ali-vilab/UniAnimate
A unified video diffusion model for consistent human image animation and dance generation.
Not currently ranked — collecting fresh signals.
star history
UniAnimate is a research project implementing a video diffusion model for animating human images from a reference photo, producing consistent motion sequences such as dances. The model uses unified diffusion architecture to handle temporal consistency across video frames. The released code includes training and inference implementations for the SCIS-2025 paper.
Frequently asked
- What is ali-vilab/UniAnimate?
- A unified video diffusion model for consistent human image animation and dance generation.
- Is UniAnimate open source?
- Yes — ali-vilab/UniAnimate is an open-source project tracked on heatdrop.
- What language is UniAnimate written in?
- ali-vilab/UniAnimate is primarily written in Python.
- How popular is UniAnimate?
- ali-vilab/UniAnimate has 1.2k stars on GitHub.
- Where can I find UniAnimate?
- ali-vilab/UniAnimate is on GitHub at https://github.com/ali-vilab/UniAnimate.