Is there any reason why you normalize only input, but not gt?
DongHwanJang opened this issue · comments
I believe that if you use the global residual connection, you need to normalize both of them.
Unless, the residual term has to learn much more complicated space.
Hi!
I have the same question. Do you understand now?
Maybe the convolutional network automatically learns the reversal normalization such that it produces the clean image without normalizing the gt.
Hi, I have the same question. Do you get any idea?
@Wangbk-dl No. I guess there was a mistake implementing the code.
Error while implementing the code.