TextBoost: Boosting Scene Text Fidelity in Ultra-low Bitrate Image Compression

2026-03-04

AI Summary

This work addresses the challenge of preserving both text legibility and overall visual quality in ultra-low-bitrate image compression, particularly for small-font scene text. The authors propose a decoder-side semantic guidance approach that leverages a lightweight OCR module to extract textual information, which is then integrated via adaptive guidance map generation, attention-based feature fusion, and a text-region consistency regularizer. Notably, this method decouples text enhancement from global compression without relying on conventional region-of-interest (ROI) coding. By incorporating minimal-cost semantic priors, it transcends traditional rate-distortion optimization frameworks and significantly improves text fidelity. Experiments on TextOCR and ICDAR 2015 demonstrate up to a 60.6% relative improvement in text recognition F1 score at comparable PSNR and bits per pixel (bpp), while maintaining high global image quality.

Technology Category

• Computer Vision: Learning & Optimization for CV
• Natural Language Processing: Sentence-level Semantics, Textual Inference, etc.
• Machine Learning: Learning on the Edge & Model Compression

Application Category

• Search and Retrieval-Augmented AI: Web learning to rank, online learning, and counterfactual learning for ranking
• Semantics and Knowledge: Methods to enhance, augment, integrate or synergize semantic models such as knowledge graphs and LLMs
• Economics, Online Markets and Human Computation: LLM based quality controls for crowd work

Abstract

Ultra-low bitrate image compression faces a critical challenge: preserving small-font scene text while maintaining overall visual quality. Region-of-interest (ROI) bit allocation can prioritize text but often degrades global fidelity, leading to a trade-off between local accuracy and overall image quality. Instead of relying on ROI coding, we incorporate auxiliary textual information extracted by OCR and transmitted with negligible overhead, enabling the decoder to leverage this semantic guidance. Our method, TextBoost, operationalizes this idea through three strategic designs: (i) adaptively filtering OCR outputs and rendering them into a guidance map; (ii) integrating this guidance with decoder features in a calibrated manner via an attention-guided fusion block; and (iii) enforcing guidance-consistent reconstruction in text regions with a regularizing loss that promotes natural blending with the scene. Extensive experiments on TextOCR and ICDAR 2015 demonstrate that TextBoost yields up to 60.6% higher text-recognition F1 at comparable Peak Signal-to-Noise Ratio (PSNR) and bits per pixel (bpp), producing sharper small-font text while preserving global image quality and effectively decoupling text enhancement from global rate-distortion optimization.

Problem

• ultra-low bitrate compression
• scene text preservation
• image compression
• text fidelity
• small-font text

Innovation

• scene text preservation
• ultra-low bitrate compression
• OCR-guided decoding
• attention-guided fusion
• rate-distortion optimization