Image-based table recognition: data, model, and evaluation

By Xu Zhong, Elaheh Shafieibavani, Antonio Jimeno-Yepes. European Conference on Computer Vision, 2019

358 citations recorded by the source corpus.

Important information that relates to a specific topic in a document is often organized in tabular format to assist readers with information retrieval and comparison, which may be difficult to provide in natural language. However, tabular data in unstructured digital documents, e.g., Portable Document Format (PDF) and images, are difficult to parse into structured machine-readable format, due to complexity and diversity in their structure and style. To facilitate image-based table recognition with deep learning, we develop the largest publicly available table recognition dataset PubTabNet (this https URL), containing 568k table images with corresponding structured HTML representation. PubTabNet is automatically generated by matching the XML and PDF representations of the scientific articles in PubMed Central Open Access Subset (PMCOA). We also propose a novel attention-based encoder-dual-decoder (EDD) architecture that converts images of tables into HTML code. The model has a structure decoder which reconstructs the table structure and helps the cell decoder to recognize cell content. In addition, we propose a new Tree-Edit-Distance-based Similarity (TEDS) metric for table recognition, which more appropriately captures multi-hop cell misalignment and OCR errors than the pre-established metric. The experiments demonstrate that the EDD model can accurately recognize complex tables solely relying on the image representation, outperforming the state-of-the-art by 9.7% absolute TEDS score.

Subfields: Computer Vision and Pattern Recognition

View Paper: https://www.semanticscholar.org/paper/99b2a05177c17dd86627863608ac45dc4cfe0446

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Computer Vision and Pattern Recognition

Citations in the corpus

BioTABQA: Instruction Learning for Biomedical Table Question Answering (2022, background): “Recently, Parmar et al. [31] believe that instruction bias in existing Natural Language Understanding (NLU) datasets can impact the instruction”
OCR-Free Document Understanding Transformer (2021, methodology): “Similar metrics are used in recent works on document IE [70,23] We use two public benchmark datasets as well as two private industrial datasets which are from our active real-world service products.”