Text-Aware Real-World Image Super-Resolution via Diffusion Model with Joint Segmentation Decoders

TL;DR

This paper presents TADiSR, a diffusion-based super-resolution framework that enhances text clarity and structural fidelity in real-world degraded images by integrating text-aware attention and joint segmentation decoders.

Contribution

The paper introduces a novel diffusion model with text-aware attention and segmentation decoders, improving text structure preservation in super-resolution of real-world images.

Findings

Achieves state-of-the-art performance in text super-resolution.

Significantly improves text legibility in super-resolved images.

Demonstrates strong generalization to real-world scenarios.

Abstract

The introduction of generative models has significantly advanced image super-resolution (SR) in handling real-world degradations. However, they often incur fidelity-related issues, particularly distorting textual structures. In this paper, we introduce a novel diffusion-based SR framework, namely TADiSR, which integrates text-aware attention and joint segmentation decoders to recover not only natural details but also the structural fidelity of text regions in degraded real-world images. Moreover, we propose a complete pipeline for synthesizing high-quality images with fine-grained full-image text masks, combining realistic foreground text regions with detailed background content. Extensive experiments demonstrate that our approach substantially enhances text legibility in super-resolved images, achieving state-of-the-art performance across multiple evaluation metrics and exhibiting strong generalization to real-world scenarios. Our code is available at \href{https://github.com/mingcv/TADiSR}{here}.