jcjohnson/neural-style
A Torch implementation of neural style transfer that applies artistic style from one image to the content of another using convolutional neural networks.

Velocity · 7d
+4.7
★ / day
Trend
→steady
star history
This repository implements the seminal neural style transfer algorithm by Gatys et al. It uses a pretrained VGG CNN to separate and recombine content and style from different images, optimizing pixel values through gradient descent to match both content features and style gram matrices. The result is an artistic image that preserves the content of one image in the style of another.