---
pipeline_tag: other
tags:
- 3d
- human-motion-generation
- diffusion
- arxiv:2605.20955
---

# DrawMotion: Generating 3D Human Motions by Freehand Drawing

<p align="center">
  <img src="figure/demo.gif" alt="DrawMotion web demo" width="70%">
</p>

DrawMotion is a diffusion-based framework for generating 3D human motions from text and freehand drawing conditions, including 2D trajectories and stickman sketches.

- **Paper:** [DrawMotion: Generating 3D Human Motions by Freehand Drawing](https://huggingface.co/papers/2605.20955)
- **Repository:** [InvertedForest/DrawMotion](https://github.com/InvertedForest/DrawMotion)
- **Demo Video:** [YouTube](https://youtu.be/sy2QTdDD09A)

## Model Assets

This Hugging Face repository hosts the public model assets used by DrawMotion. For installation, web demo reproduction, training, and evaluation instructions, please refer to the GitHub repository.

Included assets:

- `logs/human_ml3d/last.ckpt`
- `mid_feat/t2m/mid_feat.pt`
- `stickman/weight/real_init/t2m/stickman_encoder.ckpt`
- `logs/kit_ml/last.ckpt`
- `mid_feat/kit/mid_feat.pt`
- `stickman/weight/kit_ml/split_weight/stickman_encoder.ckpt`

The current public web demo exposes text and trajectory control. Stickman-conditioned interaction is coming soon.

## Citation

If you find DrawMotion useful, please cite:

```bibtex
@article{wang2026drawmotion,
  title={DrawMotion: Generating 3D Human Motions by Freehand Drawing},
  author={Wang, Tao and Jin, Lei and Wu, Zhihua and He, Qiaozhi and Chu, Jiaming and Cheng, Yu and Xing, Junliang and Zhao, Jian and Yan, Shuicheng and Wang, Li},
  journal={IEEE Transactions on Pattern Analysis and Machine Intelligence},
  year={2026},
  pages={1--17},
  doi={10.1109/TPAMI.2026.3679530}
}
```