John1231983

Me too. I did not saw any improvement using GAN. As you saw in the paper, the author also mentioned that her method only has some gain in some dataset,...

Me too. I did not see any improvement when ran the GAN+FCN and FCN only.

Hi @rogertrullo: Do you have any update results? I am wondering about the benefit of GAN for segmentation

Hi, I found your explanation about different between caffe and torch implementation for BN. I guess you tried to modify the BatchNorm layer in caffe to make it is similar...

Thanks for point out this. Could you tell me which file did you change for BatchNorm layer? I would like to test it in my caffe version by changing these...

@solomatov : Do you find the solution? I have same issue as you

@KrnTneja : Thanks for your tricks. Could you provide any code to do it? I also meet the problem of loss D goes to zero

Not yet. I did not work on it

So, could you suggest any solutiom to solve the issue?

@junyanz : Thanks. I have removed the new loss and the issue still maintain. I guess that cycleGAN has the issue when I use it on the new dataset. For...