ChromaGAN: Adversarial Picture Colorization with Semantic Class Distribution
This addresses the challenge of generating realistic color images from grayscale inputs for applications in photography and computer vision, representing a strong incremental improvement over existing methods.
The paper tackled the ill-posed problem of grayscale image colorization by proposing ChromaGAN, an adversarial learning approach that incorporates semantic information to infer chromaticity, achieving state-of-the-art results in realistic colorization.
The colorization of grayscale images is an ill-posed problem, with multiple correct solutions. In this paper, we propose an adversarial learning colorization approach coupled with semantic information. A generative network is used to infer the chromaticity of a given grayscale image conditioned to semantic clues. This network is framed in an adversarial model that learns to colorize by incorporating perceptual and semantic understanding of color and class distributions. The model is trained via a fully self-supervised strategy. Qualitative and quantitative results show the capacity of the proposed method to colorize images in a realistic way achieving state-of-the-art results.