Colorization Model for Black and White Images

Abstract

In recent years, researchers' interest in image colorization has grown significantly, particularly with regard to deep learning-based image colorization approaches (DLIC). This work takes a novel deep learning-based approach to analyze the most recent developments in image colorization approaches methodically and extensively. It tackles the issue of imagining a believable color version of a grayscale image given only the input. Previous solutions have either required a lot of user engagement or produced desaturated colorizations, but this work provides a fully automated method that results in vivid and accurate colorizations. This paper discusses the introduction and effectiveness of Open CV and Generative Adversarial Networks (GANs) for automatic picture colorization. It focuses on the introduction and effectiveness of Open CV and GANs, their structure, functionality and extent of research for image colorization. It also discusses results, several open issues of image colorization and outline future research directions. This paper can serve as a reference for researchers in image colorization and related fields.