MengXinChengXuYuan

Thank you for your response! I wonder how did you take the noisy level map as an input to the blind denoise net? Simply concat or pixel wise multiply with...

Hi thank you for your great work first! Actually it seems that you are taking the rgb values from sRGB image and putting them in the corresponding Raw RGB location,...

> > Hi thank you for your great work first! > > Actually it seems that you are taking the rgb values from sRGB image and putting them in the...

And, 10140 from sony part is also corrupted by a passing by car for the gt :)

I`m also having the problem, seems random

We just seem to have the same question! :) Which part have you trained? I`ve posted my training record of domain B, check #144

> Maybe you should remove L1 loss: > For L1 loss, model tends to generate colors which is the median number of your training space. > By the way, what...

@raywzy @zhangmozhe Hi is it possible sharing the training logs of the provided weights? And if with some training intermediate generated images, that would be great. I just want to...

I'm also looking forward to the training log In my experiment, using the defaut model setting and my own dataset When training model A, G_featD is very high around 1.5-1.7...

> Hi @MengXinChengXuYuan > What's your --k_size value? I redesign the network, for the efficiency. Most conv kernels are 3*3, transposeConvs are 4 * 4. I don't think this will...