TL;DR: A general face swapping training framework that:
✅ solves no-guidance problems
✅ enhances source identity preservation
✅ is orthogonal and compatiable with existing methods
Image Face Swapping
Results on images collected from the web.
Results on CelebA-HQ.
Results on FF+.
Video Face Swapping
What Problems We Solve
During face swapping training, the re-construction task (used when $X_{\rm{t}}=X_{\rm{s}}$) cannot be used as the proxy anymore when $X_{\rm{t}} \neq X_{\rm{s}}$, lacking pixel-wise supervison $\Gamma$.
Method
Results on CelebA-HQ.
Results on CelebA-HQ.
Reference
Coming soon.
Acknowledgements
Coming soon.