Personalized data generation

YuanxunLu / LiveSpeechPortraits

Live Speech Portraits: Real-Time Photorealistic Talking-Head Animation (SIGGRAPH Asia 2021)

MIT License

1.16k stars 198 forks source link

Personalized data generation #76

Open arvind-kr7 opened 1 year ago

arvind-kr7 commented 1 year ago

hey, I cloned the repo. Its working fine with the pretrained model and data generated. But, I wish to train model on new face, and how to generate data for the new face. Is there any source code available?? if yes, whether it is included in this repo? If not can you please provide the same.

If anyone has worked on personalized datasets. please do comment. Your help is appreciated. Thanks

qMartis commented 1 year ago

I'm looking for same thing, do you have any solution ?

samsgates commented 1 year ago

options folder ([https://github.com/YuanxunLu/LiveSpeechPortraits/tree/main/options]) contained training and testing code, but i don't know how to use this code. If anyone has knowledge of these use cases, please let me know.

komilaria commented 11 months ago

please share your train.py