Google Releases Source Code for Machine Dreaming Software, Psychedelic Images Flood the Web

Google has just released the source code for its 'Deep Dream' software on the website github and the internet has quickly started putting it to work, creating all sorts of trippy, hallucinogenic images.

'Deep Dream' was the result of an artificial intelligence experiment the search giant conducted two weeks ago to test its new photo recognition software. James Titcomb explains the methodology behind the test in his article for the Telegraph:

In an experiment, engineers at Google's research labs ran various pictures through its neural network, asking the software to identify patterns in the images and then alter that image to exaggerate the patterns.

In other words, it sees a pattern it thinks it recognises, such as a face or a door, and makes the picture look a little bit more like that thing. The neural network then repeats the process with the altered image.

If that happens enough times, it will change a picture radically.

Google is pushing heavily into the study of neural networks which is why it decided to go public with the code. In an official blog post they explain:

"We have seen a lot of interest and received some great questions, from programmers and artists alike, about the details of how these visualizations are made.We have decided to open source the code we used to generate these images in an IPython notebook, so now you can make neural network inspired images yourself!"

Now Deep Dream is free for anyone to experiment with and social media is being flooded with the otherworldly results. The hashtag "#deepdream" already has a myriad of interesting images available for viewing.

It seems that the code has a preference toward animals, as some of the more highly publicized photos show regular landscapes and panoramas transformed into the heads of various wildlife once put through the Deep Dream process.

 One such example is an image of a stream flowing through a forest. Once it was subject to Deep Dream, certain shapes found in the leaves and water were transformed into birds.

Real Time Analytics