Model taking lip video as input

joonson / syncnet_python

Out of time: automated lip sync in the wild

MIT License

682 stars 150 forks source link

Model taking lip video as input #46

Open ZhengRachel opened 2 years ago

ZhengRachel commented 2 years ago

Hi! Thank you for your excellent work in the paper! As is said in you paper, your model takes lip video as input, while this repo however only provides a model taking face video as input. Could you please provide a pre-trained model which takes the lip video as input? It will be very helpful for me!

ZhengRachel commented 2 years ago

Or could you please make it clearer how I can train the model by myself with lip videos as input?