license: apache-2.0 | |
pipeline_tag: text-to-video | |
<h1 align="center"> | |
<a href="https://arxiv.org/abs/2309.01246"><b>Motion Consistency Model: Accelerating Video Diffusion with Disentangled Motion-Appearance Distillation</b></a> | |
</h1> | |
[[Project page]](https://yhzhai.github.io/mcm/) [[Code]](https://github.com/yhZhai/mcm) [[arXiv]](https://arxiv.org/abs/2406.06890) | |
**TL;DR**: Our motion consistency model not only accelerates text2video diffusion model sampling process, but also can benefit from an additional high-quality image dataset to improve the frame quality of generated videos. | |
![Our motion consistency model not only distill the motion prior from the teacher to accelerate sampling, but also can benefit from an additional high-quality image dataset to improve the frame quality of generated videos.](https://github.com/yhZhai/mcm/blob/main/static/images/illustration.png?raw=true) | |
## Usage |