search

tencent-ailab / V-Express

V-Express aims to generate a talking head video under the control of a reference image, an audio, and a sequence of V-Kps images.
2.03k stars 250 forks source link

为什么用V-Kps frames? #28

Open Xelawk opened 1 month ago

Xelawk commented 1 month ago

有试过face mask region的方案么？ v-kps frames 对比 face mask region的优势在哪里？