Open nftblackmagic opened 3 weeks ago
The problem is still existing. Anyway here is my training implementation.
@nftblackmagic Your results here look good tho without this color shift issue? https://wandb.ai/anzhangusc/train_controlnet/runs/v34c4bin?nw=nwuseranzhangusc
I used a different strength which improved the results. But the problem is still existing
My feeling is that the ip adapter may end up causing the image hue. I saw that you zero initialize the weights for ip adapter, but the paper for ip adapter seems to initialize it with the existing Wk Wv matrices for the regular cross attention. Perhaps that's one of the reasons. I am kind of lazy, so I still end up doing a random init. Not sure how long it takes you to train or the hardware you train on, but I have only recently started training. I'll let it train for a couple days and see what happens.
Are you randomly dropping out the conditions during training? This is needed for CFG to work. Usually dropping out 10% of the time is recommended. Specifically, you need to zero out the reference features similar to what's done for the unconditional input in the tryon pipeline.
Hi there,
I recently replicate the training code. Came across a really interesting problem.
This is my overfitting inferencing test. I got the final results with color difference. It looks weird. The three rows has different guidance_scale = [0.99, 2, 5]
This is the inferencing of official weights.![old_model_0_80daf7e0e4128e3b92a8](https://github.com/yisol/IDM-VTON/assets/108916776/eb641cb7-5afb-45fc-851a-d5e92ab7981a)
The only difference between two pictures is the unet module.
Does anyone have any insight into what might be causing the color difference?