search

ashawkey / RAD-NeRF

Real-time Neural Radiance Talking Portrait Synthesis via Audio-spatial Decomposition
MIT License
878 stars 153 forks source link

Performance without audio branch? #17

Closed JuneoXIE closed 1 year ago

JuneoXIE commented 1 year ago

Hello, thank you for the great work!!

I'd like to test on my own data without audio. What would happen if I train the model without audio branch (setting enc_a to None)?

ashawkey commented 1 year ago

@JuneoXIE Hi, if your data is still a monocular talking head video, this method can only learn an "averaged" expression without audio guidance.