Axis-Aligned Document Dewarping

TL;DR

This paper introduces a novel axis-aligned geometric constraint and a new evaluation metric for document dewarping, leveraging the inherent geometric properties of physical documents to improve accuracy and robustness.

Contribution

It proposes an axis-aligned geometric constraint, an axis alignment preprocessing strategy, and a new metric, Axis-Aligned Distortion, to enhance document dewarping performance.

Findings

Achieves state-of-the-art results on multiple benchmarks.

Improves the AAD metric by 18.2% to 34.5%.

Demonstrates robustness and geometric consistency in dewarping.

Abstract

Document dewarping is crucial for many applications. However, existing learning-based methods rely heavily on supervised regression with annotated data without fully leveraging the inherent geometric properties of physical documents. Our key insight is that a well-dewarped document is defined by its axis-aligned feature lines. This property aligns with the inherent axis-aligned nature of the discrete grid geometry in planar documents. Harnessing this property, we introduce three synergistic contributions: for the training phase, we propose an axis-aligned geometric constraint to enhance document dewarping; for the inference phase, we propose an axis alignment preprocessing strategy to reduce the dewarping difficulty; and for the evaluation phase, we introduce a new metric, Axis-Aligned Distortion (AAD), that not only incorporates geometric meaning and aligns with human visual perception but also demonstrates greater robustness. As a result, our method achieves state-of-the-art performance on multiple existing benchmarks, improving the AAD metric by 18.2% to 34.5%. The code is publicly available at https://github.com/chaoyunwang/AADD.