Style transfer is the technique of recomposing images in the style of other images. These were mostly created using Justin Johnson’s code based on the paper by Gatys, Ecker, and Bethge demonstrating a method for restyling images using convolutional neural networks. Instructions here, and more details here. A gallery with all of these and more style transfers can be viewed here.
Pablo Picasso painting on glass in 1937, restyled by works from his Blue, African, and Cubist periods respectively.
Mona Lisa restyled by Egyptian hieroglyphs, the Crab Nebula, and Google Maps.
Mona Lisa restyled by Picasso, van Gogh, and Monet.
Tea party and riddle scene from Alice in Wonderland restyled by 17 iconic paintings.
A more recent implementation for video style transfer, with improved frame-to-frame stability by adding an optical-flow-based loss term to the normal content/style reconstruction loss terms. Based on the paper and code by Manuel Ruder et al. Van Gogh / Hokusai / Google Maps / Basquiat applied to a video overlooking Manhattan from the J-train.
Cubist Mirror was an installation made for alt-ai, demonstrating near real-time style transfer on the webcam feed, based on the paper by Johnson, Alahi, and Li and using the chainer implementation by Yusuke Tomoto, trained on a generic cubist painting.