#281 · Primary category: Computer Vision

colorization

automatic-colorization caffe colorization computer-vision deep-learning deep-learning-algorithms

Automatic colorization using deep neural networks. "Colorful Image Colorization." In ECCV, 2016.

Project last updated:11/27/23

GitHub Stars

3.5K

Forks

922

Contributors

6

License

BSD-2-Clause

Why we included this project

The standard starting point for adding color to old black-and-white photos, archival scans, or grayscale footage is this repo from the original ECCV 2016 paper. It ships two pretrained models, the 2016 one and the later SIGGRAPH 2017 interactive version, both loadable in a few lines of PyTorch, plus a demo script that handles the Lab-space conversion for you, so you can run a folder of images without digging into the internals. The PyTorch path is kept deliberately simple for test-time use, while the original Caffe branch still contains the full training code if you want to reproduce the research or fine-tune. This is research code, not a maintained product, so expect to do your own integration, but for the core task of giving grayscale images plausible color it remains a solid baseline.

Articles for this project

No articles for this project yet.

To suggest a topic or contribute an article, contact us.

Related projects in this category