Francis-Rings/StableAnimator
A video diffusion model that animates a reference image using pose sequences while preserving identity, published at CVPR 2025.
Not currently ranked — collecting fresh signals.
star history
StableAnimator is an end-to-end video diffusion framework that synthesizes high-quality identity-preserving videos from a reference image and pose sequences without post-processing. It addresses challenges in temporal consistency and facial fidelity during animation using a novel approach that avoids relying on face-swapping or restoration tools. The model generates videos directly from pose guidance while maintaining the subject’s identity.
Frequently asked
- What is Francis-Rings/StableAnimator?
- A video diffusion model that animates a reference image using pose sequences while preserving identity, published at CVPR 2025.
- Is StableAnimator open source?
- Yes — Francis-Rings/StableAnimator is open source, released under the MIT license.
- What language is StableAnimator written in?
- Francis-Rings/StableAnimator is primarily written in Python.
- How popular is StableAnimator?
- Francis-Rings/StableAnimator has 1.4k stars on GitHub.
- Where can I find StableAnimator?
- Francis-Rings/StableAnimator is on GitHub at https://github.com/Francis-Rings/StableAnimator.