Table Detection in the Wild: A Novel Diverse Table Detection Dataset and Method

TL;DR

This paper introduces a large, diverse dataset for table detection in documents and demonstrates that deep learning methods outperform classical approaches, advancing the development of robust table detection systems.

Contribution

The paper provides a new large-scale, diverse table detection dataset and baseline deep learning methods, addressing limitations of previous benchmarks.

Findings

Deep learning methods outperform classical computer vision approaches.

The dataset contains over 7,000 samples with diverse table structures.

Baseline results establish a foundation for future research.

Abstract

Recent deep learning approaches in table detection achieved outstanding performance and proved to be effective in identifying document layouts. Currently, available table detection benchmarks have many limitations, including the lack of samples diversity, simple table structure, the lack of training cases, and samples quality. In this paper, we introduce a diverse large-scale dataset for table detection with more than seven thousand samples containing a wide variety of table structures collected from many diverse sources. In addition to that, we also present baseline results using a convolutional neural network-based method to detect table structure in documents. Experimental results show the superiority of applying convolutional deep learning methods over classical computer vision-based methods. The introduction of this diverse table detection dataset will enable the community to develop high throughput deep learning methods for understanding document layout and tabular data processing. Dataset is available at: 1. https://www.kaggle.com/datasets/mrinalim/stdw-dataset 2. https://huggingface.co/datasets/n3011/STDW