Faceswap Forum

Hi all,

I'm pretty new to FaceSwap.
I'm making the tutorial (Trump/Cage).
I got many questions, you probably will find them stupid but I need some clarifications.

1) On the model-training part, are we forced to "link" A (Trump) to B (Cage) ? I mean, can we train A and B separately so we have 2 models (1 for A and 1 for B) ?
2) If point 1 is possible, can we then train A (Trump), B (Cage) and add a C (Obama) ? So we could choose for the convert part to use the "A-model" on B's src video, and then to use "A-model" on a C src picture.

I hope I'm clear enough. Don't hesitate to ask me if necessary.

Thanks a lot for reading me, hope to read you soon!

Fuzzy wrote: ↑Sat Sep 26, 2020 8:37 pm

1) On the model-training part, are we forced to "link" A (Trump) to B (Cage) ? I mean, can we train A and B separately so we have 2 models (1 for A and 1 for B) ?

No. The specifics of why are covered in the training guide and also answered elsewhere.