Any plan to integrate the face vid2vid model trained by Live Portrait?

OpenTalker / SadTalker

[CVPR 2023] SadTalker：Learning Realistic 3D Motion Coefficients for Stylized Audio-Driven Single Image Talking Face Animation

https://sadtalker.github.io/

Other

11.93k stars 2.22k forks source link

Any plan to integrate the face vid2vid model trained by Live Portrait? #941

Open hoveychen opened 3 months ago

hoveychen commented 3 months ago

https://github.com/KwaiVGI/LivePortrait has trained a better face vid2vid model from ground up. However, the official repo provides only video-driven generation. That's what SadTalker comes into play. Is there any plan to integrate the advantage from both SadTalker and Live Portrait?

oisilener1982 commented 3 months ago

THe main Dev of Sadtalker is busy with another project called MOFA. Most likely this project is dead.

hoveychen commented 3 months ago

what a pity. Is there any training code that I can retrain the model with better vid2vid model?