Greetings,
I wanted to understand if there was a way to avoid putting the false face on a second face during the conversion.
Basically I have a video with two different faces, one I would like to modify and the other to leave the original.
I hope I was clear.
Thank you all.
This is all covered in the Extract Guide. Specifically, cleaning your alignments file so it only contains your target faces.
Although I have read all the guides this step escaped me, I had understood that that type of cleaning included, for example, wrong alignment parts, I actually hadn't understood that you could exclude faces that you don't want to modify. I'll try to understand how it works, thanks. If I have problems with its functioning, I write here.
To selectively modify one face and leave the other unaltered during conversion, use face detection with unique identifiers and implement conditional processing.