Implementing Image Colorization Using CNN

Abstract: The main aim of this paper is to address the problem of generating a plausible colored photograph given a grayscale image and how it can be automated. Previous approaches to coloring grayscale images often heavily relied upon human input and often produce desaturated colorizations. Inspired by [1, 2, 4], we built a convolutional neural network model over a set of images to colorize images without human input. In image processing, the availability of deep learning has allowed us to build models that can automate the rigorous tasks such as detection, classification etc.