CinCGAN-pytorch
CinCGAN-pytorch copied to clipboard
sangyun884
Pytorch implementation of "Unsupervised Image Super-Resolution using Cycle-in-Cycle Generative Adversarial Networks", CVPRW 2018
about the dataset, why you use the transform operation. ```python train_transform_s = transforms.Compose([ Random90Rot(), transforms.RandomCrop((self.imsize_x,self.imsize_x)), transforms.ToTensor(), transforms.Normalize(mean=(0.5, 0.5, 0.5), std=(0.5, 0.5, 0.5)) ]) test_transform = transforms.Compose([ Crop(self.max_hw), transforms.ToTensor(), transforms.Normalize(mean=(0.5, 0.5,...
in your code 's' means blurred image and 't' means bicubic image? but I think 't' means target which hasn't the correspond hr image for it, so 't' should be...
Metadata
Owner
Metadata
Pytorch implementation of "Unsupervised Image Super-Resolution using Cycle-in-Cycle Generative Adversarial Networks", CVPRW 2018