How do GANs create an image with specific characteristics?


I've seen GANs that do things like convert an image to a painting or this GAN here that takes in a set of characteristics and generates a waifu with those characteristics.

My understanding of a GAN is that the generator upsamples random noise and the discriminator detects if an image is within the real or fake. So if the generator say, generates a waifu with the wrong hair color, how would the discriminator know?

Bryan Tan

Posted 2019-05-29T23:12:22.280

Reputation: 163

No answers