Shi Guo
The paper use scaled tanh to ensure the output pixel in the range [0,255], and 1/2 stride to upsampling, i didn't find implement details in the code, how can we...
The output should be the mask. Why the channel number is 2?
When running the test code written by myself, I can't obtain the results in your paper. Can you also release your test code? It would help a lot to check...
Hi, this work is very interesting. When I testing the performance on my own captured image, I'm confused about the input shape of MMNet. ``` p = Demosaic(image_patch.float(), M_patch.float()) p.cuda_()...
1. ```PE(pos, 2i) = sin(pos / 10000 ^ (2i/dim))```, Why the parameter is set as 10000? Does this have some meaning for this task. 2. I do not fully understand...
Some thing wrong when I use 3 channel bilinear kernel. Following is the 2 upsampling results and original images.   My init code is followings: function net...