/Style-Transfer-using-Pytorch

Using Pytorch library to design a good style transfer model.

Primary LanguageJupyter Notebook

Style-Transfer-using-Pytorch

In this notebook, i'm trying to recreate a style transfer method that is outlined in the paper, Image Style Transfer Using Convolutional Neural Networks, by Gatys in PyTorch link to Paper!.

This paper uses VGG19 model to create a new target image which should contain desired content and style components:

  • objects and their arrangement are similar to that of the content image

  • style, colors, and textures are similar to that of the style image

Test the model

Here is the content & style of the first example

And the target of applying the model over the two images

Target1

Achieving a loss of 305766.5625 which is better than the referenced example here CV Foundation

Here is the content & style of the second example

And the target of applying the model over the two images

Target2

Achieving a loss of 289863.03125 which is better than the referenced example here GitHub

Here is the content & style of the third example

And the target of applying the model over the two images

Target3

Achieving a loss of 13028.4619140625 which is better than the referenced example here Pytorch

Acknowledgement

The code benefits from outstanding prior work and their implementations including:

  • style-transfer GitHub repository (Code!)

  • Artistic Neural Style Transfer with PyTorch (Link!)

  • NEURAL TRANSFER USING PYTORCH (Link!)