On Unguided Automatic Colorization of Monochrome Images

Abstract

Image colorization is a challenging problem due to the infinite RGB solutions for a grayscale picture. Therefore, human assistance, either directly or indirectly, is essential for achieving visually plausible colorization. This paper aims to perform colorization using only a grayscale image as the data source, without any reliance on metadata or human hints. The method assumes an (arbitrary) rgb2gray model and utilizes a few simple heuristics. Despite probabilistic elements, the results are visually acceptable and repeatable, making this approach feasible (e.g. for aesthetic purposes) in domains where only monochrome visual representations exist. The paper explains the method, presents exemplary results, and discusses a few supplementary issues.