DocRes: A Generalist Model Toward Unifying Document Image Restoration Tasks

TL;DR

DocRes is a versatile model that unifies five document image restoration tasks using a novel visual prompt method, enabling multi-task learning and achieving state-of-the-art results.

Contribution

The paper introduces DocRes, a generalist model for multiple document image restoration tasks, utilizing Dynamic Task-Specific Prompt (DTSPrompt) for flexible, efficient multi-task learning.

Findings

Achieves competitive or superior performance to state-of-the-art models.

Unifies five restoration tasks into a single model.

Demonstrates flexibility and effectiveness of DTSPrompt.

Abstract

Document image restoration is a crucial aspect of Document AI systems, as the quality of document images significantly influences the overall performance. Prevailing methods address distinct restoration tasks independently, leading to intricate systems and the incapability to harness the potential synergies of multi-task learning. To overcome this challenge, we propose DocRes, a generalist model that unifies five document image restoration tasks including dewarping, deshadowing, appearance enhancement, deblurring, and binarization. To instruct DocRes to perform various restoration tasks, we propose a novel visual prompt approach called Dynamic Task-Specific Prompt (DTSPrompt). The DTSPrompt for different tasks comprises distinct prior features, which are additional characteristics extracted from the input image. Beyond its role as a cue for task-specific execution, DTSPrompt can also serve as supplementary information to enhance the model's performance. Moreover, DTSPrompt is more flexible than prior visual prompt approaches as it can be seamlessly applied and adapted to inputs with high and variable resolutions. Experimental results demonstrate that DocRes achieves competitive or superior performance compared to existing state-of-the-art task-specific models. This underscores the potential of DocRes across a broader spectrum of document image restoration tasks. The source code is publicly available at https://github.com/ZZZHANG-jx/DocRes