This repository contains two approaches to image colorization inspired by the paper by Zhang et al. (2016) which can be found here https://arxiv.org/pdf/1603.08511.pdf. The first approach (/code/classical_approach.py) takes images as an input, converts them to Lab color space and uses the L (luminance) channel as the input and the ab (color) channels as the target. The second approach (/code/classification_approach) transforms the problem to a classification task. For further details, please read our report in the /report folder.
The 'submit' files are used to distribute the scripts on our university server. If you want to run our code on your own device, you just have to change the variables that the script collects from the 'submit' files and assign your own values to them. They can be found at the beginning of the main function. Additionally, you would have to change the paths found in the scripts to save the results on your local device. That's it - have fun experimenting!