jwyang / lr-gan.pytorch

Pytorch code for our ICLR 2017 paper "Layered-Recursive GAN for image generation"
151 stars 32 forks source link

why doesn't the background layer learn everything? #11

Open laodar opened 6 years ago

laodar commented 6 years ago

It seems the background layer has enough capacity to model complex image in your paper's figures, but I didnt find any regularization term about this for G in your code. It's amaing, what do you think the possible reason is?