GL-PGENet: A Parameterized Generation Framework for Robust Document Image Enhancement

TL;DR

GL-PGENet is a novel multi-degradation document image enhancement framework that combines hierarchical global-local refinement, parametric generation, and dense feature fusion to achieve state-of-the-art results and robustness in real-world applications.

Contribution

The paper introduces a hierarchical enhancement architecture with parametric generation and a dense fusion network, improving robustness and generalization for multi-degraded document images.

Findings

Achieves state-of-the-art SSIM scores of 0.7721 on DocUNet and 0.9480 on RealDAE.

Demonstrates excellent cross-domain adaptability and efficiency on high-resolution images.

Outperforms existing methods in real-world document image enhancement tasks.

Abstract

Document Image Enhancement (DIE) serves as a critical component in Document AI systems, where its performance substantially determines the effectiveness of downstream tasks. To address the limitations of existing methods confined to single-degradation restoration or grayscale image processing, we present Global with Local Parametric Generation Enhancement Network (GL-PGENet), a novel architecture designed for multi-degraded color document images, ensuring both efficiency and robustness in real-world scenarios. Our solution incorporates three key innovations: First, a hierarchical enhancement framework that integrates global appearance correction with local refinement, enabling coarse-to-fine quality improvement. Second, a Dual-Branch Local-Refine Network with parametric generation mechanisms that replaces conventional direct prediction, producing enhanced outputs through learned intermediate parametric representations rather than pixel-wise mapping. This approach enhances local consistency while improving model generalization. Finally, a modified NestUNet architecture incorporating dense block to effectively fuse low-level pixel features and high-level semantic features, specifically adapted for document image characteristics. In addition, to enhance generalization performance, we adopt a two-stage training strategy: large-scale pretraining on a synthetic dataset of 500,000+ samples followed by task-specific fine-tuning. Extensive experiments demonstrate the superiority of GL-PGENet, achieving state-of-the-art SSIM scores of 0.7721 on DocUNet and 0.9480 on RealDAE. The model also exhibits remarkable cross-domain adaptability and maintains computational efficiency for high-resolution images without performance degradation, confirming its practical utility in real-world scenarios.