README.md

---
license: cc-by-nc-4.0
task_categories:
  - question-answering
language:
  - en
size_categories:
  - 10M<n<100M
pretty_name: SpatialForge-10M
---

# SpatialForge-10M

**SpatialForge: Bootstrapping 3D-Aware Spatial Reasoning from Open-World 2D Images**

 📑 [Paper](https://arxiv.org/html/2605.11462v1)

**Zishan Liu, Ruoxi Zang, Yanglin Zhang, Wei Liu, Yin Zhang, Jian Yao, Jiayin Zheng, Zhengzhe Liu**

Lingnan University · XPENG Robotics

---

# 📦 SpatialForge-10M

A large-scale vision-language dataset designed for **3D-aware spatial perception and reasoning from open-world 2D images**.

SpatialForge-10M contains over **10 million QA pairs** generated from **2.8 million curated real-world images**, covering both low-level spatial perception and high-level spatial reasoning tasks. The dataset is constructed from diverse open-world image sources including **Objects365**, **Pixmo** and **OpenImages**, enabling broad visual diversity across indoor, outdoor, egocentric, and internet-scale scenes.

SpatialForge-10M is designed for:

- Spatial reasoning pretraining
- Multitask VLM supervision
- 3D-aware perception learning
- Grounding and referring research
- Camera-centric and human-centric reasoning
- Spatial instruction tuning

The dataset supports a unified QA format suitable for training modern multimodal large language models such as Qwen-VL, InternVL, LLaVA, and related architectures.

---

# 📌 Important Notice

This release contains the **full SpatialForge-10M annotations**, including all question-answer pairs and task splits. You will need to obtain the corresponding images directly from the source dataset.

## Key Features

- ✅ **10M+ spatial QA pairs** spanning perception and reasoning tasks
- ✅ **2.8M curated open-world images** from diverse visual domains
- ✅ Covers both **object-centric** and **human-centric** spatial reasoning
- ✅ Includes grounding, referring, counting, depth reasoning, and perspective understanding
- ✅ Designed for scalable VLM pretraining and instruction tuning
- ✅ Bounding boxes are normalized to **[0, 1000]**, following the format used in Qwen3-VL pretraining

---

# 🧠 Task Overview

SpatialForge-10M contains six major spatial tasks divided into two categories: **Perception** and **Relation Reasoning**.

| Level | Task | Description | Count |
|---|---|---|---|
| Perception | Grounding | Localize objects from textual descriptions → bbox prediction | 3.6M |
| Perception | Referring | Generate object descriptions from regions/bboxes | 3.6M |
| Perception | Counting | Count objects satisfying semantic conditions | 495k |
| Relation | Near-Far | Determine relative depth comapring between objects | 2.6M |
| Relation | Left-Right | Infer camera-centric horizontal spatial relations | 93k |
| Relation | Perspective | Human-centric viewpoint and perspective reasoning | 8k |
|  | **Total** |  | **10.2M** |

---

# 🌍 Open-World Data Sources

SpatialForge-10M is bootstrapped from large-scale public image datasets:

- **Objects365**
- **OpenImages**
- **Pixmo**

These datasets provide rich scene diversity across:

- Indoor environments
- Outdoor scenes
- Human-object interactions
- Crowded object layouts
- Real-world internet imagery

Our pipeline automatically constructs spatial supervision signals from 2D images while preserving geometric consistency and viewpoint awareness.

<!-- # 🚀 Sample Usage

This section provides guidance on how to download the SpatialForge-10M annotations and the corresponding images.

## 1. Download Annotations from Hugging Face

First, download the annotation package from Hugging Face Hub:

```bash
# Install huggingface hub if not already
pip install huggingface-hub

# Download the full annotations (QA pairs + task splits)
huggingface-cli download SpatialForge/SpatialForge-10M \
  --repo-type dataset \
  --local-dir ./SpatialForge-10M \
  --resume-download
 -->
---

```bibtex
@article{liu2026spatialforge,
  title={SpatialForge: Bootstrapping 3D-Aware Spatial Reasoning from Open-World 2D Images},
  author={Liu, Zishan and Zang, Ruoxi and Zhang, Yanglin and Liu, Wei and Zhang, Yin and Yao, Jian and Zheng, Jiayin and Liu, Zhengzhe},
  journal={arXiv preprint arXiv:2605.11462},
  year={2026}
}