How neural networks create images
This is fascinating stuff.
watching DGN samples form. they jostle wildly through a generative space until converging on… a cheeseburger pic.twitter.com/yEKsTgMweu
— Gene Kogan (@genekogan) August 29, 2016
According to Gene Kogan on Twitter: DGN “deep generator network” is a neural net that synthesizes realistic images by learning low-dimensional encoding over an image set and samples from it to activate neurons associated w/ a class, e.g. cheeseburger.
More examples here.
This is some seriously future shit.