Read the FAQs and search the forum before posting a new topic.
This forum is for discussing tips and understanding the process involved with Training a Faceswap model.
If you have found a bug are having issues with the Training process not working, then you should post in the Training Support forum.
Please mark any answers that fixed your problems so others can find the solutions.
Hello,
Just a stupid question:
Being a fan of classic movies... most of which in Black and White... would it be reasonable to create a simpler neural network (trainer) working on a single luminance channel, instead of full RGB ?
Would it be ...let's say feasible in this way to trade the infos usually needed for the other two colors in a higher resolution/greater speed ?
Just a question... I am still a newbye, but ...curious.
Yes it's possible in theory, in reality black and white has less information to go off of and tends to require more model space to get a good result, so it's usually just as good to use RGB.
