Exploiting Style and Attention in Real-World Super-Resolution

TL;DR

This paper introduces a novel super-resolution pipeline that leverages style transfer and attention mechanisms to better handle real-world low-resolution images, improving generalization and robustness.

Contribution

It proposes a styleVAE for realistic LR image generation and a SR network with global and local attention residual blocks, enhancing super-resolution performance on real-world data.

Findings

Outperforms state-of-the-art methods quantitatively

Achieves superior qualitative image quality

Enhances robustness and generalization of SR models

Abstract

Real-world image super-resolution (SR) is a challenging image translation problem. Low-resolution (LR) images are often generated by various unknown transformations rather than by applying simple bilinear down-sampling on high-resolution (HR) images. To address this issue, this paper proposes a novel pipeline which exploits style and attention mechanism in real-world SR. Our pipeline consists of a style Variational Autoencoder (styleVAE) and a SR network incorporated with attention mechanism. To get real-world-like low-quality images paired with the HR images, we design the styleVAE to transfer the complex nuisance factors in real-world LR images to the generated LR images. We also use mutual information estimation (MI) to get better style information. For our SR network, we firstly propose a global attention residual block to learn long-range dependencies in images. Then another local attention residual block is proposed to enforce the attention of SR network moving to local areas of images in which texture detail will be filled. It is worth noticing that styleVAE can be presented in a plug-and-play manner and thus can help to improve the generalization and robustness of our SR method as well as other SR methods. Extensive experiments demonstrate that our method surpasses the state-of-the-art work, both quantitatively and qualitatively.