SCGAN: Saliency Map-guided Colorization with Generative Adversarial Network

TL;DR

SCGAN is an automatic colorization framework that jointly predicts color and saliency maps to reduce semantic confusion and color bleeding, achieving perceptually realistic results with less training data.

Contribution

It introduces a novel saliency map-guided approach with hierarchical discriminators, improving colorization quality and efficiency over existing methods.

Findings

Outperforms state-of-the-art colorization techniques on ImageNet.

Reduces semantic confusion and color bleeding in colorized images.

Requires less training data due to global feature embedding.

Abstract

Given a grayscale photograph, the colorization system estimates a visually plausible colorful image. Conventional methods often use semantics to colorize grayscale images. However, in these methods, only classification semantic information is embedded, resulting in semantic confusion and color bleeding in the final colorized image. To address these issues, we propose a fully automatic Saliency Map-guided Colorization with Generative Adversarial Network (SCGAN) framework. It jointly predicts the colorization and saliency map to minimize semantic confusion and color bleeding in the colorized image. Since the global features from pre-trained VGG-16-Gray network are embedded to the colorization encoder, the proposed SCGAN can be trained with much less data than state-of-the-art methods to achieve perceptually reasonable colorization. In addition, we propose a novel saliency map-based guidance method. Branches of the colorization decoder are used to predict the saliency map as a proxy target. Moreover, two hierarchical discriminators are utilized for the generated colorization and saliency map, respectively, in order to strengthen visual perception performance. The proposed system is evaluated on ImageNet validation set. Experimental results show that SCGAN can generate more reasonable colorized images than state-of-the-art techniques.