Faceswap Forum

Has anyone experienced this?

I got a dataset which, when trained long enough, always ends up with a difference in size and color between left & right eye on some of the converted faces
(right eye is slightly larger and sometimes a more blueish tint or sometimes a bit cloudy compared to the left eye which has the correct color/lightness/saturation)

I went through all the pictures and I can't really see where it's getting this data from...

I can honestly say I've never seen this issue. I've had models where eye-tracking has gone wonky, but not your specific example.

for some reason it was just running far behind on training for 1 side of the face in certain angles, like 1 side was perfect and after a long long time it started matching the other side with the same proportions
probably due to a lack of B training data for those angles, and it took a long time to fill in the blanks is my guess

i just gotta be more patient