Pose-Controllable Talking Face Generation by Implicitly Modularized Audio-Visual Representation

Hang Zhou; Yasheng Sun; Wayne Wu; Chen Change Loy; Xiaogang Wang; Ziwei Liu

doi:10.1109/cvpr46437.2021.00416

CVPR2021顶会

Pose-Controllable Talking Face Generation by Implicitly Modularized Audio-Visual Representation

Hang Zhou, Yasheng Sun, Wayne Wu, Chen Change Loy, Xiaogang Wang, Ziwei Liu

arXiv DOI 出版方

2021年份

124顶会引用

摘要

Our approach takes one frame as identity reference and generates audio-driven talking faces with pose controlled by another pose source video. The mouth shapes of the generated frames are matched with the first row (synced video with audio) while the pose is matched with the bottom row (pose source).