This tutorial is one of the demos used in the course “Introduction to Machine Learning” at MINES ParisTech - PSL Research University, lectured by Simon Tamayo.
Here is what Style transfer looks like:
- You have an input_image:
- You have an input_style (that you want to apply to your image):
- The output a new image with keeps the features of your input_image while changing its style:
Most of this demo was adapted from the PyTorch tutorial on Style Transfer (https://pytorch.org/tutorials/advanced/neural_style_tutorial.html), which explains how to implement the Neural-Style algorithm developed by L. A. Gatys, A. S. Ecker and M. Bethge.
Before running this code make sure you have Python3 and the packages: PyTorch, PIL, TorchVision and Matplotlib.
This tutorial uses the convolutional neural network VGG-19, which was trained on more than a million images from the ImageNet database. The network is 19 layers deep and can classify images into 1000 object categories
- PyTorch - Deep learning platform
- TorchVision - package withpopular datasets and model architectures for computer vision.
This project is licensed under the MIT License - see the LICENSE.md file for details.