An open approach towards the benchmarking of table structure recognition systems

Shahab, Asif; Shafait, Faisal; Kieninger, Thomas; Dengel, Andreas

doi:10.1145/1815330.1815345

Cited by 88 publications

(89 citation statements)

References 20 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…They can be used either to compute the loss and backpropagate through the function, or to predict the classes and form the resultant adjacency matrices. [37] defined a set of metrics for detailed evaluation of the results of table parsing and detection. They defined criteria for correct and partial detection and defined a heuristic for labeling elements as under-segmented, oversegmented and missed.…”

Section: ) Classificationmentioning

confidence: 99%

Rethinking Table Recognition using Graph Neural Networks

Qasim

Mahmood

2019

2019 International Conference on Document Analysis and Recognition (ICDAR)

Self Cite

119

View full text Add to dashboard Cite

Document structure analysis, such as zone segmentation and table recognition, is a complex problem in document processing and is an active area of research. The recent success of deep learning in solving various computer vision and machine learning problems has not been reflected in document structure analysis since conventional neural networks are not well suited to the input structure of the problem. In this paper, we propose an architecture based on graph networks as a better alternative to standard neural networks for table recognition. We argue that graph networks are a more natural choice for these problems, and explore two gradient-based graph neural networks. Our proposed architecture combines the benefits of convolutional neural networks for visual feature extraction and graph networks for dealing with the problem structure. We empirically demonstrate that our method outperforms the baseline by a significant margin. In addition, we identify the lack of large scale datasets as a major hindrance for deep learning research for structure analysis and present a new large scale synthetic dataset for the problem of table recognition. Finally, we open-source our implementation of dataset generation and the training framework of our graph networks to promote reproducible research in this direction 1 .

show abstract

Section: ) Classificationmentioning

confidence: 99%

Rethinking Table Recognition using Graph Neural Networks

Qasim

Mahmood

2019

2019 International Conference on Document Analysis and Recognition (ICDAR)

Self Cite

119

View full text Add to dashboard Cite

show abstract

“…Since our work is focused on table structure extraction rather than table detection, we cropped the tables from the images using UNLV ground truth files, resulting in 557 tables. We then evaluated our model's outputs against the ground truths provided in [21]. We did not use any image from the UNLV dataset in the training and validation process of our model and thus all the images were unseen by the model.…”

Section: F False Positive Detectionsmentioning

confidence: 99%

Table Structure Extraction with Bi-Directional Gated Recurrent Unit Networks

Khan

Khalid

Shahzad

et al. 2019

2019 International Conference on Document Analysis and Recognition (ICDAR)

Self Cite

View full text Add to dashboard Cite

Tables present summarized and structured information to the reader, which makes table's structure extraction an important part of document understanding applications. However, table structure identification is a hard problem not only because of the large variation in the table layouts and styles, but also owing to the variations in the page layouts and the noise contamination levels. A lot of research has been done to identify table structure, most of which is based on applying heuristics with the aid of optical character recognition (OCR) to hand pick layout features of the tables. These methods fail to generalize well because of the variations in the table layouts and the errors generated by OCR. In this paper, we have proposed a robust deep learning based approach to extract rows and columns from a detected table in document images with a high precision. In the proposed solution, the table images are first pre-processed and then fed to a bi-directional Recurrent Neural Network with Gated Recurrent Units (GRU) followed by a fully-connected layer with softmax activation. The network scans the images from top-to-bottom as well as left-to-right and classifies each input as either a row-separator or a column-separator. We have benchmarked our system on publicly available UNLV as well as ICDAR 2013 datasets on which it outperformed the state-of-theart table structure extraction systems by a significant margin.

show abstract

“…Especially skewing is not supported. Evaluation is performed using the precision/recall model as defined by [24] (as correct and missed detection) with an overlap threshold of TH=0.50 (we tried various values, without any impact on the comparison). For this evaluation, both CRF and ECN were trained with the full dataset1.…”

Section: ) Table Row Evaluationmentioning

confidence: 99%

Comparing Machine Learning Approaches for Table Recognition in Historical Register Books

Clinchant¹,

Dejean²,

Meunier³

et al. 2018

2018 13th IAPR International Workshop on Document Analysis Systems (DAS)

View full text Add to dashboard Cite

We present in this paper experiments on Table Recognition in hand-written registry books. We first explain how the problem of row and column detection is modelled, and then compare two Machine Learning approaches (Conditional Random Field and Graph Convolutional Network) for detecting these table elements. Evaluation was conducted on death records provided by the Archive of the Diocese of Passau. Both methods show similar results, a 89 F1 score, a quality which allows for Information Extraction. Software and dataset are open source/data.

show abstract

An open approach towards the benchmarking of table structure recognition systems

Cited by 88 publications

References 20 publications

Rethinking Table Recognition using Graph Neural Networks

Rethinking Table Recognition using Graph Neural Networks

Table Structure Extraction with Bi-Directional Gated Recurrent Unit Networks

Comparing Machine Learning Approaches for Table Recognition in Historical Register Books

Contact Info

Product

Resources

About