Thursday, 8 June 2017
If you thought artificial intelligence couldn’t get any better than voice recognition, you’re wrong. Google scientist Mike Tyka, is taking a new approach to art ー using AI to make digital portraits of non-existent people. According to Tyka: “For a while now I’ve been experimenting with ways to use generative neural nets to make portraits. Early experiments were based on deep dream-like approaches using backprop to the image but lately I’ve focused on GANs (generative adversarial networks). As always resolution and fine detail is really difficult to achieve.” Here are a few pieces of his work: I’d like to think…
This story continues at The Next Web
Or just read more coverage about: LG
