A Neural Algorithm of Artistic Style implementation – Neural Style Transfer
ArtiStyle
To quote authors Leon A. Gatys, Alexander S. Ecker, Matthias Bethge, “in light of the striking similarities between performance-optimised artificial neural networks and biological vision, our work offers a path forward to an algorithmic understanding of how humans create and perceive artistic imagery.
The idea of Neural Style Transfer is taking a white noise as an input image, changing the input in such a way that it resembles the content of the content image and the texture/artistic style of the style image to reproduce it as a new artistic stylized image.
We define two distances, one for the content that measures how different the content between the two images is, and one for style that measures how different the style between the two images is.