Applying StyleNet and the neural style project

Once we have an image recognition CNN trained, we can use the network itself for some interesting data and image processing. StyleNet is a procedure that attempts to learn an image style from one picture and apply it to a second picture while keeping the second image structure (or content) intact. To do so, we have to find intermediate CNN nodes that correlate strongly with a style, separately from the content of the image.

StyleNet is a procedure that takes two images and applies the style of one image to the content of the second image. It is based on a famous paper by Leon Gatys in 2015, A Neural Algorithm of Artistic Style (refer to the first bullet point under the next See also section). The authors found a property in some CNNs containing intermediate layers. Some of them seem to encode the style of a picture, and some others its content. To this end, if we train the style layers on the style picture...