CycleGAN
CycleGAN copied to clipboard
Blurry Image of Fake image and reconstructed image
I used the CycleGan to transfer the image of Yosemite in summer and winter, in 17093 iterations, the Fake image and reconstructed image were still blurred, should I wait for more iterations or is there some error in my code (the code was written by myself with little modifying on your original code of CycleGan implementation)
I would wait for more epochs. In our experiment, we trained the model for 200 epochs. (Each epoch has 1, 000 iterations). You also want to double-check if there is bug introduced during the modification.