 1500.19], [1568.64, 1503.22], [1568.02, 1506.22], [1567.27, 1509.2], [1566.39, 1512.14], [1565.38, 1515.04], [1564.25, 1517.89], [1563.0, 1520.7], [1561.63, 1523.44], [1560.14, 1526.13], [1558.53, 1528.74], [1556.81, 1531.29], [1554.99, 1533.75], [1553.06, 1536.14], [1551.02, 1538.44], [1548.89, 1540.65], [1546.67, 1542.77], [1544.36, 1544.78], [1541.96, 1546.7], [1539.48, 1548.51], [1536.92, 1550.21], [1534.3, 1551.8], [1531.6, 1553.27], [1528.85, 1554.63], [1526.04, 1555.87], [1523.18, 1556.98], [1520.27, 1557.97], [1517.33, 1558.83], [1514.35, 1559.56], [1511.34, 1560.16], [1508.3, 1560.63], [1505.25, 1560.97], [1502.19, 1561.18], [1499.12, 1561.26], [1496.05, 1561.2], [1492.99, 1561.01], [1489.94, 1560.69], [1486.9, 1560.23], [1483.89, 1559.65], [1480.89, 1558.98], [1477.92, 1558.2], [1474.98, 1557.32], [1472.08, 1556.31], [1469.23, 1555.18], [1466.39, 1554.0], [1463.56, 1552.83], [1460.73, 1551.63], [1457.92, 1550.4], [1455.12, 1549.13], [1452.32, 1547.88], [1449.53, 1546.6], [1446.79, 1545.22], [1444.1, 1543.73], [1441.43, 1542.23], [1438.74, 1540.75], [1436.05, 1539.27], [1433.38, 1537.75], [1430.74, 1536.18], [1428.14, 1534.55], [1425.58, 1532.86], [1423.03, 1531.15], [1420.49, 1529.43], [1417.95, 1527.7], [1415.45, 1525.92], [1413.0, 1524.07], [1410.59, 1522.16], [1408.19, 1520.26], [1405.84, 1518.28], [1403.57, 1516.21], [1401.35, 1514.1], [1399.14, 1511.97], [1396.95, 1509.81], [1394.81, 1507.61], [1392.63, 1505.45], [1390.48, 1503.26], [1388.35, 1501.05], [1386.19, 1498.87], [1384.06, 1496.66], [1381.96, 1494.42], [1379.85, 1492.19], [1377.74, 1489.96], [1375.64, 1487.73], [1373.55, 1485.48], [1371.48, 1483.21], [1369.43, 1480.92], [1367.4, 1478.62], [1365.4, 1476.29], [1363.44, 1473.93], [1361.52, 1471.54], [1359.61, 1469.13], [1357.77, 1466.68], [1355.92, 1464.22], [1354.04, 1461.8], [1352.16, 1459.37], [1350.32, 1456.91], [1348.5, 1454.44], [1346.66, 1451.99], [1344.76, 1449.57], [1342.83, 1447.19], [1340.9, 1444.8], [1338.94, 1442.43], [1336.94, 1440.11], [1334.87, 1437.85], [1332.76, 1435.61], [1330.7, 1433.33], [1328.71, 1431.0], [1326.75, 1428.64], [1324.85, 1426.23], [1322.99, 1423.79], [1321.14, 1421.34], [1319.25, 1418.91], [1317.34, 1416.51], [1315.43, 1414.11], [1313.53, 1411.7], [1311.64, 1409.28], [1309.78, 1406.84], [1307.95, 1404.38], [1306.16, 1401.88], [1304.41, 1399.35], [1302.71, 1396.8], [1301.04, 1394.23], [1299.35, 1391.67], [1297.63, 1389.12], [1295.93, 1386.56], [1294.21, 1384.02], [1292.47, 1381.49], [1290.69, 1378.99], [1288.8, 1376.57], [1286.87, 1374.19], [1284.92, 1371.82], [1282.97, 1369.45], [1281.03, 1367.07], [1279.14, 1364.65], [1277.27, 1362.22], [1275.4, 1359.78], [1273.54, 1357.34], [1271.68, 1354.9], [1269.86, 1352.42], [1268.06, 1349.94], [1266.24, 1347.46], [1264.4, 1345.01], [1262.46, 1342.63], [1260.45, 1340.31], [1258.37, 1338.05], [1256.2, 1335.88], [1253.98, 1333.76], [1251.76, 1331.64], [1249.55, 1329.51], [1247.4, 1327.32], [1245.26, 1325.12], [1243.16, 1322.88], [1241.12, 1320.58], [1239.12, 1318.26], [1237.09, 1315.96], [1235.04, 1313.67], [1232.99, 1311.38], [1230.95, 1309.09], [1228.9, 1306.8], [1226.85, 1304.52], [1224.79, 1302.25], [1222.7, 1300.0], [1220.6, 1297.76], [1218.52, 1295.5], [1216.51, 1293.18], [1214.46, 1290.89], [1212.45, 1288.58], [1210.48, 1286.22], [1208.54, 1283.85], [1206.62, 1281.45], [1204.71, 1279.05], [1202.81, 1276.63], [1200.86, 1274.26], [1198.93, 1271.88], [1197.03, 1269.47], [1195.15, 1267.04], [1193.29, 1264.6], [1191.47, 1262.12], [1189.62, 1259.67], [1187.71, 1257.27], [1185.79, 1254.87], [1183.87, 1252.48], [1181.98, 1250.07], [1180.12, 1247.62], [1178.29, 1245.15], [1176.5, 1242.67], [1174.67, 1240.2], [1172.89, 1237.7], [1171.11, 1235.2], [1169.29, 1232.73], [1167.41, 1230.3], [1165.5, 1227.89], [1163.6, 1225.48], [1161.72, 1223.06], [1159.86, 1220.62], [1157.99, 1218.18], [1156.13, 1215.74], [1154.28, 1213.29], [1152.42, 1210.85], [1150.56, 1208.4], [1148.73, 1205.94], [1146.84, 1203.52], [1144.94, 1201.11], [1142.98, 1198.74], [1141.01, 1196.39], [1139.04, 1194.04], [1137.08, 1191.68], [1135.08, 1189.34], [1133.08, 1187.02], [1131.03, 1184.74], [1128.92, 1182.5], [1126.78, 1180.3], [1124.6, 1178.15], [1122.43, 1175.97], [1120.29, 1173.77], [1118.16, 1171.56], [1116.04, 1169.35], [1113.93, 1167.11], [1111.84, 1164.86], [1109.79, 1162.58], [1107.74, 1160.29], [1105.73, 1157.98], [1103.69, 1155.68], [1101.65, 1153.38], [1099.6, 1151.1], [1097.51, 1148.86], [1095.37, 1146.65], [1093.23, 1144.45], [1091.07, 1142.27], [1088.91, 1140.09], [1086.71, 1137.95], [1084.51, 1135.8], [1082.31, 1133.67], [1080.09, 1131.55], [1077.83, 1129.47], [1075.55, 1127.42], [1073.27, 1125.36], [1071.0, 1123.3], [1068.72, 1121.24], [1066.4, 1119.23], [1064.0, 1117.32], [1061.52, 1115.51], [1058.96, 1113.81], [1056.33, 1112.22], [1053.64, 1110.75], [1050.9, 1109.36], [1048.14, 1108.03], [1045.34, 1106.77], [1042.54, 1105.5], [1039.76, 1104.21], [1036.97, 1102.92], [1034.2, 1101.6], [1031.46, 1100.21], [1028.76, 1098.75], [1026.09, 1097.25], [1023.43, 1095.71], [1020.79, 1094.14], [1018.18, 1092.52], [1015.61, 1090.84], [1013.01, 1089.22], [1010.37, 1087.66], [1007.68, 1086.16], [1004.99, 1084.69], [1002.32, 1083.17], [999.68, 1081.62], [997.04, 1080.05], [994.38, 1078.51], [991.68, 1077.05], [988.92, 1075.71], [986.11, 1074.47], [983.25, 1073.35], [980.35, 1072.34], [977.42, 1071.44], [974.46, 1070.64], [971.48, 1069.91], [968.48, 1069.24], [965.47, 1068.65], [962.47, 1068.0], [959.45, 1067.43], [956.43, 1066.9], [953.4, 1066.41], [950.37, 1065.91], [947.34, 1065.41], [944.3, 1064.95], [941.27, 1064.49], [938.23, 1064.07], [935.18, 1063.7], [932.13, 1063.34], [929.09, 1062.96], [926.06, 1062.47], [923.03, 1061.97], [920.0, 1061.48], [916.98, 1060.9], [913.98, 1060.24], [910.99, 1059.57], [908.0, 1058.87], [905.01, 1058.16], [902.02, 1057.47], [899.05, 1056.69], [896.11, 1055.83], [893.16, 1054.98], [890.23, 1054.06], [887.35, 1053.01], [884.5, 1051.86], [881.7, 1050.59], [878.95, 1049.24], [876.25, 1047.77], [873.59, 1046.25], [870.91, 1044.74], [868.24, 1043.24], [865.56, 1041.73], [862.86, 1040.28], [860.13, 1038.88], [857.36, 1037.54], [854.59, 1036.21], [851.83, 1034.88], [849.09, 1033.49], [846.39, 1032.04], [843.74, 1030.48], [841.14, 1028.85], [838.58, 1027.16], [836.01, 1025.48], [833.45, 1023.79], [830.92, 1022.05], [828.4, 1020.3], [825.87, 1018.55], [823.34, 1016.82], [820.77, 1015.14], [818.18, 1013.49], [815.57, 1011.87], [812.93, 1010.31], [810.25, 1008.82], [807.54, 1007.37], [804.82, 1005.95], [802.08, 1004.56], [799.35, 1003.16], [796.62, 1001.75], [793.9, 1000.35], [791.23, 998.82], [788.64, 997.18], [786.11, 995.44], [783.63, 993.63], [781.22, 991.73], [778.85, 989.78], [776.54, 987.75], [774.29, 985.67], [772.08, 983.53], [769.93, 981.35], [767.81, 979.12], [765.77, 976.84], [763.77, 974.5], [761.75, 972.19], [759.69, 969.92], [757.53, 967.74], [755.28, 965.65], [753.01, 963.58], [750.76, 961.49], [748.5, 959.41], [746.24, 957.34], [743.98, 955.26], [741.71, 953.19], [739.47, 951.09], [737.23, 948.99], [735.04, 946.85], [732.9, 944.65], [730.76, 942.44], [728.62, 940.25], [726.46, 938.06], [724.3, 935.88], [722.1, 933.74], [719.83, 931.68], [717.55, 929.62], [715.3, 927.53], [713.05, 925.44], [710.85, 923.3], [708.7, 921.11], [706.57, 918.9], [704.46, 916.67], [702.38, 914.41], [700.34, 912.12], [698.28, 909.85], [696.16, 907.62], [694.02, 905.43], [691.82, 903.28], [689.61, 901.16], [687.39, 899.03], [685.14, 896.94], [682.89, 894.85], [680.68, 892.72], [678.49, 890.57], [676.28, 888.45], [674.07, 886.31], [671.93, 884.11], [669.76, 881.94], [667.61, 879.75], [665.51, 877.51], [663.45, 875.23], [661.45, 872.91], [659.46, 870.57], [657.5, 868.21], [655.55, 865.83], [653.62, 863.45], [651.64, 861.1], [649.61, 858.8], [647.55, 856.52], [645.46, 854.28], [643.29, 852.1], [641.03, 850.03], [638.73, 847.99], [636.39, 846.01], [634.01, 844.07], [631.64, 842.12], [629.25, 840.19], [626.82, 838.32], [624.35, 836.49], [621.84, 834.72], [619.29, 833.02], [616.73, 831.32], [614.15, 829.67], [611.6, 827.96], [609.05, 826.25], [606.52, 824.51], [603.98, 822.78], [601.45, 821.05], [598.89, 819.35], [596.32, 817.67], [593.76, 815.97], [591.19, 814.29], [588.64, 812.59], [586.11, 810.86], [583.62, 809.05], [581.14, 807.25], [578.65, 805.46], [576.11, 803.72], [573.56, 802.02], [571.0, 800.32], [568.43, 798.64], [565.89, 796.92], [563.33, 795.23], [560.75, 793.57], [558.13, 791.97], [555.49, 790.4], [552.88, 788.78], [550.24, 787.22], [547.59, 785.67], [544.94, 784.12], [542.27, 782.6], [539.61, 781.08], [536.93, 779.58], [534.22, 778.14], [531.51, 776.7], [528.8, 775.25], [526.1, 773.8], [523.42, 772.29], [520.82, 770.67], [518.28, 768.94], [515.81, 767.12], [513.39, 765.23], [511.02, 763.28], [508.6, 761.39], [506.14, 759.56], [503.63, 757.79], [501.09, 756.06], [498.52, 754.38], [495.95, 752.71], [493.36, 751.05], [490.76, 749.42], [488.14, 747.83], [485.51, 746.24], [482.88, 744.66], [480.23, 743.11], [477.63, 741.48], [475.06, 739.81], [472.5, 738.11], [469.94, 736.42], [467.39, 734.71], [464.87, 732.95], [462.41, 731.12], [459.99, 729.22], [457.66, 727.23], [455.39, 725.16], [453.19, 723.02], [451.04, 720.83], [448.96, 718.57], [446.93, 716.27], [444.95, 713.92], [442.99, 711.56], [441.06, 709.18], [439.19, 706.74], [437.34, 704.29], [435.48, 701.85], [433.66, 699.38], [431.83, 696.91], [430.05, 694.41], [428.29, 691.9], [426.53, 689.38], [424.79, 686.85], [423.04, 684.33], [421.3, 681.8], [419.55, 679.28], [417.81, 676.75], [416.06, 674.23], [414.32, 671.7], [412.57, 669.18], [410.78, 666.69], [408.98, 664.2], [407.15, 661.73], [405.3, 659.29], [403.44, 656.84], [401.52, 654.45], [399.51, 652.12], [397.4, 649.89], [395.25, 647.71], [393.01, 645.61], [390.72, 643.56], [388.39, 641.57], [386.03, 639.6], [383.68, 637.63], [381.35, 635.63], [379.05, 633.6], [376.76, 631.55], [374.39, 629.6], [371.93, 627.76], [369.4, 626.02], [366.8, 624.4], [364.13, 622.89], [361.39, 621.5], [358.6, 620.22], [355.75, 619.07], [352.86, 618.04], [349.93, 617.14], [346.95, 616.37], [343.95, 615.72], [340.93, 615.21], [337.88, 614.83], [334.82, 614.58], [331.75, 614.46], [328.68, 614.48], [325.62, 614.63], [322.56, 614.91], [319.52, 615.32], [316.5, 615.86], [313.5, 616.54], [310.54, 617.34], [307.62, 618.27], [304.73, 619.33], [301.9, 620.51], [299.12, 621.81], [296.4, 623.24], [293.74, 624.77], [291.16, 626.43], [288.64, 628.19], [286.21, 630.06], [283.86, 632.03], [281.59, 634.1], [279.42, 636.27], [277.34, 638.53], [275.37, 640.88], [273.49, 643.31], [271.73, 645.82], [270.07, 648.41], [268.53, 651.06], [267.1, 653.78], [265.78, 656.55], [264.55, 659.36], [263.37, 662.2], [262.28, 665.07], [261.28, 667.97], [260.35, 670.89], [259.51, 673.85], [258.74, 676.82], [258.04, 679.81], [257.37, 682.8], [256.75, 685.81], [256.25, 688.84], [255.89, 691.88], [255.66, 694.95], [255.56, 698.01], [255.6, 701.08], [255.77, 704.15], [256.07, 707.2], [256.5, 710.24], [257.07, 713.26], [257.76, 716.25], [258.55, 719.21], [259.45, 722.15], [260.44, 725.06], [261.49, 727.94], [262.59, 730.8], [263.79, 733.63], [264.97, 736.47], [266.05, 739.34], [267.12, 742.22], [268.13, 745.12], [269.06, 748.04], [269.89, 751.0], [270.64, 753.97], [271.32, 756.97], [271.95, 759.97], [272.5, 762.99], [272.98, 766.02], [273.43, 769.06], [273.88, 772.09], [274.28, 775.14], [274.62, 778.19], [274.94, 781.24], [275.25, 784.3], [275.54, 787.35], [275.85, 790.41], [276.17, 793.46], [276.46, 796.51], [276.71, 799.57], [277.01, 802.63], [277.34, 805.68], [277.66, 808.73], [278.05, 811.78], [278.56, 814.81], [279.11, 817.83], [279.79, 820.82], [280.59, 823.78], [281.5, 826.71], [282.5, 829.62], [283.55, 832.5], [284.61, 835.38], [285.64, 838.27], [286.58, 841.2], [287.58, 844.1], [288.64, 846.98], [289.68, 849.87], [290.7, 852.76], [291.64, 855.69], [292.5, 858.63], [293.3, 861.59], [294.02, 864.58], [294.75, 867.56], [295.48, 870.54], [296.29, 873.5], [297.17, 876.44], [298.1, 879.37], [299.06, 882.28], [300.1, 885.17], [301.19, 888.04], [302.38, 890.87], [303.65, 893.67], [304.99, 896.43], [306.39, 899.16], [307.76, 901.91], [309.07, 904.68], [310.31, 907.49], [311.53, 910.31], [312.83, 913.09], [314.11, 915.88], [315.37, 918.68], [316.58, 921.5], [317.71, 924.35], [318.78, 927.23], [319.76, 930.14], [320.65, 933.08], [321.44, 936.04], [322.2, 939.02], [322.94, 942.0], [323.65, 944.98], [324.37, 947.97], [325.03, 950.97], [325.62, 953.98], [326.19, 956.99], [326.77, 960.01], [327.35, 963.02], [327.96, 966.03], [328.57, 969.04], [329.19, 972.05], [329.77, 975.06], [330.28, 978.09], [330.78, 981.12], [331.27, 984.15], [331.82, 987.17], [332.48, 990.16], [333.19, 993.15], [333.89, 996.14], [334.6, 999.13], [335.39, 1002.09], [336.24, 1005.04], [337.19, 1007.96], [338.22, 1010.85], [339.33, 1013.71], [340.56, 1016.53], [341.87, 1019.3], [343.24, 1022.05], [344.67, 1024.76], [346.07, 1027.5], [347.46, 1030.23], [348.88, 1032.96], [350.31, 1035.67], [351.76, 1038.38], [353.18, 1041.1], [354.58, 1043.83], [355.98, 1046.56], [357.36, 1049.31], [358.66, 1052.08], [359.9, 1054.89], [361.13, 1057.71], [362.38, 1060.51], [363.61, 1063.32], [364.82, 1066.14], [366.08, 1068.94], [367.4, 1071.71], [368.76, 1074.46], [370.15, 1077.2], [371.61, 1079.9], [373.03, 1082.62], [374.37, 1085.39], [375.6, 1088.2], [376.74, 1091.05], [377.79, 1093.93], [378.76, 1096.85], [379.64, 1099.79], [380.51, 1102.73], [381.39, 1105.67], [382.31, 1108.6], [383.18, 1111.54], [384.02, 1114.5], [384.86, 1117.45], [385.7, 1120.4], [386.51, 1123.36], [387.29, 1126.33], [388.06, 1129.3], [388.76, 1132.29], [389.39, 1135.3], [389.98, 1138.31], [390.55, 1141.32], [391.07, 1144.35], [391.56, 1147.38], [392.02, 1150.41], [392.42, 1153.46], [392.84, 1156.5], [393.36, 1159.52], [394.0, 1162.53], [394.75, 1165.5], [395.62, 1168.45], [396.61, 1171.35], [397.73, 1174.21], [398.91, 1177.05], [400.15, 1179.85], [401.45, 1182.63], [402.84, 1185.37], [404.23, 1188.11], [405.67, 1190.82], [407.15, 1193.51], [408.55, 1196.24], [409.86, 1199.02], [411.12, 1201.81], [412.36, 1204.62], [413.52, 1207.47], [414.63, 1210.33], [415.74, 1213.19], [416.85, 1216.05], [418.0, 1218.9], [419.16, 1221.74], [420.4, 1224.55], [421.7, 1227.33], [422.89, 1230.16], [423.95, 1233.04], [424.88, 1235.96], [425.69, 1238.92], [426.37, 1241.92], [426.92, 1244.94], [427.33, 1247.98], [427.62, 1251.04], [427.77, 1254.1], [427.79, 1257.17], [427.68, 1260.24], [427.43, 1263.3], [427.05, 1266.34], [426.54, 1269.37], [425.9, 1272.37], [425.13, 1275.35], [424.24, 1278.28], [423.21, 1281.18], [422.07, 1284.02], [420.8, 1286.82], [419.41, 1289.55], [417.9, 1292.23], [416.28, 1294.83], [414.54, 1297.37], [412.7, 1299.82], [410.76, 1302.2], [408.71, 1304.49], [406.57, 1306.68], [404.33, 1308.79], [402.01, 1310.79], [399.6, 1312.69], [397.11, 1314.49], [394.54, 1316.17], [391.9, 1317.75], [389.2, 1319.2], [386.44, 1320.54], [383.62, 1321.76], [380.76, 1322.86], [377.85, 1323.86], [374.91, 1324.74], [371.94, 1325.5], [368.94, 1326.14], [365.92, 1326.68], [362.88, 1327.1], [359.82, 1327.4], [356.76, 1327.63], [353.69, 1327.75], [350.62, 1327.75], [347.55, 1327.71], [344.48, 1327.69], [341.41, 1327.68], [338.34, 1327.66], [335.27, 1327.65], [332.21, 1327.67], [329.14, 1327.67], [326.07, 1327.58], [323.0, 1327.49], [319.93, 1327.35], [316.87, 1327.17], [313.81, 1326.96], [310.74, 1326.81], [307.67, 1326.73], [304.6, 1326.76], [301.53, 1326.79], [298.46, 1326.82], [295.39, 1326.89], [292.32, 1326.93], [289.25, 1326.91], [286.19, 1326.95], [283.12, 1327.03], [280.05, 1327.15], [276.98, 1327.32], [273.92, 1327.53], [270.86, 1327.74], [267.8, 1327.98], [264.74, 1328.22], [261.68, 1328.46], [258.61, 1328.61], [255.54, 1328.71], [252.48, 1328.77], [249.41, 1328.73], [246.34, 1328.68], [243.27, 1328.67], [240.2, 1328.69], [237.13, 1328.67], [234.06, 1328.6], [230.99, 1328.57], [227.92, 1328.51], [224.85, 1328.43], [221.78, 1328.36], [218.71, 1328.3], [215.65, 1328.11], [212.59, 1327.82], [209.54, 1327.53], [206.48, 1327.24], [203.43, 1326.95], [200.37, 1326.66], [197.31, 1326.37], [194.26, 1326.08], [191.2, 1325.79], [188.15, 1325.5], [185.09, 1325.21], [182.03, 1324.92], [178.98, 1324.63], [175.92, 1324.35], [172.87, 1324.06], [169.81, 1323.77], [166.76, 1323.48], [163.7, 1323.19], [160.64, 1322.9], [157.59, 1322.61], [154.53, 1322.32], [151.48, 1322.03], [148.42, 1321.74], [145.36, 1321.45], [142.31, 1321.16], [139.25, 1320.87], [136.2, 1320.58], [133.14, 1320.29], [130.08, 1320.0], [127.03, 1319.71], [123.97, 1319.42], [120.92, 1319.13], [117.86, 1318.84], [114.8, 1318.55], [111.75, 1318.26], [108.69, 1317.97], [105.64, 1317.68], [102.58, 1317.39], [99.52, 1317.1], [96.47, 1316.82], [93.41, 1316.53], [90.36, 1316.24], [87.3, 1315.95], [84.24, 1315.66], [81.19, 1315.37], [78.13, 1315.08], [75.08, 1314.79], [72.02, 1314.5], [68.96, 1314.21]], [[897.82, 70.0], [897.82, 73.74], [897.82, 77.47], [897.83, 81.21], [897.83, 84.95], [897.83, 88.69], [897.83, 92.42], [897.84, 96.16], [897.84, 99.9], [897.84, 103.64], [897.84, 107.37], [897.85, 111.11], [897.85, 114.85], [897.85, 118.59], [897.85, 122.32], [897.85, 126.06], [897.86, 129.8], [897.86, 133.54], [897.86, 137.27], [897.86, 141.01], [897.87, 144.75], [897.87, 148.49], [897.87, 152.22], [897.87, 155.96], [897.88, 159.7], [897.88, 163.43], [897.88, 167.17], [897.88, 170.91], [897.88, 174.65], [897.89, 178.38], [897.89, 182.12], [897.89, 185.86], [897.89, 189.6], [897.9, 193.33], [897.9, 197.07], [897.9, 200.81], [897.9, 204.55], [897.91, 208.28], [897.91, 212.02], [897.91, 215.76], [897.91, 219.5], [897.78, 223.23], [897.5, 226.96], [897.09, 230.67], [896.55, 234.37], [895.86, 238.04], [895.05, 241.69], [894.09, 245.31], [893.01, 248.88], [891.79, 252.42], [890.45, 255.9], [888.98, 259.34], [887.38, 262.72], [885.66, 266.03], [883.81, 269.29], [881.85, 272.47], [879.77, 275.57], [877.58, 278.6], [875.28, 281.55], [872.88, 284.41], [870.36, 287.17], [867.76, 289.85], [865.05, 292.43], [862.25, 294.91], [859.37, 297.28], [856.4, 299.55], [853.34, 301.7], [850.21, 303.74], [847.01, 305.67], [843.73, 307.47], [840.39, 309.15], [837.01, 310.74], [833.57, 312.2], [830.08, 313.53], [826.54, 314.73], [822.96, 315.81], [819.34, 316.74], [815.69, 317.55], [812.02, 318.23], [808.32, 318.78], [804.61, 319.2], [800.88, 319.52], [797.15, 319.73], [793.42, 319.93], [789.69, 320.13], [785.95, 320.29], [782.22, 320.42], [778.48, 320.52], [774.75, 320.57], [771.01, 320.57], [767.27, 320.48], [763.54, 320.3], [759.81, 320.03], [756.09, 319.71], [752.37, 319.34], [748.65, 318.97], [744.93, 318.6], [741.21, 318.22], [737.5, 317.82], [733.78, 317.46], [730.05, 317.17], [726.32, 316.94], [722.58, 316.83], [718.85, 316.75], [715.11, 316.63], [711.38, 316.51], [707.64, 316.36], [703.91, 316.18], [700.18, 315.96], [696.45, 315.7], [692.72, 315.41], [688.99, 315.16], [685.27, 314.89], [681.54, 314.57], [677.82, 314.2], [674.1, 313.84], [670.39, 313.43], [666.68, 312.99], [662.97, 312.53], [659.27, 312.03], [655.56, 311.51], [651.86, 311.01], [648.16, 310.46], [644.47, 309.87], [640.79, 309.26], [637.09, 308.67], [633.41, 308.07], [629.72, 307.46], [626.04, 306.83], [622.35, 306.2], [618.66, 305.62], [614.96, 305.12], [611.24, 304.72], [607.51, 304.43], [603.78, 304.28], [600.04, 304.18], [596.31, 304.11], [592.57, 304.05], [588.83, 304.02], [585.09, 304.05], [581.36, 304.14], [577.62, 304.21], [573.89, 304.29], [570.15, 304.35], [566.41, 304.43], [562.67, 304.48], [558.94, 304.54], [555.2, 304.61], [551.46, 304.69], [547.73, 304.76], [543.99, 304.87], [540.26, 304.96], [536.52, 305.13], [532.79, 305.37], [529.07, 305.7], [525.36, 306.15], [521.66, 306.67], [517.96, 307.22], [514.27, 307.8], [510.59, 308.43], [506.91, 309.1], [503.23, 309.77], [499.55, 310.44], [495.88, 311.1], [492.2, 311.78], [488.52, 312.45], [484.84, 313.05], [481.15, 313.66], [477.46, 314.27], [473.78, 314.9], [470.1, 315.58], [466.43, 316.28], [462.77, 317.03], [459.13, 317.87], [455.47, 318.66], [451.81, 319.41], [448.13, 320.07], [444.44, 320.64], [440.74, 321.12], [437.02, 321.51], [433.29, 321.83], [429.57, 322.11], [425.83, 322.3], [422.1, 322.39], [418.36, 322.4], [414.62, 322.44], [410.89, 322.62], [407.17, 322.93], [403.46, 323.39], [399.77, 323.98], [396.1, 324.7], [392.46, 325.56], [388.86, 326.56], [385.3, 327.68], [381.78, 328.94], [378.31, 330.33], [374.89, 331.84], [371.53, 333.48], [368.23, 335.24], [365.0, 337.12], [361.85, 339.12], [358.76, 341.23], [355.76, 343.46], [352.84, 345.79], [350.01, 348.23], [347.27, 350.78], [344.63, 353.42], [342.09, 356.16], [339.65, 358.99], [337.31, 361.91], [335.09, 364.91], [332.98, 368.0], [330.98, 371.15], [329.1, 374.38], [327.34, 377.68], [325.7, 381.04], [324.19, 384.46], [322.8, 387.93], [321.55, 391.45], [320.42, 395.01], [319.43, 398.62], [318.57, 402.25], [317.81, 405.91], [317.13, 409.59], [316.4, 413.25], [315.62, 416.91], [314.82, 420.56], [314.06, 424.22], [313.31, 427.88], [312.62, 431.55], [312.0, 435.24], [311.47, 438.94], [311.04, 442.65], [310.75, 446.38], [310.49, 450.11], [310.36, 453.84], [310.34, 457.58], [310.4, 461.32], [310.54, 465.05], [310.71, 468.78], [310.85, 472.52], [310.96, 476.25], [311.06, 479.99], [311.15, 483.73], [311.28, 487.46], [311.4, 491.2], [311.58, 494.93], [311.71, 498.67], [311.84, 502.4], [311.95, 506.14], [312.01, 509.87], [312.01, 513.61], [311.99, 517.35], [311.91, 521.08], [311.8, 524.82], [311.71, 528.56], [311.64, 532.29], [311.61, 536.03], [311.57, 539.77], [311.48, 543.5], [311.38, 547.24], [311.32, 550.98], [311.32, 554.71], [311.42, 558.45], [311.56, 562.19], [311.74, 565.92], [311.96, 569.65], [312.18, 573.38], [312.41, 577.11], [312.6, 580.84], [312.82, 584.57], [313.03, 588.31], [313.29, 592.03], [313.6, 595.76], [313.94, 599.48], [314.32, 603.2], [314.72, 606.91], [315.14, 610.63], [315.57, 614.34], [315.96, 618.06], [316.32, 621.78], [316.58, 625.51], [316.82, 629.24], [317.04, 632.97], [317.26, 636.7], [317.39, 640.43], [317.43, 644.17], [317.61, 647.9], [317.93, 651.63], [318.39, 655.34], [318.98, 659.03], [319.71, 662.69], [320.57, 666.33], [321.57, 669.93], [322.69, 673.49], [323.95, 677.01], [325.34, 680.48], [326.86, 683.9], [328.5, 687.26], [330.26, 690.55], [332.14, 693.78], [334.14, 696.94], [336.26, 700.02], [338.48, 703.02], [340.82, 705.94], [343.26, 708.77], [345.81, 711.5], [348.45, 714.14], [351.19, 716.69], [354.03, 719.13], [356.94, 721.46], [359.95, 723.68], [363.03, 725.79], [366.19, 727.79], [369.42, 729.67], [372.72, 731.43], [376.08, 733.06], [379.5, 734.57], [382.97, 735.95], [386.49, 737.21], [390.06, 738.33], [393.66, 739.32], [397.3, 740.18], [400.97, 740.91], [404.66, 741.49], [408.36, 741.98], [412.08, 742.39], [415.8, 742.73], [419.53, 743.0], [423.26, 743.24], [426.99, 743.48], [430.72, 743.7], [434.45, 743.86], [438.19, 743.98], [441.92, 744.05], [445.66, 744.09], [449.4, 744.07], [453.13, 744.0], [456.87, 743.9], [460.6, 743.75], [464.34, 743.54], [468.07, 743.3], [471.79, 743.0], [475.51, 742.62], [479.23, 742.22], [482.94, 741.81], [486.65, 741.34], [490.34, 740.77], [494.01, 740.07], [497.68, 739.34], [501.33, 738.56], [504.99, 737.79], [508.65, 737.01], [512.32, 736.3], [515.99, 735.65], [519.69, 735.09], [523.4, 734.62], [527.11, 734.22], [530.84, 733.87], [534.56, 733.59], [538.29, 733.38], [542.03, 733.26], [545.77, 733.2], [549.5, 733.2], [553.24, 733.21], [556.98, 733.23], [560.72, 733.27], [564.45, 733.29], [568.19, 733.3], [571.93, 733.25], [575.66, 733.12], [579.39, 732.92], [583.12, 732.69], [586.85, 732.43], [590.58, 732.18], [594.31, 731.86], [598.02, 731.47], [601.73, 731.03], [605.44, 730.58], [609.15, 730.12], [612.86, 729.66], [616.57, 729.21], [620.28, 728.73], [623.98, 728.2], [627.68, 727.68], [631.38, 727.18], [635.1, 726.79], [638.83, 726.49], [642.56, 726.28], [646.29, 726.14], [650.03, 726.05], [653.77, 726.0], [657.5, 725.91], [661.24, 725.76], [664.97, 725.61], [668.71, 725.53], [672.44, 725.45], [676.18, 725.39], [679.92, 725.28], [683.65, 725.11], [687.38, 724.96], [691.12, 724.76], [694.85, 724.56], [698.58, 724.32], [702.31, 724.12], [706.04, 723.98], [709.78, 723.89], [713.52, 723.85], [717.26, 723.84], [720.99, 723.8], [724.73, 723.73], [728.47, 723.62], [732.2, 723.55], [735.94, 723.48], [739.68, 723.4], [743.41, 723.37], [747.15, 723.39], [750.89, 723.43], [754.62, 723.52], [758.36, 723.65], [762.09, 723.79], [765.83, 723.94], [769.56, 724.09], [773.3, 724.25], [777.03, 724.41], [780.76, 724.56], [784.5, 724.72], [788.23, 724.9], [791.96, 725.07], [795.7, 725.23], [799.43, 725.37], [803.17, 725.47], [806.91, 725.56], [810.64, 725.59], [814.38, 725.5], [818.11, 725.3], [821.83, 724.97], [825.54, 724.51], [829.23, 723.92], [832.91, 723.23], [836.57, 722.5], [840.22, 721.67], [843.85, 720.8], [847.49, 719.94], [851.13, 719.08], [854.75, 718.19], [858.37, 717.25], [861.98, 716.29], [865.59, 715.32], [869.21, 714.37], [872.83, 713.46], [876.46, 712.57], [880.11, 711.76], [883.78, 711.02], [887.46, 710.37], [891.14, 709.76], [894.84, 709.22], [898.54, 708.7], [902.24, 708.19], [905.94, 707.64], [909.64, 707.14], [913.36, 706.69], [917.06, 706.22], [920.77, 705.73], [924.47, 705.21], [928.17, 704.7], [931.88, 704.2], [935.57, 703.64], [939.25, 702.97], [942.95, 702.44], [946.66, 702.05], [950.39, 701.79], [954.13, 701.68], [957.87, 701.7], [961.6, 701.85], [965.33, 702.15], [969.04, 702.58], [972.73, 703.15], [976.4, 703.85], [980.04, 704.69], [983.65, 705.67], [987.22, 706.77], [990.75, 708.01], [994.23, 709.37], [997.66, 710.87], [1001.02, 712.48], [1004.33, 714.23], [1007.57, 716.09], [1010.74, 718.07], [1013.84, 720.16], [1016.85, 722.37], [1019.78, 724.69], [1022.63, 727.11], [1025.38, 729.64], [1028.04, 732.27], [1030.6, 734.99], [1033.05, 737.81], [1035.41, 740.71], [1037.65, 743.7], [1039.78, 746.77], [1041.8, 749.92], [1043.7, 753.14], [1045.47, 756.43], [1047.13, 759.78], [1048.66, 763.19], [1050.07, 766.65], [1051.35, 770.16], [1052.49, 773.72], [1053.51, 777.32], [1054.39, 780.95], [1055.13, 784.61], [1055.74, 788.3], [1056.22, 792.0], [1056.56, 795.73], [1056.76, 799.46], [1056.82, 803.19], [1056.74, 806.93], [1056.53, 810.66], [1056.18, 814.38], [1055.69, 818.09], [1055.07, 821.77], [1054.31, 825.43], [1053.42, 829.06], [1052.39, 832.66], [1051.23, 836.21], [1049.94, 839.72], [1048.53, 843.18], [1046.98, 846.58], [1045.32, 849.92], [1043.53, 853.21], [1041.62, 856.42], [1039.59, 859.56], [1037.45, 862.62], [1035.19, 865.6], [1032.83, 868.5], [1030.37, 871.31], [1027.8, 874.02], [1025.13, 876.64], [1022.37, 879.16], [1019.52, 881.57], [1016.61, 883.92], [1013.63, 886.18], [1010.59, 888.35], [1007.47, 890.41], [1004.28, 892.35], [1001.02, 894.18], [997.69, 895.88], [994.33, 897.53], [990.95, 899.12], [987.55, 900.67], [984.15, 902.21], [980.73, 903.72], [977.28, 905.16], [973.83, 906.6], [970.39, 908.06], [966.97, 909.56], [963.57, 911.12], [960.21, 912.75], [956.87, 914.43], [953.52, 916.1], [950.18, 917.77], [946.84, 919.44], [943.5, 921.11], [940.17, 922.82], [936.89, 924.62], [933.64, 926.46], [930.43, 928.38], [927.25, 930.33], [924.1, 932.34], [920.96, 934.37], [917.83, 936.41], [914.67, 938.42], [911.48, 940.35], [908.24, 942.22], [904.95, 943.99], [901.62, 945.68], [898.25, 947.32], [894.9, 948.96], [891.53, 950.57], [888.17, 952.22], [884.81, 953.85], [881.42, 955.43], [878.02, 956.98], [874.6, 958.49], [871.17, 959.96], [867.73, 961.43], [864.29, 962.89], [860.84, 964.33], [857.41, 965.81], [853.96, 967.25], [850.5, 968.66], [847.05, 970.09], [843.58, 971.5], [840.1, 972.85], [836.62, 974.21], [833.12, 975.53], [829.64, 976.89], [826.18, 978.3], [822.75, 979.78], [819.32, 981.28], [815.93, 982.85], [812.57, 984.47], [809.26, 986.21], [806.01, 988.07], [802.83, 990.03], [799.7, 992.07], [796.6, 994.16], [793.55, 996.31], [790.53, 998.51], [787.56, 1000.79], [784.64, 1003.12], [781.75, 1005.49], [778.9, 1007.9], [776.09, 1010.37], [773.28, 1012.83], [770.49, 1015.32], [767.72, 1017.83], [764.98, 1020.37], [762.23, 1022.9], [759.49, 1025.45], [756.79, 1028.03], [754.08, 1030.6], [751.47, 1033.27], [748.96, 1036.04], [746.55, 1038.9], [744.25, 1041.85], [742.06, 1044.87], [739.98, 1047.98], [738.02, 1051.16], [736.18, 1054.41], [734.45, 1057.73], [732.86, 1061.11], [731.38, 1064.55], [730.04, 1068.03], [728.82, 1071.57], [727.74, 1075.14], [726.79, 1078.76], [725.97, 1082.41], [725.29, 1086.08], [724.74, 1089.78], [724.33, 1093.49], [724.06, 1097.22], [723.92, 1100.95], [723.93, 1104.69], [724.07, 1108.43], [724.35, 1112.15], [724.76, 1115.87], [725.31, 1119.56], [726.0, 1123.24], [726.82, 1126.88], [727.78, 1130.5], [728.87, 1134.07], [730.09, 1137.6], [731.44, 1141.09], [732.92, 1144.52], [734.52, 1147.9], [736.25, 1151.21], [738.09, 1154.46], [740.06, 1157.64], [742.14, 1160.74], [744.34, 1163.77], [746.64, 1166.71], [749.05, 1169.57], [751.57, 1172.33], [754.18, 1175.0], [756.89, 1177.57], [759.7, 1180.04], [762.59, 1182.41], [765.57, 1184.66], [768.63, 1186.81], [771.77, 1188.84], [774.98, 1190.75], [778.26, 1192.55], [781.6, 1194.22], [785.0, 1195.77], [788.46, 1197.19], [791.97, 1198.48], [795.52, 1199.64], [799.11, 1200.67], [802.74, 1201.57], [806.4, 1202.33], [810.08, 1202.96], [813.79, 1203.45], [817.51, 1203.81], [821.24, 1204.03], [824.98, 1204.11], [828.71, 1204.05], [832.45, 1203.88], [836.17, 1203.6], [839.89, 1203.21], [843.59, 1202.68], [847.27, 1202.01], [850.92, 1201.21], [854.54, 1200.27], [858.12, 1199.23], [861.68, 1198.09], [865.2, 1196.83], [868.68, 1195.47], [872.15, 1194.06], [875.58, 1192.59], [879.02, 1191.12], [882.44, 1189.62], [885.86, 1188.11], [889.28, 1186.61], [892.7, 1185.09], [896.12, 1183.58], [899.54, 1182.09], [902.97, 1180.59], [906.37, 1179.04], [909.75, 1177.45], [913.12, 1175.83], [916.48, 1174.2], [919.82, 1172.51], [923.12, 1170.78], [926.44, 1169.06], [929.77, 1167.35], [933.08, 1165.62], [936.4, 1163.9], [939.73, 1162.2], [943.06, 1160.51], [946.39, 1158.81], [949.71, 1157.09], [953.02, 1155.36], [956.32, 1153.61], [959.62, 1151.84], [962.92, 1150.1], [966.23, 1148.36], [969.5, 1146.56], [972.78, 1144.76], [976.02, 1142.9], [979.23, 1140.99], [982.43, 1139.04], [985.57, 1137.02], [988.69, 1134.97], [991.8, 1132.9], [994.9, 1130.8], [997.97, 1128.68], [1001.05, 1126.56], [1004.13, 1124.45], [1007.19, 1122.29], [1010.2, 1120.08], [1013.19, 1117.84], [1016.14, 1115.54], [1019.09, 1113.24], [1022.0, 1110.9], [1024.89, 1108.53], [1027.78, 1106.16], [1030.65, 1103.77], [1033.47, 1101.32], [1036.26, 1098.83], [1039.03, 1096.32], [1041.8, 1093.81], [1044.63, 1091.36], [1047.49, 1088.96], [1050.38, 1086.6], [1053.33, 1084.3], [1056.31, 1082.05], [1059.33, 1079.85], [1062.38, 1077.68], [1065.45, 1075.54], [1068.52, 1073.42], [1071.59, 1071.29], [1074.7, 1069.22], [1077.82, 1067.16], [1080.95, 1065.11], [1084.08, 1063.08], [1087.21, 1061.03], [1090.32, 1058.96], [1093.46, 1056.92], [1096.55, 1054.83], [1099.63, 1052.72], [1102.71, 1050.59], [1105.78, 1048.46], [1108.83, 1046.3], [1111.82, 1044.06], [1114.75, 1041.74], [1117.69, 1039.43], [1120.61, 1037.1], [1123.51, 1034.75], [1126.37, 1032.34], [1129.22, 1029.92], [1132.05, 1027.48], [1134.9, 1025.06], [1137.76, 1022.65], [1140.62, 1020.24], [1143.46, 1017.82], [1146.3, 1015.39], [1149.08, 1012.89], [1151.84, 1010.37], [1154.58, 1007.83], [1157.28, 1005.24], [1159.96, 1002.64], [1162.62, 1000.01], [1165.3, 997.41], [1168.01, 994.83], [1170.67, 992.21], [1173.26, 989.52], [1175.83, 986.8], [1178.49, 984.18], [1181.25, 981.65], [1184.09, 979.23], [1187.03, 976.92], [1190.05, 974.72], [1193.15, 972.63], [1196.32, 970.65], [1199.56, 968.79], [1202.87, 967.06], [1206.25, 965.45], [1209.67, 963.96], [1213.16, 962.6], [1216.68, 961.37], [1220.26, 960.27], [1223.87, 959.3], [1227.51, 958.47], [1231.18, 957.77], [1234.88, 957.21], [1238.59, 956.78], [1242.32, 956.5], [1246.05, 956.34], [1249.79, 956.33], [1253.52, 956.46], [1257.25, 956.72], [1260.97, 957.12], [1264.67, 957.65], [1268.35, 958.31], [1272.0, 959.1], [1275.62, 960.02], [1279.22, 961.02], [1282.79, 962.14], [1286.32, 963.36], [1289.82, 964.67], [1293.3, 966.03], [1296.73, 967.51], [1300.12, 969.1], [1303.44, 970.8], [1306.7, 972.63], [1309.89, 974.58], [1313.0, 976.65], [1316.04, 978.83], [1319.0, 981.12], [1321.92, 983.45], [1324.78, 985.85], [1327.65, 988.25], [1330.52, 990.64], [1333.37, 993.06], [1336.23, 995.46], [1339.11, 997.85], [1342.01, 1000.2], [1344.93, 1002.53], [1347.82, 1004.9], [1350.71, 1007.28], [1353.57, 1009.68], [1356.38, 1012.15], [1359.12, 1014.69], [1361.77, 1017.32], [1364.36, 1020.02], [1366.9, 1022.76], [1369.42, 1025.52], [1371.95, 1028.27], [1374.47, 1031.03], [1377.04, 1033.74], [1379.69, 1036.38], [1382.38, 1038.97], [1385.09, 1041.54], [1387.8, 1044.11], [1390.52, 1046.68], [1393.21, 1049.28], [1395.93, 1051.84], [1398.66, 1054.39], [1401.4, 1056.93], [1404.13, 1059.48], [1406.87, 1062.03], [1409.63, 1064.55], [1412.42, 1067.04], [1415.25, 1069.48], [1418.11, 1071.88], [1420.96, 1074.29], [1423.82, 1076.71], [1426.6, 1079.21], [1429.29, 1081.8], [1431.9, 1084.47], [1434.41, 1087.24], [1436.85, 1090.07], [1439.22, 1092.96], [1441.58, 1095.86], [1443.95, 1098.75], [1446.36, 1101.61], [1448.79, 1104.45], [1451.29, 1107.22], [1453.84, 1109.96], [1456.4, 1112.68], [1458.96, 1115.4], [1461.55, 1118.1], [1464.14, 1120.79], [1466.79, 1123.42], [1469.42, 1126.08], [1472.04, 1128.74], [1474.67, 1131.4], [1477.34, 1134.01], [1480.07, 1136.57], [1482.83, 1139.09], [1485.6, 1141.6], [1488.4, 1144.07], [1491.24, 1146.51], [1494.11, 1148.9], [1497.01, 1151.25], [1499.93, 1153.59], [1502.88, 1155.88], [1505.86, 1158.15], [1508.82, 1160.42], [1511.79, 1162.69], [1514.73, 1164.99], [1517.65, 1167.34], [1520.5, 1169.75], [1523.28, 1172.24], [1525.97, 1174.84], [1528.57, 1177.52], [1531.09, 1180.28], [1533.52, 1183.13], [1535.85, 1186.05], [1538.14, 1189.0], [1540.4, 1191.98], [1542.7, 1194.92], [1544.99, 1197.88], [1547.33, 1200.79], [1549.73, 1203.65], [1552.2, 1206.46], [1554.68, 1209.26], [1557.14, 1212.07], [1559.58, 1214.9], [1562.05, 1217.71], [1564.54, 1220.49], [1567.08, 1223.23], [1569.68, 1225.92], [1572.31, 1228.57], [1575.0, 1231.17], [1577.7, 1233.75], [1580.41, 1236.33], [1583.13, 1238.89], [1585.89, 1241.42], [1588.67, 1243.91], [1591.49, 1246.36], [1594.35, 1248.77], [1597.23, 1251.15], [1600.14, 1253.49], [1603.06, 1255.83], [1605.98, 1258.16], [1608.9, 1260.49], [1611.8, 1262.85], [1614.7, 1265.21], [1617.59, 1267.57], [1620.49, 1269.94], [1623.39, 1272.29], [1626.29, 1274.65], [1629.2, 1277.0], [1632.07, 1279.39], [1634.91, 1281.82], [1637.75, 1284.24], [1640.56, 1286.72], [1643.28, 1289.27], [1645.95, 1291.89], [1648.54, 1294.59], [1651.04, 1297.37], [1653.43, 1300.23], [1655.73, 1303.18], [1657.95, 1306.19], [1660.1, 1309.25], [1662.17, 1312.36], [1664.15, 1315.53], [1666.13, 1318.7], [1668.06, 1321.9], [1670.0, 1325.09], [1671.89, 1328.32], [1673.75, 1331.56], [1675.48, 1334.87], [1677.1, 1338.24], [1678.58, 1341.67], [1679.95, 1345.15], [1681.18, 1348.68], [1682.28, 1352.25], [1683.27, 1355.85], [1684.14, 1359.49], [1684.96, 1363.14], [1685.74, 1366.79], [1686.48, 1370.45], [1687.15, 1374.13], [1687.68, 1377.83], [1688.1, 1381.54], [1688.49, 1385.26], [1688.76, 1388.99], [1688.9, 1392.72], [1688.95, 1396.46], [1688.93, 1400.2], [1688.87, 1403.93], [1688.76, 1407.67], [1688.62, 1411.41], [1688.43, 1415.14], [1688.2, 1418.87], [1687.94, 1422.6], [1687.67, 1426.32], [1687.47, 1430.06], [1687.39, 1433.79], [1687.36, 1437.53], [1687.35, 1441.27], [1687.37, 1445.0], [1687.43, 1448.74], [1687.58, 1452.48], [1687.85, 1456.2], [1688.12, 1459.93], [1688.42, 1463.66], [1688.75, 1467.38], [1689.05, 1471.1], [1689.24, 1474.84], [1689.42, 1478.57], [1689.57, 1482.3], [1689.68, 1486.04], [1689.69, 1489.78], [1689.62, 1493.51], [1689.52, 1497.25], [1689.41, 1500.99], [1689.3, 1504.72], [1689.2, 1508.46], [1689.18, 1512.2], [1689.25, 1515.93], [1689.41, 1519.67], [1689.66, 1523.4], [1689.94, 1527.12], [1690.27, 1530.85], [1690.64, 1534.56], [1691.03, 1538.28], [1691.4, 1542.0], [1691.7, 1545.73], [1691.98, 1549.45], [1692.17, 1553.19], [1692.27, 1556.92], [1692.3, 1560.66], [1692.28, 1564.4], [1692.2, 1568.13], [1692.1, 1571.87], [1691.95, 1575.6], [1691.76, 1579.34], [1691.57, 1583.07], [1691.42, 1586.8], [1691.37, 1590.54], [1691.44, 1594.28], [1691.63, 1598.01], [1691.88, 1601.74], [1692.15, 1605.47], [1692.5, 1609.19], [1692.9, 1612.9], [1693.37, 1616.61], [1693.88, 1620.31], [1694.4, 1624.01], [1694.95, 1627.71], [1695.53, 1631.4], [1696.1, 1635.1], [1696.61, 1638.8], [1697.04, 1642.51], [1697.38, 1646.23], [1697.73, 1649.95], [1698.06, 1653.68], [1698.39, 1657.4], [1698.76, 1661.12], [1699.12, 1664.84], [1699.44, 1668.56], [1699.63, 1672.29], [1699.72, 1676.03], [1699.74, 1679.77], [1699.73, 1683.51], [1699.68, 1687.24], [1699.57, 1690.98], [1699.4, 1694.71], [1699.18, 1698.44], [1698.96, 1702.17], [1698.74, 1705.9], [1698.52, 1709.64], [1698.3, 1713.37], [1698.08, 1717.1], [1697.86, 1720.83], [1697.64, 1724.56], [1697.42, 1728.29], [1697.21, 1732.02], [1696.99, 1735.75], [1696.77, 1739.48], [1696.55, 1743.21], [1696.33, 1746.95], [1696.11, 1750.68], [1695.89, 1754.41], [1695.67, 1758.14], [1695.45, 1761.87], [1695.23, 1765.6], [1695.01, 1769.33], [1694.79, 1773.06], [1694.57, 1776.79], [1694.35, 1780.52], [1694.13, 1784.25], [1693.91, 1787.99], [1693.69, 1791.72], [1693.47, 1795.45], [1693.25, 1799.18], [1693.03, 1802.91], [1692.81, 1806.64], [1692.59, 1810.37], [1692.37, 1814.1], [1692.15, 1817.83], [1691.93, 1821.56], [1691.72, 1825.29], [1691.5, 1829.03], [1691.28, 1832.76], [1691.06, 1836.49], [1690.84, 1840.22], [1690.62, 1843.95]]], "image": "images/traverse_ordering_00000.png"} +{"width": 1916, "height": 1916, "num_extra_distractor_curves": 1, "s_curve": 0, "s_at_start": true, "s_point": {"x": 1377.22, "y": 70.0, "label_x": 1325.29, "label_y": 72.86}, "marks": [{"label": "J", "curve": 0, "on_s_curve": true, "polyline_index": 119, "x": 1558.31, "y": 419.48, "label_x": 1513.29, "label_y": 436.12}, {"label": "I", "curve": 0, "on_s_curve": true, "polyline_index": 284, "x": 1489.35, "y": 978.11, "label_x": 1441.35, "label_y": 978.7}, {"label": "A", "curve": 0, "on_s_curve": true, "polyline_index": 478, "x": 943.65, "y": 850.64, "label_x": 968.74, "label_y": 809.71}, {"label": "L", "curve": 0, "on_s_curve": true, "polyline_index": 661, "x": 373.17, "y": 587.24, "label_x": 344.55, "label_y": 548.71}, {"label": "B", "curve": 0, "on_s_curve": true, "polyline_index": 794, "x": 629.62, "y": 895.26, "label_x": 605.03, "label_y": 936.48}, {"label": "E", "curve": 0, "on_s_curve": true, "polyline_index": 1182, "x": 500.24, "y": 1623.92, "label_x": 489.99, "label_y": 1577.03}, {"label": "C", "curve": 1, "on_s_curve": false, "polyline_index": 273, "x": 1275.46, "y": 1423.22, "label_x": 1264.89, "label_y": 1376.4}, {"label": "F", "curve": 1, "on_s_curve": false, "polyline_index": 493, "x": 1344.73, "y": 804.38, "label_x": 1297.25, "label_y": 797.36}, {"label": "H", "curve": 1, "on_s_curve": false, "polyline_index": 686, "x": 1166.74, "y": 805.47, "label_x": 1118.94, "label_y": 801.18}, {"label": "G", "curve": 1, "on_s_curve": false, "polyline_index": 1045, "x": 847.15, "y": 1106.9, "label_x": 801.16, "label_y": 1093.15}, {"label": "D", "curve": 1, "on_s_curve": false, "polyline_index": 1374, "x": 729.26, "y": 1099.78, "label_x": 686.67, "label_y": 1077.64}, {"label": "K", "curve": 1, "on_s_curve": false, "polyline_index": 1689, "x": 571.76, "y": 285.37, "label_x": 589.62, "label_y": 240.81}], "num_marks_total": 12, "num_marks_on_s": 6, "num_pair_crossings": 9, "num_self_crossings": 0, "question": "The image shows two curves or strands that cross each other. Several labeled points are marked on the two curves combined. A green 'S' marks the start endpoint of one of the curves. Starting from S, follow THAT curve (the one S lies on) all the way to its other endpoint, and list the labels of the marked points you visit in order, separated by commas. Ignore any labeled points that lie on the OTHER curve. Provide your final answer enclosed in ... tags.", "answer": "J, I, A, L, B, E", "polylines": [[[1377.22, 70.0], [1377.41, 73.55], [1377.61, 77.1], [1377.8, 80.64], [1378.0, 84.19], [1378.19, 87.74], [1378.39, 91.29], [1378.58, 94.84], [1378.78, 98.38], [1378.98, 101.93], [1379.17, 105.48], [1379.37, 109.03], [1379.56, 112.58], [1379.76, 116.12], [1379.95, 119.67], [1380.15, 123.22], [1380.34, 126.77], [1380.54, 130.31], [1380.74, 133.86], [1380.93, 137.41], [1381.13, 140.96], [1381.32, 144.51], [1381.52, 148.05], [1381.71, 151.6], [1381.91, 155.15], [1382.11, 158.7], [1382.3, 162.25], [1382.5, 165.79], [1382.69, 169.34], [1382.89, 172.89], [1383.08, 176.44], [1383.28, 179.99], [1383.47, 183.53], [1383.67, 187.08], [1383.87, 190.63], [1384.06, 194.18], [1384.26, 197.73], [1384.45, 201.27], [1384.65, 204.82], [1384.84, 208.37], [1385.04, 211.92], [1385.24, 215.46], [1385.43, 219.01], [1385.77, 222.55], [1386.25, 226.07], [1386.87, 229.57], [1387.63, 233.04], [1388.52, 236.48], [1389.56, 239.88], [1390.73, 243.23], [1392.03, 246.54], [1393.47, 249.79], [1395.03, 252.98], [1396.72, 256.1], [1398.54, 259.16], [1400.48, 262.14], [1402.5, 265.06], [1404.61, 267.92], [1406.73, 270.77], [1408.9, 273.58], [1411.12, 276.36], [1413.39, 279.09], [1415.7, 281.79], [1418.07, 284.44], [1420.49, 287.04], [1422.94, 289.62], [1425.37, 292.2], [1427.83, 294.77], [1430.29, 297.34], [1432.73, 299.91], [1435.2, 302.48], [1437.68, 305.02], [1440.19, 307.54], [1442.71, 310.04], [1445.31, 312.46], [1447.97, 314.81], [1450.69, 317.1], [1453.5, 319.27], [1456.37, 321.37], [1459.32, 323.35], [1462.32, 325.25], [1465.33, 327.14], [1468.36, 329.0], [1471.4, 330.84], [1474.43, 332.69], [1477.46, 334.56], [1480.47, 336.44], [1483.48, 338.33], [1486.49, 340.21], [1489.5, 342.11], [1492.5, 344.01], [1495.48, 345.95], [1498.44, 347.91], [1501.41, 349.87], [1504.36, 351.85], [1507.3, 353.84], [1510.3, 355.75], [1513.28, 357.69], [1516.24, 359.64], [1519.19, 361.63], [1522.05, 363.73], [1524.83, 365.94], [1527.52, 368.27], [1530.11, 370.7], [1532.61, 373.23], [1535.0, 375.86], [1537.28, 378.58], [1539.45, 381.4], [1541.51, 384.29], [1543.44, 387.27], [1545.26, 390.33], [1546.96, 393.45], [1548.58, 396.61], [1550.14, 399.8], [1551.65, 403.02], [1553.09, 406.27], [1554.51, 409.52], [1555.85, 412.82], [1557.11, 416.14], [1558.31, 419.48], [1559.5, 422.83], [1560.64, 426.2], [1561.77, 429.57], [1562.84, 432.95], [1563.9, 436.34], [1564.91, 439.75], [1565.83, 443.18], [1566.69, 446.63], [1567.45, 450.1], [1568.13, 453.59], [1568.73, 457.09], [1569.21, 460.61], [1569.54, 464.15], [1569.73, 467.7], [1569.81, 471.25], [1569.8, 474.8], [1569.71, 478.36], [1569.57, 481.91], [1569.33, 485.45], [1569.06, 489.0], [1568.65, 492.53], [1568.1, 496.04], [1567.41, 499.52], [1566.58, 502.98], [1565.63, 506.4], [1564.56, 509.79], [1563.35, 513.13], [1562.01, 516.42], [1560.54, 519.66], [1558.98, 522.85], [1557.37, 526.01], [1555.7, 529.15], [1553.96, 532.25], [1552.14, 535.3], [1550.22, 538.29], [1548.18, 541.2], [1546.03, 544.03], [1543.8, 546.8], [1541.49, 549.5], [1539.08, 552.1], [1536.56, 554.61], [1533.94, 557.01], [1531.23, 559.31], [1528.43, 561.5], [1525.58, 563.62], [1522.82, 565.85], [1520.14, 568.2], [1517.57, 570.64], [1515.09, 573.19], [1512.72, 575.84], [1510.43, 578.56], [1508.26, 581.37], [1506.19, 584.26], [1504.25, 587.23], [1502.42, 590.28], [1500.72, 593.4], [1499.15, 596.59], [1497.7, 599.83], [1496.39, 603.14], [1495.21, 606.49], [1494.16, 609.88], [1493.26, 613.32], [1492.48, 616.79], [1491.8, 620.27], [1491.21, 623.78], [1490.71, 627.29], [1490.27, 630.82], [1489.94, 634.36], [1489.7, 637.9], [1489.56, 641.46], [1489.43, 645.01], [1489.33, 648.56], [1489.28, 652.11], [1489.32, 655.66], [1489.4, 659.22], [1489.49, 662.77], [1489.6, 666.32], [1489.73, 669.87], [1489.91, 673.42], [1490.15, 676.96], [1490.47, 680.5], [1490.81, 684.04], [1491.18, 687.57], [1491.6, 691.1], [1491.99, 694.63], [1492.32, 698.17], [1492.63, 701.71], [1492.89, 705.26], [1493.04, 708.81], [1493.08, 712.36], [1493.02, 715.91], [1492.84, 719.46], [1492.6, 723.01], [1492.28, 726.55], [1491.94, 730.08], [1491.6, 733.62], [1491.27, 737.16], [1490.89, 740.69], [1490.42, 744.21], [1489.97, 747.74], [1489.48, 751.26], [1488.93, 754.77], [1488.34, 758.27], [1487.79, 761.78], [1487.27, 765.3], [1486.88, 768.83], [1486.62, 772.37], [1486.46, 775.92], [1486.36, 779.47], [1486.3, 783.03], [1486.31, 786.58], [1486.36, 790.13], [1486.39, 793.69], [1486.51, 797.24], [1486.7, 800.79], [1486.92, 804.33], [1487.2, 807.87], [1487.49, 811.42], [1487.83, 814.95], [1488.24, 818.48], [1488.7, 822.01], [1489.19, 825.53], [1489.65, 829.05], [1490.09, 832.57], [1490.47, 836.11], [1490.74, 839.65], [1490.95, 843.2], [1491.06, 846.75], [1491.08, 850.3], [1491.05, 853.86], [1491.04, 857.41], [1491.03, 860.96], [1491.01, 864.52], [1491.03, 868.07], [1491.08, 871.62], [1491.12, 875.17], [1491.15, 878.73], [1491.15, 882.28], [1491.15, 885.83], [1491.17, 889.39], [1491.21, 892.94], [1491.25, 896.49], [1491.31, 900.05], [1491.35, 903.6], [1491.4, 907.15], [1491.44, 910.71], [1491.57, 914.26], [1491.67, 917.81], [1491.72, 921.36], [1491.7, 924.91], [1491.54, 928.46], [1491.32, 932.01], [1491.02, 935.55], [1490.66, 939.09], [1490.3, 942.62], [1490.0, 946.16], [1489.73, 949.7], [1489.49, 953.25], [1489.29, 956.8], [1489.2, 960.35], [1489.15, 963.9], [1489.19, 967.46], [1489.22, 971.01], [1489.29, 974.56], [1489.35, 978.11], [1489.39, 981.67], [1489.44, 985.22], [1489.51, 988.77], [1489.59, 992.33], [1489.69, 995.88], [1489.81, 999.43], [1489.95, 1002.98], [1489.94, 1006.53], [1489.79, 1010.08], [1489.5, 1013.62], [1489.06, 1017.15], [1488.49, 1020.66], [1487.77, 1024.14], [1486.92, 1027.59], [1485.93, 1031.0], [1484.8, 1034.37], [1483.54, 1037.69], [1482.14, 1040.96], [1480.62, 1044.17], [1478.97, 1047.32], [1477.19, 1050.39], [1475.3, 1053.4], [1473.28, 1056.32], [1471.15, 1059.17], [1468.9, 1061.92], [1466.55, 1064.58], [1464.09, 1067.15], [1461.54, 1069.62], [1458.88, 1071.98], [1456.13, 1074.23], [1453.3, 1076.37], [1450.38, 1078.4], [1447.38, 1080.3], [1444.31, 1082.09], [1441.17, 1083.75], [1437.96, 1085.28], [1434.7, 1086.69], [1431.38, 1087.96], [1428.01, 1089.1], [1424.6, 1090.1], [1421.16, 1090.96], [1417.68, 1091.69], [1414.17, 1092.28], [1410.65, 1092.72], [1407.11, 1093.03], [1403.56, 1093.19], [1400.0, 1093.21], [1396.45, 1093.08], [1392.91, 1092.82], [1389.38, 1092.41], [1385.87, 1091.86], [1382.38, 1091.17], [1378.93, 1090.35], [1375.5, 1089.4], [1372.09, 1088.41], [1368.68, 1087.41], [1365.29, 1086.36], [1361.89, 1085.32], [1358.5, 1084.25], [1355.12, 1083.15], [1351.74, 1082.07], [1348.34, 1081.02], [1344.95, 1079.97], [1341.54, 1078.96], [1338.13, 1077.95], [1334.74, 1076.89], [1331.35, 1075.82], [1327.96, 1074.77], [1324.56, 1073.73], [1321.15, 1072.73], [1317.74, 1071.74], [1314.32, 1070.77], [1310.9, 1069.81], [1307.48, 1068.84], [1304.06, 1067.88], [1300.64, 1066.9], [1297.25, 1065.84], [1293.89, 1064.69], [1290.53, 1063.52], [1287.21, 1062.27], [1283.92, 1060.93], [1280.66, 1059.51], [1277.42, 1058.04], [1274.23, 1056.49], [1271.1, 1054.81], [1268.01, 1053.05], [1264.97, 1051.21], [1261.98, 1049.29], [1259.05, 1047.28], [1256.16, 1045.22], [1253.28, 1043.14], [1250.39, 1041.07], [1247.45, 1039.07], [1244.52, 1037.06], [1241.6, 1035.03], [1238.71, 1032.97], [1235.83, 1030.89], [1232.98, 1028.76], [1230.17, 1026.59], [1227.35, 1024.43], [1224.49, 1022.32], [1221.61, 1020.24], [1218.74, 1018.14], [1215.89, 1016.02], [1213.07, 1013.86], [1210.27, 1011.67], [1207.52, 1009.43], [1204.79, 1007.14], [1202.05, 1004.89], [1199.24, 1002.7], [1196.42, 1000.55], [1193.53, 998.48], [1190.59, 996.49], [1187.61, 994.55], [1184.59, 992.68], [1181.55, 990.84], [1178.55, 988.94], [1175.56, 987.01], [1172.61, 985.04], [1169.65, 983.07], [1166.73, 981.04], [1163.86, 978.95], [1161.0, 976.83], [1158.14, 974.73], [1155.28, 972.62], [1152.46, 970.46], [1149.62, 968.32], [1146.73, 966.25], [1143.78, 964.28], [1140.77, 962.38], [1137.71, 960.58], [1134.61, 958.85], [1131.49, 957.15], [1128.38, 955.43], [1125.28, 953.7], [1122.2, 951.92], [1119.16, 950.08], [1116.14, 948.2], [1113.16, 946.27], [1110.16, 944.36], [1107.14, 942.5], [1104.09, 940.67], [1101.0, 938.92], [1097.88, 937.22], [1094.75, 935.54], [1091.61, 933.87], [1088.44, 932.26], [1085.27, 930.67], [1082.12, 929.02], [1078.98, 927.35], [1075.83, 925.7], [1072.67, 924.08], [1069.49, 922.5], [1066.28, 920.98], [1063.03, 919.53], [1059.78, 918.11], [1056.51, 916.72], [1053.22, 915.37], [1049.94, 913.99], [1046.68, 912.59], [1043.44, 911.12], [1040.22, 909.62], [1037.0, 908.11], [1033.79, 906.6], [1030.58, 905.07], [1027.36, 903.56], [1024.15, 902.05], [1020.94, 900.52], [1017.74, 898.97], [1014.56, 897.38], [1011.37, 895.83], [1008.17, 894.29], [1004.96, 892.76], [1001.75, 891.23], [998.58, 889.64], [995.43, 887.99], [992.3, 886.3], [989.21, 884.55], [986.18, 882.69], [983.22, 880.73], [980.32, 878.68], [977.45, 876.57], [974.64, 874.41], [971.87, 872.18], [969.09, 869.97], [966.34, 867.71], [963.62, 865.43], [960.87, 863.17], [958.11, 860.94], [955.32, 858.73], [952.51, 856.56], [949.63, 854.48], [946.68, 852.51], [943.65, 850.64], [940.58, 848.87], [937.46, 847.16], [934.3, 845.54], [931.11, 843.96], [927.9, 842.45], [924.66, 840.98], [921.41, 839.55], [918.15, 838.14], [914.87, 836.77], [911.59, 835.4], [908.29, 834.08], [904.97, 832.83], [901.63, 831.61], [898.28, 830.43], [894.92, 829.26], [891.57, 828.09], [888.22, 826.9], [884.89, 825.66], [881.57, 824.38], [878.28, 823.05], [874.98, 821.74], [871.73, 820.3], [868.54, 818.74], [865.37, 817.13], [862.21, 815.5], [859.04, 813.9], [855.84, 812.34], [852.65, 810.78], [849.45, 809.23], [846.27, 807.64], [843.13, 805.99], [840.04, 804.24], [837.0, 802.39], [833.99, 800.5], [831.03, 798.54], [828.11, 796.52], [825.19, 794.49], [822.3, 792.42], [819.45, 790.31], [816.64, 788.12], [813.89, 785.88], [811.17, 783.58], [808.5, 781.25], [805.85, 778.87], [803.22, 776.49], [800.56, 774.13], [797.88, 771.8], [795.13, 769.55], [792.32, 767.37], [789.48, 765.24], [786.65, 763.09], [783.79, 760.99], [780.95, 758.84], [778.12, 756.7], [775.32, 754.51], [772.5, 752.35], [769.66, 750.21], [766.79, 748.12], [763.89, 746.06], [760.97, 744.05], [758.02, 742.07], [755.03, 740.15], [752.04, 738.22], [749.05, 736.3], [746.06, 734.38], [743.03, 732.52], [739.98, 730.7], [736.89, 728.94], [733.75, 727.27], [730.59, 725.66], [727.4, 724.1], [724.18, 722.6], [720.95, 721.11], [717.71, 719.64], [714.48, 718.17], [711.27, 716.64], [708.12, 714.99], [705.02, 713.27], [701.94, 711.48], [698.9, 709.64], [695.88, 707.77], [692.9, 705.84], [689.95, 703.86], [687.02, 701.86], [684.11, 699.81], [681.24, 697.72], [678.39, 695.6], [675.56, 693.45], [672.77, 691.25], [670.02, 689.0], [667.25, 686.77], [664.46, 684.57], [661.62, 682.44], [658.71, 680.4], [655.73, 678.45], [652.72, 676.58], [649.65, 674.79], [646.54, 673.06], [643.43, 671.34], [640.35, 669.56], [637.3, 667.75], [634.29, 665.86], [631.33, 663.9], [628.42, 661.86], [625.55, 659.76], [622.72, 657.61], [619.91, 655.44], [617.06, 653.32], [614.15, 651.27], [611.18, 649.33], [608.14, 647.49], [605.05, 645.73], [601.93, 644.03], [598.79, 642.38], [595.62, 640.77], [592.42, 639.21], [589.22, 637.67], [586.01, 636.15], [582.82, 634.59], [579.64, 633.01], [576.46, 631.42], [573.26, 629.87], [570.06, 628.33], [566.85, 626.8], [563.64, 625.28], [560.43, 623.76], [557.22, 622.24], [553.99, 620.74], [550.75, 619.28], [547.51, 617.83], [544.28, 616.34], [541.08, 614.81], [537.86, 613.31], [534.61, 611.87], [531.35, 610.46], [528.08, 609.06], [524.82, 607.64], [521.57, 606.22], [518.31, 604.79], [515.07, 603.34], [511.83, 601.87], [508.61, 600.38], [505.38, 598.9], [502.14, 597.43], [498.89, 596.0], [495.64, 594.57], [492.39, 593.12], [489.16, 591.64], [485.96, 590.1], [482.77, 588.54], [479.62, 586.89], [476.49, 585.2], [473.4, 583.45], [470.36, 581.62], [467.25, 579.9], [464.07, 578.31], [460.83, 576.85], [457.53, 575.53], [454.19, 574.33], [450.8, 573.27], [447.36, 572.35], [443.9, 571.56], [440.4, 570.92], [436.89, 570.41], [433.35, 570.05], [429.81, 569.83], [426.25, 569.75], [422.7, 569.81], [419.15, 570.02], [415.62, 570.36], [412.1, 570.85], [408.6, 571.48], [405.13, 572.25], [401.7, 573.16], [398.3, 574.2], [394.95, 575.38], [391.65, 576.7], [388.4, 578.14], [385.21, 579.71], [382.09, 581.41], [379.04, 583.24], [376.07, 585.18], [373.17, 587.24], [370.37, 589.42], [367.65, 591.7], [365.02, 594.1], [362.49, 596.6], [360.07, 599.19], [357.75, 601.88], [355.54, 604.67], [353.44, 607.54], [351.46, 610.49], [349.6, 613.51], [347.86, 616.61], [346.25, 619.78], [344.77, 623.01], [343.41, 626.29], [342.19, 629.63], [341.11, 633.01], [340.16, 636.44], [339.35, 639.9], [338.67, 643.39], [338.14, 646.9], [337.75, 650.43], [337.5, 653.98], [337.4, 657.53], [337.43, 661.08], [337.61, 664.63], [337.93, 668.17], [338.39, 671.69], [339.0, 675.19], [339.74, 678.67], [340.62, 682.11], [341.64, 685.52], [342.79, 688.88], [344.08, 692.19], [345.5, 695.45], [347.05, 698.64], [348.73, 701.78], [350.53, 704.84], [352.45, 707.83], [354.49, 710.74], [356.64, 713.56], [358.91, 716.3], [361.28, 718.95], [363.76, 721.49], [366.27, 724.0], [368.85, 726.45], [371.49, 728.83], [374.2, 731.13], [376.92, 733.41], [379.68, 735.65], [382.5, 737.81], [385.35, 739.93], [388.18, 742.08], [391.0, 744.25], [393.78, 746.46], [396.58, 748.65], [399.44, 750.75], [402.36, 752.78], [405.32, 754.75], [408.34, 756.62], [411.4, 758.43], [414.47, 760.21], [417.56, 761.96], [420.62, 763.77], [423.63, 765.66], [426.62, 767.58], [429.62, 769.49], [432.63, 771.38], [435.65, 773.24], [438.69, 775.09], [441.72, 776.94], [444.79, 778.73], [447.92, 780.42], [451.09, 782.01], [454.32, 783.51], [457.57, 784.94], [460.8, 786.42], [463.99, 787.97], [467.18, 789.56], [470.37, 791.11], [473.58, 792.64], [476.81, 794.13], [480.06, 795.56], [483.32, 796.97], [486.6, 798.32], [489.89, 799.67], [493.17, 801.04], [496.44, 802.44], [499.71, 803.83], [502.97, 805.23], [506.24, 806.63], [509.5, 808.05], [512.74, 809.51], [515.98, 810.97], [519.2, 812.47], [522.4, 814.02], [525.56, 815.62], [528.69, 817.31], [531.78, 819.07], [534.85, 820.86], [537.88, 822.72], [540.89, 824.6], [543.87, 826.53], [546.83, 828.5], [549.77, 830.5], [552.69, 832.53], [555.59, 834.58], [558.5, 836.62], [561.4, 838.67], [564.3, 840.73], [567.2, 842.78], [570.06, 844.88], [572.87, 847.06], [575.62, 849.31], [578.39, 851.53], [581.14, 853.78], [583.83, 856.1], [586.47, 858.48], [589.09, 860.88], [591.69, 863.31], [594.25, 865.77], [596.79, 868.26], [599.31, 870.76], [601.83, 873.27], [604.38, 875.74], [606.97, 878.17], [609.6, 880.57], [612.31, 882.86], [615.05, 885.12], [617.87, 887.28], [620.71, 889.42], [623.62, 891.46], [626.59, 893.41], [629.62, 895.26], [632.71, 897.02], [635.86, 898.66], [639.05, 900.23], [642.26, 901.75], [645.5, 903.21], [648.77, 904.61], [652.05, 905.98], [655.35, 907.28], [658.68, 908.53], [662.01, 909.76], [665.32, 911.06], [668.61, 912.39], [671.89, 913.78], [675.15, 915.19], [678.39, 916.64], [681.62, 918.11], [684.85, 919.6], [688.07, 921.1], [691.31, 922.56], [694.55, 924.02], [697.81, 925.44], [701.08, 926.83], [704.37, 928.17], [707.67, 929.49], [710.98, 930.78], [714.27, 932.12], [717.56, 933.46], [720.84, 934.83], [724.11, 936.22], [727.38, 937.61], [730.65, 939.0], [733.92, 940.38], [737.17, 941.83], [740.42, 943.27], [743.66, 944.73], [746.9, 946.19], [750.11, 947.71], [753.28, 949.32], [756.44, 950.94], [759.59, 952.59], [762.75, 954.22], [765.9, 955.86], [769.03, 957.54], [772.15, 959.23], [775.27, 960.93], [778.39, 962.64], [781.47, 964.4], [784.54, 966.2], [787.59, 968.01], [790.59, 969.92], [793.55, 971.88], [796.43, 973.97], [799.25, 976.14], [802.03, 978.35], [804.79, 980.58], [807.5, 982.88], [810.17, 985.23], [812.86, 987.55], [815.55, 989.86], [818.27, 992.15], [821.0, 994.43], [823.75, 996.69], [826.55, 998.87], [829.38, 1001.01], [832.25, 1003.12], [835.12, 1005.2], [838.01, 1007.28], [840.9, 1009.34], [843.83, 1011.35], [846.78, 1013.33], [849.72, 1015.33], [852.6, 1017.4], [855.45, 1019.53], [858.27, 1021.69], [861.07, 1023.88], [863.87, 1026.07], [866.72, 1028.2], [869.6, 1030.28], [872.51, 1032.31], [875.45, 1034.3], [878.37, 1036.33], [881.27, 1038.39], [884.12, 1040.51], [886.98, 1042.61], [889.82, 1044.75], [892.61, 1046.95], [895.36, 1049.2], [898.09, 1051.47], [900.81, 1053.76], [903.51, 1056.07], [906.19, 1058.4], [908.87, 1060.73], [911.57, 1063.04], [914.3, 1065.33], [917.07, 1067.55], [919.87, 1069.74], [922.69, 1071.89], [925.53, 1074.03], [928.36, 1076.18], [931.17, 1078.35], [933.94, 1080.58], [936.68, 1082.84], [939.44, 1085.09], [942.19, 1087.33], [944.94, 1089.58], [947.75, 1091.76], [950.56, 1093.93], [953.38, 1096.09], [956.23, 1098.22], [959.09, 1100.32], [961.98, 1102.39], [964.88, 1104.45], [967.79, 1106.49], [970.7, 1108.52], [973.62, 1110.55], [976.55, 1112.56], [979.49, 1114.56], [982.44, 1116.54], [985.39, 1118.51], [988.35, 1120.48], [991.33, 1122.41], [994.31, 1124.35], [997.22, 1126.38], [1000.05, 1128.54], [1002.82, 1130.77], [1005.51, 1133.09], [1008.11, 1135.51], [1010.6, 1138.04], [1013.02, 1140.64], [1015.37, 1143.31], [1017.67, 1146.02], [1019.89, 1148.79], [1022.01, 1151.64], [1024.08, 1154.53], [1026.13, 1157.44], [1028.16, 1160.35], [1030.23, 1163.24], [1032.33, 1166.11], [1034.44, 1168.96], [1036.58, 1171.8], [1038.75, 1174.62], [1040.93, 1177.42], [1043.13, 1180.21], [1045.39, 1182.95], [1047.69, 1185.66], [1049.99, 1188.37], [1052.27, 1191.09], [1054.56, 1193.81], [1056.84, 1196.54], [1059.14, 1199.24], [1061.44, 1201.96], [1063.65, 1204.74], [1065.82, 1207.55], [1067.94, 1210.4], [1070.08, 1213.24], [1072.17, 1216.11], [1074.21, 1219.02], [1076.18, 1221.98], [1078.09, 1224.98], [1079.96, 1228.0], [1081.82, 1231.03], [1083.72, 1234.03], [1085.65, 1237.01], [1087.6, 1239.98], [1089.55, 1242.95], [1091.57, 1245.88], [1093.64, 1248.76], [1095.7, 1251.66], [1097.75, 1254.56], [1099.81, 1257.46], [1101.74, 1260.44], [1103.56, 1263.49], [1105.25, 1266.62], [1106.81, 1269.81], [1108.25, 1273.06], [1109.55, 1276.36], [1110.72, 1279.72], [1111.76, 1283.12], [1112.66, 1286.56], [1113.41, 1290.03], [1114.03, 1293.53], [1114.51, 1297.05], [1114.85, 1300.59], [1115.05, 1304.13], [1115.1, 1307.69], [1115.01, 1311.24], [1114.78, 1314.78], [1114.41, 1318.32], [1113.89, 1321.83], [1113.24, 1325.33], [1112.44, 1328.79], [1111.51, 1332.22], [1110.44, 1335.61], [1109.24, 1338.95], [1107.9, 1342.24], [1106.43, 1345.48], [1104.83, 1348.65], [1103.11, 1351.76], [1101.27, 1354.8], [1099.3, 1357.76], [1097.22, 1360.64], [1095.02, 1363.43], [1092.71, 1366.13], [1090.3, 1368.74], [1087.79, 1371.25], [1085.18, 1373.67], [1082.54, 1376.05], [1079.86, 1378.37], [1077.12, 1380.64], [1074.36, 1382.88], [1071.53, 1385.03], [1068.65, 1387.11], [1065.73, 1389.14], [1062.82, 1391.17], [1059.89, 1393.19], [1056.97, 1395.21], [1054.03, 1397.2], [1051.08, 1399.18], [1048.08, 1401.08], [1045.05, 1402.94], [1042.0, 1404.77], [1038.93, 1406.55], [1035.82, 1408.27], [1032.7, 1409.97], [1029.56, 1411.63], [1026.41, 1413.28], [1023.28, 1414.96], [1020.16, 1416.66], [1017.03, 1418.35], [1013.86, 1419.95], [1010.65, 1421.48], [1007.42, 1422.95], [1004.14, 1424.31], [1000.82, 1425.58], [997.48, 1426.8], [994.12, 1427.96], [990.75, 1429.08], [987.35, 1430.11], [983.93, 1431.07], [980.49, 1431.98], [977.05, 1432.87], [973.6, 1433.73], [970.16, 1434.6], [966.7, 1435.39], [963.22, 1436.13], [959.74, 1436.86], [956.25, 1437.53], [952.75, 1438.14], [949.25, 1438.72], [945.75, 1439.35], [942.27, 1440.06], [938.8, 1440.85], [935.34, 1441.66], [931.88, 1442.44], [928.41, 1443.22], [924.94, 1444.0], [921.48, 1444.78], [918.01, 1445.57], [914.55, 1446.35], [911.07, 1447.07], [907.57, 1447.73], [904.07, 1448.33], [900.56, 1448.85], [897.03, 1449.29], [893.5, 1449.67], [889.96, 1449.98], [886.42, 1450.25], [882.87, 1450.48], [879.32, 1450.72], [875.78, 1450.96], [872.24, 1451.27], [868.71, 1451.67], [865.2, 1452.21], [861.71, 1452.88], [858.24, 1453.66], [854.8, 1454.53], [851.38, 1455.5], [847.96, 1456.48], [844.54, 1457.43], [841.12, 1458.39], [837.71, 1459.39], [834.31, 1460.44], [830.95, 1461.59], [827.61, 1462.79], [824.28, 1464.04], [820.97, 1465.33], [817.67, 1466.64], [814.36, 1467.95], [811.05, 1469.24], [807.74, 1470.53], [804.44, 1471.85], [801.14, 1473.16], [797.83, 1474.46], [794.53, 1475.78], [791.22, 1477.06], [787.9, 1478.34], [784.57, 1479.57], [781.22, 1480.74], [777.85, 1481.88], [774.47, 1482.99], [771.1, 1484.1], [767.77, 1485.35], [764.49, 1486.71], [761.25, 1488.16], [758.04, 1489.69], [754.88, 1491.32], [751.79, 1493.08], [748.76, 1494.93], [745.79, 1496.88], [742.88, 1498.93], [740.03, 1501.05], [737.21, 1503.2], [734.4, 1505.38], [731.62, 1507.59], [728.83, 1509.79], [726.07, 1512.02], [723.34, 1514.3], [720.69, 1516.67], [718.06, 1519.06], [715.47, 1521.5], [712.9, 1523.95], [710.34, 1526.41], [707.73, 1528.82], [705.02, 1531.12], [702.23, 1533.32], [699.36, 1535.42], [696.5, 1537.52], [693.6, 1539.57], [690.66, 1541.58], [687.73, 1543.59], [684.82, 1545.62], [681.9, 1547.65], [678.93, 1549.59], [675.94, 1551.52], [672.94, 1553.42], [669.88, 1555.22], [666.79, 1556.98], [663.69, 1558.72], [660.56, 1560.41], [657.42, 1562.07], [654.3, 1563.77], [651.24, 1565.57], [648.21, 1567.43], [645.2, 1569.32], [642.2, 1571.23], [639.26, 1573.21], [636.31, 1575.2], [633.33, 1577.13], [630.29, 1578.98], [627.2, 1580.73], [624.08, 1582.43], [620.94, 1584.1], [617.79, 1585.74], [614.61, 1587.33], [611.41, 1588.87], [608.21, 1590.41], [605.02, 1591.97], [601.81, 1593.51], [598.61, 1595.05], [595.42, 1596.61], [592.23, 1598.17], [589.05, 1599.75], [585.86, 1601.32], [582.67, 1602.89], [579.47, 1604.43], [576.21, 1605.84], [572.89, 1607.12], [569.53, 1608.27], [566.12, 1609.28], [562.69, 1610.18], [559.23, 1611.02], [555.78, 1611.87], [552.32, 1612.67], [548.85, 1613.43], [545.36, 1614.12], [541.88, 1614.8], [538.38, 1615.42], [534.88, 1616.05], [531.39, 1616.73], [527.91, 1617.42], [524.42, 1618.12], [520.95, 1618.86], [517.49, 1619.68], [514.04, 1620.54], [510.59, 1621.37], [507.14, 1622.22], [503.69, 1623.07], [500.24, 1623.92], [496.78, 1624.74], [493.31, 1625.51], [489.83, 1626.23], [486.34, 1626.88], [482.83, 1627.44], [479.31, 1627.96], [475.79, 1628.4], [472.26, 1628.79], [468.73, 1629.25], [465.21, 1629.73], [461.69, 1630.19], [458.16, 1630.57], [454.61, 1630.86], [451.07, 1631.16], [447.53, 1631.47], [444.0, 1631.8], [440.46, 1632.14], [436.92, 1632.4], [433.37, 1632.65], [429.82, 1632.86], [426.27, 1633.01], [422.72, 1633.1], [419.17, 1633.22], [415.62, 1633.37], [412.07, 1633.54], [408.52, 1633.7], [404.98, 1633.97], [401.44, 1634.3], [397.91, 1634.74], [394.4, 1635.27], [390.9, 1635.89], [387.42, 1636.58], [383.92, 1637.22], [380.41, 1637.8], [376.89, 1638.28], [373.37, 1638.73], [369.84, 1639.1], [366.3, 1639.4], [362.76, 1639.71], [359.23, 1640.12], [355.7, 1640.57], [352.18, 1641.04], [348.66, 1641.51], [345.14, 1642.01], [341.61, 1642.43], [338.08, 1642.8], [334.54, 1643.13], [331.0, 1643.47], [327.47, 1643.84], [323.94, 1644.27], [320.42, 1644.73], [316.9, 1645.22], [313.38, 1645.73], [309.87, 1646.28], [306.37, 1646.89], [302.86, 1647.47], [299.35, 1647.99], [295.84, 1648.51], [292.32, 1649.02], [288.79, 1649.45], [285.26, 1649.85], [281.73, 1650.26], [278.2, 1650.65], [274.66, 1651.0], [271.12, 1651.31], [267.58, 1651.54], [264.03, 1651.75], [260.48, 1651.94], [256.93, 1652.08], [253.38, 1652.19], [249.83, 1652.28], [246.28, 1652.42], [242.73, 1652.54], [239.17, 1652.67], [235.62, 1652.78], [232.07, 1652.9], [228.52, 1653.06], [224.98, 1653.31], [221.43, 1653.56], [217.89, 1653.8], [214.34, 1654.06], [210.8, 1654.28], [207.25, 1654.48], [203.7, 1654.68], [200.16, 1654.89], [196.61, 1655.19], [193.08, 1655.56], [189.55, 1655.95], [186.02, 1656.4], [182.51, 1656.9], [178.99, 1657.4], [175.47, 1657.89], [171.95, 1658.39], [168.44, 1658.96], [164.95, 1659.59], [161.47, 1660.31], [158.0, 1661.09], [154.56, 1661.97], [151.12, 1662.87], [147.7, 1663.83], [144.3, 1664.86], [140.86, 1665.75], [137.39, 1666.51], [133.89, 1667.12], [130.37, 1667.6], [126.83, 1667.93], [123.28, 1668.12], [119.73, 1668.3], [116.18, 1668.47], [112.63, 1668.65], [109.08, 1668.82], [105.53, 1669.0], [101.99, 1669.17], [98.44, 1669.35], [94.89, 1669.52], [91.34, 1669.7], [87.79, 1669.87], [84.24, 1670.05], [80.69, 1670.22], [77.14, 1670.4], [73.59, 1670.58], [70.04, 1670.75], [66.5, 1670.93]], [[1846.0, 866.13], [1842.77, 865.7], [1839.53, 865.26], [1836.3, 864.83], [1833.06, 864.4], [1829.83, 863.96], [1826.59, 863.53], [1823.36, 863.1], [1820.12, 862.66], [1816.89, 862.23], [1813.65, 861.8], [1810.42, 861.36], [1807.18, 860.93], [1803.95, 860.5], [1800.71, 860.07], [1797.48, 859.63], [1794.24, 859.2], [1791.01, 858.77], [1787.77, 858.33], [1784.54, 857.9], [1781.31, 857.47], [1778.07, 857.03], [1774.84, 856.6], [1771.6, 856.17], [1768.37, 855.73], [1765.13, 855.3], [1761.9, 854.87], [1758.66, 854.43], [1755.43, 854.0], [1752.19, 853.57], [1748.96, 853.13], [1745.72, 852.7], [1742.49, 852.27], [1739.25, 851.83], [1736.02, 851.4], [1732.78, 850.97], [1729.55, 850.53], [1726.32, 850.1], [1723.08, 849.67], [1719.85, 849.23], [1716.61, 848.8], [1713.38, 848.37], [1710.14, 847.93], [1706.91, 847.5], [1703.67, 847.07], [1700.44, 846.63], [1697.19, 846.33], [1693.93, 846.16], [1690.67, 846.12], [1687.4, 846.21], [1684.15, 846.43], [1680.9, 846.78], [1677.67, 847.26], [1674.47, 847.88], [1671.29, 848.62], [1668.14, 849.48], [1665.03, 850.48], [1661.97, 851.59], [1658.95, 852.83], [1655.97, 854.17], [1653.04, 855.6], [1650.14, 857.11], [1647.29, 858.7], [1644.47, 860.34], [1641.68, 862.02], [1638.92, 863.77], [1636.2, 865.56], [1633.53, 867.45], [1630.91, 869.39], [1628.31, 871.36], [1625.75, 873.39], [1623.22, 875.45], [1620.75, 877.59], [1618.31, 879.76], [1615.96, 882.02], [1613.63, 884.31], [1611.36, 886.65], [1609.18, 889.07], [1607.07, 891.57], [1605.03, 894.12], [1603.02, 896.69], [1601.05, 899.29], [1599.18, 901.96], [1597.39, 904.69], [1595.71, 907.49], [1594.11, 910.34], [1592.61, 913.23], [1591.19, 916.17], [1589.86, 919.15], [1588.6, 922.16], [1587.36, 925.18], [1586.12, 928.2], [1584.87, 931.21], [1583.64, 934.24], [1582.4, 937.26], [1581.17, 940.28], [1579.97, 943.31], [1578.76, 946.34], [1577.52, 949.36], [1576.29, 952.39], [1575.09, 955.42], [1573.92, 958.47], [1572.72, 961.5], [1571.54, 964.54], [1570.36, 967.59], [1569.19, 970.64], [1567.97, 973.66], [1566.73, 976.68], [1565.49, 979.7], [1564.27, 982.73], [1563.04, 985.75], [1561.77, 988.75], [1560.49, 991.76], [1559.23, 994.77], [1557.96, 997.78], [1556.66, 1000.77], [1555.34, 1003.75], [1553.97, 1006.71], [1552.53, 1009.64], [1551.06, 1012.56], [1549.49, 1015.42], [1547.85, 1018.24], [1546.12, 1021.01], [1544.36, 1023.75], [1542.52, 1026.45], [1540.66, 1029.13], [1538.74, 1031.77], [1536.83, 1034.42], [1534.89, 1037.04], [1532.93, 1039.65], [1530.96, 1042.26], [1529.05, 1044.9], [1527.21, 1047.6], [1525.47, 1050.36], [1523.83, 1053.18], [1522.29, 1056.06], [1520.81, 1058.97], [1519.41, 1061.91], [1518.02, 1064.87], [1516.67, 1067.84], [1515.27, 1070.79], [1513.8, 1073.7], [1512.32, 1076.61], [1510.83, 1079.51], [1509.33, 1082.41], [1507.83, 1085.31], [1506.28, 1088.18], [1504.64, 1091.01], [1502.94, 1093.79], [1501.17, 1096.53], [1499.43, 1099.29], [1497.67, 1102.04], [1495.86, 1104.75], [1493.98, 1107.43], [1492.08, 1110.08], [1490.13, 1112.7], [1488.13, 1115.28], [1486.1, 1117.83], [1484.05, 1120.37], [1481.99, 1122.9], [1479.92, 1125.43], [1477.86, 1127.95], [1475.79, 1130.48], [1473.71, 1132.99], [1471.65, 1135.52], [1469.62, 1138.08], [1467.6, 1140.64], [1465.6, 1143.23], [1463.64, 1145.83], [1461.66, 1148.42], [1459.68, 1151.02], [1457.68, 1153.6], [1455.74, 1156.23], [1453.91, 1158.93], [1452.12, 1161.66], [1450.32, 1164.38], [1448.5, 1167.09], [1446.66, 1169.78], [1444.83, 1172.48], [1443.01, 1175.19], [1441.18, 1177.9], [1439.36, 1180.6], [1437.56, 1183.33], [1435.74, 1186.03], [1433.87, 1188.71], [1432.01, 1191.39], [1430.13, 1194.06], [1428.24, 1196.72], [1426.32, 1199.36], [1424.44, 1202.03], [1422.56, 1204.7], [1420.77, 1207.43], [1419.09, 1210.23], [1417.49, 1213.07], [1415.94, 1215.94], [1414.41, 1218.82], [1412.85, 1221.69], [1411.25, 1224.54], [1409.6, 1227.35], [1407.88, 1230.12], [1406.14, 1232.88], [1404.37, 1235.63], [1402.6, 1238.37], [1400.82, 1241.1], [1399.03, 1243.83], [1397.23, 1246.55], [1395.39, 1249.25], [1393.52, 1251.93], [1391.66, 1254.61], [1389.82, 1257.3], [1387.98, 1260.0], [1386.13, 1262.69], [1384.26, 1265.36], [1382.39, 1268.04], [1380.58, 1270.75], [1378.85, 1273.52], [1377.18, 1276.32], [1375.61, 1279.18], [1374.07, 1282.06], [1372.63, 1284.99], [1371.22, 1287.93], [1369.85, 1290.9], [1368.56, 1293.89], [1367.27, 1296.89], [1365.99, 1299.89], [1364.75, 1302.91], [1363.53, 1305.94], [1362.29, 1308.96], [1361.07, 1311.99], [1359.82, 1315.0], [1358.53, 1318.0], [1357.2, 1320.98], [1355.81, 1323.93], [1354.44, 1326.89], [1353.08, 1329.86], [1351.69, 1332.82], [1350.29, 1335.76], [1348.85, 1338.69], [1347.46, 1341.64], [1346.09, 1344.61], [1344.75, 1347.58], [1343.42, 1350.56], [1342.14, 1353.56], [1340.88, 1356.57], [1339.59, 1359.57], [1338.28, 1362.56], [1336.91, 1365.53], [1335.54, 1368.49], [1334.17, 1371.45], [1332.85, 1374.43], [1331.57, 1377.43], [1330.17, 1380.38], [1328.65, 1383.27], [1327.01, 1386.09], [1325.27, 1388.85], [1323.41, 1391.54], [1321.45, 1394.14], [1319.38, 1396.67], [1317.22, 1399.11], [1314.95, 1401.46], [1312.6, 1403.72], [1310.16, 1405.89], [1307.63, 1407.95], [1305.02, 1409.91], [1302.33, 1411.76], [1299.57, 1413.51], [1296.75, 1415.14], [1293.86, 1416.66], [1290.91, 1418.05], [1287.91, 1419.33], [1284.85, 1420.49], [1281.76, 1421.53], [1278.63, 1422.44], [1275.46, 1423.22], [1272.26, 1423.87], [1269.04, 1424.4], [1265.8, 1424.79], [1262.55, 1425.06], [1259.28, 1425.19], [1256.02, 1425.2], [1252.76, 1425.07], [1249.51, 1424.81], [1246.27, 1424.42], [1243.04, 1423.9], [1239.85, 1423.25], [1236.68, 1422.47], [1233.54, 1421.57], [1230.44, 1420.54], [1227.39, 1419.39], [1224.39, 1418.11], [1221.44, 1416.72], [1218.54, 1415.21], [1215.71, 1413.58], [1212.95, 1411.84], [1210.26, 1409.99], [1207.65, 1408.04], [1205.12, 1405.98], [1202.67, 1403.82], [1200.31, 1401.56], [1198.05, 1399.21], [1195.88, 1396.78], [1193.81, 1394.25], [1191.84, 1391.65], [1189.98, 1388.97], [1188.23, 1386.21], [1186.59, 1383.39], [1185.06, 1380.51], [1183.65, 1377.56], [1182.37, 1374.56], [1181.2, 1371.51], [1180.11, 1368.44], [1179.11, 1365.33], [1178.2, 1362.2], [1177.37, 1359.04], [1176.6, 1355.87], [1175.92, 1352.68], [1175.29, 1349.47], [1174.73, 1346.26], [1174.23, 1343.04], [1173.77, 1339.8], [1173.37, 1336.56], [1173.02, 1333.32], [1172.8, 1330.06], [1172.67, 1326.8], [1172.64, 1323.54], [1172.7, 1320.28], [1172.84, 1317.02], [1173.11, 1313.76], [1173.51, 1310.52], [1174.04, 1307.3], [1174.67, 1304.1], [1175.42, 1300.93], [1176.29, 1297.78], [1177.22, 1294.65], [1178.13, 1291.52], [1179.07, 1288.39], [1180.02, 1285.27], [1181.02, 1282.16], [1182.06, 1279.07], [1183.11, 1275.98], [1184.18, 1272.9], [1185.28, 1269.82], [1186.41, 1266.76], [1187.54, 1263.7], [1188.7, 1260.65], [1189.89, 1257.61], [1191.11, 1254.58], [1192.36, 1251.57], [1193.68, 1248.58], [1195.05, 1245.62], [1196.42, 1242.66], [1197.82, 1239.71], [1199.21, 1236.76], [1200.61, 1233.81], [1202.03, 1230.87], [1203.51, 1227.96], [1205.01, 1225.06], [1206.54, 1222.18], [1208.11, 1219.32], [1209.66, 1216.45], [1211.2, 1213.57], [1212.71, 1210.68], [1214.14, 1207.74], [1215.54, 1204.8], [1216.92, 1201.84], [1218.28, 1198.87], [1219.59, 1195.88], [1220.81, 1192.86], [1221.96, 1189.8], [1223.07, 1186.73], [1224.11, 1183.64], [1225.18, 1180.55], [1226.24, 1177.47], [1227.2, 1174.35], [1228.11, 1171.22], [1228.97, 1168.07], [1229.77, 1164.9], [1230.48, 1161.72], [1231.14, 1158.52], [1231.74, 1155.31], [1232.34, 1152.11], [1232.86, 1148.89], [1233.37, 1145.66], [1233.82, 1142.43], [1234.25, 1139.19], [1234.69, 1135.96], [1235.18, 1132.73], [1235.74, 1129.52], [1236.34, 1126.31], [1236.98, 1123.11], [1237.69, 1119.93], [1238.44, 1116.75], [1239.16, 1113.56], [1239.86, 1110.38], [1240.56, 1107.19], [1241.24, 1104.0], [1241.95, 1100.81], [1242.72, 1097.64], [1243.53, 1094.48], [1244.39, 1091.33], [1245.27, 1088.19], [1246.1, 1085.03], [1246.87, 1081.86], [1247.63, 1078.69], [1248.42, 1075.52], [1249.21, 1072.35], [1250.04, 1069.2], [1250.92, 1066.05], [1251.84, 1062.92], [1252.72, 1059.78], [1253.53, 1056.62], [1254.3, 1053.45], [1255.07, 1050.28], [1255.84, 1047.11], [1256.56, 1043.92], [1257.29, 1040.74], [1258.01, 1037.56], [1258.75, 1034.38], [1259.55, 1031.21], [1260.38, 1028.06], [1261.23, 1024.91], [1262.12, 1021.77], [1263.1, 1018.66], [1264.03, 1015.52], [1264.94, 1012.39], [1265.88, 1009.27], [1266.87, 1006.16], [1267.91, 1003.06], [1268.98, 999.98], [1270.08, 996.91], [1271.16, 993.83], [1272.22, 990.74], [1273.32, 987.67], [1274.51, 984.63], [1275.76, 981.62], [1277.05, 978.61], [1278.41, 975.65], [1279.81, 972.7], [1281.24, 969.77], [1282.7, 966.85], [1284.18, 963.94], [1285.64, 961.02], [1287.17, 958.14], [1288.73, 955.27], [1290.34, 952.43], [1291.98, 949.61], [1293.65, 946.81], [1295.34, 944.01], [1297.05, 941.24], [1298.76, 938.46], [1300.44, 935.66], [1302.07, 932.83], [1303.7, 930.01], [1305.32, 927.17], [1306.96, 924.35], [1308.62, 921.54], [1310.29, 918.74], [1311.9, 915.9], [1313.45, 913.03], [1315.05, 910.18], [1316.67, 907.34], [1318.33, 904.54], [1319.99, 901.73], [1321.61, 898.89], [1323.16, 896.02], [1324.61, 893.1], [1325.94, 890.12], [1327.15, 887.09], [1328.23, 884.01], [1329.28, 880.92], [1330.26, 877.8], [1331.26, 874.7], [1332.27, 871.6], [1333.25, 868.48], [1334.19, 865.35], [1335.04, 862.2], [1335.88, 859.05], [1336.75, 855.9], [1337.61, 852.76], [1338.42, 849.59], [1339.16, 846.42], [1339.82, 843.22], [1340.4, 840.01], [1340.91, 836.79], [1341.39, 833.56], [1341.83, 830.32], [1342.26, 827.09], [1342.65, 823.85], [1343.01, 820.6], [1343.36, 817.36], [1343.74, 814.12], [1344.09, 810.87], [1344.38, 807.62], [1344.73, 804.38], [1345.2, 801.15], [1345.74, 797.93], [1346.33, 794.72], [1347.03, 791.53], [1347.71, 788.34], [1348.37, 785.14], [1349.03, 781.95], [1349.75, 778.77], [1350.54, 775.6], [1351.4, 772.45], [1352.37, 769.33], [1353.43, 766.25], [1354.58, 763.19], [1355.82, 760.17], [1357.12, 757.18], [1358.43, 754.19], [1359.81, 751.23], [1361.23, 748.3], [1362.67, 745.37], [1364.11, 742.44], [1365.57, 739.52], [1367.02, 736.59], [1368.45, 733.66], [1369.9, 730.73], [1371.37, 727.82], [1372.9, 724.94], [1374.47, 722.08], [1376.03, 719.21], [1377.59, 716.35], [1379.17, 713.49], [1380.76, 710.64], [1382.36, 707.8], [1383.96, 704.95], [1385.55, 702.1], [1387.15, 699.26], [1388.76, 696.42], [1390.39, 693.59], [1392.0, 690.75], [1393.57, 687.89], [1395.06, 684.99], [1396.53, 682.07], [1397.98, 679.15], [1399.36, 676.19], [1400.69, 673.21], [1401.97, 670.21], [1403.19, 667.18], [1404.35, 664.13], [1405.38, 661.04], [1406.34, 657.92], [1407.2, 654.77], [1407.98, 651.6], [1408.63, 648.4], [1409.16, 645.18], [1409.55, 641.94], [1409.81, 638.69], [1409.94, 635.43], [1409.94, 632.16], [1409.8, 628.9], [1409.54, 625.65], [1409.15, 622.41], [1408.62, 619.19], [1407.97, 615.99], [1407.19, 612.82], [1406.28, 609.69], [1405.25, 606.59], [1404.09, 603.54], [1402.81, 600.53], [1401.42, 597.59], [1399.9, 594.7], [1398.27, 591.87], [1396.53, 589.11], [1394.68, 586.42], [1392.72, 583.81], [1390.66, 581.28], [1388.49, 578.84], [1386.24, 576.48], [1383.88, 574.22], [1381.44, 572.05], [1378.92, 569.98], [1376.31, 568.02], [1373.63, 566.16], [1370.87, 564.41], [1368.05, 562.78], [1365.16, 561.26], [1362.21, 559.85], [1359.21, 558.57], [1356.16, 557.41], [1353.07, 556.37], [1349.94, 555.45], [1346.77, 554.67], [1343.57, 554.01], [1340.35, 553.48], [1337.11, 553.08], [1333.86, 552.81], [1330.6, 552.67], [1327.34, 552.66], [1324.08, 552.78], [1320.82, 553.04], [1317.58, 553.42], [1314.36, 553.94], [1311.16, 554.58], [1307.99, 555.35], [1304.85, 556.25], [1301.75, 557.28], [1298.7, 558.43], [1295.69, 559.7], [1292.74, 561.09], [1289.84, 562.59], [1287.01, 564.22], [1284.25, 565.95], [1281.56, 567.8], [1278.94, 569.75], [1276.41, 571.8], [1273.95, 573.96], [1271.59, 576.21], [1269.32, 578.56], [1267.15, 580.99], [1265.08, 583.51], [1263.1, 586.11], [1261.24, 588.79], [1259.48, 591.54], [1257.84, 594.36], [1256.31, 597.24], [1254.9, 600.19], [1253.59, 603.18], [1252.31, 606.18], [1251.07, 609.2], [1249.91, 612.25], [1248.81, 615.32], [1247.74, 618.4], [1246.68, 621.49], [1245.55, 624.55], [1244.38, 627.6], [1243.17, 630.63], [1241.89, 633.63], [1240.6, 636.63], [1239.29, 639.62], [1237.93, 642.59], [1236.54, 645.54], [1235.11, 648.47], [1233.69, 651.41], [1232.27, 654.35], [1230.81, 657.26], [1229.29, 660.15], [1227.76, 663.04], [1226.19, 665.9], [1224.64, 668.77], [1223.1, 671.65], [1221.52, 674.5], [1219.9, 677.34], [1218.26, 680.16], [1216.63, 682.99], [1214.97, 685.79], [1213.29, 688.6], [1211.62, 691.4], [1209.92, 694.19], [1208.22, 696.97], [1206.58, 699.79], [1204.99, 702.64], [1203.41, 705.5], [1201.82, 708.35], [1200.22, 711.19], [1198.62, 714.04], [1197.02, 716.88], [1195.41, 719.72], [1193.83, 722.58], [1192.25, 725.43], [1190.67, 728.29], [1189.09, 731.15], [1187.54, 734.02], [1186.03, 736.91], [1184.57, 739.83], [1183.1, 742.74], [1181.67, 745.67], [1180.28, 748.63], [1178.97, 751.62], [1177.78, 754.65], [1176.67, 757.73], [1175.58, 760.8], [1174.54, 763.89], [1173.56, 767.01], [1172.62, 770.13], [1171.73, 773.27], [1170.92, 776.43], [1170.16, 779.61], [1169.49, 782.8], [1168.91, 786.01], [1168.4, 789.24], [1167.95, 792.47], [1167.57, 795.71], [1167.24, 798.96], [1166.97, 802.21], [1166.74, 805.47], [1166.52, 808.72], [1166.32, 811.98], [1166.14, 815.24], [1165.91, 818.49], [1165.66, 821.75], [1165.41, 825.0], [1165.13, 828.25], [1164.76, 831.5], [1164.34, 834.73], [1163.81, 837.95], [1163.26, 841.17], [1162.72, 844.39], [1162.19, 847.61], [1161.68, 850.83], [1161.23, 854.06], [1160.85, 857.31], [1160.5, 860.55], [1160.2, 863.8], [1159.88, 867.05], [1159.56, 870.3], [1159.27, 873.55], [1159.03, 876.8], [1158.79, 880.06], [1158.53, 883.31], [1158.26, 886.56], [1158.04, 889.82], [1157.79, 893.07], [1157.5, 896.32], [1157.25, 899.58], [1157.03, 902.83], [1156.79, 906.09], [1156.58, 909.34], [1156.4, 912.6], [1156.15, 915.86], [1155.88, 919.11], [1155.63, 922.36], [1155.45, 925.62], [1155.3, 928.88], [1155.18, 932.14], [1155.12, 935.41], [1155.07, 938.67], [1155.0, 941.93], [1154.92, 945.2], [1154.87, 948.46], [1154.82, 951.72], [1154.77, 954.99], [1154.74, 958.25], [1154.74, 961.51], [1154.74, 964.78], [1154.73, 968.04], [1154.69, 971.3], [1154.6, 974.57], [1154.46, 977.83], [1154.26, 981.08], [1154.01, 984.34], [1153.72, 987.59], [1153.38, 990.83], [1153.07, 994.08], [1152.77, 997.33], [1152.51, 1000.59], [1152.26, 1003.84], [1151.98, 1007.09], [1151.7, 1010.34], [1151.35, 1013.59], [1150.91, 1016.82], [1150.36, 1020.04], [1149.67, 1023.23], [1148.88, 1026.4], [1147.97, 1029.53], [1146.96, 1032.63], [1145.91, 1035.72], [1144.85, 1038.81], [1143.78, 1041.89], [1142.73, 1044.98], [1141.74, 1048.09], [1140.8, 1051.22], [1139.94, 1054.37], [1139.13, 1057.53], [1138.37, 1060.7], [1137.66, 1063.89], [1136.97, 1067.08], [1136.28, 1070.27], [1135.57, 1073.45], [1134.94, 1076.66], [1134.31, 1079.86], [1133.68, 1083.06], [1133.04, 1086.26], [1132.36, 1089.45], [1131.67, 1092.64], [1131.01, 1095.84], [1130.35, 1099.03], [1129.73, 1102.24], [1129.17, 1105.45], [1128.66, 1108.68], [1128.16, 1111.9], [1127.65, 1115.13], [1127.13, 1118.35], [1126.62, 1121.57], [1126.16, 1124.8], [1125.72, 1128.03], [1125.24, 1131.26], [1124.75, 1134.49], [1124.26, 1137.72], [1123.81, 1140.95], [1123.37, 1144.18], [1122.95, 1147.42], [1122.55, 1150.66], [1122.18, 1153.9], [1121.78, 1157.14], [1121.41, 1160.38], [1121.05, 1163.63], [1120.68, 1166.87], [1120.34, 1170.11], [1120.01, 1173.36], [1119.68, 1176.61], [1119.3, 1179.85], [1118.92, 1183.09], [1118.54, 1186.33], [1118.17, 1189.57], [1117.79, 1192.82], [1117.4, 1196.06], [1116.98, 1199.29], [1116.47, 1202.52], [1116.01, 1205.75], [1115.53, 1208.98], [1115.07, 1212.21], [1114.59, 1215.44], [1114.1, 1218.66], [1113.61, 1221.89], [1113.14, 1225.12], [1112.67, 1228.35], [1112.19, 1231.58], [1111.72, 1234.81], [1111.16, 1238.02], [1110.5, 1241.22], [1109.71, 1244.38], [1108.79, 1247.51], [1107.82, 1250.63], [1106.73, 1253.71], [1105.52, 1256.74], [1104.19, 1259.72], [1102.74, 1262.64], [1101.17, 1265.5], [1099.49, 1268.3], [1097.69, 1271.03], [1095.79, 1273.68], [1093.79, 1276.25], [1091.68, 1278.74], [1089.47, 1281.15], [1087.17, 1283.46], [1084.78, 1285.68], [1082.3, 1287.81], [1079.73, 1289.83], [1077.09, 1291.74], [1074.38, 1293.55], [1071.59, 1295.25], [1068.73, 1296.83], [1065.82, 1298.3], [1062.85, 1299.65], [1059.82, 1300.88], [1056.75, 1301.98], [1053.64, 1302.97], [1050.49, 1303.82], [1047.31, 1304.55], [1044.1, 1305.15], [1040.87, 1305.62], [1037.63, 1305.96], [1034.37, 1306.17], [1031.11, 1306.25], [1027.85, 1306.2], [1024.59, 1306.02], [1021.34, 1305.7], [1018.11, 1305.26], [1014.89, 1304.68], [1011.71, 1303.98], [1008.55, 1303.15], [1005.43, 1302.2], [1002.35, 1301.11], [999.32, 1299.91], [996.33, 1298.59], [993.41, 1297.14], [990.54, 1295.58], [987.74, 1293.91], [985.01, 1292.12], [982.35, 1290.23], [979.77, 1288.23], [977.27, 1286.13], [974.86, 1283.93], [972.54, 1281.63], [970.32, 1279.25], [968.19, 1276.77], [966.16, 1274.21], [964.24, 1271.58], [962.42, 1268.87], [960.72, 1266.08], [959.12, 1263.23], [957.65, 1260.32], [956.29, 1257.35], [955.05, 1254.33], [953.94, 1251.27], [952.95, 1248.16], [952.08, 1245.01], [951.35, 1241.83], [950.74, 1238.63], [950.26, 1235.4], [949.91, 1232.15], [949.69, 1228.9], [949.6, 1225.63], [949.64, 1222.37], [949.82, 1219.11], [950.12, 1215.86], [950.56, 1212.63], [951.12, 1209.41], [951.82, 1206.22], [952.64, 1203.07], [953.59, 1199.94], [954.66, 1196.86], [955.86, 1193.82], [957.17, 1190.84], [958.59, 1187.9], [960.12, 1185.02], [961.72, 1182.17], [963.4, 1179.37], [965.15, 1176.62], [966.96, 1173.9], [968.84, 1171.24], [970.8, 1168.63], [972.85, 1166.08], [974.99, 1163.62], [977.19, 1161.21], [979.41, 1158.82], [981.7, 1156.5], [984.07, 1154.25], [986.52, 1152.1], [989.02, 1150.0], [991.58, 1147.98], [994.2, 1146.02], [996.85, 1144.12], [999.49, 1142.2], [1002.1, 1140.24], [1004.68, 1138.24], [1007.23, 1136.2], [1009.69, 1134.06], [1012.07, 1131.82], [1014.35, 1129.49], [1016.54, 1127.07], [1018.63, 1124.56], [1020.61, 1121.97], [1022.49, 1119.31], [1024.27, 1116.57], [1025.93, 1113.76], [1027.47, 1110.88], [1028.9, 1107.95], [1030.21, 1104.96], [1031.41, 1101.92], [1032.48, 1098.84], [1033.43, 1095.72], [1034.25, 1092.56], [1034.94, 1089.37], [1035.5, 1086.15], [1035.97, 1082.92], [1036.32, 1079.68], [1036.54, 1076.42], [1036.63, 1073.16], [1036.65, 1069.9], [1036.57, 1066.63], [1036.4, 1063.37], [1036.16, 1060.12], [1035.85, 1056.87], [1035.4, 1053.64], [1034.82, 1050.43], [1034.12, 1047.24], [1033.29, 1044.08], [1032.33, 1040.96], [1031.25, 1037.89], [1030.04, 1034.85], [1028.71, 1031.87], [1027.27, 1028.95], [1025.7, 1026.08], [1024.03, 1023.28], [1022.24, 1020.55], [1020.35, 1017.89], [1018.34, 1015.31], [1016.24, 1012.82], [1014.04, 1010.41], [1011.74, 1008.09], [1009.36, 1005.87], [1006.88, 1003.74], [1004.32, 1001.71], [1001.68, 999.79], [998.97, 997.98], [996.19, 996.28], [993.34, 994.69], [990.42, 993.21], [987.45, 991.86], [984.43, 990.62], [981.37, 989.51], [978.26, 988.52], [975.11, 987.66], [971.93, 986.92], [968.72, 986.32], [965.49, 985.84], [962.25, 985.49], [958.99, 985.27], [955.73, 985.19], [952.47, 985.23], [949.21, 985.41], [945.96, 985.72], [942.72, 986.16], [939.51, 986.72], [936.32, 987.42], [933.16, 988.24], [930.04, 989.19], [926.96, 990.27], [923.92, 991.47], [920.94, 992.78], [918.01, 994.22], [915.14, 995.78], [912.33, 997.44], [909.6, 999.22], [906.94, 1001.11], [904.35, 1003.11], [901.85, 1005.2], [899.39, 1007.35], [897.0, 1009.57], [894.69, 1011.87], [892.47, 1014.27], [890.35, 1016.74], [888.33, 1019.31], [886.33, 1021.89], [884.36, 1024.49], [882.38, 1027.09], [880.41, 1029.69], [878.49, 1032.32], [876.61, 1035.0], [874.77, 1037.69], [872.97, 1040.41], [871.2, 1043.16], [869.45, 1045.91], [867.75, 1048.69], [866.07, 1051.49], [864.5, 1054.36], [863.01, 1057.26], [861.61, 1060.21], [860.28, 1063.19], [858.99, 1066.18], [857.8, 1069.22], [856.73, 1072.31], [855.75, 1075.42], [854.83, 1078.55], [853.96, 1081.7], [853.1, 1084.85], [852.25, 1087.99], [851.37, 1091.14], [850.5, 1094.28], [849.65, 1097.44], [848.83, 1100.59], [847.98, 1103.74], [847.15, 1106.9], [846.28, 1110.05], [845.37, 1113.18], [844.35, 1116.28], [843.29, 1119.37], [842.18, 1122.44], [841.04, 1125.49], [839.9, 1128.55], [838.78, 1131.62], [837.64, 1134.68], [836.45, 1137.72], [835.27, 1140.76], [834.05, 1143.79], [832.82, 1146.81], [831.56, 1149.82], [830.27, 1152.82], [829.06, 1155.85], [827.88, 1158.89], [826.7, 1161.93], [825.56, 1164.99], [824.38, 1168.03], [823.19, 1171.07], [822.01, 1174.12], [820.84, 1177.16], [819.66, 1180.2], [818.51, 1183.26], [817.41, 1186.33], [816.42, 1189.44], [815.5, 1192.57], [814.66, 1195.73], [813.9, 1198.9], [813.19, 1202.09], [812.56, 1205.29], [812.0, 1208.5], [811.5, 1211.73], [811.04, 1214.96], [810.56, 1218.19], [810.13, 1221.42], [809.75, 1224.66], [809.44, 1227.91], [809.19, 1231.17], [808.97, 1234.42], [808.82, 1237.68], [808.65, 1240.94], [808.5, 1244.2], [808.39, 1247.46], [808.32, 1250.73], [808.32, 1253.99], [808.34, 1257.25], [808.4, 1260.52], [808.35, 1263.78], [808.2, 1267.04], [807.97, 1270.3], [807.61, 1273.54], [807.21, 1276.78], [806.77, 1280.01], [806.36, 1283.25], [805.97, 1286.49], [805.59, 1289.73], [805.23, 1292.98], [804.93, 1296.23], [804.65, 1299.48], [804.37, 1302.73], [804.11, 1305.98], [803.87, 1309.24], [803.63, 1312.49], [803.42, 1315.75], [803.26, 1319.01], [803.11, 1322.27], [802.97, 1325.53], [802.89, 1328.79], [802.86, 1332.06], [802.78, 1335.32], [802.66, 1338.58], [802.55, 1341.84], [802.44, 1345.1], [802.35, 1348.36], [802.32, 1351.63], [802.3, 1354.89], [802.28, 1358.16], [802.2, 1361.42], [802.09, 1364.68], [801.98, 1367.94], [801.9, 1371.2], [801.83, 1374.47], [801.82, 1377.73], [801.74, 1380.99], [801.65, 1384.26], [801.55, 1387.52], [801.42, 1390.78], [801.34, 1394.04], [801.26, 1397.3], [801.18, 1400.57], [801.12, 1403.83], [801.02, 1407.09], [800.92, 1410.35], [800.83, 1413.62], [800.72, 1416.88], [800.63, 1420.14], [800.52, 1423.4], [800.39, 1426.66], [800.25, 1429.92], [800.06, 1433.18], [799.84, 1436.44], [799.57, 1439.69], [799.2, 1442.93], [798.79, 1446.17], [798.34, 1449.4], [797.9, 1452.64], [797.45, 1455.87], [797.02, 1459.1], [796.61, 1462.34], [796.21, 1465.58], [795.85, 1468.83], [795.48, 1472.07], [795.1, 1475.31], [794.68, 1478.55], [794.26, 1481.78], [793.8, 1485.01], [793.28, 1488.24], [792.75, 1491.46], [792.25, 1494.68], [791.71, 1497.9], [791.05, 1501.1], [790.41, 1504.3], [789.79, 1507.5], [789.16, 1510.7], [788.57, 1513.91], [787.85, 1517.1], [787.01, 1520.25], [786.04, 1523.36], [784.94, 1526.44], [783.72, 1529.47], [782.39, 1532.44], [780.93, 1535.36], [779.36, 1538.22], [777.67, 1541.02], [775.87, 1543.74], [773.97, 1546.39], [771.96, 1548.96], [769.84, 1551.45], [767.63, 1553.85], [765.33, 1556.16], [762.93, 1558.38], [760.45, 1560.49], [757.88, 1562.51], [755.24, 1564.42], [752.52, 1566.22], [749.73, 1567.92], [746.87, 1569.5], [743.95, 1570.96], [740.98, 1572.3], [737.95, 1573.53], [734.88, 1574.63], [731.77, 1575.61], [728.62, 1576.46], [725.43, 1577.18], [722.22, 1577.77], [718.99, 1578.24], [715.75, 1578.58], [712.49, 1578.78], [709.23, 1578.85], [705.97, 1578.8], [702.71, 1578.61], [699.46, 1578.29], [696.23, 1577.84], [693.01, 1577.26], [689.83, 1576.55], [686.67, 1575.71], [683.56, 1574.75], [680.48, 1573.67], [677.45, 1572.46], [674.47, 1571.13], [671.54, 1569.68], [668.68, 1568.11], [665.88, 1566.43], [663.15, 1564.64], [660.5, 1562.75], [657.92, 1560.74], [655.43, 1558.64], [653.02, 1556.43], [650.7, 1554.13], [648.48, 1551.74], [646.36, 1549.27], [644.33, 1546.7], [642.42, 1544.06], [640.6, 1541.35], [638.9, 1538.56], [637.32, 1535.71], [635.85, 1532.8], [634.49, 1529.83], [633.26, 1526.81], [632.15, 1523.74], [631.17, 1520.62], [630.28, 1517.49], [629.51, 1514.31], [628.84, 1511.12], [628.19, 1507.92], [627.6, 1504.71], [627.02, 1501.5], [626.48, 1498.28], [625.99, 1495.05], [625.54, 1491.82], [625.15, 1488.58], [624.8, 1485.34], [624.56, 1482.08], [624.39, 1478.82], [624.27, 1475.56], [624.22, 1472.3], [624.23, 1469.03], [624.31, 1465.77], [624.46, 1462.51], [624.65, 1459.25], [624.83, 1455.99], [625.01, 1452.74], [625.18, 1449.48], [625.34, 1446.22], [625.49, 1442.96], [625.63, 1439.7], [625.77, 1436.44], [625.88, 1433.17], [625.97, 1429.91], [626.05, 1426.65], [626.14, 1423.39], [626.29, 1420.13], [626.5, 1416.87], [626.76, 1413.62], [627.1, 1410.37], [627.4, 1407.12], [627.75, 1403.88], [628.12, 1400.63], [628.46, 1397.39], [628.81, 1394.14], [629.17, 1390.9], [629.54, 1387.66], [629.97, 1384.42], [630.47, 1381.2], [630.95, 1377.97], [631.43, 1374.74], [631.87, 1371.51], [632.28, 1368.27], [632.65, 1365.03], [633.0, 1361.78], [633.32, 1358.53], [633.65, 1355.29], [634.01, 1352.04], [634.36, 1348.8], [634.7, 1345.55], [635.02, 1342.3], [635.35, 1339.06], [635.69, 1335.81], [636.03, 1332.57], [636.37, 1329.32], [636.69, 1326.07], [637.02, 1322.83], [637.36, 1319.58], [637.71, 1316.33], [638.12, 1313.1], [638.58, 1309.87], [639.01, 1306.63], [639.44, 1303.4], [639.9, 1300.16], [640.41, 1296.94], [641.05, 1293.74], [641.79, 1290.56], [642.62, 1287.4], [643.51, 1284.26], [644.47, 1281.15], [645.53, 1278.06], [646.7, 1275.01], [647.97, 1272.01], [649.32, 1269.04], [650.77, 1266.11], [652.31, 1263.23], [653.89, 1260.38], [655.56, 1257.57], [657.26, 1254.79], [659.0, 1252.03], [660.77, 1249.29], [662.55, 1246.55], [664.32, 1243.81], [666.13, 1241.09], [667.96, 1238.39], [669.8, 1235.69], [671.68, 1233.02], [673.54, 1230.35], [675.44, 1227.69], [677.38, 1225.07], [679.33, 1222.45], [681.26, 1219.82], [683.18, 1217.18], [685.09, 1214.53], [686.99, 1211.88], [688.86, 1209.21], [690.67, 1206.49], [692.44, 1203.74], [694.15, 1200.97], [695.75, 1198.12], [697.28, 1195.24], [698.69, 1192.3], [699.99, 1189.3], [701.19, 1186.27], [702.33, 1183.21], [703.46, 1180.15], [704.54, 1177.07], [705.56, 1173.97], [706.5, 1170.84], [707.43, 1167.72], [708.29, 1164.57], [709.11, 1161.41], [709.85, 1158.23], [710.55, 1155.04], [711.2, 1151.84], [711.82, 1148.64], [712.41, 1145.43], [713.02, 1142.22], [713.67, 1139.02], [714.38, 1135.84], [715.15, 1132.67], [716.03, 1129.52], [717.0, 1126.41], [718.07, 1123.33], [719.2, 1120.26], [720.4, 1117.23], [721.69, 1114.23], [723.09, 1111.28], [724.56, 1108.37], [726.09, 1105.49], [727.67, 1102.63], [729.26, 1099.78], [730.91, 1096.96], [732.56, 1094.15], [734.16, 1091.31], [735.66, 1088.4], [737.09, 1085.47], [738.45, 1082.51], [739.78, 1079.53], [741.03, 1076.51], [742.19, 1073.46], [743.34, 1070.41], [744.43, 1067.33], [745.5, 1064.25], [746.53, 1061.15], [747.52, 1058.04], [748.51, 1054.93], [749.46, 1051.81], [750.38, 1048.68], [751.25, 1045.53], [752.05, 1042.37], [752.81, 1039.19], [753.51, 1036.01], [754.18, 1032.81], [754.83, 1029.61], [755.46, 1026.41], [756.04, 1023.2], [756.61, 1019.99], [757.11, 1016.76], [757.55, 1013.53], [757.86, 1010.28], [758.12, 1007.03], [758.31, 1003.77], [758.43, 1000.51], [758.5, 997.24], [758.59, 993.98], [758.68, 990.72], [758.8, 987.46], [758.95, 984.2], [759.01, 980.93], [759.0, 977.67], [758.98, 974.41], [758.98, 971.14], [758.92, 967.88], [758.83, 964.62], [758.76, 961.35], [758.72, 958.09], [758.72, 954.83], [758.72, 951.56], [758.77, 948.3], [758.9, 945.04], [759.05, 941.78], [759.16, 938.52], [759.26, 935.26], [759.37, 931.99], [759.52, 928.73], [759.77, 925.48], [760.12, 922.23], [760.56, 919.0], [761.05, 915.77], [761.58, 912.55], [762.13, 909.34], [762.69, 906.12], [763.31, 902.92], [764.0, 899.73], [764.77, 896.56], [765.63, 893.41], [766.5, 890.26], [767.4, 887.13], [768.36, 884.01], [769.36, 880.9], [770.38, 877.8], [771.42, 874.71], [772.51, 871.63], [773.64, 868.57], [774.76, 865.5], [775.9, 862.44], [777.07, 859.4], [778.28, 856.37], [779.5, 853.34], [780.72, 850.31], [781.97, 847.3], [783.24, 844.29], [784.51, 841.28], [785.8, 838.29], [787.1, 835.29], [788.37, 832.29], [789.64, 829.28], [790.92, 826.28], [792.19, 823.27], [793.46, 820.27], [794.75, 817.27], [796.03, 814.27], [797.32, 811.27], [798.61, 808.27], [799.9, 805.27], [801.2, 802.28], [802.52, 799.3], [803.84, 796.31], [805.2, 793.34], [806.54, 790.37], [807.9, 787.4], [809.21, 784.41], [810.53, 781.42], [811.85, 778.44], [813.14, 775.44], [814.42, 772.44], [815.7, 769.44], [816.99, 766.44], [818.27, 763.44], [819.58, 760.45], [820.92, 757.47], [822.18, 754.46], [823.37, 751.43], [824.47, 748.35], [825.49, 745.25], [826.46, 742.14], [827.41, 739.01], [828.37, 735.89], [829.35, 732.78], [830.32, 729.66], [831.25, 726.54], [832.12, 723.39], [832.86, 720.21], [833.52, 717.02], [834.07, 713.8], [834.48, 710.56], [834.79, 707.31], [835.04, 704.06], [835.23, 700.8], [835.37, 697.54], [835.46, 694.28], [835.48, 691.01], [835.43, 687.75], [835.35, 684.49], [835.2, 681.23], [835.02, 677.97], [834.79, 674.71], [834.54, 671.46], [834.23, 668.21], [833.92, 664.96], [833.57, 661.72], [833.16, 658.48], [832.73, 655.25], [832.3, 652.01], [831.9, 648.77], [831.51, 645.53], [831.17, 642.28], [830.84, 639.04], [830.56, 635.79], [830.32, 632.53], [830.05, 629.28], [829.76, 626.03], [829.5, 622.77], [829.18, 619.53], [828.87, 616.28], [828.67, 613.02], [828.58, 609.76], [828.62, 606.5], [828.76, 603.23], [828.97, 599.98], [829.16, 596.72], [829.33, 593.46], [829.48, 590.2], [829.64, 586.94], [829.8, 583.68], [829.97, 580.42], [830.16, 577.16], [830.41, 573.91], [830.74, 570.66], [831.16, 567.43], [831.62, 564.19], [832.14, 560.97], [832.72, 557.76], [833.37, 554.56], [834.04, 551.37], [834.74, 548.18], [835.46, 545.0], [836.24, 541.83], [837.08, 538.67], [837.91, 535.52], [838.73, 532.36], [839.5, 529.19], [840.24, 526.01], [841.04, 522.85], [841.86, 519.69], [842.71, 516.54], [843.59, 513.39], [844.42, 510.24], [845.2, 507.07], [845.87, 503.87], [846.4, 500.65], [846.81, 497.42], [847.09, 494.16], [847.27, 490.91], [847.38, 487.64], [847.46, 484.38], [847.51, 481.12], [847.53, 477.85], [847.46, 474.59], [847.33, 471.33], [847.17, 468.07], [846.92, 464.82], [846.64, 461.57], [846.32, 458.32], [846.0, 455.07], [845.67, 451.82], [845.34, 448.58], [844.97, 445.33], [844.63, 442.09], [844.3, 438.84], [844.02, 435.59], [843.76, 432.34], [843.45, 429.09], [843.11, 425.84], [842.73, 422.6], [842.32, 419.36], [841.95, 416.12], [841.56, 412.88], [841.04, 409.66], [840.4, 406.46], [839.62, 403.29], [838.72, 400.15], [837.7, 397.05], [836.55, 394.0], [835.27, 390.99], [833.88, 388.04], [832.37, 385.15], [830.75, 382.32], [829.01, 379.55], [827.17, 376.86], [825.21, 374.25], [823.16, 371.71], [821.0, 369.26], [818.75, 366.9], [816.42, 364.61], [814.03, 362.39], [811.56, 360.26], [809.01, 358.22], [806.42, 356.23], [803.82, 354.26], [801.21, 352.3], [798.61, 350.34], [795.98, 348.39], [793.37, 346.44], [790.74, 344.5], [788.12, 342.56], [785.49, 340.62], [782.86, 338.69], [780.22, 336.78], [777.57, 334.87], [774.91, 332.99], [772.2, 331.16], [769.44, 329.41], [766.63, 327.76], [763.77, 326.18], [760.87, 324.69], [757.91, 323.31], [754.9, 322.05], [751.84, 320.91], [748.75, 319.87], [745.63, 318.92], [742.48, 318.07], [739.32, 317.26], [736.15, 316.46], [732.99, 315.65], [729.8, 314.96], [726.6, 314.34], [723.38, 313.77], [720.16, 313.24], [716.94, 312.74], [713.7, 312.3], [710.47, 311.89], [707.23, 311.5], [703.98, 311.16], [700.74, 310.79], [697.5, 310.37], [694.28, 309.86], [691.06, 309.3], [687.84, 308.76], [684.62, 308.24], [681.4, 307.7], [678.18, 307.18], [674.96, 306.68], [671.73, 306.19], [668.49, 305.81], [665.23, 305.56], [661.97, 305.44], [658.71, 305.44], [655.45, 305.53], [652.19, 305.68], [648.93, 305.91], [645.68, 306.24], [642.43, 306.5], [639.17, 306.71], [635.91, 306.84], [632.65, 306.83], [629.39, 306.7], [626.14, 306.43], [622.9, 306.04], [619.67, 305.53], [616.47, 304.88], [613.3, 304.11], [610.17, 303.21], [607.07, 302.19], [604.01, 301.05], [600.99, 299.79], [598.04, 298.42], [595.13, 296.93], [592.27, 295.36], [589.42, 293.76], [586.6, 292.12], [583.72, 290.59], [580.78, 289.18], [577.79, 287.86], [574.78, 286.6], [571.76, 285.37], [568.71, 284.2], [565.64, 283.09], [562.55, 282.05], [559.45, 281.03], [556.32, 280.1], [553.18, 279.23], [550.01, 278.45], [546.82, 277.75], [543.62, 277.11], [540.42, 276.45], [537.24, 275.72], [534.07, 274.96], [530.91, 274.15], [527.75, 273.32], [524.6, 272.48], [521.45, 271.62], [518.29, 270.8], [515.13, 269.99], [511.96, 269.19], [508.8, 268.4], [505.62, 267.65], [502.44, 266.9], [499.25, 266.24], [496.04, 265.65], [492.82, 265.12], [489.59, 264.63], [486.37, 264.12], [483.14, 263.66], [479.9, 263.28], [476.65, 262.95], [473.4, 262.69], [470.14, 262.47], [466.88, 262.3], [463.62, 262.14], [460.36, 262.03], [457.1, 261.94], [453.83, 261.88], [450.57, 261.8], [447.31, 261.73], [444.04, 261.67], [440.78, 261.6], [437.52, 261.48], [434.26, 261.32], [431.0, 261.14], [427.75, 260.94], [424.49, 260.72], [421.24, 260.47], [417.99, 260.11], [414.76, 259.66], [411.53, 259.18], [408.31, 258.66], [405.09, 258.1], [401.88, 257.5], [398.68, 256.89], [395.48, 256.23], [392.3, 255.51], [389.12, 254.79], [385.93, 254.07], [382.75, 253.35], [379.56, 252.64], [376.37, 251.98], [373.17, 251.34], [369.96, 250.74], [366.74, 250.23], [363.51, 249.74], [360.27, 249.33], [357.04, 248.87], [353.81, 248.43], [350.58, 247.97], [347.34, 247.58], [344.09, 247.22], [340.84, 246.93], [337.58, 246.74], [334.32, 246.63], [331.06, 246.56], [327.8, 246.51], [324.53, 246.49], [321.27, 246.48], [318.01, 246.47], [314.74, 246.43], [311.48, 246.28], [308.23, 246.06], [304.98, 245.73], [301.74, 245.33], [298.5, 244.92], [295.26, 244.51], [292.04, 244.02], [288.82, 243.47], [285.61, 242.88], [282.4, 242.3], [279.2, 241.67], [275.98, 241.12], [272.77, 240.53], [269.57, 239.91], [266.37, 239.27], [263.17, 238.62], [259.97, 237.95], [256.77, 237.32], [253.58, 236.65], [250.39, 235.94], [247.22, 235.17], [244.05, 234.39], [240.88, 233.61], [237.72, 232.82], [234.55, 232.04], [231.38, 231.24], [228.21, 230.48], [225.02, 229.78], [221.82, 229.12], [218.61, 228.55], [215.38, 228.08], [212.14, 227.73], [208.88, 227.52], [205.62, 227.33], [202.36, 227.15], [199.1, 226.96], [195.85, 226.78], [192.59, 226.59], [189.33, 226.41], [186.07, 226.22], [182.81, 226.03], [179.55, 225.85], [176.3, 225.66], [173.04, 225.48], [169.78, 225.29], [166.52, 225.11], [163.26, 224.92], [160.0, 224.73], [156.75, 224.55], [153.49, 224.36], [150.23, 224.18], [146.97, 223.99], [143.71, 223.81], [140.45, 223.62], [137.2, 223.43], [133.94, 223.25], [130.68, 223.06], [127.42, 222.88], [124.16, 222.69], [120.9, 222.51], [117.65, 222.32], [114.39, 222.13], [111.13, 221.95], [107.87, 221.76], [104.61, 221.58], [101.35, 221.39], [98.1, 221.21], [94.84, 221.02], [91.58, 220.83], [88.32, 220.65], [85.06, 220.46], [81.8, 220.28], [78.55, 220.09], [75.29, 219.91], [72.03, 219.72], [68.77, 219.53]], [[1248.21, 1846.0], [1248.34, 1842.33], [1248.46, 1838.65], [1248.59, 1834.98], [1248.71, 1831.31], [1248.84, 1827.63], [1248.96, 1823.96], [1249.09, 1820.29], [1249.21, 1816.61], [1249.34, 1812.94], [1249.46, 1809.27], [1249.59, 1805.59], [1249.71, 1801.92], [1249.83, 1798.25], [1249.96, 1794.57], [1250.08, 1790.9], [1250.21, 1787.23], [1250.33, 1783.55], [1250.46, 1779.88], [1250.58, 1776.21], [1250.71, 1772.53], [1250.83, 1768.86], [1250.96, 1765.19], [1251.08, 1761.51], [1251.21, 1757.84], [1251.33, 1754.17], [1251.46, 1750.49], [1251.58, 1746.82], [1251.71, 1743.15], [1251.83, 1739.47], [1251.96, 1735.8], [1252.08, 1732.13], [1252.21, 1728.45], [1252.33, 1724.78], [1252.45, 1721.11], [1252.58, 1717.43], [1252.7, 1713.76], [1252.83, 1710.09], [1252.95, 1706.41], [1253.08, 1702.74], [1253.2, 1699.07], [1253.46, 1695.4], [1253.85, 1691.75], [1254.36, 1688.11], [1255.01, 1684.49], [1255.79, 1680.9], [1256.69, 1677.33], [1257.72, 1673.81], [1258.88, 1670.32], [1260.16, 1666.87], [1261.56, 1663.47], [1263.08, 1660.13], [1264.72, 1656.84], [1266.48, 1653.61], [1268.35, 1650.45], [1270.34, 1647.36], [1272.43, 1644.34], [1274.63, 1641.39], [1276.93, 1638.53], [1279.34, 1635.75], [1281.84, 1633.06], [1284.44, 1630.46], [1287.13, 1627.95], [1289.9, 1625.54], [1292.76, 1623.23], [1295.71, 1621.03], [1298.72, 1618.93], [1301.81, 1616.93], [1304.96, 1615.04], [1308.18, 1613.27], [1311.46, 1611.61], [1314.79, 1610.07], [1318.18, 1608.64], [1321.61, 1607.31], [1325.06, 1606.04], [1328.54, 1604.85], [1332.05, 1603.78], [1335.6, 1602.83], [1339.16, 1601.93], [1342.74, 1601.07], [1346.33, 1600.27], [1349.92, 1599.49], [1353.52, 1598.74], [1357.12, 1598.04], [1360.73, 1597.35], [1364.35, 1596.71], [1367.98, 1596.09], [1371.6, 1595.48], [1375.23, 1594.89], [1378.87, 1594.4], [1382.53, 1594.04], [1386.19, 1593.75], [1389.86, 1593.55], [1393.54, 1593.4], [1397.21, 1593.38], [1400.89, 1593.42], [1404.56, 1593.53], [1408.23, 1593.68], [1411.9, 1593.82], [1415.58, 1593.92], [1419.25, 1594.0], [1422.93, 1594.05], [1426.6, 1594.1], [1430.28, 1594.12], [1433.95, 1594.12], [1437.63, 1594.2], [1441.3, 1594.26], [1444.98, 1594.29], [1448.65, 1594.3], [1452.33, 1594.29], [1456.01, 1594.27], [1459.68, 1594.26], [1463.36, 1594.28], [1467.03, 1594.31], [1470.71, 1594.39], [1474.38, 1594.5], [1478.05, 1594.61], [1481.72, 1594.79], [1485.4, 1594.95], [1489.07, 1594.98], [1492.75, 1594.88], [1496.41, 1594.65], [1500.07, 1594.28], [1503.71, 1593.79], [1507.34, 1593.17], [1510.93, 1592.41], [1514.5, 1591.53], [1518.04, 1590.52], [1521.53, 1589.39], [1524.99, 1588.13], [1528.39, 1586.75], [1531.75, 1585.25], [1535.05, 1583.63], [1538.28, 1581.89], [1541.46, 1580.04], [1544.57, 1578.08], [1547.6, 1576.0], [1550.56, 1573.82], [1553.44, 1571.54], [1556.23, 1569.15], [1558.94, 1566.66], [1561.56, 1564.08], [1564.08, 1561.41], [1566.51, 1558.65], [1568.83, 1555.81], [1571.06, 1552.88], [1573.17, 1549.87], [1575.18, 1546.8], [1577.08, 1543.65], [1578.86, 1540.43], [1580.53, 1537.16], [1582.08, 1533.82], [1583.5, 1530.44], [1584.81, 1527.0], [1585.99, 1523.52], [1587.05, 1520.0], [1587.98, 1516.45], [1588.79, 1512.86], [1589.46, 1509.25], [1590.01, 1505.61], [1590.42, 1501.96], [1590.71, 1498.3], [1590.86, 1494.63], [1590.88, 1490.95], [1590.77, 1487.28], [1590.53, 1483.61], [1590.16, 1479.95], [1589.66, 1476.31], [1589.03, 1472.69], [1588.27, 1469.09], [1587.38, 1465.53], [1586.36, 1462.0], [1585.22, 1458.5], [1583.96, 1455.05], [1582.57, 1451.65], [1581.06, 1448.3], [1579.43, 1445.0], [1577.69, 1441.77], [1575.83, 1438.6], [1573.86, 1435.49], [1571.78, 1432.46], [1569.59, 1429.51], [1567.3, 1426.64], [1564.9, 1423.85], [1562.41, 1421.15], [1559.83, 1418.53], [1557.15, 1416.02], [1554.38, 1413.6], [1551.53, 1411.28], [1548.6, 1409.06], [1545.59, 1406.95], [1542.51, 1404.95], [1539.35, 1403.06], [1536.15, 1401.26], [1532.88, 1399.57], [1529.56, 1398.0], [1526.18, 1396.56], [1522.75, 1395.23], [1519.28, 1394.03], [1515.77, 1392.95], [1512.22, 1392.0], [1508.63, 1391.17], [1505.04, 1390.39], [1501.43, 1389.73], [1497.8, 1389.12], [1494.17, 1388.58], [1490.51, 1388.18], [1486.85, 1387.87], [1483.19, 1387.62], [1479.51, 1387.45], [1475.84, 1387.31], [1472.17, 1387.18], [1468.49, 1387.09], [1464.82, 1387.0], [1461.15, 1386.91], [1457.47, 1386.94], [1453.8, 1387.08], [1450.13, 1387.28], [1446.46, 1387.5], [1442.78, 1387.59], [1439.11, 1387.55], [1435.44, 1387.38], [1431.77, 1387.08], [1428.12, 1386.65], [1424.49, 1386.09], [1420.88, 1385.4], [1417.3, 1384.58], [1413.75, 1383.63], [1410.23, 1382.55], [1406.76, 1381.35], [1403.33, 1380.03], [1399.95, 1378.59], [1396.62, 1377.02], [1393.36, 1375.34], [1390.15, 1373.54], [1387.01, 1371.63], [1383.94, 1369.61], [1380.95, 1367.48], [1378.03, 1365.24], [1375.19, 1362.9], [1372.45, 1360.47], [1369.78, 1357.93], [1367.22, 1355.3], [1364.74, 1352.58], [1362.37, 1349.78], [1360.1, 1346.89], [1357.93, 1343.92], [1355.87, 1340.87], [1353.92, 1337.76], [1352.08, 1334.58], [1350.36, 1331.33], [1348.72, 1328.04], [1347.18, 1324.7], [1345.76, 1321.31], [1344.41, 1317.89], [1343.13, 1314.45], [1341.95, 1310.97], [1340.8, 1307.48], [1339.75, 1303.96], [1338.82, 1300.4], [1338.02, 1296.81], [1337.35, 1293.2], [1336.8, 1289.56], [1336.39, 1285.91], [1336.11, 1282.25], [1335.87, 1278.58], [1335.7, 1274.91], [1335.56, 1271.24], [1335.29, 1267.57], [1334.89, 1263.92], [1334.37, 1260.28], [1333.71, 1256.66], [1332.92, 1253.07], [1332.01, 1249.51], [1330.97, 1245.99], [1329.81, 1242.5], [1328.53, 1239.05], [1327.12, 1235.66], [1325.63, 1232.3], [1324.06, 1228.98], [1322.36, 1225.72], [1320.55, 1222.52], [1318.62, 1219.39], [1316.62, 1216.31], [1314.6, 1213.24], [1312.55, 1210.19], [1310.47, 1207.16], [1308.39, 1204.13], [1306.26, 1201.13], [1304.11, 1198.15], [1301.91, 1195.2], [1299.65, 1192.31], [1297.3, 1189.48], [1294.85, 1186.74], [1292.32, 1184.07], [1289.73, 1181.47], [1287.07, 1178.93], [1284.33, 1176.48], [1281.5, 1174.13], [1278.68, 1171.77], [1275.85, 1169.43], [1273.05, 1167.05], [1270.28, 1164.64], [1267.5, 1162.23], [1264.75, 1159.8], [1261.99, 1157.37], [1259.25, 1154.91], [1256.52, 1152.45], [1253.76, 1150.03], [1250.99, 1147.61], [1248.19, 1145.23], [1245.33, 1142.93], [1242.38, 1140.73], [1239.38, 1138.61], [1236.33, 1136.55], [1233.22, 1134.59], [1230.09, 1132.68], [1226.95, 1130.76], [1223.78, 1128.9], [1220.58, 1127.1], [1217.34, 1125.36], [1214.07, 1123.68], [1210.76, 1122.08], [1207.44, 1120.51], [1204.08, 1119.01], [1200.72, 1117.51], [1197.37, 1116.02], [1194.02, 1114.49], [1190.68, 1112.96], [1187.4, 1111.31], [1184.18, 1109.54], [1181.0, 1107.69], [1177.87, 1105.76], [1174.79, 1103.75], [1171.77, 1101.67], [1168.79, 1099.51], [1165.87, 1097.28], [1162.98, 1095.01], [1160.11, 1092.71], [1157.22, 1090.44], [1154.28, 1088.23], [1151.31, 1086.07], [1148.34, 1083.9], [1145.37, 1081.75], [1142.38, 1079.61], [1139.34, 1077.54], [1136.31, 1075.45], [1133.3, 1073.36], [1130.29, 1071.24], [1127.24, 1069.19], [1124.14, 1067.21], [1120.99, 1065.32], [1117.8, 1063.5], [1114.56, 1061.76], [1111.29, 1060.08], [1107.99, 1058.46], [1104.68, 1056.88], [1101.34, 1055.34], [1097.98, 1053.85], [1094.6, 1052.4], [1091.25, 1050.89], [1087.91, 1049.36], [1084.62, 1047.71], [1081.4, 1045.94], [1078.24, 1044.06], [1075.15, 1042.07], [1072.14, 1039.97], [1069.2, 1037.76], [1066.34, 1035.45], [1063.56, 1033.04], [1060.88, 1030.53], [1058.28, 1027.93], [1055.78, 1025.24], [1053.38, 1022.46], [1051.08, 1019.59], [1048.88, 1016.64], [1046.79, 1013.62], [1044.81, 1010.53], [1042.94, 1007.36], [1041.18, 1004.13], [1039.55, 1000.84], [1038.03, 997.5], [1036.63, 994.1], [1035.35, 990.65], [1034.2, 987.16], [1033.17, 983.63], [1032.27, 980.07], [1031.5, 976.48], [1030.85, 972.86], [1030.34, 969.22], [1029.89, 965.57], [1029.45, 961.92], [1029.15, 958.26], [1028.95, 954.59], [1028.87, 950.91], [1028.92, 947.24], [1029.1, 943.57], [1029.41, 939.9], [1029.86, 936.26], [1030.43, 932.63], [1031.11, 929.01], [1031.9, 925.42], [1032.8, 921.86], [1033.8, 918.33], [1034.83, 914.8], [1035.92, 911.29], [1037.07, 907.8], [1038.26, 904.32], [1039.47, 900.85], [1040.68, 897.38], [1041.86, 893.9], [1042.97, 890.39], [1043.95, 886.85], [1044.84, 883.28], [1045.64, 879.7], [1046.37, 876.09], [1047.07, 872.49], [1047.85, 868.89], [1048.71, 865.32], [1049.64, 861.76], [1050.63, 858.23], [1051.67, 854.7], [1052.77, 851.19], [1053.85, 847.68], [1054.94, 844.17], [1056.05, 840.67], [1057.18, 837.17], [1058.36, 833.69], [1059.56, 830.21], [1060.75, 826.74], [1061.94, 823.26], [1063.11, 819.78], [1064.24, 816.28], [1065.37, 812.78], [1066.47, 809.27], [1067.62, 805.78], [1068.79, 802.3], [1069.98, 798.82], [1071.09, 795.32], [1072.23, 791.82], [1073.4, 788.34], [1074.6, 784.86], [1075.82, 781.4], [1077.07, 777.94], [1078.26, 774.46], [1079.41, 770.97], [1080.59, 767.49], [1081.75, 764.0], [1082.94, 760.53], [1084.1, 757.04], [1085.19, 753.53], [1086.2, 749.99], [1087.19, 746.45], [1088.08, 742.89], [1088.95, 739.32], [1089.81, 735.74], [1090.62, 732.16], [1091.31, 728.55], [1091.88, 724.92], [1092.34, 721.27], [1092.73, 717.62], [1093.0, 713.95], [1093.2, 710.28], [1093.3, 706.61], [1093.36, 702.93], [1093.35, 699.26], [1093.29, 695.58], [1093.19, 691.91], [1093.03, 688.24], [1092.83, 684.57], [1092.67, 680.9], [1092.52, 677.22], [1092.39, 673.55], [1092.19, 669.88], [1091.96, 666.21], [1091.74, 662.54], [1091.49, 658.88], [1091.24, 655.21], [1090.98, 651.54], [1090.67, 647.88], [1090.37, 644.22], [1090.06, 640.56], [1089.74, 636.89], [1089.48, 633.23], [1089.35, 629.55], [1089.31, 625.88], [1089.39, 622.2], [1089.57, 618.53], [1089.71, 614.86], [1089.88, 611.19], [1090.08, 607.52], [1090.37, 603.86], [1090.8, 600.2], [1091.32, 596.57], [1091.92, 592.94], [1092.56, 589.32], [1093.23, 585.71], [1094.0, 582.11], [1094.85, 578.54], [1095.82, 574.99], [1096.84, 571.46], [1097.87, 567.93], [1098.96, 564.42], [1100.07, 560.92], [1101.23, 557.43], [1102.44, 553.96], [1103.66, 550.49], [1104.85, 547.02], [1106.01, 543.53], [1107.05, 540.0], [1107.95, 536.44], [1108.75, 532.85], [1109.47, 529.25], [1110.14, 525.63], [1110.78, 522.02], [1111.41, 518.39], [1111.95, 514.76], [1112.39, 511.11], [1112.72, 507.45], [1112.92, 503.78], [1112.99, 500.11], [1112.93, 496.43], [1112.74, 492.76], [1112.42, 489.1], [1111.96, 485.45], [1111.38, 481.82], [1110.66, 478.22], [1109.82, 474.64], [1108.85, 471.09], [1107.76, 467.59], [1106.54, 464.12], [1105.19, 460.7], [1103.73, 457.33], [1102.14, 454.01], [1100.44, 450.75], [1098.62, 447.56], [1096.69, 444.43], [1094.65, 441.37], [1092.5, 438.39], [1090.25, 435.49], [1087.89, 432.67], [1085.44, 429.93], [1082.89, 427.29], [1080.24, 424.74], [1077.51, 422.28], [1074.69, 419.92], [1071.78, 417.67], [1068.8, 415.52], [1065.75, 413.48], [1062.62, 411.55], [1059.42, 409.73], [1056.17, 408.03], [1052.85, 406.44], [1049.48, 404.98], [1046.06, 403.63], [1042.59, 402.41], [1039.08, 401.32], [1035.54, 400.35], [1031.96, 399.5], [1028.36, 398.79], [1024.73, 398.21], [1021.09, 397.65], [1017.45, 397.13], [1013.8, 396.75], [1010.13, 396.49], [1006.46, 396.37], [1002.78, 396.37], [999.11, 396.51], [995.45, 396.78], [991.79, 397.15], [988.14, 397.6], [984.51, 398.17], [980.9, 398.85], [977.31, 399.66], [973.75, 400.58], [970.22, 401.6], [966.71, 402.68], [963.2, 403.78], [959.68, 404.82], [956.14, 405.8], [952.59, 406.75], [949.01, 407.62], [945.43, 408.41], [941.81, 409.09], [938.19, 409.71], [934.57, 410.33], [930.95, 410.98], [927.34, 411.66], [923.72, 412.29], [920.11, 412.99], [916.5, 413.7], [912.9, 414.4], [909.28, 415.09], [905.68, 415.79], [902.06, 416.42], [898.42, 416.99], [894.78, 417.48], [891.13, 417.92], [887.48, 418.31], [883.82, 418.66], [880.16, 418.99], [876.51, 419.41], [872.86, 419.88], [869.21, 420.33], [865.56, 420.76], [861.92, 421.27], [858.29, 421.8], [854.66, 422.37], [851.02, 422.88], [847.37, 423.36], [843.75, 423.98], [840.14, 424.67], [836.53, 425.37], [832.92, 426.05], [829.3, 426.68], [825.67, 427.26], [822.03, 427.78], [818.4, 428.34], [814.76, 428.85], [811.11, 429.33], [807.47, 429.77], [803.82, 430.25], [800.18, 430.76], [796.54, 431.26], [792.9, 431.73], [789.26, 432.26], [785.64, 432.91], [782.02, 433.55], [778.43, 434.32], [774.86, 435.21], [771.33, 436.22], [767.83, 437.33], [764.36, 438.55], [760.91, 439.83], [757.46, 441.09], [754.01, 442.37], [750.6, 443.73], [747.22, 445.18], [743.84, 446.62], [740.49, 448.12], [737.16, 449.68], [733.86, 451.3], [730.54, 452.88], [727.19, 454.4], [723.8, 455.83], [720.38, 457.17], [716.92, 458.39], [713.43, 459.54], [709.92, 460.65], [706.39, 461.66], [702.85, 462.64], [699.3, 463.61], [695.74, 464.53], [692.17, 465.4], [688.6, 466.27], [685.02, 467.11], [681.43, 467.91], [677.84, 468.69], [674.26, 469.51], [670.69, 470.39], [667.14, 471.33], [663.62, 472.39], [660.13, 473.55], [656.67, 474.79], [653.24, 476.11], [649.85, 477.52], [646.48, 479.0], [643.11, 480.47], [639.7, 481.84], [636.26, 483.13], [632.8, 484.35], [629.3, 485.49], [625.78, 486.53], [622.25, 487.56], [618.71, 488.56], [615.17, 489.55], [611.64, 490.55], [608.09, 491.52], [604.53, 492.42], [600.95, 493.28], [597.38, 494.13], [593.79, 494.92], [590.19, 495.69], [586.6, 496.45], [582.99, 497.17], [579.39, 497.88], [575.79, 498.63], [572.22, 499.49], [568.65, 500.38], [565.11, 501.35], [561.59, 502.42], [558.09, 503.54], [554.6, 504.7], [551.12, 505.88], [547.64, 507.08], [544.22, 508.4], [540.8, 509.77], [537.42, 511.2], [534.08, 512.73], [530.71, 514.2], [527.34, 515.68], [523.96, 517.13], [520.54, 518.45], [517.06, 519.65], [513.64, 520.98], [510.26, 522.42], [506.93, 523.99], [503.67, 525.68], [500.46, 527.48], [497.32, 529.39], [494.26, 531.41], [491.26, 533.55], [488.35, 535.79], [485.51, 538.13], [482.77, 540.57], [480.11, 543.11], [477.54, 545.74], [475.07, 548.46], [472.7, 551.27], [470.43, 554.16], [468.26, 557.13], [466.21, 560.17], [464.26, 563.29], [462.42, 566.47], [460.7, 569.72], [459.08, 573.02], [457.53, 576.35], [456.06, 579.72], [454.72, 583.14], [453.43, 586.58], [452.22, 590.05], [451.14, 593.57], [450.18, 597.12], [449.35, 600.7], [448.65, 604.3], [448.08, 607.94], [447.64, 611.58], [447.27, 615.24], [446.91, 618.9], [446.63, 622.56], [446.47, 626.24], [446.44, 629.91], [446.52, 633.58], [446.66, 637.26], [446.92, 640.92], [447.24, 644.59], [447.63, 648.24], [447.97, 651.9], [448.18, 655.57], [448.26, 659.24], [448.2, 662.92], [448.02, 666.59], [447.7, 670.25], [447.26, 673.9], [446.68, 677.53], [445.98, 681.14], [445.14, 684.72], [444.18, 688.26], [443.1, 691.78], [441.88, 695.24], [440.55, 698.67], [439.09, 702.04], [437.51, 705.36], [435.82, 708.63], [434.01, 711.82], [432.09, 714.96], [430.05, 718.02], [427.91, 721.0], [425.66, 723.91], [423.31, 726.74], [420.87, 729.48], [418.32, 732.13], [415.68, 734.69], [412.95, 737.15], [410.14, 739.52], [407.24, 741.78], [404.26, 743.93], [401.21, 745.98], [398.09, 747.92], [394.9, 749.74], [391.65, 751.45], [388.33, 753.05], [384.97, 754.52], [381.55, 755.87], [378.11, 757.17], [374.66, 758.44], [371.21, 759.72], [367.77, 760.99], [364.31, 762.25], [360.85, 763.48], [357.38, 764.7], [353.9, 765.87], [350.37, 766.9], [346.81, 767.82], [343.23, 768.66], [339.64, 769.42], [336.04, 770.15], [332.44, 770.9], [328.85, 771.7], [325.28, 772.56], [321.72, 773.47], [318.17, 774.43], [314.62, 775.37], [311.08, 776.36], [307.55, 777.4], [304.03, 778.45], [300.51, 779.51], [296.98, 780.55], [293.47, 781.64], [289.98, 782.78], [286.48, 783.91], [283.0, 785.09], [279.52, 786.27], [276.04, 787.45], [272.57, 788.65], [269.09, 789.84], [265.6, 790.99], [262.11, 792.16], [258.62, 793.32], [255.14, 794.48], [251.65, 795.63], [248.17, 796.81], [244.68, 797.99], [241.19, 799.14], [237.69, 800.25], [234.17, 801.3], [230.65, 802.35], [227.13, 803.43], [223.62, 804.51], [220.08, 805.51], [216.54, 806.5], [213.0, 807.47], [209.42, 808.31], [205.81, 809.02], [202.18, 809.6], [198.55, 810.12], [194.91, 810.64], [191.27, 811.16], [187.63, 811.68], [183.99, 812.2], [180.35, 812.72], [176.72, 813.24], [173.08, 813.77], [169.44, 814.29], [165.8, 814.81], [162.16, 815.33], [158.52, 815.85], [154.89, 816.37], [151.25, 816.89], [147.61, 817.41], [143.97, 817.93], [140.33, 818.46], [136.69, 818.98], [133.06, 819.5], [129.42, 820.02], [125.78, 820.54], [122.14, 821.06], [118.5, 821.58], [114.86, 822.1], [111.23, 822.62], [107.59, 823.14], [103.95, 823.67], [100.31, 824.19], [96.67, 824.71], [93.03, 825.23], [89.4, 825.75], [85.76, 826.27], [82.12, 826.79], [78.48, 827.31], [74.84, 827.83], [71.21, 828.35], [67.57, 828.88], [63.93, 829.4]]], "image": "images/traverse_ordering_00001.png"} +{"width": 1916, "height": 1916, "num_extra_distractor_curves": 1, "s_curve": 0, "s_at_start": true, "s_point": {"x": 803.98, "y": 70.0, "label_x": 751.98, "label_y": 69.56}, "marks": [{"label": "I", "curve": 0, "on_s_curve": true, "polyline_index": 149, "x": 1087.3, "y": 362.58, "label_x": 1080.8, "label_y": 410.13}, {"label": "A", "curve": 0, "on_s_curve": true, "polyline_index": 348, "x": 768.54, "y": 385.45, "label_x": 773.7, "label_y": 337.73}, {"label": "E", "curve": 0, "on_s_curve": true, "polyline_index": 563, "x": 1064.48, "y": 849.23, "label_x": 1032.63, "label_y": 885.15}, {"label": "F", "curve": 0, "on_s_curve": true, "polyline_index": 852, "x": 1091.76, "y": 1265.95, "label_x": 1115.72, "label_y": 1224.36}, {"label": "C", "curve": 0, "on_s_curve": true, "polyline_index": 989, "x": 684.18, "y": 1112.39, "label_x": 679.76, "label_y": 1064.59}, {"label": "H", "curve": 0, "on_s_curve": true, "polyline_index": 1136, "x": 380.56, "y": 1394.45, "label_x": 334.66, "label_y": 1380.43}, {"label": "G", "curve": 1, "on_s_curve": false, "polyline_index": 128, "x": 1490.22, "y": 1321.8, "label_x": 1508.92, "label_y": 1366.01}, {"label": "L", "curve": 1, "on_s_curve": false, "polyline_index": 332, "x": 1704.74, "y": 962.08, "label_x": 1656.76, "label_y": 961.06}, {"label": "K", "curve": 1, "on_s_curve": false, "polyline_index": 545, "x": 1465.32, "y": 588.49, "label_x": 1473.34, "label_y": 541.17}, {"label": "D", "curve": 1, "on_s_curve": false, "polyline_index": 722, "x": 1074.78, "y": 1103.6, "label_x": 1031.47, "label_y": 1082.92}, {"label": "B", "curve": 1, "on_s_curve": false, "polyline_index": 959, "x": 837.33, "y": 973.02, "label_x": 881.3, "label_y": 992.27}, {"label": "J", "curve": 1, "on_s_curve": false, "polyline_index": 1330, "x": 314.32, "y": 913.45, "label_x": 332.33, "label_y": 868.95}], "num_marks_total": 12, "num_marks_on_s": 6, "num_pair_crossings": 9, "num_self_crossings": 1, "question": "The image shows two curves or strands that cross each other. Several labeled points are marked on the two curves combined. A green 'S' marks the start endpoint of one of the curves. Starting from S, follow THAT curve (the one S lies on) all the way to its other endpoint, and list the labels of the marked points you visit in order, separated by commas. Ignore any labeled points that lie on the OTHER curve. Provide your final answer enclosed in ... tags.", "answer": "I, A, E, F, C, H", "polylines": [[[803.98, 70.0], [803.95, 73.3], [803.92, 76.61], [803.9, 79.91], [803.87, 83.22], [803.84, 86.52], [803.81, 89.82], [803.79, 93.13], [803.76, 96.43], [803.73, 99.74], [803.7, 103.04], [803.68, 106.34], [803.65, 109.65], [803.62, 112.95], [803.59, 116.26], [803.56, 119.56], [803.54, 122.86], [803.51, 126.17], [803.48, 129.47], [803.45, 132.78], [803.43, 136.08], [803.4, 139.38], [803.37, 142.69], [803.34, 145.99], [803.32, 149.3], [803.29, 152.6], [803.26, 155.9], [803.23, 159.21], [803.2, 162.51], [803.18, 165.82], [803.15, 169.12], [803.12, 172.42], [803.09, 175.73], [803.07, 179.03], [803.04, 182.34], [803.01, 185.64], [802.98, 188.94], [802.96, 192.25], [802.93, 195.55], [802.9, 198.86], [802.87, 202.16], [802.84, 205.46], [802.82, 208.77], [802.79, 212.07], [802.76, 215.38], [802.73, 218.68], [802.83, 221.98], [803.04, 225.28], [803.37, 228.57], [803.81, 231.84], [804.38, 235.1], [805.06, 238.33], [805.86, 241.54], [806.77, 244.71], [807.8, 247.85], [808.94, 250.95], [810.19, 254.01], [811.55, 257.02], [813.02, 259.98], [814.6, 262.89], [816.27, 265.73], [818.05, 268.52], [819.93, 271.23], [821.91, 273.88], [823.98, 276.46], [826.12, 278.97], [828.36, 281.4], [830.68, 283.75], [833.09, 286.02], [835.57, 288.2], [838.13, 290.28], [840.77, 292.28], [843.47, 294.17], [846.25, 295.97], [849.08, 297.67], [851.97, 299.27], [854.92, 300.76], [857.91, 302.17], [860.94, 303.47], [864.03, 304.66], [867.15, 305.74], [870.31, 306.72], [873.5, 307.57], [876.7, 308.37], [879.93, 309.09], [883.17, 309.74], [886.43, 310.3], [889.69, 310.8], [892.97, 311.23], [896.25, 311.63], [899.53, 312.02], [902.8, 312.45], [906.07, 312.95], [909.32, 313.55], [912.55, 314.23], [915.76, 315.03], [918.95, 315.89], [922.12, 316.83], [925.28, 317.8], [928.41, 318.85], [931.52, 319.94], [934.62, 321.09], [937.7, 322.29], [940.79, 323.48], [943.89, 324.6], [947.02, 325.66], [950.16, 326.69], [953.32, 327.66], [956.51, 328.53], [959.71, 329.35], [962.92, 330.12], [966.14, 330.88], [969.35, 331.64], [972.58, 332.35], [975.81, 333.03], [979.04, 333.73], [982.27, 334.45], [985.5, 335.15], [988.73, 335.83], [991.97, 336.5], [995.2, 337.15], [998.42, 337.9], [1001.63, 338.69], [1004.83, 339.53], [1008.0, 340.44], [1011.16, 341.41], [1014.3, 342.45], [1017.43, 343.51], [1020.56, 344.56], [1023.68, 345.66], [1026.79, 346.76], [1029.9, 347.87], [1033.0, 349.01], [1036.09, 350.2], [1039.19, 351.34], [1042.29, 352.47], [1045.39, 353.61], [1048.52, 354.68], [1051.66, 355.71], [1054.83, 356.64], [1058.02, 357.5], [1061.23, 358.29], [1064.45, 359.02], [1067.68, 359.7], [1070.94, 360.29], [1074.2, 360.8], [1077.47, 361.3], [1080.74, 361.76], [1084.01, 362.18], [1087.3, 362.58], [1090.58, 362.93], [1093.87, 363.28], [1097.14, 363.7], [1100.42, 364.13], [1103.7, 364.55], [1106.96, 365.07], [1110.2, 365.7], [1113.42, 366.46], [1116.61, 367.32], [1119.76, 368.31], [1122.88, 369.4], [1125.95, 370.61], [1128.98, 371.93], [1131.96, 373.35], [1134.89, 374.89], [1137.76, 376.52], [1140.57, 378.26], [1143.32, 380.1], [1145.99, 382.04], [1148.6, 384.07], [1151.13, 386.2], [1153.58, 388.42], [1155.95, 390.72], [1158.23, 393.1], [1160.43, 395.57], [1162.54, 398.12], [1164.55, 400.73], [1166.47, 403.42], [1168.29, 406.18], [1170.01, 409.0], [1171.63, 411.89], [1173.14, 414.82], [1174.54, 417.81], [1175.84, 420.85], [1177.02, 423.94], [1178.1, 427.06], [1179.06, 430.22], [1179.9, 433.42], [1180.63, 436.64], [1181.25, 439.89], [1181.74, 443.15], [1182.12, 446.44], [1182.38, 449.73], [1182.52, 453.03], [1182.54, 456.34], [1182.44, 459.64], [1182.22, 462.93], [1181.89, 466.22], [1181.43, 469.49], [1180.86, 472.75], [1180.17, 475.98], [1179.37, 479.18], [1178.45, 482.36], [1177.41, 485.5], [1176.27, 488.59], [1175.01, 491.65], [1173.64, 494.66], [1172.17, 497.61], [1170.59, 500.52], [1168.9, 503.36], [1167.12, 506.14], [1165.23, 508.85], [1163.25, 511.5], [1161.18, 514.07], [1159.01, 516.56], [1156.76, 518.98], [1154.41, 521.31], [1151.99, 523.56], [1149.49, 525.71], [1146.91, 527.78], [1144.26, 529.75], [1141.54, 531.63], [1138.75, 533.4], [1135.9, 535.07], [1132.99, 536.64], [1130.03, 538.11], [1127.02, 539.46], [1123.96, 540.71], [1120.86, 541.84], [1117.71, 542.87], [1114.54, 543.77], [1111.33, 544.57], [1108.1, 545.25], [1104.84, 545.81], [1101.56, 546.25], [1098.28, 546.57], [1094.98, 546.78], [1091.68, 546.86], [1088.37, 546.83], [1085.07, 546.69], [1081.77, 546.48], [1078.48, 546.17], [1075.2, 545.77], [1071.93, 545.31], [1068.67, 544.81], [1065.41, 544.25], [1062.16, 543.63], [1058.93, 542.94], [1055.71, 542.2], [1052.49, 541.45], [1049.29, 540.66], [1046.08, 539.84], [1042.9, 538.95], [1039.72, 538.06], [1036.54, 537.16], [1033.37, 536.23], [1030.2, 535.29], [1027.06, 534.28], [1023.92, 533.24], [1020.81, 532.14], [1017.73, 530.93], [1014.69, 529.64], [1011.69, 528.25], [1008.71, 526.83], [1005.75, 525.36], [1002.85, 523.78], [999.98, 522.14], [997.11, 520.51], [994.21, 518.92], [991.3, 517.36], [988.41, 515.76], [985.52, 514.15], [982.65, 512.51], [979.79, 510.86], [976.95, 509.17], [974.18, 507.37], [971.45, 505.51], [968.73, 503.64], [966.0, 501.77], [963.28, 499.9], [960.56, 498.02], [957.82, 496.18], [955.03, 494.41], [952.21, 492.68], [949.37, 491.0], [946.49, 489.38], [943.61, 487.76], [940.78, 486.05], [938.01, 484.25], [935.27, 482.4], [932.55, 480.52], [929.84, 478.63], [927.12, 476.75], [924.4, 474.88], [921.67, 473.02], [918.92, 471.19], [916.14, 469.41], [913.32, 467.68], [910.48, 466.0], [907.61, 464.35], [904.74, 462.73], [901.83, 461.15], [898.92, 459.6], [895.99, 458.06], [893.06, 456.54], [890.12, 455.04], [887.15, 453.57], [884.2, 452.09], [881.24, 450.63], [878.26, 449.19], [875.29, 447.76], [872.31, 446.33], [869.32, 444.91], [866.34, 443.48], [863.38, 442.03], [860.45, 440.49], [857.57, 438.88], [854.74, 437.17], [851.97, 435.37], [849.26, 433.48], [846.59, 431.54], [843.96, 429.53], [841.4, 427.44], [838.89, 425.29], [836.43, 423.09], [833.98, 420.87], [831.52, 418.66], [829.11, 416.41], [826.66, 414.19], [824.19, 412.0], [821.71, 409.81], [819.2, 407.66], [816.64, 405.57], [814.04, 403.53], [811.38, 401.58], [808.64, 399.73], [805.84, 397.98], [802.97, 396.33], [800.05, 394.78], [797.08, 393.35], [794.05, 392.02], [790.98, 390.8], [787.87, 389.69], [784.72, 388.69], [781.54, 387.81], [778.32, 387.04], [775.08, 386.4], [771.82, 385.86], [768.54, 385.45], [765.25, 385.15], [761.95, 384.98], [758.65, 384.92], [755.35, 384.98], [752.05, 385.16], [748.76, 385.46], [745.48, 385.88], [742.22, 386.41], [738.98, 387.07], [735.77, 387.83], [732.58, 388.72], [729.43, 389.72], [726.32, 390.83], [723.25, 392.05], [720.23, 393.39], [717.26, 394.83], [714.34, 396.38], [711.47, 398.03], [708.67, 399.78], [705.94, 401.64], [703.27, 403.59], [700.68, 405.63], [698.16, 407.77], [695.72, 410.0], [693.36, 412.31], [691.09, 414.71], [688.9, 417.19], [686.81, 419.75], [684.81, 422.38], [682.9, 425.08], [681.1, 427.84], [679.39, 430.67], [677.79, 433.56], [676.29, 436.51], [674.9, 439.51], [673.62, 442.55], [672.45, 445.64], [671.4, 448.77], [670.45, 451.94], [669.62, 455.14], [668.91, 458.36], [668.31, 461.61], [667.83, 464.88], [667.47, 468.17], [667.23, 471.46], [667.11, 474.76], [667.11, 478.07], [667.22, 481.37], [667.45, 484.66], [667.81, 487.95], [668.28, 491.22], [668.87, 494.47], [669.57, 497.7], [670.39, 500.9], [671.33, 504.07], [672.38, 507.2], [673.54, 510.29], [674.82, 513.34], [676.2, 516.34], [677.69, 519.29], [679.23, 522.21], [680.87, 525.09], [682.61, 527.9], [684.45, 530.64], [686.38, 533.32], [688.41, 535.92], [690.54, 538.45], [692.75, 540.91], [695.05, 543.28], [697.41, 545.59], [699.79, 547.88], [702.23, 550.12], [704.68, 552.33], [707.14, 554.53], [709.62, 556.72], [712.1, 558.9], [714.66, 560.99], [717.28, 563.0], [719.97, 564.92], [722.68, 566.81], [725.42, 568.66], [728.18, 570.47], [730.96, 572.26], [733.76, 574.02], [736.58, 575.75], [739.41, 577.45], [742.26, 579.12], [745.11, 580.78], [747.97, 582.43], [750.84, 584.07], [753.73, 585.68], [756.61, 587.31], [759.48, 588.93], [762.35, 590.57], [765.23, 592.19], [768.12, 593.8], [771.02, 595.38], [773.93, 596.95], [776.82, 598.55], [779.71, 600.15], [782.61, 601.73], [785.48, 603.36], [788.37, 604.97], [791.23, 606.63], [794.07, 608.31], [796.92, 609.99], [799.77, 611.65], [802.6, 613.36], [805.37, 615.16], [808.13, 616.98], [810.85, 618.85], [813.5, 620.82], [816.09, 622.88], [818.66, 624.95], [821.16, 627.12], [823.6, 629.35], [825.98, 631.63], [828.32, 633.96], [830.62, 636.34], [832.85, 638.78], [835.03, 641.26], [837.19, 643.76], [839.31, 646.3], [841.41, 648.85], [843.52, 651.39], [845.63, 653.93], [847.79, 656.44], [849.96, 658.92], [852.18, 661.37], [854.44, 663.78], [856.74, 666.16], [859.07, 668.5], [861.41, 670.83], [863.79, 673.12], [866.21, 675.37], [868.64, 677.61], [871.13, 679.78], [873.65, 681.91], [876.22, 683.99], [878.84, 686.01], [881.48, 688.0], [884.15, 689.94], [886.85, 691.84], [889.58, 693.71], [892.33, 695.54], [895.09, 697.35], [897.85, 699.17], [900.6, 701.0], [903.35, 702.84], [906.07, 704.71], [908.77, 706.61], [911.44, 708.56], [914.06, 710.57], [916.61, 712.67], [919.1, 714.84], [921.52, 717.09], [923.89, 719.4], [926.23, 721.73], [928.58, 724.06], [930.9, 726.41], [933.27, 728.71], [935.66, 730.99], [938.07, 733.25], [940.5, 735.49], [942.98, 737.68], [945.48, 739.83], [947.99, 741.98], [950.5, 744.12], [953.06, 746.22], [955.58, 748.35], [958.11, 750.48], [960.61, 752.64], [963.12, 754.79], [965.67, 756.89], [968.22, 758.99], [970.83, 761.02], [973.42, 763.07], [975.98, 765.15], [978.57, 767.22], [981.14, 769.29], [983.67, 771.42], [986.17, 773.57], [988.63, 775.78], [991.05, 778.03], [993.43, 780.32], [995.78, 782.65], [998.06, 785.03], [1000.31, 787.46], [1002.51, 789.92], [1004.67, 792.42], [1006.84, 794.91], [1009.05, 797.37], [1011.25, 799.83], [1013.47, 802.28], [1015.72, 804.7], [1018.05, 807.04], [1020.41, 809.35], [1022.82, 811.62], [1025.28, 813.82], [1027.79, 815.97], [1030.35, 818.06], [1032.93, 820.12], [1035.52, 822.17], [1038.09, 824.26], [1040.6, 826.41], [1043.05, 828.62], [1045.47, 830.86], [1047.9, 833.1], [1050.29, 835.38], [1052.63, 837.72], [1054.97, 840.05], [1057.33, 842.36], [1059.7, 844.67], [1062.07, 846.97], [1064.48, 849.23], [1066.93, 851.44], [1069.44, 853.6], [1071.99, 855.69], [1074.6, 857.73], [1077.24, 859.7], [1079.92, 861.65], [1082.59, 863.59], [1085.3, 865.48], [1088.05, 867.32], [1090.83, 869.1], [1093.63, 870.85], [1096.44, 872.58], [1099.29, 874.25], [1102.15, 875.91], [1105.0, 877.58], [1107.83, 879.29], [1110.6, 881.09], [1113.31, 882.99], [1115.94, 884.98], [1118.53, 887.03], [1121.04, 889.18], [1123.5, 891.39], [1125.9, 893.66], [1128.3, 895.93], [1130.67, 898.23], [1133.03, 900.55], [1135.41, 902.84], [1137.82, 905.1], [1140.28, 907.3], [1142.77, 909.47], [1145.33, 911.56], [1147.92, 913.61], [1150.55, 915.62], [1153.21, 917.58], [1155.9, 919.5], [1158.56, 921.45], [1161.22, 923.42], [1163.81, 925.47], [1166.32, 927.61], [1168.78, 929.82], [1171.19, 932.08], [1173.59, 934.35], [1175.97, 936.64], [1178.34, 938.95], [1180.69, 941.27], [1183.04, 943.6], [1185.41, 945.89], [1187.81, 948.16], [1190.22, 950.43], [1192.66, 952.66], [1195.11, 954.87], [1197.63, 957.0], [1200.15, 959.14], [1202.71, 961.24], [1205.24, 963.36], [1207.74, 965.52], [1210.24, 967.68], [1212.75, 969.83], [1215.25, 971.99], [1217.72, 974.18], [1220.2, 976.36], [1222.75, 978.47], [1225.33, 980.53], [1227.96, 982.53], [1230.6, 984.51], [1233.26, 986.47], [1235.9, 988.47], [1238.5, 990.5], [1241.11, 992.53], [1243.69, 994.59], [1246.25, 996.68], [1248.76, 998.83], [1251.25, 1001.0], [1253.7, 1003.22], [1256.18, 1005.41], [1258.67, 1007.57], [1261.2, 1009.71], [1263.7, 1011.87], [1266.14, 1014.09], [1268.51, 1016.39], [1270.83, 1018.74], [1273.16, 1021.09], [1275.48, 1023.44], [1277.78, 1025.81], [1280.03, 1028.23], [1282.25, 1030.68], [1284.45, 1033.14], [1286.64, 1035.61], [1288.82, 1038.1], [1291.01, 1040.57], [1293.18, 1043.07], [1295.27, 1045.62], [1297.31, 1048.22], [1299.33, 1050.84], [1301.34, 1053.46], [1303.36, 1056.08], [1305.43, 1058.65], [1307.55, 1061.19], [1309.66, 1063.73], [1311.76, 1066.28], [1313.81, 1068.87], [1315.87, 1071.45], [1317.95, 1074.02], [1320.07, 1076.55], [1322.21, 1079.08], [1324.35, 1081.59], [1326.47, 1084.13], [1328.6, 1086.65], [1330.77, 1089.14], [1332.96, 1091.62], [1335.13, 1094.1], [1337.34, 1096.56], [1339.58, 1098.99], [1341.86, 1101.38], [1344.16, 1103.76], [1346.49, 1106.1], [1348.82, 1108.44], [1351.2, 1110.74], [1353.59, 1113.01], [1356.02, 1115.25], [1358.48, 1117.45], [1360.98, 1119.62], [1363.46, 1121.81], [1365.88, 1124.05], [1368.22, 1126.38], [1370.51, 1128.77], [1372.71, 1131.23], [1374.88, 1133.72], [1376.99, 1136.26], [1379.07, 1138.83], [1381.14, 1141.41], [1383.19, 1144.0], [1385.21, 1146.62], [1387.16, 1149.28], [1389.1, 1151.95], [1391.01, 1154.66], [1392.85, 1157.4], [1394.62, 1160.19], [1396.32, 1163.02], [1397.96, 1165.89], [1399.6, 1168.76], [1401.24, 1171.62], [1402.93, 1174.47], [1404.62, 1177.31], [1406.33, 1180.13], [1408.02, 1182.97], [1409.72, 1185.81], [1411.43, 1188.63], [1413.17, 1191.44], [1414.97, 1194.21], [1416.79, 1196.97], [1418.51, 1199.79], [1420.13, 1202.67], [1421.65, 1205.6], [1423.06, 1208.59], [1424.35, 1211.63], [1425.54, 1214.71], [1426.62, 1217.84], [1427.58, 1221.0], [1428.43, 1224.19], [1429.16, 1227.41], [1429.78, 1230.66], [1430.28, 1233.93], [1430.66, 1237.21], [1430.92, 1240.5], [1431.07, 1243.8], [1431.09, 1247.11], [1430.99, 1250.41], [1430.78, 1253.71], [1430.45, 1256.99], [1430.0, 1260.27], [1429.43, 1263.52], [1428.74, 1266.75], [1427.94, 1269.96], [1427.02, 1273.13], [1425.99, 1276.27], [1424.85, 1279.37], [1423.59, 1282.43], [1422.23, 1285.44], [1420.75, 1288.4], [1419.18, 1291.3], [1417.5, 1294.14], [1415.71, 1296.93], [1413.83, 1299.64], [1411.85, 1302.29], [1409.78, 1304.86], [1407.62, 1307.36], [1405.36, 1309.77], [1403.02, 1312.11], [1400.6, 1314.36], [1398.1, 1316.52], [1395.53, 1318.58], [1392.88, 1320.56], [1390.16, 1322.44], [1387.37, 1324.21], [1384.52, 1325.89], [1381.62, 1327.46], [1378.66, 1328.93], [1375.65, 1330.29], [1372.59, 1331.54], [1369.48, 1332.67], [1366.34, 1333.7], [1363.17, 1334.61], [1359.96, 1335.41], [1356.73, 1336.09], [1353.47, 1336.66], [1350.21, 1337.14], [1346.92, 1337.5], [1343.63, 1337.75], [1340.32, 1337.87], [1337.02, 1337.93], [1333.72, 1337.87], [1330.42, 1337.72], [1327.12, 1337.46], [1323.83, 1337.22], [1320.53, 1336.98], [1317.24, 1336.74], [1313.94, 1336.51], [1310.65, 1336.26], [1307.36, 1335.94], [1304.07, 1335.64], [1300.78, 1335.34], [1297.5, 1334.95], [1294.22, 1334.5], [1290.95, 1334.03], [1287.69, 1333.53], [1284.42, 1333.02], [1281.17, 1332.44], [1277.91, 1331.87], [1274.66, 1331.31], [1271.4, 1330.74], [1268.16, 1330.12], [1264.93, 1329.39], [1261.73, 1328.58], [1258.54, 1327.74], [1255.33, 1326.96], [1252.11, 1326.19], [1248.89, 1325.45], [1245.67, 1324.7], [1242.47, 1323.88], [1239.3, 1322.95], [1236.16, 1321.93], [1233.03, 1320.88], [1229.9, 1319.82], [1226.76, 1318.79], [1223.62, 1317.77], [1220.49, 1316.7], [1217.35, 1315.66], [1214.21, 1314.64], [1211.07, 1313.61], [1207.93, 1312.58], [1204.8, 1311.53], [1201.66, 1310.48], [1198.57, 1309.33], [1195.5, 1308.11], [1192.48, 1306.77], [1189.49, 1305.37], [1186.54, 1303.87], [1183.59, 1302.38], [1180.63, 1300.92], [1177.63, 1299.53], [1174.6, 1298.22], [1171.55, 1296.95], [1168.48, 1295.72], [1165.4, 1294.53], [1162.29, 1293.4], [1159.17, 1292.31], [1156.04, 1291.25], [1152.9, 1290.25], [1149.73, 1289.31], [1146.55, 1288.4], [1143.37, 1287.51], [1140.18, 1286.64], [1136.99, 1285.79], [1133.8, 1284.92], [1130.62, 1284.01], [1127.47, 1283.04], [1124.35, 1281.95], [1121.26, 1280.78], [1118.21, 1279.5], [1115.18, 1278.18], [1112.19, 1276.78], [1109.21, 1275.34], [1106.27, 1273.85], [1103.33, 1272.34], [1100.4, 1270.82], [1097.49, 1269.25], [1094.61, 1267.63], [1091.76, 1265.95], [1088.91, 1264.28], [1086.05, 1262.63], [1083.22, 1260.92], [1080.42, 1259.17], [1077.64, 1257.38], [1074.88, 1255.57], [1072.13, 1253.73], [1069.41, 1251.86], [1066.69, 1249.99], [1063.99, 1248.08], [1061.32, 1246.13], [1058.67, 1244.16], [1055.99, 1242.23], [1053.3, 1240.31], [1050.59, 1238.42], [1047.9, 1236.5], [1045.23, 1234.56], [1042.57, 1232.6], [1039.9, 1230.64], [1037.24, 1228.69], [1034.55, 1226.76], [1031.86, 1224.85], [1029.15, 1222.95], [1026.4, 1221.13], [1023.63, 1219.32], [1020.88, 1217.5], [1018.13, 1215.67], [1015.38, 1213.83], [1012.66, 1211.96], [1009.94, 1210.08], [1007.23, 1208.19], [1004.5, 1206.32], [1001.78, 1204.45], [999.06, 1202.57], [996.35, 1200.68], [993.64, 1198.79], [990.92, 1196.92], [988.19, 1195.06], [985.46, 1193.19], [982.74, 1191.33], [980.02, 1189.45], [977.32, 1187.54], [974.62, 1185.63], [971.91, 1183.74], [969.2, 1181.86], [966.49, 1179.97], [963.77, 1178.09], [961.05, 1176.22], [958.3, 1174.38], [955.55, 1172.55], [952.81, 1170.71], [950.05, 1168.88], [947.3, 1167.05], [944.56, 1165.21], [941.81, 1163.38], [939.05, 1161.56], [936.3, 1159.73], [933.53, 1157.94], [930.75, 1156.15], [927.98, 1154.34], [925.22, 1152.53], [922.47, 1150.7], [919.69, 1148.92], [916.89, 1147.15], [914.06, 1145.46], [911.2, 1143.79], [908.33, 1142.15], [905.44, 1140.56], [902.49, 1139.07], [899.55, 1137.56], [896.57, 1136.13], [893.57, 1134.76], [890.57, 1133.36], [887.58, 1131.96], [884.6, 1130.53], [881.63, 1129.09], [878.65, 1127.65], [875.7, 1126.18], [872.75, 1124.68], [869.78, 1123.24], [866.76, 1121.89], [863.72, 1120.6], [860.67, 1119.33], [857.6, 1118.11], [854.51, 1116.93], [851.39, 1115.85], [848.24, 1114.85], [845.07, 1113.91], [841.9, 1113.0], [838.72, 1112.07], [835.54, 1111.21], [832.35, 1110.35], [829.14, 1109.53], [825.91, 1108.84], [822.66, 1108.26], [819.39, 1107.79], [816.1, 1107.45], [812.81, 1107.22], [809.5, 1107.12], [806.2, 1107.13], [802.9, 1107.26], [799.6, 1107.51], [796.32, 1107.88], [793.03, 1108.17], [789.73, 1108.39], [786.43, 1108.56], [783.13, 1108.67], [779.83, 1108.71], [776.52, 1108.74], [773.22, 1108.71], [769.92, 1108.62], [766.61, 1108.53], [763.31, 1108.46], [760.01, 1108.41], [756.7, 1108.36], [753.4, 1108.33], [750.09, 1108.34], [746.79, 1108.38], [743.49, 1108.47], [740.18, 1108.55], [736.88, 1108.65], [733.58, 1108.76], [730.28, 1108.9], [726.98, 1109.09], [723.68, 1109.26], [720.38, 1109.45], [717.08, 1109.64], [713.78, 1109.82], [710.48, 1109.94], [707.18, 1110.06], [703.88, 1110.2], [700.58, 1110.41], [697.29, 1110.72], [694.01, 1111.11], [690.73, 1111.53], [687.46, 1111.97], [684.18, 1112.39], [680.9, 1112.81], [677.62, 1113.15], [674.32, 1113.41], [671.02, 1113.59], [667.72, 1113.74], [664.42, 1113.85], [661.12, 1113.92], [657.81, 1113.93], [654.51, 1113.87], [651.21, 1113.76], [647.91, 1113.59], [644.61, 1113.37], [641.32, 1113.1], [638.02, 1112.87], [634.73, 1112.63], [631.44, 1112.35], [628.14, 1112.06], [624.85, 1111.75], [621.57, 1111.44], [618.27, 1111.14], [614.98, 1110.92], [611.68, 1110.78], [608.37, 1110.77], [605.07, 1110.84], [601.77, 1110.97], [598.47, 1111.11], [595.17, 1111.29], [591.87, 1111.55], [588.58, 1111.79], [585.28, 1112.05], [582.0, 1112.43], [578.73, 1112.85], [575.45, 1113.31], [572.19, 1113.83], [568.93, 1114.39], [565.69, 1115.0], [562.44, 1115.63], [559.2, 1116.24], [555.94, 1116.83], [552.68, 1117.34], [549.41, 1117.83], [546.15, 1118.34], [542.87, 1118.78], [539.59, 1119.19], [536.31, 1119.57], [533.03, 1119.91], [529.74, 1120.24], [526.45, 1120.6], [523.19, 1121.09], [519.94, 1121.69], [516.71, 1122.4], [513.51, 1123.24], [510.35, 1124.18], [507.22, 1125.25], [504.13, 1126.42], [501.09, 1127.7], [498.09, 1129.1], [495.15, 1130.6], [492.26, 1132.2], [489.43, 1133.91], [486.67, 1135.72], [483.97, 1137.63], [481.34, 1139.63], [478.79, 1141.73], [476.31, 1143.92], [473.92, 1146.19], [471.61, 1148.56], [469.38, 1151.0], [467.25, 1153.52], [465.2, 1156.12], [463.26, 1158.79], [461.41, 1161.52], [459.65, 1164.33], [458.0, 1167.18], [456.44, 1170.1], [455.0, 1173.07], [453.66, 1176.09], [452.42, 1179.15], [451.29, 1182.26], [450.27, 1185.4], [449.33, 1188.57], [448.47, 1191.76], [447.68, 1194.97], [446.94, 1198.19], [446.26, 1201.42], [445.62, 1204.66], [445.03, 1207.91], [444.5, 1211.17], [443.94, 1214.43], [443.39, 1217.69], [442.85, 1220.95], [442.32, 1224.21], [441.84, 1227.48], [441.29, 1230.74], [440.72, 1233.99], [440.11, 1237.24], [439.44, 1240.47], [438.7, 1243.7], [437.92, 1246.91], [437.1, 1250.11], [436.24, 1253.3], [435.31, 1256.47], [434.31, 1259.62], [433.21, 1262.73], [432.08, 1265.84], [430.98, 1268.95], [429.86, 1272.06], [428.7, 1275.15], [427.56, 1278.25], [426.35, 1281.33], [425.09, 1284.38], [423.78, 1287.42], [422.48, 1290.46], [421.19, 1293.5], [419.86, 1296.52], [418.58, 1299.57], [417.26, 1302.6], [415.9, 1305.61], [414.52, 1308.61], [413.1, 1311.59], [411.66, 1314.56], [410.17, 1317.52], [408.67, 1320.46], [407.18, 1323.41], [405.68, 1326.35], [404.15, 1329.28], [402.65, 1332.22], [401.19, 1335.19], [399.84, 1338.21], [398.54, 1341.24], [397.34, 1344.32], [396.18, 1347.41], [395.06, 1350.52], [393.96, 1353.64], [392.88, 1356.76], [391.82, 1359.89], [390.79, 1363.03], [389.74, 1366.17], [388.72, 1369.31], [387.66, 1372.44], [386.71, 1375.6], [385.77, 1378.77], [384.79, 1381.92], [383.76, 1385.06], [382.72, 1388.2], [381.67, 1391.33], [380.56, 1394.45], [379.5, 1397.57], [378.52, 1400.73], [377.58, 1403.9], [376.71, 1407.08], [375.9, 1410.29], [375.14, 1413.5], [374.48, 1416.74], [373.89, 1419.99], [373.34, 1423.25], [372.87, 1426.52], [372.4, 1429.79], [371.93, 1433.06], [371.39, 1436.32], [370.87, 1439.58], [370.3, 1442.84], [369.66, 1446.08], [368.95, 1449.31], [368.13, 1452.51], [367.26, 1455.7], [366.49, 1458.91], [365.77, 1462.13], [365.12, 1465.37], [364.54, 1468.62], [364.05, 1471.89], [363.63, 1475.17], [363.32, 1478.46], [363.12, 1481.76], [363.03, 1485.06], [363.05, 1488.36], [363.16, 1491.67], [363.4, 1494.96], [363.75, 1498.25], [364.2, 1501.52], [364.76, 1504.78], [365.44, 1508.01], [366.23, 1511.22], [367.1, 1514.41], [368.07, 1517.56], [369.12, 1520.7], [370.23, 1523.81], [371.39, 1526.9], [372.56, 1529.99], [373.76, 1533.07], [374.94, 1536.16], [376.06, 1539.27], [377.13, 1542.39], [378.15, 1545.54], [379.08, 1548.71], [379.98, 1551.89], [380.84, 1555.07], [381.69, 1558.27], [382.55, 1561.46], [383.47, 1564.63], [384.41, 1567.8], [385.44, 1570.94], [386.51, 1574.06], [387.61, 1577.18], [388.73, 1580.29], [389.89, 1583.38], [391.06, 1586.47], [392.28, 1589.55], [393.56, 1592.59], [394.9, 1595.61], [396.2, 1598.65], [397.57, 1601.65], [398.95, 1604.66], [400.27, 1607.68], [401.56, 1610.73], [402.86, 1613.76], [404.16, 1616.8], [405.47, 1619.84], [406.73, 1622.89], [408.02, 1625.93], [409.29, 1628.98], [410.54, 1632.04], [411.79, 1635.1], [413.02, 1638.17], [414.23, 1641.24], [415.45, 1644.31], [416.68, 1647.38], [417.91, 1650.45], [419.03, 1653.55], [420.07, 1656.69], [421.02, 1659.85], [421.96, 1663.02], [422.89, 1666.19], [423.72, 1669.39], [424.43, 1672.62], [425.02, 1675.87], [425.49, 1679.14], [425.85, 1682.42], [426.08, 1685.72], [426.2, 1689.02], [426.2, 1692.32], [426.08, 1695.63], [425.84, 1698.92], [425.48, 1702.21], [425.0, 1705.48], [424.41, 1708.73], [423.7, 1711.95], [422.87, 1715.15], [421.93, 1718.32], [420.87, 1721.45], [419.7, 1724.54], [418.43, 1727.59], [417.04, 1730.59], [415.54, 1733.53], [413.94, 1736.42], [412.24, 1739.25], [410.44, 1742.02], [408.53, 1744.72], [406.53, 1747.35], [404.44, 1749.91], [402.26, 1752.39], [399.99, 1754.79], [397.63, 1757.11], [395.19, 1759.34], [392.67, 1761.48], [390.08, 1763.52], [387.42, 1765.48], [384.68, 1767.33], [381.88, 1769.09], [379.02, 1770.74], [376.1, 1772.29], [373.13, 1773.74], [370.11, 1775.07], [367.04, 1776.3], [363.94, 1777.43], [360.81, 1778.49], [357.65, 1779.46], [354.48, 1780.39], [351.29, 1781.26], [348.09, 1782.09], [344.88, 1782.85], [341.64, 1783.5], [338.38, 1784.07], [335.11, 1784.54], [331.84, 1784.96], [328.55, 1785.31], [325.26, 1785.57], [321.96, 1785.8], [318.66, 1785.96], [315.36, 1786.05], [312.05, 1786.08], [308.75, 1786.08], [305.45, 1786.04], [302.14, 1785.99], [298.84, 1785.95], [295.54, 1785.86], [292.23, 1785.74], [288.93, 1785.64], [285.63, 1785.58], [282.32, 1785.57], [279.02, 1785.59], [275.72, 1785.68], [272.42, 1785.81], [269.12, 1785.98], [265.82, 1786.18], [262.52, 1786.34], [259.22, 1786.48], [255.91, 1786.57], [252.61, 1786.62], [249.31, 1786.62], [246.0, 1786.6], [242.7, 1786.54], [239.4, 1786.43], [236.09, 1786.37], [232.79, 1786.4], [229.48, 1786.46], [226.18, 1786.52], [222.88, 1786.57], [219.58, 1786.71], [216.27, 1786.8], [212.97, 1786.88], [209.67, 1786.96], [206.36, 1787.05], [203.06, 1787.13], [199.76, 1787.22], [196.45, 1787.3], [193.15, 1787.38], [189.85, 1787.47], [186.55, 1787.55], [183.24, 1787.64], [179.94, 1787.72], [176.64, 1787.81], [173.33, 1787.89], [170.03, 1787.97], [166.73, 1788.06], [163.42, 1788.14], [160.12, 1788.23], [156.82, 1788.31], [153.52, 1788.39], [150.21, 1788.48], [146.91, 1788.56], [143.61, 1788.65], [140.3, 1788.73], [137.0, 1788.82], [133.7, 1788.9], [130.39, 1788.98], [127.09, 1789.07], [123.79, 1789.15], [120.49, 1789.24], [117.18, 1789.32], [113.88, 1789.4], [110.58, 1789.49], [107.27, 1789.57], [103.97, 1789.66], [100.67, 1789.74], [97.36, 1789.83], [94.06, 1789.91], [90.76, 1789.99], [87.45, 1790.08], [84.15, 1790.16], [80.85, 1790.25], [77.55, 1790.33], [74.24, 1790.41], [70.94, 1790.5], [67.64, 1790.58]], [[1846.0, 1045.65], [1842.25, 1046.26], [1838.51, 1046.87], [1834.76, 1047.48], [1831.02, 1048.09], [1827.27, 1048.7], [1823.53, 1049.31], [1819.78, 1049.92], [1816.04, 1050.54], [1812.29, 1051.15], [1808.55, 1051.76], [1804.8, 1052.37], [1801.06, 1052.98], [1797.31, 1053.59], [1793.57, 1054.2], [1789.82, 1054.81], [1786.08, 1055.43], [1782.33, 1056.04], [1778.59, 1056.65], [1774.84, 1057.26], [1771.1, 1057.87], [1767.35, 1058.48], [1763.61, 1059.09], [1759.86, 1059.71], [1756.12, 1060.32], [1752.37, 1060.93], [1748.63, 1061.54], [1744.88, 1062.15], [1741.14, 1062.76], [1737.39, 1063.37], [1733.65, 1063.98], [1729.9, 1064.6], [1726.16, 1065.21], [1722.41, 1065.82], [1718.67, 1066.43], [1714.92, 1067.04], [1711.18, 1067.65], [1707.43, 1068.26], [1703.69, 1068.87], [1699.94, 1069.49], [1696.23, 1070.26], [1692.55, 1071.2], [1688.92, 1072.3], [1685.33, 1073.54], [1681.79, 1074.91], [1678.32, 1076.43], [1674.91, 1078.11], [1671.58, 1079.93], [1668.3, 1081.84], [1665.11, 1083.9], [1661.98, 1086.04], [1658.89, 1088.24], [1655.85, 1090.51], [1652.89, 1092.88], [1649.99, 1095.33], [1647.17, 1097.86], [1644.45, 1100.51], [1641.82, 1103.25], [1639.28, 1106.07], [1636.86, 1108.99], [1634.57, 1112.02], [1632.42, 1115.15], [1630.31, 1118.3], [1628.21, 1121.46], [1626.08, 1124.6], [1623.95, 1127.74], [1621.8, 1130.87], [1619.64, 1133.98], [1617.45, 1137.09], [1615.22, 1140.16], [1612.97, 1143.22], [1610.71, 1146.26], [1608.41, 1149.28], [1606.07, 1152.27], [1603.77, 1155.29], [1601.6, 1158.4], [1599.57, 1161.61], [1597.64, 1164.87], [1595.8, 1168.19], [1594.02, 1171.54], [1592.31, 1174.93], [1590.59, 1178.31], [1588.84, 1181.68], [1587.11, 1185.05], [1585.42, 1188.46], [1583.77, 1191.87], [1582.08, 1195.27], [1580.35, 1198.64], [1578.54, 1201.98], [1576.67, 1205.28], [1574.72, 1208.54], [1572.82, 1211.82], [1570.98, 1215.14], [1569.2, 1218.49], [1567.48, 1221.87], [1565.76, 1225.26], [1564.06, 1228.65], [1562.27, 1232.0], [1560.48, 1235.34], [1558.62, 1238.65], [1556.72, 1241.94], [1554.77, 1245.19], [1552.76, 1248.41], [1550.74, 1251.62], [1548.67, 1254.8], [1546.63, 1258.0], [1544.63, 1261.23], [1542.7, 1264.5], [1540.85, 1267.8], [1539.04, 1271.14], [1537.24, 1274.48], [1535.44, 1277.82], [1533.67, 1281.18], [1531.88, 1284.53], [1529.96, 1287.8], [1527.88, 1290.98], [1525.68, 1294.06], [1523.33, 1297.05], [1520.86, 1299.93], [1518.27, 1302.7], [1515.55, 1305.35], [1512.73, 1307.88], [1509.79, 1310.28], [1506.75, 1312.55], [1503.61, 1314.69], [1500.39, 1316.69], [1497.08, 1318.54], [1493.69, 1320.24], [1490.22, 1321.8], [1486.7, 1323.2], [1483.11, 1324.44], [1479.48, 1325.53], [1475.8, 1326.46], [1472.08, 1327.22], [1468.33, 1327.82], [1464.56, 1328.25], [1460.78, 1328.52], [1456.98, 1328.62], [1453.19, 1328.55], [1449.4, 1328.32], [1445.63, 1327.92], [1441.88, 1327.35], [1438.15, 1326.63], [1434.46, 1325.73], [1430.82, 1324.68], [1427.22, 1323.47], [1423.68, 1322.1], [1420.21, 1320.58], [1416.8, 1318.9], [1413.47, 1317.08], [1410.23, 1315.11], [1407.07, 1313.0], [1404.01, 1310.76], [1401.06, 1308.38], [1398.21, 1305.88], [1395.47, 1303.25], [1392.85, 1300.5], [1390.35, 1297.65], [1387.98, 1294.68], [1385.75, 1291.62], [1383.65, 1288.46], [1381.69, 1285.2], [1379.88, 1281.87], [1378.21, 1278.46], [1376.7, 1274.98], [1375.34, 1271.44], [1374.14, 1267.84], [1373.09, 1264.19], [1372.21, 1260.5], [1371.49, 1256.77], [1370.94, 1253.02], [1370.55, 1249.25], [1370.33, 1245.46], [1370.27, 1241.66], [1370.38, 1237.87], [1370.66, 1234.09], [1371.07, 1230.31], [1371.59, 1226.55], [1372.16, 1222.8], [1372.77, 1219.06], [1373.42, 1215.32], [1374.1, 1211.59], [1374.81, 1207.86], [1375.53, 1204.13], [1376.29, 1200.41], [1377.1, 1196.71], [1377.97, 1193.02], [1378.96, 1189.35], [1380.09, 1185.73], [1381.37, 1182.16], [1382.72, 1178.61], [1384.18, 1175.11], [1385.76, 1171.66], [1387.34, 1168.21], [1388.86, 1164.73], [1390.35, 1161.24], [1391.82, 1157.74], [1393.23, 1154.22], [1394.6, 1150.68], [1395.92, 1147.12], [1397.19, 1143.55], [1398.44, 1139.96], [1399.64, 1136.36], [1400.8, 1132.75], [1401.92, 1129.13], [1403.12, 1125.53], [1404.37, 1121.94], [1405.62, 1118.36], [1406.9, 1114.79], [1408.2, 1111.22], [1409.52, 1107.66], [1410.86, 1104.11], [1412.26, 1100.58], [1413.67, 1097.06], [1415.17, 1093.58], [1416.76, 1090.13], [1418.47, 1086.75], [1420.3, 1083.42], [1422.26, 1080.17], [1424.3, 1076.97], [1426.47, 1073.86], [1428.74, 1070.82], [1431.08, 1067.83], [1433.51, 1064.92], [1436.04, 1062.09], [1438.56, 1059.25], [1441.16, 1056.48], [1443.81, 1053.77], [1446.48, 1051.07], [1449.16, 1048.39], [1451.83, 1045.69], [1454.54, 1043.04], [1457.3, 1040.43], [1460.09, 1037.86], [1462.88, 1035.28], [1465.62, 1032.66], [1468.3, 1029.97], [1470.91, 1027.22], [1473.39, 1024.35], [1475.75, 1021.37], [1477.98, 1018.31], [1480.15, 1015.19], [1482.29, 1012.06], [1484.46, 1008.95], [1486.62, 1005.83], [1488.8, 1002.72], [1490.94, 999.59], [1493.04, 996.43], [1495.05, 993.21], [1496.99, 989.95], [1498.96, 986.7], [1500.93, 983.46], [1502.89, 980.21], [1504.84, 976.96], [1506.72, 973.66], [1508.5, 970.31], [1510.23, 966.93], [1511.92, 963.53], [1513.62, 960.14], [1515.28, 956.73], [1516.92, 953.31], [1518.63, 949.92], [1520.38, 946.55], [1522.19, 943.22], [1524.04, 939.9], [1525.93, 936.61], [1527.91, 933.38], [1530.02, 930.22], [1532.21, 927.12], [1534.5, 924.1], [1536.87, 921.13], [1539.3, 918.21], [1541.79, 915.36], [1544.35, 912.55], [1546.95, 909.79], [1549.63, 907.1], [1552.31, 904.42], [1554.98, 901.72], [1557.61, 898.99], [1560.36, 896.37], [1563.23, 893.88], [1566.2, 891.52], [1569.27, 889.29], [1572.43, 887.2], [1575.69, 885.25], [1579.03, 883.44], [1582.44, 881.78], [1585.92, 880.28], [1589.47, 878.93], [1593.07, 877.73], [1596.72, 876.7], [1600.41, 875.82], [1604.14, 875.11], [1607.9, 874.57], [1611.67, 874.19], [1615.46, 873.97], [1619.25, 873.92], [1623.05, 874.04], [1626.83, 874.33], [1630.6, 874.78], [1634.34, 875.4], [1638.06, 876.18], [1641.73, 877.12], [1645.36, 878.23], [1648.94, 879.49], [1652.46, 880.91], [1655.91, 882.48], [1659.3, 884.2], [1662.6, 886.07], [1665.81, 888.08], [1668.94, 890.24], [1671.97, 892.52], [1674.89, 894.94], [1677.71, 897.48], [1680.41, 900.15], [1682.99, 902.93], [1685.45, 905.82], [1687.77, 908.82], [1689.97, 911.92], [1692.02, 915.11], [1693.93, 918.39], [1695.7, 921.74], [1697.32, 925.18], [1698.78, 928.68], [1700.09, 932.24], [1701.25, 935.85], [1702.24, 939.52], [1703.07, 943.22], [1703.73, 946.95], [1704.24, 950.72], [1704.57, 954.5], [1704.74, 958.29], [1704.74, 962.08], [1704.58, 965.87], [1704.25, 969.65], [1703.75, 973.42], [1703.09, 977.15], [1702.27, 980.86], [1701.28, 984.52], [1700.14, 988.14], [1698.83, 991.7], [1697.38, 995.21], [1695.82, 998.66], [1694.12, 1002.06], [1692.35, 1005.41], [1690.42, 1008.69], [1688.36, 1011.87], [1686.16, 1014.96], [1683.83, 1017.96], [1681.48, 1020.93], [1679.08, 1023.87], [1676.64, 1026.78], [1674.14, 1029.64], [1671.62, 1032.47], [1669.04, 1035.26], [1666.41, 1037.99], [1663.72, 1040.67], [1660.99, 1043.31], [1658.19, 1045.87], [1655.37, 1048.41], [1652.54, 1050.93], [1649.61, 1053.35], [1646.61, 1055.67], [1643.55, 1057.9], [1640.42, 1060.05], [1637.24, 1062.12], [1634.04, 1064.16], [1630.83, 1066.18], [1627.59, 1068.16], [1624.34, 1070.12], [1621.09, 1072.08], [1617.85, 1074.06], [1614.6, 1076.02], [1611.35, 1077.98], [1608.1, 1079.93], [1604.83, 1081.86], [1601.53, 1083.73], [1598.14, 1085.45], [1594.69, 1087.02], [1591.17, 1088.43], [1587.59, 1089.7], [1583.96, 1090.8], [1580.28, 1091.74], [1576.57, 1092.52], [1572.83, 1093.14], [1569.06, 1093.59], [1565.27, 1093.87], [1561.48, 1093.99], [1557.69, 1093.94], [1553.9, 1093.72], [1550.12, 1093.34], [1546.37, 1092.8], [1542.64, 1092.08], [1538.95, 1091.21], [1535.3, 1090.17], [1531.7, 1088.98], [1528.15, 1087.62], [1524.67, 1086.12], [1521.26, 1084.46], [1517.92, 1082.65], [1514.66, 1080.7], [1511.5, 1078.6], [1508.43, 1076.37], [1505.46, 1074.01], [1502.6, 1071.52], [1499.85, 1068.9], [1497.22, 1066.17], [1494.71, 1063.33], [1492.32, 1060.37], [1490.07, 1057.32], [1487.96, 1054.16], [1485.99, 1050.92], [1484.16, 1047.6], [1482.48, 1044.2], [1480.95, 1040.72], [1479.57, 1037.19], [1478.35, 1033.59], [1477.29, 1029.95], [1476.39, 1026.27], [1475.64, 1022.54], [1475.07, 1018.79], [1474.65, 1015.02], [1474.4, 1011.24], [1474.22, 1007.45], [1474.03, 1003.65], [1473.84, 999.87], [1473.63, 996.08], [1473.4, 992.29], [1473.24, 988.5], [1473.23, 984.7], [1473.37, 980.91], [1473.64, 977.12], [1474.06, 973.35], [1474.61, 969.6], [1475.29, 965.87], [1476.1, 962.16], [1477.01, 958.48], [1477.96, 954.8], [1478.93, 951.13], [1479.97, 947.48], [1480.98, 943.83], [1482.0, 940.17], [1483.04, 936.52], [1484.02, 932.85], [1484.96, 929.18], [1485.8, 925.48], [1486.59, 921.77], [1487.31, 918.04], [1487.95, 914.3], [1488.56, 910.55], [1489.16, 906.81], [1489.77, 903.06], [1490.33, 899.31], [1490.85, 895.55], [1491.35, 891.79], [1491.84, 888.03], [1492.33, 884.26], [1492.8, 880.5], [1493.26, 876.73], [1493.74, 872.97], [1494.19, 869.2], [1494.59, 865.42], [1494.97, 861.65], [1495.29, 857.87], [1495.56, 854.08], [1495.77, 850.29], [1496.0, 846.51], [1496.25, 842.72], [1496.57, 838.94], [1496.9, 835.16], [1497.28, 831.38], [1497.68, 827.61], [1498.19, 823.85], [1498.79, 820.1], [1499.49, 816.37], [1500.24, 812.65], [1501.05, 808.95], [1501.91, 805.25], [1502.79, 801.56], [1503.65, 797.86], [1504.52, 794.17], [1505.4, 790.48], [1506.27, 786.78], [1507.05, 783.07], [1507.77, 779.35], [1508.46, 775.61], [1509.1, 771.87], [1509.82, 768.15], [1510.56, 764.42], [1511.34, 760.71], [1512.19, 757.01], [1513.1, 753.33], [1514.07, 749.66], [1515.17, 746.03], [1516.38, 742.43], [1517.71, 738.88], [1519.13, 735.36], [1520.61, 731.86], [1522.07, 728.36], [1523.52, 724.85], [1524.93, 721.33], [1526.31, 717.8], [1527.66, 714.25], [1529.03, 710.71], [1530.42, 707.18], [1531.82, 703.66], [1533.06, 700.07], [1534.15, 696.43], [1535.07, 692.75], [1535.83, 689.03], [1536.43, 685.29], [1536.86, 681.52], [1537.12, 677.73], [1537.22, 673.94], [1537.15, 670.14], [1536.92, 666.36], [1536.51, 662.58], [1535.95, 658.83], [1535.22, 655.11], [1534.32, 651.42], [1533.27, 647.77], [1532.05, 644.18], [1530.68, 640.64], [1529.15, 637.17], [1527.48, 633.76], [1525.65, 630.44], [1523.68, 627.19], [1521.57, 624.04], [1519.33, 620.98], [1516.95, 618.02], [1514.44, 615.17], [1511.81, 612.44], [1509.07, 609.82], [1506.21, 607.33], [1503.24, 604.96], [1500.17, 602.72], [1497.01, 600.63], [1493.76, 598.67], [1490.42, 596.86], [1487.01, 595.2], [1483.53, 593.69], [1479.99, 592.33], [1476.39, 591.13], [1472.74, 590.09], [1469.05, 589.21], [1465.32, 588.49], [1461.57, 587.94], [1457.79, 587.55], [1454.0, 587.33], [1450.21, 587.28], [1446.42, 587.39], [1442.63, 587.67], [1438.86, 588.12], [1435.12, 588.73], [1431.39, 589.44], [1427.69, 590.28], [1424.03, 591.27], [1420.41, 592.43], [1416.86, 593.75], [1413.34, 595.19], [1409.88, 596.74], [1406.46, 598.38], [1403.08, 600.11], [1399.76, 601.94], [1396.51, 603.91], [1393.35, 606.01], [1390.29, 608.25], [1387.33, 610.62], [1384.47, 613.12], [1381.73, 615.74], [1379.04, 618.42], [1376.44, 621.18], [1373.95, 624.05], [1371.57, 627.0], [1369.3, 630.04], [1367.15, 633.17], [1365.06, 636.33], [1363.02, 639.53], [1361.03, 642.76], [1359.11, 646.04], [1357.26, 649.35], [1355.48, 652.7], [1353.73, 656.07], [1352.01, 659.45], [1350.32, 662.85], [1348.66, 666.26], [1347.03, 669.69], [1345.48, 673.15], [1343.92, 676.61], [1342.37, 680.08], [1340.87, 683.56], [1339.39, 687.06], [1337.91, 690.55], [1336.35, 694.01], [1334.67, 697.41], [1332.87, 700.75], [1330.95, 704.02], [1328.91, 707.22], [1326.74, 710.34], [1324.5, 713.41], [1322.21, 716.42], [1319.89, 719.43], [1317.48, 722.36], [1315.02, 725.25], [1312.54, 728.12], [1310.04, 730.98], [1307.51, 733.8], [1304.94, 736.6], [1302.34, 739.36], [1299.7, 742.09], [1297.1, 744.85], [1294.53, 747.64], [1292.01, 750.48], [1289.55, 753.37], [1287.17, 756.32], [1284.86, 759.33], [1282.61, 762.39], [1280.41, 765.49], [1278.22, 768.58], [1276.04, 771.69], [1273.85, 774.79], [1271.66, 777.88], [1269.47, 780.98], [1267.25, 784.07], [1265.1, 787.19], [1262.98, 790.33], [1260.87, 793.49], [1258.87, 796.72], [1256.93, 799.98], [1255.1, 803.3], [1253.35, 806.67], [1251.67, 810.07], [1250.06, 813.51], [1248.5, 816.97], [1246.97, 820.44], [1245.46, 823.92], [1244.0, 827.42], [1242.58, 830.94], [1241.21, 834.48], [1239.85, 838.02], [1238.53, 841.58], [1237.23, 845.15], [1235.93, 848.71], [1234.58, 852.26], [1233.14, 855.77], [1231.61, 859.24], [1230.03, 862.69], [1228.41, 866.12], [1226.77, 869.54], [1225.04, 872.92], [1223.32, 876.3], [1221.56, 879.67], [1219.81, 883.03], [1218.12, 886.43], [1216.5, 889.86], [1214.93, 893.32], [1213.34, 896.76], [1211.8, 900.23], [1210.22, 903.68], [1208.56, 907.09], [1206.82, 910.46], [1205.02, 913.81], [1203.13, 917.09], [1201.19, 920.36], [1199.25, 923.62], [1197.27, 926.86], [1195.28, 930.08], [1193.23, 933.28], [1191.06, 936.39], [1188.8, 939.44], [1186.45, 942.42], [1184.03, 945.34], [1181.55, 948.22], [1179.01, 951.04], [1176.48, 953.86], [1173.9, 956.64], [1171.33, 959.44], [1168.75, 962.22], [1166.16, 965.0], [1163.51, 967.71], [1160.8, 970.36], [1158.07, 973.0], [1155.36, 975.65], [1152.74, 978.4], [1150.21, 981.23], [1147.81, 984.17], [1145.41, 987.11], [1143.04, 990.07], [1140.68, 993.05], [1138.33, 996.03], [1136.02, 999.03], [1133.69, 1002.03], [1131.39, 1005.05], [1129.15, 1008.11], [1126.95, 1011.2], [1124.75, 1014.3], [1122.55, 1017.39], [1120.39, 1020.51], [1118.27, 1023.65], [1116.23, 1026.86], [1114.27, 1030.1], [1112.41, 1033.41], [1110.58, 1036.73], [1108.78, 1040.07], [1106.97, 1043.41], [1105.1, 1046.71], [1103.17, 1049.98], [1101.25, 1053.25], [1099.28, 1056.5], [1097.28, 1059.72], [1095.34, 1062.98], [1093.46, 1066.28], [1091.64, 1069.61], [1089.86, 1072.96], [1088.12, 1076.33], [1086.43, 1079.73], [1084.78, 1083.15], [1083.08, 1086.54], [1081.4, 1089.94], [1079.75, 1093.36], [1078.08, 1096.77], [1076.43, 1100.18], [1074.78, 1103.6], [1073.13, 1107.02], [1071.48, 1110.44], [1069.87, 1113.87], [1068.28, 1117.32], [1066.69, 1120.76], [1065.16, 1124.23], [1063.52, 1127.66], [1061.76, 1131.02], [1059.85, 1134.3], [1057.83, 1137.51], [1055.7, 1140.65], [1053.49, 1143.74], [1051.15, 1146.72], [1048.76, 1149.67], [1046.3, 1152.56], [1043.81, 1155.42], [1041.35, 1158.31], [1038.94, 1161.24], [1036.51, 1164.16], [1034.08, 1167.07], [1031.63, 1169.97], [1029.19, 1172.87], [1026.7, 1175.74], [1024.23, 1178.62], [1021.79, 1181.52], [1019.39, 1184.47], [1017.04, 1187.44], [1014.71, 1190.44], [1012.39, 1193.44], [1010.09, 1196.46], [1007.86, 1199.53], [1005.7, 1202.65], [1003.63, 1205.83], [1001.66, 1209.07], [999.78, 1212.37], [997.95, 1215.7], [996.18, 1219.05], [994.43, 1222.42], [992.73, 1225.81], [991.07, 1229.22], [989.48, 1232.67], [987.89, 1236.11], [986.31, 1239.56], [984.72, 1243.01], [983.05, 1246.42], [981.25, 1249.76], [979.34, 1253.04], [977.35, 1256.27], [975.23, 1259.41], [973.03, 1262.51], [970.75, 1265.54], [968.44, 1268.55], [966.15, 1271.58], [963.9, 1274.63], [961.64, 1277.68], [959.43, 1280.77], [957.28, 1283.89], [955.16, 1287.04], [953.09, 1290.22], [951.09, 1293.44], [949.09, 1296.67], [947.13, 1299.92], [945.17, 1303.17], [943.21, 1306.42], [941.19, 1309.63], [939.16, 1312.84], [937.13, 1316.04], [935.13, 1319.27], [933.17, 1322.52], [931.25, 1325.79], [929.36, 1329.08], [927.53, 1332.41], [925.73, 1335.75], [923.95, 1339.1], [922.2, 1342.46], [920.41, 1345.81], [918.62, 1349.16], [916.86, 1352.52], [915.1, 1355.88], [913.3, 1359.22], [911.44, 1362.53], [909.51, 1365.8], [907.44, 1368.98], [905.23, 1372.06], [902.89, 1375.05], [900.42, 1377.93], [897.82, 1380.69], [895.11, 1383.34], [892.28, 1385.87], [889.34, 1388.28], [886.3, 1390.55], [883.16, 1392.68], [879.93, 1394.68], [876.62, 1396.53], [873.23, 1398.23], [869.77, 1399.79], [866.24, 1401.18], [862.66, 1402.43], [859.02, 1403.51], [855.34, 1404.44], [851.62, 1405.2], [847.87, 1405.79], [844.1, 1406.23], [840.32, 1406.49], [836.53, 1406.59], [832.73, 1406.52], [828.94, 1406.29], [825.17, 1405.89], [821.42, 1405.32], [817.69, 1404.59], [814.01, 1403.7], [810.36, 1402.64], [806.77, 1401.43], [803.23, 1400.06], [799.75, 1398.54], [796.35, 1396.86], [793.02, 1395.04], [789.78, 1393.07], [786.62, 1390.96], [783.56, 1388.71], [780.61, 1386.33], [777.76, 1383.83], [775.02, 1381.2], [772.4, 1378.45], [769.9, 1375.6], [767.54, 1372.63], [765.3, 1369.56], [763.2, 1366.4], [761.25, 1363.15], [759.42, 1359.83], [757.73, 1356.43], [756.18, 1352.96], [754.78, 1349.44], [753.5, 1345.87], [752.31, 1342.26], [751.21, 1338.63], [750.18, 1334.98], [749.27, 1331.29], [748.51, 1327.58], [747.84, 1323.84], [747.21, 1320.1], [746.69, 1316.34], [746.23, 1312.57], [745.83, 1308.8], [745.52, 1305.02], [745.32, 1301.23], [745.19, 1297.44], [745.13, 1293.64], [745.07, 1289.85], [745.03, 1286.05], [745.07, 1282.26], [745.17, 1278.46], [745.32, 1274.67], [745.51, 1270.88], [745.74, 1267.1], [745.99, 1263.31], [746.21, 1259.52], [746.45, 1255.73], [746.74, 1251.95], [747.02, 1248.17], [747.32, 1244.38], [747.58, 1240.6], [747.82, 1236.81], [748.08, 1233.02], [748.41, 1229.24], [748.71, 1225.46], [749.01, 1221.68], [749.4, 1217.9], [749.87, 1214.14], [750.49, 1210.39], [751.23, 1206.67], [752.1, 1202.98], [753.11, 1199.32], [754.26, 1195.7], [755.55, 1192.14], [756.89, 1188.59], [758.32, 1185.07], [759.71, 1181.54], [761.06, 1178.0], [762.45, 1174.46], [763.89, 1170.95], [765.4, 1167.47], [766.91, 1163.99], [768.47, 1160.53], [770.07, 1157.09], [771.79, 1153.7], [773.58, 1150.36], [775.44, 1147.06], [777.37, 1143.78], [779.29, 1140.51], [781.27, 1137.28], [783.3, 1134.07], [785.28, 1130.83], [787.2, 1127.56], [788.99, 1124.21], [790.63, 1120.79], [792.12, 1117.3], [793.5, 1113.77], [794.82, 1110.21], [796.04, 1106.62], [797.16, 1102.99], [798.29, 1099.37], [799.43, 1095.75], [800.54, 1092.12], [801.74, 1088.52], [803.03, 1084.95], [804.4, 1081.41], [805.77, 1077.87], [807.09, 1074.32], [808.36, 1070.74], [809.49, 1067.12], [810.51, 1063.46], [811.44, 1059.78], [812.32, 1056.09], [813.1, 1052.38], [813.78, 1048.65], [814.4, 1044.9], [815.01, 1041.16], [815.59, 1037.41], [816.14, 1033.65], [816.75, 1029.91], [817.38, 1026.17], [818.11, 1022.44], [819.0, 1018.75], [820.02, 1015.1], [821.19, 1011.49], [822.42, 1007.9], [823.73, 1004.34], [825.08, 1000.79], [826.51, 997.28], [827.98, 993.78], [829.42, 990.27], [830.96, 986.8], [832.56, 983.36], [834.14, 979.91], [835.75, 976.47], [837.33, 973.02], [838.86, 969.55], [840.34, 966.05], [841.8, 962.55], [843.28, 959.06], [844.77, 955.57], [846.24, 952.07], [847.65, 948.54], [849.03, 945.01], [850.4, 941.47], [851.75, 937.93], [853.08, 934.37], [854.35, 930.8], [855.57, 927.2], [856.76, 923.6], [857.87, 919.97], [858.92, 916.33], [859.93, 912.67], [860.86, 908.99], [861.75, 905.3], [862.56, 901.59], [863.35, 897.88], [864.14, 894.17], [864.97, 890.47], [865.79, 886.76], [866.63, 883.06], [867.47, 879.36], [868.25, 875.65], [869.05, 871.94], [869.83, 868.22], [870.58, 864.5], [871.27, 860.77], [871.89, 857.03], [872.51, 853.28], [873.14, 849.54], [873.75, 845.8], [874.33, 842.05], [874.88, 838.29], [875.38, 834.53], [875.85, 830.76], [876.26, 826.99], [876.66, 823.22], [877.03, 819.44], [877.38, 815.66], [877.7, 811.88], [878.03, 808.1], [878.52, 804.34], [879.17, 800.6], [879.86, 796.87], [880.58, 793.14], [881.36, 789.43], [882.27, 785.75], [883.35, 782.11], [884.55, 778.51], [885.89, 774.96], [887.34, 771.45], [888.9, 767.99], [890.56, 764.58], [892.23, 761.17], [893.94, 757.79], [895.65, 754.4], [897.32, 750.99], [899.06, 747.62], [900.89, 744.29], [902.79, 741.01], [904.77, 737.77], [906.8, 734.57], [908.85, 731.37], [910.91, 728.18], [912.94, 724.98], [914.95, 721.76], [916.89, 718.5], [918.79, 715.21], [920.73, 711.95], [922.76, 708.75], [924.82, 705.56], [926.91, 702.39], [929.01, 699.23], [931.14, 696.09], [933.29, 692.97], [935.41, 689.82], [937.46, 686.63], [939.44, 683.39], [941.33, 680.1], [943.15, 676.77], [944.94, 673.42], [946.7, 670.06], [948.34, 666.64], [949.84, 663.15], [951.26, 659.63], [952.56, 656.07], [953.76, 652.47], [954.89, 648.84], [955.9, 645.19], [956.82, 641.51], [957.67, 637.81], [958.5, 634.11], [959.31, 630.4], [960.07, 626.68], [960.77, 622.95], [961.47, 619.22], [962.18, 615.49], [962.86, 611.76], [963.5, 608.02], [964.1, 604.27], [964.71, 600.53], [965.32, 596.78], [965.94, 593.04], [966.53, 589.29], [967.11, 585.54], [967.68, 581.79], [968.28, 578.04], [968.88, 574.29], [969.51, 570.55], [970.22, 566.82], [970.91, 563.09], [971.59, 559.36], [972.26, 555.62], [972.94, 551.89], [973.57, 548.15], [974.2, 544.41], [974.84, 540.67], [975.58, 536.95], [976.47, 533.26], [977.19, 529.53], [977.74, 525.78], [978.13, 522.0], [978.36, 518.21], [978.42, 514.42], [978.31, 510.63], [978.03, 506.84], [977.59, 503.07], [976.98, 499.33], [976.21, 495.61], [975.28, 491.93], [974.19, 488.3], [972.93, 484.72], [971.52, 481.19], [969.96, 477.74], [968.25, 474.35], [966.39, 471.04], [964.39, 467.82], [962.24, 464.69], [959.96, 461.65], [957.55, 458.72], [955.02, 455.9], [952.36, 453.19], [949.59, 450.6], [946.7, 448.14], [943.71, 445.8], [940.62, 443.6], [937.43, 441.54], [934.16, 439.62], [930.81, 437.84], [927.38, 436.21], [923.88, 434.74], [920.33, 433.42], [916.71, 432.26], [913.05, 431.26], [909.35, 430.41], [905.62, 429.74], [901.86, 429.23], [898.08, 428.88], [894.29, 428.7], [890.5, 428.69], [886.7, 428.84], [882.92, 429.16], [879.16, 429.65], [875.42, 430.3], [871.71, 431.11], [868.05, 432.09], [864.43, 433.22], [860.86, 434.52], [857.35, 435.96], [853.9, 437.54], [850.48, 439.19], [847.1, 440.9], [843.74, 442.68], [840.42, 444.52], [837.13, 446.41], [833.89, 448.37], [830.69, 450.42], [827.54, 452.53], [824.42, 454.7], [821.39, 456.98], [818.44, 459.36], [815.57, 461.85], [812.72, 464.35], [809.9, 466.89], [807.08, 469.43], [804.21, 471.91], [801.34, 474.4], [798.47, 476.87], [795.63, 479.4], [792.82, 481.95], [790.03, 484.51], [787.26, 487.11], [784.46, 489.67], [781.65, 492.22], [778.79, 494.71], [775.87, 497.14], [772.86, 499.45], [769.82, 501.73], [766.78, 503.99], [763.75, 506.28], [760.72, 508.56], [757.64, 510.78], [754.51, 512.93], [751.36, 515.04], [748.2, 517.14], [745.05, 519.25], [741.88, 521.34], [738.67, 523.36], [735.47, 525.4], [732.32, 527.52], [729.24, 529.74], [726.21, 532.02], [723.25, 534.4], [720.31, 536.79], [717.36, 539.19], [714.37, 541.52], [711.33, 543.79], [708.26, 546.02], [705.16, 548.2], [702.04, 550.37], [698.9, 552.5], [695.73, 554.59], [692.61, 556.75], [689.54, 558.97], [686.49, 561.24], [683.5, 563.57], [680.56, 565.97], [677.67, 568.43], [674.83, 570.94], [672.01, 573.48], [669.16, 575.99], [666.31, 578.49], [663.48, 581.02], [660.64, 583.54], [657.81, 586.07], [655.02, 588.64], [652.32, 591.31], [649.73, 594.08], [647.22, 596.92], [644.77, 599.82], [642.42, 602.8], [640.13, 605.82], [637.91, 608.9], [635.78, 612.04], [633.68, 615.21], [631.64, 618.41], [629.65, 621.64], [627.7, 624.89], [625.73, 628.14], [623.73, 631.36], [621.76, 634.61], [619.79, 637.85], [617.82, 641.09], [615.86, 644.34], [613.88, 647.58], [611.95, 650.84], [610.02, 654.11], [608.09, 657.38], [606.16, 660.65], [604.22, 663.91], [602.2, 667.12], [600.14, 670.31], [598.1, 673.51], [596.11, 676.73], [594.16, 679.99], [592.2, 683.24], [590.16, 686.44], [588.13, 689.65], [586.14, 692.88], [584.11, 696.09], [582.14, 699.33], [580.19, 702.58], [578.27, 705.86], [576.36, 709.13], [574.39, 712.38], [572.4, 715.61], [570.43, 718.85], [568.47, 722.1], [566.53, 725.36], [564.55, 728.6], [562.53, 731.81], [560.43, 734.97], [558.3, 738.11], [556.1, 741.21], [553.87, 744.28], [551.61, 747.32], [549.31, 750.34], [546.96, 753.32], [544.48, 756.19], [541.94, 759.01], [539.35, 761.78], [536.7, 764.5], [534.05, 767.21], [531.43, 769.96], [528.83, 772.73], [526.27, 775.53], [523.75, 778.37], [521.24, 781.21], [518.76, 784.09], [516.27, 786.95], [513.82, 789.84], [511.4, 792.76], [509.01, 795.71], [506.61, 798.66], [504.25, 801.62], [501.85, 804.57], [499.44, 807.5], [497.0, 810.41], [494.53, 813.29], [492.05, 816.16], [489.53, 818.99], [487.0, 821.82], [484.45, 824.63], [481.92, 827.46], [479.4, 830.29], [476.91, 833.16], [474.33, 835.95], [471.66, 838.63], [468.93, 841.27], [466.17, 843.88], [463.39, 846.46], [460.55, 848.98], [457.64, 851.42], [454.69, 853.8], [451.72, 856.16], [448.77, 858.55], [445.8, 860.92], [442.8, 863.24], [439.82, 865.58], [436.84, 867.93], [433.87, 870.3], [430.93, 872.7], [428.04, 875.15], [425.18, 877.65], [422.34, 880.17], [419.54, 882.73], [416.78, 885.33], [414.09, 888.01], [411.44, 890.72], [408.77, 893.42], [406.08, 896.09], [403.27, 898.64], [400.35, 901.07], [397.32, 903.36], [394.2, 905.51], [390.99, 907.53], [387.69, 909.41], [384.31, 911.13], [380.86, 912.71], [377.34, 914.14], [373.76, 915.4], [370.13, 916.51], [366.46, 917.46], [362.75, 918.25], [359.0, 918.87], [355.24, 919.33], [351.45, 919.62], [347.66, 919.75], [343.87, 919.71], [340.08, 919.5], [336.3, 919.12], [332.55, 918.58], [328.82, 917.88], [325.12, 917.01], [321.47, 915.98], [317.87, 914.79], [314.32, 913.45], [310.83, 911.95], [307.42, 910.3], [304.08, 908.5], [300.82, 906.55], [297.65, 904.46], [294.58, 902.24], [291.6, 899.88], [288.74, 897.39], [285.98, 894.79], [283.34, 892.06], [280.83, 889.22], [278.44, 886.27], [276.18, 883.22], [274.06, 880.07], [272.08, 876.83], [270.25, 873.51], [268.56, 870.11], [267.02, 866.64], [265.64, 863.11], [264.42, 859.52], [263.35, 855.88], [262.44, 852.19], [261.7, 848.47], [261.12, 844.72], [260.71, 840.95], [260.45, 837.16], [260.3, 833.37], [260.27, 829.57], [260.42, 825.78], [260.72, 822.0], [261.13, 818.23], [261.68, 814.47], [262.39, 810.75], [263.23, 807.04], [264.22, 803.38], [265.38, 799.77], [266.66, 796.2], [268.01, 792.65], [269.32, 789.09], [270.52, 785.49], [271.57, 781.84], [272.57, 778.18], [273.53, 774.51], [274.39, 770.81], [275.25, 767.12], [276.15, 763.43], [277.1, 759.76], [278.12, 756.1], [279.19, 752.46], [280.32, 748.84], [281.55, 745.25], [282.89, 741.7], [284.32, 738.19], [285.83, 734.7], [287.42, 731.26], [289.08, 727.84], [290.81, 724.47], [292.61, 721.12], [294.47, 717.82], [296.39, 714.55], [298.32, 711.28], [300.23, 708.0], [302.11, 704.7], [303.85, 701.33], [305.47, 697.9], [306.97, 694.41], [308.34, 690.88], [309.65, 687.31], [310.87, 683.72], [312.01, 680.1], [313.09, 676.46], [314.1, 672.81], [315.07, 669.14], [315.98, 665.46], [316.88, 661.77], [317.68, 658.06], [318.43, 654.34], [319.17, 650.62], [319.87, 646.89], [320.63, 643.17], [321.44, 639.46], [322.37, 635.78], [323.36, 632.12], [324.37, 628.46], [325.43, 624.82], [326.51, 621.18], [327.64, 617.56], [328.83, 613.96], [329.93, 610.33], [330.94, 606.67], [331.87, 602.99], [332.76, 599.3], [333.6, 595.6], [334.38, 591.88], [335.11, 588.16], [335.83, 584.44], [336.55, 580.71], [337.23, 576.98], [337.89, 573.24], [338.53, 569.5], [339.17, 565.76], [339.82, 562.02], [340.63, 558.31], [341.6, 554.64], [342.69, 551.01], [343.83, 547.39], [344.94, 543.76], [345.98, 540.11], [346.95, 536.44], [347.89, 532.77], [348.8, 529.08], [349.67, 525.39], [350.53, 521.69], [351.36, 517.99], [352.22, 514.29], [353.04, 510.59], [353.82, 506.87], [354.57, 503.16], [355.34, 499.44], [356.12, 495.73], [356.95, 492.02], [357.88, 488.35], [358.92, 484.69], [360.06, 481.08], [361.32, 477.5], [362.64, 473.94], [364.01, 470.4], [365.38, 466.86], [366.83, 463.36], [368.34, 459.87], [369.94, 456.43], [371.57, 453.0], [373.26, 449.61], [374.98, 446.22], [376.72, 442.85], [378.43, 439.46], [380.05, 436.03], [381.6, 432.57], [383.07, 429.07], [384.54, 425.58], [385.86, 422.02], [387.02, 418.4], [388.02, 414.74], [388.85, 411.04], [389.52, 407.3], [390.03, 403.54], [390.37, 399.77], [390.55, 395.97], [390.56, 392.18], [390.4, 388.39], [390.08, 384.61], [389.59, 380.84], [388.93, 377.11], [388.11, 373.4], [387.13, 369.74], [385.99, 366.12], [384.69, 362.55], [383.24, 359.05], [381.63, 355.61], [379.87, 352.25], [377.97, 348.96], [375.93, 345.77], [373.74, 342.66], [371.42, 339.66], [368.98, 336.76], [366.4, 333.97], [363.71, 331.29], [360.9, 328.74], [357.99, 326.31], [354.97, 324.02], [351.85, 321.86], [348.64, 319.83], [345.34, 317.96], [341.96, 316.22], [338.51, 314.64], [335.0, 313.21], [331.42, 311.94], [327.8, 310.82], [324.12, 309.87], [320.41, 309.08], [316.67, 308.45], [312.9, 307.98], [309.12, 307.69], [305.34, 307.44], [301.54, 307.31], [297.75, 307.32], [293.96, 307.49], [290.18, 307.82], [286.41, 308.31], [282.67, 308.92], [278.92, 309.54], [275.19, 310.19], [271.44, 310.8], [267.7, 311.45], [263.96, 312.05], [260.21, 312.68], [256.48, 313.38], [252.76, 314.13], [249.04, 314.88], [245.31, 315.58], [241.57, 316.21], [237.82, 316.78], [234.06, 317.31], [230.29, 317.75], [226.52, 318.15], [222.75, 318.56], [218.97, 318.93], [215.19, 319.26], [211.41, 319.6], [207.63, 319.92], [203.86, 320.3], [200.08, 320.66], [196.3, 320.98], [192.51, 321.24], [188.73, 321.53], [184.95, 321.86], [181.18, 322.29], [177.41, 322.78], [173.65, 323.25], [169.88, 323.68], [166.11, 324.16], [162.34, 324.58], [158.57, 324.95], [154.79, 325.37], [151.03, 325.87], [147.28, 326.41], [143.53, 327.03], [139.81, 327.78], [136.11, 328.61], [132.4, 329.4], [128.71, 330.28], [125.02, 331.17], [121.34, 332.11], [117.67, 333.08], [114.0, 334.03], [110.33, 334.99], [106.64, 335.86], [102.92, 336.64], [99.21, 337.43], [95.5, 338.21], [91.78, 338.96], [88.05, 339.69], [84.33, 340.43], [80.61, 341.19], [76.9, 341.99], [73.21, 342.85]], [[1846.0, 238.79], [1842.34, 238.87], [1838.69, 238.94], [1835.03, 239.01], [1831.37, 239.08], [1827.72, 239.15], [1824.06, 239.22], [1820.4, 239.29], [1816.75, 239.36], [1813.09, 239.43], [1809.43, 239.5], [1805.77, 239.57], [1802.12, 239.65], [1798.46, 239.72], [1794.8, 239.79], [1791.15, 239.86], [1787.49, 239.93], [1783.83, 240.0], [1780.18, 240.07], [1776.52, 240.14], [1772.86, 240.21], [1769.21, 240.28], [1765.55, 240.35], [1761.89, 240.43], [1758.24, 240.5], [1754.58, 240.57], [1750.92, 240.64], [1747.26, 240.71], [1743.61, 240.78], [1739.95, 240.85], [1736.29, 240.92], [1732.64, 240.99], [1728.98, 241.06], [1725.32, 241.13], [1721.67, 241.21], [1718.01, 241.28], [1714.35, 241.35], [1710.7, 241.42], [1707.04, 241.49], [1703.38, 241.56], [1699.73, 241.63], [1696.07, 241.7], [1692.42, 241.92], [1688.78, 242.3], [1685.16, 242.82], [1681.56, 243.49], [1678.0, 244.31], [1674.47, 245.28], [1670.99, 246.39], [1667.55, 247.64], [1664.17, 249.04], [1660.84, 250.54], [1657.57, 252.19], [1654.35, 253.93], [1651.2, 255.79], [1648.1, 257.72], [1644.98, 259.63], [1641.89, 261.58], [1638.84, 263.61], [1635.87, 265.74], [1632.91, 267.89], [1629.97, 270.06], [1627.07, 272.29], [1624.18, 274.53], [1621.33, 276.83], [1618.48, 279.13], [1615.69, 281.48], [1612.96, 283.92], [1610.29, 286.42], [1607.66, 288.97], [1605.09, 291.56], [1602.55, 294.2], [1600.05, 296.87], [1597.61, 299.59], [1595.21, 302.35], [1592.81, 305.11], [1590.4, 307.86], [1588.0, 310.62], [1585.55, 313.33], [1583.07, 316.02], [1580.56, 318.69], [1577.97, 321.27], [1575.37, 323.85], [1572.78, 326.43], [1570.22, 329.03], [1567.68, 331.67], [1565.14, 334.3], [1562.55, 336.89], [1559.92, 339.42], [1557.23, 341.9], [1554.53, 344.37], [1551.8, 346.8], [1549.01, 349.18], [1546.2, 351.51], [1543.38, 353.84], [1540.53, 356.13], [1537.65, 358.39], [1534.75, 360.61], [1531.8, 362.78], [1528.78, 364.84], [1525.69, 366.8], [1522.55, 368.68], [1519.39, 370.52], [1516.21, 372.32], [1513.01, 374.1], [1509.75, 375.75], [1506.42, 377.26], [1503.04, 378.66], [1499.6, 379.92], [1496.15, 381.13], [1492.7, 382.35], [1489.24, 383.54], [1485.76, 384.64], [1482.25, 385.68], [1478.7, 386.57], [1475.13, 387.36], [1471.54, 388.07], [1467.93, 388.63], [1464.31, 389.12], [1460.67, 389.51], [1457.02, 389.83], [1453.37, 390.01], [1449.71, 390.06], [1446.06, 390.02], [1442.4, 389.88], [1438.75, 389.63], [1435.12, 389.25], [1431.48, 388.84], [1427.86, 388.34], [1424.24, 387.83], [1420.61, 387.32], [1416.99, 386.8], [1413.39, 386.19], [1409.78, 385.59], [1406.16, 385.05], [1402.54, 384.56], [1398.91, 384.08], [1395.28, 383.68], [1391.63, 383.34], [1387.99, 383.05], [1384.34, 382.83], [1380.69, 382.61], [1377.03, 382.41], [1373.38, 382.25], [1369.73, 382.05], [1366.08, 381.78], [1362.43, 381.52], [1358.78, 381.25], [1355.13, 381.02], [1351.48, 380.9], [1347.82, 380.83], [1344.17, 380.72], [1340.51, 380.62], [1336.85, 380.62], [1333.2, 380.73], [1329.54, 380.84], [1325.89, 381.03], [1322.24, 381.26], [1318.59, 381.52], [1314.94, 381.76], [1311.29, 381.98], [1307.64, 382.19], [1303.98, 382.29], [1300.32, 382.32], [1296.67, 382.25], [1293.01, 382.08], [1289.37, 381.78], [1285.73, 381.4], [1282.1, 380.96], [1278.48, 380.42], [1274.87, 379.84], [1271.24, 379.42], [1267.59, 379.15], [1263.93, 379.02], [1260.28, 379.05], [1256.62, 379.23], [1252.98, 379.56], [1249.35, 380.04], [1245.75, 380.67], [1242.18, 381.44], [1238.64, 382.37], [1235.13, 383.42], [1231.68, 384.61], [1228.26, 385.91], [1224.88, 387.31], [1221.54, 388.81], [1218.24, 390.39], [1215.02, 392.11], [1211.84, 393.91], [1208.67, 395.75], [1205.57, 397.69], [1202.5, 399.68], [1199.48, 401.74], [1196.52, 403.89], [1193.58, 406.06], [1190.71, 408.33], [1187.91, 410.68], [1185.2, 413.15], [1182.57, 415.68], [1179.97, 418.25], [1177.43, 420.89], [1174.95, 423.57], [1172.43, 426.23], [1169.95, 428.92], [1167.5, 431.63], [1165.04, 434.33], [1162.59, 437.05], [1160.2, 439.82], [1157.85, 442.63], [1155.53, 445.45], [1153.27, 448.33], [1151.05, 451.24], [1148.94, 454.22], [1146.95, 457.29], [1145.07, 460.43], [1143.27, 463.61], [1141.58, 466.85], [1139.99, 470.15], [1138.43, 473.46], [1136.9, 476.78], [1135.47, 480.15], [1134.08, 483.53], [1132.7, 486.92], [1131.28, 490.29], [1129.81, 493.64], [1128.28, 496.96], [1126.72, 500.27], [1125.15, 503.57], [1123.61, 506.89], [1122.08, 510.21], [1120.6, 513.56], [1119.14, 516.91], [1117.77, 520.3], [1116.48, 523.72], [1115.27, 527.17], [1114.08, 530.63], [1112.95, 534.11], [1111.75, 537.57], [1110.52, 541.01], [1109.27, 544.45], [1107.93, 547.85], [1106.51, 551.22], [1105.05, 554.58], [1103.59, 557.93], [1102.04, 561.24], [1100.43, 564.53], [1098.74, 567.77], [1096.94, 570.95], [1095.07, 574.1], [1093.13, 577.2], [1091.16, 580.28], [1089.21, 583.37], [1087.27, 586.47], [1085.35, 589.59], [1083.49, 592.73], [1081.69, 595.92], [1079.93, 599.13], [1078.23, 602.37], [1076.62, 605.65], [1075.09, 608.97], [1073.57, 612.3], [1072.15, 615.67], [1070.78, 619.06], [1069.49, 622.48], [1068.25, 625.92], [1067.08, 629.39], [1065.91, 632.86], [1064.71, 636.31], [1063.57, 639.78], [1062.43, 643.26], [1061.28, 646.73], [1060.15, 650.21], [1058.98, 653.68], [1057.74, 657.12], [1056.45, 660.54], [1055.1, 663.94], [1053.71, 667.32], [1052.29, 670.69], [1050.79, 674.03], [1049.21, 677.33], [1047.55, 680.59], [1045.81, 683.8], [1043.98, 686.97], [1042.14, 690.13], [1040.25, 693.26], [1038.4, 696.42], [1036.64, 699.63], [1034.96, 702.87], [1033.34, 706.16], [1031.84, 709.49], [1030.38, 712.84], [1029.0, 716.23], [1027.67, 719.64], [1026.4, 723.07], [1025.14, 726.5], [1023.93, 729.95], [1022.76, 733.42], [1021.65, 736.9], [1020.63, 740.42], [1019.58, 743.92], [1018.52, 747.42], [1017.45, 750.92], [1016.43, 754.43], [1015.44, 757.95], [1014.47, 761.48], [1013.43, 764.98], [1012.27, 768.45], [1011.08, 771.91], [1009.78, 775.33], [1008.49, 778.75], [1007.14, 782.15], [1005.69, 785.51], [1004.19, 788.84], [1002.6, 792.14], [1001.02, 795.44], [999.42, 798.73], [997.88, 802.04], [996.31, 805.35], [994.71, 808.64], [993.02, 811.88], [991.19, 815.05], [989.28, 818.17], [987.35, 821.27], [985.35, 824.34], [983.28, 827.35], [981.17, 830.34], [979.03, 833.31], [976.88, 836.26], [974.65, 839.16], [972.35, 842.01], [970.05, 844.85], [967.76, 847.71], [965.51, 850.59], [963.26, 853.47], [961.03, 856.37], [958.68, 859.17], [956.21, 861.87], [953.64, 864.47], [950.96, 866.96], [948.18, 869.34], [945.3, 871.6], [942.33, 873.73], [939.28, 875.75], [936.15, 877.63], [932.94, 879.39], [929.66, 881.01], [926.31, 882.49], [922.91, 883.84], [919.46, 885.04], [915.96, 886.1], [912.42, 887.01], [908.84, 887.78], [905.24, 888.4], [901.61, 888.87], [897.96, 889.19], [894.31, 889.35], [890.65, 889.37], [887.0, 889.23], [883.35, 888.95], [879.72, 888.51], [876.11, 887.93], [872.52, 887.2], [868.96, 886.39], [865.42, 885.46], [861.9, 884.48], [858.39, 883.45], [854.9, 882.34], [851.47, 881.08], [848.09, 879.69], [844.76, 878.17], [841.47, 876.57], [838.25, 874.84], [835.08, 873.01], [831.97, 871.09], [828.89, 869.12], [825.86, 867.07], [822.86, 864.97], [819.86, 862.88], [816.82, 860.85], [813.73, 858.89], [810.61, 856.97], [807.46, 855.13], [804.25, 853.36], [801.01, 851.66], [797.76, 850.0], [794.47, 848.4], [791.16, 846.83], [787.83, 845.32], [784.48, 843.86], [781.11, 842.44], [777.73, 841.04], [774.33, 839.69], [770.95, 838.29], [767.62, 836.78], [764.33, 835.19], [761.06, 833.54], [757.82, 831.85], [754.59, 830.13], [751.34, 828.45], [748.09, 826.77], [744.81, 825.16], [741.5, 823.6], [738.21, 822.01], [734.94, 820.36], [731.7, 818.66], [728.52, 816.85], [725.41, 814.93], [722.38, 812.89], [719.39, 810.77], [716.47, 808.57], [713.59, 806.32], [710.76, 804.0], [707.98, 801.63], [705.24, 799.2], [702.54, 796.74], [699.87, 794.23], [697.22, 791.71], [694.59, 789.17], [691.98, 786.61], [689.38, 784.04], [686.77, 781.47], [684.2, 778.87], [681.68, 776.22], [679.13, 773.6], [676.61, 770.95], [674.1, 768.28], [671.63, 765.59], [669.17, 762.88], [666.7, 760.18], [664.16, 757.55], [661.51, 755.03], [658.76, 752.62], [655.97, 750.26], [653.13, 747.95], [650.25, 745.7], [647.3, 743.54], [644.3, 741.44], [641.25, 739.42], [638.17, 737.46], [635.02, 735.59], [631.85, 733.78], [628.62, 732.06], [625.36, 730.4], [622.11, 728.72], [618.86, 727.04], [615.61, 725.36], [612.39, 723.63], [609.17, 721.89], [605.95, 720.16], [602.69, 718.49], [599.39, 716.91], [596.1, 715.33], [592.8, 713.74], [589.49, 712.2], [586.2, 710.6], [582.94, 708.93], [579.75, 707.15], [576.58, 705.31], [573.44, 703.45], [570.27, 701.62], [567.08, 699.82], [563.91, 698.01], [560.73, 696.19], [557.55, 694.4], [554.4, 692.53], [551.29, 690.61], [548.21, 688.63], [545.16, 686.61], [542.18, 684.5], [539.24, 682.32], [536.35, 680.08], [533.51, 677.77], [530.72, 675.41], [527.96, 673.01], [525.19, 670.61], [522.48, 668.17], [519.8, 665.68], [517.17, 663.13], [514.57, 660.56], [511.95, 658.01], [509.27, 655.52], [506.52, 653.11], [503.68, 650.8], [500.85, 648.48], [497.99, 646.21], [495.05, 644.03], [492.04, 641.95], [488.98, 639.95], [485.88, 638.0], [482.74, 636.13], [479.54, 634.35], [476.34, 632.6], [473.11, 630.88], [469.85, 629.22], [466.56, 627.62], [463.25, 626.06], [459.93, 624.52], [456.61, 623.0], [453.27, 621.5], [449.92, 620.04], [446.58, 618.55], [443.22, 617.1], [439.85, 615.67], [436.47, 614.28], [433.07, 612.92], [429.67, 611.57], [426.28, 610.2], [422.93, 608.73], [419.63, 607.16], [416.34, 605.57], [413.05, 603.97], [409.81, 602.27], [406.63, 600.46], [403.52, 598.53], [400.48, 596.51], [397.46, 594.43], [394.44, 592.38], [391.45, 590.27], [388.47, 588.14], [385.48, 586.04], [382.48, 583.95], [379.43, 581.92], [376.34, 579.98], [373.25, 578.02], [370.08, 576.19], [366.84, 574.5], [363.53, 572.94], [360.16, 571.52], [356.73, 570.24], [353.26, 569.1], [349.74, 568.1], [346.18, 567.26], [342.59, 566.56], [338.97, 566.01], [335.34, 565.61], [331.69, 565.36], [328.03, 565.26], [324.38, 565.31], [320.72, 565.51], [317.08, 565.87], [313.46, 566.37], [309.86, 567.02], [306.29, 567.83], [302.76, 568.77], [299.27, 569.87], [295.83, 571.11], [292.44, 572.48], [289.11, 574.0], [285.85, 575.65], [282.66, 577.44], [279.54, 579.36], [276.51, 581.4], [273.56, 583.57], [270.71, 585.86], [267.95, 588.26], [265.3, 590.78], [262.75, 593.4], [260.31, 596.13], [257.99, 598.95], [255.78, 601.87], [253.7, 604.88], [251.74, 607.97], [249.92, 611.13], [248.22, 614.38], [246.66, 617.68], [245.24, 621.05], [243.96, 624.48], [242.82, 627.96], [241.83, 631.48], [240.98, 635.04], [240.29, 638.63], [239.74, 642.24], [239.34, 645.88], [239.09, 649.53], [238.99, 653.18], [239.04, 656.84], [239.24, 660.49], [239.6, 664.13], [240.1, 667.75], [240.76, 671.35], [241.56, 674.92], [242.51, 678.45], [243.6, 681.94], [244.84, 685.39], [246.22, 688.77], [247.74, 692.1], [249.39, 695.36], [251.18, 698.55], [253.09, 701.67], [255.14, 704.7], [257.31, 707.65], [259.6, 710.5], [262.0, 713.26], [264.52, 715.91], [267.14, 718.46], [269.87, 720.9], [272.69, 723.22], [275.61, 725.43], [278.53, 727.63], [281.44, 729.84], [284.32, 732.1], [287.17, 734.39], [290.0, 736.71], [292.76, 739.11], [295.47, 741.57], [298.09, 744.11], [300.61, 746.77], [303.04, 749.5], [305.38, 752.32], [307.64, 755.19], [309.93, 758.04], [312.24, 760.88], [314.55, 763.71], [316.81, 766.59], [319.06, 769.47], [321.29, 772.37], [323.46, 775.32], [325.55, 778.32], [327.63, 781.33], [329.68, 784.36], [331.8, 787.33], [334.01, 790.25], [336.28, 793.11], [338.59, 795.95], [340.94, 798.76], [343.31, 801.54], [345.73, 804.28], [348.2, 806.99], [350.7, 809.65], [353.24, 812.28], [355.84, 814.86], [358.47, 817.4], [361.1, 819.94], [363.7, 822.51], [366.24, 825.14], [368.68, 827.86], [371.08, 830.63], [373.41, 833.45], [375.65, 836.34], [377.82, 839.28], [379.95, 842.26], [381.99, 845.29], [384.01, 848.34], [385.98, 851.43], [387.89, 854.54], [389.77, 857.68], [391.6, 860.85], [393.36, 864.05], [395.05, 867.3], [396.68, 870.57], [398.32, 873.84], [399.97, 877.1], [401.64, 880.36], [403.34, 883.6], [405.06, 886.82], [406.84, 890.02], [408.69, 893.17], [410.56, 896.32], [412.44, 899.46], [414.33, 902.59], [416.23, 905.71], [418.17, 908.82], [420.11, 911.92], [422.05, 915.02], [424.04, 918.08], [426.1, 921.1], [428.23, 924.08], [430.41, 927.02], [432.55, 929.98], [434.65, 932.98], [436.73, 935.99], [438.73, 939.05], [440.64, 942.17], [442.5, 945.32], [444.34, 948.48], [446.22, 951.61], [448.1, 954.75], [449.95, 957.91], [451.73, 961.1], [453.42, 964.35], [455.05, 967.62], [456.68, 970.9], [458.28, 974.18], [459.94, 977.44], [461.63, 980.69], [463.32, 983.93], [465.0, 987.18], [466.65, 990.44], [468.32, 993.7], [469.98, 996.95], [471.61, 1000.23], [473.23, 1003.51], [474.79, 1006.82], [476.31, 1010.14], [477.8, 1013.49], [479.29, 1016.83], [480.75, 1020.18], [482.2, 1023.54], [483.73, 1026.86], [485.31, 1030.16], [486.89, 1033.46], [488.54, 1036.72], [490.24, 1039.96], [491.98, 1043.18], [493.77, 1046.36], [495.6, 1049.53], [497.53, 1052.64], [499.58, 1055.67], [501.72, 1058.63], [503.96, 1061.52], [506.3, 1064.34], [508.73, 1067.07], [511.19, 1069.78], [513.69, 1072.45], [516.23, 1075.08], [518.76, 1077.72], [521.32, 1080.33], [523.88, 1082.95], [526.42, 1085.58], [528.96, 1088.21], [531.5, 1090.84], [534.02, 1093.49], [536.53, 1096.15], [538.96, 1098.89], [541.27, 1101.72], [543.47, 1104.64], [545.57, 1107.64], [547.57, 1110.7], [549.46, 1113.83], [551.28, 1117.01], [553.04, 1120.21], [554.73, 1123.45], [556.35, 1126.73], [557.83, 1130.08], [559.17, 1133.48], [560.37, 1136.94], [561.42, 1140.44], [562.33, 1143.98], [563.1, 1147.56], [563.71, 1151.17], [564.17, 1154.79], [564.49, 1158.44], [564.65, 1162.09], [564.66, 1165.75], [564.52, 1169.4], [564.23, 1173.05], [563.79, 1176.68], [563.2, 1180.29], [562.49, 1183.88], [561.63, 1187.43], [560.63, 1190.95], [559.48, 1194.42], [558.18, 1197.84], [556.75, 1201.21], [555.21, 1204.53], [553.65, 1207.83], [551.96, 1211.08], [550.23, 1214.3], [548.41, 1217.47], [546.49, 1220.59], [544.49, 1223.65], [542.44, 1226.68], [540.32, 1229.66], [538.16, 1232.61], [536.02, 1235.58], [533.9, 1238.55], [531.65, 1241.44], [529.28, 1244.23], [526.8, 1246.91], [524.21, 1249.5], [521.51, 1251.97], [518.72, 1254.33], [515.83, 1256.57], [512.85, 1258.69], [509.79, 1260.69], [506.64, 1262.56], [503.42, 1264.3], [500.14, 1265.9], [496.78, 1267.36], [493.38, 1268.69], [489.91, 1269.87], [486.41, 1270.91], [482.86, 1271.8], [479.28, 1272.55], [475.67, 1273.15], [472.04, 1273.6], [468.4, 1273.91], [464.74, 1274.07], [461.09, 1274.14], [457.43, 1274.16], [453.77, 1274.24], [450.12, 1274.23], [446.46, 1274.21], [442.8, 1274.21], [439.14, 1274.21], [435.49, 1274.22], [431.83, 1274.26], [428.17, 1274.33], [424.51, 1274.38], [420.86, 1274.41], [417.2, 1274.39], [413.54, 1274.39], [409.89, 1274.49], [406.23, 1274.72], [402.59, 1275.06], [398.95, 1275.42], [395.31, 1275.72], [391.66, 1275.97], [388.01, 1276.18], [384.36, 1276.41], [380.71, 1276.68], [377.07, 1277.05], [373.44, 1277.48], [369.81, 1277.93], [366.18, 1278.42], [362.56, 1278.95], [358.95, 1279.49], [355.33, 1280.05], [351.72, 1280.61], [348.11, 1281.22], [344.51, 1281.86], [340.9, 1282.46], [337.29, 1283.02], [333.68, 1283.59], [330.06, 1284.15], [326.44, 1284.67], [322.82, 1285.15], [319.19, 1285.61], [315.56, 1286.07], [311.93, 1286.52], [308.29, 1286.89], [304.64, 1287.16], [300.99, 1287.36], [297.34, 1287.5], [293.68, 1287.52], [290.02, 1287.51], [286.36, 1287.51], [282.71, 1287.53], [279.05, 1287.53], [275.39, 1287.52], [271.73, 1287.5], [268.08, 1287.49], [264.42, 1287.52], [260.76, 1287.52], [257.1, 1287.49], [253.45, 1287.39], [249.79, 1287.22], [246.14, 1287.04], [242.49, 1286.83], [238.84, 1286.65], [235.18, 1286.46], [231.53, 1286.28], [227.88, 1286.06], [224.22, 1285.91], [220.57, 1285.81], [216.91, 1285.7], [213.26, 1285.59], [209.6, 1285.41], [205.95, 1285.25], [202.29, 1285.15], [198.64, 1285.13], [194.98, 1285.26], [191.33, 1285.42], [187.68, 1285.68], [184.03, 1285.95], [180.39, 1286.26], [176.74, 1286.61], [173.11, 1286.97], [169.47, 1287.41], [165.85, 1287.88], [162.22, 1288.36], [158.58, 1288.69], [154.92, 1288.86], [151.27, 1288.89], [147.61, 1288.77], [143.96, 1288.49], [140.33, 1288.12], [136.69, 1287.74], [133.05, 1287.37], [129.41, 1286.99], [125.77, 1286.62], [122.13, 1286.24], [118.5, 1285.87], [114.86, 1285.5], [111.22, 1285.12], [107.58, 1284.75], [103.94, 1284.37], [100.3, 1284.0], [96.67, 1283.62], [93.03, 1283.25], [89.39, 1282.88], [85.75, 1282.5], [82.11, 1282.13], [78.47, 1281.75], [74.84, 1281.38], [71.2, 1281.0]]], "image": "images/traverse_ordering_00002.png"} +{"width": 1916, "height": 1916, "num_extra_distractor_curves": 1, "s_curve": 1, "s_at_start": true, "s_point": {"x": 1267.64, "y": 1846.0, "label_x": 1319.64, "label_y": 1846.22}, "marks": [{"label": "B", "curve": 1, "on_s_curve": true, "polyline_index": 233, "x": 982.13, "y": 1362.2, "label_x": 1029.36, "label_y": 1370.77}, {"label": "H", "curve": 1, "on_s_curve": true, "polyline_index": 493, "x": 1367.72, "y": 849.38, "label_x": 1368.0, "label_y": 801.38}, {"label": "D", "curve": 1, "on_s_curve": true, "polyline_index": 710, "x": 653.6, "y": 1211.63, "label_x": 626.98, "label_y": 1171.69}, {"label": "A", "curve": 1, "on_s_curve": true, "polyline_index": 887, "x": 624.01, "y": 874.64, "label_x": 662.12, "label_y": 903.83}, {"label": "F", "curve": 1, "on_s_curve": true, "polyline_index": 1053, "x": 1082.78, "y": 466.23, "label_x": 1053.56, "label_y": 428.14}, {"label": "G", "curve": 1, "on_s_curve": true, "polyline_index": 1236, "x": 838.44, "y": 518.29, "label_x": 817.22, "label_y": 475.24}, {"label": "I", "curve": 0, "on_s_curve": false, "polyline_index": 263, "x": 1242.2, "y": 619.66, "label_x": 1238.27, "label_y": 571.82}, {"label": "E", "curve": 0, "on_s_curve": false, "polyline_index": 686, "x": 652.45, "y": 651.46, "label_x": 624.2, "label_y": 612.65}, {"label": "C", "curve": 0, "on_s_curve": false, "polyline_index": 874, "x": 999.93, "y": 1112.4, "label_x": 1019.35, "label_y": 1068.51}, {"label": "K", "curve": 0, "on_s_curve": false, "polyline_index": 1033, "x": 1460.7, "y": 1392.5, "label_x": 1425.5, "label_y": 1425.14}, {"label": "L", "curve": 0, "on_s_curve": false, "polyline_index": 1247, "x": 926.29, "y": 1537.95, "label_x": 932.62, "label_y": 1490.37}, {"label": "J", "curve": 0, "on_s_curve": false, "polyline_index": 1410, "x": 428.01, "y": 1638.39, "label_x": 380.24, "label_y": 1633.65}], "num_marks_total": 12, "num_marks_on_s": 6, "num_pair_crossings": 9, "num_self_crossings": 1, "question": "The image shows two curves or strands that cross each other. Several labeled points are marked on the two curves combined. A green 'S' marks the start endpoint of one of the curves. Starting from S, follow THAT curve (the one S lies on) all the way to its other endpoint, and list the labels of the marked points you visit in order, separated by commas. Ignore any labeled points that lie on the OTHER curve. Provide your final answer enclosed in ... tags.", "answer": "B, H, D, A, F, G", "polylines": [[[1041.46, 70.0], [1040.93, 73.39], [1040.4, 76.77], [1039.87, 80.16], [1039.35, 83.54], [1038.82, 86.93], [1038.29, 90.31], [1037.76, 93.7], [1037.23, 97.08], [1036.7, 100.47], [1036.18, 103.85], [1035.65, 107.24], [1035.12, 110.62], [1034.59, 114.01], [1034.06, 117.39], [1033.54, 120.78], [1033.01, 124.16], [1032.48, 127.55], [1031.95, 130.93], [1031.42, 134.32], [1030.89, 137.7], [1030.37, 141.09], [1029.84, 144.47], [1029.31, 147.86], [1028.78, 151.24], [1028.25, 154.63], [1027.72, 158.01], [1027.2, 161.4], [1026.67, 164.78], [1026.14, 168.17], [1025.61, 171.55], [1025.08, 174.94], [1024.55, 178.32], [1024.03, 181.71], [1023.5, 185.09], [1022.97, 188.48], [1022.44, 191.86], [1021.91, 195.25], [1021.39, 198.63], [1020.86, 202.02], [1020.33, 205.4], [1019.8, 208.79], [1019.27, 212.17], [1018.74, 215.56], [1018.35, 218.96], [1018.1, 222.38], [1017.99, 225.8], [1018.02, 229.23], [1018.18, 232.65], [1018.49, 236.06], [1018.93, 239.46], [1019.51, 242.84], [1020.23, 246.19], [1021.08, 249.51], [1022.07, 252.79], [1023.19, 256.02], [1024.44, 259.21], [1025.82, 262.35], [1027.33, 265.43], [1028.96, 268.44], [1030.71, 271.39], [1032.56, 274.27], [1034.5, 277.09], [1036.53, 279.85], [1038.63, 282.56], [1040.79, 285.22], [1043.01, 287.83], [1045.29, 290.39], [1047.63, 292.89], [1050.01, 295.35], [1052.49, 297.72], [1055.0, 300.04], [1057.59, 302.29], [1060.23, 304.47], [1062.95, 306.56], [1065.7, 308.6], [1068.49, 310.59], [1071.32, 312.52], [1074.19, 314.39], [1077.08, 316.24], [1079.98, 318.05], [1082.92, 319.82], [1085.86, 321.57], [1088.83, 323.29], [1091.8, 325.0], [1094.78, 326.69], [1097.75, 328.39], [1100.69, 330.14], [1103.64, 331.9], [1106.59, 333.64], [1109.56, 335.34], [1112.55, 337.01], [1115.52, 338.72], [1118.48, 340.44], [1121.44, 342.18], [1124.38, 343.93], [1127.33, 345.68], [1130.27, 347.43], [1133.22, 349.18], [1136.18, 350.91], [1139.13, 352.64], [1142.09, 354.38], [1145.02, 356.14], [1147.97, 357.89], [1150.92, 359.63], [1153.86, 361.38], [1156.8, 363.14], [1159.76, 364.88], [1162.69, 366.64], [1165.64, 368.39], [1168.61, 370.1], [1171.59, 371.79], [1174.58, 373.47], [1177.58, 375.12], [1180.63, 376.68], [1183.74, 378.11], [1186.9, 379.43], [1190.11, 380.62], [1193.37, 381.7], [1196.65, 382.67], [1199.96, 383.58], [1203.29, 384.37], [1206.65, 385.06], [1210.02, 385.67], [1213.4, 386.22], [1216.77, 386.81], [1220.15, 387.41], [1223.52, 388.01], [1226.9, 388.59], [1230.27, 389.18], [1233.63, 389.88], [1236.96, 390.67], [1240.27, 391.54], [1243.56, 392.49], [1246.83, 393.53], [1250.07, 394.65], [1253.29, 395.81], [1256.5, 397.0], [1259.69, 398.25], [1262.87, 399.54], [1266.04, 400.82], [1269.23, 402.09], [1272.4, 403.39], [1275.56, 404.7], [1278.72, 406.03], [1281.87, 407.37], [1285.03, 408.7], [1288.17, 410.07], [1291.32, 411.42], [1294.48, 412.74], [1297.68, 413.96], [1300.9, 415.14], [1304.12, 416.31], [1307.35, 417.44], [1310.59, 418.57], [1313.83, 419.67], [1317.09, 420.74], [1320.35, 421.77], [1323.61, 422.84], [1326.85, 423.95], [1330.09, 425.05], [1333.36, 426.08], [1336.62, 427.13], [1339.9, 428.12], [1343.19, 429.09], [1346.48, 430.04], [1349.77, 430.98], [1353.06, 431.96], [1356.3, 433.07], [1359.49, 434.31], [1362.63, 435.68], [1365.72, 437.17], [1368.74, 438.79], [1371.69, 440.53], [1374.57, 442.39], [1377.37, 444.36], [1380.08, 446.45], [1382.71, 448.65], [1385.25, 450.95], [1387.7, 453.35], [1390.04, 455.85], [1392.28, 458.44], [1394.41, 461.12], [1396.44, 463.89], [1398.34, 466.73], [1400.13, 469.65], [1401.81, 472.64], [1403.35, 475.7], [1404.77, 478.82], [1406.07, 481.99], [1407.23, 485.21], [1408.27, 488.48], [1409.16, 491.78], [1409.93, 495.12], [1410.56, 498.49], [1411.04, 501.88], [1411.4, 505.29], [1411.61, 508.71], [1411.68, 512.14], [1411.62, 515.56], [1411.41, 518.98], [1411.07, 522.39], [1410.59, 525.78], [1409.97, 529.15], [1409.21, 532.49], [1408.32, 535.8], [1407.29, 539.07], [1406.14, 542.29], [1404.85, 545.47], [1403.44, 548.59], [1401.9, 551.65], [1400.23, 554.65], [1398.45, 557.57], [1396.55, 560.42], [1394.53, 563.19], [1392.41, 565.88], [1390.17, 568.48], [1387.84, 570.99], [1385.41, 573.4], [1382.89, 575.72], [1380.27, 577.93], [1377.56, 580.03], [1374.78, 582.02], [1371.93, 583.93], [1369.04, 585.77], [1366.12, 587.56], [1363.15, 589.27], [1360.16, 590.94], [1357.16, 592.6], [1354.14, 594.21], [1351.08, 595.76], [1348.01, 597.28], [1344.91, 598.74], [1341.79, 600.16], [1338.65, 601.53], [1335.48, 602.83], [1332.3, 604.1], [1329.11, 605.34], [1325.91, 606.56], [1322.7, 607.78], [1319.5, 608.99], [1316.29, 610.18], [1313.07, 611.37], [1309.85, 612.52], [1306.62, 613.66], [1303.38, 614.8], [1300.11, 615.8], [1296.79, 616.67], [1293.45, 617.4], [1290.07, 618.0], [1286.68, 618.45], [1283.27, 618.77], [1279.85, 618.97], [1276.42, 619.07], [1273.0, 619.06], [1269.57, 618.98], [1266.15, 618.87], [1262.72, 618.77], [1259.3, 618.72], [1255.87, 618.76], [1252.45, 618.9], [1249.03, 619.11], [1245.61, 619.36], [1242.2, 619.66], [1238.79, 620.02], [1235.39, 620.44], [1231.99, 620.88], [1228.59, 621.3], [1225.19, 621.72], [1221.79, 622.13], [1218.39, 622.53], [1214.98, 622.86], [1211.56, 623.09], [1208.14, 623.2], [1204.71, 623.2], [1201.28, 623.17], [1197.86, 623.19], [1194.43, 623.21], [1191.01, 623.2], [1187.58, 623.21], [1184.15, 623.23], [1180.73, 623.2], [1177.3, 623.21], [1173.88, 623.2], [1170.45, 623.15], [1167.02, 623.15], [1163.6, 623.22], [1160.18, 623.38], [1156.76, 623.61], [1153.34, 623.91], [1149.94, 624.28], [1146.53, 624.66], [1143.13, 625.08], [1139.74, 625.54], [1136.35, 626.03], [1132.96, 626.56], [1129.57, 627.04], [1126.17, 627.48], [1122.77, 627.9], [1119.37, 628.27], [1115.96, 628.63], [1112.55, 628.99], [1109.15, 629.38], [1105.75, 629.81], [1102.35, 630.2], [1098.94, 630.58], [1095.54, 630.96], [1092.12, 631.24], [1088.7, 631.43], [1085.27, 631.48], [1081.85, 631.4], [1078.43, 631.25], [1075.01, 631.03], [1071.6, 630.71], [1068.2, 630.29], [1064.8, 629.84], [1061.42, 629.3], [1058.05, 628.69], [1054.69, 628.01], [1051.34, 627.26], [1048.01, 626.47], [1044.67, 625.71], [1041.32, 624.98], [1037.97, 624.25], [1034.64, 623.47], [1031.31, 622.67], [1027.96, 621.94], [1024.61, 621.23], [1021.25, 620.54], [1017.9, 619.82], [1014.56, 619.05], [1011.23, 618.27], [1007.88, 617.53], [1004.52, 616.86], [1001.15, 616.24], [997.77, 615.69], [994.37, 615.26], [990.96, 614.94], [987.55, 614.68], [984.13, 614.4], [980.72, 614.07], [977.32, 613.66], [973.91, 613.28], [970.52, 612.84], [967.11, 612.48], [963.7, 612.16], [960.28, 611.93], [956.86, 611.82], [953.43, 611.8], [950.01, 611.85], [946.58, 611.94], [943.16, 612.09], [939.73, 612.23], [936.31, 612.35], [932.89, 612.52], [929.46, 612.64], [926.04, 612.69], [922.61, 612.64], [919.19, 612.52], [915.77, 612.3], [912.36, 612.02], [908.94, 611.72], [905.54, 611.35], [902.13, 611.02], [898.71, 610.77], [895.29, 610.53], [891.87, 610.34], [888.45, 610.25], [885.02, 610.25], [881.59, 610.31], [878.17, 610.38], [874.74, 610.43], [871.32, 610.42], [867.89, 610.42], [864.47, 610.43], [861.04, 610.4], [857.61, 610.34], [854.19, 610.29], [850.76, 610.25], [847.34, 610.19], [843.91, 610.1], [840.49, 609.91], [837.08, 609.64], [833.67, 609.28], [830.27, 608.82], [826.9, 608.25], [823.5, 607.81], [820.09, 607.51], [816.66, 607.35], [813.24, 607.33], [809.81, 607.45], [806.4, 607.71], [802.99, 608.1], [799.61, 608.64], [796.25, 609.31], [792.92, 610.12], [789.63, 611.06], [786.37, 612.14], [783.17, 613.34], [780.01, 614.68], [776.91, 616.14], [773.88, 617.73], [770.91, 619.44], [768.01, 621.27], [765.19, 623.22], [762.46, 625.28], [759.8, 627.44], [757.24, 629.72], [754.77, 632.1], [752.4, 634.57], [750.14, 637.14], [747.98, 639.8], [745.93, 642.54], [743.99, 645.37], [742.17, 648.27], [740.47, 651.25], [738.89, 654.29], [737.44, 657.39], [736.11, 660.55], [734.91, 663.76], [733.85, 667.02], [732.92, 670.31], [732.12, 673.64], [731.46, 677.01], [730.93, 680.39], [730.55, 683.8], [730.3, 687.21], [730.19, 690.64], [730.22, 694.06], [730.39, 697.48], [730.7, 700.9], [731.15, 704.29], [731.73, 707.67], [732.46, 711.02], [733.31, 714.34], [734.31, 717.61], [735.43, 720.85], [736.69, 724.04], [738.07, 727.17], [739.58, 730.25], [741.21, 733.26], [742.97, 736.2], [744.84, 739.07], [746.83, 741.86], [748.92, 744.57], [751.06, 747.25], [753.24, 749.89], [755.5, 752.47], [757.86, 754.95], [760.32, 757.34], [762.86, 759.64], [765.49, 761.83], [768.21, 763.92], [771.01, 765.89], [773.89, 767.75], [776.8, 769.55], [779.68, 771.41], [782.53, 773.3], [785.39, 775.2], [788.23, 777.12], [791.06, 779.04], [793.91, 780.95], [796.75, 782.86], [799.6, 784.76], [802.45, 786.67], [805.28, 788.6], [808.1, 790.54], [810.92, 792.49], [813.72, 794.46], [816.55, 796.39], [819.39, 798.3], [822.28, 800.15], [825.24, 801.87], [828.27, 803.48], [831.36, 804.95], [834.48, 806.37], [837.59, 807.8], [840.73, 809.18], [843.87, 810.54], [846.98, 811.98], [850.09, 813.43], [853.19, 814.89], [856.3, 816.32], [859.43, 817.72], [862.55, 819.12], [865.7, 820.47], [868.88, 821.76], [872.08, 822.98], [875.3, 824.14], [878.57, 825.17], [881.87, 826.07], [885.2, 826.9], [888.53, 827.71], [891.87, 828.48], [895.21, 829.25], [898.54, 830.02], [901.87, 830.86], [905.16, 831.81], [908.43, 832.82], [911.7, 833.85], [914.97, 834.86], [918.24, 835.9], [921.5, 836.93], [924.76, 837.99], [928.02, 839.04], [931.29, 840.06], [934.55, 841.13], [937.79, 842.25], [941.07, 843.24], [944.38, 844.1], [947.73, 844.82], [951.11, 845.4], [954.5, 845.85], [957.92, 846.16], [961.34, 846.33], [964.76, 846.36], [968.19, 846.25], [971.61, 846.0], [975.01, 845.61], [978.4, 845.09], [981.76, 844.43], [985.09, 843.63], [988.39, 842.69], [991.64, 841.63], [994.85, 840.43], [998.01, 839.1], [1001.11, 837.65], [1004.15, 836.07], [1007.13, 834.37], [1010.03, 832.55], [1012.85, 830.61], [1015.6, 828.56], [1018.26, 826.4], [1020.83, 824.13], [1023.3, 821.76], [1025.68, 819.29], [1027.95, 816.73], [1030.12, 814.08], [1032.18, 811.34], [1034.12, 808.52], [1035.95, 805.62], [1037.66, 802.65], [1039.25, 799.62], [1040.71, 796.52], [1042.05, 793.36], [1043.25, 790.16], [1044.33, 786.9], [1045.27, 783.61], [1046.08, 780.28], [1046.75, 776.92], [1047.28, 773.54], [1047.68, 770.13], [1047.93, 766.72], [1048.05, 763.29], [1048.03, 759.87], [1047.87, 756.44], [1047.57, 753.03], [1047.13, 749.63], [1046.56, 746.25], [1045.84, 742.9], [1045.0, 739.58], [1044.01, 736.3], [1042.9, 733.06], [1041.65, 729.87], [1040.28, 726.73], [1038.78, 723.65], [1037.15, 720.64], [1035.41, 717.69], [1033.54, 714.81], [1031.56, 712.02], [1029.47, 709.3], [1027.27, 706.68], [1024.97, 704.14], [1022.56, 701.71], [1020.06, 699.37], [1017.46, 697.13], [1014.77, 695.0], [1012.01, 692.99], [1009.16, 691.08], [1006.24, 689.29], [1003.27, 687.57], [1000.28, 685.91], [997.24, 684.32], [994.2, 682.74], [991.13, 681.21], [988.06, 679.7], [984.97, 678.22], [981.86, 676.78], [978.74, 675.36], [975.59, 674.03], [972.41, 672.75], [969.2, 671.56], [965.99, 670.37], [962.78, 669.15], [959.57, 667.97], [956.36, 666.77], [953.13, 665.62], [949.88, 664.52], [946.62, 663.47], [943.34, 662.5], [940.04, 661.58], [936.72, 660.72], [933.39, 659.92], [930.05, 659.17], [926.71, 658.41], [923.38, 657.58], [920.07, 656.71], [916.76, 655.81], [913.47, 654.87], [910.18, 653.91], [906.9, 652.93], [903.62, 651.92], [900.37, 650.85], [897.11, 649.79], [893.86, 648.7], [890.62, 647.59], [887.38, 646.49], [884.13, 645.4], [880.88, 644.32], [877.62, 643.25], [874.36, 642.21], [871.07, 641.26], [867.75, 640.4], [864.4, 639.68], [861.03, 639.07], [857.64, 638.55], [854.24, 638.16], [850.82, 637.91], [847.4, 637.73], [843.97, 637.67], [840.55, 637.68], [837.12, 637.75], [833.7, 637.83], [830.27, 637.89], [826.85, 637.91], [823.42, 637.83], [820.0, 637.62], [816.59, 637.34], [813.18, 636.99], [809.78, 636.58], [806.37, 636.2], [802.96, 635.91], [799.54, 635.71], [796.11, 635.61], [792.69, 635.59], [789.26, 635.66], [785.84, 635.78], [782.42, 635.97], [779.0, 636.2], [775.59, 636.48], [772.18, 636.83], [768.77, 637.19], [765.36, 637.52], [761.95, 637.82], [758.53, 638.02], [755.1, 638.19], [751.68, 638.34], [748.26, 638.4], [744.83, 638.39], [741.4, 638.31], [737.98, 638.14], [734.57, 637.86], [731.16, 637.52], [727.75, 637.22], [724.33, 636.91], [720.92, 636.6], [717.51, 636.3], [714.09, 636.05], [710.68, 635.77], [707.26, 635.49], [703.84, 635.34], [700.41, 635.34], [696.99, 635.47], [693.58, 635.75], [690.17, 636.16], [686.79, 636.71], [683.44, 637.4], [680.11, 638.22], [676.82, 639.18], [673.57, 640.27], [670.37, 641.5], [667.23, 642.85], [664.14, 644.33], [661.11, 645.93], [658.15, 647.65], [655.26, 649.5], [652.45, 651.46], [649.72, 653.53], [647.08, 655.71], [644.53, 658.0], [642.07, 660.38], [639.71, 662.87], [637.46, 665.45], [635.31, 668.12], [633.28, 670.88], [631.35, 673.71], [629.55, 676.62], [627.86, 679.6], [626.29, 682.65], [624.86, 685.76], [623.54, 688.93], [622.36, 692.14], [621.31, 695.4], [620.4, 698.71], [619.62, 702.04], [618.97, 705.41], [618.46, 708.79], [618.09, 712.2], [617.86, 715.62], [617.74, 719.04], [617.76, 722.47], [617.92, 725.89], [618.21, 729.3], [618.65, 732.7], [619.22, 736.08], [619.93, 739.43], [620.74, 742.76], [621.64, 746.07], [622.66, 749.34], [623.79, 752.57], [625.03, 755.77], [626.39, 758.91], [627.88, 762.0], [629.49, 765.02], [631.21, 767.98], [633.05, 770.87], [634.96, 773.72], [636.95, 776.5], [638.99, 779.26], [641.1, 781.96], [643.24, 784.63], [645.42, 787.28], [647.58, 789.93], [649.73, 792.6], [651.83, 795.31], [653.94, 798.01], [656.05, 800.71], [658.11, 803.44], [660.14, 806.21], [662.16, 808.98], [664.15, 811.76], [666.2, 814.51], [668.35, 817.18], [670.59, 819.77], [672.93, 822.27], [675.33, 824.71], [677.76, 827.13], [680.22, 829.51], [682.67, 831.91], [685.15, 834.27], [687.62, 836.65], [690.09, 839.03], [692.59, 841.37], [695.08, 843.72], [697.61, 846.03], [700.14, 848.34], [702.66, 850.66], [705.16, 853.0], [707.67, 855.33], [710.15, 857.7], [712.61, 860.08], [715.07, 862.47], [717.57, 864.81], [720.12, 867.1], [722.71, 869.34], [725.32, 871.56], [727.98, 873.72], [730.64, 875.88], [733.32, 878.01], [736.03, 880.12], [738.7, 882.25], [741.38, 884.4], [744.07, 886.51], [746.79, 888.59], [749.54, 890.64], [752.28, 892.69], [755.03, 894.75], [757.73, 896.85], [760.35, 899.05], [762.92, 901.32], [765.39, 903.7], [767.78, 906.16], [770.09, 908.68], [772.35, 911.26], [774.51, 913.92], [776.6, 916.63], [778.64, 919.38], [780.64, 922.16], [782.62, 924.96], [784.57, 927.78], [786.45, 930.64], [788.3, 933.53], [790.19, 936.39], [792.03, 939.27], [793.85, 942.18], [795.63, 945.1], [797.37, 948.05], [799.05, 951.04], [800.78, 954.0], [802.58, 956.91], [804.44, 959.79], [806.34, 962.64], [808.24, 965.49], [810.09, 968.37], [811.96, 971.25], [813.86, 974.1], [815.76, 976.95], [817.72, 979.76], [819.69, 982.56], [821.7, 985.34], [823.72, 988.1], [825.76, 990.86], [827.79, 993.61], [829.84, 996.36], [831.88, 999.11], [833.94, 1001.85], [835.98, 1004.6], [838.06, 1007.33], [840.2, 1010.0], [842.39, 1012.64], [844.64, 1015.22], [846.99, 1017.72], [849.44, 1020.11], [851.95, 1022.44], [854.53, 1024.69], [857.15, 1026.9], [859.83, 1029.04], [862.56, 1031.1], [865.34, 1033.1], [868.12, 1035.11], [870.92, 1037.08], [873.74, 1039.03], [876.54, 1041.01], [879.34, 1042.98], [882.15, 1044.94], [884.93, 1046.94], [887.67, 1048.99], [890.36, 1051.12], [893.01, 1053.3], [895.61, 1055.52], [898.16, 1057.81], [900.68, 1060.14], [903.16, 1062.49], [905.66, 1064.83], [908.26, 1067.07], [910.88, 1069.28], [913.57, 1071.39], [916.35, 1073.4], [919.19, 1075.32], [922.07, 1077.17], [924.98, 1078.98], [927.9, 1080.77], [930.86, 1082.51], [933.83, 1084.21], [936.84, 1085.85], [939.84, 1087.5], [942.88, 1089.07], [945.94, 1090.61], [949.04, 1092.09], [952.14, 1093.54], [955.27, 1094.93], [958.43, 1096.26], [961.6, 1097.55], [964.78, 1098.84], [967.96, 1100.1], [971.15, 1101.35], [974.35, 1102.57], [977.56, 1103.78], [980.77, 1104.98], [983.98, 1106.17], [987.21, 1107.32], [990.44, 1108.46], [993.65, 1109.65], [996.81, 1110.96], [999.93, 1112.4], [1003.0, 1113.91], [1006.05, 1115.48], [1009.1, 1117.04], [1012.15, 1118.59], [1015.23, 1120.09], [1018.31, 1121.6], [1021.38, 1123.12], [1024.44, 1124.65], [1027.51, 1126.17], [1030.6, 1127.65], [1033.69, 1129.13], [1036.8, 1130.56], [1039.93, 1131.97], [1043.06, 1133.36], [1046.21, 1134.7], [1049.36, 1136.05], [1052.52, 1137.38], [1055.69, 1138.69], [1058.85, 1140.0], [1062.02, 1141.3], [1065.21, 1142.56], [1068.4, 1143.8], [1071.61, 1145.0], [1074.82, 1146.19], [1078.05, 1147.35], [1081.27, 1148.51], [1084.5, 1149.66], [1087.73, 1150.81], [1090.97, 1151.91], [1094.22, 1152.99], [1097.46, 1154.11], [1100.68, 1155.28], [1103.84, 1156.59], [1106.98, 1157.97], [1110.1, 1159.39], [1113.24, 1160.76], [1116.38, 1162.13], [1119.5, 1163.54], [1122.63, 1164.93], [1125.78, 1166.29], [1128.93, 1167.63], [1132.05, 1169.05], [1135.14, 1170.52], [1138.24, 1171.99], [1141.3, 1173.54], [1144.31, 1175.17], [1147.28, 1176.88], [1150.23, 1178.62], [1153.17, 1180.38], [1156.11, 1182.13], [1159.06, 1183.88], [1162.01, 1185.61], [1165.0, 1187.29], [1168.02, 1188.91], [1171.01, 1190.59], [1173.97, 1192.31], [1176.91, 1194.07], [1179.83, 1195.85], [1182.74, 1197.66], [1185.69, 1199.41], [1188.66, 1201.12], [1191.66, 1202.76], [1194.68, 1204.39], [1197.7, 1206.0], [1200.73, 1207.6], [1203.78, 1209.16], [1206.83, 1210.73], [1209.89, 1212.28], [1212.93, 1213.84], [1215.95, 1215.46], [1218.91, 1217.19], [1221.81, 1219.01], [1224.66, 1220.91], [1227.49, 1222.84], [1230.35, 1224.74], [1233.18, 1226.67], [1235.99, 1228.62], [1238.82, 1230.55], [1241.68, 1232.44], [1244.58, 1234.27], [1247.5, 1236.05], [1250.46, 1237.79], [1253.42, 1239.51], [1256.38, 1241.23], [1259.36, 1242.93], [1262.36, 1244.59], [1265.37, 1246.22], [1268.38, 1247.85], [1271.42, 1249.43], [1274.46, 1251.01], [1277.49, 1252.61], [1280.5, 1254.24], [1283.51, 1255.88], [1286.52, 1257.52], [1289.54, 1259.14], [1292.58, 1260.72], [1295.64, 1262.27], [1298.7, 1263.8], [1301.76, 1265.34], [1304.83, 1266.86], [1307.9, 1268.38], [1310.94, 1269.96], [1313.98, 1271.54], [1317.02, 1273.13], [1320.06, 1274.71], [1323.09, 1276.29], [1326.13, 1277.88], [1329.15, 1279.5], [1332.17, 1281.12], [1335.16, 1282.78], [1338.11, 1284.53], [1341.02, 1286.34], [1343.92, 1288.16], [1346.84, 1289.96], [1349.75, 1291.77], [1352.65, 1293.58], [1355.56, 1295.39], [1358.45, 1297.23], [1361.33, 1299.09], [1364.19, 1300.98], [1367.01, 1302.93], [1369.79, 1304.93], [1372.52, 1307.0], [1375.19, 1309.14], [1377.79, 1311.38], [1380.32, 1313.69], [1382.76, 1316.09], [1385.14, 1318.55], [1387.47, 1321.07], [1389.73, 1323.64], [1391.96, 1326.25], [1394.13, 1328.89], [1396.27, 1331.57], [1398.35, 1334.29], [1400.42, 1337.02], [1402.49, 1339.75], [1404.64, 1342.42], [1406.8, 1345.08], [1409.07, 1347.65], [1411.44, 1350.12], [1413.86, 1352.55], [1416.35, 1354.9], [1418.89, 1357.19], [1421.49, 1359.42], [1424.14, 1361.6], [1426.81, 1363.75], [1429.5, 1365.87], [1432.22, 1367.96], [1434.98, 1369.98], [1437.76, 1371.98], [1440.53, 1374.0], [1443.23, 1376.11], [1445.9, 1378.25], [1448.51, 1380.47], [1451.06, 1382.76], [1453.55, 1385.11], [1455.99, 1387.52], [1458.38, 1389.98], [1460.7, 1392.5], [1462.97, 1395.06], [1465.2, 1397.66], [1467.41, 1400.28], [1469.58, 1402.93], [1471.71, 1405.62], [1473.8, 1408.33], [1475.86, 1411.07], [1477.93, 1413.8], [1480.01, 1416.52], [1482.12, 1419.22], [1484.23, 1421.92], [1486.39, 1424.58], [1488.43, 1427.33], [1490.36, 1430.16], [1492.18, 1433.06], [1493.88, 1436.04], [1495.45, 1439.09], [1496.9, 1442.19], [1498.22, 1445.35], [1499.41, 1448.56], [1500.47, 1451.82], [1501.39, 1455.12], [1502.18, 1458.45], [1502.84, 1461.82], [1503.36, 1465.2], [1503.74, 1468.61], [1503.98, 1472.03], [1504.08, 1475.45], [1504.04, 1478.88], [1503.86, 1482.3], [1503.55, 1485.71], [1503.09, 1489.11], [1502.5, 1492.48], [1501.77, 1495.83], [1500.91, 1499.14], [1499.91, 1502.42], [1498.78, 1505.66], [1497.52, 1508.84], [1496.13, 1511.97], [1494.62, 1515.05], [1492.98, 1518.05], [1491.22, 1520.99], [1489.34, 1523.86], [1487.35, 1526.65], [1485.24, 1529.35], [1483.03, 1531.96], [1480.71, 1534.49], [1478.29, 1536.91], [1475.78, 1539.24], [1473.17, 1541.46], [1470.48, 1543.58], [1467.7, 1545.58], [1464.84, 1547.47], [1461.91, 1549.24], [1458.94, 1550.96], [1455.93, 1552.59], [1452.9, 1554.19], [1449.84, 1555.73], [1446.75, 1557.21], [1443.64, 1558.65], [1440.49, 1560.01], [1437.35, 1561.36], [1434.2, 1562.71], [1431.03, 1564.01], [1427.85, 1565.3], [1424.66, 1566.54], [1421.45, 1567.74], [1418.24, 1568.95], [1415.01, 1570.09], [1411.75, 1571.13], [1408.46, 1572.09], [1405.17, 1573.06], [1401.87, 1573.97], [1398.55, 1574.82], [1395.22, 1575.61], [1391.87, 1576.33], [1388.51, 1577.03], [1385.16, 1577.71], [1381.8, 1578.4], [1378.44, 1579.06], [1375.08, 1579.72], [1371.73, 1580.45], [1368.37, 1581.13], [1365.01, 1581.77], [1361.64, 1582.38], [1358.27, 1583.01], [1354.9, 1583.62], [1351.53, 1584.27], [1348.17, 1584.95], [1344.82, 1585.63], [1341.45, 1586.27], [1338.07, 1586.83], [1334.68, 1587.3], [1331.28, 1587.74], [1327.87, 1588.11], [1324.46, 1588.41], [1321.05, 1588.69], [1317.63, 1588.99], [1314.22, 1589.34], [1310.82, 1589.71], [1307.41, 1590.09], [1304.01, 1590.51], [1300.62, 1590.97], [1297.23, 1591.47], [1293.84, 1591.97], [1290.44, 1592.41], [1287.03, 1592.76], [1283.62, 1593.01], [1280.19, 1593.13], [1276.77, 1593.13], [1273.34, 1593.03], [1269.92, 1592.84], [1266.51, 1592.56], [1263.1, 1592.24], [1259.69, 1591.85], [1256.3, 1591.39], [1252.91, 1590.88], [1249.53, 1590.3], [1246.16, 1589.7], [1242.8, 1589.04], [1239.45, 1588.32], [1236.1, 1587.58], [1232.77, 1586.79], [1229.44, 1585.97], [1226.13, 1585.11], [1222.83, 1584.19], [1219.53, 1583.25], [1216.24, 1582.28], [1212.97, 1581.28], [1209.7, 1580.25], [1206.44, 1579.21], [1203.17, 1578.18], [1199.9, 1577.14], [1196.64, 1576.09], [1193.37, 1575.07], [1190.08, 1574.11], [1186.77, 1573.24], [1183.43, 1572.49], [1180.07, 1571.82], [1176.7, 1571.2], [1173.32, 1570.61], [1169.94, 1570.1], [1166.53, 1569.7], [1163.12, 1569.43], [1159.7, 1569.19], [1156.28, 1569.01], [1152.85, 1568.92], [1149.43, 1568.91], [1146.0, 1568.93], [1142.57, 1568.96], [1139.15, 1569.03], [1135.72, 1569.11], [1132.3, 1569.19], [1128.87, 1569.24], [1125.45, 1569.3], [1122.02, 1569.35], [1118.6, 1569.38], [1115.17, 1569.37], [1111.74, 1569.33], [1108.32, 1569.27], [1104.89, 1569.16], [1101.47, 1568.98], [1098.06, 1568.75], [1094.63, 1568.56], [1091.22, 1568.32], [1087.8, 1568.03], [1084.39, 1567.79], [1080.97, 1567.52], [1077.55, 1567.25], [1074.15, 1566.91], [1070.74, 1566.53], [1067.33, 1566.17], [1063.93, 1565.78], [1060.53, 1565.32], [1057.14, 1564.83], [1053.76, 1564.28], [1050.39, 1563.69], [1047.03, 1563.03], [1043.66, 1562.36], [1040.31, 1561.66], [1036.96, 1560.95], [1033.62, 1560.17], [1030.29, 1559.37], [1026.96, 1558.56], [1023.62, 1557.8], [1020.28, 1557.05], [1016.93, 1556.34], [1013.57, 1555.66], [1010.21, 1554.98], [1006.84, 1554.35], [1003.47, 1553.74], [1000.1, 1553.16], [996.72, 1552.56], [993.36, 1551.93], [989.99, 1551.28], [986.62, 1550.66], [983.26, 1550.0], [979.89, 1549.37], [976.52, 1548.75], [973.16, 1548.1], [969.8, 1547.4], [966.46, 1546.68], [963.12, 1545.91], [959.78, 1545.12], [956.45, 1544.34], [953.11, 1543.57], [949.77, 1542.79], [946.43, 1542.03], [943.09, 1541.3], [939.74, 1540.58], [936.39, 1539.85], [933.04, 1539.13], [929.67, 1538.5], [926.29, 1537.95], [922.9, 1537.48], [919.49, 1537.09], [916.08, 1536.83], [912.65, 1536.68], [909.23, 1536.61], [905.8, 1536.63], [902.38, 1536.78], [898.96, 1537.03], [895.55, 1537.35], [892.15, 1537.77], [888.76, 1538.25], [885.37, 1538.78], [882.0, 1539.38], [878.62, 1539.97], [875.25, 1540.59], [871.89, 1541.25], [868.53, 1541.9], [865.16, 1542.53], [861.79, 1543.14], [858.43, 1543.78], [855.05, 1544.39], [851.69, 1545.05], [848.34, 1545.75], [844.99, 1546.47], [841.65, 1547.22], [838.31, 1548.02], [834.98, 1548.8], [831.64, 1549.56], [828.31, 1550.35], [824.97, 1551.12], [821.61, 1551.83], [818.26, 1552.53], [814.91, 1553.26], [811.56, 1553.99], [808.22, 1554.75], [804.88, 1555.5], [801.53, 1556.2], [798.17, 1556.86], [794.8, 1557.5], [791.43, 1558.09], [788.04, 1558.64], [784.66, 1559.17], [781.26, 1559.64], [777.87, 1560.08], [774.47, 1560.53], [771.07, 1560.95], [767.66, 1561.29], [764.24, 1561.51], [760.82, 1561.72], [757.4, 1561.85], [753.97, 1561.92], [750.55, 1561.89], [747.12, 1561.77], [743.7, 1561.57], [740.29, 1561.28], [736.88, 1560.92], [733.48, 1560.53], [730.08, 1560.13], [726.67, 1559.74], [723.26, 1559.39], [719.85, 1559.05], [716.45, 1558.7], [713.04, 1558.31], [709.64, 1557.9], [706.25, 1557.41], [702.87, 1556.87], [699.48, 1556.33], [696.1, 1555.82], [692.71, 1555.29], [689.33, 1554.77], [685.95, 1554.18], [682.59, 1553.53], [679.22, 1552.9], [675.85, 1552.29], [672.47, 1551.72], [669.08, 1551.22], [665.68, 1550.78], [662.28, 1550.41], [658.86, 1550.09], [655.45, 1549.78], [652.04, 1549.47], [648.63, 1549.21], [645.21, 1548.96], [641.79, 1548.75], [638.37, 1548.57], [634.95, 1548.38], [631.52, 1548.27], [628.1, 1548.26], [624.67, 1548.34], [621.25, 1548.53], [617.84, 1548.82], [614.43, 1549.16], [611.02, 1549.5], [607.6, 1549.77], [604.19, 1550.05], [600.77, 1550.32], [597.35, 1550.53], [593.93, 1550.7], [590.51, 1550.85], [587.08, 1550.91], [583.66, 1550.89], [580.23, 1550.79], [576.81, 1550.63], [573.39, 1550.35], [569.98, 1550.02], [566.58, 1549.62], [563.19, 1549.16], [559.8, 1548.64], [556.42, 1548.08], [553.05, 1547.48], [549.68, 1546.87], [546.31, 1546.26], [542.94, 1545.62], [539.58, 1544.96], [536.22, 1544.27], [532.87, 1543.58], [529.49, 1543.02], [526.09, 1542.59], [522.67, 1542.31], [519.25, 1542.17], [515.82, 1542.16], [512.4, 1542.29], [508.98, 1542.57], [505.58, 1542.98], [502.2, 1543.53], [498.84, 1544.21], [495.52, 1545.04], [492.23, 1545.99], [488.98, 1547.08], [485.78, 1548.3], [482.63, 1549.65], [479.54, 1551.13], [476.51, 1552.73], [473.55, 1554.45], [470.66, 1556.3], [467.85, 1558.25], [465.12, 1560.33], [462.48, 1562.51], [459.92, 1564.79], [457.47, 1567.18], [455.11, 1569.66], [452.85, 1572.24], [450.68, 1574.89], [448.58, 1577.6], [446.55, 1580.36], [444.64, 1583.2], [442.84, 1586.11], [441.15, 1589.1], [439.56, 1592.13], [438.08, 1595.22], [436.68, 1598.35], [435.4, 1601.53], [434.26, 1604.76], [433.25, 1608.03], [432.37, 1611.34], [431.63, 1614.69], [430.99, 1618.05], [430.43, 1621.43], [429.88, 1624.81], [429.33, 1628.2], [428.82, 1631.58], [428.39, 1634.98], [428.01, 1638.39], [427.7, 1641.8], [427.43, 1645.22], [427.25, 1648.64], [427.12, 1652.06], [427.02, 1655.48], [426.94, 1658.91], [426.94, 1662.34], [427.07, 1665.76], [427.3, 1669.18], [427.64, 1672.59], [428.04, 1675.99], [428.52, 1679.38], [429.09, 1682.76], [429.72, 1686.13], [430.42, 1689.48], [431.11, 1692.84], [431.75, 1696.2], [432.33, 1699.58], [432.91, 1702.96], [433.51, 1706.33], [434.04, 1709.71], [434.54, 1713.1], [435.08, 1716.49], [435.57, 1719.88], [435.92, 1723.29], [436.13, 1726.71], [436.2, 1730.13], [436.13, 1733.56], [435.93, 1736.98], [435.58, 1740.39], [435.1, 1743.78], [434.48, 1747.15], [433.72, 1750.49], [432.83, 1753.8], [431.8, 1757.06], [430.65, 1760.29], [429.36, 1763.46], [427.94, 1766.58], [426.4, 1769.64], [424.74, 1772.64], [422.95, 1775.56], [421.05, 1778.41], [419.03, 1781.18], [416.9, 1783.87], [414.67, 1786.46], [412.33, 1788.97], [409.89, 1791.37], [407.36, 1793.68], [404.73, 1795.88], [402.02, 1797.97], [399.22, 1799.95], [396.35, 1801.81], [393.4, 1803.56], [390.38, 1805.18], [387.3, 1806.69], [384.16, 1808.06], [380.97, 1809.31], [377.73, 1810.42], [374.45, 1811.4], [371.13, 1812.25], [367.78, 1812.96], [364.4, 1813.54], [361.0, 1813.98], [357.59, 1814.28], [354.17, 1814.44], [350.74, 1814.46], [347.32, 1814.34], [343.9, 1814.12], [340.49, 1813.84], [337.08, 1813.5], [333.67, 1813.1], [330.28, 1812.64], [326.89, 1812.12], [323.51, 1811.55], [320.14, 1810.95], [316.76, 1810.39], [313.39, 1809.78], [310.02, 1809.18], [306.65, 1808.54], [303.29, 1807.86], [299.94, 1807.14], [296.59, 1806.42], [293.26, 1805.63], [289.93, 1804.82], [286.61, 1803.99], [283.27, 1803.21], [279.93, 1802.46], [276.59, 1801.71], [273.23, 1801.03], [269.85, 1800.45], [266.46, 1799.95], [263.06, 1799.56], [259.65, 1799.24], [256.23, 1798.98], [252.81, 1798.78], [249.39, 1798.61], [245.97, 1798.44], [242.54, 1798.27], [239.12, 1798.07], [235.7, 1797.86], [232.28, 1797.65], [228.86, 1797.53], [225.43, 1797.48], [222.01, 1797.48], [218.58, 1797.5], [215.16, 1797.38], [211.74, 1797.12], [208.34, 1796.72], [204.94, 1796.27], [201.55, 1795.82], [198.15, 1795.37], [194.75, 1794.92], [191.36, 1794.47], [187.96, 1794.02], [184.56, 1793.57], [181.17, 1793.12], [177.77, 1792.67], [174.38, 1792.22], [170.98, 1791.77], [167.58, 1791.32], [164.19, 1790.87], [160.79, 1790.42], [157.39, 1789.97], [154.0, 1789.52], [150.6, 1789.07], [147.2, 1788.62], [143.81, 1788.17], [140.41, 1787.72], [137.02, 1787.27], [133.62, 1786.82], [130.22, 1786.37], [126.83, 1785.92], [123.43, 1785.47], [120.03, 1785.02], [116.64, 1784.57], [113.24, 1784.12], [109.84, 1783.67], [106.45, 1783.21], [103.05, 1782.76], [99.66, 1782.31], [96.26, 1781.86], [92.86, 1781.41], [89.47, 1780.96], [86.07, 1780.51], [82.67, 1780.06], [79.28, 1779.61], [75.88, 1779.16], [72.48, 1778.71]], [[1267.64, 1846.0], [1267.66, 1842.28], [1267.67, 1838.56], [1267.69, 1834.84], [1267.7, 1831.12], [1267.72, 1827.41], [1267.73, 1823.69], [1267.75, 1819.97], [1267.76, 1816.25], [1267.78, 1812.53], [1267.8, 1808.81], [1267.81, 1805.09], [1267.83, 1801.37], [1267.84, 1797.65], [1267.86, 1793.94], [1267.87, 1790.22], [1267.89, 1786.5], [1267.91, 1782.78], [1267.92, 1779.06], [1267.94, 1775.34], [1267.95, 1771.62], [1267.97, 1767.9], [1267.98, 1764.18], [1268.0, 1760.47], [1268.02, 1756.75], [1268.03, 1753.03], [1268.05, 1749.31], [1268.06, 1745.59], [1268.08, 1741.87], [1268.09, 1738.15], [1268.11, 1734.43], [1268.13, 1730.71], [1268.14, 1727.0], [1268.16, 1723.28], [1268.17, 1719.56], [1268.19, 1715.84], [1268.2, 1712.12], [1268.22, 1708.4], [1268.23, 1704.68], [1268.25, 1700.96], [1268.27, 1697.25], [1268.42, 1693.53], [1268.71, 1689.82], [1269.14, 1686.13], [1269.7, 1682.45], [1270.4, 1678.8], [1271.24, 1675.18], [1272.21, 1671.59], [1273.31, 1668.03], [1274.54, 1664.52], [1275.9, 1661.06], [1277.39, 1657.66], [1279.0, 1654.3], [1280.74, 1651.02], [1282.58, 1647.78], [1284.49, 1644.59], [1286.5, 1641.46], [1288.54, 1638.35], [1290.56, 1635.23], [1292.57, 1632.1], [1294.59, 1628.98], [1296.6, 1625.85], [1298.6, 1622.72], [1300.62, 1619.59], [1302.62, 1616.46], [1304.6, 1613.31], [1306.55, 1610.14], [1308.47, 1606.96], [1310.37, 1603.76], [1312.28, 1600.57], [1314.2, 1597.39], [1316.12, 1594.2], [1318.03, 1591.01], [1319.97, 1587.84], [1321.89, 1584.65], [1323.9, 1581.52], [1326.02, 1578.47], [1328.19, 1575.45], [1330.35, 1572.42], [1332.41, 1569.32], [1334.34, 1566.15], [1336.16, 1562.9], [1337.86, 1559.6], [1339.43, 1556.23], [1340.88, 1552.8], [1342.2, 1549.32], [1343.39, 1545.8], [1344.45, 1542.23], [1345.37, 1538.63], [1346.17, 1535.0], [1346.82, 1531.34], [1347.34, 1527.66], [1347.73, 1523.96], [1347.97, 1520.25], [1348.08, 1516.53], [1348.05, 1512.81], [1347.89, 1509.1], [1347.58, 1505.39], [1347.14, 1501.7], [1346.56, 1498.02], [1345.85, 1494.37], [1345.0, 1490.75], [1344.02, 1487.16], [1342.91, 1483.62], [1341.66, 1480.11], [1340.29, 1476.66], [1338.79, 1473.25], [1337.16, 1469.91], [1335.42, 1466.63], [1333.55, 1463.41], [1331.56, 1460.27], [1329.46, 1457.2], [1327.25, 1454.21], [1324.92, 1451.31], [1322.49, 1448.49], [1319.96, 1445.77], [1317.33, 1443.14], [1314.61, 1440.61], [1311.79, 1438.18], [1308.88, 1435.86], [1305.89, 1433.65], [1302.82, 1431.55], [1299.68, 1429.57], [1296.46, 1427.7], [1293.18, 1425.95], [1289.83, 1424.33], [1286.43, 1422.83], [1282.97, 1421.46], [1279.46, 1420.22], [1275.91, 1419.11], [1272.33, 1418.13], [1268.7, 1417.29], [1265.05, 1416.58], [1261.38, 1416.0], [1257.69, 1415.56], [1253.98, 1415.26], [1250.26, 1415.1], [1246.55, 1415.07], [1242.83, 1415.18], [1239.12, 1415.41], [1235.41, 1415.77], [1231.73, 1416.26], [1228.06, 1416.88], [1224.41, 1417.61], [1220.79, 1418.42], [1217.18, 1419.34], [1213.61, 1420.38], [1210.07, 1421.5], [1206.54, 1422.69], [1203.05, 1423.97], [1199.58, 1425.32], [1196.14, 1426.73], [1192.75, 1428.26], [1189.39, 1429.84], [1186.04, 1431.46], [1182.7, 1433.08], [1179.39, 1434.78], [1176.13, 1436.58], [1172.95, 1438.5], [1169.81, 1440.5], [1166.69, 1442.52], [1163.58, 1444.56], [1160.49, 1446.63], [1157.42, 1448.73], [1154.36, 1450.85], [1151.35, 1453.02], [1148.32, 1455.19], [1145.3, 1457.35], [1142.29, 1459.54], [1139.31, 1461.76], [1136.24, 1463.86], [1133.1, 1465.86], [1129.89, 1467.73], [1126.61, 1469.49], [1123.27, 1471.12], [1119.87, 1472.62], [1116.42, 1474.0], [1112.91, 1475.26], [1109.37, 1476.38], [1105.78, 1477.36], [1102.16, 1478.22], [1098.51, 1478.94], [1094.84, 1479.52], [1091.15, 1479.97], [1087.44, 1480.28], [1083.73, 1480.46], [1080.01, 1480.49], [1076.29, 1480.39], [1072.58, 1480.15], [1068.88, 1479.77], [1065.2, 1479.26], [1061.54, 1478.61], [1057.9, 1477.83], [1054.3, 1476.91], [1050.73, 1475.85], [1047.21, 1474.67], [1043.73, 1473.36], [1040.3, 1471.92], [1036.93, 1470.35], [1033.61, 1468.66], [1030.37, 1466.85], [1027.19, 1464.92], [1024.08, 1462.87], [1021.06, 1460.71], [1018.11, 1458.43], [1015.26, 1456.05], [1012.49, 1453.57], [1009.81, 1450.99], [1007.24, 1448.3], [1004.76, 1445.53], [1002.39, 1442.66], [1000.13, 1439.71], [997.98, 1436.68], [995.94, 1433.57], [994.02, 1430.39], [992.21, 1427.13], [990.53, 1423.82], [988.98, 1420.44], [987.54, 1417.01], [986.24, 1413.52], [985.07, 1409.99], [984.03, 1406.42], [983.12, 1402.82], [982.35, 1399.18], [981.71, 1395.51], [981.21, 1391.83], [980.84, 1388.13], [980.61, 1384.42], [980.52, 1380.7], [980.57, 1376.98], [980.75, 1373.27], [981.08, 1369.56], [981.54, 1365.87], [982.13, 1362.2], [982.86, 1358.55], [983.73, 1354.94], [984.73, 1351.35], [985.86, 1347.81], [987.12, 1344.31], [988.51, 1340.86], [990.03, 1337.47], [991.67, 1334.13], [993.43, 1330.86], [995.32, 1327.65], [997.32, 1324.52], [999.44, 1321.46], [1001.67, 1318.48], [1004.0, 1315.59], [1006.45, 1312.79], [1008.98, 1310.07], [1011.62, 1307.44], [1014.35, 1304.92], [1017.17, 1302.5], [1020.09, 1300.19], [1023.08, 1297.98], [1026.15, 1295.89], [1029.3, 1293.91], [1032.52, 1292.04], [1035.8, 1290.3], [1039.13, 1288.64], [1042.46, 1286.97], [1045.82, 1285.38], [1049.22, 1283.87], [1052.67, 1282.48], [1056.14, 1281.15], [1059.64, 1279.91], [1063.17, 1278.73], [1066.69, 1277.52], [1070.21, 1276.33], [1073.74, 1275.17], [1077.3, 1274.09], [1080.89, 1273.11], [1084.46, 1272.08], [1088.05, 1271.08], [1091.66, 1270.2], [1095.27, 1269.31], [1098.86, 1268.33], [1102.41, 1267.24], [1105.94, 1266.08], [1109.45, 1264.84], [1112.94, 1263.56], [1116.44, 1262.29], [1119.92, 1260.98], [1123.38, 1259.62], [1126.82, 1258.2], [1130.21, 1256.68], [1133.59, 1255.12], [1136.94, 1253.51], [1140.27, 1251.85], [1143.58, 1250.17], [1146.87, 1248.43], [1150.14, 1246.65], [1153.4, 1244.86], [1156.64, 1243.03], [1159.86, 1241.17], [1163.06, 1239.29], [1166.26, 1237.38], [1169.43, 1235.45], [1172.61, 1233.51], [1175.81, 1231.63], [1179.06, 1229.81], [1182.36, 1228.11], [1185.73, 1226.53], [1189.16, 1225.08], [1192.61, 1223.69], [1196.08, 1222.37], [1199.6, 1221.15], [1203.12, 1219.97], [1206.67, 1218.86], [1210.23, 1217.79], [1213.81, 1216.75], [1217.37, 1215.71], [1220.91, 1214.54], [1224.41, 1213.3], [1227.9, 1212.0], [1231.37, 1210.67], [1234.81, 1209.25], [1238.23, 1207.8], [1241.65, 1206.33], [1245.05, 1204.84], [1248.44, 1203.31], [1251.81, 1201.73], [1255.16, 1200.11], [1258.5, 1198.49], [1261.83, 1196.83], [1265.15, 1195.15], [1268.44, 1193.41], [1271.72, 1191.67], [1275.04, 1189.99], [1278.41, 1188.41], [1281.8, 1186.89], [1285.22, 1185.42], [1288.63, 1183.95], [1292.04, 1182.45], [1295.45, 1180.97], [1298.87, 1179.52], [1302.3, 1178.06], [1305.7, 1176.58], [1309.11, 1175.07], [1312.51, 1173.59], [1315.95, 1172.17], [1319.41, 1170.79], [1322.85, 1169.39], [1326.29, 1167.98], [1329.74, 1166.59], [1333.18, 1165.17], [1336.61, 1163.74], [1340.05, 1162.31], [1343.49, 1160.9], [1346.93, 1159.5], [1350.37, 1158.09], [1353.82, 1156.69], [1357.27, 1155.3], [1360.72, 1153.92], [1364.22, 1152.65], [1367.73, 1151.44], [1371.27, 1150.3], [1374.82, 1149.18], [1378.4, 1148.17], [1382.0, 1147.25], [1385.62, 1146.4], [1389.26, 1145.63], [1392.92, 1144.96], [1396.59, 1144.38], [1400.28, 1143.88], [1403.96, 1143.38], [1407.65, 1142.93], [1411.35, 1142.5], [1415.05, 1142.15], [1418.75, 1141.81], [1422.46, 1141.49], [1426.17, 1141.19], [1429.87, 1140.83], [1433.56, 1140.43], [1437.25, 1139.96], [1440.94, 1139.5], [1444.63, 1139.03], [1448.32, 1138.56], [1452.01, 1138.05], [1455.69, 1137.56], [1459.37, 1137.02], [1463.04, 1136.43], [1466.69, 1135.71], [1470.31, 1134.85], [1473.9, 1133.87], [1477.44, 1132.74], [1480.94, 1131.49], [1484.4, 1130.11], [1487.8, 1128.6], [1491.14, 1126.97], [1494.42, 1125.22], [1497.63, 1123.34], [1500.77, 1121.35], [1503.83, 1119.24], [1506.82, 1117.02], [1509.72, 1114.69], [1512.53, 1112.26], [1515.25, 1109.72], [1517.87, 1107.08], [1520.39, 1104.35], [1522.81, 1101.53], [1525.13, 1098.62], [1527.33, 1095.62], [1529.43, 1092.55], [1531.4, 1089.4], [1533.26, 1086.18], [1535.0, 1082.89], [1536.62, 1079.54], [1538.11, 1076.13], [1539.47, 1072.67], [1540.7, 1069.17], [1541.81, 1065.62], [1542.78, 1062.03], [1543.62, 1058.4], [1544.32, 1054.75], [1544.89, 1051.07], [1545.32, 1047.38], [1545.61, 1043.67], [1545.77, 1039.96], [1545.79, 1036.24], [1545.67, 1032.52], [1545.45, 1028.81], [1545.14, 1025.1], [1544.71, 1021.41], [1544.15, 1017.73], [1543.45, 1014.08], [1542.62, 1010.46], [1541.65, 1006.86], [1540.55, 1003.31], [1539.32, 999.8], [1537.96, 996.34], [1536.48, 992.93], [1534.87, 989.58], [1533.14, 986.29], [1531.41, 982.99], [1529.66, 979.71], [1527.85, 976.46], [1526.04, 973.21], [1524.22, 969.97], [1522.4, 966.73], [1520.54, 963.51], [1518.57, 960.35], [1516.49, 957.27], [1514.34, 954.23], [1512.11, 951.26], [1509.84, 948.32], [1507.57, 945.37], [1505.24, 942.47], [1502.82, 939.65], [1500.34, 936.87], [1497.87, 934.1], [1495.37, 931.34], [1492.83, 928.63], [1490.24, 925.96], [1487.6, 923.34], [1484.97, 920.71], [1482.29, 918.13], [1479.56, 915.61], [1476.8, 913.12], [1474.0, 910.67], [1471.21, 908.2], [1468.4, 905.77], [1465.56, 903.37], [1462.7, 900.99], [1459.84, 898.61], [1456.96, 896.27], [1454.15, 893.82], [1451.33, 891.4], [1448.56, 888.92], [1445.89, 886.33], [1443.31, 883.65], [1440.81, 880.9], [1438.2, 878.25], [1435.5, 875.69], [1432.71, 873.23], [1429.83, 870.88], [1426.86, 868.64], [1423.81, 866.51], [1420.69, 864.49], [1417.49, 862.6], [1414.23, 860.82], [1410.9, 859.16], [1407.51, 857.63], [1404.07, 856.22], [1400.57, 854.94], [1397.04, 853.8], [1393.46, 852.78], [1389.84, 851.9], [1386.2, 851.15], [1382.53, 850.54], [1378.85, 850.04], [1375.15, 849.68], [1371.43, 849.46], [1367.72, 849.38], [1364.0, 849.42], [1360.28, 849.6], [1356.58, 849.91], [1352.88, 850.35], [1349.21, 850.92], [1345.56, 851.64], [1341.94, 852.47], [1338.34, 853.41], [1334.75, 854.38], [1331.19, 855.48], [1327.66, 856.65], [1324.15, 857.85], [1320.67, 859.16], [1317.23, 860.59], [1313.82, 862.07], [1310.43, 863.6], [1307.04, 865.12], [1303.65, 866.66], [1300.26, 868.18], [1296.83, 869.62], [1293.4, 871.05], [1289.94, 872.43], [1286.46, 873.74], [1282.98, 875.04], [1279.5, 876.35], [1276.01, 877.64], [1272.55, 879.01], [1269.09, 880.36], [1265.59, 881.64], [1262.07, 882.84], [1258.55, 884.02], [1254.99, 885.12], [1251.44, 886.21], [1247.88, 887.3], [1244.31, 888.33], [1240.74, 889.36], [1237.16, 890.36], [1233.58, 891.38], [1230.01, 892.41], [1226.44, 893.45], [1222.9, 894.61], [1219.41, 895.88], [1215.96, 897.28], [1212.55, 898.76], [1209.17, 900.31], [1205.83, 901.95], [1202.53, 903.66], [1199.26, 905.44], [1196.04, 907.3], [1192.86, 909.23], [1189.7, 911.18], [1186.57, 913.2], [1183.46, 915.23], [1180.31, 917.22], [1177.13, 919.14], [1173.88, 920.94], [1170.57, 922.65], [1167.21, 924.23], [1163.81, 925.75], [1160.38, 927.17], [1156.9, 928.5], [1153.39, 929.72], [1149.87, 930.92], [1146.37, 932.19], [1142.9, 933.52], [1139.44, 934.89], [1136.01, 936.31], [1132.58, 937.76], [1129.17, 939.24], [1125.78, 940.77], [1122.4, 942.32], [1119.03, 943.88], [1115.69, 945.52], [1112.4, 947.25], [1109.14, 949.05], [1105.93, 950.93], [1102.77, 952.9], [1099.66, 954.92], [1096.54, 956.94], [1093.42, 958.97], [1090.28, 960.97], [1087.14, 962.96], [1083.96, 964.9], [1080.76, 966.79], [1077.51, 968.59], [1074.21, 970.31], [1070.9, 972.01], [1067.62, 973.75], [1064.32, 975.47], [1060.98, 977.11], [1057.61, 978.68], [1054.21, 980.19], [1050.79, 981.64], [1047.36, 983.07], [1043.91, 984.46], [1040.49, 985.93], [1037.09, 987.44], [1033.75, 989.08], [1030.48, 990.83], [1027.23, 992.65], [1023.96, 994.42], [1020.68, 996.17], [1017.39, 997.91], [1014.07, 999.59], [1010.71, 1001.19], [1007.33, 1002.74], [1003.94, 1004.25], [1000.51, 1005.7], [997.06, 1007.07], [993.61, 1008.48], [990.2, 1009.97], [986.85, 1011.58], [983.54, 1013.27], [980.25, 1015.0], [976.96, 1016.73], [973.68, 1018.49], [970.4, 1020.23], [967.12, 1021.99], [963.84, 1023.74], [960.54, 1025.45], [957.25, 1027.2], [953.96, 1028.94], [950.64, 1030.6], [947.29, 1032.22], [943.93, 1033.82], [940.57, 1035.4], [937.23, 1037.04], [933.95, 1038.8], [930.73, 1040.65], [927.57, 1042.61], [924.47, 1044.66], [921.41, 1046.79], [918.4, 1048.96], [915.41, 1051.18], [912.45, 1053.43], [909.49, 1055.67], [906.45, 1057.82], [903.34, 1059.87], [900.18, 1061.82], [896.98, 1063.72], [893.76, 1065.58], [890.49, 1067.36], [887.21, 1069.1], [883.9, 1070.8], [880.56, 1072.44], [877.2, 1074.03], [873.83, 1075.59], [870.46, 1077.18], [867.13, 1078.82], [863.84, 1080.56], [860.62, 1082.41], [857.46, 1084.38], [854.35, 1086.42], [851.3, 1088.55], [848.3, 1090.74], [845.33, 1092.98], [842.39, 1095.27], [839.51, 1097.61], [836.65, 1099.99], [833.78, 1102.36], [830.91, 1104.72], [828.04, 1107.09], [825.23, 1109.52], [822.43, 1111.97], [819.65, 1114.44], [816.91, 1116.95], [814.16, 1119.45], [811.37, 1121.91], [808.57, 1124.36], [805.76, 1126.81], [802.87, 1129.14], [799.89, 1131.37], [796.87, 1133.53], [793.8, 1135.64], [790.66, 1137.63], [787.5, 1139.58], [784.29, 1141.47], [781.06, 1143.31], [777.85, 1145.19], [774.65, 1147.09], [771.5, 1149.05], [768.35, 1151.03], [765.21, 1153.03], [762.09, 1155.06], [758.99, 1157.1], [755.89, 1159.16], [752.81, 1161.25], [749.79, 1163.42], [746.75, 1165.55], [743.65, 1167.61], [740.5, 1169.59], [737.31, 1171.5], [734.06, 1173.31], [730.78, 1175.06], [727.48, 1176.77], [724.14, 1178.41], [720.77, 1179.98], [717.39, 1181.53], [714.01, 1183.1], [710.66, 1184.7], [707.28, 1186.26], [703.89, 1187.77], [700.47, 1189.24], [697.04, 1190.68], [693.62, 1192.14], [690.19, 1193.59], [686.74, 1194.98], [683.28, 1196.32], [679.84, 1197.74], [676.42, 1199.21], [673.0, 1200.67], [669.63, 1202.23], [666.31, 1203.91], [663.06, 1205.72], [659.87, 1207.63], [656.72, 1209.6], [653.6, 1211.63], [650.54, 1213.75], [647.55, 1215.95], [644.59, 1218.2], [641.68, 1220.52], [638.81, 1222.89], [636.0, 1225.32], [633.2, 1227.77], [630.37, 1230.18], [627.47, 1232.51], [624.49, 1234.73], [621.45, 1236.88], [618.35, 1238.93], [615.21, 1240.93], [612.06, 1242.9], [608.91, 1244.88], [605.79, 1246.9], [602.64, 1248.89], [599.47, 1250.83], [596.28, 1252.73], [593.1, 1254.66], [589.96, 1256.66], [586.89, 1258.76], [583.74, 1260.74], [580.52, 1262.6], [577.23, 1264.34], [573.88, 1265.95], [570.48, 1267.44], [567.02, 1268.81], [563.51, 1270.04], [559.96, 1271.15], [556.37, 1272.12], [552.75, 1272.96], [549.09, 1273.66], [545.42, 1274.23], [541.73, 1274.66], [538.02, 1274.96], [534.3, 1275.11], [530.58, 1275.13], [526.87, 1275.01], [523.16, 1274.76], [519.46, 1274.37], [515.78, 1273.84], [512.12, 1273.17], [508.49, 1272.37], [504.89, 1271.43], [501.33, 1270.36], [497.81, 1269.17], [494.33, 1267.84], [490.91, 1266.38], [487.54, 1264.8], [484.24, 1263.09], [481.0, 1261.27], [477.83, 1259.32], [474.74, 1257.26], [471.72, 1255.08], [468.79, 1252.8], [465.94, 1250.41], [463.18, 1247.91], [460.52, 1245.31], [457.95, 1242.62], [455.49, 1239.83], [453.13, 1236.96], [450.88, 1234.0], [448.75, 1230.96], [446.72, 1227.84], [444.81, 1224.64], [443.03, 1221.38], [441.36, 1218.06], [439.82, 1214.67], [438.4, 1211.23], [437.12, 1207.74], [435.96, 1204.21], [434.93, 1200.64], [434.04, 1197.03], [433.28, 1193.38], [432.66, 1189.72], [432.18, 1186.03], [431.83, 1182.33], [431.62, 1178.62], [431.54, 1174.9], [431.6, 1171.18], [431.79, 1167.46], [432.1, 1163.76], [432.52, 1160.06], [433.08, 1156.39], [433.77, 1152.73], [434.6, 1149.11], [435.53, 1145.51], [436.6, 1141.94], [437.8, 1138.42], [439.12, 1134.95], [440.58, 1131.53], [442.16, 1128.16], [443.86, 1124.86], [445.69, 1121.62], [447.59, 1118.42], [449.57, 1115.27], [451.58, 1112.14], [453.64, 1109.05], [455.74, 1105.98], [457.84, 1102.9], [459.95, 1099.85], [462.1, 1096.81], [464.33, 1093.83], [466.56, 1090.86], [468.74, 1087.85], [470.87, 1084.8], [472.98, 1081.74], [475.1, 1078.68], [477.21, 1075.62], [479.3, 1072.54], [481.36, 1069.44], [483.41, 1066.34], [485.4, 1063.2], [487.4, 1060.07], [489.41, 1056.94], [491.45, 1053.83], [493.46, 1050.7], [495.43, 1047.54], [497.38, 1044.37], [499.27, 1041.18], [501.16, 1037.97], [503.1, 1034.8], [505.16, 1031.7], [507.3, 1028.66], [509.51, 1025.67], [511.67, 1022.64], [513.8, 1019.59], [515.9, 1016.52], [517.94, 1013.42], [519.99, 1010.31], [522.13, 1007.27], [524.31, 1004.25], [526.58, 1001.31], [528.95, 998.45], [531.39, 995.64], [533.9, 992.89], [536.39, 990.13], [538.87, 987.36], [541.32, 984.56], [543.77, 981.76], [546.17, 978.93], [548.51, 976.04], [550.82, 973.12], [553.16, 970.23], [555.47, 967.31], [557.75, 964.38], [560.06, 961.47], [562.38, 958.56], [564.66, 955.62], [566.89, 952.64], [569.1, 949.65], [571.33, 946.68], [573.6, 943.73], [575.85, 940.77], [578.07, 937.78], [580.28, 934.79], [582.45, 931.77], [584.6, 928.74], [586.72, 925.69], [588.89, 922.66], [591.09, 919.67], [593.3, 916.67], [595.47, 913.65], [597.62, 910.62], [599.75, 907.57], [601.87, 904.52], [603.97, 901.44], [606.1, 898.4], [608.27, 895.38], [610.46, 892.37], [612.72, 889.42], [614.94, 886.43], [617.18, 883.47], [619.46, 880.53], [621.75, 877.6], [624.01, 874.64], [626.24, 871.66], [628.47, 868.69], [630.72, 865.73], [633.0, 862.79], [635.29, 859.86], [637.59, 856.94], [639.88, 854.01], [642.15, 851.06], [644.46, 848.15], [646.78, 845.24], [649.13, 842.36], [651.55, 839.54], [654.03, 836.76], [656.5, 833.99], [658.96, 831.2], [661.42, 828.4], [663.91, 825.65], [666.47, 822.94], [669.1, 820.32], [671.79, 817.76], [674.58, 815.29], [677.38, 812.84], [680.23, 810.46], [683.09, 808.08], [685.96, 805.71], [688.85, 803.37], [691.78, 801.09], [694.75, 798.85], [697.73, 796.62], [700.72, 794.41], [703.69, 792.17], [706.66, 789.93], [709.61, 787.66], [712.52, 785.35], [715.35, 782.94], [718.15, 780.49], [720.93, 778.02], [723.74, 775.58], [726.56, 773.16], [729.41, 770.78], [732.32, 768.46], [735.24, 766.15], [738.14, 763.82], [741.04, 761.5], [743.99, 759.23], [746.97, 757.01], [749.98, 754.83], [753.03, 752.69], [756.12, 750.62], [759.23, 748.58], [762.32, 746.51], [765.43, 744.48], [768.56, 742.46], [771.7, 740.48], [774.83, 738.47], [777.99, 736.51], [781.15, 734.54], [784.31, 732.59], [787.49, 730.65], [790.68, 728.75], [793.87, 726.84], [797.01, 724.84], [800.1, 722.77], [803.18, 720.69], [806.24, 718.57], [809.25, 716.39], [812.18, 714.11], [815.04, 711.72], [817.83, 709.26], [820.56, 706.74], [823.23, 704.15], [825.85, 701.51], [828.43, 698.83], [830.98, 696.12], [833.49, 693.38], [836.03, 690.66], [838.59, 687.97], [841.23, 685.35], [843.94, 682.8], [846.71, 680.32], [849.46, 677.82], [852.23, 675.33], [855.05, 672.92], [857.93, 670.55], [860.83, 668.23], [863.72, 665.9], [866.65, 663.59], [869.57, 661.29], [872.5, 659.0], [875.42, 656.7], [878.3, 654.35], [881.11, 651.92], [883.87, 649.42], [886.56, 646.85], [889.29, 644.32], [892.02, 641.8], [894.76, 639.29], [897.49, 636.76], [900.17, 634.19], [902.9, 631.66], [905.63, 629.14], [908.43, 626.69], [911.3, 624.32], [914.22, 622.02], [917.14, 619.71], [920.04, 617.39], [922.86, 614.97], [925.66, 612.51], [928.47, 610.08], [931.31, 607.68], [934.2, 605.33], [937.09, 603.0], [940.01, 600.7], [942.96, 598.43], [945.92, 596.18], [948.89, 593.94], [951.8, 591.62], [954.66, 589.25], [957.48, 586.82], [960.31, 584.42], [963.17, 582.03], [965.99, 579.61], [968.86, 577.25], [971.75, 574.9], [974.66, 572.59], [977.61, 570.32], [980.6, 568.1], [983.6, 565.91], [986.59, 563.7], [989.56, 561.47], [992.45, 559.12], [995.25, 556.68], [998.01, 554.18], [1000.71, 551.63], [1003.38, 549.04], [1006.07, 546.47], [1008.75, 543.89], [1011.35, 541.23], [1013.89, 538.52], [1016.37, 535.75], [1018.77, 532.91], [1021.12, 530.02], [1023.4, 527.08], [1025.69, 524.15], [1027.96, 521.2], [1030.22, 518.26], [1032.44, 515.27], [1034.61, 512.25], [1036.77, 509.23], [1039.01, 506.25], [1041.31, 503.33], [1043.71, 500.49], [1046.2, 497.73], [1048.79, 495.06], [1051.43, 492.44], [1054.14, 489.89], [1056.83, 487.33], [1059.54, 484.78], [1062.3, 482.29], [1065.12, 479.86], [1067.98, 477.49], [1070.89, 475.17], [1073.82, 472.88], [1076.78, 470.63], [1079.76, 468.4], [1082.78, 466.23], [1085.82, 464.09], [1088.82, 461.89], [1091.78, 459.64], [1094.72, 457.36], [1097.64, 455.06], [1100.52, 452.7], [1103.37, 450.31], [1106.15, 447.85], [1108.89, 445.33], [1111.62, 442.81], [1114.39, 440.32], [1117.2, 437.89], [1120.02, 435.47], [1122.86, 433.07], [1125.74, 430.71], [1128.64, 428.39], [1131.55, 426.07], [1134.37, 423.65], [1137.11, 421.13], [1139.74, 418.5], [1142.28, 415.78], [1144.71, 412.97], [1147.04, 410.07], [1149.26, 407.09], [1151.37, 404.03], [1153.36, 400.89], [1155.24, 397.67], [1156.99, 394.39], [1158.62, 391.05], [1160.13, 387.65], [1161.51, 384.2], [1162.76, 380.7], [1163.88, 377.15], [1164.87, 373.57], [1165.73, 369.95], [1166.45, 366.3], [1167.03, 362.63], [1167.48, 358.93], [1167.79, 355.23], [1167.96, 351.51], [1168.0, 347.79], [1167.9, 344.08], [1167.66, 340.37], [1167.28, 336.67], [1166.77, 332.98], [1166.12, 329.32], [1165.34, 325.69], [1164.42, 322.08], [1163.37, 318.52], [1162.18, 314.99], [1160.87, 311.51], [1159.43, 308.08], [1157.86, 304.71], [1156.17, 301.4], [1154.36, 298.15], [1152.43, 294.97], [1150.38, 291.87], [1148.22, 288.84], [1145.95, 285.9], [1143.57, 283.04], [1141.09, 280.27], [1138.5, 277.59], [1135.82, 275.02], [1133.04, 272.54], [1130.18, 270.17], [1127.23, 267.91], [1124.2, 265.76], [1121.08, 263.72], [1117.9, 261.8], [1114.65, 259.99], [1111.33, 258.31], [1107.95, 256.75], [1104.52, 255.32], [1101.04, 254.02], [1097.51, 252.85], [1093.94, 251.81], [1090.33, 250.9], [1086.69, 250.12], [1083.03, 249.49], [1079.35, 248.98], [1075.65, 248.62], [1071.93, 248.39], [1068.22, 248.3], [1064.5, 248.34], [1060.78, 248.53], [1057.08, 248.85], [1053.39, 249.31], [1049.72, 249.91], [1046.07, 250.64], [1042.45, 251.5], [1038.87, 252.5], [1035.33, 253.63], [1031.83, 254.89], [1028.38, 256.28], [1024.99, 257.8], [1021.65, 259.44], [1018.37, 261.21], [1015.17, 263.09], [1012.03, 265.09], [1008.98, 267.21], [1006.0, 269.44], [1003.11, 271.77], [1000.3, 274.22], [997.59, 276.76], [994.94, 279.37], [992.39, 282.07], [989.94, 284.87], [987.59, 287.76], [985.35, 290.73], [983.23, 293.78], [981.19, 296.89], [979.23, 300.05], [977.23, 303.19], [975.26, 306.34], [973.36, 309.54], [971.5, 312.76], [969.67, 315.99], [967.82, 319.22], [966.04, 322.49], [964.24, 325.74], [962.47, 329.01], [960.74, 332.3], [959.0, 335.59], [957.33, 338.91], [955.71, 342.26], [954.15, 345.64], [952.58, 349.01], [950.94, 352.35], [949.26, 355.66], [947.56, 358.97], [945.9, 362.3], [944.23, 365.62], [942.6, 368.97], [941.0, 372.32], [939.45, 375.7], [937.94, 379.1], [936.46, 382.51], [934.99, 385.93], [933.51, 389.34], [932.03, 392.75], [930.54, 396.16], [928.95, 399.52], [927.32, 402.86], [925.66, 406.19], [923.95, 409.49], [922.22, 412.79], [920.46, 416.06], [918.72, 419.35], [916.85, 422.56], [914.88, 425.72], [912.9, 428.87], [910.95, 432.03], [909.06, 435.23], [907.19, 438.45], [905.38, 441.7], [903.6, 444.97], [901.8, 448.22], [900.01, 451.48], [898.23, 454.75], [896.48, 458.03], [894.77, 461.33], [893.02, 464.61], [891.21, 467.86], [889.31, 471.05], [887.38, 474.23], [885.5, 477.44], [883.5, 480.57], [881.38, 483.63], [879.15, 486.61], [876.81, 489.5], [874.37, 492.31], [871.82, 495.02], [869.18, 497.63], [866.44, 500.15], [863.61, 502.56], [860.69, 504.87], [857.69, 507.06], [854.62, 509.16], [851.47, 511.14], [848.29, 513.06], [845.05, 514.9], [841.77, 516.64], [838.44, 518.29], [835.06, 519.84], [831.63, 521.28], [828.18, 522.68], [824.72, 524.04], [821.25, 525.38], [817.79, 526.73], [814.32, 528.08], [810.84, 529.38], [807.33, 530.62], [803.78, 531.73], [800.19, 532.72], [796.57, 533.57], [792.93, 534.29], [789.25, 534.87], [785.58, 535.45], [781.89, 535.95], [778.19, 536.32], [774.49, 536.62], [770.78, 536.87], [767.06, 537.07], [763.35, 537.23], [759.63, 537.3], [755.91, 537.28], [752.19, 537.16], [748.48, 537.0], [744.76, 536.84], [741.05, 536.65], [737.33, 536.46], [733.62, 536.27], [729.91, 536.03], [726.2, 535.72], [722.5, 535.36], [718.81, 534.95], [715.12, 534.48], [711.41, 534.13], [707.71, 533.83], [704.01, 533.48], [700.31, 533.05], [696.62, 532.61], [692.93, 532.17], [689.23, 531.73], [685.54, 531.27], [681.84, 530.91], [678.13, 530.7], [674.41, 530.52], [670.7, 530.44], [666.98, 530.41], [663.26, 530.4], [659.54, 530.43], [655.82, 530.43], [652.1, 530.38], [648.38, 530.34], [644.66, 530.32], [640.95, 530.23], [637.23, 530.18], [633.51, 530.2], [629.79, 530.36], [626.09, 530.65], [622.39, 531.06], [618.71, 531.57], [615.02, 532.07], [611.34, 532.57], [607.64, 532.95], [603.93, 533.29], [600.22, 533.48], [596.5, 533.57], [592.78, 533.56], [589.06, 533.49], [585.35, 533.36], [581.63, 533.19], [577.91, 533.06], [574.2, 532.99], [570.48, 532.94], [566.76, 532.89], [563.04, 532.86], [559.33, 532.69], [555.62, 532.42], [551.91, 532.11], [548.22, 531.68], [544.53, 531.19], [540.85, 530.62], [537.19, 529.99], [533.53, 529.35], [529.87, 528.66], [526.22, 527.99], [522.55, 527.36], [518.88, 526.73], [515.22, 526.1], [511.56, 525.42], [507.91, 524.74], [504.23, 524.16], [500.55, 523.64], [496.86, 523.17], [493.17, 522.75], [489.47, 522.35], [485.76, 522.04], [482.05, 521.82], [478.33, 521.68], [474.62, 521.58], [470.9, 521.5], [467.18, 521.45], [463.46, 521.45], [459.74, 521.4], [456.03, 521.25], [452.31, 521.04], [448.61, 520.74], [444.91, 520.36], [441.22, 519.91], [437.52, 519.47], [433.83, 519.07], [430.12, 518.74], [426.41, 518.49], [422.7, 518.22], [418.99, 518.0], [415.28, 517.71], [411.58, 517.37], [407.88, 517.04], [404.18, 516.65], [400.47, 516.31], [396.77, 515.99], [393.06, 515.76], [389.34, 515.59], [385.63, 515.44], [381.91, 515.33], [378.19, 515.28], [374.47, 515.24], [370.75, 515.22], [367.03, 515.24], [363.31, 515.29], [359.6, 515.36], [355.88, 515.46], [352.16, 515.51], [348.44, 515.58], [344.72, 515.67], [341.01, 515.78], [337.29, 515.86], [333.57, 515.89], [329.85, 515.9], [326.13, 515.81], [322.42, 515.61], [318.71, 515.29], [315.02, 514.88], [311.33, 514.37], [307.65, 513.86], [303.97, 513.35], [300.28, 512.87], [296.58, 512.47], [292.88, 512.16], [289.17, 511.92], [285.45, 511.71], [281.74, 511.54], [278.02, 511.42], [274.3, 511.3], [270.59, 511.16], [266.87, 511.06], [263.15, 510.98], [259.43, 510.92], [255.71, 510.9], [252.0, 510.88], [248.28, 510.86], [244.56, 510.87], [240.84, 510.88], [237.12, 510.92], [233.4, 510.99], [229.68, 511.11], [225.97, 511.24], [222.25, 511.22], [218.53, 511.21], [214.81, 511.05], [211.11, 510.76], [207.41, 510.33], [203.73, 509.8], [200.05, 509.26], [196.37, 508.73], [192.69, 508.2], [189.01, 507.66], [185.33, 507.13], [181.65, 506.6], [177.97, 506.07], [174.29, 505.53], [170.61, 505.0], [166.93, 504.47], [163.25, 503.94], [159.57, 503.4], [155.89, 502.87], [152.2, 502.34], [148.52, 501.8], [144.84, 501.27], [141.16, 500.74], [137.48, 500.21], [133.8, 499.67], [130.12, 499.14], [126.44, 498.61], [122.76, 498.07], [119.08, 497.54], [115.4, 497.01], [111.72, 496.48], [108.04, 495.94], [104.36, 495.41], [100.68, 494.88], [97.0, 494.35], [93.32, 493.81], [89.64, 493.28], [85.96, 492.75], [82.27, 492.21], [78.59, 491.68], [74.91, 491.15], [71.23, 490.62], [67.55, 490.08]], [[1846.0, 1094.88], [1842.67, 1094.32], [1839.35, 1093.76], [1836.02, 1093.2], [1832.69, 1092.64], [1829.36, 1092.08], [1826.04, 1091.52], [1822.71, 1090.96], [1819.38, 1090.4], [1816.06, 1089.84], [1812.73, 1089.28], [1809.4, 1088.72], [1806.07, 1088.16], [1802.75, 1087.6], [1799.42, 1087.04], [1796.09, 1086.48], [1792.76, 1085.92], [1789.44, 1085.36], [1786.11, 1084.8], [1782.78, 1084.24], [1779.46, 1083.68], [1776.13, 1083.12], [1772.8, 1082.56], [1769.47, 1082.0], [1766.15, 1081.44], [1762.82, 1080.88], [1759.49, 1080.32], [1756.17, 1079.76], [1752.84, 1079.2], [1749.51, 1078.64], [1746.18, 1078.08], [1742.86, 1077.52], [1739.53, 1076.96], [1736.2, 1076.4], [1732.88, 1075.84], [1729.55, 1075.28], [1726.22, 1074.72], [1722.89, 1074.16], [1719.57, 1073.6], [1716.24, 1073.04], [1712.91, 1072.48], [1709.58, 1071.92], [1706.26, 1071.36], [1702.93, 1070.8], [1699.6, 1070.24], [1696.26, 1069.81], [1692.9, 1069.51], [1689.53, 1069.33], [1686.15, 1069.29], [1682.78, 1069.37], [1679.41, 1069.57], [1676.05, 1069.9], [1672.71, 1070.36], [1669.39, 1070.94], [1666.08, 1071.62], [1662.8, 1072.38], [1659.53, 1073.21], [1656.27, 1074.1], [1653.06, 1075.12], [1649.88, 1076.26], [1646.75, 1077.53], [1643.68, 1078.91], [1640.65, 1080.41], [1637.69, 1082.02], [1634.79, 1083.75], [1631.95, 1085.57], [1629.18, 1087.5], [1626.46, 1089.49], [1623.8, 1091.56], [1621.2, 1093.71], [1618.67, 1095.95], [1616.2, 1098.25], [1613.74, 1100.55], [1611.33, 1102.92], [1608.96, 1105.32], [1606.59, 1107.72], [1604.25, 1110.15], [1601.91, 1112.59], [1599.62, 1115.06], [1597.35, 1117.56], [1595.14, 1120.1], [1592.92, 1122.65], [1590.76, 1125.24], [1588.6, 1127.83], [1586.38, 1130.37], [1584.13, 1132.88], [1581.81, 1135.34], [1579.46, 1137.76], [1577.07, 1140.14], [1574.67, 1142.5], [1572.21, 1144.82], [1569.7, 1147.08], [1567.11, 1149.23], [1564.45, 1151.31], [1561.71, 1153.28], [1558.93, 1155.2], [1556.11, 1157.05], [1553.25, 1158.84], [1550.41, 1160.66], [1547.61, 1162.54], [1544.85, 1164.47], [1542.13, 1166.47], [1539.43, 1168.5], [1536.77, 1170.58], [1534.16, 1172.72], [1531.59, 1174.89], [1529.04, 1177.11], [1526.54, 1179.37], [1524.06, 1181.66], [1521.59, 1183.95], [1519.08, 1186.21], [1516.55, 1188.45], [1514.07, 1190.73], [1511.61, 1193.04], [1509.16, 1195.37], [1506.72, 1197.69], [1504.28, 1200.02], [1501.85, 1202.37], [1499.42, 1204.7], [1496.99, 1207.04], [1494.56, 1209.39], [1492.16, 1211.76], [1489.74, 1214.11], [1487.23, 1216.36], [1484.63, 1218.52], [1481.96, 1220.57], [1479.21, 1222.53], [1476.42, 1224.43], [1473.63, 1226.32], [1470.78, 1228.14], [1467.89, 1229.88], [1464.97, 1231.56], [1462.03, 1233.23], [1459.11, 1234.92], [1456.21, 1236.64], [1453.31, 1238.35], [1450.36, 1240.0], [1447.39, 1241.6], [1444.4, 1243.16], [1441.38, 1244.66], [1438.33, 1246.11], [1435.31, 1247.62], [1432.27, 1249.08], [1429.21, 1250.49], [1426.13, 1251.88], [1423.05, 1253.24], [1419.97, 1254.63], [1416.93, 1256.09], [1413.91, 1257.6], [1410.9, 1259.12], [1407.9, 1260.66], [1404.94, 1262.28], [1402.02, 1263.97], [1399.12, 1265.7], [1396.24, 1267.45], [1393.33, 1269.16], [1390.41, 1270.85], [1387.46, 1272.49], [1384.56, 1274.22], [1381.71, 1276.02], [1378.87, 1277.85], [1376.04, 1279.69], [1373.2, 1281.5], [1370.34, 1283.29], [1367.49, 1285.1], [1364.64, 1286.91], [1361.82, 1288.75], [1359.04, 1290.67], [1356.33, 1292.68], [1353.7, 1294.79], [1351.12, 1296.97], [1348.62, 1299.23], [1346.18, 1301.56], [1343.79, 1303.95], [1341.45, 1306.38], [1339.11, 1308.81], [1336.83, 1311.29], [1334.61, 1313.83], [1332.44, 1316.42], [1330.3, 1319.02], [1328.17, 1321.64], [1326.02, 1324.24], [1323.86, 1326.84], [1321.72, 1329.45], [1319.63, 1332.09], [1317.51, 1334.71], [1315.37, 1337.33], [1313.2, 1339.91], [1311.02, 1342.48], [1308.84, 1345.06], [1306.62, 1347.6], [1304.32, 1350.07], [1301.95, 1352.47], [1299.51, 1354.8], [1296.99, 1357.04], [1294.38, 1359.18], [1291.72, 1361.25], [1289.0, 1363.25], [1286.23, 1365.18], [1283.4, 1367.02], [1280.52, 1368.78], [1277.65, 1370.55], [1274.71, 1372.2], [1271.7, 1373.74], [1268.65, 1375.16], [1265.53, 1376.47], [1262.38, 1377.65], [1259.17, 1378.72], [1255.93, 1379.66], [1252.66, 1380.47], [1249.36, 1381.16], [1246.03, 1381.72], [1242.68, 1382.16], [1239.32, 1382.46], [1235.95, 1382.64], [1232.58, 1382.69], [1229.21, 1382.61], [1225.84, 1382.39], [1222.48, 1382.05], [1219.14, 1381.59], [1215.82, 1380.99], [1212.52, 1380.27], [1209.26, 1379.42], [1206.03, 1378.45], [1202.84, 1377.35], [1199.69, 1376.14], [1196.59, 1374.8], [1193.55, 1373.34], [1190.56, 1371.77], [1187.64, 1370.09], [1184.78, 1368.3], [1181.99, 1366.4], [1179.28, 1364.39], [1176.65, 1362.28], [1174.1, 1360.07], [1171.63, 1357.76], [1169.26, 1355.37], [1166.98, 1352.88], [1164.8, 1350.31], [1162.71, 1347.65], [1160.73, 1344.92], [1158.86, 1342.12], [1157.1, 1339.24], [1155.38, 1336.33], [1153.75, 1333.38], [1152.12, 1330.42], [1150.59, 1327.42], [1149.12, 1324.38], [1147.75, 1321.3], [1146.46, 1318.18], [1145.29, 1315.02], [1144.22, 1311.82], [1143.24, 1308.59], [1142.3, 1305.35], [1141.48, 1302.08], [1140.77, 1298.78], [1140.17, 1295.46], [1139.6, 1292.13], [1139.07, 1288.8], [1138.56, 1285.46], [1138.05, 1282.13], [1137.56, 1278.79], [1137.06, 1275.45], [1136.56, 1272.12], [1136.1, 1268.77], [1135.62, 1265.43], [1135.17, 1262.09], [1134.71, 1258.75], [1134.23, 1255.41], [1133.79, 1252.06], [1133.34, 1248.72], [1132.91, 1245.37], [1132.56, 1242.02], [1132.24, 1238.66], [1131.91, 1235.3], [1131.57, 1231.94], [1131.16, 1228.6], [1130.66, 1225.26], [1130.12, 1221.93], [1129.53, 1218.61], [1128.9, 1215.29], [1128.2, 1211.99], [1127.43, 1208.71], [1126.62, 1205.43], [1125.74, 1202.17], [1124.81, 1198.93], [1123.83, 1195.7], [1122.77, 1192.5], [1121.72, 1189.29], [1120.68, 1186.08], [1119.61, 1182.88], [1118.59, 1179.67], [1117.58, 1176.45], [1116.6, 1173.22], [1115.64, 1169.98], [1114.66, 1166.75], [1113.67, 1163.53], [1112.69, 1160.3], [1111.73, 1157.07], [1110.76, 1153.83], [1109.82, 1150.6], [1108.96, 1147.33], [1108.17, 1144.05], [1107.37, 1140.77], [1106.54, 1137.51], [1105.7, 1134.24], [1104.81, 1130.98], [1103.91, 1127.73], [1103.04, 1124.47], [1102.25, 1121.19], [1101.5, 1117.9], [1100.77, 1114.61], [1100.14, 1111.29], [1099.6, 1107.96], [1099.03, 1104.64], [1098.44, 1101.31], [1097.8, 1098.0], [1097.1, 1094.7], [1096.35, 1091.41], [1095.55, 1088.13], [1094.72, 1084.86], [1093.87, 1081.6], [1093.01, 1078.34], [1092.12, 1075.08], [1091.23, 1071.83], [1090.32, 1068.58], [1089.39, 1065.34], [1088.46, 1062.09], [1087.49, 1058.86], [1086.51, 1055.63], [1085.51, 1052.41], [1084.49, 1049.19], [1083.48, 1045.98], [1082.47, 1042.76], [1081.48, 1039.53], [1080.48, 1036.31], [1079.52, 1033.07], [1078.56, 1029.84], [1077.61, 1026.6], [1076.64, 1023.37], [1075.66, 1020.14], [1074.7, 1016.91], [1073.79, 1013.66], [1072.93, 1010.4], [1072.09, 1007.13], [1071.31, 1003.84], [1070.6, 1000.54], [1070.02, 997.22], [1069.57, 993.88], [1069.23, 990.52], [1068.99, 987.16], [1068.83, 983.79], [1068.68, 980.41], [1068.5, 977.05], [1068.33, 973.68], [1068.17, 970.31], [1067.97, 966.94], [1067.84, 963.57], [1067.7, 960.19], [1067.59, 956.82], [1067.51, 953.45], [1067.49, 950.08], [1067.57, 946.7], [1067.72, 943.33], [1067.94, 939.97], [1068.22, 936.6], [1068.57, 933.25], [1068.86, 929.89], [1069.22, 926.53], [1069.61, 923.18], [1069.99, 919.83], [1070.37, 916.47], [1070.78, 913.13], [1071.17, 909.77], [1071.51, 906.42], [1071.77, 903.05], [1071.99, 899.69], [1072.11, 896.31], [1072.16, 892.94], [1072.09, 889.57], [1071.96, 886.2], [1071.74, 882.83], [1071.46, 879.47], [1071.13, 876.11], [1070.8, 872.75], [1070.42, 869.4], [1069.99, 866.05], [1069.52, 862.71], [1069.0, 859.38], [1068.44, 856.05], [1067.94, 852.71], [1067.5, 849.37], [1067.12, 846.02], [1066.8, 842.66], [1066.62, 839.29], [1066.54, 835.91], [1066.5, 832.54], [1066.44, 829.17], [1066.36, 825.79], [1066.26, 822.42], [1066.08, 819.05], [1065.84, 815.69], [1065.55, 812.33], [1065.29, 808.96], [1065.02, 805.6], [1064.75, 802.24], [1064.44, 798.88], [1064.09, 795.52], [1063.61, 792.18], [1063.0, 788.86], [1062.27, 785.57], [1061.41, 782.31], [1060.42, 779.08], [1059.32, 775.89], [1058.09, 772.75], [1056.74, 769.66], [1055.28, 766.62], [1053.7, 763.64], [1052.0, 760.72], [1050.2, 757.87], [1048.29, 755.09], [1046.27, 752.38], [1044.15, 749.76], [1041.93, 747.21], [1039.62, 744.76], [1037.21, 742.39], [1034.72, 740.12], [1032.14, 737.95], [1029.48, 735.87], [1026.74, 733.9], [1023.92, 732.04], [1021.04, 730.29], [1018.09, 728.64], [1015.09, 727.11], [1012.02, 725.7], [1008.91, 724.41], [1005.74, 723.23], [1002.54, 722.18], [999.29, 721.25], [996.02, 720.45], [992.71, 719.77], [989.38, 719.22], [986.04, 718.8], [982.67, 718.51], [979.31, 718.34], [975.93, 718.3], [972.56, 718.4], [969.19, 718.62], [965.84, 718.97], [962.5, 719.45], [959.18, 720.06], [955.88, 720.79], [952.62, 721.65], [949.39, 722.64], [946.21, 723.74], [943.06, 724.97], [939.97, 726.32], [936.93, 727.78], [933.95, 729.36], [931.03, 731.06], [928.18, 732.86], [925.4, 734.77], [922.7, 736.79], [920.07, 738.91], [917.53, 741.13], [915.07, 743.44], [912.71, 745.85], [910.44, 748.34], [908.26, 750.92], [906.19, 753.58], [904.22, 756.32], [902.35, 759.13], [900.6, 762.02], [898.96, 764.96], [897.43, 767.97], [896.01, 771.03], [894.72, 774.15], [893.55, 777.31], [892.49, 780.52], [891.56, 783.76], [890.76, 787.04], [890.07, 790.34], [889.51, 793.67], [889.08, 797.02], [888.78, 800.38], [888.59, 803.75], [888.53, 807.12], [888.54, 810.49], [888.66, 813.87], [888.84, 817.23], [889.15, 820.59], [889.58, 823.94], [890.14, 827.27], [890.83, 830.57], [891.63, 833.85], [892.5, 837.11], [893.43, 840.35], [894.38, 843.59], [895.34, 846.82], [896.3, 850.06], [897.39, 853.25], [898.59, 856.4], [899.82, 859.54], [901.07, 862.68], [902.33, 865.81], [903.64, 868.92], [904.99, 872.01], [906.38, 875.09], [907.89, 878.1], [909.5, 881.07], [911.19, 883.99], [912.96, 886.86], [914.8, 889.69], [916.7, 892.47], [918.64, 895.24], [920.62, 897.97], [922.62, 900.69], [924.54, 903.46], [926.4, 906.28], [928.24, 909.11], [930.08, 911.93], [931.96, 914.74], [933.91, 917.49], [935.89, 920.22], [937.87, 922.95], [939.9, 925.65], [941.93, 928.34], [943.94, 931.05], [945.87, 933.82], [947.73, 936.64], [949.54, 939.48], [951.29, 942.37], [952.99, 945.28], [954.65, 948.22], [956.33, 951.14], [958.05, 954.04], [959.85, 956.9], [961.66, 959.74], [963.54, 962.55], [965.43, 965.34], [967.36, 968.11], [969.29, 970.88], [971.29, 973.6], [973.26, 976.34], [975.21, 979.09], [977.15, 981.85], [979.03, 984.65], [980.88, 987.47], [982.68, 990.33], [984.42, 993.21], [986.14, 996.12], [987.83, 999.04], [989.48, 1001.98], [991.12, 1004.93], [992.79, 1007.86], [994.44, 1010.8], [996.06, 1013.77], [997.64, 1016.74], [999.2, 1019.74], [1000.87, 1022.67], [1002.66, 1025.53], [1004.55, 1028.33], [1006.54, 1031.05], [1008.64, 1033.69], [1010.84, 1036.25], [1013.14, 1038.72], [1015.53, 1041.1], [1018.0, 1043.39], [1020.57, 1045.59], [1023.21, 1047.68], [1025.94, 1049.67], [1028.74, 1051.55], [1031.61, 1053.33], [1034.54, 1055.0], [1037.54, 1056.55], [1040.59, 1057.98], [1043.7, 1059.3], [1046.85, 1060.5], [1050.05, 1061.57], [1053.28, 1062.53], [1056.56, 1063.35], [1059.86, 1064.06], [1063.18, 1064.63], [1066.52, 1065.08], [1069.88, 1065.4], [1073.25, 1065.59], [1076.63, 1065.65], [1080.0, 1065.58], [1083.37, 1065.38], [1086.72, 1065.06], [1090.07, 1064.6], [1093.39, 1064.02], [1096.69, 1063.31], [1099.96, 1062.48], [1103.19, 1061.52], [1106.39, 1060.43], [1109.54, 1059.23], [1112.64, 1057.9], [1115.69, 1056.46], [1118.69, 1054.9], [1121.62, 1053.23], [1124.48, 1051.45], [1127.28, 1049.56], [1130.0, 1047.56], [1132.64, 1045.46], [1135.2, 1043.26], [1137.67, 1040.97], [1140.05, 1038.58], [1142.34, 1036.1], [1144.54, 1033.54], [1146.63, 1030.89], [1148.62, 1028.17], [1150.51, 1025.37], [1152.28, 1022.5], [1153.95, 1019.57], [1155.5, 1016.57], [1156.93, 1013.52], [1158.25, 1010.41], [1159.45, 1007.26], [1160.53, 1004.06], [1161.48, 1000.82], [1162.31, 997.55], [1163.01, 994.25], [1163.59, 990.93], [1164.03, 987.58], [1164.35, 984.23], [1164.54, 980.86], [1164.6, 977.48], [1164.54, 974.11], [1164.34, 970.74], [1164.01, 967.38], [1163.56, 964.04], [1162.98, 960.72], [1162.27, 957.42], [1161.44, 954.15], [1160.49, 950.91], [1159.41, 947.71], [1158.21, 944.56], [1156.89, 941.45], [1155.48, 938.39], [1153.94, 935.39], [1152.3, 932.44], [1150.58, 929.54], [1148.8, 926.67], [1146.97, 923.84], [1145.19, 920.97], [1143.43, 918.09], [1141.73, 915.18], [1140.06, 912.24], [1138.45, 909.28], [1136.84, 906.31], [1135.27, 903.33], [1133.75, 900.32], [1132.2, 897.32], [1130.6, 894.35], [1129.03, 891.36], [1127.45, 888.38], [1125.89, 885.39], [1124.29, 882.42], [1122.57, 879.51], [1120.79, 876.65], [1118.92, 873.84], [1116.99, 871.08], [1114.98, 868.36], [1112.94, 865.68], [1110.93, 862.96], [1108.98, 860.21], [1107.09, 857.42], [1105.21, 854.62], [1103.3, 851.84], [1101.37, 849.07], [1099.46, 846.29], [1097.55, 843.5], [1095.69, 840.69], [1093.82, 837.89], [1091.98, 835.05], [1090.18, 832.2], [1088.37, 829.35], [1086.49, 826.55], [1084.55, 823.79], [1082.55, 821.07], [1080.49, 818.4], [1078.38, 815.77], [1076.2, 813.2], [1073.94, 810.69], [1071.61, 808.25], [1069.18, 805.91], [1066.67, 803.66], [1064.07, 801.5], [1061.39, 799.45], [1058.64, 797.5], [1055.81, 795.66], [1052.91, 793.93], [1049.98, 792.27], [1047.01, 790.67], [1044.0, 789.14], [1040.93, 787.72], [1037.82, 786.43], [1034.69, 785.17], [1031.55, 783.94], [1028.37, 782.8], [1025.17, 781.75], [1021.94, 780.75], [1018.71, 779.79], [1015.45, 778.91], [1012.17, 778.12], [1008.89, 777.34], [1005.6, 776.59], [1002.3, 775.9], [998.99, 775.25], [995.65, 774.73], [992.3, 774.33], [988.94, 774.07], [985.57, 773.93], [982.19, 773.92], [978.82, 774.02], [975.45, 774.21], [972.09, 774.52], [968.75, 774.95], [965.4, 775.42], [962.08, 775.98], [958.76, 776.6], [955.45, 777.28], [952.16, 778.0], [948.87, 778.73], [945.58, 779.49], [942.29, 780.26], [939.01, 781.03], [935.71, 781.74], [932.39, 782.33], [929.05, 782.8], [925.7, 783.19], [922.34, 783.56], [918.98, 783.87], [915.62, 784.1], [912.25, 784.26], [908.87, 784.33], [905.5, 784.38], [902.13, 784.46], [898.75, 784.53], [895.38, 784.6], [892.01, 784.71], [888.64, 784.84], [885.27, 785.04], [881.9, 785.31], [878.54, 785.61], [875.18, 785.91], [871.83, 786.24], [868.48, 786.66], [865.14, 787.12], [861.8, 787.65], [858.47, 788.15], [855.14, 788.7], [851.82, 789.3], [848.51, 789.95], [845.21, 790.66], [841.91, 791.37], [838.63, 792.14], [835.35, 792.94], [832.06, 793.71], [828.79, 794.54], [825.53, 795.42], [822.28, 796.32], [819.04, 797.24], [815.79, 798.15], [812.51, 798.97], [809.21, 799.65], [805.89, 800.23], [802.54, 800.69], [799.22, 801.27], [795.92, 801.99], [792.66, 802.83], [789.42, 803.79], [786.2, 804.79], [783.0, 805.87], [779.84, 807.05], [776.7, 808.28], [773.59, 809.58], [770.49, 810.92], [767.41, 812.3], [764.34, 813.7], [761.3, 815.17], [758.29, 816.68], [755.33, 818.31], [752.4, 819.98], [749.51, 821.72], [746.63, 823.48], [743.79, 825.3], [740.98, 827.16], [738.2, 829.08], [735.46, 831.05], [732.8, 833.12], [730.21, 835.29], [727.71, 837.55], [725.28, 839.89], [722.92, 842.31], [720.62, 844.78], [718.37, 847.29], [716.11, 849.79], [713.88, 852.32], [711.65, 854.85], [709.4, 857.37], [707.19, 859.92], [705.03, 862.51], [702.83, 865.07], [700.65, 867.64], [698.44, 870.2], [696.18, 872.7], [693.9, 875.18], [691.59, 877.64], [689.29, 880.11], [687.02, 882.61], [684.77, 885.13], [682.5, 887.62], [680.24, 890.12], [678.02, 892.66], [675.81, 895.22], [673.65, 897.81], [671.6, 900.48], [669.64, 903.23], [667.71, 906.0], [665.82, 908.8], [663.94, 911.6], [662.09, 914.42], [660.3, 917.28], [658.55, 920.17], [656.82, 923.06], [655.15, 925.99], [653.46, 928.91], [651.76, 931.82], [650.06, 934.74], [648.39, 937.67], [646.7, 940.59], [644.99, 943.5], [643.31, 946.43], [641.67, 949.37], [640.03, 952.32], [638.41, 955.28], [636.84, 958.27], [635.33, 961.29], [633.86, 964.33], [632.4, 967.37], [630.97, 970.42], [629.58, 973.5], [628.24, 976.59], [626.87, 979.68], [625.53, 982.77], [624.19, 985.87], [622.86, 988.97], [621.58, 992.09], [620.29, 995.21], [618.98, 998.32], [617.68, 1001.43], [616.39, 1004.55], [615.06, 1007.65], [613.73, 1010.75], [612.36, 1013.84], [610.93, 1016.89], [609.45, 1019.92], [607.87, 1022.91], [606.19, 1025.83], [604.39, 1028.69], [602.49, 1031.47], [600.59, 1034.26], [598.77, 1037.1], [596.99, 1039.97], [595.24, 1042.85], [593.5, 1045.75], [591.76, 1048.63], [590.01, 1051.52], [588.28, 1054.42], [586.58, 1057.33], [584.81, 1060.2], [583.02, 1063.06], [581.17, 1065.88], [579.21, 1068.63], [577.18, 1071.33], [575.09, 1073.97], [572.89, 1076.53], [570.63, 1079.04], [568.35, 1081.53], [566.1, 1084.04], [563.83, 1086.54], [561.46, 1088.94], [559.01, 1091.25], [556.47, 1093.47], [553.84, 1095.59], [551.13, 1097.61], [548.35, 1099.52], [545.5, 1101.32], [542.58, 1103.01], [539.6, 1104.59], [536.56, 1106.06], [533.47, 1107.42], [530.34, 1108.66], [527.15, 1109.78], [523.93, 1110.78], [520.67, 1111.65], [517.38, 1112.4], [514.07, 1113.02], [510.73, 1113.52], [507.38, 1113.89], [504.01, 1114.16], [500.64, 1114.3], [497.27, 1114.41], [493.9, 1114.46], [490.52, 1114.44], [487.15, 1114.38], [483.78, 1114.26], [480.4, 1114.18], [477.03, 1114.12], [473.66, 1114.15], [470.28, 1114.23], [466.91, 1114.34], [463.54, 1114.55], [460.18, 1114.78], [456.81, 1114.96], [453.44, 1115.16], [450.08, 1115.42], [446.72, 1115.76], [443.37, 1116.16], [440.02, 1116.59], [436.68, 1117.06], [433.35, 1117.57], [430.0, 1117.99], [426.65, 1118.38], [423.29, 1118.72], [419.94, 1119.09], [416.59, 1119.51], [413.24, 1119.95], [409.91, 1120.44], [406.56, 1120.91], [403.22, 1121.38], [399.88, 1121.85], [396.54, 1122.3], [393.19, 1122.74], [389.85, 1123.17], [386.51, 1123.65], [383.17, 1124.17], [379.83, 1124.63], [376.48, 1124.99], [373.11, 1125.23], [369.74, 1125.43], [366.37, 1125.57], [363.0, 1125.73], [359.63, 1125.94], [356.27, 1126.15], [352.9, 1126.34], [349.53, 1126.53], [346.16, 1126.72], [342.79, 1126.88], [339.42, 1127.04], [336.05, 1127.18], [332.68, 1127.34], [329.31, 1127.48], [325.94, 1127.54], [322.56, 1127.55], [319.19, 1127.48], [315.82, 1127.37], [312.45, 1127.19], [309.08, 1126.97], [305.71, 1126.75], [302.34, 1126.56], [298.98, 1126.38], [295.61, 1126.21], [292.24, 1126.05], [288.86, 1125.96], [285.49, 1125.95], [282.12, 1126.0], [278.74, 1126.11], [275.37, 1126.3], [272.01, 1126.55], [268.65, 1126.87], [265.3, 1127.23], [261.94, 1127.54], [258.57, 1127.82], [255.21, 1128.06], [251.84, 1128.24], [248.47, 1128.37], [245.09, 1128.42], [241.72, 1128.42], [238.35, 1128.35], [234.97, 1128.24], [231.6, 1128.1], [228.23, 1127.92], [224.86, 1127.79], [221.49, 1127.71], [218.12, 1127.75], [214.74, 1127.87], [211.38, 1128.05], [208.01, 1128.23], [204.63, 1128.35], [201.26, 1128.39], [197.89, 1128.47], [194.51, 1128.52], [191.14, 1128.55], [187.77, 1128.58], [184.39, 1128.52], [181.02, 1128.44], [177.65, 1128.34], [174.27, 1128.27], [170.9, 1128.24], [167.53, 1128.17], [164.15, 1128.13], [160.78, 1128.2], [157.41, 1128.33], [154.04, 1128.46], [150.66, 1128.56], [147.29, 1128.66], [143.92, 1128.85], [140.56, 1129.08], [137.18, 1129.17], [133.81, 1129.14], [130.44, 1128.98], [127.07, 1128.78], [123.7, 1128.59], [120.33, 1128.4], [116.97, 1128.21], [113.6, 1128.02], [110.23, 1127.83], [106.86, 1127.64], [103.49, 1127.45], [100.12, 1127.26], [96.75, 1127.07], [93.39, 1126.87], [90.02, 1126.68], [86.65, 1126.49], [83.28, 1126.3], [79.91, 1126.11], [76.54, 1125.92], [73.17, 1125.73], [69.81, 1125.54]]], "image": "images/traverse_ordering_00003.png"} +{"width": 1916, "height": 1916, "num_extra_distractor_curves": 1, "s_curve": 0, "s_at_start": false, "s_point": {"x": 432.14, "y": 1846.13, "label_x": 380.14, "label_y": 1846.67}, "marks": [{"label": "L", "curve": 0, "on_s_curve": true, "polyline_index": 166, "x": 905.16, "y": 326.68, "label_x": 888.87, "label_y": 371.83}, {"label": "B", "curve": 0, "on_s_curve": true, "polyline_index": 449, "x": 1310.96, "y": 695.6, "label_x": 1328.49, "label_y": 740.28}, {"label": "C", "curve": 0, "on_s_curve": true, "polyline_index": 682, "x": 697.79, "y": 970.6, "label_x": 669.67, "label_y": 1009.5}, {"label": "A", "curve": 0, "on_s_curve": true, "polyline_index": 866, "x": 742.07, "y": 1150.41, "label_x": 706.58, "label_y": 1118.09}, {"label": "F", "curve": 0, "on_s_curve": true, "polyline_index": 1048, "x": 1231.46, "y": 1344.01, "label_x": 1221.21, "label_y": 1390.9}, {"label": "D", "curve": 0, "on_s_curve": true, "polyline_index": 1414, "x": 498.18, "y": 1406.56, "label_x": 519.22, "label_y": 1449.7}, {"label": "J", "curve": 1, "on_s_curve": false, "polyline_index": 226, "x": 1139.76, "y": 670.96, "label_x": 1128.68, "label_y": 624.26}, {"label": "E", "curve": 1, "on_s_curve": false, "polyline_index": 456, "x": 1267.02, "y": 1288.96, "label_x": 1307.63, "label_y": 1263.38}, {"label": "K", "curve": 1, "on_s_curve": false, "polyline_index": 699, "x": 994.34, "y": 1170.81, "label_x": 1029.75, "label_y": 1138.4}, {"label": "H", "curve": 1, "on_s_curve": false, "polyline_index": 873, "x": 630.29, "y": 676.49, "label_x": 670.2, "label_y": 649.82}, {"label": "G", "curve": 1, "on_s_curve": false, "polyline_index": 1038, "x": 389.24, "y": 762.52, "label_x": 346.58, "label_y": 784.53}, {"label": "I", "curve": 1, "on_s_curve": false, "polyline_index": 1240, "x": 367.67, "y": 1312.73, "label_x": 392.43, "label_y": 1271.61}], "num_marks_total": 12, "num_marks_on_s": 6, "num_pair_crossings": 9, "num_self_crossings": 0, "question": "The image shows two curves or strands that cross each other. Several labeled points are marked on the two curves combined. A green 'S' marks the start endpoint of one of the curves. Starting from S, follow THAT curve (the one S lies on) all the way to its other endpoint, and list the labels of the marked points you visit in order, separated by commas. Ignore any labeled points that lie on the OTHER curve. Provide your final answer enclosed in ... tags.", "answer": "D, F, A, C, B, L", "polylines": [[[527.45, 70.0], [527.61, 73.4], [527.77, 76.8], [527.93, 80.2], [528.09, 83.6], [528.24, 87.0], [528.4, 90.4], [528.56, 93.8], [528.72, 97.2], [528.88, 100.6], [529.04, 104.0], [529.2, 107.4], [529.36, 110.81], [529.52, 114.21], [529.68, 117.61], [529.83, 121.01], [529.99, 124.41], [530.15, 127.81], [530.31, 131.21], [530.47, 134.61], [530.63, 138.01], [530.79, 141.41], [530.95, 144.81], [531.11, 148.21], [531.27, 151.61], [531.42, 155.01], [531.58, 158.41], [531.74, 161.81], [531.9, 165.21], [532.06, 168.61], [532.22, 172.01], [532.38, 175.41], [532.54, 178.81], [532.7, 182.21], [532.86, 185.61], [533.01, 189.02], [533.17, 192.42], [533.33, 195.82], [533.49, 199.22], [533.65, 202.62], [533.81, 206.02], [533.97, 209.42], [534.13, 212.82], [534.29, 216.22], [534.45, 219.62], [534.75, 223.01], [535.2, 226.38], [535.8, 229.73], [536.55, 233.06], [537.43, 236.34], [538.46, 239.59], [539.63, 242.79], [540.93, 245.93], [542.37, 249.02], [543.94, 252.03], [545.65, 254.98], [547.47, 257.86], [549.42, 260.65], [551.49, 263.35], [553.68, 265.96], [555.97, 268.47], [558.37, 270.89], [560.88, 273.19], [563.48, 275.39], [566.17, 277.47], [568.95, 279.43], [571.82, 281.27], [574.76, 282.99], [577.77, 284.57], [580.85, 286.03], [583.99, 287.35], [587.18, 288.53], [590.4, 289.61], [593.66, 290.61], [596.95, 291.48], [600.27, 292.24], [603.6, 292.94], [606.93, 293.65], [610.24, 294.43], [613.56, 295.21], [616.86, 296.01], [620.17, 296.81], [623.47, 297.67], [626.75, 298.57], [630.04, 299.47], [633.33, 300.33], [636.62, 301.19], [639.91, 302.07], [643.18, 303.03], [646.45, 303.95], [649.74, 304.82], [653.02, 305.73], [656.3, 306.66], [659.58, 307.58], [662.86, 308.49], [666.13, 309.44], [669.39, 310.4], [672.67, 311.32], [675.95, 312.22], [679.26, 313.04], [682.58, 313.78], [685.91, 314.46], [689.26, 315.11], [692.59, 315.8], [695.93, 316.44], [699.29, 317.02], [702.65, 317.54], [706.02, 318.05], [709.38, 318.58], [712.76, 318.99], [716.15, 319.3], [719.55, 319.52], [722.95, 319.7], [726.35, 319.83], [729.75, 319.95], [733.15, 320.07], [736.55, 320.18], [739.96, 320.23], [743.36, 320.22], [746.77, 320.15], [750.17, 320.03], [753.57, 319.84], [756.96, 319.63], [760.36, 319.36], [763.75, 319.11], [767.15, 318.86], [770.54, 318.61], [773.94, 318.38], [777.34, 318.19], [780.74, 317.98], [784.13, 317.75], [787.53, 317.52], [790.92, 317.3], [794.32, 317.06], [797.72, 316.86], [801.12, 316.65], [804.52, 316.46], [807.92, 316.29], [811.32, 316.19], [814.72, 316.17], [818.13, 316.16], [821.53, 316.16], [824.93, 316.2], [828.34, 316.2], [831.74, 316.14], [835.15, 316.1], [838.55, 316.05], [841.95, 316.06], [845.36, 316.07], [848.76, 316.04], [852.17, 315.98], [855.57, 315.92], [858.97, 315.95], [862.37, 316.13], [865.77, 316.39], [869.15, 316.79], [872.51, 317.29], [875.87, 317.89], [879.2, 318.59], [882.52, 319.35], [885.82, 320.16], [889.1, 321.08], [892.35, 322.08], [895.59, 323.13], [898.8, 324.28], [901.98, 325.49], [905.16, 326.68], [908.36, 327.87], [911.54, 329.07], [914.7, 330.35], [917.87, 331.59], [921.03, 332.84], [924.23, 334.02], [927.44, 335.13], [930.66, 336.23], [933.9, 337.29], [937.16, 338.28], [940.45, 339.15], [943.76, 339.94], [947.1, 340.59], [950.45, 341.19], [953.82, 341.71], [957.19, 342.15], [960.56, 342.61], [963.95, 343.01], [967.33, 343.33], [970.73, 343.62], [974.12, 343.9], [977.52, 344.12], [980.91, 344.31], [984.32, 344.45], [987.72, 344.58], [991.12, 344.72], [994.52, 344.84], [997.92, 344.96], [1001.32, 345.11], [1004.72, 345.3], [1008.12, 345.52], [1011.52, 345.71], [1014.92, 345.85], [1018.32, 346.02], [1021.72, 346.23], [1025.11, 346.52], [1028.5, 346.86], [1031.88, 347.24], [1035.26, 347.62], [1038.65, 347.95], [1042.03, 348.33], [1045.41, 348.74], [1048.8, 349.12], [1052.17, 349.57], [1055.54, 350.08], [1058.9, 350.61], [1062.25, 351.23], [1065.58, 351.92], [1068.9, 352.67], [1072.19, 353.53], [1075.46, 354.49], [1078.69, 355.56], [1081.91, 356.67], [1085.12, 357.81], [1088.31, 359.0], [1091.49, 360.21], [1094.68, 361.4], [1097.88, 362.55], [1101.1, 363.66], [1104.34, 364.7], [1107.6, 365.68], [1110.88, 366.61], [1114.14, 367.59], [1117.37, 368.64], [1120.58, 369.79], [1123.78, 370.95], [1126.98, 372.12], [1130.15, 373.35], [1133.32, 374.59], [1136.5, 375.81], [1139.69, 376.99], [1142.9, 378.14], [1146.12, 379.23], [1149.34, 380.35], [1152.55, 381.45], [1155.78, 382.53], [1159.03, 383.56], [1162.28, 384.57], [1165.56, 385.49], [1168.85, 386.34], [1172.16, 387.17], [1175.47, 387.93], [1178.79, 388.68], [1182.11, 389.44], [1185.43, 390.22], [1188.74, 390.98], [1192.08, 391.67], [1195.42, 392.32], [1198.77, 392.9], [1202.14, 393.41], [1205.52, 393.84], [1208.9, 394.23], [1212.29, 394.56], [1215.68, 394.82], [1219.08, 395.03], [1222.47, 395.26], [1225.87, 395.48], [1229.27, 395.65], [1232.67, 395.79], [1236.07, 395.94], [1239.47, 396.1], [1242.88, 396.21], [1246.28, 396.35], [1249.68, 396.47], [1253.08, 396.57], [1256.48, 396.7], [1259.88, 396.89], [1263.27, 397.17], [1266.66, 397.54], [1270.04, 397.95], [1273.41, 398.44], [1276.77, 398.99], [1280.11, 399.61], [1283.45, 400.29], [1286.78, 400.99], [1290.09, 401.79], [1293.4, 402.6], [1296.68, 403.51], [1299.92, 404.53], [1303.14, 405.64], [1306.33, 406.84], [1309.49, 408.1], [1312.62, 409.44], [1315.73, 410.81], [1318.82, 412.25], [1321.89, 413.72], [1324.93, 415.25], [1327.97, 416.78], [1331.02, 418.31], [1334.07, 419.82], [1337.12, 421.32], [1340.16, 422.86], [1343.18, 424.42], [1346.18, 426.03], [1349.17, 427.66], [1352.18, 429.26], [1355.19, 430.83], [1358.23, 432.38], [1361.29, 433.87], [1364.36, 435.34], [1367.43, 436.81], [1370.52, 438.24], [1373.66, 439.55], [1376.84, 440.75], [1380.07, 441.85], [1383.33, 442.83], [1386.61, 443.74], [1389.91, 444.55], [1393.24, 445.28], [1396.57, 445.96], [1399.91, 446.61], [1403.26, 447.23], [1406.62, 447.82], [1409.97, 448.39], [1413.31, 449.05], [1416.63, 449.79], [1419.93, 450.63], [1423.21, 451.55], [1426.46, 452.56], [1429.68, 453.68], [1432.88, 454.84], [1436.06, 456.04], [1439.22, 457.31], [1442.35, 458.65], [1445.44, 460.08], [1448.52, 461.52], [1451.59, 463.0], [1454.63, 464.53], [1457.65, 466.1], [1460.64, 467.72], [1463.63, 469.36], [1466.63, 470.96], [1469.66, 472.51], [1472.68, 474.09], [1475.69, 475.68], [1478.71, 477.25], [1481.73, 478.81], [1484.72, 480.45], [1487.63, 482.2], [1490.47, 484.09], [1493.22, 486.09], [1495.88, 488.22], [1498.45, 490.45], [1500.92, 492.79], [1503.28, 495.24], [1505.54, 497.79], [1507.68, 500.44], [1509.71, 503.17], [1511.62, 505.99], [1513.4, 508.89], [1515.05, 511.87], [1516.58, 514.91], [1517.97, 518.02], [1519.23, 521.18], [1520.34, 524.4], [1521.32, 527.66], [1522.16, 530.96], [1522.85, 534.29], [1523.39, 537.65], [1523.8, 541.03], [1524.05, 544.43], [1524.16, 547.83], [1524.11, 551.23], [1523.93, 554.63], [1523.59, 558.02], [1523.11, 561.39], [1522.48, 564.73], [1521.71, 568.05], [1520.79, 571.33], [1519.74, 574.56], [1518.54, 577.75], [1517.21, 580.89], [1515.74, 583.96], [1514.15, 586.96], [1512.42, 589.9], [1510.57, 592.75], [1508.59, 595.53], [1506.51, 598.22], [1504.32, 600.82], [1502.04, 603.35], [1499.73, 605.85], [1497.34, 608.28], [1494.91, 610.66], [1492.42, 612.98], [1489.84, 615.2], [1487.17, 617.31], [1484.41, 619.3], [1481.56, 621.17], [1478.65, 622.94], [1475.71, 624.65], [1472.72, 626.28], [1469.69, 627.83], [1466.65, 629.37], [1463.58, 630.84], [1460.49, 632.27], [1457.35, 633.58], [1454.18, 634.8], [1450.97, 635.95], [1447.73, 636.98], [1444.45, 637.92], [1441.17, 638.83], [1437.88, 639.67], [1434.58, 640.53], [1431.26, 641.29], [1427.94, 642.03], [1424.63, 642.82], [1421.32, 643.64], [1418.02, 644.45], [1414.71, 645.25], [1411.43, 646.15], [1408.16, 647.11], [1404.92, 648.17], [1401.69, 649.24], [1398.49, 650.38], [1395.31, 651.6], [1392.14, 652.85], [1389.0, 654.16], [1385.88, 655.53], [1382.8, 656.96], [1379.73, 658.45], [1376.69, 659.98], [1373.66, 661.52], [1370.66, 663.14], [1367.68, 664.78], [1364.71, 666.45], [1361.76, 668.15], [1358.81, 669.85], [1355.86, 671.55], [1352.92, 673.25], [1349.98, 674.98], [1347.05, 676.7], [1344.14, 678.48], [1341.24, 680.25], [1338.35, 682.05], [1335.45, 683.84], [1332.54, 685.6], [1329.6, 687.32], [1326.6, 688.93], [1323.55, 690.43], [1320.45, 691.85], [1317.32, 693.19], [1314.16, 694.44], [1310.96, 695.6], [1307.72, 696.67], [1304.5, 697.77], [1301.29, 698.89], [1298.07, 700.01], [1294.85, 701.11], [1291.63, 702.22], [1288.44, 703.38], [1285.25, 704.58], [1282.07, 705.81], [1278.93, 707.1], [1275.81, 708.47], [1272.69, 709.84], [1269.58, 711.22], [1266.48, 712.62], [1263.39, 714.06], [1260.32, 715.53], [1257.27, 717.04], [1254.23, 718.56], [1251.14, 720.0], [1248.02, 721.37], [1244.88, 722.69], [1241.73, 723.96], [1238.57, 725.23], [1235.41, 726.49], [1232.26, 727.8], [1229.14, 729.15], [1226.04, 730.55], [1222.95, 731.99], [1219.88, 733.46], [1216.84, 734.99], [1213.8, 736.51], [1210.77, 738.06], [1207.74, 739.62], [1204.72, 741.2], [1201.7, 742.77], [1198.68, 744.34], [1195.68, 745.94], [1192.65, 747.5], [1189.61, 749.02], [1186.52, 750.46], [1183.39, 751.79], [1180.2, 752.99], [1176.97, 754.05], [1173.7, 755.02], [1170.41, 755.89], [1167.11, 756.71], [1163.79, 757.45], [1160.46, 758.15], [1157.12, 758.82], [1153.77, 759.47], [1150.42, 760.05], [1147.06, 760.6], [1143.71, 761.2], [1140.38, 761.92], [1137.07, 762.69], [1133.78, 763.57], [1130.52, 764.55], [1127.28, 765.58], [1124.06, 766.69], [1120.85, 767.82], [1117.65, 768.99], [1114.45, 770.16], [1111.25, 771.31], [1108.04, 772.44], [1104.81, 773.53], [1101.58, 774.61], [1098.34, 775.64], [1095.1, 776.68], [1091.86, 777.73], [1088.61, 778.73], [1085.34, 779.69], [1082.08, 780.66], [1078.84, 781.7], [1075.6, 782.77], [1072.36, 783.82], [1069.13, 784.88], [1065.89, 785.92], [1062.66, 786.99], [1059.42, 788.05], [1056.18, 789.07], [1052.94, 790.13], [1049.71, 791.2], [1046.5, 792.33], [1043.3, 793.49], [1040.1, 794.66], [1036.89, 795.8], [1033.66, 796.86], [1030.41, 797.87], [1027.14, 798.83], [1023.87, 799.79], [1020.63, 800.81], [1017.37, 801.8], [1014.1, 802.76], [1010.84, 803.71], [1007.56, 804.66], [1004.27, 805.51], [1000.96, 806.29], [997.62, 806.98], [994.28, 807.65], [990.96, 808.38], [987.65, 809.17], [984.36, 810.04], [981.09, 810.99], [977.81, 811.9], [974.54, 812.86], [971.27, 813.8], [968.02, 814.8], [964.78, 815.86], [961.54, 816.9], [958.31, 817.98], [955.08, 819.03], [951.8, 819.97], [948.52, 820.86], [945.19, 821.6], [941.86, 822.27], [938.52, 822.93], [935.15, 823.46], [931.78, 823.88], [928.38, 824.18], [924.99, 824.4], [921.59, 824.52], [918.18, 824.58], [914.78, 824.6], [911.37, 824.53], [907.97, 824.44], [904.57, 824.32], [901.17, 824.13], [897.78, 823.87], [894.39, 823.54], [891.0, 823.19], [887.62, 822.82], [884.24, 822.44], [880.86, 822.01], [877.48, 821.59], [874.1, 821.22], [870.7, 820.93], [867.31, 820.66], [863.92, 820.39], [860.52, 820.11], [857.13, 819.85], [853.74, 819.57], [850.35, 819.27], [846.96, 818.98], [843.56, 818.7], [840.17, 818.46], [836.77, 818.22], [833.37, 818.09], [829.97, 818.02], [826.56, 818.0], [823.16, 818.0], [819.75, 818.09], [816.35, 818.22], [812.95, 818.4], [809.55, 818.55], [806.15, 818.67], [802.75, 818.74], [799.34, 818.79], [795.94, 818.86], [792.54, 818.99], [789.14, 819.2], [785.74, 819.41], [782.34, 819.58], [778.95, 819.78], [775.55, 819.99], [772.15, 820.23], [768.76, 820.46], [765.36, 820.72], [761.97, 821.01], [758.58, 821.35], [755.2, 821.77], [751.83, 822.25], [748.47, 822.8], [745.12, 823.39], [741.78, 824.05], [738.47, 824.81], [735.17, 825.66], [731.87, 826.51], [728.58, 827.36], [725.29, 828.25], [722.0, 829.13], [718.75, 830.14], [715.54, 831.27], [712.37, 832.5], [709.25, 833.87], [706.18, 835.34], [703.18, 836.95], [700.25, 838.68], [697.39, 840.54], [694.62, 842.52], [691.94, 844.61], [689.36, 846.83], [686.87, 849.15], [684.48, 851.58], [682.2, 854.1], [680.03, 856.73], [677.98, 859.45], [676.05, 862.25], [674.24, 865.13], [672.56, 868.09], [671.01, 871.12], [669.59, 874.22], [668.3, 877.37], [667.16, 880.57], [666.15, 883.82], [665.28, 887.12], [664.56, 890.44], [663.99, 893.8], [663.55, 897.18], [663.27, 900.57], [663.13, 903.97], [663.14, 907.37], [663.3, 910.77], [663.61, 914.16], [664.06, 917.54], [664.65, 920.89], [665.4, 924.21], [666.28, 927.5], [667.31, 930.74], [668.47, 933.94], [669.78, 937.09], [671.22, 940.17], [672.79, 943.19], [674.49, 946.14], [676.31, 949.01], [678.26, 951.81], [680.33, 954.51], [682.52, 957.12], [684.81, 959.64], [687.21, 962.05], [689.71, 964.36], [692.32, 966.55], [695.01, 968.63], [697.79, 970.6], [700.64, 972.46], [703.53, 974.26], [706.45, 976.0], [709.41, 977.69], [712.4, 979.31], [715.41, 980.9], [718.47, 982.39], [721.55, 983.84], [724.65, 985.25], [727.76, 986.63], [730.9, 987.95], [734.06, 989.22], [737.25, 990.4], [740.46, 991.54], [743.68, 992.66], [746.93, 993.67], [750.18, 994.67], [753.46, 995.57], [756.77, 996.4], [760.1, 997.08], [763.46, 997.62], [766.84, 998.03], [770.23, 998.33], [773.63, 998.49], [777.04, 998.52], [780.44, 998.47], [783.84, 998.29], [787.23, 997.99], [790.6, 997.55], [793.97, 997.02], [797.32, 996.42], [800.65, 995.7], [803.96, 994.92], [807.26, 994.08], [810.55, 993.22], [813.84, 992.34], [817.13, 991.46], [820.42, 990.59], [823.72, 989.76], [827.04, 988.98], [830.36, 988.24], [833.69, 987.57], [837.04, 986.93], [840.38, 986.29], [843.72, 985.62], [847.05, 984.92], [850.36, 984.14], [853.67, 983.31], [856.96, 982.43], [860.24, 981.54], [863.52, 980.63], [866.79, 979.67], [870.06, 978.73], [873.32, 977.76], [876.58, 976.76], [879.83, 975.75], [883.12, 974.89], [886.45, 974.17], [889.8, 973.59], [893.18, 973.16], [896.57, 972.87], [899.97, 972.73], [903.38, 972.74], [906.78, 972.9], [910.17, 973.2], [913.54, 973.65], [916.89, 974.25], [920.21, 974.99], [923.5, 975.87], [926.75, 976.9], [929.95, 978.06], [933.09, 979.37], [936.18, 980.8], [939.2, 982.37], [942.15, 984.07], [945.02, 985.9], [947.81, 987.85], [950.52, 989.91], [953.13, 992.1], [955.65, 994.39], [958.06, 996.79], [960.37, 999.29], [962.56, 1001.89], [964.65, 1004.59], [966.61, 1007.37], [968.45, 1010.23], [970.17, 1013.17], [971.75, 1016.18], [973.21, 1019.26], [974.53, 1022.4], [975.71, 1025.59], [976.75, 1028.83], [977.66, 1032.11], [978.42, 1035.43], [979.03, 1038.78], [979.5, 1042.15], [979.82, 1045.54], [980.0, 1048.94], [980.03, 1052.34], [979.91, 1055.75], [979.64, 1059.14], [979.23, 1062.52], [978.67, 1065.88], [977.96, 1069.21], [977.12, 1072.5], [976.13, 1075.76], [975.0, 1078.97], [973.73, 1082.13], [972.32, 1085.23], [970.79, 1088.27], [969.12, 1091.24], [967.33, 1094.13], [965.41, 1096.94], [963.37, 1099.67], [961.22, 1102.31], [958.95, 1104.85], [956.58, 1107.29], [954.1, 1109.62], [951.53, 1111.85], [948.86, 1113.96], [946.1, 1115.96], [943.25, 1117.83], [940.33, 1119.58], [937.34, 1121.2], [934.28, 1122.68], [931.15, 1124.04], [927.98, 1125.26], [924.75, 1126.34], [921.47, 1127.27], [918.16, 1128.07], [914.82, 1128.72], [911.46, 1129.25], [908.08, 1129.64], [904.68, 1129.9], [901.28, 1130.03], [897.88, 1130.03], [894.48, 1129.94], [891.08, 1129.72], [887.69, 1129.4], [884.3, 1129.04], [880.92, 1128.65], [877.55, 1128.16], [874.2, 1127.59], [870.84, 1127.01], [867.48, 1126.49], [864.11, 1126.03], [860.72, 1125.7], [857.33, 1125.43], [853.93, 1125.24], [850.52, 1125.12], [847.12, 1125.13], [843.72, 1125.15], [840.31, 1125.17], [836.91, 1125.19], [833.5, 1125.2], [830.1, 1125.21], [826.7, 1125.23], [823.29, 1125.23], [819.89, 1125.2], [816.48, 1125.14], [813.08, 1125.1], [809.68, 1125.03], [806.27, 1124.89], [802.87, 1124.8], [799.47, 1124.75], [796.07, 1124.86], [792.67, 1125.11], [789.29, 1125.5], [785.93, 1126.05], [782.6, 1126.74], [779.3, 1127.57], [776.03, 1128.54], [772.82, 1129.66], [769.65, 1130.91], [766.54, 1132.3], [763.5, 1133.82], [760.52, 1135.48], [757.62, 1137.26], [754.8, 1139.16], [752.06, 1141.19], [749.42, 1143.33], [746.87, 1145.58], [744.42, 1147.95], [742.07, 1150.41], [739.83, 1152.98], [737.71, 1155.64], [735.7, 1158.39], [733.82, 1161.22], [732.05, 1164.13], [730.42, 1167.12], [728.92, 1170.18], [727.55, 1173.29], [726.32, 1176.47], [725.22, 1179.69], [724.27, 1182.96], [723.46, 1186.26], [722.79, 1189.6], [722.27, 1192.97], [721.9, 1196.35], [721.67, 1199.75], [721.59, 1203.15], [721.65, 1206.55], [721.87, 1209.95], [722.23, 1213.33], [722.73, 1216.7], [723.38, 1220.04], [724.18, 1223.35], [725.12, 1226.62], [726.2, 1229.85], [727.42, 1233.03], [728.77, 1236.15], [730.26, 1239.22], [731.88, 1242.21], [733.63, 1245.13], [735.5, 1247.97], [737.5, 1250.73], [739.61, 1253.4], [741.84, 1255.98], [744.17, 1258.45], [746.61, 1260.83], [749.1, 1263.15], [751.64, 1265.42], [754.27, 1267.58], [757.0, 1269.62], [759.81, 1271.54], [762.7, 1273.33], [765.67, 1275.0], [768.69, 1276.56], [771.78, 1277.99], [774.91, 1279.34], [778.08, 1280.57], [781.31, 1281.66], [784.57, 1282.61], [787.88, 1283.44], [791.2, 1284.17], [794.55, 1284.78], [797.92, 1285.27], [801.3, 1285.66], [804.69, 1286.03], [808.08, 1286.31], [811.47, 1286.52], [814.87, 1286.7], [818.28, 1286.83], [821.68, 1286.97], [825.08, 1287.12], [828.48, 1287.22], [831.88, 1287.3], [835.29, 1287.41], [838.68, 1287.65], [842.07, 1287.98], [845.46, 1288.34], [848.85, 1288.65], [852.23, 1288.98], [855.62, 1289.36], [858.99, 1289.8], [862.35, 1290.34], [865.7, 1290.96], [869.03, 1291.66], [872.36, 1292.37], [875.69, 1293.09], [879.0, 1293.86], [882.3, 1294.72], [885.57, 1295.65], [888.84, 1296.62], [892.09, 1297.62], [895.34, 1298.63], [898.58, 1299.67], [901.82, 1300.71], [905.06, 1301.77], [908.31, 1302.79], [911.58, 1303.73], [914.88, 1304.56], [918.2, 1305.3], [921.54, 1305.98], [924.88, 1306.62], [928.23, 1307.23], [931.6, 1307.73], [934.97, 1308.19], [938.34, 1308.64], [941.71, 1309.12], [945.08, 1309.63], [948.44, 1310.15], [951.8, 1310.75], [955.13, 1311.41], [958.46, 1312.13], [961.78, 1312.88], [965.11, 1313.61], [968.44, 1314.33], [971.78, 1314.97], [975.13, 1315.55], [978.5, 1316.02], [981.88, 1316.47], [985.24, 1316.99], [988.59, 1317.62], [991.93, 1318.27], [995.28, 1318.9], [998.63, 1319.5], [1001.99, 1320.04], [1005.35, 1320.57], [1008.72, 1321.06], [1012.09, 1321.52], [1015.46, 1322.02], [1018.82, 1322.57], [1022.17, 1323.17], [1025.52, 1323.78], [1028.88, 1324.33], [1032.24, 1324.85], [1035.61, 1325.35], [1038.98, 1325.85], [1042.34, 1326.35], [1045.71, 1326.85], [1049.08, 1327.33], [1052.45, 1327.79], [1055.82, 1328.26], [1059.21, 1328.66], [1062.6, 1328.95], [1065.99, 1329.22], [1069.39, 1329.41], [1072.79, 1329.55], [1076.19, 1329.7], [1079.59, 1329.85], [1082.99, 1330.08], [1086.38, 1330.41], [1089.75, 1330.87], [1093.1, 1331.45], [1096.45, 1332.09], [1099.79, 1332.72], [1103.14, 1333.33], [1106.5, 1333.88], [1109.88, 1334.33], [1113.26, 1334.7], [1116.65, 1334.96], [1120.05, 1335.25], [1123.44, 1335.51], [1126.84, 1335.71], [1130.24, 1335.82], [1133.64, 1335.85], [1137.05, 1335.78], [1140.45, 1335.66], [1143.85, 1335.51], [1147.25, 1335.38], [1150.65, 1335.25], [1154.05, 1335.05], [1157.45, 1334.85], [1160.85, 1334.6], [1164.24, 1334.36], [1167.64, 1334.24], [1171.05, 1334.25], [1174.45, 1334.38], [1177.85, 1334.58], [1181.24, 1334.9], [1184.62, 1335.26], [1188.0, 1335.7], [1191.36, 1336.23], [1194.7, 1336.88], [1198.04, 1337.56], [1201.37, 1338.27], [1204.7, 1338.94], [1208.04, 1339.62], [1211.39, 1340.24], [1214.74, 1340.84], [1218.09, 1341.43], [1221.44, 1342.02], [1224.79, 1342.65], [1228.13, 1343.31], [1231.46, 1344.01], [1234.79, 1344.73], [1238.11, 1345.48], [1241.4, 1346.33], [1244.69, 1347.21], [1247.98, 1348.11], [1251.27, 1348.96], [1254.56, 1349.83], [1257.86, 1350.66], [1261.16, 1351.5], [1264.46, 1352.36], [1267.75, 1353.22], [1271.06, 1354.0], [1274.39, 1354.72], [1277.72, 1355.42], [1281.06, 1356.08], [1284.41, 1356.7], [1287.75, 1357.34], [1291.1, 1357.96], [1294.46, 1358.52], [1297.82, 1359.03], [1301.2, 1359.48], [1304.59, 1359.79], [1307.98, 1360.07], [1311.37, 1360.32], [1314.77, 1360.61], [1318.15, 1360.95], [1321.53, 1361.35], [1324.91, 1361.81], [1328.27, 1362.33], [1331.62, 1362.92], [1334.97, 1363.53], [1338.32, 1364.16], [1341.65, 1364.86], [1344.97, 1365.6], [1348.29, 1366.38], [1351.6, 1367.15], [1354.91, 1367.95], [1358.22, 1368.76], [1361.51, 1369.62], [1364.8, 1370.48], [1368.1, 1371.32], [1371.4, 1372.17], [1374.7, 1373.0], [1377.96, 1373.97], [1381.18, 1375.08], [1384.35, 1376.33], [1387.46, 1377.72], [1390.5, 1379.24], [1393.48, 1380.89], [1396.38, 1382.67], [1399.21, 1384.57], [1401.94, 1386.59], [1404.59, 1388.73], [1407.14, 1390.99], [1409.6, 1393.35], [1411.95, 1395.81], [1414.18, 1398.37], [1416.31, 1401.03], [1418.32, 1403.78], [1420.21, 1406.61], [1421.97, 1409.53], [1423.61, 1412.51], [1425.11, 1415.56], [1426.48, 1418.68], [1427.72, 1421.85], [1428.82, 1425.07], [1429.77, 1428.34], [1430.58, 1431.65], [1431.25, 1434.99], [1431.78, 1438.35], [1432.16, 1441.73], [1432.39, 1445.13], [1432.47, 1448.53], [1432.41, 1451.93], [1432.2, 1455.33], [1431.84, 1458.72], [1431.34, 1462.08], [1430.69, 1465.43], [1429.89, 1468.74], [1428.96, 1472.01], [1427.88, 1475.24], [1426.66, 1478.42], [1425.31, 1481.54], [1423.82, 1484.6], [1422.21, 1487.6], [1420.46, 1490.52], [1418.59, 1493.37], [1416.6, 1496.13], [1414.49, 1498.8], [1412.26, 1501.37], [1409.93, 1503.85], [1407.49, 1506.23], [1404.95, 1508.5], [1402.32, 1510.65], [1399.59, 1512.69], [1396.78, 1514.61], [1393.89, 1516.4], [1390.92, 1518.07], [1387.88, 1519.61], [1384.78, 1521.02], [1381.63, 1522.3], [1378.42, 1523.45], [1375.17, 1524.45], [1371.88, 1525.31], [1368.55, 1526.03], [1365.2, 1526.61], [1361.82, 1527.04], [1358.43, 1527.32], [1355.03, 1527.46], [1351.62, 1527.45], [1348.22, 1527.29], [1344.83, 1527.01], [1341.44, 1526.66], [1338.07, 1526.22], [1334.7, 1525.74], [1331.35, 1525.14], [1328.01, 1524.48], [1324.68, 1523.77], [1321.36, 1523.02], [1318.05, 1522.23], [1314.74, 1521.42], [1311.44, 1520.6], [1308.13, 1519.8], [1304.81, 1519.05], [1301.49, 1518.3], [1298.15, 1517.64], [1294.81, 1516.99], [1291.46, 1516.33], [1288.14, 1515.63], [1284.83, 1514.79], [1281.56, 1513.85], [1278.3, 1512.88], [1275.06, 1511.85], [1271.8, 1510.85], [1268.52, 1509.96], [1265.21, 1509.16], [1261.88, 1508.45], [1258.54, 1507.8], [1255.19, 1507.18], [1251.85, 1506.53], [1248.52, 1505.83], [1245.19, 1505.13], [1241.86, 1504.41], [1238.55, 1503.61], [1235.26, 1502.74], [1231.98, 1501.81], [1228.72, 1500.83], [1225.49, 1499.77], [1222.26, 1498.71], [1219.02, 1497.65], [1215.81, 1496.53], [1212.61, 1495.34], [1209.44, 1494.11], [1206.29, 1492.82], [1203.14, 1491.54], [1199.98, 1490.26], [1196.85, 1488.94], [1193.71, 1487.62], [1190.54, 1486.38], [1187.36, 1485.15], [1184.17, 1483.97], [1180.99, 1482.76], [1177.81, 1481.53], [1174.62, 1480.35], [1171.43, 1479.15], [1168.26, 1477.93], [1165.08, 1476.7], [1161.94, 1475.39], [1158.82, 1474.03], [1155.71, 1472.64], [1152.62, 1471.21], [1149.56, 1469.73], [1146.51, 1468.21], [1143.46, 1466.7], [1140.37, 1465.27], [1137.26, 1463.9], [1134.12, 1462.57], [1130.99, 1461.22], [1127.85, 1459.92], [1124.7, 1458.63], [1121.52, 1457.42], [1118.3, 1456.3], [1115.09, 1455.17], [1111.86, 1454.11], [1108.61, 1453.1], [1105.35, 1452.12], [1102.08, 1451.16], [1098.81, 1450.21], [1095.55, 1449.22], [1092.3, 1448.21], [1089.04, 1447.24], [1085.76, 1446.33], [1082.46, 1445.51], [1079.13, 1444.8], [1075.78, 1444.16], [1072.43, 1443.6], [1069.06, 1443.09], [1065.68, 1442.69], [1062.29, 1442.37], [1058.89, 1442.13], [1055.5, 1441.9], [1052.1, 1441.69], [1048.7, 1441.57], [1045.29, 1441.51], [1041.89, 1441.5], [1038.49, 1441.5], [1035.08, 1441.5], [1031.68, 1441.44], [1028.28, 1441.37], [1024.87, 1441.24], [1021.47, 1441.07], [1018.07, 1440.9], [1014.68, 1440.67], [1011.28, 1440.43], [1007.89, 1440.18], [1004.49, 1439.91], [1001.1, 1439.61], [997.71, 1439.3], [994.32, 1439.04], [990.92, 1438.83], [987.52, 1438.69], [984.12, 1438.54], [980.72, 1438.35], [977.32, 1438.17], [973.92, 1437.99], [970.53, 1437.75], [967.13, 1437.51], [963.73, 1437.33], [960.33, 1437.13], [956.93, 1437.01], [953.53, 1436.93], [950.12, 1436.94], [946.72, 1437.01], [943.32, 1437.14], [939.92, 1437.32], [936.52, 1437.51], [933.12, 1437.71], [929.72, 1437.92], [926.32, 1438.09], [922.92, 1438.22], [919.52, 1438.3], [916.11, 1438.35], [912.71, 1438.41], [909.31, 1438.49], [905.9, 1438.59], [902.5, 1438.67], [899.1, 1438.7], [895.69, 1438.63], [892.3, 1438.45], [888.9, 1438.2], [885.51, 1437.86], [882.14, 1437.44], [878.76, 1436.99], [875.39, 1436.49], [872.04, 1435.93], [868.68, 1435.34], [865.32, 1434.84], [861.95, 1434.35], [858.58, 1433.86], [855.2, 1433.46], [851.82, 1433.04], [848.44, 1432.66], [845.04, 1432.37], [841.65, 1432.17], [838.24, 1432.06], [834.84, 1432.02], [831.44, 1432.0], [828.03, 1431.9], [824.63, 1431.79], [821.23, 1431.64], [817.83, 1431.46], [814.43, 1431.25], [811.04, 1430.95], [807.67, 1430.52], [804.29, 1430.05], [800.93, 1429.51], [797.58, 1428.93], [794.22, 1428.35], [790.86, 1427.82], [787.5, 1427.29], [784.13, 1426.81], [780.76, 1426.34], [777.38, 1425.91], [774.01, 1425.44], [770.64, 1424.97], [767.27, 1424.47], [763.9, 1424.0], [760.52, 1423.58], [757.14, 1423.19], [753.76, 1422.8], [750.38, 1422.35], [747.02, 1421.84], [743.66, 1421.27], [740.32, 1420.64], [736.98, 1419.96], [733.65, 1419.28], [730.32, 1418.57], [726.99, 1417.84], [723.65, 1417.16], [720.33, 1416.43], [717.01, 1415.66], [713.71, 1414.85], [710.39, 1414.07], [707.07, 1413.34], [703.74, 1412.63], [700.4, 1411.97], [697.05, 1411.35], [693.72, 1410.67], [690.38, 1409.99], [687.05, 1409.28], [683.73, 1408.55], [680.4, 1407.84], [677.04, 1407.28], [673.66, 1406.86], [670.28, 1406.51], [666.88, 1406.21], [663.49, 1405.97], [660.09, 1405.83], [656.68, 1405.74], [653.28, 1405.72], [649.88, 1405.75], [646.47, 1405.79], [643.07, 1405.83], [639.66, 1405.85], [636.26, 1405.87], [632.86, 1405.85], [629.45, 1405.8], [626.05, 1405.67], [622.65, 1405.49], [619.26, 1405.25], [615.86, 1405.0], [612.47, 1404.74], [609.07, 1404.52], [605.68, 1404.26], [602.29, 1403.94], [598.9, 1403.64], [595.51, 1403.32], [592.12, 1402.97], [588.74, 1402.6], [585.35, 1402.25], [581.96, 1401.92], [578.57, 1401.62], [575.18, 1401.35], [571.78, 1401.09], [568.39, 1400.87], [564.99, 1400.65], [561.59, 1400.42], [558.19, 1400.24], [554.79, 1400.08], [551.4, 1399.88], [548.0, 1399.64], [544.61, 1399.36], [541.21, 1399.08], [537.82, 1398.78], [534.42, 1398.63], [531.02, 1398.63], [527.62, 1398.77], [524.23, 1399.06], [520.85, 1399.5], [517.5, 1400.08], [514.17, 1400.81], [510.88, 1401.68], [507.63, 1402.69], [504.43, 1403.84], [501.28, 1405.13], [498.18, 1406.56], [495.16, 1408.12], [492.2, 1409.81], [489.32, 1411.62], [486.52, 1413.56], [483.81, 1415.61], [481.19, 1417.79], [478.64, 1420.04], [476.15, 1422.36], [473.75, 1424.78], [471.43, 1427.27], [469.18, 1429.83], [467.01, 1432.45], [464.9, 1435.12], [462.84, 1437.83], [460.83, 1440.57], [458.86, 1443.35], [456.92, 1446.15], [455.05, 1448.99], [453.29, 1451.91], [451.65, 1454.89], [450.12, 1457.93], [448.65, 1461.0], [447.24, 1464.1], [445.85, 1467.21], [444.53, 1470.35], [443.2, 1473.48], [441.87, 1476.61], [440.54, 1479.75], [439.31, 1482.92], [438.21, 1486.14], [437.18, 1489.39], [436.15, 1492.63], [435.18, 1495.9], [434.36, 1499.2], [433.67, 1502.53], [433.13, 1505.89], [432.7, 1509.27], [432.4, 1512.66], [432.22, 1516.06], [432.18, 1519.46], [432.22, 1522.87], [432.36, 1526.27], [432.54, 1529.67], [432.71, 1533.07], [432.86, 1536.47], [432.98, 1539.87], [433.12, 1543.27], [433.35, 1546.67], [433.52, 1550.07], [433.69, 1553.47], [433.76, 1556.87], [433.81, 1560.28], [433.79, 1563.68], [433.72, 1567.08], [433.58, 1570.48], [433.4, 1573.88], [433.23, 1577.28], [433.06, 1580.68], [432.88, 1584.08], [432.73, 1587.48], [432.63, 1590.89], [432.55, 1594.29], [432.52, 1597.69], [432.53, 1601.1], [432.59, 1604.5], [432.65, 1607.91], [432.7, 1611.31], [432.77, 1614.71], [432.8, 1618.12], [432.86, 1621.52], [432.83, 1624.92], [432.74, 1628.33], [432.61, 1631.73], [432.43, 1635.13], [432.26, 1638.53], [432.11, 1641.93], [431.91, 1645.33], [431.68, 1648.72], [431.46, 1652.12], [431.24, 1655.52], [431.09, 1658.92], [430.92, 1662.32], [430.8, 1665.72], [430.72, 1669.12], [430.7, 1672.53], [430.66, 1675.93], [430.57, 1679.33], [430.49, 1682.74], [430.47, 1686.14], [430.5, 1689.55], [430.55, 1692.95], [430.57, 1696.35], [430.61, 1699.76], [430.65, 1703.16], [430.68, 1706.57], [430.72, 1709.97], [430.75, 1713.37], [430.79, 1716.78], [430.82, 1720.18], [430.86, 1723.59], [430.89, 1726.99], [430.93, 1730.39], [430.97, 1733.8], [431.0, 1737.2], [431.04, 1740.61], [431.07, 1744.01], [431.11, 1747.41], [431.14, 1750.82], [431.18, 1754.22], [431.21, 1757.63], [431.25, 1761.03], [431.28, 1764.43], [431.32, 1767.84], [431.36, 1771.24], [431.39, 1774.65], [431.43, 1778.05], [431.46, 1781.45], [431.5, 1784.86], [431.53, 1788.26], [431.57, 1791.67], [431.6, 1795.07], [431.64, 1798.47], [431.68, 1801.88], [431.71, 1805.28], [431.75, 1808.68], [431.78, 1812.09], [431.82, 1815.49], [431.85, 1818.9], [431.89, 1822.3], [431.92, 1825.7], [431.96, 1829.11], [431.99, 1832.51], [432.03, 1835.92], [432.07, 1839.32], [432.1, 1842.72], [432.14, 1846.13]], [[1428.67, 70.0], [1428.15, 73.5], [1427.64, 77.0], [1427.12, 80.5], [1426.61, 84.0], [1426.09, 87.5], [1425.57, 91.0], [1425.06, 94.5], [1424.54, 98.0], [1424.03, 101.5], [1423.51, 105.0], [1423.0, 108.51], [1422.48, 112.01], [1421.96, 115.51], [1421.45, 119.01], [1420.93, 122.51], [1420.42, 126.01], [1419.9, 129.51], [1419.39, 133.01], [1418.87, 136.51], [1418.35, 140.01], [1417.84, 143.51], [1417.32, 147.01], [1416.81, 150.51], [1416.29, 154.01], [1415.78, 157.51], [1415.26, 161.01], [1414.75, 164.51], [1414.23, 168.01], [1413.71, 171.51], [1413.2, 175.01], [1412.68, 178.51], [1412.17, 182.01], [1411.65, 185.52], [1411.14, 189.02], [1410.62, 192.52], [1410.1, 196.02], [1409.59, 199.52], [1409.07, 203.02], [1408.56, 206.52], [1408.04, 210.02], [1407.53, 213.52], [1407.01, 217.02], [1406.37, 220.5], [1405.62, 223.96], [1404.75, 227.39], [1403.76, 230.78], [1402.65, 234.14], [1401.43, 237.46], [1400.09, 240.74], [1398.65, 243.97], [1397.31, 247.25], [1396.09, 250.57], [1394.99, 253.93], [1393.99, 257.32], [1393.08, 260.74], [1392.29, 264.19], [1391.62, 267.67], [1391.07, 271.16], [1390.59, 274.67], [1390.16, 278.18], [1389.77, 281.7], [1389.43, 285.22], [1389.17, 288.75], [1388.97, 292.28], [1388.78, 295.81], [1388.63, 299.35], [1388.48, 302.88], [1388.29, 306.42], [1388.1, 309.95], [1387.89, 313.48], [1387.68, 317.01], [1387.5, 320.55], [1387.33, 324.08], [1387.19, 327.62], [1386.99, 331.15], [1386.79, 334.68], [1386.64, 338.22], [1386.49, 341.75], [1386.36, 345.29], [1386.23, 348.82], [1386.11, 352.36], [1385.97, 355.89], [1385.93, 359.43], [1385.91, 362.97], [1385.86, 366.51], [1385.82, 370.05], [1385.86, 373.58], [1385.97, 377.12], [1386.15, 380.65], [1386.43, 384.18], [1386.81, 387.7], [1387.31, 391.2], [1387.93, 394.69], [1388.6, 398.16], [1389.32, 401.62], [1390.07, 405.08], [1390.84, 408.54], [1391.61, 411.99], [1392.32, 415.45], [1392.96, 418.93], [1393.62, 422.41], [1394.34, 425.87], [1395.11, 429.33], [1395.92, 432.77], [1396.75, 436.21], [1397.63, 439.64], [1398.55, 443.06], [1399.48, 446.47], [1400.4, 449.89], [1401.29, 453.31], [1402.09, 456.76], [1402.79, 460.23], [1403.42, 463.71], [1403.98, 467.2], [1404.48, 470.7], [1404.9, 474.22], [1405.27, 477.74], [1405.61, 481.26], [1405.97, 484.78], [1406.31, 488.3], [1406.59, 491.83], [1406.85, 495.36], [1407.16, 498.88], [1407.46, 502.41], [1407.73, 505.93], [1407.98, 509.46], [1408.27, 512.99], [1408.58, 516.51], [1408.87, 520.04], [1409.15, 523.57], [1409.3, 527.1], [1409.32, 530.64], [1409.23, 534.18], [1409.02, 537.71], [1408.68, 541.23], [1408.23, 544.74], [1407.65, 548.23], [1406.96, 551.7], [1406.14, 555.14], [1405.21, 558.56], [1404.16, 561.94], [1403.0, 565.28], [1401.72, 568.58], [1400.33, 571.83], [1398.82, 575.03], [1397.21, 578.18], [1395.49, 581.28], [1393.67, 584.31], [1391.74, 587.28], [1389.72, 590.18], [1387.59, 593.01], [1385.37, 595.76], [1383.06, 598.44], [1380.65, 601.03], [1378.16, 603.54], [1375.58, 605.96], [1372.94, 608.32], [1370.23, 610.59], [1367.45, 612.78], [1364.61, 614.9], [1361.72, 616.93], [1358.78, 618.9], [1355.78, 620.78], [1352.72, 622.57], [1349.62, 624.26], [1346.47, 625.87], [1343.3, 627.44], [1340.08, 628.92], [1336.84, 630.33], [1333.56, 631.67], [1330.25, 632.9], [1326.89, 634.04], [1323.53, 635.12], [1320.15, 636.18], [1316.78, 637.25], [1313.41, 638.33], [1310.02, 639.34], [1306.6, 640.27], [1303.17, 641.11], [1299.71, 641.87], [1296.23, 642.52], [1292.75, 643.12], [1289.26, 643.75], [1285.78, 644.37], [1282.29, 644.96], [1278.81, 645.58], [1275.34, 646.28], [1271.87, 646.96], [1268.39, 647.61], [1264.91, 648.25], [1261.44, 648.92], [1257.96, 649.57], [1254.48, 650.24], [1251.01, 650.89], [1247.53, 651.58], [1244.07, 652.28], [1240.6, 652.96], [1237.13, 653.69], [1233.68, 654.47], [1230.25, 655.33], [1226.84, 656.27], [1223.43, 657.21], [1220.01, 658.14], [1216.58, 658.99], [1213.13, 659.8], [1209.68, 660.55], [1206.21, 661.24], [1202.72, 661.86], [1199.23, 662.41], [1195.72, 662.89], [1192.21, 663.33], [1188.69, 663.71], [1185.17, 664.02], [1181.65, 664.36], [1178.13, 664.7], [1174.6, 665.04], [1171.08, 665.38], [1167.56, 665.74], [1164.05, 666.16], [1160.54, 666.65], [1157.06, 667.25], [1153.59, 667.94], [1150.12, 668.64], [1146.66, 669.38], [1143.21, 670.18], [1139.76, 670.96], [1136.33, 671.82], [1132.92, 672.77], [1129.51, 673.72], [1126.1, 674.65], [1122.67, 675.54], [1119.25, 676.44], [1115.85, 677.4], [1112.43, 678.31], [1109.01, 679.22], [1105.58, 680.09], [1102.13, 680.87], [1098.66, 681.58], [1095.18, 682.22], [1091.69, 682.81], [1088.19, 683.32], [1084.68, 683.78], [1081.18, 684.24], [1077.67, 684.73], [1074.16, 685.16], [1070.65, 685.58], [1067.14, 686.06], [1063.64, 686.56], [1060.14, 687.11], [1056.66, 687.76], [1053.21, 688.51], [1049.77, 689.35], [1046.36, 690.3], [1042.99, 691.37], [1039.66, 692.56], [1036.36, 693.86], [1033.12, 695.27], [1029.93, 696.79], [1026.79, 698.43], [1023.7, 700.16], [1020.68, 702.01], [1017.73, 703.95], [1014.84, 706.0], [1012.03, 708.14], [1009.29, 710.38], [1006.63, 712.71], [1004.05, 715.14], [1001.55, 717.64], [999.15, 720.24], [996.83, 722.91], [994.61, 725.67], [992.48, 728.49], [990.45, 731.39], [988.52, 734.36], [986.7, 737.39], [984.98, 740.48], [983.36, 743.63], [981.86, 746.83], [980.47, 750.09], [979.19, 753.38], [978.02, 756.72], [976.97, 760.1], [976.03, 763.52], [975.22, 766.96], [974.52, 770.43], [973.94, 773.92], [973.48, 777.43], [973.15, 780.95], [972.93, 784.48], [972.84, 788.02], [972.86, 791.55], [973.01, 795.09], [973.28, 798.62], [973.67, 802.13], [974.18, 805.64], [974.8, 809.12], [975.53, 812.58], [976.36, 816.02], [977.29, 819.43], [978.35, 822.81], [979.52, 826.15], [980.8, 829.45], [982.19, 832.7], [983.69, 835.91], [985.25, 839.08], [986.91, 842.21], [988.68, 845.27], [990.51, 848.3], [992.4, 851.29], [994.36, 854.24], [996.4, 857.13], [998.48, 859.99], [1000.59, 862.83], [1002.73, 865.64], [1004.89, 868.45], [1007.1, 871.21], [1009.34, 873.95], [1011.62, 876.66], [1013.94, 879.33], [1016.26, 882.0], [1018.55, 884.7], [1020.85, 887.38], [1023.17, 890.06], [1025.46, 892.75], [1027.71, 895.49], [1029.89, 898.27], [1031.97, 901.13], [1033.99, 904.04], [1035.91, 907.01], [1037.79, 910.01], [1039.63, 913.03], [1041.47, 916.05], [1043.3, 919.08], [1045.14, 922.11], [1046.95, 925.15], [1048.76, 928.18], [1050.58, 931.22], [1052.47, 934.21], [1054.43, 937.16], [1056.41, 940.09], [1058.41, 943.01], [1060.43, 945.91], [1062.47, 948.8], [1064.54, 951.68], [1066.63, 954.53], [1068.71, 957.39], [1070.8, 960.25], [1072.81, 963.16], [1074.78, 966.09], [1076.68, 969.08], [1078.51, 972.11], [1080.3, 975.16], [1082.08, 978.22], [1083.82, 981.3], [1085.53, 984.4], [1087.22, 987.5], [1088.89, 990.62], [1090.48, 993.79], [1092.06, 996.95], [1093.69, 1000.09], [1095.41, 1003.19], [1097.17, 1006.25], [1099.02, 1009.27], [1100.87, 1012.29], [1102.71, 1015.31], [1104.48, 1018.37], [1106.19, 1021.47], [1107.82, 1024.61], [1109.39, 1027.78], [1110.94, 1030.96], [1112.46, 1034.15], [1113.91, 1037.38], [1115.3, 1040.64], [1116.69, 1043.89], [1118.05, 1047.16], [1119.4, 1050.43], [1120.72, 1053.71], [1122.09, 1056.97], [1123.48, 1060.23], [1124.91, 1063.46], [1126.37, 1066.69], [1127.83, 1069.91], [1129.24, 1073.15], [1130.66, 1076.39], [1132.06, 1079.65], [1133.41, 1082.92], [1134.79, 1086.17], [1136.19, 1089.42], [1137.61, 1092.67], [1139.04, 1095.9], [1140.49, 1099.13], [1141.92, 1102.37], [1143.29, 1105.63], [1144.64, 1108.9], [1145.95, 1112.19], [1147.25, 1115.47], [1148.56, 1118.76], [1149.85, 1122.06], [1151.13, 1125.35], [1152.42, 1128.65], [1153.7, 1131.95], [1154.97, 1135.25], [1156.27, 1138.54], [1157.61, 1141.82], [1158.99, 1145.07], [1160.42, 1148.31], [1161.93, 1151.51], [1163.56, 1154.65], [1165.28, 1157.74], [1167.08, 1160.79], [1168.95, 1163.79], [1170.85, 1166.78], [1172.81, 1169.72], [1174.82, 1172.64], [1176.88, 1175.51], [1179.01, 1178.33], [1181.19, 1181.12], [1183.39, 1183.89], [1185.58, 1186.68], [1187.71, 1189.5], [1189.79, 1192.36], [1191.85, 1195.24], [1193.87, 1198.14], [1195.89, 1201.05], [1197.91, 1203.95], [1199.96, 1206.84], [1202.01, 1209.72], [1204.08, 1212.59], [1206.19, 1215.43], [1208.36, 1218.22], [1210.56, 1221.0], [1212.79, 1223.74], [1215.03, 1226.48], [1217.27, 1229.22], [1219.54, 1231.94], [1221.84, 1234.62], [1224.16, 1237.29], [1226.49, 1239.96], [1228.86, 1242.58], [1231.22, 1245.22], [1233.6, 1247.84], [1235.96, 1250.48], [1238.33, 1253.1], [1240.75, 1255.69], [1243.12, 1258.31], [1245.48, 1260.95], [1247.81, 1263.61], [1250.11, 1266.3], [1252.37, 1269.02], [1254.64, 1271.74], [1256.82, 1274.52], [1258.99, 1277.32], [1261.14, 1280.13], [1263.19, 1283.01], [1265.14, 1285.96], [1267.02, 1288.96], [1268.83, 1292.0], [1270.63, 1295.05], [1272.37, 1298.13], [1274.06, 1301.24], [1275.65, 1304.4], [1277.21, 1307.58], [1278.77, 1310.75], [1280.32, 1313.93], [1281.85, 1317.12], [1283.35, 1320.33], [1284.8, 1323.55], [1286.24, 1326.78], [1287.65, 1330.03], [1289.05, 1333.28], [1290.48, 1336.51], [1292.0, 1339.71], [1293.56, 1342.88], [1295.21, 1346.02], [1296.94, 1349.1], [1298.72, 1352.16], [1300.48, 1355.23], [1302.2, 1358.32], [1303.87, 1361.44], [1305.47, 1364.6], [1307.04, 1367.77], [1308.56, 1370.97], [1310.04, 1374.18], [1311.51, 1377.4], [1312.93, 1380.64], [1314.33, 1383.89], [1315.71, 1387.14], [1317.1, 1390.4], [1318.46, 1393.66], [1319.82, 1396.93], [1321.14, 1400.21], [1322.5, 1403.48], [1323.86, 1406.75], [1325.3, 1409.98], [1326.8, 1413.18], [1328.36, 1416.36], [1329.97, 1419.51], [1331.47, 1422.71], [1332.87, 1425.97], [1334.15, 1429.26], [1335.31, 1432.61], [1336.36, 1435.98], [1337.29, 1439.4], [1338.11, 1442.84], [1338.8, 1446.31], [1339.38, 1449.8], [1339.84, 1453.31], [1340.17, 1456.83], [1340.39, 1460.36], [1340.48, 1463.9], [1340.45, 1467.44], [1340.3, 1470.97], [1340.03, 1474.5], [1339.64, 1478.02], [1339.13, 1481.52], [1338.5, 1485.0], [1337.74, 1488.46], [1336.87, 1491.89], [1335.89, 1495.28], [1334.78, 1498.65], [1333.57, 1501.97], [1332.23, 1505.25], [1330.79, 1508.48], [1329.24, 1511.66], [1327.58, 1514.78], [1325.81, 1517.84], [1323.94, 1520.85], [1321.96, 1523.78], [1319.89, 1526.65], [1317.72, 1529.44], [1315.45, 1532.16], [1313.09, 1534.8], [1310.65, 1537.35], [1308.11, 1539.82], [1305.5, 1542.21], [1302.8, 1544.5], [1300.02, 1546.69], [1297.18, 1548.79], [1294.26, 1550.79], [1291.27, 1552.69], [1288.22, 1554.49], [1285.12, 1556.18], [1281.95, 1557.76], [1278.73, 1559.23], [1275.47, 1560.59], [1272.16, 1561.84], [1268.81, 1562.98], [1265.42, 1564.0], [1262.0, 1564.9], [1258.54, 1565.68], [1255.07, 1566.34], [1251.57, 1566.89], [1248.06, 1567.31], [1244.54, 1567.61], [1241.0, 1567.8], [1237.46, 1567.86], [1233.93, 1567.8], [1230.39, 1567.61], [1226.87, 1567.31], [1223.35, 1566.88], [1219.86, 1566.34], [1216.38, 1565.67], [1212.93, 1564.89], [1209.51, 1563.99], [1206.12, 1562.97], [1202.77, 1561.83], [1199.46, 1560.58], [1196.2, 1559.22], [1192.98, 1557.75], [1189.82, 1556.17], [1186.71, 1554.48], [1183.66, 1552.68], [1180.67, 1550.78], [1177.76, 1548.78], [1174.91, 1546.68], [1172.14, 1544.48], [1169.44, 1542.19], [1166.82, 1539.81], [1164.29, 1537.34], [1161.84, 1534.78], [1159.49, 1532.14], [1157.22, 1529.42], [1155.05, 1526.63], [1152.98, 1523.76], [1151.01, 1520.82], [1149.13, 1517.82], [1147.37, 1514.76], [1145.71, 1511.63], [1144.15, 1508.45], [1142.71, 1505.22], [1141.37, 1501.95], [1140.11, 1498.64], [1138.92, 1495.31], [1137.85, 1491.93], [1136.9, 1488.53], [1136.07, 1485.09], [1135.34, 1481.63], [1134.65, 1478.16], [1134.0, 1474.68], [1133.4, 1471.19], [1132.87, 1467.69], [1132.4, 1464.19], [1131.99, 1460.67], [1131.67, 1457.15], [1131.45, 1453.62], [1131.29, 1450.08], [1131.2, 1446.54], [1131.13, 1443.01], [1131.0, 1439.47], [1130.87, 1435.93], [1130.67, 1432.4], [1130.42, 1428.87], [1130.19, 1425.34], [1129.95, 1421.81], [1129.7, 1418.28], [1129.45, 1414.75], [1129.22, 1411.22], [1129.04, 1407.69], [1128.9, 1404.15], [1128.82, 1400.62], [1128.76, 1397.08], [1128.76, 1393.54], [1128.79, 1390.0], [1128.83, 1386.46], [1128.9, 1382.93], [1129.0, 1379.39], [1129.12, 1375.85], [1129.2, 1372.32], [1129.27, 1368.78], [1129.35, 1365.24], [1129.38, 1361.7], [1129.28, 1358.17], [1129.07, 1354.63], [1128.73, 1351.11], [1128.27, 1347.6], [1127.69, 1344.11], [1126.99, 1340.64], [1126.18, 1337.2], [1125.26, 1333.78], [1124.23, 1330.4], [1123.12, 1327.04], [1121.97, 1323.69], [1120.82, 1320.35], [1119.65, 1317.01], [1118.45, 1313.68], [1117.24, 1310.36], [1115.96, 1307.06], [1114.65, 1303.77], [1113.24, 1300.53], [1111.72, 1297.33], [1110.13, 1294.17], [1108.5, 1291.03], [1106.76, 1287.95], [1104.96, 1284.9], [1103.08, 1281.9], [1101.1, 1278.97], [1099.02, 1276.11], [1096.88, 1273.29], [1094.66, 1270.54], [1092.39, 1267.82], [1090.08, 1265.14], [1087.74, 1262.49], [1085.35, 1259.88], [1082.92, 1257.31], [1080.46, 1254.77], [1077.97, 1252.25], [1075.45, 1249.77], [1072.94, 1247.27], [1070.43, 1244.78], [1067.91, 1242.29], [1065.41, 1239.79], [1062.93, 1237.26], [1060.43, 1234.76], [1057.93, 1232.26], [1055.4, 1229.79], [1052.89, 1227.29], [1050.37, 1224.81], [1047.83, 1222.35], [1045.26, 1219.92], [1042.67, 1217.51], [1040.06, 1215.11], [1037.46, 1212.71], [1034.83, 1210.35], [1032.2, 1207.98], [1029.56, 1205.62], [1026.93, 1203.26], [1024.35, 1200.83], [1021.79, 1198.39], [1019.24, 1195.94], [1016.71, 1193.47], [1014.21, 1190.96], [1011.75, 1188.42], [1009.28, 1185.88], [1006.81, 1183.36], [1004.31, 1180.85], [1001.8, 1178.36], [999.29, 1175.87], [996.79, 1173.36], [994.34, 1170.81], [991.92, 1168.22], [989.57, 1165.58], [987.26, 1162.9], [984.99, 1160.18], [982.79, 1157.41], [980.65, 1154.6], [978.56, 1151.74], [976.52, 1148.85], [974.47, 1145.97], [972.44, 1143.07], [970.45, 1140.14], [968.46, 1137.22], [966.47, 1134.29], [964.48, 1131.36], [962.53, 1128.41], [960.63, 1125.43], [958.7, 1122.46], [956.78, 1119.49], [954.91, 1116.48], [953.07, 1113.46], [951.25, 1110.43], [949.45, 1107.38], [947.65, 1104.34], [945.83, 1101.31], [943.95, 1098.31], [942.05, 1095.32], [940.15, 1092.34], [938.2, 1089.38], [936.2, 1086.47], [934.11, 1083.61], [931.97, 1080.8], [929.76, 1078.03], [927.52, 1075.29], [925.21, 1072.61], [922.88, 1069.94], [920.65, 1067.2], [918.47, 1064.41], [916.34, 1061.59], [914.29, 1058.71], [912.29, 1055.79], [910.33, 1052.84], [908.4, 1049.88], [906.46, 1046.91], [904.53, 1043.95], [902.56, 1041.01], [900.53, 1038.12], [898.41, 1035.28], [896.21, 1032.51], [893.96, 1029.78], [891.68, 1027.07], [889.36, 1024.4], [886.99, 1021.77], [884.58, 1019.18], [882.14, 1016.62], [879.71, 1014.05], [877.32, 1011.44], [874.99, 1008.78], [872.71, 1006.07], [870.52, 1003.29], [868.39, 1000.47], [866.31, 997.6], [864.26, 994.72], [862.23, 991.82], [860.24, 988.9], [858.31, 985.93], [856.43, 982.93], [854.54, 979.95], [852.67, 976.94], [850.77, 973.96], [848.87, 970.97], [846.98, 967.98], [845.09, 964.99], [843.23, 961.98], [841.38, 958.96], [839.51, 955.96], [837.61, 952.98], [835.7, 950.0], [833.7, 947.07], [831.64, 944.2], [829.55, 941.34], [827.45, 938.5], [825.28, 935.7], [823.06, 932.95], [820.78, 930.25], [818.46, 927.57], [816.1, 924.94], [813.69, 922.35], [811.23, 919.8], [808.74, 917.28], [806.26, 914.77], [803.75, 912.26], [801.28, 909.73], [798.85, 907.16], [796.45, 904.56], [794.1, 901.92], [791.84, 899.2], [789.67, 896.4], [787.59, 893.54], [785.57, 890.63], [783.6, 887.7], [781.65, 884.74], [779.76, 881.75], [777.91, 878.73], [776.04, 875.73], [774.15, 872.74], [772.26, 869.75], [770.34, 866.78], [768.38, 863.83], [766.45, 860.87], [764.51, 857.9], [762.62, 854.92], [760.68, 851.95], [758.74, 849.0], [756.74, 846.08], [754.7, 843.19], [752.59, 840.35], [750.39, 837.57], [748.21, 834.79], [746.05, 831.99], [743.87, 829.2], [741.61, 826.48], [739.32, 823.78], [736.99, 821.12], [734.66, 818.45], [732.37, 815.76], [730.06, 813.08], [727.76, 810.39], [725.47, 807.69], [723.16, 805.01], [720.82, 802.35], [718.44, 799.74], [716.04, 797.14], [713.63, 794.55], [711.19, 791.99], [708.72, 789.45], [706.23, 786.94], [703.74, 784.42], [701.24, 781.92], [698.78, 779.38], [696.35, 776.81], [693.97, 774.19], [691.64, 771.53], [689.37, 768.81], [687.15, 766.06], [685.0, 763.25], [682.91, 760.39], [680.88, 757.49], [678.96, 754.52], [677.06, 751.54], [675.17, 748.55], [673.3, 745.54], [671.45, 742.53], [669.6, 739.51], [667.76, 736.49], [665.87, 733.5], [663.94, 730.53], [662.0, 727.57], [660.07, 724.61], [658.13, 721.65], [656.16, 718.71], [654.18, 715.77], [652.21, 712.83], [650.29, 709.86], [648.39, 706.88], [646.51, 703.88], [644.68, 700.85], [642.9, 697.8], [641.13, 694.73], [639.39, 691.65], [637.67, 688.56], [635.94, 685.48], [634.14, 682.43], [632.25, 679.44], [630.29, 676.49], [628.27, 673.59], [626.19, 670.73], [624.09, 667.88], [621.94, 665.06], [619.77, 662.27], [617.58, 659.49], [615.36, 656.73], [613.15, 653.97], [610.95, 651.2], [608.79, 648.4], [606.7, 645.54], [604.68, 642.64], [602.73, 639.69], [600.87, 636.68], [599.07, 633.63], [597.32, 630.56], [595.63, 627.45], [593.94, 624.34], [592.27, 621.22], [590.55, 618.13], [588.75, 615.08], [586.88, 612.08], [584.97, 609.1], [583.03, 606.14], [581.07, 603.2], [579.16, 600.22], [577.28, 597.22], [575.45, 594.19], [573.66, 591.14], [571.94, 588.05], [570.2, 584.96], [568.46, 581.88], [566.65, 578.84], [564.73, 575.87], [562.74, 572.95], [560.64, 570.1], [558.45, 567.32], [556.16, 564.62], [553.78, 562.0], [551.32, 559.46], [548.77, 557.01], [546.13, 554.65], [543.42, 552.38], [540.63, 550.21], [537.76, 548.13], [534.83, 546.15], [531.83, 544.27], [528.77, 542.49], [525.68, 540.78], [522.54, 539.13], [519.4, 537.51], [516.2, 535.99], [513.0, 534.49], [509.78, 533.02], [506.55, 531.58], [503.3, 530.19], [500.01, 528.88], [496.7, 527.62], [493.39, 526.39], [490.06, 525.18], [486.73, 524.0], [483.38, 522.86], [480.02, 521.74], [476.65, 520.66], [473.27, 519.63], [469.88, 518.62], [466.46, 517.68], [463.03, 516.82], [459.57, 516.09], [456.09, 515.46], [452.59, 514.95], [449.07, 514.56], [445.54, 514.29], [442.01, 514.12], [438.47, 514.07], [434.93, 514.13], [431.4, 514.32], [427.87, 514.63], [424.36, 515.06], [420.87, 515.61], [417.39, 516.28], [413.94, 517.07], [410.52, 517.98], [407.14, 519.0], [403.79, 520.14], [400.48, 521.4], [397.21, 522.76], [394.0, 524.24], [390.84, 525.83], [387.73, 527.52], [384.69, 529.32], [381.71, 531.23], [378.79, 533.23], [375.95, 535.34], [373.18, 537.54], [370.48, 539.84], [367.87, 542.22], [365.34, 544.7], [362.9, 547.26], [360.55, 549.9], [358.28, 552.62], [356.12, 555.42], [354.05, 558.29], [352.08, 561.23], [350.21, 564.23], [348.45, 567.3], [346.79, 570.43], [345.25, 573.61], [343.81, 576.84], [342.48, 580.12], [341.27, 583.45], [340.17, 586.81], [339.19, 590.21], [338.33, 593.64], [337.58, 597.1], [336.96, 600.58], [336.44, 604.08], [336.04, 607.6], [335.75, 611.12], [335.59, 614.66], [335.55, 618.2], [335.63, 621.73], [335.83, 625.27], [336.06, 628.8], [336.38, 632.32], [336.77, 635.84], [337.24, 639.35], [337.79, 642.84], [338.43, 646.32], [339.19, 649.78], [340.03, 653.21], [340.95, 656.63], [341.96, 660.02], [343.07, 663.38], [344.21, 666.73], [345.29, 670.1], [346.32, 673.49], [347.34, 676.87], [348.37, 680.26], [349.44, 683.63], [350.58, 686.98], [351.82, 690.29], [353.16, 693.57], [354.56, 696.82], [356.07, 700.02], [357.65, 703.18], [359.29, 706.32], [360.96, 709.44], [362.63, 712.56], [364.29, 715.68], [365.98, 718.79], [367.64, 721.91], [369.27, 725.05], [370.86, 728.21], [372.48, 731.36], [374.12, 734.5], [375.8, 737.61], [377.49, 740.72], [379.16, 743.84], [380.85, 746.94], [382.51, 750.07], [384.16, 753.2], [385.84, 756.32], [387.54, 759.42], [389.24, 762.52], [390.93, 765.63], [392.58, 768.76], [394.16, 771.92], [395.71, 775.11], [397.17, 778.33], [398.61, 781.56], [400.0, 784.82], [401.36, 788.08], [402.7, 791.36], [404.02, 794.64], [405.34, 797.92], [406.7, 801.19], [408.09, 804.44], [409.48, 807.7], [410.87, 810.95], [412.31, 814.18], [413.84, 817.37], [415.34, 820.58], [416.83, 823.79], [418.31, 827.0], [419.76, 830.23], [421.23, 833.45], [422.69, 836.67], [424.08, 839.92], [425.5, 843.16], [426.99, 846.37], [428.49, 849.58], [429.97, 852.79], [431.43, 856.01], [432.84, 859.26], [434.25, 862.5], [435.64, 865.76], [437.03, 869.01], [438.45, 872.25], [439.89, 875.49], [441.37, 878.7], [442.83, 881.92], [444.26, 885.16], [445.63, 888.42], [446.96, 891.7], [448.25, 894.99], [449.48, 898.31], [450.68, 901.64], [451.89, 904.96], [453.15, 908.27], [454.41, 911.58], [455.7, 914.87], [456.97, 918.18], [458.3, 921.45], [459.67, 924.72], [461.08, 927.96], [462.53, 931.19], [464.06, 934.38], [465.64, 937.54], [467.23, 940.71], [468.8, 943.87], [470.44, 947.01], [472.06, 950.16], [473.65, 953.32], [475.22, 956.49], [476.78, 959.66], [478.35, 962.83], [479.97, 965.98], [481.64, 969.1], [483.33, 972.21], [484.98, 975.34], [486.56, 978.5], [488.11, 981.69], [489.57, 984.91], [490.94, 988.17], [492.25, 991.46], [493.52, 994.76], [494.8, 998.06], [496.07, 1001.36], [497.31, 1004.67], [498.51, 1008.0], [499.69, 1011.34], [500.82, 1014.69], [501.93, 1018.05], [503.01, 1021.42], [504.09, 1024.79], [505.17, 1028.16], [506.24, 1031.53], [507.29, 1034.91], [508.32, 1038.3], [509.32, 1041.69], [510.26, 1045.1], [511.2, 1048.51], [512.15, 1051.92], [513.12, 1055.32], [514.1, 1058.72], [515.09, 1062.12], [516.1, 1065.51], [517.17, 1068.88], [518.32, 1072.23], [519.55, 1075.55], [520.89, 1078.82], [522.31, 1082.06], [523.77, 1085.29], [525.19, 1088.52], [526.56, 1091.79], [527.89, 1095.07], [529.23, 1098.34], [530.55, 1101.62], [531.92, 1104.89], [533.28, 1108.16], [534.69, 1111.4], [536.15, 1114.62], [537.6, 1117.85], [538.94, 1121.13], [540.16, 1124.45], [541.27, 1127.81], [542.26, 1131.2], [543.13, 1134.63], [543.89, 1138.09], [544.53, 1141.57], [545.05, 1145.07], [545.44, 1148.58], [545.72, 1152.11], [545.87, 1155.65], [545.91, 1159.18], [545.82, 1162.72], [545.61, 1166.25], [545.28, 1169.78], [544.83, 1173.28], [544.27, 1176.78], [543.62, 1180.26], [542.84, 1183.71], [542.05, 1187.16], [541.16, 1190.58], [540.24, 1194.0], [539.24, 1197.39], [538.2, 1200.77], [537.05, 1204.12], [535.83, 1207.44], [534.52, 1210.73], [533.15, 1213.99], [531.79, 1217.26], [530.44, 1220.53], [529.15, 1223.82], [527.92, 1227.14], [526.76, 1230.48], [525.67, 1233.85], [524.64, 1237.23], [523.56, 1240.6], [522.49, 1243.98], [521.42, 1247.35], [520.4, 1250.74], [519.44, 1254.14], [518.37, 1257.51], [517.17, 1260.84], [515.87, 1264.13], [514.45, 1267.37], [512.92, 1270.57], [511.28, 1273.7], [509.54, 1276.78], [507.69, 1279.8], [505.74, 1282.75], [503.69, 1285.63], [501.54, 1288.44], [499.3, 1291.18], [496.96, 1293.83], [494.53, 1296.41], [492.02, 1298.9], [489.42, 1301.3], [486.74, 1303.61], [483.98, 1305.83], [481.15, 1307.95], [478.25, 1309.98], [475.28, 1311.9], [472.25, 1313.72], [469.15, 1315.43], [466.0, 1317.04], [462.79, 1318.54], [459.54, 1319.93], [456.24, 1321.2], [452.89, 1322.36], [449.51, 1323.4], [446.1, 1324.33], [442.66, 1325.14], [439.19, 1325.83], [435.69, 1326.41], [432.18, 1326.86], [428.66, 1327.19], [425.13, 1327.4], [421.59, 1327.49], [418.05, 1327.45], [414.52, 1327.3], [410.99, 1327.02], [407.48, 1326.62], [403.98, 1326.11], [400.5, 1325.47], [397.04, 1324.71], [393.61, 1323.84], [390.21, 1322.84], [386.85, 1321.74], [383.53, 1320.51], [380.26, 1319.18], [377.03, 1317.73], [373.85, 1316.17], [370.73, 1314.5], [367.67, 1312.73], [364.67, 1310.85], [361.74, 1308.88], [358.87, 1306.8], [356.08, 1304.62], [353.37, 1302.35], [350.73, 1299.99], [348.18, 1297.54], [345.71, 1295.0], [343.34, 1292.38], [341.05, 1289.68], [338.86, 1286.91], [336.76, 1284.05], [334.77, 1281.13], [332.87, 1278.15], [331.08, 1275.09], [329.39, 1271.98], [327.82, 1268.82], [326.35, 1265.6], [324.99, 1262.33], [323.75, 1259.02], [322.62, 1255.66], [321.61, 1252.27], [320.71, 1248.85], [319.93, 1245.4], [319.27, 1241.92], [318.74, 1238.42], [318.32, 1234.91], [318.02, 1231.39], [317.84, 1227.85], [317.79, 1224.31], [317.85, 1220.78], [318.04, 1217.24], [318.35, 1213.72], [318.78, 1210.21], [319.33, 1206.71], [320.0, 1203.24], [320.79, 1199.79], [321.7, 1196.37], [322.72, 1192.98], [323.86, 1189.63], [325.12, 1186.32], [326.48, 1183.06], [327.96, 1179.84], [329.55, 1176.68], [331.24, 1173.58], [333.04, 1170.53], [334.95, 1167.55], [336.95, 1164.63], [339.06, 1161.79], [341.26, 1159.02], [343.56, 1156.33], [345.76, 1153.56], [347.86, 1150.71], [349.87, 1147.8], [351.77, 1144.82], [353.57, 1141.77], [355.27, 1138.66], [356.86, 1135.5], [358.33, 1132.29], [359.7, 1129.02], [360.95, 1125.72], [362.09, 1122.37], [363.12, 1118.98], [364.03, 1115.56], [364.81, 1112.11], [365.48, 1108.64], [366.03, 1105.14], [366.46, 1101.63], [366.77, 1098.1], [366.96, 1094.57], [367.03, 1091.03], [366.97, 1087.5], [366.8, 1083.96], [366.5, 1080.44], [366.08, 1076.92], [365.54, 1073.43], [364.88, 1069.95], [364.1, 1066.5], [363.21, 1063.07], [362.2, 1059.68], [361.07, 1056.33], [359.82, 1053.02], [358.47, 1049.75], [357.0, 1046.53], [355.42, 1043.36], [353.74, 1040.25], [351.94, 1037.2], [350.05, 1034.21], [348.05, 1031.29], [345.96, 1028.44], [343.77, 1025.66], [341.48, 1022.96], [339.1, 1020.34], [336.63, 1017.81], [334.08, 1015.36], [331.45, 1012.99], [328.73, 1010.72], [325.94, 1008.55], [323.08, 1006.47], [320.15, 1004.49], [317.15, 1002.62], [314.08, 1000.84], [310.96, 999.18], [307.79, 997.62], [304.56, 996.17], [301.28, 994.83], [297.96, 993.61], [294.6, 992.5], [291.21, 991.51], [287.78, 990.64], [284.32, 989.88], [280.84, 989.24], [277.34, 988.72], [273.82, 988.33], [270.3, 988.05], [266.76, 987.9], [263.22, 987.86], [259.69, 987.95], [256.15, 988.16], [252.63, 988.49], [249.12, 988.94], [245.63, 989.49], [242.15, 990.15], [238.7, 990.94], [235.28, 991.84], [231.89, 992.86], [228.54, 993.98], [225.21, 995.2], [221.94, 996.53], [218.7, 997.96], [215.51, 999.49], [212.36, 1001.11], [209.24, 1002.76], [206.13, 1004.46], [203.02, 1006.15], [199.9, 1007.81], [196.75, 1009.43], [193.59, 1011.02], [190.42, 1012.59], [187.23, 1014.12], [184.01, 1015.6], [180.78, 1017.02], [177.53, 1018.43], [174.29, 1019.84], [171.02, 1021.21], [167.75, 1022.55], [164.47, 1023.89], [161.2, 1025.23], [157.91, 1026.54], [154.62, 1027.85], [151.34, 1029.17], [148.07, 1030.53], [144.81, 1031.88], [141.53, 1033.21], [138.24, 1034.53], [134.97, 1035.87], [131.72, 1037.27], [128.49, 1038.71], [125.25, 1040.13], [122.0, 1041.53], [118.78, 1042.99], [115.54, 1044.43], [112.3, 1045.84], [109.04, 1047.23], [105.78, 1048.6], [102.52, 1049.96], [99.29, 1051.42], [96.09, 1052.94], [92.93, 1054.53], [89.8, 1056.18], [86.69, 1057.86], [83.6, 1059.57], [80.52, 1061.31], [77.47, 1063.11], [74.43, 1064.93], [71.42, 1066.79], [68.41, 1068.64], [65.34, 1070.41], [62.27, 1072.17], [59.2, 1073.93], [56.14, 1075.7], [53.05, 1077.43]], [[1462.59, 70.0], [1462.88, 73.14], [1463.16, 76.27], [1463.44, 79.41], [1463.73, 82.54], [1464.01, 85.68], [1464.3, 88.81], [1464.58, 91.95], [1464.86, 95.08], [1465.15, 98.22], [1465.43, 101.35], [1465.72, 104.49], [1466.0, 107.62], [1466.28, 110.76], [1466.57, 113.89], [1466.85, 117.03], [1467.13, 120.16], [1467.42, 123.3], [1467.7, 126.43], [1467.99, 129.57], [1468.27, 132.7], [1468.55, 135.84], [1468.84, 138.97], [1469.12, 142.11], [1469.41, 145.24], [1469.69, 148.38], [1469.97, 151.51], [1470.26, 154.65], [1470.54, 157.78], [1470.82, 160.92], [1471.11, 164.05], [1471.39, 167.19], [1471.68, 170.32], [1471.96, 173.46], [1472.24, 176.59], [1472.53, 179.73], [1472.81, 182.86], [1473.1, 186.0], [1473.38, 189.13], [1473.66, 192.27], [1473.95, 195.4], [1474.23, 198.54], [1474.51, 201.67], [1474.8, 204.81], [1475.08, 207.94], [1475.37, 211.08], [1475.65, 214.21], [1475.93, 217.35], [1476.11, 220.49], [1476.18, 223.64], [1476.19, 226.79], [1476.16, 229.93], [1476.07, 233.08], [1475.95, 236.23], [1475.8, 239.37], [1475.64, 242.51], [1475.46, 245.66], [1475.23, 248.8], [1474.89, 251.93], [1474.45, 255.04], [1473.95, 258.15], [1473.42, 261.25], [1472.85, 264.35], [1472.23, 267.43], [1471.59, 270.52], [1470.95, 273.6], [1470.27, 276.67], [1469.54, 279.73], [1468.77, 282.79], [1468.05, 285.85], [1467.33, 288.92], [1466.61, 291.98], [1465.84, 295.03], [1465.05, 298.08], [1464.21, 301.11], [1463.31, 304.13], [1462.35, 307.13], [1461.39, 310.13], [1460.49, 313.14], [1459.59, 316.16], [1458.74, 319.19], [1457.93, 322.23], [1457.14, 325.28], [1456.3, 328.31], [1455.45, 331.34], [1454.58, 334.37], [1453.67, 337.38], [1452.68, 340.37], [1451.64, 343.34], [1450.57, 346.3], [1449.45, 349.24], [1448.27, 352.16], [1447.02, 355.05], [1445.71, 357.92], [1444.33, 360.74], [1442.88, 363.54], [1441.4, 366.31], [1439.86, 369.06], [1438.26, 371.77], [1436.62, 374.46], [1434.92, 377.11], [1433.25, 379.78], [1431.59, 382.45], [1429.89, 385.1], [1428.15, 387.73], [1426.4, 390.34], [1424.54, 392.88], [1422.59, 395.35], [1420.55, 397.75], [1418.42, 400.07], [1416.21, 402.31], [1413.93, 404.48], [1411.57, 406.56], [1409.19, 408.62], [1406.79, 410.66], [1404.39, 412.69], [1401.97, 414.71], [1399.56, 416.74], [1397.12, 418.72], [1394.67, 420.7], [1392.2, 422.65], [1389.74, 424.62], [1387.26, 426.55], [1384.77, 428.48], [1382.27, 430.39], [1379.73, 432.25], [1377.19, 434.11], [1374.65, 435.97], [1372.12, 437.84], [1369.57, 439.69], [1367.02, 441.53], [1364.46, 443.37], [1361.89, 445.19], [1359.31, 446.99], [1356.73, 448.79], [1354.13, 450.57], [1351.54, 452.35], [1348.91, 454.08], [1346.25, 455.77], [1343.56, 457.4], [1340.83, 458.98], [1338.05, 460.45], [1335.27, 461.92], [1332.46, 463.33], [1329.6, 464.66], [1326.7, 465.88], [1323.76, 467.0], [1320.8, 468.07], [1317.81, 469.07], [1314.8, 469.99], [1311.77, 470.85], [1308.73, 471.65], [1305.67, 472.39], [1302.61, 473.12], [1299.55, 473.85], [1296.49, 474.6], [1293.43, 475.35], [1290.38, 476.12], [1287.32, 476.87], [1284.27, 477.65], [1281.23, 478.46], [1278.17, 479.2], [1275.1, 479.9], [1272.02, 480.56], [1268.94, 481.18], [1265.85, 481.81], [1262.77, 482.47], [1259.7, 483.13], [1256.63, 483.84], [1253.58, 484.63], [1250.54, 485.46], [1247.51, 486.29], [1244.47, 487.11], [1241.42, 487.9], [1238.38, 488.71], [1235.34, 489.53], [1232.3, 490.35], [1229.27, 491.19], [1226.23, 492.02], [1223.2, 492.87], [1220.19, 493.77], [1217.2, 494.77], [1214.24, 495.84], [1211.31, 497.0], [1208.41, 498.23], [1205.55, 499.53], [1202.72, 500.92], [1199.91, 502.34], [1197.13, 503.8], [1194.35, 505.28], [1191.59, 506.79], [1188.82, 508.29], [1186.03, 509.75], [1183.19, 511.11], [1180.33, 512.41], [1177.44, 513.66], [1174.52, 514.84], [1171.57, 515.95], [1168.6, 516.98], [1165.6, 517.96], [1162.6, 518.88], [1159.57, 519.75], [1156.53, 520.58], [1153.48, 521.36], [1150.42, 522.1], [1147.36, 522.81], [1144.28, 523.49], [1141.2, 524.14], [1138.12, 524.79], [1135.05, 525.47], [1131.98, 526.17], [1128.92, 526.89], [1125.86, 527.65], [1122.82, 528.45], [1119.78, 529.29], [1116.77, 530.2], [1113.77, 531.16], [1110.79, 532.17], [1107.82, 533.22], [1104.86, 534.29], [1101.89, 535.33], [1098.91, 536.35], [1095.93, 537.35], [1092.93, 538.32], [1089.94, 539.28], [1086.92, 540.19], [1083.91, 541.1], [1080.89, 541.98], [1077.84, 542.79], [1074.79, 543.56], [1071.73, 544.29], [1068.67, 545.02], [1065.6, 545.73], [1062.54, 546.46], [1059.47, 547.16], [1056.4, 547.87], [1053.34, 548.6], [1050.29, 549.36], [1047.23, 550.11], [1044.18, 550.89], [1041.14, 551.69], [1038.09, 552.47], [1035.04, 553.25], [1031.99, 554.05], [1028.96, 554.9], [1025.94, 555.77], [1022.93, 556.69], [1019.95, 557.71], [1016.98, 558.78], [1014.03, 559.86], [1011.08, 560.96], [1008.14, 562.08], [1005.19, 563.19], [1002.23, 564.25], [999.25, 565.26], [996.25, 566.23], [993.24, 567.14], [990.22, 568.02], [987.19, 568.88], [984.15, 569.7], [981.11, 570.51], [978.07, 571.35], [975.03, 572.16], [971.99, 572.99], [968.97, 573.85], [965.94, 574.72], [962.93, 575.62], [959.93, 576.58], [956.96, 577.64], [954.03, 578.78], [951.12, 579.99], [948.26, 581.3], [945.42, 582.67], [942.62, 584.1], [939.84, 585.57], [937.05, 587.03], [934.29, 588.54], [931.54, 590.08], [928.83, 591.68], [926.14, 593.32], [923.49, 595.01], [920.84, 596.7], [918.2, 598.42], [915.55, 600.13], [912.89, 601.81], [910.19, 603.43], [907.47, 605.01], [904.72, 606.53], [901.93, 608.0], [899.1, 609.38], [896.24, 610.69], [893.35, 611.95], [890.44, 613.14], [887.51, 614.29], [884.57, 615.41], [881.62, 616.51], [878.68, 617.65], [875.77, 618.85], [872.91, 620.16], [870.09, 621.55], [867.29, 622.99], [864.52, 624.49], [861.77, 626.02], [859.03, 627.57], [856.3, 629.14], [853.59, 630.75], [850.91, 632.38], [848.23, 634.04], [845.53, 635.65], [842.8, 637.23], [840.04, 638.75], [837.28, 640.25], [834.5, 641.74], [831.73, 643.23], [828.98, 644.76], [826.22, 646.27], [823.48, 647.82], [820.76, 649.41], [818.08, 651.06], [815.39, 652.7], [812.69, 654.32], [810.0, 655.95], [807.31, 657.58], [804.63, 659.23], [801.95, 660.89], [799.3, 662.59], [796.68, 664.33], [794.09, 666.12], [791.52, 667.93], [788.94, 669.75], [786.38, 671.57], [783.83, 673.42], [781.29, 675.28], [778.77, 677.16], [776.24, 679.04], [773.72, 680.92], [771.2, 682.81], [768.69, 684.7], [766.18, 686.6], [763.64, 688.46], [761.1, 690.33], [758.55, 692.17], [755.99, 694.01], [753.42, 695.83], [750.83, 697.61], [748.24, 699.41], [745.65, 701.19], [743.04, 702.95], [740.44, 704.73], [737.85, 706.51], [735.26, 708.3], [732.66, 710.07], [730.06, 711.86], [727.45, 713.61], [724.84, 715.37], [722.24, 717.14], [719.62, 718.89], [717.0, 720.63], [714.35, 722.33], [711.77, 724.13], [709.25, 726.03], [706.78, 727.97], [704.34, 729.96], [701.92, 731.98], [699.52, 734.02], [697.14, 736.08], [694.83, 738.21], [692.6, 740.43], [690.45, 742.74], [688.39, 745.12], [686.43, 747.58], [684.55, 750.11], [682.78, 752.71], [681.1, 755.37], [679.49, 758.07], [677.98, 760.84], [676.54, 763.64], [675.18, 766.47], [673.89, 769.35], [672.66, 772.24], [671.5, 775.17], [670.4, 778.12], [669.36, 781.09], [668.37, 784.08], [667.37, 787.06], [666.45, 790.07], [665.58, 793.1], [664.74, 796.13], [663.93, 799.18], [663.14, 802.22], [662.36, 805.27], [661.57, 808.32], [660.81, 811.37], [660.09, 814.44], [659.34, 817.49], [658.57, 820.55], [657.76, 823.59], [656.93, 826.62], [656.06, 829.65], [655.2, 832.68], [654.39, 835.72], [653.6, 838.77], [652.81, 841.82], [652.07, 844.88], [651.35, 847.94], [650.64, 851.01], [649.96, 854.08], [649.28, 857.15], [648.6, 860.23], [647.93, 863.3], [647.25, 866.38], [646.55, 869.45], [645.8, 872.5], [645.01, 875.55], [644.23, 878.6], [643.44, 881.65], [642.67, 884.7], [641.9, 887.75], [641.1, 890.8], [640.25, 893.83], [639.34, 896.84], [638.38, 899.84], [637.36, 902.82], [636.29, 905.78], [635.12, 908.7], [633.85, 911.58], [632.57, 914.45], [631.27, 917.32], [629.93, 920.17], [628.52, 922.99], [627.06, 925.77], [625.58, 928.55], [624.02, 931.29], [622.39, 933.98], [620.71, 936.64], [618.98, 939.27], [617.24, 941.89], [615.48, 944.5], [613.75, 947.13], [612.04, 949.78], [610.36, 952.44], [608.74, 955.14], [607.13, 957.84], [605.57, 960.58], [604.04, 963.33], [602.54, 966.09], [601.08, 968.89], [599.68, 971.7], [598.3, 974.53], [596.95, 977.38], [595.59, 980.22], [594.2, 983.04], [592.81, 985.87], [591.36, 988.66], [589.92, 991.46], [588.58, 994.31], [587.25, 997.16], [585.94, 1000.02], [584.63, 1002.88], [583.31, 1005.74], [582.0, 1008.6], [580.66, 1011.45], [579.32, 1014.3], [577.99, 1017.15], [576.61, 1019.98], [575.21, 1022.8], [573.81, 1025.62], [572.39, 1028.43], [570.99, 1031.25], [569.52, 1034.03], [568.0, 1036.79], [566.4, 1039.5], [564.7, 1042.15], [562.95, 1044.77], [561.18, 1047.37], [559.47, 1050.01], [557.75, 1052.65], [555.99, 1055.26], [554.21, 1057.86], [552.4, 1060.43], [550.56, 1062.98], [548.66, 1065.5], [546.75, 1068.0], [544.86, 1070.52], [543.0, 1073.06], [541.17, 1075.62], [539.3, 1078.15], [537.38, 1080.64], [535.45, 1083.13], [533.48, 1085.59], [531.46, 1088.0], [529.37, 1090.35], [527.25, 1092.68], [525.14, 1095.01], [523.06, 1097.37], [520.99, 1099.75], [518.99, 1102.18], [517.03, 1104.65], [515.12, 1107.15], [513.23, 1109.67], [511.39, 1112.22], [509.58, 1114.8], [507.82, 1117.4], [506.1, 1120.04], [504.36, 1122.66], [502.57, 1125.26], [500.76, 1127.83], [498.87, 1130.35], [496.9, 1132.8], [494.87, 1135.21], [492.8, 1137.58], [490.73, 1139.95], [488.63, 1142.29], [486.53, 1144.64], [484.45, 1147.0], [482.37, 1149.37], [480.29, 1151.72], [478.24, 1154.12], [476.2, 1156.51], [474.2, 1158.94], [472.24, 1161.41], [470.33, 1163.91], [468.4, 1166.39], [466.47, 1168.89], [464.55, 1171.38], [462.67, 1173.9], [460.8, 1176.43], [458.91, 1178.95], [457.06, 1181.5], [455.24, 1184.07], [453.42, 1186.64], [451.55, 1189.17], [449.66, 1191.69], [447.76, 1194.2], [445.87, 1196.71], [444.07, 1199.3], [442.37, 1201.94], [440.77, 1204.66], [439.27, 1207.42], [437.88, 1210.25], [436.6, 1213.12], [435.43, 1216.04], [434.36, 1219.01], [433.41, 1222.01], [432.58, 1225.04], [431.86, 1228.11], [431.25, 1231.2], [430.76, 1234.31], [430.39, 1237.43], [430.14, 1240.57], [430.01, 1243.72], [429.99, 1246.86], [430.1, 1250.01], [430.32, 1253.15], [430.66, 1256.28], [431.12, 1259.39], [431.69, 1262.49], [432.39, 1265.56], [433.19, 1268.6], [434.12, 1271.61], [435.15, 1274.58], [436.3, 1277.52], [437.55, 1280.4], [438.92, 1283.24], [440.39, 1286.02], [441.96, 1288.75], [443.64, 1291.41], [445.41, 1294.01], [447.28, 1296.54], [449.25, 1299.0], [451.31, 1301.38], [453.45, 1303.69], [455.68, 1305.91], [458.0, 1308.04], [460.39, 1310.09], [462.81, 1312.1], [465.28, 1314.05], [467.79, 1315.96], [470.32, 1317.83], [472.91, 1319.62], [475.54, 1321.34], [478.21, 1323.01], [480.93, 1324.6], [483.68, 1326.13], [486.46, 1327.61], [489.25, 1329.06], [492.1, 1330.4], [495.0, 1331.63], [497.93, 1332.77], [500.89, 1333.86], [503.87, 1334.85], [506.88, 1335.78], [509.91, 1336.63], [512.97, 1337.37], [516.06, 1337.99], [519.16, 1338.52], [522.28, 1338.92], [525.42, 1339.23], [528.56, 1339.45], [531.7, 1339.62], [534.85, 1339.72], [537.99, 1339.75], [541.14, 1339.69], [544.28, 1339.54], [547.43, 1339.35], [550.56, 1339.07], [553.69, 1338.74], [556.82, 1338.34], [559.93, 1337.87], [563.03, 1337.36], [566.13, 1336.8], [569.22, 1336.19], [572.3, 1335.56], [575.38, 1334.89], [578.44, 1334.16], [581.49, 1333.39], [584.53, 1332.57], [587.58, 1331.76], [590.6, 1330.9], [593.64, 1330.07], [596.69, 1329.29], [599.76, 1328.58], [602.84, 1327.95], [605.95, 1327.44], [609.07, 1327.03], [612.2, 1326.73], [615.34, 1326.52], [618.49, 1326.38], [621.63, 1326.3], [624.78, 1326.26], [627.93, 1326.29], [631.08, 1326.32], [634.22, 1326.31], [637.37, 1326.24], [640.52, 1326.18], [643.67, 1326.13], [646.81, 1326.07], [649.96, 1325.99], [653.11, 1325.88], [656.25, 1325.76], [659.4, 1325.61], [662.54, 1325.42], [665.68, 1325.21], [668.82, 1325.0], [671.96, 1324.79], [675.11, 1324.65], [678.25, 1324.58], [681.4, 1324.56], [684.55, 1324.62], [687.69, 1324.77], [690.83, 1324.96], [693.98, 1325.17], [697.11, 1325.41], [700.25, 1325.7], [703.39, 1325.96], [706.52, 1326.19], [709.67, 1326.36], [712.81, 1326.53], [715.96, 1326.65], [719.1, 1326.71], [722.25, 1326.73], [725.4, 1326.74], [728.55, 1326.81], [731.69, 1326.92], [734.84, 1327.09], [737.98, 1327.3], [741.12, 1327.54], [744.25, 1327.84], [747.38, 1328.15], [750.51, 1328.5], [753.64, 1328.85], [756.77, 1329.18], [759.9, 1329.5], [763.04, 1329.78], [766.18, 1329.98], [769.32, 1330.13], [772.47, 1330.2], [775.62, 1330.2], [778.76, 1330.11], [781.91, 1329.99], [785.05, 1329.8], [788.19, 1329.56], [791.33, 1329.28], [794.46, 1329.04], [797.6, 1328.73], [800.73, 1328.39], [803.85, 1328.03], [806.97, 1327.6], [810.07, 1327.06], [813.16, 1326.46], [816.23, 1325.76], [819.29, 1325.02], [822.34, 1324.22], [825.36, 1323.33], [828.36, 1322.39], [831.35, 1321.39], [834.33, 1320.4], [837.32, 1319.4], [840.27, 1318.31], [843.21, 1317.19], [846.14, 1316.04], [849.07, 1314.88], [852.0, 1313.74], [854.95, 1312.64], [857.93, 1311.6], [860.89, 1310.56], [863.85, 1309.47], [866.8, 1308.37], [869.74, 1307.26], [872.68, 1306.14], [875.61, 1304.98], [878.53, 1303.81], [881.45, 1302.62], [884.37, 1301.44], [887.29, 1300.27], [890.2, 1299.08], [893.12, 1297.9], [896.06, 1296.76], [898.97, 1295.57], [901.88, 1294.37], [904.8, 1293.18], [907.71, 1291.98], [910.62, 1290.79], [913.53, 1289.59], [916.46, 1288.45], [919.43, 1287.38], [922.39, 1286.33], [925.37, 1285.3], [928.35, 1284.29], [931.33, 1283.28], [934.3, 1282.24], [937.3, 1281.28], [940.32, 1280.38], [943.35, 1279.55], [946.41, 1278.79], [949.48, 1278.13], [952.58, 1277.57], [955.7, 1277.11], [958.82, 1276.73], [961.95, 1276.38], [965.08, 1276.04], [968.21, 1275.73], [971.35, 1275.48], [974.49, 1275.3], [977.64, 1275.18], [980.78, 1275.06], [983.93, 1274.95], [987.08, 1274.85], [990.22, 1274.8], [993.37, 1274.79], [996.52, 1274.78], [999.67, 1274.78], [1002.81, 1274.8], [1005.96, 1274.84], [1009.11, 1274.9], [1012.26, 1275.0], [1015.4, 1275.06], [1018.55, 1275.14], [1021.7, 1275.22], [1024.84, 1275.27], [1027.99, 1275.35], [1031.14, 1275.42], [1034.29, 1275.49], [1037.43, 1275.55], [1040.58, 1275.54], [1043.73, 1275.55], [1046.88, 1275.55], [1050.02, 1275.59], [1053.17, 1275.62], [1056.32, 1275.68], [1059.47, 1275.76], [1062.61, 1275.86], [1065.76, 1275.99], [1068.9, 1276.15], [1072.05, 1276.25], [1075.19, 1276.26], [1078.34, 1276.15], [1081.48, 1275.99], [1084.62, 1275.77], [1087.76, 1275.48], [1090.88, 1275.07], [1093.99, 1274.55], [1097.08, 1274.0], [1100.17, 1273.38], [1103.24, 1272.69], [1106.3, 1271.93], [1109.33, 1271.08], [1112.35, 1270.21], [1115.36, 1269.27], [1118.36, 1268.32], [1121.37, 1267.41], [1124.39, 1266.51], [1127.39, 1265.58], [1130.38, 1264.58], [1133.36, 1263.57], [1136.35, 1262.57], [1139.32, 1261.53], [1142.28, 1260.46], [1145.22, 1259.34], [1148.15, 1258.2], [1151.07, 1257.01], [1153.97, 1255.79], [1156.86, 1254.54], [1159.75, 1253.29], [1162.63, 1252.03], [1165.51, 1250.76], [1168.39, 1249.48], [1171.26, 1248.19], [1174.14, 1246.91], [1177.01, 1245.62], [1179.87, 1244.3], [1182.72, 1242.97], [1185.57, 1241.63], [1188.42, 1240.3], [1191.28, 1238.98], [1194.14, 1237.66], [1197.0, 1236.36], [1199.87, 1235.07], [1202.74, 1233.76], [1205.6, 1232.46], [1208.49, 1231.2], [1211.37, 1229.93], [1214.25, 1228.66], [1217.13, 1227.4], [1220.04, 1226.19], [1222.96, 1225.02], [1225.89, 1223.87], [1228.83, 1222.73], [1231.77, 1221.62], [1234.73, 1220.54], [1237.7, 1219.51], [1240.7, 1218.56], [1243.72, 1217.64], [1246.75, 1216.79], [1249.79, 1215.98], [1252.84, 1215.22], [1255.91, 1214.5], [1258.99, 1213.86], [1262.08, 1213.27], [1265.17, 1212.68], [1268.27, 1212.09], [1271.37, 1211.55], [1274.47, 1211.04], [1277.58, 1210.55], [1280.68, 1209.99], [1283.77, 1209.36], [1286.84, 1208.67], [1289.9, 1207.94], [1292.97, 1207.24], [1296.04, 1206.56], [1299.12, 1205.9], [1302.21, 1205.29], [1305.3, 1204.73], [1308.4, 1204.17], [1311.5, 1203.59], [1314.59, 1202.99], [1317.67, 1202.37], [1320.75, 1201.7], [1323.81, 1200.99], [1326.87, 1200.23], [1329.94, 1199.52], [1333.0, 1198.79], [1336.06, 1198.05], [1339.11, 1197.28], [1342.16, 1196.5], [1345.21, 1195.73], [1348.27, 1194.98], [1351.33, 1194.24], [1354.39, 1193.51], [1357.48, 1192.91], [1360.59, 1192.42], [1363.72, 1192.04], [1366.85, 1191.79], [1370.0, 1191.65], [1373.15, 1191.63], [1376.29, 1191.73], [1379.43, 1191.95], [1382.56, 1192.29], [1385.68, 1192.75], [1388.77, 1193.32], [1391.84, 1194.01], [1394.89, 1194.82], [1397.9, 1195.74], [1400.87, 1196.77], [1403.8, 1197.91], [1406.69, 1199.17], [1409.53, 1200.53], [1412.31, 1202.0], [1415.04, 1203.57], [1417.71, 1205.24], [1420.31, 1207.01], [1422.84, 1208.88], [1425.3, 1210.85], [1427.69, 1212.9], [1429.99, 1215.04], [1432.22, 1217.27], [1434.36, 1219.58], [1436.41, 1221.97], [1438.37, 1224.43], [1440.24, 1226.96], [1442.04, 1229.55], [1443.84, 1232.12], [1445.71, 1234.66], [1447.63, 1237.16], [1449.59, 1239.62], [1451.51, 1242.11], [1453.44, 1244.6], [1455.33, 1247.12], [1457.15, 1249.68], [1458.9, 1252.3], [1460.57, 1254.97], [1462.16, 1257.69], [1463.69, 1260.44], [1465.16, 1263.22], [1466.56, 1266.04], [1467.95, 1268.87], [1469.3, 1271.71], [1470.61, 1274.57], [1471.89, 1277.45], [1473.13, 1280.34], [1474.38, 1283.23], [1475.64, 1286.11], [1476.88, 1289.01], [1478.1, 1291.91], [1479.29, 1294.82], [1480.45, 1297.75], [1481.63, 1300.67], [1482.79, 1303.6], [1483.9, 1306.54], [1485.01, 1309.49], [1486.11, 1312.43], [1487.22, 1315.38], [1488.43, 1318.29], [1489.72, 1321.16], [1491.05, 1324.01], [1492.38, 1326.86], [1493.69, 1329.73], [1495.02, 1332.58], [1496.33, 1335.44], [1497.65, 1338.3], [1498.99, 1341.15], [1500.42, 1343.95], [1501.89, 1346.74], [1503.4, 1349.5], [1504.81, 1352.32], [1506.1, 1355.18], [1507.29, 1358.1], [1508.39, 1361.05], [1509.37, 1364.04], [1510.3, 1367.05], [1511.13, 1370.08], [1511.89, 1373.14], [1512.62, 1376.2], [1513.28, 1379.28], [1513.93, 1382.36], [1514.54, 1385.45], [1515.07, 1388.55], [1515.53, 1391.66], [1515.97, 1394.78], [1516.37, 1397.9], [1516.76, 1401.03], [1517.15, 1404.15], [1517.46, 1407.28], [1517.66, 1410.42], [1517.73, 1413.57], [1517.69, 1416.72], [1517.53, 1419.86], [1517.25, 1423.0], [1516.89, 1426.12], [1516.46, 1429.24], [1515.94, 1432.35], [1515.31, 1435.43], [1514.56, 1438.49], [1513.72, 1441.52], [1512.77, 1444.53], [1511.74, 1447.5], [1510.65, 1450.45], [1509.51, 1453.39], [1508.31, 1456.3], [1507.05, 1459.18], [1505.78, 1462.06], [1504.51, 1464.94], [1503.22, 1467.81], [1501.9, 1470.67], [1500.58, 1473.53], [1499.22, 1476.37], [1497.82, 1479.19], [1496.43, 1482.01], [1495.01, 1484.82], [1493.6, 1487.63], [1492.16, 1490.44], [1490.73, 1493.24], [1489.26, 1496.02], [1487.84, 1498.83], [1486.49, 1501.68], [1485.23, 1504.56], [1484.08, 1507.49], [1483.02, 1510.46], [1481.96, 1513.42], [1480.85, 1516.36], [1479.73, 1519.3], [1478.57, 1522.23], [1477.41, 1525.16], [1476.24, 1528.08], [1475.18, 1531.05], [1474.23, 1534.05], [1473.4, 1537.08], [1472.68, 1540.15], [1472.08, 1543.24], [1471.6, 1546.35], [1471.23, 1549.47], [1470.98, 1552.61], [1470.85, 1555.76], [1470.84, 1558.91], [1470.95, 1562.05], [1471.18, 1565.19], [1471.52, 1568.32], [1471.98, 1571.43], [1472.56, 1574.53], [1473.26, 1577.6], [1474.07, 1580.64], [1475.0, 1583.65], [1476.03, 1586.62], [1477.18, 1589.55], [1478.44, 1592.44], [1479.81, 1595.27], [1481.28, 1598.05], [1482.86, 1600.78], [1484.54, 1603.44], [1486.32, 1606.04], [1488.19, 1608.57], [1490.16, 1611.02], [1492.22, 1613.4], [1494.37, 1615.7], [1496.6, 1617.92], [1498.92, 1620.05], [1501.31, 1622.1], [1503.78, 1624.05], [1506.33, 1625.9], [1508.94, 1627.66], [1511.61, 1629.32], [1514.35, 1630.88], [1517.14, 1632.33], [1519.98, 1633.68], [1522.88, 1634.92], [1525.82, 1636.05], [1528.8, 1637.07], [1531.81, 1637.97], [1534.86, 1638.76], [1537.93, 1639.43], [1541.03, 1639.99], [1544.15, 1640.43], [1547.28, 1640.76], [1550.42, 1640.99], [1553.56, 1641.14], [1556.71, 1641.26], [1559.86, 1641.28], [1563.0, 1641.27], [1566.15, 1641.18], [1569.29, 1640.97], [1572.43, 1640.67], [1575.55, 1640.32], [1578.68, 1639.97], [1581.81, 1639.64], [1584.94, 1639.31], [1588.08, 1639.01], [1591.21, 1638.75], [1594.35, 1638.46], [1597.48, 1638.15], [1600.61, 1637.85], [1603.75, 1637.58], [1606.89, 1637.33], [1610.03, 1637.07], [1613.17, 1636.86], [1616.31, 1636.7], [1619.46, 1636.57], [1622.6, 1636.47], [1625.75, 1636.36], [1628.89, 1636.22], [1632.04, 1636.1], [1635.19, 1636.0], [1638.33, 1635.91], [1641.48, 1635.82], [1644.63, 1635.75], [1647.77, 1635.69], [1650.92, 1635.59], [1654.06, 1635.46], [1657.21, 1635.32], [1660.35, 1635.13], [1663.49, 1634.89], [1666.63, 1634.68], [1669.77, 1634.44], [1672.9, 1634.16], [1676.03, 1633.77], [1679.14, 1633.3], [1682.24, 1632.77], [1685.34, 1632.18], [1688.41, 1631.51], [1691.48, 1630.83], [1694.55, 1630.1], [1697.6, 1629.32], [1700.65, 1628.55], [1703.73, 1627.89], [1706.83, 1627.34], [1709.94, 1626.85], [1713.05, 1626.36], [1716.16, 1625.88], [1719.27, 1625.39], [1722.38, 1624.9], [1725.49, 1624.41], [1728.6, 1623.92], [1731.7, 1623.43], [1734.81, 1622.95], [1737.92, 1622.46], [1741.03, 1621.97], [1744.14, 1621.48], [1747.25, 1620.99], [1750.36, 1620.51], [1753.47, 1620.02], [1756.58, 1619.53], [1759.69, 1619.04], [1762.8, 1618.55], [1765.91, 1618.06], [1769.02, 1617.58], [1772.13, 1617.09], [1775.24, 1616.6], [1778.35, 1616.11], [1781.46, 1615.62], [1784.57, 1615.13], [1787.68, 1614.65], [1790.79, 1614.16], [1793.9, 1613.67], [1797.01, 1613.18], [1800.12, 1612.69], [1803.23, 1612.2], [1806.34, 1611.72], [1809.45, 1611.23], [1812.56, 1610.74], [1815.67, 1610.25], [1818.78, 1609.76], [1821.89, 1609.27], [1825.0, 1608.79], [1828.11, 1608.3], [1831.22, 1607.81], [1834.33, 1607.32], [1837.44, 1606.83], [1840.55, 1606.35], [1843.66, 1605.86], [1846.77, 1605.37]]], "image": "images/traverse_ordering_00004.png"} diff --git a/data/images/AC-0.png b/data/images/AC-0.png new file mode 100644 index 0000000000000000000000000000000000000000..8920097e153e56541d7ea57d289bd12cae9a4810 --- /dev/null +++ b/data/images/AC-0.png @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:7417166a81381ac73660a2764bda82600a82ab566b3395df2219ad25d2db4e36 +size 2415918 diff --git a/data/images/AC-1.png b/data/images/AC-1.png new file mode 100644 index 0000000000000000000000000000000000000000..895fa15c1d027018d6a6ff98a8915757d70a01c5 --- /dev/null +++ b/data/images/AC-1.png @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:3e58261808a2e435d27e9c008860796489e37647458cc30bd25bcbc610dd7f27 +size 2136140 diff --git a/data/images/AC-2.png b/data/images/AC-2.png new file mode 100644 index 0000000000000000000000000000000000000000..2ccfa2c7b3a72443c66c8b65d5b176c705833cd0 --- /dev/null +++ b/data/images/AC-2.png @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:1b9f82255b54c3c326ef98f0f9b5e274da5854b9dc1d87a1b2ba4d438a2d89ac +size 1189974 diff --git a/data/images/AC-3.png b/data/images/AC-3.png new file mode 100644 index 0000000000000000000000000000000000000000..98657211813c029cf7d34096077151bb4da4a0bd --- /dev/null +++ b/data/images/AC-3.png @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:dcdf8a1800e5dfca85ea527e178298d17030edc4003bfcf2092e8ee576fa8047 +size 2307013 diff --git a/data/images/AC-4.png b/data/images/AC-4.png new file mode 100644 index 0000000000000000000000000000000000000000..c969d6d5086e62e22063b0b6fc0f1f6945eace68 --- /dev/null +++ b/data/images/AC-4.png @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:951a6ebde1307c2eca593506d25aac4c044226508422d80375ed96492c74dd9c +size 2851237 diff --git a/data/images/AGC-0.png b/data/images/AGC-0.png new file mode 100644 index 0000000000000000000000000000000000000000..6d2b141e23105022a8ecce2adeed698bc4ddb5b1 --- /dev/null +++ b/data/images/AGC-0.png @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:d82f8b770b02791aa925d5f6e4c085fb2241574f0d6831e2bf4cd2271cd0b253 +size 2707256 diff --git a/data/images/AGC-1.png b/data/images/AGC-1.png new file mode 100644 index 0000000000000000000000000000000000000000..bdfa02cc1b3170fc959c26377538318c21de2dff --- /dev/null +++ b/data/images/AGC-1.png @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:50ff27f0e307d0352e1eeb5601acbf2ba24bbf9cd169d7caab86a47cd9cc663b +size 2445782 diff --git a/data/images/AGC-2.png b/data/images/AGC-2.png new file mode 100644 index 0000000000000000000000000000000000000000..ec3706afdf8bc42365977fb775eceeb45f630bfa --- /dev/null +++ b/data/images/AGC-2.png @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:7d05fa89c6a44d47b43b9f7c24fe3af4cf22c2b126dc1596a9ee0960981fb101 +size 2626029 diff --git a/data/images/AGC-3.png b/data/images/AGC-3.png new file mode 100644 index 0000000000000000000000000000000000000000..d093a11c576cbb708b6d6fcb0103b0af77e475db --- /dev/null +++ b/data/images/AGC-3.png @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:bd07263069e58898db048120e93e1079057e7e5e4de039f70e1c199df394c6b5 +size 2058324 diff --git a/data/images/AGC-4.png b/data/images/AGC-4.png new file mode 100644 index 0000000000000000000000000000000000000000..1623b45750baf707388a4e6da378a7652ee3dcb8 --- /dev/null +++ b/data/images/AGC-4.png @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:d0611a29bce166345a4409bd34c1158840e4ffc98b9214ad154b9357f5bd4ea1 +size 2468930 diff --git a/data/images/BFC-0.png b/data/images/BFC-0.png new file mode 100644 index 0000000000000000000000000000000000000000..a7defe921e544a6144c7829b2dff32af8c78a10e --- /dev/null +++ b/data/images/BFC-0.png @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:3a249fbde7c1911df0d5f2f1f6fa294188c94a8517dbd20966d1d8bcc2d1efaa +size 2970682 diff --git a/data/images/BFC-1.png b/data/images/BFC-1.png new file mode 100644 index 0000000000000000000000000000000000000000..6dc4594fc566cc3566bc5c2b39c24085020c03d5 --- /dev/null +++ b/data/images/BFC-1.png @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:1a271517410294e0cef4642f2a4244fb06b8725953e0d2a988ed670c1fe3f880 +size 2819013 diff --git a/data/images/BFC-2.png b/data/images/BFC-2.png new file mode 100644 index 0000000000000000000000000000000000000000..a2096e571f2ac1b77c65fd4c246ef01ac6c010d0 --- /dev/null +++ b/data/images/BFC-2.png @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:252f724e27d6d8b6f95764c25efc26b05c3e4d979b870c22f4fe768a9a683b92 +size 2758272 diff --git a/data/images/BFC-3.png b/data/images/BFC-3.png new file mode 100644 index 0000000000000000000000000000000000000000..d6b9f861cf9ca6a1334a7443bedb07c3320732af --- /dev/null +++ b/data/images/BFC-3.png @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:9a1ce169b39704bb71e74d7236505504bf7bc786a6e28f8d2b9430d623fd73eb +size 3047706 diff --git a/data/images/BFC-4.png b/data/images/BFC-4.png new file mode 100644 index 0000000000000000000000000000000000000000..69c93d0e12e7b68ff582aa76b6aa987cb318faed --- /dev/null +++ b/data/images/BFC-4.png @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:a0e9da7e438cb5ca0f21a5d734e95f3d3b199f2dad4df7066eb65d7bfc3ea729 +size 2886636 diff --git a/data/images/CCC-0.png b/data/images/CCC-0.png new file mode 100644 index 0000000000000000000000000000000000000000..0394abda7397eea9243642bd0dafc886bc3b6cb9 --- /dev/null +++ b/data/images/CCC-0.png @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:456cf0bd3b3a832dd6b71e1610e1c324e5ffbc4bd5cdf242e89b3b74989c8cf4 +size 2762884 diff --git a/data/images/CCC-1.png b/data/images/CCC-1.png new file mode 100644 index 0000000000000000000000000000000000000000..f494443d5ab338002b6079eb7a3eb732231b3bff --- /dev/null +++ b/data/images/CCC-1.png @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:bd91dfb264fe66cb942e6da65048f8f745f542a7f648b38728603434669d720c +size 2791126 diff --git a/data/images/CCC-2.png b/data/images/CCC-2.png new file mode 100644 index 0000000000000000000000000000000000000000..ad39c54a05f182931a14b900efb3f2cf3ade0946 --- /dev/null +++ b/data/images/CCC-2.png @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:b42a2ac6131b69280700b9f615abe5108b0ee4601d6e819c8ab05defd3c0ff30 +size 2819239 diff --git a/data/images/CCC-3.png b/data/images/CCC-3.png new file mode 100644 index 0000000000000000000000000000000000000000..87d8f68381c3c7ad2d45b649d95527f62e491f20 --- /dev/null +++ b/data/images/CCC-3.png @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:c236ae836fd47ef1d5795a6d9d110d8fb20900004eb0f29482ad03e4042a9efc +size 3099845 diff --git a/data/images/CCC-4.png b/data/images/CCC-4.png new file mode 100644 index 0000000000000000000000000000000000000000..41a3ac8f0ba86abde0015ea9b8df10c4069fa857 --- /dev/null +++ b/data/images/CCC-4.png @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:a326a583f1a1d5c3a40f71e0f7ae157cb404a16bc42284fcc97f6c2ea02b360e +size 2769262 diff --git a/data/images/CMC-0.png b/data/images/CMC-0.png new file mode 100644 index 0000000000000000000000000000000000000000..62946139726108ece97a28ffc91d13a5b318d08f --- /dev/null +++ b/data/images/CMC-0.png @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:5f929a6fea5039fae7174bd6913c431052e366f284482c92496ef6def6b13877 +size 1383124 diff --git a/data/images/CMC-1.png b/data/images/CMC-1.png new file mode 100644 index 0000000000000000000000000000000000000000..e6fa5d8e67d462b73329c01aa1aac8d34d4288b5 --- /dev/null +++ b/data/images/CMC-1.png @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:18f56e2d6618ebc9a0d310b79d2cce471567ee9b2c4c9ca529aa1b2802691751 +size 1717806 diff --git a/data/images/CMC-2.png b/data/images/CMC-2.png new file mode 100644 index 0000000000000000000000000000000000000000..231411d23522b2cbdf37be378ed2a75f1b24ee6a --- /dev/null +++ b/data/images/CMC-2.png @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:f39221f5dd4c3a532ad331cfb87a1e032e91f31115e8fe871b464f7fd423a54a +size 1211652 diff --git a/data/images/CMC-3.png b/data/images/CMC-3.png new file mode 100644 index 0000000000000000000000000000000000000000..28367a71960836b40c05307f6015cc1f8d18fab3 --- /dev/null +++ b/data/images/CMC-3.png @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:09de618818bd66d57a79d75bf18d2c1f85f731cb3db5f6da915cd8bf5e544a10 +size 1709099 diff --git a/data/images/CMC-4.png b/data/images/CMC-4.png new file mode 100644 index 0000000000000000000000000000000000000000..923fc37d70937cd963e8d03ec0d94abc09b8fbe7 --- /dev/null +++ b/data/images/CMC-4.png @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:7cb4f3e981a45bca948a79e54c7a68aa9141c116d28436f47bda1e594b564485 +size 2064530 diff --git a/data/images/CR-0.png b/data/images/CR-0.png new file mode 100644 index 0000000000000000000000000000000000000000..f519a61a219758acccae5a0a9559207812e480a3 --- /dev/null +++ b/data/images/CR-0.png @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:0ec1b5a70118a429ba78f324a46d7546a233ff36348cba4ac616ef52ce998b8f +size 2951866 diff --git a/data/images/CR-1.png b/data/images/CR-1.png new file mode 100644 index 0000000000000000000000000000000000000000..380f2ad9de785ec5148f3e2ccc751bceaa03b14c --- /dev/null +++ b/data/images/CR-1.png @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:9c33ca1e80167e758322379f390c3941c443d58ba49c28f9f40e7aee97ca0e58 +size 2303024 diff --git a/data/images/CR-2.png b/data/images/CR-2.png new file mode 100644 index 0000000000000000000000000000000000000000..ecd200498eb7646fa27d882fd0a313693a69e0da --- /dev/null +++ b/data/images/CR-2.png @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:1dd4b90d468b151f6cb322321b59de35cb50f7d7c0bb2a91433b488df3a1461d +size 3011937 diff --git a/data/images/CR-3.png b/data/images/CR-3.png new file mode 100644 index 0000000000000000000000000000000000000000..d309fb6f916801a6116c92441eca90691c71d07f --- /dev/null +++ b/data/images/CR-3.png @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:cc2bce4441e9fdd6a9583915dd337926d73a7ca0dc1f14d9ddf16314ef4dae58 +size 2239189 diff --git a/data/images/CR-4.png b/data/images/CR-4.png new file mode 100644 index 0000000000000000000000000000000000000000..ae18918b35107a23eb03a583553fd6ac0a62d636 --- /dev/null +++ b/data/images/CR-4.png @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:40460bb780fc5b2a74edf666b725b8895179e97dfdaf5663a6563be906c4e572 +size 2726754 diff --git a/data/images/CSC-0.png b/data/images/CSC-0.png new file mode 100644 index 0000000000000000000000000000000000000000..a1a9f1c2287e4f54802f03f40a8b378841229a63 --- /dev/null +++ b/data/images/CSC-0.png @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:b650e26df2081f8e62d66dafc33486cf90549a0e10d1660cda1ed83cc900ed05 +size 2428684 diff --git a/data/images/CSC-1.png b/data/images/CSC-1.png new file mode 100644 index 0000000000000000000000000000000000000000..f4d0d7a9bf95626248b6ae41fe0c25da659966d6 --- /dev/null +++ b/data/images/CSC-1.png @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:034e4e90dfd985f8dfc2e105cf3de8137b6e2cf2642566e855688a3cef29d70b +size 2102726 diff --git a/data/images/CSC-2.png b/data/images/CSC-2.png new file mode 100644 index 0000000000000000000000000000000000000000..f6230c7bf552a65c3988370ae3742226765bc3c1 --- /dev/null +++ b/data/images/CSC-2.png @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:e516d6b0daddb0bf1197ec78fe3b082d94f00ad321f8dd3a5ed4750c8ed70f4e +size 2930287 diff --git a/data/images/CSC-3.png b/data/images/CSC-3.png new file mode 100644 index 0000000000000000000000000000000000000000..230221c0bbb82e8c4ccd1daf2de896299b790e19 --- /dev/null +++ b/data/images/CSC-3.png @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:60ea7ea5a0166726afa95812d7c848dcd0d1d7faeab3cc7a5aa265024dd353b4 +size 2667101 diff --git a/data/images/CSC-4.png b/data/images/CSC-4.png new file mode 100644 index 0000000000000000000000000000000000000000..f765532e49eddff04309dd5ce3dc5728f9de7b8a --- /dev/null +++ b/data/images/CSC-4.png @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:f38c3ecaf3ecaec06633806d8cfa1b1b67cb79effc7b89c2743ead0236ace0a1 +size 3030148 diff --git a/data/images/CZ-0.png b/data/images/CZ-0.png new file mode 100644 index 0000000000000000000000000000000000000000..7ed20120532a8bc6cc1c3f1473f2f421f66ee521 --- /dev/null +++ b/data/images/CZ-0.png @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:88b50223c38c896244842a8f41d0282c683e4abe5d4b8280c8167d9ef242de45 +size 2437135 diff --git a/data/images/CZ-1.png b/data/images/CZ-1.png new file mode 100644 index 0000000000000000000000000000000000000000..6cc61f66b62e6075ee71d9e3d13e7aa64d453e30 --- /dev/null +++ b/data/images/CZ-1.png @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:b4cde60a8e87fd1a7328d35e0dc0d9c302fe6a38f74f086f2a2ee4866dd2a4b6 +size 3098367 diff --git a/data/images/CZ-2.png b/data/images/CZ-2.png new file mode 100644 index 0000000000000000000000000000000000000000..f4e095cffda3a633fb27f9becbb21c59505d0af9 --- /dev/null +++ b/data/images/CZ-2.png @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:7251a813552432d3bc01544cc3b2fe25014b2b47d9c79a95a00f0dd6ae139032 +size 2150437 diff --git a/data/images/CZ-3.png b/data/images/CZ-3.png new file mode 100644 index 0000000000000000000000000000000000000000..191defb05f956831e553439a52e155d915786337 --- /dev/null +++ b/data/images/CZ-3.png @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:40f669624624fa3481909c9ed943d1d49aaa062fa85dd3f4cd047d1e792243b6 +size 1823517 diff --git a/data/images/CZ-4.png b/data/images/CZ-4.png new file mode 100644 index 0000000000000000000000000000000000000000..f4df472d86960e4e1ffba54e9af46f2b49083d4e --- /dev/null +++ b/data/images/CZ-4.png @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:8b1d2112f7125b7999981c58da367935bbe42b7a244aaf4b3463d5f3daca7059 +size 2118485 diff --git a/data/images/LI-0.png b/data/images/LI-0.png new file mode 100644 index 0000000000000000000000000000000000000000..4d2a97bed9bed7174261a9210fc3e6f5382ff8dd --- /dev/null +++ b/data/images/LI-0.png @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:7d7c33dd7d09f1b8abfc346a63b2dac969beb465d0f03f1301140cdfd492d133 +size 2539944 diff --git a/data/images/LI-1.png b/data/images/LI-1.png new file mode 100644 index 0000000000000000000000000000000000000000..1102edc5ab2a6a8455ec3f3daa3378b0a7d82c33 --- /dev/null +++ b/data/images/LI-1.png @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:1504a6fd0ec76e2513a23ad71be4e274e197c470b3b3d9d5d282ed8d94439467 +size 2366749 diff --git a/data/images/LI-2.png b/data/images/LI-2.png new file mode 100644 index 0000000000000000000000000000000000000000..43dbb3d49db4e2b29bd66f2b6d7deb77576d1e88 --- /dev/null +++ b/data/images/LI-2.png @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:d95525d70708fd7213fdb5981ae9a5f522cb36fbe3f7340078243fdf9f7ac23a +size 2711327 diff --git a/data/images/LI-3.png b/data/images/LI-3.png new file mode 100644 index 0000000000000000000000000000000000000000..4f869f06c369074fa5cbe5e4d4681722498c67a0 --- /dev/null +++ b/data/images/LI-3.png @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:4328215d10303c8a3dfcf81b6b5fa5738cd46e51568766205b0d3d2e08b96751 +size 2454968 diff --git a/data/images/LI-4.png b/data/images/LI-4.png new file mode 100644 index 0000000000000000000000000000000000000000..5ee23d207f54d8724792079535cba15058189d7b --- /dev/null +++ b/data/images/LI-4.png @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:ffcd8c76dad10674c3584a6fda965598cfb1b4c8dcf95311ccc3cd80be85f4df +size 2611352 diff --git a/data/images/MZ-0.png b/data/images/MZ-0.png new file mode 100644 index 0000000000000000000000000000000000000000..cd6f7f1c5db0373418c9c80dad6b171931698128 --- /dev/null +++ b/data/images/MZ-0.png @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:8c5d171a2367fe52fcb00d0b54f2da6b03bdb6895e5ffb17512d69e274e3d5fb +size 2690751 diff --git a/data/images/MZ-1.png b/data/images/MZ-1.png new file mode 100644 index 0000000000000000000000000000000000000000..04b40758ee4eb293d891434479b310d0397fc901 --- /dev/null +++ b/data/images/MZ-1.png @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:30f0d0e0a9a0f5c63d119cafc794dff22632f61823194c47e1fd97de941e44e3 +size 2422908 diff --git a/data/images/MZ-2.png b/data/images/MZ-2.png new file mode 100644 index 0000000000000000000000000000000000000000..ab956f025ae627a9147b1ab81ccc9057d8fd7744 --- /dev/null +++ b/data/images/MZ-2.png @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:b94e9b199c054e6ff90c251be8ba9b0df17189dd9647bbdd61d5dc692942b12b +size 2720778 diff --git a/data/images/MZ-3.png b/data/images/MZ-3.png new file mode 100644 index 0000000000000000000000000000000000000000..612c5bd4e34b40b8c19ca3e5fcbf93b93cf94fcf --- /dev/null +++ b/data/images/MZ-3.png @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:a165f4c4c84a16660f11aea536c50555f7637114973fc89d5283955a39ac4314 +size 2107101 diff --git a/data/images/MZ-4.png b/data/images/MZ-4.png new file mode 100644 index 0000000000000000000000000000000000000000..957a8eadd126f770e357cde3525da7cc842ebccf --- /dev/null +++ b/data/images/MZ-4.png @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:08177b3d624709e9a53a733fec5efd8f1fcd43a1c3a74b2dc0ba8e4d6dcadb85 +size 2132675 diff --git a/data/images/SCD-0.png b/data/images/SCD-0.png new file mode 100644 index 0000000000000000000000000000000000000000..c7ff63e97175a6a8c3d134382ce673fbf14be941 --- /dev/null +++ b/data/images/SCD-0.png @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:3bf0b8dabd91d3cf952ecafa2818fd7758f2feeb6ffd54299bb471c83eb4005d +size 2826811 diff --git a/data/images/SCD-1.png b/data/images/SCD-1.png new file mode 100644 index 0000000000000000000000000000000000000000..513466192c3ce992ec49ef19c887c573c3b2cccb --- /dev/null +++ b/data/images/SCD-1.png @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:14ab657991686763224af9c32d45ca07cc5cdb746bb037b71621da4bcab79d97 +size 2363237 diff --git a/data/images/SCD-2.png b/data/images/SCD-2.png new file mode 100644 index 0000000000000000000000000000000000000000..a196ed78983a5d6dd8ebfe12e105740e99711603 --- /dev/null +++ b/data/images/SCD-2.png @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:80f6908839fbb8f8787b04e88634bfdaa3713a9f55030c63ee6f472178530096 +size 2919743 diff --git a/data/images/SCD-3.png b/data/images/SCD-3.png new file mode 100644 index 0000000000000000000000000000000000000000..65a9febca1f36e1a568ea16e508a4c418b1d30a9 --- /dev/null +++ b/data/images/SCD-3.png @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:731301ded0afe036b66e41d5696f970151032ebc25a22d594a3af272745d9067 +size 2922337 diff --git a/data/images/SCD-4.png b/data/images/SCD-4.png new file mode 100644 index 0000000000000000000000000000000000000000..17a16e97111e1ce8fb8433b9bf21228489eb5845 --- /dev/null +++ b/data/images/SCD-4.png @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:b75e993758e0967a41def272396d62ab00a3969e5752c82fcb75dab13e399c3e +size 3201772 diff --git a/data/images/SFD-0.png b/data/images/SFD-0.png new file mode 100644 index 0000000000000000000000000000000000000000..c13538aa29d57880cdd4e76b7200937e9cf67cef --- /dev/null +++ b/data/images/SFD-0.png @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:9bd58bde7410c52abb44e401103af4ab81fda4c9cce1f5e26f761ae64b4b5c93 +size 2477023 diff --git a/data/images/SFD-1.png b/data/images/SFD-1.png new file mode 100644 index 0000000000000000000000000000000000000000..b4258d8a00c26bdc3d9bfe12d21825f0f7ce5794 --- /dev/null +++ b/data/images/SFD-1.png @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:1460a81f6979fb4c9cc91422f7b2ef5c28c9fad4b4ceafacf3f2f3c4785523d5 +size 2197455 diff --git a/data/images/SFD-2.png b/data/images/SFD-2.png new file mode 100644 index 0000000000000000000000000000000000000000..ddb7d6ef048ca4b5a03168529691dccd5d1c3b4e --- /dev/null +++ b/data/images/SFD-2.png @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:13b5ac75ddc28310e2796ebdf259765ca00f352df17d85eebb70f3754cea1575 +size 2098145 diff --git a/data/images/SFD-3.png b/data/images/SFD-3.png new file mode 100644 index 0000000000000000000000000000000000000000..d31258866cb3c760f8cd5b06e6d6c17e4c2cba8f --- /dev/null +++ b/data/images/SFD-3.png @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:2b351ff7e5344eae263f953ab5a78d225b6df81596dd18b3eea24021e9d9a519 +size 2304805 diff --git a/data/images/SFD-4.png b/data/images/SFD-4.png new file mode 100644 index 0000000000000000000000000000000000000000..4d06b2417ab0cec118399440b4cd977455b6342e --- /dev/null +++ b/data/images/SFD-4.png @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:2aec3568530c4eb4d83d84015818c3eaeb4f61c13e6ac2ac05029f715a16afbd +size 1984282 diff --git a/data/images/SGC-0.png b/data/images/SGC-0.png new file mode 100644 index 0000000000000000000000000000000000000000..4c8b9dffec7d11ebfe348396df1952d582be67df --- /dev/null +++ b/data/images/SGC-0.png @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:07cc07427e72e28ad81cf374d5f7029cb2c31177dc1bc8593615d8692b5913fe +size 2218473 diff --git a/data/images/SGC-1.png b/data/images/SGC-1.png new file mode 100644 index 0000000000000000000000000000000000000000..1f946fe3f946077078129c1961f4d61d9ae5e28c --- /dev/null +++ b/data/images/SGC-1.png @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:2069c73768f3aa6d010048fdffab71b50af1584ba00bf6226341c9cfba8409d8 +size 2816176 diff --git a/data/images/SGC-2.png b/data/images/SGC-2.png new file mode 100644 index 0000000000000000000000000000000000000000..026328a9fda0900bb7d95d6c12d8491c5e1bdb51 --- /dev/null +++ b/data/images/SGC-2.png @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:d63c140840e20bec90d4b1fbe6d250784a2aba9db3bd523917fc94c9d2d8dd62 +size 1968621 diff --git a/data/images/SGC-3.png b/data/images/SGC-3.png new file mode 100644 index 0000000000000000000000000000000000000000..0995f3f89f3f5d44376775a3b3cb874631bbcdbd --- /dev/null +++ b/data/images/SGC-3.png @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:4d3e00a1593a3fc94182c43d5b2f4e0cfd06b710af44174f0d860b081f1807e2 +size 2565931 diff --git a/data/images/SGC-4.png b/data/images/SGC-4.png new file mode 100644 index 0000000000000000000000000000000000000000..94cefec6dad98fe7e0d42f0a7c805e857a90ec24 --- /dev/null +++ b/data/images/SGC-4.png @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:90fbdc38b1d0dcbbeab0ec4c1750303fb8058f5dcee9052ae3435e0e4855f1bb +size 2715357 diff --git a/data/images/SSD-0.png b/data/images/SSD-0.png new file mode 100644 index 0000000000000000000000000000000000000000..e9705afe85847ca3f632f35d8c40ba7a86208332 --- /dev/null +++ b/data/images/SSD-0.png @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:5101f298b4eff88220e6556800c6aa5cdcbabb39f3d929ae1661aa735108c55b +size 1928847 diff --git a/data/images/SSD-1.png b/data/images/SSD-1.png new file mode 100644 index 0000000000000000000000000000000000000000..15f0082adcda9b427c75b999093fb16c93a1a140 --- /dev/null +++ b/data/images/SSD-1.png @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:cc3743c0ba8a6310a562f221e750d8d918ff305762a5bd1b3a44c149a3789252 +size 1647987 diff --git a/data/images/SSD-2.png b/data/images/SSD-2.png new file mode 100644 index 0000000000000000000000000000000000000000..47369d447a889ef7fdf7c75b3c6d5243590031cf --- /dev/null +++ b/data/images/SSD-2.png @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:6050738551a82fb9659e42b7ebf5877a789794e1b7e127706dd338e0aceab086 +size 1996845 diff --git a/data/images/SSD-3.png b/data/images/SSD-3.png new file mode 100644 index 0000000000000000000000000000000000000000..a4018d3f2885ac651882ddaa8223b37e1cdf22f0 --- /dev/null +++ b/data/images/SSD-3.png @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:4d700c3e6be20af7fde17ab05616ebde070a00f5345385528ec686e7a16177a3 +size 1401827 diff --git a/data/images/SSD-4.png b/data/images/SSD-4.png new file mode 100644 index 0000000000000000000000000000000000000000..5be56b82dedf936bdaba52bb88800700d69a41e6 --- /dev/null +++ b/data/images/SSD-4.png @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:3387cf3ef346cd9cfebb554d0cabfc1980b0b768c1e1cee75a5c55c8fdd7b03c +size 2645494 diff --git a/data/images/STD-0.png b/data/images/STD-0.png new file mode 100644 index 0000000000000000000000000000000000000000..71e4cead6073ffb145191e29b6b81226f6daee80 --- /dev/null +++ b/data/images/STD-0.png @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:df712b3d672531c1a4e96f6593ded0dfe760dc8b9748a7a3c6cee8dd6440270f +size 2678962 diff --git a/data/images/STD-1.png b/data/images/STD-1.png new file mode 100644 index 0000000000000000000000000000000000000000..2ee6f47858157c2d945ebce8a1e983cc1791e8c2 --- /dev/null +++ b/data/images/STD-1.png @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:8af263de4e70c44aafed2a3db9c42bb2808d65d9a11c7c8a1452bbf71acbbb69 +size 3016187 diff --git a/data/images/STD-2.png b/data/images/STD-2.png new file mode 100644 index 0000000000000000000000000000000000000000..3f4256bcf1406648533d9e3315de4fbc668437b8 --- /dev/null +++ b/data/images/STD-2.png @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:dd26b142ae3f324d713c4ebde1630d2cbeddcf0758682bf32b60dc2c6288de0a +size 2176110 diff --git a/data/images/STD-3.png b/data/images/STD-3.png new file mode 100644 index 0000000000000000000000000000000000000000..6d6f786fa16d21770b54e322ee089e7d5a2125db --- /dev/null +++ b/data/images/STD-3.png @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:bade76b8a07ab6805437b9952a3a438f88152fdfe446c2f2f2818c5047fc35e3 +size 2797966 diff --git a/data/images/STD-4.png b/data/images/STD-4.png new file mode 100644 index 0000000000000000000000000000000000000000..fed207160a2d284cb3b3da4a2e2098add81d948d --- /dev/null +++ b/data/images/STD-4.png @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:8a526d7f5ab2f6429db542939c74abb109a8c0b651640291a7e1b0be0dcb91e7 +size 2899999 diff --git a/data/images/TL-0.png b/data/images/TL-0.png new file mode 100644 index 0000000000000000000000000000000000000000..af3f55f11014ea06d122cf6920609ceb12896d5c --- /dev/null +++ b/data/images/TL-0.png @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:39d650e832cc6edd4af73344d31e9c5fbffdd9fabf903b290c7cfa10707ce1c0 +size 2350226 diff --git a/data/images/TL-1.png b/data/images/TL-1.png new file mode 100644 index 0000000000000000000000000000000000000000..0bfd2bdabe16c67866315cfe24a4fdfb23f11aa6 --- /dev/null +++ b/data/images/TL-1.png @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:8a4873a9620ebf807e52df9d86aa31de3810c283f4541e97553a74058d766511 +size 2853924 diff --git a/data/images/TL-2.png b/data/images/TL-2.png new file mode 100644 index 0000000000000000000000000000000000000000..e5b9404630fc7a1814b65351dc77ce3ea5a875ab --- /dev/null +++ b/data/images/TL-2.png @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:7614b2cc36b63e66cf47b0af3780f98c829a9a7e58c54309bfe30a7fa66b7cef +size 2364510 diff --git a/data/images/TL-3.png b/data/images/TL-3.png new file mode 100644 index 0000000000000000000000000000000000000000..85fef68084c69f547c8085e925507c7de0f49c44 --- /dev/null +++ b/data/images/TL-3.png @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:1d3d0b0d031949467381b78eba6b3e7ae38f49d3a98281b484018c2c47a5bb0c +size 2837255 diff --git a/data/images/TL-4.png b/data/images/TL-4.png new file mode 100644 index 0000000000000000000000000000000000000000..8e960fa5467fb781ade5d9a483a27b0940978e58 --- /dev/null +++ b/data/images/TL-4.png @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:1f425f36465b1affbabf42d6b4368f2d5d229bf3ea8f0c231ae2384e0710bd7f +size 2371176 diff --git a/data/images/TO-0.png b/data/images/TO-0.png new file mode 100644 index 0000000000000000000000000000000000000000..cbc6da82b84a53d00a24a393db66cdd441ff75a0 --- /dev/null +++ b/data/images/TO-0.png @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:a55eaaab754a98dcc7c2a72c62aef5dcfd3a713a9d4d82d92e7642d855cc39c3 +size 2621503 diff --git a/data/images/TO-1.png b/data/images/TO-1.png new file mode 100644 index 0000000000000000000000000000000000000000..43cd619d4a5946e4f0873aab07c6779a33e9face --- /dev/null +++ b/data/images/TO-1.png @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:802513182beca8954adb481bf7351958cb7c74909e80a94496bcc11cae8fd313 +size 2568419 diff --git a/data/images/TO-2.png b/data/images/TO-2.png new file mode 100644 index 0000000000000000000000000000000000000000..55ccb8dead56fec3d293cea9fe7ad1e365e7f1be --- /dev/null +++ b/data/images/TO-2.png @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:616488cce7610f749e86bc449415960cdba9a60d8c479a1a928bf87ba9a2cd36 +size 2543311 diff --git a/data/images/TO-3.png b/data/images/TO-3.png new file mode 100644 index 0000000000000000000000000000000000000000..3d84b5ee3a8c44f29ec65c96237e9260701d2fda --- /dev/null +++ b/data/images/TO-3.png @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:4f275ac1edda0bfc2581550d1d4bab670d15a050399e207f8b5776bf69e8b7c4 +size 2485899 diff --git a/data/images/TO-4.png b/data/images/TO-4.png new file mode 100644 index 0000000000000000000000000000000000000000..c9f1cb30ebad0320ed9641bd0691af2fea488617 --- /dev/null +++ b/data/images/TO-4.png @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:3d08aa963f8aa0560967a28a8f6bee9a27ff8a4994654ba215080bd8998f8625 +size 2663938 diff --git a/data/manifest.json b/data/manifest.json new file mode 100644 index 0000000000000000000000000000000000000000..b63a2c7f43ff52eb536122aef9aae06668172772 --- /dev/null +++ b/data/manifest.json @@ -0,0 +1,852 @@ +[ + { + "id": "AGC-0", + "task": "attribute_group_counting", + "category": "distributed_scanning", + "image": "images/AGC-0.png", + "image_sha256": "d82f8b770b02791aa925d5f6e4c085fb2241574f0d6831e2bf4cd2271cd0b253", + "image_source_filename": "attribute_group_counting_00000.gptimage_v1_stones.png", + "question": "How many distinct shapes are in the image? The image contains many small pieces scattered across the canvas. Two pieces belong to the same type if and only if they have exactly the same silhouette. Pieces may differ in surface texture or colour tone — only the silhouette outline matters. Count the total number of distinct silhouette types and report the count as a positive integer. Provide your final answer enclosed in ... tags.", + "answer": 3 + }, + { + "id": "AGC-1", + "task": "attribute_group_counting", + "category": "distributed_scanning", + "image": "images/AGC-1.png", + "image_sha256": "50ff27f0e307d0352e1eeb5601acbf2ba24bbf9cd169d7caab86a47cd9cc663b", + "image_source_filename": "attribute_group_counting_00001.gptimage_v2_cookies.png", + "question": "How many distinct shapes are in the image? The image contains many small pieces scattered across the canvas. Two pieces belong to the same type if and only if they have exactly the same silhouette. Pieces may differ in surface texture or colour tone — only the silhouette outline matters. Count the total number of distinct silhouette types and report the count as a positive integer. Provide your final answer enclosed in ... tags.", + "answer": 5 + }, + { + "id": "AGC-2", + "task": "attribute_group_counting", + "category": "distributed_scanning", + "image": "images/AGC-2.png", + "image_sha256": "7d05fa89c6a44d47b43b9f7c24fe3af4cf22c2b126dc1596a9ee0960981fb101", + "image_source_filename": "attribute_group_counting_00002.gptimage_v3_leaves.png", + "question": "How many distinct shapes are in the image? The image contains many small pieces scattered across the canvas. Two pieces belong to the same type if and only if they have exactly the same silhouette. Pieces may differ in surface texture or colour tone — only the silhouette outline matters. Count the total number of distinct silhouette types and report the count as a positive integer. Provide your final answer enclosed in ... tags.", + "answer": 6 + }, + { + "id": "AGC-3", + "task": "attribute_group_counting", + "category": "distributed_scanning", + "image": "images/AGC-3.png", + "image_sha256": "bd07263069e58898db048120e93e1079057e7e5e4de039f70e1c199df394c6b5", + "image_source_filename": "attribute_group_counting_00003.gptimage_v4_teacups.png", + "question": "How many distinct shapes are in the image? The image contains many small pieces scattered across the canvas. Two pieces belong to the same type if and only if they have exactly the same silhouette. Pieces may differ in surface texture or colour tone — only the silhouette outline matters. Count the total number of distinct silhouette types and report the count as a positive integer. Provide your final answer enclosed in ... tags.", + "answer": 8 + }, + { + "id": "AGC-4", + "task": "attribute_group_counting", + "category": "distributed_scanning", + "image": "images/AGC-4.png", + "image_sha256": "d0611a29bce166345a4409bd34c1158840e4ffc98b9214ad154b9357f5bd4ea1", + "image_source_filename": "attribute_group_counting_00004.gptimage_v5_claytiles.png", + "question": "How many distinct shapes are in the image? The image contains many small pieces scattered across the canvas. Two pieces belong to the same type if and only if they have exactly the same silhouette. Pieces may differ in surface texture or colour tone — only the silhouette outline matters. Count the total number of distinct silhouette types and report the count as a positive integer. Provide your final answer enclosed in ... tags.", + "answer": 10 + }, + { + "id": "BFC-0", + "task": "bounded_faces_counting", + "category": "distributed_scanning", + "image": "images/BFC-0.png", + "image_sha256": "3a249fbde7c1911df0d5f2f1f6fa294188c94a8517dbd20966d1d8bcc2d1efaa", + "image_source_filename": "bounded_faces_counting_00000.gptimage_v1_pool1_driftwood.png", + "question": "How many distinct enclosed regions (bounded faces) are visible in this image? An enclosed region is a maximal area that is fully surrounded by drawn lines or strands on every side, with no opening to the outside background. The unbounded outside area does not count. Each enclosed region should be counted exactly once. Provide your final answer enclosed in ... tags.", + "answer": 5 + }, + { + "id": "BFC-1", + "task": "bounded_faces_counting", + "category": "distributed_scanning", + "image": "images/BFC-1.png", + "image_sha256": "1a271517410294e0cef4642f2a4244fb06b8725953e0d2a988ed670c1fe3f880", + "image_source_filename": "bounded_faces_counting_00001.gptimage_v2_pool2_rebar.png", + "question": "How many distinct enclosed regions (bounded faces) are visible in this image? An enclosed region is a maximal area that is fully surrounded by drawn lines or strands on every side, with no opening to the outside background. The unbounded outside area does not count. Each enclosed region should be counted exactly once. Provide your final answer enclosed in ... tags.", + "answer": 8 + }, + { + "id": "BFC-2", + "task": "bounded_faces_counting", + "category": "distributed_scanning", + "image": "images/BFC-2.png", + "image_sha256": "252f724e27d6d8b6f95764c25efc26b05c3e4d979b870c22f4fe768a9a683b92", + "image_source_filename": "bounded_faces_counting_00002.gptimage_v3_pool3_mossstone.png", + "question": "How many distinct enclosed regions (bounded faces) are visible in this image? An enclosed region is a maximal area that is fully surrounded by drawn lines or strands on every side, with no opening to the outside background. The unbounded outside area does not count. Each enclosed region should be counted exactly once. Provide your final answer enclosed in ... tags.", + "answer": 10 + }, + { + "id": "BFC-3", + "task": "bounded_faces_counting", + "category": "distributed_scanning", + "image": "images/BFC-3.png", + "image_sha256": "9a1ce169b39704bb71e74d7236505504bf7bc786a6e28f8d2b9430d623fd73eb", + "image_source_filename": "bounded_faces_counting_00003.gptimage_v4_pool4_tentstakes.png", + "question": "How many distinct enclosed regions (bounded faces) are visible in this image? An enclosed region is a maximal area that is fully surrounded by drawn lines or strands on every side, with no opening to the outside background. The unbounded outside area does not count. Each enclosed region should be counted exactly once. Provide your final answer enclosed in ... tags.", + "answer": 12 + }, + { + "id": "BFC-4", + "task": "bounded_faces_counting", + "category": "distributed_scanning", + "image": "images/BFC-4.png", + "image_sha256": "a0e9da7e438cb5ca0f21a5d734e95f3d3b199f2dad4df7066eb65d7bfc3ea729", + "image_source_filename": "bounded_faces_counting_00004.gptimage_v5_pool5_telegraph.png", + "question": "How many distinct enclosed regions (bounded faces) are visible in this image? An enclosed region is a maximal area that is fully surrounded by drawn lines or strands on every side, with no opening to the outside background. The unbounded outside area does not count. Each enclosed region should be counted exactly once. Provide your final answer enclosed in ... tags.", + "answer": 15 + }, + { + "id": "CCC-0", + "task": "counting_connected_components", + "category": "distributed_scanning", + "image": "images/CCC-0.png", + "image_sha256": "456cf0bd3b3a832dd6b71e1610e1c324e5ffbc4bd5cdf242e89b3b74989c8cf4", + "image_source_filename": "counting_connected_components_00000.gptimage_v1_driftwood.png", + "question": "How many connected components are there in the image? A connected component is a maximal group of anchor points such that any two are linked by a path of one or more strands (directly or through other anchors in the same group). Anchors not linked by any chain belong to different components, even if they appear visually close. Count every connected component and report the total. Provide your final answer enclosed in ... tags.", + "answer": 10 + }, + { + "id": "CCC-1", + "task": "counting_connected_components", + "category": "distributed_scanning", + "image": "images/CCC-1.png", + "image_sha256": "bd91dfb264fe66cb942e6da65048f8f745f542a7f648b38728603434669d720c", + "image_source_filename": "counting_connected_components_00001.gptimage_v2_rebar.png", + "question": "How many connected components are there in the image? A connected component is a maximal group of anchor points such that any two are linked by a path of one or more strands (directly or through other anchors in the same group). Anchors not linked by any chain belong to different components, even if they appear visually close. Count every connected component and report the total. Provide your final answer enclosed in ... tags.", + "answer": 12 + }, + { + "id": "CCC-2", + "task": "counting_connected_components", + "category": "distributed_scanning", + "image": "images/CCC-2.png", + "image_sha256": "b42a2ac6131b69280700b9f615abe5108b0ee4601d6e819c8ab05defd3c0ff30", + "image_source_filename": "counting_connected_components_00002.gptimage_v3_mossstone.png", + "question": "How many connected components are there in the image? A connected component is a maximal group of anchor points such that any two are linked by a path of one or more strands (directly or through other anchors in the same group). Anchors not linked by any chain belong to different components, even if they appear visually close. Count every connected component and report the total. Provide your final answer enclosed in ... tags.", + "answer": 15 + }, + { + "id": "CCC-3", + "task": "counting_connected_components", + "category": "distributed_scanning", + "image": "images/CCC-3.png", + "image_sha256": "c236ae836fd47ef1d5795a6d9d110d8fb20900004eb0f29482ad03e4042a9efc", + "image_source_filename": "counting_connected_components_00003.gptimage_v4_tentstakes.png", + "question": "How many connected components are there in the image? A connected component is a maximal group of anchor points such that any two are linked by a path of one or more strands (directly or through other anchors in the same group). Anchors not linked by any chain belong to different components, even if they appear visually close. Count every connected component and report the total. Provide your final answer enclosed in ... tags.", + "answer": 18 + }, + { + "id": "CCC-4", + "task": "counting_connected_components", + "category": "distributed_scanning", + "image": "images/CCC-4.png", + "image_sha256": "a326a583f1a1d5c3a40f71e0f7ae157cb404a16bc42284fcc97f6c2ea02b360e", + "image_source_filename": "counting_connected_components_00004.gptimage_v5_telegraph.png", + "question": "How many connected components are there in the image? A connected component is a maximal group of anchor points such that any two are linked by a path of one or more strands (directly or through other anchors in the same group). Anchors not linked by any chain belong to different components, even if they appear visually close. Count every connected component and report the total. Provide your final answer enclosed in ... tags.", + "answer": 20 + }, + { + "id": "CR-0", + "task": "counting_regions", + "category": "distributed_scanning", + "image": "images/CR-0.png", + "image_sha256": "0ec1b5a70118a429ba78f324a46d7546a233ff36348cba4ac616ef52ce998b8f", + "image_source_filename": "counting_regions_00000.gptimage_v8_animaltrails.png", + "question": "How many separated regions are inside the image? A region is a maximal area filled with one continuous fill. Boundaries between regions may appear as continuous lines OR as a chain of small individual markers — discrete dots, footprints, dashes, or short segments aligned along a curve — that you must visually link together to recognize as a single boundary. Two locations belong to the same region if you can travel between them without crossing such a boundary (continuous or chained). Count every distinct region and report the total as a positive integer. Provide your final answer enclosed in ... tags.", + "answer": 10 + }, + { + "id": "CR-1", + "task": "counting_regions", + "category": "distributed_scanning", + "image": "images/CR-1.png", + "image_sha256": "9c33ca1e80167e758322379f390c3941c443d58ba49c28f9f40e7aee97ca0e58", + "image_source_filename": "counting_regions_00001.gptimage_v9_footprintsnow.png", + "question": "How many separated regions are inside the image? A region is a maximal area filled with one continuous fill. Boundaries between regions may appear as continuous lines OR as a chain of small individual markers — discrete dots, footprints, dashes, or short segments aligned along a curve — that you must visually link together to recognize as a single boundary. Two locations belong to the same region if you can travel between them without crossing such a boundary (continuous or chained). Count every distinct region and report the total as a positive integer. Provide your final answer enclosed in ... tags.", + "answer": 12 + }, + { + "id": "CR-2", + "task": "counting_regions", + "category": "distributed_scanning", + "image": "images/CR-2.png", + "image_sha256": "1dd4b90d468b151f6cb322321b59de35cb50f7d7c0bb2a91433b488df3a1461d", + "image_source_filename": "counting_regions_00002.gptimage_v13_antmarches.png", + "question": "How many separated regions are inside the image? A region is a maximal area filled with one continuous fill. Boundaries between regions may appear as continuous lines OR as a chain of small individual markers — discrete dots, footprints, dashes, or short segments aligned along a curve — that you must visually link together to recognize as a single boundary. Two locations belong to the same region if you can travel between them without crossing such a boundary (continuous or chained). Count every distinct region and report the total as a positive integer. Provide your final answer enclosed in ... tags.", + "answer": 15 + }, + { + "id": "CR-3", + "task": "counting_regions", + "category": "distributed_scanning", + "image": "images/CR-3.png", + "image_sha256": "cc2bce4441e9fdd6a9583915dd337926d73a7ca0dc1f14d9ddf16314ef4dae58", + "image_source_filename": "counting_regions_00003.gptimage_v14_parchmentink.png", + "question": "How many separated regions are inside the image? A region is a maximal area filled with one continuous fill. Boundaries between regions may appear as continuous lines OR as a chain of small individual markers — discrete dots, footprints, dashes, or short segments aligned along a curve — that you must visually link together to recognize as a single boundary. Two locations belong to the same region if you can travel between them without crossing such a boundary (continuous or chained). Count every distinct region and report the total as a positive integer. Provide your final answer enclosed in ... tags.", + "answer": 18 + }, + { + "id": "CR-4", + "task": "counting_regions", + "category": "distributed_scanning", + "image": "images/CR-4.png", + "image_sha256": "40460bb780fc5b2a74edf666b725b8895179e97dfdaf5663a6563be906c4e572", + "image_source_filename": "counting_regions_00004.gptimage_v15_beachrope.png", + "question": "How many separated regions are inside the image? A region is a maximal area filled with one continuous fill. Boundaries between regions may appear as continuous lines OR as a chain of small individual markers — discrete dots, footprints, dashes, or short segments aligned along a curve — that you must visually link together to recognize as a single boundary. Two locations belong to the same region if you can travel between them without crossing such a boundary (continuous or chained). Count every distinct region and report the total as a positive integer. Provide your final answer enclosed in ... tags.", + "answer": 20 + }, + { + "id": "TL-0", + "task": "tangled_loops", + "category": "distributed_scanning", + "image": "images/TL-0.png", + "image_sha256": "39d650e832cc6edd4af73344d31e9c5fbffdd9fabf903b290c7cfa10707ce1c0", + "image_source_filename": "tangled_loops_00000.gptimage_v1_pool1_oldrope.png", + "question": "How many distinct closed loops are tangled together in this image? Each loop is a continuous closed curve (rope, cord, cable, etc.) with no loose endpoints anywhere. Loops may cross other loops or themselves freely. Count the total number of distinct closed loops and report the count as a positive integer. Provide your final answer enclosed in ... tags.", + "answer": 10 + }, + { + "id": "TL-1", + "task": "tangled_loops", + "category": "distributed_scanning", + "image": "images/TL-1.png", + "image_sha256": "8a4873a9620ebf807e52df9d86aa31de3810c283f4541e97553a74058d766511", + "image_source_filename": "tangled_loops_00001.gptimage_v2_pool2_cords.png", + "question": "How many distinct closed loops are tangled together in this image? Each loop is a continuous closed curve (rope, cord, cable, etc.) with no loose endpoints anywhere. Loops may cross other loops or themselves freely. Count the total number of distinct closed loops and report the count as a positive integer. Provide your final answer enclosed in ... tags.", + "answer": 12 + }, + { + "id": "TL-2", + "task": "tangled_loops", + "category": "distributed_scanning", + "image": "images/TL-2.png", + "image_sha256": "7614b2cc36b63e66cf47b0af3780f98c829a9a7e58c54309bfe30a7fa66b7cef", + "image_source_filename": "tangled_loops_00002.gptimage_v3_pool3_chaos.png", + "question": "How many distinct closed loops are tangled together in this image? Each loop is a continuous closed curve (rope, cord, cable, etc.) with no loose endpoints anywhere. Loops may cross other loops or themselves freely. Count the total number of distinct closed loops and report the count as a positive integer. Provide your final answer enclosed in ... tags.", + "answer": 15 + }, + { + "id": "TL-3", + "task": "tangled_loops", + "category": "distributed_scanning", + "image": "images/TL-3.png", + "image_sha256": "1d3d0b0d031949467381b78eba6b3e7ae38f49d3a98281b484018c2c47a5bb0c", + "image_source_filename": "tangled_loops_00003.gptimage_v4_pool4_climbing.png", + "question": "How many distinct closed loops are tangled together in this image? Each loop is a continuous closed curve (rope, cord, cable, etc.) with no loose endpoints anywhere. Loops may cross other loops or themselves freely. Count the total number of distinct closed loops and report the count as a positive integer. Provide your final answer enclosed in ... tags.", + "answer": 18 + }, + { + "id": "TL-4", + "task": "tangled_loops", + "category": "distributed_scanning", + "image": "images/TL-4.png", + "image_sha256": "1f425f36465b1affbabf42d6b4368f2d5d229bf3ea8f0c231ae2384e0710bd7f", + "image_source_filename": "tangled_loops_00004.gptimage_v5_pool5_stagecables.png", + "question": "How many distinct closed loops are tangled together in this image? Each loop is a continuous closed curve (rope, cord, cable, etc.) with no loose endpoints anywhere. Loops may cross other loops or themselves freely. Count the total number of distinct closed loops and report the count as a positive integer. Provide your final answer enclosed in ... tags.", + "answer": 20 + }, + { + "id": "AC-0", + "task": "arrow_chain", + "category": "sequential_traversal", + "image": "images/AC-0.png", + "image_sha256": "7417166a81381ac73660a2764bda82600a82ab566b3395df2219ad25d2db4e36", + "image_source_filename": "arrow_chain_00000.gptimage_v1_foot.png", + "question": "The image shows many small pieces scattered across the canvas, each clearly oriented in a specific direction, plus several larger labeled terminus markers (each marker bears a distinct letter). The piece highlighted by a green ring is the starting point. From that piece, cast an infinitely thin ray (a mathematical half-line with zero width) from its centre along the direction it points; the next step is the FIRST other marker this ray enters. A marker only counts if the ray actually crosses its boundary — grazing nearby without entering does not count. Continue until the ray enters a labeled terminus marker and report its letter. Answer with a single letter. Provide your final answer enclosed in ... tags.", + "answer": "H" + }, + { + "id": "AC-1", + "task": "arrow_chain", + "category": "sequential_traversal", + "image": "images/AC-1.png", + "image_sha256": "3e58261808a2e435d27e9c008860796489e37647458cc30bd25bcbc610dd7f27", + "image_source_filename": "arrow_chain_00001.gptimage_v2_fish.png", + "question": "The image shows many small pieces scattered across the canvas, each clearly oriented in a specific direction, plus several larger labeled terminus markers (each marker bears a distinct letter). The piece highlighted by a green ring is the starting point. From that piece, cast an infinitely thin ray (a mathematical half-line with zero width) from its centre along the direction it points; the next step is the FIRST other marker this ray enters. A marker only counts if the ray actually crosses its boundary — grazing nearby without entering does not count. Continue until the ray enters a labeled terminus marker and report its letter. Answer with a single letter. Provide your final answer enclosed in ... tags.", + "answer": "B" + }, + { + "id": "AC-2", + "task": "arrow_chain", + "category": "sequential_traversal", + "image": "images/AC-2.png", + "image_sha256": "1b9f82255b54c3c326ef98f0f9b5e274da5854b9dc1d87a1b2ba4d438a2d89ac", + "image_source_filename": "arrow_chain_00002.gptimage_v3_bird.png", + "question": "The image shows many small pieces scattered across the canvas, each clearly oriented in a specific direction, plus several larger labeled terminus markers (each marker bears a distinct letter). The piece highlighted by a green ring is the starting point. From that piece, cast an infinitely thin ray (a mathematical half-line with zero width) from its centre along the direction it points; the next step is the FIRST other marker this ray enters. A marker only counts if the ray actually crosses its boundary — grazing nearby without entering does not count. Continue until the ray enters a labeled terminus marker and report its letter. Answer with a single letter. Provide your final answer enclosed in ... tags.", + "answer": "B" + }, + { + "id": "AC-3", + "task": "arrow_chain", + "category": "sequential_traversal", + "image": "images/AC-3.png", + "image_sha256": "dcdf8a1800e5dfca85ea527e178298d17030edc4003bfcf2092e8ee576fa8047", + "image_source_filename": "arrow_chain_00003.gptimage_v4_key.png", + "question": "The image shows many small pieces scattered across the canvas, each clearly oriented in a specific direction, plus several larger labeled terminus markers (each marker bears a distinct letter). The piece highlighted by a green ring is the starting point. From that piece, cast an infinitely thin ray (a mathematical half-line with zero width) from its centre along the direction it points; the next step is the FIRST other marker this ray enters. A marker only counts if the ray actually crosses its boundary — grazing nearby without entering does not count. Continue until the ray enters a labeled terminus marker and report its letter. Answer with a single letter. Provide your final answer enclosed in ... tags.", + "answer": "J" + }, + { + "id": "AC-4", + "task": "arrow_chain", + "category": "sequential_traversal", + "image": "images/AC-4.png", + "image_sha256": "951a6ebde1307c2eca593506d25aac4c044226508422d80375ed96492c74dd9c", + "image_source_filename": "arrow_chain_00004.gptimage_v5_leaf.png", + "question": "The image shows many small pieces scattered across the canvas, each clearly oriented in a specific direction, plus several larger labeled terminus markers (each marker bears a distinct letter). The piece highlighted by a green ring is the starting point. From that piece, cast an infinitely thin ray (a mathematical half-line with zero width) from its centre along the direction it points; the next step is the FIRST other marker this ray enters. A marker only counts if the ray actually crosses its boundary — grazing nearby without entering does not count. Continue until the ray enters a labeled terminus marker and report its letter. Answer with a single letter. Provide your final answer enclosed in ... tags.", + "answer": "C" + }, + { + "id": "CZ-0", + "task": "color_zone_sequence", + "category": "sequential_traversal", + "image": "images/CZ-0.png", + "image_sha256": "88b50223c38c896244842a8f41d0282c683e4abe5d4b8280c8167d9ef242de45", + "image_source_filename": "color_zone_00000.gptimage_demo.png", + "question": "The image shows several distinct regions, each carrying a single-letter label and distinguished by its own fill (colour, texture, or material). A single smooth curve starts at the position marked 'Start' and travels through several regions, potentially re-entering earlier ones. Follow the curve from start to end and list the region labels it visits in order, separated by commas. Provide your final answer enclosed in ... tags.", + "answer": "H, D, F, B, I, A, C, J, A, J, H" + }, + { + "id": "CZ-1", + "task": "color_zone_sequence", + "category": "sequential_traversal", + "image": "images/CZ-1.png", + "image_sha256": "b4cde60a8e87fd1a7328d35e0dc0d9c302fe6a38f74f086f2a2ee4866dd2a4b6", + "image_source_filename": "color_zone_00001.gptimage_v2_nautical.png", + "question": "The image shows several distinct regions, each carrying a single-letter label and distinguished by its own fill (colour, texture, or material). A single smooth curve starts at the position marked 'Start' and travels through several regions, potentially re-entering earlier ones. Follow the curve from start to end and list the region labels it visits in order, separated by commas. Provide your final answer enclosed in ... tags.", + "answer": "B, D, C, G, C, E, I, A, J, F, C" + }, + { + "id": "CZ-2", + "task": "color_zone_sequence", + "category": "sequential_traversal", + "image": "images/CZ-2.png", + "image_sha256": "7251a813552432d3bc01544cc3b2fe25014b2b47d9c79a95a00f0dd6ae139032", + "image_source_filename": "color_zone_00002.gptimage_v3_storybook.png", + "question": "The image shows several distinct regions, each carrying a single-letter label and distinguished by its own fill (colour, texture, or material). A single smooth curve starts at the position marked 'Start' and travels through several regions, potentially re-entering earlier ones. Follow the curve from start to end and list the region labels it visits in order, separated by commas. Provide your final answer enclosed in ... tags.", + "answer": "F, E, F, I, C, D, A, J, H, G, E" + }, + { + "id": "CZ-3", + "task": "color_zone_sequence", + "category": "sequential_traversal", + "image": "images/CZ-3.png", + "image_sha256": "40f669624624fa3481909c9ed943d1d49aaa062fa85dd3f4cd047d1e792243b6", + "image_source_filename": "color_zone_00003.gptimage_v4_transit.png", + "question": "The image shows several distinct regions, each carrying a single-letter label and distinguished by its own fill (colour, texture, or material). A single smooth curve starts at the position marked 'Start' and travels through several regions, potentially re-entering earlier ones. Follow the curve from start to end and list the region labels it visits in order, separated by commas. Provide your final answer enclosed in ... tags.", + "answer": "G, B, D, G, A, I, C, I, A, G, D, F" + }, + { + "id": "CZ-4", + "task": "color_zone_sequence", + "category": "sequential_traversal", + "image": "images/CZ-4.png", + "image_sha256": "8b1d2112f7125b7999981c58da367935bbe42b7a244aaf4b3463d5f3daca7059", + "image_source_filename": "color_zone_00004.gptimage_v5_floorplan.png", + "question": "The image shows several distinct regions, each carrying a single-letter label and distinguished by its own fill (colour, texture, or material). A single smooth curve starts at the position marked 'Start' and travels through several regions, potentially re-entering earlier ones. Follow the curve from start to end and list the region labels it visits in order, separated by commas. Provide your final answer enclosed in ... tags.", + "answer": "C, A, G, A, F, E, D, J, B, C, H, I, G" + }, + { + "id": "LI-0", + "task": "line_intersections", + "category": "sequential_traversal", + "image": "images/LI-0.png", + "image_sha256": "7d7c33dd7d09f1b8abfc346a63b2dac969beb465d0f03f1301140cdfd492d133", + "image_source_filename": "line_intersections_00000.gptimage_v1_beach.png", + "question": "The image shows several curves or strands tangled together. Each curve has a single labeled endpoint on the perimeter, marked with a distinct letter. Starting from the endpoint labeled A, follow that curve through the interior tangle. At every crossing you pass through, record the label of the OTHER curve involved — when curve A crosses itself, record A. Report the ordered sequence of labels, separated by commas. Provide your final answer enclosed in ... tags.", + "answer": "C, C, D, C, D, D, C, C, C, B, C" + }, + { + "id": "LI-1", + "task": "line_intersections", + "category": "sequential_traversal", + "image": "images/LI-1.png", + "image_sha256": "1504a6fd0ec76e2513a23ad71be4e274e197c470b3b3d9d5d282ed8d94439467", + "image_source_filename": "line_intersections_00001.gptimage_v2_workshop.png", + "question": "The image shows several curves or strands tangled together. Each curve has a single labeled endpoint on the perimeter, marked with a distinct letter. Starting from the endpoint labeled A, follow that curve through the interior tangle. At every crossing you pass through, record the label of the OTHER curve involved — when curve A crosses itself, record A. Report the ordered sequence of labels, separated by commas. Provide your final answer enclosed in ... tags.", + "answer": "D, B, B, C, C, B, B, B, C" + }, + { + "id": "LI-2", + "task": "line_intersections", + "category": "sequential_traversal", + "image": "images/LI-2.png", + "image_sha256": "d95525d70708fd7213fdb5981ae9a5f522cb36fbe3f7340078243fdf9f7ac23a", + "image_source_filename": "line_intersections_00002.gptimage_v3_forest.png", + "question": "The image shows several curves or strands tangled together. Each curve has a single labeled endpoint on the perimeter, marked with a distinct letter. Starting from the endpoint labeled A, follow that curve through the interior tangle. At every crossing you pass through, record the label of the OTHER curve involved — when curve A crosses itself, record A. Report the ordered sequence of labels, separated by commas. Provide your final answer enclosed in ... tags.", + "answer": "D, C, C, B, B, D, B, D, C" + }, + { + "id": "LI-3", + "task": "line_intersections", + "category": "sequential_traversal", + "image": "images/LI-3.png", + "image_sha256": "4328215d10303c8a3dfcf81b6b5fa5738cd46e51568766205b0d3d2e08b96751", + "image_source_filename": "line_intersections_00003.gptimage_v4_hoses.png", + "question": "The image shows several curves or strands tangled together. Each curve has a single labeled endpoint on the perimeter, marked with a distinct letter. Starting from the endpoint labeled A, follow that curve through the interior tangle. At every crossing you pass through, record the label of the OTHER curve involved — when curve A crosses itself, record A. Report the ordered sequence of labels, separated by commas. Provide your final answer enclosed in ... tags.", + "answer": "D, C, B, C, D, C, D, C, C" + }, + { + "id": "LI-4", + "task": "line_intersections", + "category": "sequential_traversal", + "image": "images/LI-4.png", + "image_sha256": "ffcd8c76dad10674c3584a6fda965598cfb1b4c8dcf95311ccc3cd80be85f4df", + "image_source_filename": "line_intersections_00004.gptimage_v5_ivy.png", + "question": "The image shows several curves or strands tangled together. Each curve has a single labeled endpoint on the perimeter, marked with a distinct letter. Starting from the endpoint labeled A, follow that curve through the interior tangle. At every crossing you pass through, record the label of the OTHER curve involved — when curve A crosses itself, record A. Report the ordered sequence of labels, separated by commas. Provide your final answer enclosed in ... tags.", + "answer": "C, C, B, B, C, B, C, B, C, D, B" + }, + { + "id": "MZ-0", + "task": "maze", + "category": "sequential_traversal", + "image": "images/MZ-0.png", + "image_sha256": "8c5d171a2367fe52fcb00d0b54f2da6b03bdb6895e5ffb17512d69e274e3d5fb", + "image_source_filename": "maze_00000.gptimage_v1_hedge.png", + "question": "The image shows a maze with several labeled openings on its border. Exactly one pair of openings is connected by a path through the maze; all other pairs are blocked by walls. Identify the connected pair. A path exists only if you can travel from one opening to another through corridors without crossing any wall. Report the connected pair in alphabetical order, formatted as X-Y. Provide your final answer enclosed in ... tags.", + "answer": "E-F" + }, + { + "id": "MZ-1", + "task": "maze", + "category": "sequential_traversal", + "image": "images/MZ-1.png", + "image_sha256": "30f0d0e0a9a0f5c63d119cafc794dff22632f61823194c47e1fd97de941e44e3", + "image_source_filename": "maze_00001.gptimage_v2_dungeon.png", + "question": "The image shows a maze with several labeled openings on its border. Exactly one pair of openings is connected by a path through the maze; all other pairs are blocked by walls. Identify the connected pair. A path exists only if you can travel from one opening to another through corridors without crossing any wall. Report the connected pair in alphabetical order, formatted as X-Y. Provide your final answer enclosed in ... tags.", + "answer": "C-E" + }, + { + "id": "MZ-2", + "task": "maze", + "category": "sequential_traversal", + "image": "images/MZ-2.png", + "image_sha256": "b94e9b199c054e6ff90c251be8ba9b0df17189dd9647bbdd61d5dc692942b12b", + "image_source_filename": "maze_00002.gptimage_v3_cornfield.png", + "question": "The image shows a maze with several labeled openings on its border. Exactly one pair of openings is connected by a path through the maze; all other pairs are blocked by walls. Identify the connected pair. A path exists only if you can travel from one opening to another through corridors without crossing any wall. Report the connected pair in alphabetical order, formatted as X-Y. Provide your final answer enclosed in ... tags.", + "answer": "A-G" + }, + { + "id": "MZ-3", + "task": "maze", + "category": "sequential_traversal", + "image": "images/MZ-3.png", + "image_sha256": "a165f4c4c84a16660f11aea536c50555f7637114973fc89d5283955a39ac4314", + "image_source_filename": "maze_00003.gptimage_v4_library.png", + "question": "The image shows a maze with several labeled openings on its border. Exactly one pair of openings is connected by a path through the maze; all other pairs are blocked by walls. Identify the connected pair. A path exists only if you can travel from one opening to another through corridors without crossing any wall. Report the connected pair in alphabetical order, formatted as X-Y. Provide your final answer enclosed in ... tags.", + "answer": "C-G" + }, + { + "id": "MZ-4", + "task": "maze", + "category": "sequential_traversal", + "image": "images/MZ-4.png", + "image_sha256": "08177b3d624709e9a53a733fec5efd8f1fcd43a1c3a74b2dc0ba8e4d6dcadb85", + "image_source_filename": "maze_00004.gptimage_v5_snowmaze.png", + "question": "The image shows a maze with several labeled openings on its border. Exactly one pair of openings is connected by a path through the maze; all other pairs are blocked by walls. Identify the connected pair. A path exists only if you can travel from one opening to another through corridors without crossing any wall. Report the connected pair in alphabetical order, formatted as X-Y. Provide your final answer enclosed in ... tags.", + "answer": "B-C" + }, + { + "id": "TO-0", + "task": "traverse_ordering", + "category": "sequential_traversal", + "image": "images/TO-0.png", + "image_sha256": "a55eaaab754a98dcc7c2a72c62aef5dcfd3a713a9d4d82d92e7642d855cc39c3", + "image_source_filename": "traverse_ordering_00000.gptimage_v1_beach.png", + "question": "The image shows two curves or strands that cross each other. Several labeled points are marked on the two curves combined. A green 'S' marks the start endpoint of one of the curves. Starting from S, follow THAT curve (the one S lies on) all the way to its other endpoint, and list the labels of the marked points you visit in order, separated by commas. Ignore any labeled points that lie on the OTHER curve. Provide your final answer enclosed in ... tags.", + "answer": "L, G, H, F, C, K" + }, + { + "id": "TO-1", + "task": "traverse_ordering", + "category": "sequential_traversal", + "image": "images/TO-1.png", + "image_sha256": "802513182beca8954adb481bf7351958cb7c74909e80a94496bcc11cae8fd313", + "image_source_filename": "traverse_ordering_00001.gptimage_v2_workshop.png", + "question": "The image shows two curves or strands that cross each other. Several labeled points are marked on the two curves combined. A green 'S' marks the start endpoint of one of the curves. Starting from S, follow THAT curve (the one S lies on) all the way to its other endpoint, and list the labels of the marked points you visit in order, separated by commas. Ignore any labeled points that lie on the OTHER curve. Provide your final answer enclosed in ... tags.", + "answer": "J, I, A, L, B, E" + }, + { + "id": "TO-2", + "task": "traverse_ordering", + "category": "sequential_traversal", + "image": "images/TO-2.png", + "image_sha256": "616488cce7610f749e86bc449415960cdba9a60d8c479a1a928bf87ba9a2cd36", + "image_source_filename": "traverse_ordering_00002.gptimage_v3_forest.png", + "question": "The image shows two curves or strands that cross each other. Several labeled points are marked on the two curves combined. A green 'S' marks the start endpoint of one of the curves. Starting from S, follow THAT curve (the one S lies on) all the way to its other endpoint, and list the labels of the marked points you visit in order, separated by commas. Ignore any labeled points that lie on the OTHER curve. Provide your final answer enclosed in ... tags.", + "answer": "I, A, E, F, C, H" + }, + { + "id": "TO-3", + "task": "traverse_ordering", + "category": "sequential_traversal", + "image": "images/TO-3.png", + "image_sha256": "4f275ac1edda0bfc2581550d1d4bab670d15a050399e207f8b5776bf69e8b7c4", + "image_source_filename": "traverse_ordering_00003.gptimage_v4_hoses.png", + "question": "The image shows two curves or strands that cross each other. Several labeled points are marked on the two curves combined. A green 'S' marks the start endpoint of one of the curves. Starting from S, follow THAT curve (the one S lies on) all the way to its other endpoint, and list the labels of the marked points you visit in order, separated by commas. Ignore any labeled points that lie on the OTHER curve. Provide your final answer enclosed in ... tags.", + "answer": "B, H, D, A, F, G" + }, + { + "id": "TO-4", + "task": "traverse_ordering", + "category": "sequential_traversal", + "image": "images/TO-4.png", + "image_sha256": "3d08aa963f8aa0560967a28a8f6bee9a27ff8a4994654ba215080bd8998f8625", + "image_source_filename": "traverse_ordering_00004.gptimage_v5_ivy.png", + "question": "The image shows two curves or strands that cross each other. Several labeled points are marked on the two curves combined. A green 'S' marks the start endpoint of one of the curves. Starting from S, follow THAT curve (the one S lies on) all the way to its other endpoint, and list the labels of the marked points you visit in order, separated by commas. Ignore any labeled points that lie on the OTHER curve. Provide your final answer enclosed in ... tags.", + "answer": "D, F, A, C, B, L" + }, + { + "id": "CMC-0", + "task": "constellation_match_count", + "category": "visual_attribute_transfer", + "image": "images/CMC-0.png", + "image_sha256": "5f929a6fea5039fae7174bd6913c431052e366f284482c92496ef6def6b13877", + "image_source_filename": "constellation_match_count_00000.gptimage_v1_stars.png", + "question": "This image has two panels separated by a thin vertical divider. The left panel shows the Template: a small constellation of marker points. The right panel shows the Field: a larger scene with many marker points. Both panels are drawn at the SAME pixel scale. Count how many copies of the Template pattern appear in the Field. A copy may be rotated by a small angle (up to about 20°) and individual point positions may be jittered slightly, but the overall pattern of relative positions must be preserved. No mirroring or scaling. Report the count as a non-negative integer. Provide your final answer enclosed in ... tags.", + "answer": 5 + }, + { + "id": "CMC-1", + "task": "constellation_match_count", + "category": "visual_attribute_transfer", + "image": "images/CMC-1.png", + "image_sha256": "18f56e2d6618ebc9a0d310b79d2cce471567ee9b2c4c9ca529aa1b2802691751", + "image_source_filename": "constellation_match_count_00001.gptimage_v2_gemstones.png", + "question": "This image has two panels separated by a thin vertical divider. The left panel shows the Template: a small constellation of marker points. The right panel shows the Field: a larger scene with many marker points. Both panels are drawn at the SAME pixel scale. Count how many copies of the Template pattern appear in the Field. A copy may be rotated by a small angle (up to about 20°) and individual point positions may be jittered slightly, but the overall pattern of relative positions must be preserved. No mirroring or scaling. Report the count as a non-negative integer. Provide your final answer enclosed in ... tags.", + "answer": 6 + }, + { + "id": "CMC-2", + "task": "constellation_match_count", + "category": "visual_attribute_transfer", + "image": "images/CMC-2.png", + "image_sha256": "f39221f5dd4c3a532ad331cfb87a1e032e91f31115e8fe871b464f7fd423a54a", + "image_source_filename": "constellation_match_count_00002.gptimage_v3_lights.png", + "question": "This image has two panels separated by a thin vertical divider. The left panel shows the Template: a small constellation of marker points. The right panel shows the Field: a larger scene with many marker points. Both panels are drawn at the SAME pixel scale. Count how many copies of the Template pattern appear in the Field. A copy may be rotated by a small angle (up to about 20°) and individual point positions may be jittered slightly, but the overall pattern of relative positions must be preserved. No mirroring or scaling. Report the count as a non-negative integer. Provide your final answer enclosed in ... tags.", + "answer": 8 + }, + { + "id": "CMC-3", + "task": "constellation_match_count", + "category": "visual_attribute_transfer", + "image": "images/CMC-3.png", + "image_sha256": "09de618818bd66d57a79d75bf18d2c1f85f731cb3db5f6da915cd8bf5e544a10", + "image_source_filename": "constellation_match_count_00003.gptimage_v4_candy.png", + "question": "This image has two panels separated by a thin vertical divider. The left panel shows the Template: a small constellation of marker points. The right panel shows the Field: a larger scene with many marker points. Both panels are drawn at the SAME pixel scale. Count how many copies of the Template pattern appear in the Field. A copy may be rotated by a small angle (up to about 20°) and individual point positions may be jittered slightly, but the overall pattern of relative positions must be preserved. No mirroring or scaling. Report the count as a non-negative integer. Provide your final answer enclosed in ... tags.", + "answer": 9 + }, + { + "id": "CMC-4", + "task": "constellation_match_count", + "category": "visual_attribute_transfer", + "image": "images/CMC-4.png", + "image_sha256": "7cb4f3e981a45bca948a79e54c7a68aa9141c116d28436f47bda1e594b564485", + "image_source_filename": "constellation_match_count_00004.gptimage_v5_shells.png", + "question": "This image has two panels separated by a thin vertical divider. The left panel shows the Template: a small constellation of marker points. The right panel shows the Field: a larger scene with many marker points. Both panels are drawn at the SAME pixel scale. Count how many copies of the Template pattern appear in the Field. A copy may be rotated by a small angle (up to about 20°) and individual point positions may be jittered slightly, but the overall pattern of relative positions must be preserved. No mirroring or scaling. Report the count as a non-negative integer. Provide your final answer enclosed in ... tags.", + "answer": 10 + }, + { + "id": "CSC-0", + "task": "contour_silhouette_count", + "category": "visual_attribute_transfer", + "image": "images/CSC-0.png", + "image_sha256": "b650e26df2081f8e62d66dafc33486cf90549a0e10d1660cda1ed83cc900ed05", + "image_source_filename": "contour_silhouette_count_00000.gptimage_v1_pool1_kitchen.png", + "question": "This image has two panels separated by a thin vertical divider. The left panel shows the Template: a single closed silhouette. The right panel shows the Field: many closed silhouettes scattered across the canvas, all at the SAME pixel scale as the Template. Count the number of Field silhouettes that are exact copies of the Template (translation only — same size, same orientation, no rotation or mirroring) and report the integer count. Provide your final answer enclosed in ... tags.", + "answer": "5" + }, + { + "id": "CSC-1", + "task": "contour_silhouette_count", + "category": "visual_attribute_transfer", + "image": "images/CSC-1.png", + "image_sha256": "034e4e90dfd985f8dfc2e105cf3de8137b6e2cf2642566e855688a3cef29d70b", + "image_source_filename": "contour_silhouette_count_00001.gptimage_v2_pool2_jewellery.png", + "question": "This image has two panels separated by a thin vertical divider. The left panel shows the Template: a single closed silhouette. The right panel shows the Field: many closed silhouettes scattered across the canvas, all at the SAME pixel scale as the Template. Count the number of Field silhouettes that are exact copies of the Template (translation only — same size, same orientation, no rotation or mirroring) and report the integer count. Provide your final answer enclosed in ... tags.", + "answer": "8" + }, + { + "id": "CSC-2", + "task": "contour_silhouette_count", + "category": "visual_attribute_transfer", + "image": "images/CSC-2.png", + "image_sha256": "e516d6b0daddb0bf1197ec78fe3b082d94f00ad321f8dd3a5ed4750c8ed70f4e", + "image_source_filename": "contour_silhouette_count_00002.gptimage_v3_pool3_fruits.png", + "question": "This image has two panels separated by a thin vertical divider. The left panel shows the Template: a single closed silhouette. The right panel shows the Field: many closed silhouettes scattered across the canvas, all at the SAME pixel scale as the Template. Count the number of Field silhouettes that are exact copies of the Template (translation only — same size, same orientation, no rotation or mirroring) and report the integer count. Provide your final answer enclosed in ... tags.", + "answer": "10" + }, + { + "id": "CSC-3", + "task": "contour_silhouette_count", + "category": "visual_attribute_transfer", + "image": "images/CSC-3.png", + "image_sha256": "60ea7ea5a0166726afa95812d7c848dcd0d1d7faeab3cc7a5aa265024dd353b4", + "image_source_filename": "contour_silhouette_count_00003.gptimage_v4_pool4_industrial.png", + "question": "This image has two panels separated by a thin vertical divider. The left panel shows the Template: a single closed silhouette. The right panel shows the Field: many closed silhouettes scattered across the canvas, all at the SAME pixel scale as the Template. Count the number of Field silhouettes that are exact copies of the Template (translation only — same size, same orientation, no rotation or mirroring) and report the integer count. Provide your final answer enclosed in ... tags.", + "answer": "12" + }, + { + "id": "CSC-4", + "task": "contour_silhouette_count", + "category": "visual_attribute_transfer", + "image": "images/CSC-4.png", + "image_sha256": "f38c3ecaf3ecaec06633806d8cfa1b1b67cb79effc7b89c2743ead0236ace0a1", + "image_source_filename": "contour_silhouette_count_00004.gptimage_v5_pool5_textile.png", + "question": "This image has two panels separated by a thin vertical divider. The left panel shows the Template: a single closed silhouette. The right panel shows the Field: many closed silhouettes scattered across the canvas, all at the SAME pixel scale as the Template. Count the number of Field silhouettes that are exact copies of the Template (translation only — same size, same orientation, no rotation or mirroring) and report the integer count. Provide your final answer enclosed in ... tags.", + "answer": "15" + }, + { + "id": "SCD-0", + "task": "spot_the_contour_diff", + "category": "visual_attribute_transfer", + "image": "images/SCD-0.png", + "image_sha256": "3bf0b8dabd91d3cf952ecafa2818fd7758f2feeb6ffd54299bb471c83eb4005d", + "image_source_filename": "contour_diff_00000.gptimage_v1_pool1_kitchen.png", + "question": "Two panels are shown side by side. Each panel contains the same set of closed silhouettes at the same positions. Some silhouettes in the right panel have a different shape from the corresponding silhouette in the left panel. Count the number of pairs whose shape differs between the two panels and report the integer count. Provide your final answer enclosed in ... tags.", + "answer": "5" + }, + { + "id": "SCD-1", + "task": "spot_the_contour_diff", + "category": "visual_attribute_transfer", + "image": "images/SCD-1.png", + "image_sha256": "14ab657991686763224af9c32d45ca07cc5cdb746bb037b71621da4bcab79d97", + "image_source_filename": "contour_diff_00001.gptimage_v2_pool2_jewellery.png", + "question": "Two panels are shown side by side. Each panel contains the same set of closed silhouettes at the same positions. Some silhouettes in the right panel have a different shape from the corresponding silhouette in the left panel. Count the number of pairs whose shape differs between the two panels and report the integer count. Provide your final answer enclosed in ... tags.", + "answer": "8" + }, + { + "id": "SCD-2", + "task": "spot_the_contour_diff", + "category": "visual_attribute_transfer", + "image": "images/SCD-2.png", + "image_sha256": "80f6908839fbb8f8787b04e88634bfdaa3713a9f55030c63ee6f472178530096", + "image_source_filename": "contour_diff_00002.gptimage_v3_pool3_fruits.png", + "question": "Two panels are shown side by side. Each panel contains the same set of closed silhouettes at the same positions. Some silhouettes in the right panel have a different shape from the corresponding silhouette in the left panel. Count the number of pairs whose shape differs between the two panels and report the integer count. Provide your final answer enclosed in ... tags.", + "answer": "10" + }, + { + "id": "SCD-3", + "task": "spot_the_contour_diff", + "category": "visual_attribute_transfer", + "image": "images/SCD-3.png", + "image_sha256": "731301ded0afe036b66e41d5696f970151032ebc25a22d594a3af272745d9067", + "image_source_filename": "contour_diff_00003.gptimage_v4_pool4_industrial.png", + "question": "Two panels are shown side by side. Each panel contains the same set of closed silhouettes at the same positions. Some silhouettes in the right panel have a different shape from the corresponding silhouette in the left panel. Count the number of pairs whose shape differs between the two panels and report the integer count. Provide your final answer enclosed in ... tags.", + "answer": "12" + }, + { + "id": "SCD-4", + "task": "spot_the_contour_diff", + "category": "visual_attribute_transfer", + "image": "images/SCD-4.png", + "image_sha256": "b75e993758e0967a41def272396d62ab00a3969e5752c82fcb75dab13e399c3e", + "image_source_filename": "contour_diff_00004.gptimage_v5_pool5_textile.png", + "question": "Two panels are shown side by side. Each panel contains the same set of closed silhouettes at the same positions. Some silhouettes in the right panel have a different shape from the corresponding silhouette in the left panel. Count the number of pairs whose shape differs between the two panels and report the integer count. Provide your final answer enclosed in ... tags.", + "answer": "15" + }, + { + "id": "SFD-0", + "task": "spot_the_field_diff", + "category": "visual_attribute_transfer", + "image": "images/SFD-0.png", + "image_sha256": "9bd58bde7410c52abb44e401103af4ab81fda4c9cce1f5e26f761ae64b4b5c93", + "image_source_filename": "field_diff_00000.gptimage_v1_corkboard.png", + "question": "Three panels are shown side by side. The leftmost panel is an INDEX key: a grid of cells numbered in row-major order (top-left = 1, increasing left-to-right then top-to-bottom). The middle and right panels are 2D color fields, each divided into the same grid. Compare the middle (Field A) and right (Field B) panels cell by cell. List the numbers of all cells that DIFFER between Field A and Field B, in ascending order, separated by commas. If no cells differ, write \"none\". Provide your final answer enclosed in ... tags.", + "answer": "6, 7, 8, 11" + }, + { + "id": "SFD-1", + "task": "spot_the_field_diff", + "category": "visual_attribute_transfer", + "image": "images/SFD-1.png", + "image_sha256": "1460a81f6979fb4c9cc91422f7b2ef5c28c9fad4b4ceafacf3f2f3c4785523d5", + "image_source_filename": "field_diff_00001.gptimage_v2_woodendesk.png", + "question": "Three panels are shown side by side. The leftmost panel is an INDEX key: a grid of cells numbered in row-major order (top-left = 1, increasing left-to-right then top-to-bottom). The middle and right panels are 2D color fields, each divided into the same grid. Compare the middle (Field A) and right (Field B) panels cell by cell. List the numbers of all cells that DIFFER between Field A and Field B, in ascending order, separated by commas. If no cells differ, write \"none\". Provide your final answer enclosed in ... tags.", + "answer": "6" + }, + { + "id": "SFD-2", + "task": "spot_the_field_diff", + "category": "visual_attribute_transfer", + "image": "images/SFD-2.png", + "image_sha256": "13b5ac75ddc28310e2796ebdf259765ca00f352df17d85eebb70f3754cea1575", + "image_source_filename": "field_diff_00002.gptimage_v3_clipboard.png", + "question": "Three panels are shown side by side. The leftmost panel is an INDEX key: a grid of cells numbered in row-major order (top-left = 1, increasing left-to-right then top-to-bottom). The middle and right panels are 2D color fields, each divided into the same grid. Compare the middle (Field A) and right (Field B) panels cell by cell. List the numbers of all cells that DIFFER between Field A and Field B, in ascending order, separated by commas. If no cells differ, write \"none\". Provide your final answer enclosed in ... tags.", + "answer": "3, 4, 8, 9, 10, 12, 13" + }, + { + "id": "SFD-3", + "task": "spot_the_field_diff", + "category": "visual_attribute_transfer", + "image": "images/SFD-3.png", + "image_sha256": "2b351ff7e5344eae263f953ab5a78d225b6df81596dd18b3eea24021e9d9a519", + "image_source_filename": "field_diff_00003.gptimage_v4_blanket.png", + "question": "Three panels are shown side by side. The leftmost panel is an INDEX key: a grid of cells numbered in row-major order (top-left = 1, increasing left-to-right then top-to-bottom). The middle and right panels are 2D color fields, each divided into the same grid. Compare the middle (Field A) and right (Field B) panels cell by cell. List the numbers of all cells that DIFFER between Field A and Field B, in ascending order, separated by commas. If no cells differ, write \"none\". Provide your final answer enclosed in ... tags.", + "answer": "3, 5, 6, 7, 8, 9, 11, 13, 14, 15" + }, + { + "id": "SFD-4", + "task": "spot_the_field_diff", + "category": "visual_attribute_transfer", + "image": "images/SFD-4.png", + "image_sha256": "2aec3568530c4eb4d83d84015818c3eaeb4f61c13e6ac2ac05029f715a16afbd", + "image_source_filename": "field_diff_00004.gptimage_v5_folder.png", + "question": "Three panels are shown side by side. The leftmost panel is an INDEX key: a grid of cells numbered in row-major order (top-left = 1, increasing left-to-right then top-to-bottom). The middle and right panels are 2D color fields, each divided into the same grid. Compare the middle (Field A) and right (Field B) panels cell by cell. List the numbers of all cells that DIFFER between Field A and Field B, in ascending order, separated by commas. If no cells differ, write \"none\". Provide your final answer enclosed in ... tags.", + "answer": "1, 2, 3, 5, 7, 8, 9, 10, 11, 12, 13, 14" + }, + { + "id": "SSD-0", + "task": "spot_the_signal_diff", + "category": "visual_attribute_transfer", + "image": "images/SSD-0.png", + "image_sha256": "5101f298b4eff88220e6556800c6aa5cdcbabb39f3d929ae1661aa735108c55b", + "image_source_filename": "signal_diff_00000.gptimage_v1_lab.png", + "question": "Two 1D signals are shown side by side. Each signal's x-axis is divided into equal segments by thin vertical gridlines, and each segment is numbered above the gridlines (the two signals share the same numbering). Compare the left (Signal A) and right (Signal B) signals segment by segment. List the numbers of all segments that DIFFER between Signal A and Signal B, in ascending order, separated by commas. If no segments differ, write \"none\". Provide your final answer enclosed in ... tags.", + "answer": "7, 8, 9" + }, + { + "id": "SSD-1", + "task": "spot_the_signal_diff", + "category": "visual_attribute_transfer", + "image": "images/SSD-1.png", + "image_sha256": "cc3743c0ba8a6310a562f221e750d8d918ff305762a5bd1b3a44c149a3789252", + "image_source_filename": "signal_diff_00001.gptimage_v2_whiteboard.png", + "question": "Two 1D signals are shown side by side. Each signal's x-axis is divided into equal segments by thin vertical gridlines, and each segment is numbered above the gridlines (the two signals share the same numbering). Compare the left (Signal A) and right (Signal B) signals segment by segment. List the numbers of all segments that DIFFER between Signal A and Signal B, in ascending order, separated by commas. If no segments differ, write \"none\". Provide your final answer enclosed in ... tags.", + "answer": "2, 5, 7, 8, 10" + }, + { + "id": "SSD-2", + "task": "spot_the_signal_diff", + "category": "visual_attribute_transfer", + "image": "images/SSD-2.png", + "image_sha256": "6050738551a82fb9659e42b7ebf5877a789794e1b7e127706dd338e0aceab086", + "image_source_filename": "signal_diff_00002.gptimage_v3_workbench.png", + "question": "Two 1D signals are shown side by side. Each signal's x-axis is divided into equal segments by thin vertical gridlines, and each segment is numbered above the gridlines (the two signals share the same numbering). Compare the left (Signal A) and right (Signal B) signals segment by segment. List the numbers of all segments that DIFFER between Signal A and Signal B, in ascending order, separated by commas. If no segments differ, write \"none\". Provide your final answer enclosed in ... tags.", + "answer": "8" + }, + { + "id": "SSD-3", + "task": "spot_the_signal_diff", + "category": "visual_attribute_transfer", + "image": "images/SSD-3.png", + "image_sha256": "4d700c3e6be20af7fde17ab05616ebde070a00f5345385528ec686e7a16177a3", + "image_source_filename": "signal_diff_00003.gptimage_v4_monitor.png", + "question": "Two 1D signals are shown side by side. Each signal's x-axis is divided into equal segments by thin vertical gridlines, and each segment is numbered above the gridlines (the two signals share the same numbering). Compare the left (Signal A) and right (Signal B) signals segment by segment. List the numbers of all segments that DIFFER between Signal A and Signal B, in ascending order, separated by commas. If no segments differ, write \"none\". Provide your final answer enclosed in ... tags.", + "answer": "3, 4, 6, 9" + }, + { + "id": "SSD-4", + "task": "spot_the_signal_diff", + "category": "visual_attribute_transfer", + "image": "images/SSD-4.png", + "image_sha256": "3387cf3ef346cd9cfebb554d0cabfc1980b0b768c1e1cee75a5c55c8fdd7b03c", + "image_source_filename": "signal_diff_00004.gptimage_v5_fabric.png", + "question": "Two 1D signals are shown side by side. Each signal's x-axis is divided into equal segments by thin vertical gridlines, and each segment is numbered above the gridlines (the two signals share the same numbering). Compare the left (Signal A) and right (Signal B) signals segment by segment. List the numbers of all segments that DIFFER between Signal A and Signal B, in ascending order, separated by commas. If no segments differ, write \"none\". Provide your final answer enclosed in ... tags.", + "answer": "2, 3, 5, 6, 10" + }, + { + "id": "STD-0", + "task": "spot_the_stroke_diff", + "category": "visual_attribute_transfer", + "image": "images/STD-0.png", + "image_sha256": "df712b3d672531c1a4e96f6593ded0dfe760dc8b9748a7a3c6cee8dd6440270f", + "image_source_filename": "stroke_diff_00000.gptimage_v1_pool1_twigs.png", + "question": "Two panels are shown side by side. Each panel contains the same number of elongated pieces (strands, twigs, threads, etc.) at the same positions. Some pieces in the right panel have a different shape from the corresponding piece in the left panel. Count the number of pairs whose shape differs between the two panels and report the integer count. Provide your final answer enclosed in ... tags.", + "answer": "5" + }, + { + "id": "STD-1", + "task": "spot_the_stroke_diff", + "category": "visual_attribute_transfer", + "image": "images/STD-1.png", + "image_sha256": "8af263de4e70c44aafed2a3db9c42bb2808d65d9a11c7c8a1452bbf71acbbb69", + "image_source_filename": "stroke_diff_00001.gptimage_v2_pool2_sewing.png", + "question": "Two panels are shown side by side. Each panel contains the same number of elongated pieces (strands, twigs, threads, etc.) at the same positions. Some pieces in the right panel have a different shape from the corresponding piece in the left panel. Count the number of pairs whose shape differs between the two panels and report the integer count. Provide your final answer enclosed in ... tags.", + "answer": "8" + }, + { + "id": "STD-2", + "task": "spot_the_stroke_diff", + "category": "visual_attribute_transfer", + "image": "images/STD-2.png", + "image_sha256": "dd26b142ae3f324d713c4ebde1630d2cbeddcf0758682bf32b60dc2c6288de0a", + "image_source_filename": "stroke_diff_00002.gptimage_v3_pool3_noodles.png", + "question": "Two panels are shown side by side. Each panel contains the same number of elongated pieces (strands, twigs, threads, etc.) at the same positions. Some pieces in the right panel have a different shape from the corresponding piece in the left panel. Count the number of pairs whose shape differs between the two panels and report the integer count. Provide your final answer enclosed in ... tags.", + "answer": "10" + }, + { + "id": "STD-3", + "task": "spot_the_stroke_diff", + "category": "visual_attribute_transfer", + "image": "images/STD-3.png", + "image_sha256": "bade76b8a07ab6805437b9952a3a438f88152fdfe446c2f2f2818c5047fc35e3", + "image_source_filename": "stroke_diff_00003.gptimage_v4_pool4_wire.png", + "question": "Two panels are shown side by side. Each panel contains the same number of elongated pieces (strands, twigs, threads, etc.) at the same positions. Some pieces in the right panel have a different shape from the corresponding piece in the left panel. Count the number of pairs whose shape differs between the two panels and report the integer count. Provide your final answer enclosed in ... tags.", + "answer": "12" + }, + { + "id": "STD-4", + "task": "spot_the_stroke_diff", + "category": "visual_attribute_transfer", + "image": "images/STD-4.png", + "image_sha256": "8a526d7f5ab2f6429db542939c74abb109a8c0b651640291a7e1b0be0dcb91e7", + "image_source_filename": "stroke_diff_00004.gptimage_v5_pool5_organic.png", + "question": "Two panels are shown side by side. Each panel contains the same number of elongated pieces (strands, twigs, threads, etc.) at the same positions. Some pieces in the right panel have a different shape from the corresponding piece in the left panel. Count the number of pairs whose shape differs between the two panels and report the integer count. Provide your final answer enclosed in ... tags.", + "answer": "15" + }, + { + "id": "SGC-0", + "task": "stroke_gesture_count", + "category": "visual_attribute_transfer", + "image": "images/SGC-0.png", + "image_sha256": "07cc07427e72e28ad81cf374d5f7029cb2c31177dc1bc8593615d8692b5913fe", + "image_source_filename": "stroke_gesture_count_00000.gptimage_v1_pool1_twigs.png", + "question": "This image has two panels separated by a thin vertical divider. The left panel shows the Template: a single elongated piece with a specific curvature profile. The right panel shows the Field: many elongated pieces of similar thickness scattered across the canvas. Both panels use the SAME pixel scale. Count the number of pieces in the Field that have the exact same shape as the Template (translation only — same curvature, same orientation, no rotation or mirroring) and report the integer count. Provide your final answer enclosed in ... tags.", + "answer": "5" + }, + { + "id": "SGC-1", + "task": "stroke_gesture_count", + "category": "visual_attribute_transfer", + "image": "images/SGC-1.png", + "image_sha256": "2069c73768f3aa6d010048fdffab71b50af1584ba00bf6226341c9cfba8409d8", + "image_source_filename": "stroke_gesture_count_00001.gptimage_v2_pool2_sewing.png", + "question": "This image has two panels separated by a thin vertical divider. The left panel shows the Template: a single white brush stroke with a specific curvature profile. The right panel shows the Field: many white brush strokes scattered across the canvas, all at the same thickness. Both panels use the SAME pixel scale and stroke width. Count the number of strokes in the Field that have the exact same shape as the Template (translation only — same curvature, same orientation, no rotation or mirroring) and report the integer count. Provide your final answer enclosed in ... tags.", + "answer": "8" + }, + { + "id": "SGC-2", + "task": "stroke_gesture_count", + "category": "visual_attribute_transfer", + "image": "images/SGC-2.png", + "image_sha256": "d63c140840e20bec90d4b1fbe6d250784a2aba9db3bd523917fc94c9d2d8dd62", + "image_source_filename": "stroke_gesture_count_00002.gptimage_v3_pool3_noodles.png", + "question": "This image has two panels separated by a thin vertical divider. The left panel shows the Template: a single elongated piece with a specific curvature profile. The right panel shows the Field: many elongated pieces of similar thickness scattered across the canvas. Both panels use the SAME pixel scale. Count the number of pieces in the Field that have the exact same shape as the Template (translation only — same curvature, same orientation, no rotation or mirroring) and report the integer count. Provide your final answer enclosed in ... tags.", + "answer": "10" + }, + { + "id": "SGC-3", + "task": "stroke_gesture_count", + "category": "visual_attribute_transfer", + "image": "images/SGC-3.png", + "image_sha256": "4d3e00a1593a3fc94182c43d5b2f4e0cfd06b710af44174f0d860b081f1807e2", + "image_source_filename": "stroke_gesture_count_00003.gptimage_v4_pool4_wire.png", + "question": "This image has two panels separated by a thin vertical divider. The left panel shows the Template: a single elongated piece with a specific curvature profile. The right panel shows the Field: many elongated pieces of similar thickness scattered across the canvas. Both panels use the SAME pixel scale. Count the number of pieces in the Field that have the exact same shape as the Template (translation only — same curvature, same orientation, no rotation or mirroring) and report the integer count. Provide your final answer enclosed in ... tags.", + "answer": "12" + }, + { + "id": "SGC-4", + "task": "stroke_gesture_count", + "category": "visual_attribute_transfer", + "image": "images/SGC-4.png", + "image_sha256": "90fbdc38b1d0dcbbeab0ec4c1750303fb8058f5dcee9052ae3435e0e4855f1bb", + "image_source_filename": "stroke_gesture_count_00004.gptimage_v5_pool5_organic.png", + "question": "This image has two panels separated by a thin vertical divider. The left panel shows the Template: a single elongated piece with a specific curvature profile. The right panel shows the Field: many elongated pieces of similar thickness scattered across the canvas. Both panels use the SAME pixel scale. Count the number of pieces in the Field that have the exact same shape as the Template (translation only — same curvature, same orientation, no rotation or mirroring) and report the integer count. Provide your final answer enclosed in ... tags.", + "answer": "15" + } +] \ No newline at end of file