Abhishek Aich

I am late to the party, but here are my inputs: - You **do** require a mask to indicate which parts are missing in the given damaged image. Blind inpainting...

> @abhishekaich27 how do you explain the mask in the loss function? > > it seems intuitive to me that the mask are 1s for missing pixels because we want...

You need a mask to indicate which pixels are missing. This is not blind inpainting (much difficult problem).

Same question. @StevenLiuWen, @zachluo -> any inputs?

Can you provide the golf dataset's pickle file?

options: [https://github.com/krishnakanthnakka/Transferable_Perturbations/blob/main/pix2pix/options/train_options1.py](url) models: [https://github.com/krishnakanthnakka/Transferable_Perturbations/tree/main/pix2pix/models](url)

@gdwang08 @donggong1 @lyn1874 In the testing script for a given video, why don't we compare the score frame-wise? We can always save the reconstruction error and hence the score for...

I observed similar issue for Shanghai-Tech as well. Although I trained the model by myself.

It's now given by the authors. Closing this issue.

@wusize the link you provide for detectron2 based code doesn't show anything. Can you please make the repo public?