Closed ShungJhon closed 8 months ago
I think sd1.5 unet config.json is ok, BTW,Is your stage 1 training result is ok? please share your insight thx
I think sd1.5 unet config.json is ok, BTW,Is your stage 1 training result is ok? please share your insight thx
I'm using ubc dataset for the first stage, the result is a static video, I'm not sure if the result is correct, the output looks very blur.
you can change a different pose compared with input images, to prove the pose condition is valid,
to animate the stage2 result, the code using a config.json for the pretrained model, how to get it?