On Cropped versus Uncropped Training Sets in Tabular Structure Detection

TL;DR

This paper compares the effectiveness of cropped versus uncropped datasets in table structure detection, revealing that cropping improves detection performance especially at higher IoU thresholds, with minimal impact at lower thresholds.

Contribution

It provides the first systematic analysis of how dataset cropping affects deep learning-based table structure detection performance.

Findings

Cropped datasets improve detection by up to 9% in average precision and recall.

Impact of cropping is negligible at IoU 50%-70%.

Cropped datasets outperform uncropped ones at IoU above 70%.

Abstract

Automated document processing for tabular information extraction is highly desired in many organizations, from industry to government. Prior works have addressed this problem under table detection and table structure detection tasks. Proposed solutions leveraging deep learning approaches have been giving promising results in these tasks. However, the impact of dataset structures on table structure detection has not been investigated. In this study, we provide a comparison of table structure detection performance with cropped and uncropped datasets. The cropped set consists of only table images that are cropped from documents assuming tables are detected perfectly. The uncropped set consists of regular document images. Experiments show that deep learning models can improve the detection performance by up to 9% in average precision and average recall on the cropped versions. Furthermore, the impact of cropped images is negligible under the Intersection over Union (IoU) values of 50%-70% when compared to the uncropped versions. However, beyond 70% IoU thresholds, cropped datasets provide significantly higher detection performance.