Home

Awesome

Awesome-Table-Recognition

A curated list of resources dedicated to table recognition

1. Papers

Conf.DateTitleHighlightcode
IJCAI2023Divide Rows and Conquer Cells: Towards Structure Recognition for Large TablesSequenceNo
AAAI2022LORE: Logical Location Regression Network for Table Structure RecognitionDetection*CODE<br>
ACM-MM2022TSRFormer: Table Structure Recognition with TransformersDetectionNo
CVPR2022TableFormer: Table Structure Understanding with Transformers.SequenceNo
CVPR2022Neural Collaborative Graph Machines for Table Structure RecognitionGNNNo
CVPR2022PubTables-1M: Towards comprehensive table extraction from unstructured documentsDataset*CODE<br>
arXiv2021/5/23Multi-Type-TD-TSR -- Extracting Tables from Document Images using a Multi-stage Pipeline for Table Detection and Table Structure Recognition: from OCR to Structured Table RepresentationsOthers*CODE<br>
ACM-MM2021Show, Read and Reason: Table Structure Recognition with Flexible Context AggregatorGNNNo
ICCV2021Parsing Table Structures in the WildDetectionNo
ICCV2021TGRNet: A Table Graph Reconstruction Network for Table Structure RecognitionGNN*CODE<br>
ICDAR Competition2021ICDAR 2021 Competition on Scientific Literature ParsingDataset*CODE<br>
ICDAR Competition2021PingAn-VCGroup's Solution for ICDAR 2021 Competition on Scientific Literature Parsing Task B: Table Recognition to HTMLSequence*CODE<br>
ICDAR Competition2021LGPMA: Complicated Table Structure Recognition with Local and Global Pyramid Mask AlignmentOthers*CODE<br>
WACV2021Global table extractor (gte): A framework for joint table identification and cell structure recognition using visual contextOthersNo
CVPR Workshop2020CascadeTabNet: An approach for end to end table detection and structure recognition from image-based documentsOthers*CODE<br>
ECCV2020Image-based table recognition: data, model, and evaluationDataset*CODE<br>
ECCV2020Table structure recognition using top-down and bottom-up cuesOthers*CODE<br>
LREC2020TableBank: A Benchmark Dataset for Table Detection and RecognitionDataset*CODE<br>
arXiv2019/8/28Complicated table structure recognitionOthers*CODE<br>
ICDAR2019Rethinking Table Recognition using Graph Neural NetworksGNN*CODE<br>
ICDAR2019Tablenet: Deep learning model for end-to-end table detection and tabular data extraction from scanned document imagesOthersNo
ICDAR2019Res2tim: Reconstruct syntactic structures from table images.Others*CODE<br>
ICDAR2017Deepdesrt: Deep learning for detection and structure recognition of tables in document imagesOthersNo

2. Datasets

2.1 Introduction

DatasetDescriptiondataset link
TableBankEnglish TableBank is a new image-based table detection and recognition dataset built with novel weak supervision from Word and Latex documents on the internet, contains 417K high-quality labeled tables.It only contain cell Topology groudtruthTableBank
SciTSR*English SciTSR is a large-scale table structure recognition dataset, which contains 15,000 tables in PDF format and their corresponding structure labels obtained from LaTeX source files.It contain cell Topology, cell content groudtruthSciTSR
PubTabNetEnglish PubTabNet is a large dataset for image-based table recognition, containing 568k+ images of tabular data annotated with the corresponding HTML representation of the tables.It contain cell Topology, cell content and non-blank cell location groudtruthPubTabNet
FinTabNetEnglish This dataset contains complex tables from the annual reports of S&P 500 companies with detailed table structure annotations to help train and test structure recognition.FinTabNet
PubTables-1MEnglish A large, detailed, high-quality dataset for training and evaluating a wide variety of models for the tasks of table detection, table structure recognition, and functional analysis.PubTables-1M
WTWEnglish and Chinese WTW-Dataset is the first wild table dataset for table detection and table structure recongnition tasks, which is constructed from photoing, scanning and web pages, covers 7 challenging cases like: (1)Inclined tables, (2) Curved tables, (3) Occluded tables or blurredtables (4) Extreme aspect ratio tables (5) Overlaid tables, (6) Multi-color tables and (7) Irregular tables in table structure recognition.It contain cell Topology, all cell location groudtruthWTW
TNCREnglish a new table dataset with varying image quality collected from open access websites.TNCR contains 9428 labeled tables with approximately 6621 images.their classification into 5 different classes(Full Lined,Merged Cells,No lines,Partial Lined,Partial Lined Merged Cells).TNCR
TAL_OCR_TABLEChinese TAL_OCR_TABLE dataset come from TAL Form Recognition Technology Challenge.The data of comes from the real homework of students in the education scene and the scene of the test paper. It contain 16k train image and 4k test imageIt contain cell Topology, cell content and all cell location groudtruthTAL_OCR_TABLE

2.2 Comparison of datasets for table structure recognition.

DatasetCell TopologyCell contentCell LocationTable Location
TableBank✓✕✕✓
SciTSR✓✓✕✓
PubTabNet✓✓✓<sup>†✓
FinTabNet✓✓✓<sup>†✓
PubTables-1M✓✓✓✓
WTW✓✕✓✓
TNCR✕✕✕✓
TAL_OCR_TABLE✓✓✓✓

<sup>†</sup> For these datasets, cell bounding boxes are given for non-blank cells only and exclude any non-text portion of a cell.

3. Other technical solutions

PRCV2021 Table Recognition Technology Challenge

ICDAR 2021 Competition on Scientfic Literature Parsing TaskB: Table Recognition to HTML