search

ashawkey / RAD-NeRF

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

Q about another langauge #53

Open suted2 opened 1 year ago

suted2 commented 1 year ago

Hello, I train model from scratch ( audio feature select with Kor wec2vec) with kor .mp4 . but result show just head , not bg or torso (even i give --torso , --bg_img )

Do you have any idea? thx for reading

lsvagusa commented 1 year ago

Hello, I train model from scratch ( audio feature select with Kor wec2vec) with kor .mp4 . but result show just head , not bg or torso (even i give --torso , --bg_img )

Do you have any idea? thx for reading

To include the torso in the output video you need to use the .pth file which is in the trial_videoname_torso/checkpoints folder.