The final srgan in its paper uses the feature map obtained by the 4-th convolution (after activation) before the 5-th maxpooling layer within the VGG19 network.
Also no tanh at the end of conv3 for the generator, just a convolution
The final srgan in its paper uses the feature map obtained by the 4-th convolution (after activation) before the 5-th maxpooling layer within the VGG19 network.
Also no tanh at the end of conv3 for the generator, just a convolution