yxt132

Results 21 comments of yxt132

How can we run demo without Adobe training set?

> The training data is not needed and you only have to provide an input RGB image, a trimap (320x320) and a trained model. You can modify the corresponding codes...

ValueError: expected 2D or 3D input (got 4D input)

no solution for this issue?

Error in loading Checkpoint file : Missing keys error

i have the same problem but have not found a solution. @Hangz-nju-cuhk can you update the repo to support higher version of pytorch?

Strange patterns in the output image

same problem here too. my input is 4-channel RGBA and output is also 4-channel RGBA. No label map nor instance map. I used pretty much the same setting as default...

system memory slowly increase

any solution?

system memory slowly increase

Any progress?

error in loading pre-trained model to train on multi-gpu

Nope. I used strict = False to avoid this issue. It seemed it does not affect the results. But I still don't know why those keys are missing.

Do you plan to release the Camera Motion LoRA model and code in the future ?

me too

White-Box model results are not good

thanks for your quick response! I have not started the training yet. I will let you know if I figure out something. By the way, which superpixel method did you...

White-Box model results are not good

Not bad. I noticed some strange color in the generated pictures though. ![image](https://user-images.githubusercontent.com/43161328/99024674-07292a00-25a2-11eb-808c-227dc044975c.png) how is the pytorch version's training? any progress?