richzhang/colorization
A deep neural network that automatically colorizes grayscale images into plausible color versions.
Not currently ranked — collecting fresh signals.
star history
The project implements deep learning models for automatic image colorization, converting images to Lab color space, applying neural network colorization at 256x256 resolution, and upscaling back to full resolution. It includes both the original ECCV 2016 model and the interactive SIGGRAPH 2017 extension. The repo now offers a PyTorch implementation alongside the historical Caffe code.
Frequently asked
- What is richzhang/colorization?
- A deep neural network that automatically colorizes grayscale images into plausible color versions.
- Is colorization open source?
- Yes — richzhang/colorization is open source, released under the BSD-2-Clause license.
- What language is colorization written in?
- richzhang/colorization is primarily written in Python.
- How popular is colorization?
- richzhang/colorization has 3.5k stars on GitHub.
- Where can I find colorization?
- richzhang/colorization is on GitHub at https://github.com/richzhang/colorization.