via towardsdatascience.com

Style Transfer – Styling Images with Convolutional Neural Networks

Creating Beautiful Image Effects

by Greg Surma

In today’s article, we are going to create remarkable style transfer effects. In order to do so, we will have to get a deeper understanding of how Convolutional Neural Networks and its layers work. By the end of this article, you will be able to create a style transfer application that is able to apply a new style to an image while still preserving its original content.

Style Transfer

Before we go to our Style Transfer application, let’s clarify what we are striving to achieve.

Let’s define a style transfer as a process of modifying the style of an image while still preserving its content.

Given an input image and a style image, we can compute an output image with the original content but a new style. It was outlined in Leon A. Gatys’ paper, A Neural Algorithm of Artistic Style, which is a great publication, and you should definitely check it out.

Input Image + Style Image -> Output Image (Styled Input)

How does it work?

  1. We take input image and style images and resize them to equal shapes.
  2. We load a pre-trained Convolutional Neural Network (VGG16).
  3. Knowing that we can distinguish layers that are responsible for the style (basic shapes, colors etc.) and the ones responsible for the content (image-specific features), we can separate the layers to independently work on the content and style.
  4. Then we set our task as an optimization problem where we are going to minimize:
  • content loss (distance between the input and output images – we strive to preserve the content)
  • style loss (distance between the style and output images – we strive to apply a new style)
  • total variation loss (regularization – spatial smoothness to denoise the output image)

5. Finally, we set our gradients and optimize with the L-BFGS algorithm.


While above high-level overview may seem confusing, let’s go straight to the code!

“What I cannot create, I do not understand.” – Richard Feynman

READ THE ARTICLE