Recognition-Guided Diffusion Model for Scene Text Image Super-Resolution

TL;DR

This paper introduces RGDiffSR, a diffusion-based super-resolution model guided by recognition cues, significantly improving the quality and recognition accuracy of low-resolution scene text images, especially in challenging cases.

Contribution

The paper proposes a novel recognition-guided diffusion model for scene text super-resolution, enhancing generative diversity and fidelity over previous CNN-based methods.

Findings

Outperforms state-of-the-art methods on TextZoom dataset

Achieves higher recognition accuracy in low-quality images

Demonstrates robustness in severely blurred scenarios

Abstract

Scene Text Image Super-Resolution (STISR) aims to enhance the resolution and legibility of text within low-resolution (LR) images, consequently elevating recognition accuracy in Scene Text Recognition (STR). Previous methods predominantly employ discriminative Convolutional Neural Networks (CNNs) augmented with diverse forms of text guidance to address this issue. Nevertheless, they remain deficient when confronted with severely blurred images, due to their insufficient generation capability when little structural or semantic information can be extracted from original images. Therefore, we introduce RGDiffSR, a Recognition-Guided Diffusion model for scene text image Super-Resolution, which exhibits great generative diversity and fidelity even in challenging scenarios. Moreover, we propose a Recognition-Guided Denoising Network, to guide the diffusion model generating LR-consistent results through succinct semantic guidance. Experiments on the TextZoom dataset demonstrate the superiority of RGDiffSR over prior state-of-the-art methods in both text recognition accuracy and image fidelity.