March 30, 2019

173 words 1 min read



Style transfer, deep learning, feature transform

repo name NVIDIA/FastPhotoStyle
repo link
language Python
size (curr.) 51946 kB
stars (curr.) 10136
created 2018-02-14
license Other

License CC BY-NC-SA 4.0 Python 2.7 Python 3.5



Copyright (C) 2018 NVIDIA Corporation. All rights reserved. Licensed under the CC BY-NC-SA 4.0 license (

What’s new

Date News
2018-07-25 Migrate to pytorch 0.4.0. For pytorch 0.3.0 user, check out FastPhotoStyle for pytorch 0.3.0.
Add a tutorial showing 3 ways of using the FastPhotoStyle algorithm.
2018-07-10 Our paper is accepted by the ECCV 2018 conference!!!


Given a content photo and a style photo, the code can transfer the style of the style photo to the content photo. The details of the algorithm behind the code is documented in our arxiv paper. Please cite the paper if this code repository is used in your publications.

A Closed-form Solution to Photorealistic Image Stylization Yijun Li (UC Merced), Ming-Yu Liu (NVIDIA), Xueting Li (UC Merced), Ming-Hsuan Yang (NVIDIA, UC Merced), Jan Kautz (NVIDIA) European Conference on Computer Vision (ECCV), 2018


Please check out the tutorial.

comments powered by Disqus