RankSRGAN: Super Resolution Generative Adversarial Networks with Learning to Rank
This work addresses the challenge of improving perceptual quality in super-resolution for applications like image enhancement, though it is incremental as it builds on existing GAN methods.
The authors tackled the problem of optimizing indifferentiable perceptual metrics in single image super-resolution by proposing RankSRGAN, which uses a Ranker to learn metric behavior and a rank-content loss, achieving state-of-the-art performance in perceptual metrics and visual quality.
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