Incorporating Uncertainty-Guided and Top-k Codebook Matching for Real-World Blind Image Super-Resolution

TL;DR

This paper introduces UGTSR, a novel super-resolution framework that leverages uncertainty-guided and Top-k codebook matching to improve texture detail and reconstruction accuracy in real-world images.

Contribution

The paper proposes a new super-resolution method combining uncertainty learning, Top-k feature matching, and an Align-Attention module for better texture and detail reconstruction.

Findings

Significant improvements in texture realism over existing methods

Enhanced feature matching accuracy with Top-k strategy

Better alignment of LR and HR features through the Align-Attention module

Abstract

Recent advancements in codebook-based real image super-resolution (SR) have shown promising results in real-world applications. The core idea involves matching high-quality image features from a codebook based on low-resolution (LR) image features. However, existing methods face two major challenges: inaccurate feature matching with the codebook and poor texture detail reconstruction. To address these issues, we propose a novel Uncertainty-Guided and Top-k Codebook Matching SR (UGTSR) framework, which incorporates three key components: (1) an uncertainty learning mechanism that guides the model to focus on texture-rich regions, (2) a Top-k feature matching strategy that enhances feature matching accuracy by fusing multiple candidate features, and (3) an Align-Attention module that enhances the alignment of information between LR and HR features. Experimental results demonstrate significant improvements in texture realism and reconstruction fidelity compared to existing methods. We will release the code upon formal publication.