Visual Understanding of Complex Table Structures from Document Images

Raja, Sachin; Mondal, Ajoy; Jawahar, C. V.

doi:10.1109/wacv51458.2022.00260

Cited by 15 publications

(8 citation statements)

References 34 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Early works (Schreiber et al 2017;Siddiqui et (Qasim, Mahmood, and Shafait 2019;Raja, Mondal, andJawahar 2020, 2022;Liu et al 2021Liu et al , 2022. However, there is still a gap between the set of relation triplets and the global table structure.…”

Section: Related Workmentioning

confidence: 99%

“…Details of datasets are available in section 2 of the supplementary. It should be noted that ICDAR-2013 provides no training data, so we extend it to the partial version for cross validation following previous works (Raja, Mondal, and Jawahar 2020;Liu et al 2022Liu et al , 2021. And when training LORE on the PubTabNet, we randomly choose 20,000 images from its training set for efficiency.…”

Section: Datasetsmentioning

confidence: 99%

See 1 more Smart Citation

LORE: Logical Location Regression Network for Table Structure Recognition

Xing

Gao

Long

et al. 2023

AAAI

View full text Add to dashboard Cite

Table structure recognition (TSR) aims at extracting tables in images into machine-understandable formats. Recent methods solve this problem by predicting the adjacency relations of detected cell boxes, or learning to generate the corresponding markup sequences from the table images. However, they either count on additional heuristic rules to recover the table structures, or require a huge amount of training data and time-consuming sequential decoders. In this paper, we propose an alternative paradigm. We model TSR as a logical location regression problem and propose a new TSR framework called LORE, standing for LOgical location REgression network, which for the first time combines logical location regression together with spatial location regression of table cells. Our proposed LORE is conceptually simpler, easier to train and more accurate than previous TSR models of other paradigms. Experiments on standard benchmarks demonstrate that LORE consistently outperforms prior arts. Code is available at https:// github.com/AlibabaResearch/AdvancedLiterateMachinery/tree/main/DocumentUnderstanding/LORE-TSR.

show abstract

Section: Related Workmentioning

confidence: 99%

Section: Datasetsmentioning

confidence: 99%

LORE: Logical Location Regression Network for Table Structure Recognition

Xing

Gao

Long

et al. 2023

AAAI

View full text Add to dashboard Cite

show abstract

“…S Raja [121] suggests a novel object-detection-based deep model that is tailored for quick optimization and captures the natural alignments of cells inside tables. Dense table recognition may still be problematic even with precise cell detection because multi-row/column spanning cells make it difficult to capture long-range row/column relationships.…”

Section: Table Structure Recognition Modelsmentioning

confidence: 99%

“…Uses OCR to read words from images Not language agnostic S Raja [121] object detection Better detection of empty cells Fails for very sparse tables where most of the cells are empty present. Second, based on the first stage proposals, it predicts the object's class, refines the bounding box, and creates a mask at the pixel level of the object.…”

Section: Recognizing Complex Table Structures Having Multi-span Rows/...mentioning

confidence: 99%

Deep learning for table detection and structure recognition: A survey

Kasem¹,

Abdallah²,

Berendeyev³

et al. 2022

Preprint

View full text Add to dashboard Cite

Tables are everywhere, from scientific journals, papers, websites, and newspapers all the way to items we buy at the supermarket. Detecting them is thus of utmost importance to automatically understanding the content of a document. The performance of table detection has substantially increased thanks to the rapid development of deep learning networks. The goals of this survey are to provide a profound comprehension of the major developments in the field of Table Detection, offer insight into the different methodologies, and provide a systematic taxonomy of the different approaches. Furthermore, we provide an analysis of both classic and new applications in the field. Lastly, the datasets and source code of the existing models are organized to provide the reader with a compass on this vast literature. Finally, we go over the architecture of utilizing various object detection and table structure recognition methods to create an effective and efficient system, as well as a set of development trends to keep up with state-of-the-art algorithms and future research. We have also set up a public GitHub repository where we will be updating the most recent publications, open data, and source code. The GitHub repository is available at https://github.com/abdoelsayed2016/ table-detection-structure-recognition.

show abstract

“…However, this process is challenging, notably due to the deteriorating condition of historical documents, handwriting variation, and visual artifacts such as fading and decaying spots. The process is further complicated by the need for precise, high-quality annotations, tables with densely packed cells, uneven layouts, and non-linear handwriting that crosses into adjacent cell regions [8,10]. This paper emphasizes the importance of TSR in ensuring the accuracy and reliability of digitized content, informing robust climate models, understanding historical climate dynamics, and highlighting the urgent need for innovative approaches to overcome the significant challenges of the digitization process.…”

Section: Introductionmentioning

confidence: 99%

Data Rescue for Historical Document Tables Using Semi-Supervised Learning

Singh,

Middleton

2024

Preprint

View full text Add to dashboard Cite

This study uses a novel semi-supervised learning framework to explore Tabular Structure Recognition (TSR) for digitizing historical documents, specifically employing the CascadeTabNet model. TSR is crucial for transforming archival tabular data into digital formats, enhancing accessibility and analysis across various research fields. Challenges like physical degradation, inconsistent lighting, and non-standard handwriting hinder the generation of high-quality annotations of historical documents needed for effective model training. To address these issues, this research explores two research questions: (i) Can a semi-supervised training approach reduce the need for expensive data annotations? and (ii) Does semi-supervised training improve model robustness? We applied our methodology across three datasets: the GloSAT and ICDAR-2019 datasets based on historical documents, and the predominantly modern documents PubTabNet dataset. Our results indicate that semi-supervised learning substantially increases TSR accuracy and decreases dependency on extensive labelled datasets, providing a robust solution for large-scale digitization initiatives and contributing to the preservation and improved accessibility of historical data. All code from this paper is freely available on GitHub.

show abstract

Visual Understanding of Complex Table Structures from Document Images

Cited by 15 publications

References 34 publications

LORE: Logical Location Regression Network for Table Structure Recognition

LORE: Logical Location Regression Network for Table Structure Recognition

Deep learning for table detection and structure recognition: A survey

Data Rescue for Historical Document Tables Using Semi-Supervised Learning

Contact Info

Product

Resources

About