Abstract: In this paper, we propose a generative face video compression framework based on affine transformations to better represent large movements without parameter transmission. It mainly consists ...
Some results have been hidden because they may be inaccessible to you
Show inaccessible results
Feedback