About us
Stay in touch
Deep convolutional GANs, published in 2015 by Alec Radford, Luke Metz and Soumith Chintala. They made GANs stable enough to train that artists began using them widely.
So DCGAN was able to be more serious about having multiple convolution layers, really taking a convolution network and turning it upside down completely and having a fully convolutional generator. They were able to get much better images. That's when GANs really started to take off.
Ian Goodfellow, Ian Goodfellow on Inventing GANs
The DCGAN was crude then, almost laughably so, but there was something profound in watching it hallucinate new petals from a learned latent space.
Sarah Meyohas, Sarah Meyohas on Irreducibly Human