HistoGAN: Controlling Colors of GAN-Generated and Real Images via Color Histograms

TL;DR

HistoGAN introduces a color histogram-based method to control the colors of GAN-generated and real images, enabling intuitive color manipulation while preserving image content, with applications in artistic and graphic design.

Contribution

The paper presents HistoGAN, a novel approach that modifies StyleGAN to control image colors via histograms and extends it to real images with ReHistoGAN for unsupervised recoloring.

Findings

HistoGAN effectively controls colors of generated images using histograms.

ReHistoGAN enables unsupervised recoloring of real images while preserving content.

The histogram-based method outperforms existing strategies in color control quality.

Abstract

While generative adversarial networks (GANs) can successfully produce high-quality images, they can be challenging to control. Simplifying GAN-based image generation is critical for their adoption in graphic design and artistic work. This goal has led to significant interest in methods that can intuitively control the appearance of images generated by GANs. In this paper, we present HistoGAN, a color histogram-based method for controlling GAN-generated images' colors. We focus on color histograms as they provide an intuitive way to describe image color while remaining decoupled from domain-specific semantics. Specifically, we introduce an effective modification of the recent StyleGAN architecture to control the colors of GAN-generated images specified by a target color histogram feature. We then describe how to expand HistoGAN to recolor real images. For image recoloring, we jointly train an encoder network along with HistoGAN. The recoloring model, ReHistoGAN, is an unsupervised approach trained to encourage the network to keep the original image's content while changing the colors based on the given target histogram. We show that this histogram-based approach offers a better way to control GAN-generated and real images' colors while producing more compelling results compared to existing alternative strategies.