Motion Forcing: A Decoupled Framework for Robust Video Generation in Motion Dynamics

arXiv Project Page GitHub

Usage

git clone https://github.com/Tianshuo-Xu/Motion-Forcing
cd Motion-Forcing
pip install -r requirements.txt
python gradio_demo.py

The model weights are downloaded automatically from this repo via diffusers. See the GitHub repository for full setup instructions and CLI options.

Demo & Results

Visit the project page for video comparisons across driving scenes, physics simulations, and embodied AI tasks.

Citation

@misc{xu2026motion,
      title={Motion Forcing: A Decoupled Framework for Robust Video Generation in Motion Dynamics}, 
      author={Tianshuo Xu and Zhifei Chen and Leyi Wu and Hao Lu and Ying-cong Chen},
      year={2026},
      eprint={2603.10408},
      archivePrefix={arXiv},
      primaryClass={cs.CV},
      url={https://arxiv.org/abs/2603.10408}, 
}
Downloads last month

-

Downloads are not tracked for this model. How to track
Safetensors
Model size
6B params
Tensor type
BF16
·
Inference Providers NEW
This model isn't deployed by any Inference Provider. 🙋 Ask for provider support

Model tree for TSXu/MotionForcing_driving

Finetuned
(4)
this model

Paper for TSXu/MotionForcing_driving