D2Dewarp: Dual Dimensions Geometric Representation Learning Based Document Image Dewarping

TL;DR

D2Dewarp introduces a dual-dimension geometric learning approach for document image dewarping, utilizing horizontal and vertical line features and a new annotated dataset to improve rectification accuracy.

Contribution

The paper presents a novel dual-dimension deformation perception model and an automatic annotation method, enhancing document dewarping performance over existing single-dimension approaches.

Findings

Outperforms state-of-the-art methods on benchmark datasets.

Achieves superior rectification quality both quantitatively and qualitatively.

Introduces a large-scale annotated dataset for training and evaluation.

Abstract

Document image dewarping remains a challenging task in the deep learning era. While existing methods have improved by leveraging text line awareness, they typically focus only on a single horizontal dimension. In this paper, we propose a fine-grained deformation perception model that focuses on Dual Dimensions of document horizontal-vertical-lines to improve document Dewarping called D2Dewarp. It can perceive distortion trends in different directions across document details. To combine the horizontal and vertical granularity features, an effective fusion module based on X and Y coordinate is designed to facilitate interaction and constraint between the two dimensions for feature complementarity. Due to the lack of annotated line features in current public dewarping datasets, we also propose an automatic fine-grained annotation method using public document texture images and automatic rendering engine to build a new large-scale distortion training dataset named DocDewarpHV. On three public Chinese and English benchmarks, both quantitative and qualitative results show that our method achieves better rectification results compared with the state-of-the-art methods. The code and dataset are available at https://github.com/xiaomore/D2Dewarp.