Scene Text Image Super-resolution based on Text-conditional Diffusion Models

TL;DR

This paper introduces a novel scene text image super-resolution framework using text-conditional diffusion models, significantly improving image quality and dataset synthesis for better scene text recognition.

Contribution

It leverages text-conditional diffusion models for super-resolution and dataset synthesis, surpassing existing methods and enhancing STISR performance.

Findings

Text-conditional DMs outperform existing STISR methods.

Synthesized LR-HR image pairs improve STISR training.

Proposed framework enhances scene text recognition accuracy.

Abstract

Scene Text Image Super-resolution (STISR) has recently achieved great success as a preprocessing method for scene text recognition. STISR aims to transform blurred and noisy low-resolution (LR) text images in real-world settings into clear high-resolution (HR) text images suitable for scene text recognition. In this study, we leverage text-conditional diffusion models (DMs), known for their impressive text-to-image synthesis capabilities, for STISR tasks. Our experimental results revealed that text-conditional DMs notably surpass existing STISR methods. Especially when texts from LR text images are given as input, the text-conditional DMs are able to produce superior quality super-resolution text images. Utilizing this capability, we propose a novel framework for synthesizing LR-HR paired text image datasets. This framework consists of three specialized text-conditional DMs, each dedicated to text image synthesis, super-resolution, and image degradation. These three modules are vital for synthesizing distinct LR and HR paired images, which are more suitable for training STISR methods. Our experiments confirmed that these synthesized image pairs significantly enhance the performance of STISR methods in the TextZoom evaluation.