Quasimondo ♾

@FabLabMuc@twitter.com @BL_Labs@twitter.com @Lowpolybot@twitter.com @pixelsorter@twitter.com @googlearts@twitter.com 2017 is when things got serious. I totally missed out on #DCGAN - maybe it was because I lacked the necessary hardware to train my own models. So #pix2pix became my door to the world of #GANs. I started with some next-frame-prediction experiments.
twitter.com/quasimondo/status/

Mario Klingemann on Twitter

“Here's what happens if you badly train a neural network…

Twitter
Quasimondo ♾

Both models are shallow ResNets derived from #Pix2PixHD. I first tried #pix2pix UNets, but there the models learned to cheat very quickly and just abused the first skip connection to pass the information almost uncompressed.