Deep-adversarial-decomposition
Deep-adversarial-decomposition copied to clipboard
jiupinjia
Pytorch implementation of the paper: "A Unified Framework for Separating Superimposed Images", in CVPR 2020.
Hello, does the D1 D2 D3 in the discriminator mean multi-scale discriminator? self.net_D1 = cycnet.define_D(input_nc=6, ndf=64, netD='n_layers', n_layers_D=2).to(device) self.net_D2 = cycnet.define_D(input_nc=6, ndf=64, netD='n_layers', n_layers_D=2).to(device) self.net_D3 = cycnet.define_D(input_nc=6, ndf=64, netD='n_layers', n_layers_D=3).to(device)...
Hello, I want to train Rain100H so how should I place the data, I want to know how to place the data I make errors in every round of evaluation...
Hi, I computed the PSNR using your pre-trained checkpoint file on dogsflowers dataset. I'm able to get the same PSNR on the validation set. However, I find that PSNR on...
Dear professor: I follow your program to do colour unmixing. Now I unmix an image with two colours to two image with a single colour. But can I decompose an...
Would you please upload the datasets for training and testing on superimposed image separation? It will help me a lot for both convenient and fair experimental comparison.
As mentioned in your paper, gt2 uses ‘zero image’ in these datasets. Does this problem degenerate to standard pixel-wise l1 loss and discriminator loss? So why is your method better...
