Upload README.md with huggingface_hub
Browse files
README.md
ADDED
|
@@ -0,0 +1,89 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
---
|
| 2 |
+
license: mit
|
| 3 |
+
task_categories:
|
| 4 |
+
- robotics
|
| 5 |
+
- object-detection
|
| 6 |
+
language:
|
| 7 |
+
- en
|
| 8 |
+
tags:
|
| 9 |
+
- robotics
|
| 10 |
+
- grasping
|
| 11 |
+
- affordance
|
| 12 |
+
- hand-object-interaction
|
| 13 |
+
- contact-detection
|
| 14 |
+
- YCB
|
| 15 |
+
- DexYCB
|
| 16 |
+
- HO3D
|
| 17 |
+
- FoundationPose
|
| 18 |
+
- MANO
|
| 19 |
+
pretty_name: Affordance2Grasp Preprocessed Assets
|
| 20 |
+
size_categories:
|
| 21 |
+
- 1G<n<10G
|
| 22 |
+
---
|
| 23 |
+
|
| 24 |
+
# Affordance2Grasp — Preprocessed Assets
|
| 25 |
+
|
| 26 |
+
Preprocessed object meshes and initialization files for the
|
| 27 |
+
[Affordance2Grasp](https://github.com/stzabl-png/UCB_Project) pipeline.
|
| 28 |
+
|
| 29 |
+
This repository contains assets you download **once** before running the pipeline.
|
| 30 |
+
The pipeline then generates per-dataset intermediate data (depth maps, MANO estimates,
|
| 31 |
+
poses, contact labels) from the raw videos, which are not included here due to size and
|
| 32 |
+
licensing constraints.
|
| 33 |
+
|
| 34 |
+
---
|
| 35 |
+
|
| 36 |
+
## Contents
|
| 37 |
+
|
| 38 |
+
| Folder | Size | Description |
|
| 39 |
+
|---|---|---|
|
| 40 |
+
| `obj_meshes/ycb/` | ~1 GB | YCB object meshes + `scale.json` (metric scale) for all objects used in DexYCB and HO3D v3 |
|
| 41 |
+
| `obj_recon_input/ycb/` | ~50 MB | FoundationPose initialization masks (reference images + bounding boxes) for each YCB object |
|
| 42 |
+
| `training_fp/dexycb/` | ~1 MB | Validated contact labels for DexYCB `ycb_dex_01` and `ycb_dex_02` (sanity-check reference) |
|
| 43 |
+
|
| 44 |
+
---
|
| 45 |
+
|
| 46 |
+
## Usage
|
| 47 |
+
|
| 48 |
+
```bash
|
| 49 |
+
pip install huggingface_hub
|
| 50 |
+
|
| 51 |
+
python -c "
|
| 52 |
+
from huggingface_hub import snapshot_download
|
| 53 |
+
snapshot_download(
|
| 54 |
+
repo_id='StZaBL/Affordance2Grasp-ProcessedData',
|
| 55 |
+
repo_type='dataset',
|
| 56 |
+
local_dir='data_hub/ProcessedData',
|
| 57 |
+
)
|
| 58 |
+
"
|
| 59 |
+
```
|
| 60 |
+
|
| 61 |
+
Then follow the pipeline in the
|
| 62 |
+
[main repository README](https://github.com/stzabl-png/UCB_Project).
|
| 63 |
+
|
| 64 |
+
---
|
| 65 |
+
|
| 66 |
+
## Supported Datasets
|
| 67 |
+
|
| 68 |
+
Both datasets use the same YCB objects — the meshes here cover all required objects.
|
| 69 |
+
|
| 70 |
+
| Dataset | Objects covered |
|
| 71 |
+
|---|---|
|
| 72 |
+
| **DexYCB** | 20 YCB objects (`ycb_dex_01` – `ycb_dex_20`) |
|
| 73 |
+
| **HO3D v3** | `003_cracker_box`, `004_sugar_box`, `006_mustard_bottle`, `010_potted_meat_can`, `011_banana`, `021_bleach_cleanser`, `035_power_drill`, `052_extra_large_clamp` |
|
| 74 |
+
|
| 75 |
+
---
|
| 76 |
+
|
| 77 |
+
## Raw Data
|
| 78 |
+
|
| 79 |
+
Raw videos must be downloaded separately from the official sources:
|
| 80 |
+
|
| 81 |
+
- **DexYCB**: https://dex-ycb.github.io
|
| 82 |
+
- **HO3D v3**: https://www.tugraz.at/institute/icg/research/team-lepetit/research-projects/hand-object-3d-pose-annotation
|
| 83 |
+
|
| 84 |
+
---
|
| 85 |
+
|
| 86 |
+
## Related
|
| 87 |
+
|
| 88 |
+
- **Code**: [stzabl-png/UCB_Project](https://github.com/stzabl-png/UCB_Project)
|
| 89 |
+
- **Pipeline**: Depth Pro → HaPTIC → FoundationPose → Contact Labels → Affordance Model → Robot Grasp
|