1 paper
Shuyuan Tu, Zhen Xing, Xintong Han +4
Current diffusion models for human image animation struggle to ensure identity (ID) consistency. This paper presents StableAnimator, the first end-to-end ID-preserving video diffus…