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

Schreiber, Sebastian; Agne, Stefan; Wolf, Ivo; Dengel, Andreas; Ahmed, Sheraz

doi:10.1109/icdar.2017.192

Cited by 296 publications

(284 citation statements)

References 25 publications

Supporting

Mentioning

282

Contrasting

Unclassified

Order By: Relevance

“…Output of the model is shown in Figure 2. Additionally, Experiment 3 was carried out to compare TableNet with the closest deep-learning based solution, DeepDSert [8]. In DeepDSert, separate models are made for Table detection As done in DeepDSert, we also randomly chose 34 images for testing and used the rest of the data images for fine-tuning our model.…”

Section: Experiments and Resultsmentioning

confidence: 99%

See 1 more Smart Citation

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

Paliwal

Vishwanath²,

Rahul³

et al. 2019

2019 International Conference on Document Analysis and Recognition (ICDAR)

142

View full text Add to dashboard Cite

With the widespread use of mobile phones and scanners to photograph and upload documents, the need for extracting the information trapped in unstructured document images such as retail receipts, insurance claim forms and financial invoices is becoming more acute. A major hurdle to this objective is that these images often contain information in the form of tables and extracting data from tabular sub-images presents a unique set of challenges. This includes accurate detection of the tabular region within an image, and subsequently detecting and extracting information from the rows and columns of the detected table. While some progress has been made in table detection, extracting the table contents is still a challenge since this involves more fine grained table structure(rows & columns) recognition. Prior approaches have attempted to solve the table detection and structure recognition problems independently using two separate models. In this paper, we propose TableNet: a novel end-toend deep learning model for both table detection and structure recognition. The model exploits the interdependence between the twin tasks of table detection and table structure recognition to segment out the table and column regions. This is followed by semantic rule-based row extraction from the identified tabular sub-regions. The proposed model and extraction approach was evaluated on the publicly available ICDAR 2013 and Marmot Table datasets obtaining state of the art results. Additionally, we demonstrate that feeding additional semantic features further improves model performance and that the model exhibits transfer learning across datasets. Another contribution of this paper is to provide additional table structure annotations for the Marmot data, which currently only has annotations for table detection.

show abstract

Section: Experiments and Resultsmentioning

confidence: 99%

“…Recently, DeepDeSRT [8] was proposed which uses deep learning for both table detection and table structure recognition, i.e. identifying rows, columns, and cell positions in the detected tables.…”

Section: Related Workmentioning

confidence: 99%

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

Paliwal

Vishwanath²,

Rahul³

et al. 2019

2019 International Conference on Document Analysis and Recognition (ICDAR)

142

View full text Add to dashboard Cite

show abstract

“…Reference [16] uses Fast R-CNN architecture with a novel idea of Euclidean distance feature to detect tables (which was compared to Tesseract). Reference [17] also uses (pretrained) Fast R-CNN and FCN semantic segmentation model for table extraction problem. In [8] work has been done on detection problem bottom up using the Hough transform, and extraction was solved with Markov networks and features from the cell positions.…”

Section: Previous Workmentioning

confidence: 99%

Table Understanding in Structured Documents

Holeček

Hoskovec²,

Baudis³

et al. 2019

2019 International Conference on Document Analysis and Recognition Workshops (ICDARW)

View full text Add to dashboard Cite

Table detection and extraction has been studied in the context of documents like reports, where tables are clearly outlined and stand out from the document structure visually. We study this topic in a rather more challenging domain of layoutheavy business documents, particularly invoices. Invoices present the novel challenges of tables being often without outlines -either in the form of borders or surrounding text flow -with ragged columns and widely varying data content. We will also show, that we can extract specific information from structurally different tables or table-like structures with one model. We present a comprehensive representation of a page using graph over word boxes, positional embeddings, trainable textual features and rephrase the table detection as a text box labeling problem. We will work on our newly presented dataset of pro forma invoices, invoices and debit note documents using this representation and propose multiple baselines to solve this labeling problem. We then propose a novel neural network model that achieves strong, practical results on the presented dataset and analyze the model performance and effects of graph convolutions and self-attention in detail.

show abstract

“…To encompass the various dimensionality of the structure of the tables, the latest Deep Learning techniques shall be used on the spatial distribution of words between rows and columns of boundaries. We can enhance the novel deep learning-based approach for table structure detection, DeepDeSRT [14], as proposed by Sebastian Schreiber et al The PDF documents can be converted to image format. In addition to the preprocessing technique(s) mentioned by Sebastian et al, further processing can be done by creating Local Binary Patterns (LBP) [15], [16] of the image.…”

Section: A Identifying Rows and Columns Using Deep Learningmentioning

confidence: 99%

Discovery of Structured Data Using Unsupervised Spatial Clustering and Human Supervision

Rachapudi¹,

Ganesh²,

Sekar³

et al. 2019

IJMLC

View full text Add to dashboard Cite

Commercial data has been preserved digitally in portable document format (PDF) for its ease of encapsulating multiple data formats. In this digitization era, there comes a need to capture and store this data in structured format to facilitate its access for automated b2b services and business intelligence. In this paper, we propose a framework that automates discovery and extraction of tabular data incorporating both artificial and human intelligence. The framework involves clustering and heuristics to group cartesian location of text and spaces in a page to determine a table. The discovered table is then validated by the user using a user-interface designed to moderate the determined boundaries and fed back to the layout knowledge repository. The table data obtained is extracted as JSON key-value pairs which can then be loaded into any database. The framework thus provides enhanced accuracy and continuous human assisted learning for the automated document digitization process. The knowledge repository is further used to train the machine to generate document templates to be used for processing unseen documents. However, this paper concentrates on the discovery of structured data alone.

show abstract

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

Cited by 296 publications

References 25 publications

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

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

Table Understanding in Structured Documents

Discovery of Structured Data Using Unsupervised Spatial Clustering and Human Supervision

Contact Info

Product

Resources

About