USTC3DV / FlashAvatar-code

[CVPR 2024] The official repo for FlashAvatar
MIT License
143 stars 18 forks source link

About paper content #17

Closed lzzbutphh closed 4 months ago

lzzbutphh commented 4 months ago

Hello, I have read your paper, and I understand that you control the actions based on the expression coefficients of FLAME. However, from my understanding, these coefficients do not include descriptions of jaw/lip movements. Could you please explain what controls these particular actions?

xiangjun-xj commented 4 months ago

The expression code we use includes expression coefficients and jaw pose. But of course it is still hard to control extremely slight motions, which are not well tracked.