1 paper · 1 filter
Hao Wu, Xiangyang Luo, Hao Wang +3
With the rapid advancement of diffusion models, talking face generation has made remarkable progress. However, existing diffusion-based methods still require task-specific fine-tun…