Table recognition refers to the process of automatically identifying and extracting tabular structures from unstructured data sources such as text documents, images, or scanned documents. The goal of table recognition is to accurately detect the presence of tables within the data and extract their contents, including rows, columns, headers, and cell values.
데이터셋
출처에 같은 태스크 이름으로 연결된 데이터셋입니다. 동명 태스크가 포함될 수 있으므로, 비교할 때는 평가 결과의 조건도 확인하세요.
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. The table images are extracted from the scientific publications included in the PubMed Central Open Access Subset (commercial use collection). Table regions are identified by matching the PDF format and the XML format of the articles in the PubMed Central Open Access Subset. More details are available in our paper "Image-based table recognition: data, model, and evaluation".
This dataset contains complex tables from the annual reports of S&P 500 companies with detailed table structure annotations to help table structure recognition and table data extraction. The dataset consists of 89,646 pages comprising 112,887 tables with cell structure annotated from IBM Research. This dataset contains cell structure labels generated through token matching between the PDF and HTML version of each article. Financial tables often have diverse styles when compared to ones in scientific and government documents, with fewer graphical lines and larger gaps within each table and more colour variations. These features are reflected in the dataset. Source: IBM Developer Image Source: DAX Dataset Preview
WTW (Wired Table in the Wild) is a large-scale dataset which includes well-annotated structure parsing of multiple style tables in several scenes like the photo, scanning files, web pages. WTW dataset has 10970 training samples and 3611 testing ones. The test images are divided into 7 challenging categories. Dataset for trains and test contain images and labels. The label is in XML format, which has cell bbox and the structure label, includes start row, end row, start col, end col, and table id. In addition, the test set also contains separate file descripts sub-classification information for each image.
The goal of PubTables-1M is to create 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. It contains: - 460,589 annotated document pages containing tables for table detection. - 947,642 fully annotated tables including text content and complete location (bounding box) information for table structure recognition and functional analysis. - Full bounding boxes in both image and PDF coordinates for all table rows, columns, and cells (including blank cells), as well as other annotated structures such as column headers and projected row headers. - Rendered images of all tables and pages. - Bounding boxes and text for all words appearing in each table and page image. - Additional cell properties not used in the current model training. Additionally, cells in the headers are canonicalized and we implement multiple quality control steps to ensure the annotations are as free of noise as possible. For more details, please see our paper.
We present TNCR, a new table dataset with varying image quality collected from free open source websites. TNCR dataset can be used for table detection in scanned document images and their classification into 5 different classes. TNCR contains 9428 high-quality labeled images. In this paper, we have implemented state-of-the-art deep learning-based methods for table detection to create several strong baselines. Cascade Mask R-CNN with ResNeXt-101-64x4d Backbone Network achieves the highest performance compared to other methods with a precision of 79.7%, recall of 89.8%, and f1 score of 84.4% on the TNCR dataset. We have made TNCR open source in the hope of encouraging more deep learning approaches to table detection, classification and structure recognition. Image source: https://github.com/abdoelsayed2016/TNCR_Dataset
WikiTableSet is a large publicly available image-based table recognition dataset in three languages built from Wikipedia. WikiTableSet contains nearly 4 million English table images, 590K Japanese table images, 640k French table images with corresponding HTML representation, and cell bounding boxes. We build a Wikipedia table extractor WTabHTML and use this to extract tables (in HTML code format) from the 2022-03-01 dump of Wikipedia. In this study, we select Wikipedia tables from three representative languages, i.e., English, Japanese, and French; however, the dataset could be extended to around 300 languages with 17M tables using our table extractor. Second, we normalize the HTML tables following the PubTabNet format (separating table headers and table data, removing CSS and style tags). Finally, we use Chrome and Selenium to render table images from table HTML codes. This dataset provides a standard benchmark for studying table recognition algorithms in different languages or even multilingual table recognition algorithms. You can click here for more details about this dataset.
The Construction Industry Steel Ordering Lists (CISOL) dataset comprises table-centric, real-world documents from the construction industry, annotated to facilitate the testing and training of deep learning models for table detection (TD) and table structure recognition (TSR). Access to the CISOL Leaderboard is provided via https://eval.ai/web/challenges/challenge-page/2257