Guided Table Structure Recognition through Anchor Optimization

TL;DR

This paper introduces a guided anchor-based method for table structure recognition that outperforms existing approaches by accurately locating rows and columns in tabular images, achieving state-of-the-art results on benchmark datasets.

Contribution

It proposes a novel guided anchor optimization technique for table structure recognition, differing from naive object detection methods, and demonstrates its effectiveness on public datasets.

Findings

Achieved 95.05% F-Measure on ICDAR-2013 dataset.

Surpassed baseline results on TabStructDB dataset.

Improved accuracy in locating table rows and columns.

Abstract

This paper presents the novel approach towards table structure recognition by leveraging the guided anchors. The concept differs from current state-of-the-art approaches for table structure recognition that naively apply object detection methods. In contrast to prior techniques, first, we estimate the viable anchors for table structure recognition. Subsequently, these anchors are exploited to locate the rows and columns in tabular images. Furthermore, the paper introduces a simple and effective method that improves the results by using tabular layouts in realistic scenarios. The proposed method is exhaustively evaluated on the two publicly available datasets of table structure recognition i.e ICDAR-2013 and TabStructDB. We accomplished state-of-the-art results on the ICDAR-2013 dataset with an average F-Measure of 95.05$\%$ (94.6$\%$ for rows and 96.32$\%$ for columns) and surpassed the baseline results on the TabStructDB dataset with an average F-Measure of 94.17$\%$ (94.08$\%$ for rows and 95.06$\%$ for columns).