Artificial intelligence code is open-sourced by Google developers, with internet users sharing psychedelic images under #deepdream hashtag
Two weeks ago, Google engineers released the bizarre results of an artificial intelligence experiment, which saw photos interpreted and edited by the company’s “neural network”.
The “DeepDream” software, based on code used to detect faces and other patterns in images, would find such a pattern, edit the image slightly to make it look more like that pattern, and repeat.
Over the course of multiple iterations, everyday photos would morph into psychedelic, abstract, images ranging from the beautiful to the grotesque. Many of the edited images featured animal eyes and faces, since that is what the software had been “trained” to recognise.
Although Google’s algorithms can land the company it hot water – this week, it emerged that its photo software assigned black people the “gorilla” tag – it can also be used to produce art, engineers suggested.
Advertisement
“The techniques presented here help us understand and visualize how neural networks are able to carry out difficult classification tasks, improve network architecture, and check what the network has learned during training,” engineers wrote at the time.
“It also makes us wonder whether neural networks could become a tool for artists—a new way to remix visual concepts—or perhaps even shed a little light on the roots of the creative process in general.”
On Wednesday night, Google made the code for the tool public, posting it on open source site Github so that anybody can download and try it out.
The results
Internet users took to Facebook, Twitter and Google+ to publish their own images under the #deepdream hashtag.
They ranged from the strange
To the trippy
And, obviously, cat pictures
To the downright weird
You can read more about the DeepDream software, and find out how to download it yourself, at Google’s Research blog.
Sent from my Tricorder