adding face-parsing as a mask option

[phill-98]

Swap-Mukham uses face-parsing and it works very well from my uses of it i think adding in it as an option to use would be great as clip2seg can be 50/50. and maybe theirs a way of combining the 2 masks to get a better output so if theirs object on the target face
face-parsing can do the face and clip2seg can deal more with objects over the face area
https://github.com/zllrunning/face-parsing.PyTorch

[C0untFloyd]

I know about it but I'm not sure how this would help with occluding stuff. If there's e.g. a cup overlapping the mouth, the face parser wouldn't detect the mouth on the target image and the area would so small that clip2seg wouldn't detect parts of the cup. In the long run trained x-seg masks will probably be best but it needs time to implement. You could of course do a PR 😋