Image Colorization Using a Deep Convolutional Neural Network

TL;DR

This paper introduces a deep learning-based method for colorizing grayscale images by leveraging pre-trained convolutional neural networks to transfer style and content, demonstrated on ukiyo-e art images.

Contribution

It presents a novel approach that combines content and style transfer using pre-trained CNNs specifically for image colorization, including a new application to ukiyo-e art.

Findings

Effective colorization of ukiyo-e images

Demonstrates potential for computer-assisted art

Utilizes pre-trained CNNs for style transfer

Abstract

In this paper, we present a novel approach that uses deep learning techniques for colorizing grayscale images. By utilizing a pre-trained convolutional neural network, which is originally designed for image classification, we are able to separate content and style of different images and recombine them into a single image. We then propose a method that can add colors to a grayscale image by combining its content with style of a color image having semantic similarity with the grayscale one. As an application, to our knowledge the first of its kind, we use the proposed method to colorize images of ukiyo-e a genre of Japanese painting?and obtain interesting results, showing the potential of this method in the growing field of computer assisted art.