Open johndpope opened 3 weeks ago
try new main code - Jay @hazard-10 spotted an error with cosface in training - and claude fixed it. thats on top of these fixes.
the discriminator i've drafted code to take it to multiscale patch gan. maybe also boost image quality... https://github.com/johndpope/MegaPortrait-hack/issues/46
the leakage - im seeing with my overfitted videos. i think the es is source of problems. when I worked on Emote paper - https://github.com/johndpope/Emote-hack/blob/main/train_stage_1_referencenet.py
UPDATE - from re-reading above - i understand adding more losses - maybe counterproductive. that said - https://arxiv.org/pdf/2404.10667 - i put DPE losses from VASA paper into training code. it doesn't seem to be hurting. https://github.com/johndpope/MegaPortrait-hack/pull/51
Dear CommitCrew -
I bring you a cleaner / faster / smarter way to disentangle images using 3x resnet50 backbones. https://arxiv.org/pdf/2405.07257
https://github.com/johndpope/speak-hack i just start training 5 minutes ago - so far.... not converging.
@Kwentar @flyingshan how are your progresses now?
had incorrectly configured to overfit - updated now https://github.com/johndpope/SPEAK-hack/issues/1
i used another of the videos as driving - and it's (almost) obviously not rotating the head past the point where the original movie went - see below.
cross_reenacted_image_57
pred_frame_191
tomorrow i plug in bigger dataset.
UPDATE - https://github.com/johndpope/MegaPortrait-hack/pull/37
when I normalize the images - i end up with this - looks bad - I add some code in train.py to un-normalize - happy with current results....
fyi - this is the frames dump out from mp4 - head cropped / maybe some warping.