Has anyone tried it? How does this compare to ControlNext-SVD?

[FangSen9000]

I am looking for a high-quality pose2video model. If anyone is interested, please feel free to share your thoughts.

[akshayparate123]

Looking at the result of controlnext-svd, I believe unianimate is better.
try mimicmotion https://github.com/Tencent/MimicMotion