Page 1 of 1

using a diferent video for the convert process

Posted: Sat Mar 09, 2024 1:44 am
by raven_soul

I want to swap a face in a movie scene. It lasts 1 min with a shot per face (my model's face and extras' faces). So I was thinking on editing the original scene (where I extracted the faces) to chop the parts where only my actor's face appears and use that video for the convert process, instead of using the original. Is that possible? Or should I convert the whole original video?


Re: using a diferent video for the convert process

Posted: Sat Mar 09, 2024 11:45 am
by torzdf

This should be fine, assuming your data quality is good.

You don't train a faceswap model to swap a scene. You train a model to swap 2 identities.

Once you have a trained model on 2 identities, you can use that model to swap the faces in any scene where the A face appears.