Image-based table recognition: data, model, and evaluation (2019-11-25T00:00:00.000000Z)

TL;DR

The largest publicly available table recognition dataset PubTabNet is developed, containing 568k table images with corresponding structured HTML representation, and a new Tree-Edit-Distance-based Similarity (TEDS) metric for table recognition is proposed, which more appropriately captures multi-hop cell misalignment and OCR errors than the pre-established metric.

Abstract

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.

Authors

Xu Zhong

3 papers

Antonio Jimeno-Yepes

4 papers

Elaheh Shafieibavani

1 papers

TL;DR

Abstract

Authors

References42 items

ICDAR 2019 Competition on Table Detection and Recognition (cTDaR)

Challenges in End-to-End Neural Scientific Table Recognition

Table Detection in Invoice Documents by Graph Neural Networks

Deep Splitting and Merging for Table Structure Decomposition

TableNet: Deep Learning Model for End-to-end Table Detection and Tabular Data Extraction from Scanned Document Images

PubLayNet: Largest Dataset Ever for Document Layout Analysis

BranchGAN: Unsupervised Mutual Image-to-Image Transfer With A Single Encoder and Dual Decoders

Rethinking Table Recognition using Graph Neural Networks

Learning Regularity in Skeleton Trajectories for Anomaly Detection in Videos

Corpus Conversion Service: A Machine Learning Platform to Ingest Documents at Scale

A Saliency-based Convolutional Neural Network for Table and Chart Detection in Digitized Documents

Extracting Scientific Figures with Distantly Supervised Neural Networks

DeepDeSRT: Deep Learning for Detection and Structure Recognition of Tables in Document Images

Table Detection Using Deep Learning

Multi-Scale Multi-Task FCN for Semantic Page Segmentation and Table Detection

Mask R-CNN

Image-to-Markup Generation with Coarse-to-Fine Attention

A Table Detection Method for PDF Documents Based on Convolutional Neural Networks

Tree edit distance: Robust and memory-efficient

Deep Residual Learning for Image Recognition

Table Region Detection on Large-scale PDF Files without Labeled Data

You Only Look Once: Unified, Real-Time Object Detection

Faster R-CNN: Towards Real-Time Object Detection with Region Proposal Networks

Show, Attend and Tell: Neural Image Caption Generation with Visual Attention

Adam: A Method for Stochastic Optimization

Literature mining of genetic variants for curation: quantifying the importance of supplementary material

Learning to Detect Tables in Scanned Document Images Using Line Information

ICDAR 2013 Table Competition

Dataset, Ground-Truth and Performance Metrics for Table Detection Evaluation

An open approach towards the benchmarking of table structure recognition systems

Table detection in heterogeneous documents

Learning Rich Hidden Markov Models in Document Analysis: Table Location

Automatic Table Detection in Document Images

A constraint-based approach to table structure derivation

Object recognition supported by user interaction for service robots

Trainable table location in document images

Medium-independent table detection

The T-Recs Table Recognition and Analysis System

A method for table structure analysis using DP matching

Mediumindependent table detection,” in Document Recognition and Retrieval VII, vol. 3967

Extracting Tabular Information From Text Files

Binary codes capable of correcting deletions, insertions, and reversals

Field of Study

Venue Information

Name

Type

URL

Alternate Names