RankSRGAN: Super Resolution Generative Adversarial Networks with Learning to Rank

TL;DR

RankSRGAN introduces a novel approach that trains a ranker to optimize super-resolution GANs directly for perceptual metrics, leading to improved visual quality and state-of-the-art results in perceptual quality assessment.

Contribution

The paper proposes a new method that learns to optimize super-resolution GANs for perceptual metrics by using a ranker and rank-content loss, enhancing visual quality.

Findings

Achieves state-of-the-art perceptual quality in super-resolution.

Effectively combines multiple perceptual metrics for better results.

Produces visually pleasing super-resolved images.

Abstract

Generative Adversarial Networks (GAN) have demonstrated the potential to recover realistic details for single image super-resolution (SISR). To further improve the visual quality of super-resolved results, PIRM2018-SR Challenge employed perceptual metrics to assess the perceptual quality, such as PI, NIQE, and Ma. However, existing methods cannot directly optimize these indifferentiable perceptual metrics, which are shown to be highly correlated with human ratings. To address the problem, we propose Super-Resolution Generative Adversarial Networks with Ranker (RankSRGAN) to optimize generator in the direction of different perceptual metrics. Specifically, we first train a Ranker which can learn the behaviour of perceptual metrics and then introduce a novel rank-content loss to optimize the perceptual quality. The most appealing part is that the proposed method can combine the strengths of different SR methods to generate better results. Furthermore, we extend our method to multiple Rankers to provide multi-dimension constraints for the generator. Extensive experiments show that RankSRGAN achieves visually pleasing results and reaches state-of-the-art performance in perceptual metrics and quality. Project page: https://wenlongzhang0517.github.io/Projects/RankSRGAN