ShaDocNet: Learning Spatial-Aware Tokens in Transformer for Document Shadow Removal

TL;DR

This paper introduces ShaDocNet, a Transformer-based model that effectively removes shadows from digital documents by encoding shadow context and refining results through a coarse-to-fine process, outperforming existing methods.

Contribution

The paper presents a novel Transformer architecture tailored for document shadow removal, incorporating shadow context encoding, detection, and pixel-level enhancement.

Findings

Competitive with state-of-the-art methods on benchmarks

Effective shadow context encoding improves removal quality

Coarse-to-fine process enhances detail and accuracy

Abstract

Shadow removal improves the visual quality and legibility of digital copies of documents. However, document shadow removal remains an unresolved subject. Traditional techniques rely on heuristics that vary from situation to situation. Given the quality and quantity of current public datasets, the majority of neural network models are ill-equipped for this task. In this paper, we propose a Transformer-based model for document shadow removal that utilizes shadow context encoding and decoding in both shadow and shadow-free regions. Additionally, shadow detection and pixel-level enhancement are included in the whole coarse-to-fine process. On the basis of comprehensive benchmark evaluations, it is competitive with state-of-the-art methods.