Find research datasets worth reusing
Search datasets from major research repositories and use ShareScore to quickly assess how well each record supports discovery, access, and reuse.
24
datasets available to search
ShareScore release 0.9.0
Dataset results
24 results for “ICDAR”
ICDAR'15 SMARTPHONE DOCUMENT CAPTURE AND OCR COMPETITION (SmartDoc) - Challenge 1 (original version)
<p><strong>CHALLENGE 1: SMARTPHONE DOCUMENT CAPTURE COMPETITION</strong></p> <p><strong>Smartphones are replacing personal scanners.</strong> They are portable, connected, powerful and affordable. They are on their way to become the new entry point in business processing applications like document archival, ID scanning, check digitization, just to name a few. In order keep our workflows streamlined, <strong>we need to make those new capture device as reliable as batch scanners</strong>.</p> <p>We believe an efficient capture process should be able to:</p> <ol> <li><em>detect and segment</em> the relevant document object during the preview phase;</li> <li><em>assess the quality</em> of the capture conditions and help the user improve them;</li> <li>optionally, <em>trigger the capture</em> at the perfect moment;</li> <li>and <em>produce a high-quality, controlled output</em> based on the high resolution captured image.</li> </ol> <p>This competition is focused on the first step of this process:<strong> </strong><strong>efficiently detect and segment document regions</strong>, as illustrated by following video showing the ideal output for the preview phase of some acquisition session: <a href="https://youtu.be/WNsI0R_rpO0">Click here to watch the video.</a> This video shows the ideal document object detection ‎(ie the ground truth, as a red frame)‎.</p> <p>For this challenge, the <strong>input</strong> consists in a set of <strong>videoclips containing a document</strong> from a predefined set, and the <strong>output</strong> should be an <strong>xml file containing the quadrilateral coordinates</strong> in which we can find the document per each frame of the video. Click <a href="https://sites.google.com/site/icdar15smartdoc/challenge-1/challenge1dataset">here</a> for detailed information about the dataset. </p> <p> </p> <p><strong>Licence</strong> for the dataset of challenge 1 (page outline detection in preview frames) :</p> <p>This work is licensed under a <strong>Creative Commons Attribution 4.0 International License</strong> <<a href="https://www.google.com/url?q=http://creativecommons.org/licenses/by/4.0/&sa=D&ust=1524734857667000&usg=AFQjCNEt4YXnUv2nXCFwkeOuBDqxDpvknQ">http://creativecommons.org/licenses/by/4.0/</a>>. Author attribution should be given by citing the following conference paper: Jean-Christophe Burie, Joseph Chazalon, Mickaël Coustaty, Sébastien Eskenazi, Muhammad Muzzamil Luqman, Maroua Mehri, Nibal Nayef, Jean-Marc OGIER, Sophea Prum and Marçal Rusinol: “ICDAR2015 Competition on Smartphone Document Capture and OCR (SmartDoc)”, In 13th International Conference on Document Analysis and Recognition (ICDAR), 2015.</p> <p><strong>If you use this dataset, please send us a short email at <icdar.smartdoc (at) gmail.com> to tell us why it was useful to you, and whether you have results or publications we can reference on our website. Thank you!</strong></p>
ScriptNet: ICDAR 2017 Competition on Baseline Detection in Archival Documents (cBAD)
<p>This dataset contains the training and test set for the ICDAR 2017 Competition on Baseline Detection in Archival Documents (cBAD).</p> <p>A newly created freely available real world dataset consisting of 2035 annotated document page images that are collected from 9 different archives and form the basis of cBAD. Two competition tracks test different characteristics of the methods submitted. Track A [Simple Documents] is published with annotated text regions and tests therefore a method's quality of text line segmentation. The more challenging Track B [Complex Documents] provides only the page area. Hence, baseline detection algorithms need to correctly locate text lines in the presence of marginalia, tables, and noise.</p> <p>The dataset comprises images with additional PAGE XMLs. The PAGE XMLs contain text regions and baseline annotations.</p> <p>Competition Website: https://scriptnet.iit.demokritos.gr/competitions/5/</p> <p>Version 3 is the version of the cBad competition</p> <p>Version 4 contains also the page region and in case of a double-page the page split as separator.</p>
The ICDAR 2003 Informal Competition for the Recognition of On-line Words: The Unipen-ICROW-03 benchmark set - Version 0.0
<p>Proposal for an informal benchmark on word recognition. See for the related ImUnipen collection<br> of word images from on-line vectorial handwriting data: https://zenodo.org/record/1195059</p> <p>At the time (ICDAR 2003) there was not a lot of interest so the project was not pursued.</p> <p>Lambert Schomaker - February 2023</p> <p>_______________________________________________________________________________</p> <p>The ICDAR 2003 Informal Competition for the Recognition of On-line Words:<br> The Unipen-ICROW-03 benchmark set <br> Version 0.0</p> <p>Lambert Schomaker / International Unipen Foundation</p> <p>The ICROW suite of test files for the recognition of isolated on-line<br> free-style (handprint, mixed and cursive) words has been<br> composed. Different tablets, nationalities and languages<br> are involved. Only the ASCII set is used within word labels.</p> <p>The set contains:</p> <p> 13119 written words<br> 884 unique lexical word entries<br> 72 writers </p> <p>Language: Dutch, English, Italian.<br> Nationalities: Dutch, Irish, Italian, + mixed</p> <p>The benchmark test is a good estimator for <br> "walk-up" recognition performance.</p> <p>[Note: some of the writers (NIC-Pc95*.dat set) are present in the<br> UNIPEN R01/V07 distribution, but the actual words are unseen <br> outside of the Int. Unipen Foundation.]</p> <p>Please note the Copyright notice in the <br> accompanying file 'Copyright'</p> <p>Wed Jul 16 21:20:10 CEST 2003</p> <p>Lambert Schomaker</p> <p>---------------------------------------------------------------------------</p> <p>Instructions for the ICDAR 2003 informal competition for<br> the recognition of on-line words.</p> <p>1 - unpack the .tgz file<br> 2 - use the UNIPEN files as input for your recognizer.<br> 3 - report, for each writer, a file <writer-id>.res</p> <p> Example: do-my-recognizer < NIC-Hi93b-marc.dat > NIC-Hi93b-marc.res</p> <p>Format of the .res file.</p> <p>No XML for this moment: simplicity does it.</p> <p>We assume that the recognizer is able to produce a top-10 list<br> of likely words, sorted from most likely to least likely.<br> The output for each word is on a single line. The correct<br> target word is in the first column.</p> <p><targetword 1> <best word hyp.> <2nd-best word hyp.> ... <10th-best word hyp><br> <targetword 2> <best word hyp.> <2nd-best word hyp.> ... <10th-best word hyp></p> <p>Example with two words:</p> <p>summertime slumbertime slipknot summertime somatome spumante simulative semitone schoolmate sermonette semimature<br> Aberdeen Adamson Aberdeen Addison Armageddon Abyssinian Araban Albanian Alabamian Abraham Adelaide</p> <p><br> 4 - pack the *.res files in a .tgz or .zip file and send them<br> to schomaker@ai.rug.nl<br> All *.dat files need to be processed.</p> <p>LS.<br> </p> <p> </p>
ICDAR 2019 Competition on Baseline Detection (cBAD)
<p>This dataset contains the training, evaluation, and test set for the ICDAR 2019 Competition on Baseline Detection (cBAD).</p> <p>A newly created freely available real world dataset consisting of 3021 annotated document page images that are collected from seven European archives and form the basis of cBAD. The baselines in all images were manually annotated. The training and the evaluation sets contain PAGE XMLs with annotated text regions and baselines.</p> <p>Competition Website: https://scriptnet.iit.demokritos.gr/competitions/11/</p>
ICDAR 2015 Competition HTRtS: Handwritten Text Recognition on the tranScriptorium Dataset
<p>This dataset comprises the dataset used for the ICDAR 2015 Competition on Handwritten Text Recognition on the tranScriptorium Dataset. The handwritten images for this contest were drawn from the English “Bentham collection” dataset used in the TRAN SCRIPTORIUM project. The selected data has been written by several hands and entails significant variabilities and difficulties regarding the quality of text images, writing styles and crossed-out text. This contest is clearly more difficult than the the first edition both for training and for testing. A portion of the training dataset and the full test dataset were provided in the form of carefully segmented line images, along with the corresponding transcripts. Another portion of the training dataset was provided as raw images and their corresponding transcripts at region level.<br> </p> <p>ICDAR 2015 competition HTRtS: handwritten text recognition on the tranScriptorium dataset<br> JA Sánchez, AH Toselli, V Romero, E Vidal. In International Conference on Document Analysis and Recognition (ICDAR), pp. 1166-1170, 2015.</p>
Dataset for ICDAR 21 paper "Vectorization of Historical Maps Using Deep Edge Filtering and Closed Shape Extraction"
<p>This is the dataset of the ICDAR 2021 conference paper "Vectorization of Historical Maps Using Deep Edge Filtering and Closed Shape Extraction".</p>
IGN Train and Validation Data for ICDAR'24 MapText Competition
<p>Data set of 2Kx2K image tiles cropped from Napoleonic Cadastre maps of the <a href="https://archives.valdemarne.fr/recherches/archives-en-ligne/cadastre-napoleonien">Val de Marne Archive</a> for the <a href="https://rrc.cvc.uab.es/?ch=28">ICDAR'24 Competition on Historical Map Text Detection, Recognition, and Linking</a>.</p> <p>Annotations and images follow the format described at the competition website and can be evaluated using the official <a href="https://github.com/icdar-maptext/evaluation">evaluation repository</a> script.</p> <table> <tbody> <tr> <td> </td> <td><strong>Train</strong></td> <td><strong>Validation</strong></td> </tr> <tr> <td>Annotations</td> <td><code>ign_train.json</code></td> <td><code>ign_val.json</code></td> </tr> <tr> <td>Images</td> <td><code>train.zip</code></td> <td><code>val.zip</code></td> </tr> <tr> <td>Files</td> <td><code>ign/train/*.jpg</code></td> <td><code>ign/val/*.jpg</code></td> </tr> <tr> <td>Tiles</td> <td>80</td> <td>15</td> </tr> <tr> <td>Map Sheets</td> <td>37</td> <td>9</td> </tr> <tr> <td>Words</td> <td>8,096</td> <td>1,801</td> </tr> <tr> <td>Label Groups</td> <td>7,449</td> <td>1,661</td> </tr> <tr> <td>Illegible Words</td> <td>563</td> <td>217</td> </tr> <tr> <td>Truncated Words</td> <td>371</td> <td>91</td> </tr> <tr> <td>Valid Words</td> <td>7,533</td> <td>1,584</td> </tr> </tbody> </table> <p><em>Original images available at <a href="https://archives.valdemarne.fr/recherches/archives-en-ligne/cadastre-napoleonien" target="_blank" rel="noopener">https://archives.valdemarne.fr/recherches/archives-en-ligne/cadastre-napoleonien</a> as of 1 Feb. 2024.</em></p>
ICDAR 2019 Competition on Image Retrieval for Historical Handwritten Documents [HisIR19] Dataset
<p>This dataset contains the training and test set used in the ICDAR 2019 Competition on Image Retrieval for Historical Handwritten Documents.</p> <p>This competition investigates the performance of large-scale retrieval of historical document images based on<br> writing style. Based on large image data sets provided by cultural heritage institutions and digital libraries, providing<br> a total of 20 000 document images representing about 10 000 writers, divided in three types: writers of (i) manuscript books, (ii) letters, (iii) charters and legal documents. We focus on the task of automatic image retrieval to simulate common scenarios of humanities research, such as writer retrieval.</p> <p>The training data set encompasses images from (i) Letters A, where each writer contributed one or three images; (ii) Manuscripts, where each writer was represented by five consecutive images from a single book.<br> In total, it contains 300 writers contributing one page, 100 writers contributing three pages, and 120 writers contributing five pages resulting in 1200 images of 520 writers.</p> <p>The test data set contains 20 000 images: About 7 500 pages stem from isolated documents (partially anonymous writers, contributing one page each), and about 12 500 pages are from writers that contributed three or five pages.</p> <p> </p> <p>If you use this dataset, please cite:</p> <p>V. Christlein, A. Nicolaou, M. Seuret, D. Stutzmann, A. Maier: "ICDAR 2019 Competition on Image Retrieval for Historical Handwritten Documents", in 15th International Conference on Document Analysis and Recognition, 2019, Sydney, Australia</p> <p> </p>
Dataset of ICDAR 2019 Competition on Post-OCR Text Correction
<p><strong>Corpus for the ICDAR2019 Competition on Post-OCR Text Correction (October 2019)</strong><br> Christophe Rigaud, Antoine Doucet, Mickael Coustaty, Jean-Philippe Moreux<br> <a href="http://l3i.univ-larochelle.fr/ICDAR2019PostOCR">http://l3i.univ-larochelle.fr/ICDAR2019PostOCR</a><br> -------------------------------------------------------------------------------</p> <p>These are the supplementary materials for the ICDAR 2019 paper <em><a href="https://zenodo.org/record/3459116">ICDAR 2019 Competition on Post-OCR Text Correction</a></em></p> <p>Please use the following citation:</p> <pre><code>@inproceedings{rigaud2019pocr,</code> <code> </code><code>title="ICDAR 2019 Competition on Post-OCR Text Correction",</code> <code> </code><code>author={Rigaud, Christophe and Doucet, Antoine and Coustaty, Mickael and Moreux, Jean-Philippe},</code> <code> </code><code>year={2019},</code> <code> </code><code>booktitle={Proceedings of the 15th International Conference on Document Analysis and Recognition (2019)}</code> <code> </code><code>}</code></pre> <p> </p> <p><strong>Description</strong><br> The corpus accounts for 22M OCRed characters along with the corresponding Gold Standard (GS). The documents come from different digital collections available, among others, at the National Library of France (BnF) and the British Library (BL). The corresponding GS comes both from BnF's internal projects and external initiatives such as Europeana Newspapers, IMPACT, Project Gutenberg, Perseus and Wikisource.</p> <p><strong>Repartition of the dataset</strong><br> - <em>ICDAR2019_Post_OCR_correction_training_18M.zip</em>: 80% of the full dataset, provided to train participants' methods.<br> - <em>ICDAR2019_Post_OCR_correction_evaluation_4M</em>: 20% of the full dataset used for the evaluation (with Gold Standard made publicly after the competition).<br> - <em>ICDAR2019_Post_OCR_correction_full_22M</em>: full dataset made publicly available after the competition.</p> <p><strong>Special case for Finnish language</strong><br> Material from the National Library of Finland (<em>Finnish dataset FI > FI1</em>) are not allowed to be re-shared on other website. Please follow these guidelines to get and format the data from the original website.</p> <p>1. Go to <a href="https://digi.kansalliskirjasto.fi/opendata/submit?set_language=en">https://digi.kansalliskirjasto.fi/opendata/submit?set_language=en</a>;<br> 2. Download <em>OCR Ground Truth Pages (Finnish Fraktur) [v1](4.8GB)</em> from <em>Digitalia (2015-17)</em> package;<br> 3. Convert the Excel file "<em>~/metadata/nlf_ocr_gt_tescomb5_2017.xlsx</em>" as Comma Separated Format (.csv) by using <em>save as</em> function in a spreadsheet software (e.g. Excel, Calc) and copy it into "<em>FI/FI1/HOWTO_get_data/input/</em>";<br> 4. Go to "<em>FI/FI1/HOWTO_get_data/</em>" and run "<em>script_1.py</em>" to generate the <em>full</em> "<em>FI1</em>" dataset in "<em>output/full/</em>";<br> 4. Run "<em>script_2.py</em>" to split the "<em>output/full/</em>" dataset into "<em>output/training/</em>" and "<em>output/evaluation/</em>" sub sets.<br> At the end of the process, you should have a "<em>training</em>", "<em>evaluation</em>" and "<em>full</em>" folder with 1579528, 380817 and 1960345 characters respectively.</p> <p><br> <strong>Licenses: free to use for non-commercial uses, according to sources in details</strong><br> - BG1: IMPACT - National Library of Bulgaria: CC BY NC ND<br> - CZ1: IMPACT - National Library of the Czech Republic: CC BY NC SA<br> - DE1: Front pages of Swiss newspaper NZZ: Creative Commons Attribution 4.0 International (<a href="https://zenodo.org/record/3333627">https://zenodo.org/record/3333627</a>)<br> - DE2: IMPACT - German National Library: CC BY NC ND<br> - DE3: GT4Hist-dta19 dataset: CC-BY-SA 4.0 (<a href="https://zenodo.org/record/1344132">https://zenodo.org/record/1344132</a>)<br> - DE4: GT4Hist - EarlyModernLatin: CC-BY-SA 4.0 (<a href="https://zenodo.org/record/1344132">https://zenodo.org/record/1344132</a>)<br> - DE5: GT4Hist - Kallimachos: CC-BY-SA 4.0 (<a href="https://zenodo.org/record/1344132">https://zenodo.org/record/1344132</a>)<br> - DE6: GT4Hist - RefCorpus-ENHG-Incunabula: CC-BY-SA 4.0 (<a href="https://zenodo.org/record/1344132">https://zenodo.org/record/1344132</a>)<br> - DE7: GT4Hist - RIDGES-Fraktur: CC-BY-SA 4.0 (<a href="https://zenodo.org/record/1344132">https://zenodo.org/record/1344132</a>)<br> - EN1: IMPACT - British Library: CC BY NC SA 3.0<br> - ES1: IMPACT - National Library of Spain: CC BY NC SA<br> - FI1: National Library of Finland: no re-sharing allowed, follow the above section to get the data. (<a href="https://digi.kansalliskirjasto.fi/opendata">https://digi.kansalliskirjasto.fi/opendata</a>)<br> - FR1: HIMANIS Project: CC0 (<a href="https://www.himanis.org/">https://www.himanis.org</a>)<br> - FR2: IMPACT - National Library of France: CC BY NC SA 3.0<br> - FR3: RECEIPT dataset: CC0 (<a href="http://findit.univ-lr.fr/">http://findit.univ-lr.fr</a>)<br> - NL1: IMPACT - National library of the Netherlands: CC BY<br> - PL1: IMPACT - National Library of Poland: CC BY<br> - SL1: IMPACT - Slovak National Library: CC BY NC</p> <p>Text post-processing such as cleaning and alignment have been applied on the resources mentioned above, so that the Gold Standard and the OCRs provided are not necessarily identical to the originals.</p> <p><br> <strong>Structure</strong><br> - **Content** [<em>./lang_type/sub_folder/#.txt</em>]<br> - "<em>[OCR_toInput] </em>" => Raw OCRed text to be de-noised.<br> - "<em>[OCR_aligned] </em>" => Aligned OCRed text.<br> - "<em>[ GS_aligned] </em>" => Aligned Gold Standard text.</p> <p>The aligned OCRed/GS texts are provided for training and test purposes. The alignment was made at the character level using "<em>@</em>" symbols. "<em>#</em>" symbols correspond to the absence of GS either related to alignment uncertainties or related to unreadable characters in the source document. For a better view of the alignment, make sure to disable the "word wrap" option in your text editor.</p> <p>The Error Rate and the quality of the alignment vary according to the nature and the state of degradation of the source documents. Periodicals (mostly historical newspapers) for example, due to their complex layout and their original fonts have been reported to be especially challenging. In addition, it should be mentioned that the quality of Gold Standard also varies as the dataset aggregates resources from different projects that have their own annotation procedure, and obviously contains some errors.</p> <p><br> <strong>ICDAR2019 competition</strong><br> Information related to the tasks, formats and the evaluation metrics are details on :<br> <a href="https://sites.google.com/view/icdar2019-postcorrectionocr/evaluation">https://sites.google.com/view/icdar2019-postcorrectionocr/evaluation</a></p> <p><br> <strong>References</strong><br> - IMPACT, European Commission's 7th Framework Program, grant agreement 215064<br> - Uwe Springmann, Christian Reul, Stefanie Dipper, Johannes Baiter (2018). Ground Truth for training OCR engines on historical documents in German Fraktur and Early Modern Latin.<br> - <a href="https://digi.nationallibrary.fi/">https://digi.nationallibrary.fi</a> , Wiipuri, 31.12.1904, Digital Collections of National Library of Finland<br> - EU Horizon 2020 research and innovation programme grant agreement No 770299</p> <p><br> <strong>Contact</strong><br> - christophe.rigaud(at)univ-lr.fr<br> - antoine.doucet(at)univ-lr.fr<br> - mickael.coustaty(at)univ-lr.fr<br> - jean-philippe.moreux(at)bnf.fr</p> <p>L3i - University of la Rochelle, <a href="http://l3i.univ-larochelle.fr/">http://l3i.univ-larochelle.fr</a><br> BnF - French National Library, <a href="http://www.bnf.fr/">http://www.bnf.fr</a></p>
ICDAR 2021 Historical Document Classification Test Dataset for Task 1 - Font Groups
<p>Test set for the font group classification task of the ICDAR 2021 Competition on Historical Document Classification competition. The tar.gz file contains the images and a ground truth CSV file. The csv file indicates for each image from which document and page it corresponds to, as well as whether augmentations have been applied.</p>
Dataset Versatile Layout Understanding via Conjugate Graph (ICDAR 2019)
<p>Contains the datasets with tables used in the following paper:</p> <p><a href="https://europe.naverlabs.com/research/publications/versatile-layout-understanding-via-conjugate-graph/">Versatile Layout Understanding via Conjugate Graph</a>. Animesh Prasad, Hervé Déjean and Jean-Luc Meunier, International Conference on Document Analysis and Recognition (ICDAR), Sydney, Australia, 20–25 September 2019.</p>
Sample Data for ICDAR'24 MapText Competition
<p>Data set of 353 image tiles (2Kx2K) cropped from 31 historical maps taken from nine (9) atlases for the <a href="https://rrc.cvc.uab.es/?ch=28">ICDAR'24 Competition on Historical Map Text Detection, Recognition, and Linking</a>.</p> <p>Annotations (<code>sample.json</code>) and images (<code>sample/*.png</code> within <code>sample.zip</code>) follow the format described at the competition website and can be evaluated using the official <a href="https://github.com/icdar-maptext/evaluation">evaluation repository</a> script.</p> <p>The data set contains 34,882 words in 31,200 label phrase groups; annotations result from automatically merging two distinct prior manual annotations (<a href="https://doi.org/11084/19349">Boateng Asante et al. 2017</a> and <a href="https://doi.org/11084/23295">Ray et al. 2018</a>). As a result, all text on the maps has been marked, but—unlike the official competition data—the annotations likely contain false negatives among the grouping (meaning, some words that should be linked into groups are not), and some words may divided into separate annotations, even individual characters.</p>
ICDAR 2015 Competition HTRtS: Handwritten Text Recognition on the tranScriptorium Dataset Rerelease
<p>A new release of the dataset used in the ICDAR 2015 HTR competition in which all Page XML files are based on the same 2013-07-15 schema. It only contains page level images, Page XML files for train and test (including the ground truth transcripts for the test and train batch 1) and plain text files for train batch 2 that have the page level ground truth transcripts. The original version of this dataset can be found at http://doi.org/10.5281/zenodo.248733<br> </p>
Rumsey Train and Validation Data for ICDAR'24 MapText Competition
<p>Data set of 2Kx2K image tiles cropped from maps of the <a href="https://davidrumsey.com">David Rumsey collection</a> for the <a href="https://rrc.cvc.uab.es/?ch=28">ICDAR'24 Competition on Historical Map Text Detection, Recognition, and Linking</a>.</p> <p>Annotations and images follow the format described at the competition website and can be evaluated using the official <a href="https://github.com/icdar-maptext/evaluation">evaluation repository</a> script.</p> <p><strong>Important</strong>: v1.1 fixes an image channel order error, superseding the prior version. v1.2 corrects group links among several annotations. v1.3 strips markup that inadvertently remained in some annotation transcriptions.</p> <table> <tbody> <tr> <td> </td> <td><strong>Train</strong></td> <td><strong>Validation</strong></td> </tr> <tr> <td>Annotations</td> <td><code>rumsey_train.json</code></td> <td><code>rumsey_val.json</code></td> </tr> <tr> <td>Images</td> <td><code>train.zip</code></td> <td><code>val.zip</code></td> </tr> <tr> <td>Files</td> <td><code>rumsey/train/*.png</code></td> <td><code>rumsey/val/*.png</code></td> </tr> <tr> <td>Tiles</td> <td>200</td> <td>40</td> </tr> <tr> <td>Map Sheets</td> <td>196</td> <td>40</td> </tr> <tr> <td>Words</td> <td>34,518</td> <td>5,544</td> </tr> <tr> <td>Label Groups</td> <td>21,205</td> <td>3,502</td> </tr> <tr> <td>Illegible Words</td> <td>1,870</td> <td>313</td> </tr> <tr> <td>Truncated Words</td> <td>3,582</td> <td>628</td> </tr> <tr> <td>Valid Words</td> <td>30,563</td> <td>4,860</td> </tr> </tbody> </table> <p> </p> <p><strong>Annotations</strong>: Copyright 2024 UMN Knowledge Computing Lab, <a href="https://creativecommons.org/licenses/by-nc-sa/4.0/">CC-BY-NC-SA 4.0 International.</a><br><strong>Images</strong>: David Rumsey Map Collection, David Rumsey Map Center, Stanford Libraries. <a href="https://creativecommons.org/licenses/by-nc-sa/3.0/">CC-BY-NC-SA 3.0 Unported</a>.</p>
ICDAR 2019 Competition on Table Detection and Recognition (cTDaR)
<p>The aim of this competition is to evaluate the performance of state of the art methods for table detection (TRACK A) and table recognition (TRACK B). For the first track, document images containing one or several tables are provided. For TRACK B two subtracks exist: the first subtrack (B.1) provides the table region. Thus, only the table structure recognition must be performed. The second subtrack (B.2) provides no a-priori information. This means, the table region and table structure detection has to be done. The Ground Truth is provided in a similar format as for the ICDAR 2013 competition (see [2]):</p> <p><?xml version="1.0" encoding="UTF-8"?></p> <p><<strong>document</strong> filename='filename.jpg'></p> <p> <<strong>table</strong> id='Table_1540517170416_3'></p> <p><strong> <Coords points="180,160 4354,160 4354,3287 180,3287"/></strong></p> <p> <<strong>cell</strong> id='TableCell_1540517477147_58' <strong>start-row</strong>='0' <strong>start-col</strong>='0' <strong>end-row</strong>='1' <strong>end-col</strong>='2'></p> <p> <<strong>Coords</strong> <strong>points</strong>="180,160 177,456 614,456 615,163"/></p> <p> </cell></p> <p> ...</p> <p> </table></p> <p> ...</p> <p></document></p> <p> </p> <p>The difference to Gobel et al. [2] is the Coords tag which defines a table/cell as a polygon specified by a list of coordinates. For B.1 the table and its coordinates is given together with the input image.</p> <p>Important Note:</p> <p>For the modern dataset, the convex hull of the content describes a cell region. For the historical dataset, it is requested that the output region of a cell is the cell boundary. This is necessary due to the characteristics of handwritten text, which is often overlapping with different cells.</p> <p>See also: http://sac.founderit.com/tasks.html</p> <p>The evaluation tool is available at github: https://github.com/cndplab-founder/ctdar_measurement_tool</p>
ICDAR 2021 Historical Document Classification Dataset for Task 2 - Dating
<p>Dataset for the localization classification task of the ICDAR 2021 Competition on Historical Document Classification competition. The tar.gz file contains the images and ground truth file (CSV format). The metadata csv file contains information, such as where the image comes from.</p>
ICDAR 2023 CROHME: Competition on Recognition of Handwritten Mathematical Expressions
<p>Here is the datasets collected for the Competitionon Recognition of Online Handwritten Mathematical Expressions in competition session of ICDAR 2023. <br> 3 tasks are proposed with different modalities, there are on-line, off-line and bi-modal. <br> For on-line task, we provide .inkml file (contain trace information, mathML and LaTeX string), and also symbol level label graph (SymLG) as ground truth. Except the new data and previous CROHME data, we also provide huge amount of artificial on-line data in the train set. <br> For off-line task, the .png images (scanned from paper or rendering from inkml) and symbol level label graph (SymLG) are provided. Except the new data and previous CROHME data, we use off-line images from OffHME to increase the size of train set. <br> For bi-modal task, both .inkml file and ,png images are provided as 2 channels input, and SymLG as ground truth. </p> <p>All the 3 tasks inherited the data collected from the previous 6 CROHME, and also the new collection 2023 in 3 sites, Nantes (France), Luleå (Sweden) and Tokyo (Japan).</p>
Figures with raw prediction samples from submissions to ICDAR'24 MapText Competition
<h2>Raw predictions sample from MapText submissions</h2> <p><em>Generated on June 6th, 2024</em></p> <p>Files structure: <code>{TASK}/{SUBSET}/{SELECTION}/{IMAGEID}.pdf</code> where</p> <ul> <li><code>{TASK}</code> is "task1", "task2", "task3" or "task4"</li> <li><code>{SUBSET}</code> is "rumsey", or "ign" (for French land registers dataset)</li> <li><code>{SELECTION}</code> is "random", "easy" or "hard"</li> <li><code>{IMAGEID}</code> is the image id in the dataset</li> </ul> <p>Example: <code>20-raw-predictions/task2/ign/hard/000016.pdf</code></p> <p>For each image, the PDF contains a comparison of the raw predictions of each submission, with the ground truth.</p> <p>For each task, the random images are the same, but the easy and hard images are different: their selection is based on the mean performance of all submissions regarding the main evaluation metric of the task.</p> <p>Sorting strategies for each task:</p> <ul> <li>Task 1: by detection quality (Panoptic Quality) for isolated words</li> <li>Task 2: by detection quality (Panoptic Quality) for word groups</li> <li>Task 3: by detection quality (Panoptic Quality) for isolated words while constraining matches between the ground truth and predictions to have exactly the same transcription</li> <li>Task 4: by character quality (Panoptic Character Quality) for word groups</li> </ul> <p>Task details and legend for each task:</p> <ul> <li><code>task1/</code>: detections for isolated words. Red means cropped or ignored regions, blue means prediction, green is ground truth.</li> <li><code>task2/</code>: detections for word groups. Groups are colored with different colors, and a link is drawn between successive group members. Ground truth is displayed separately.</li> <li><code>task3/</code>: detections and transcriptions of isolated words. Groups are colored with different colors, and the transcription for each isolated work is overlaid as black text. Ground truth is displayed separately.</li> <li><code>task4/</code>: detection and transcription of word groups. Same visualization as task3, but different sorting for easy and hard images.</li> </ul> <p><br>An additional directory <code>extrafigs/</code> contains a couple of figures with a different layout for inclusion in the report.</p> <p><em>These figures were produced by MapText'24 organizers.</em></p>
Evaluations of public submissions to ICDAR'24 MapText Competition
<p>Evaluations of submissions to the <a href="https://rrc.cvc.uab.es/?ch=28">ICDAR'24 Competition on Historical Map Text Detection, Recognition, and Linking</a>.</p> <p>Files in the archive (<code>evaluations.tar.bz2</code>) are stored in <code>ch28/tY/fZ/W.json</code> where <code>Y</code> is the task number (1–4), <code>Z</code> is the file number (1–2), and <code>W</code> is the submission ID.</p> <p>Tasks are:</p> <ul> <li><code>1</code>: Word Detection</li> <li><code>2</code>: Phrase Detection (Word Detection and Grouping)</li> <li><code>3</code>: Word Detection and Recognition</li> <li><code>4</code>: Phrase Detection and Recognition</li> </ul> <p>Files are:</p> <ul> <li><code>1</code>: Rumsey data set</li> <li><code>2</code>: IGN (French Land Register) data set</li> </ul> <p>The JSON output format results from the <a href="https://github.com/icdar-maptext/evaluation">official competition evaluation script</a> (commit-id: <code>816da23d6d4e44f7fd89ee07278ebbcd12d50d4d</code>).</p> <p>Version 2 updates results to run in an environment consistent with the RRC server.</p>
ICDAR 2017 Competition on the Classification of Medieval Handwritings in Latin Script - Dataset
<p>The ICDAR2017 Competition on the Classification of Medieval Handwritings in Latin Script (CLaMM), jointly organized by Computer Scientists and Humanists (paleographers) followed a competition at ICFHR2016 and provided a rich annotated database of European medieval manuscripts to the community on Handwriting Analysis and Recognition, containing information on date of production and class of script.</p> <p>If you use this upload, please cite:</p> <p>Florence Cloppet, Véronique Eglin, Marlène Helias-Baron, van Cuong Kieu, Dominique Stutzmann, Nicole Vincent, "ICDAR 2017 Competition on the Classification of Medieval Handwritings in Latin Script", in <em>14th IAPR International Conference on Document Analysis and Recognition</em>. ICDAR 2017, 1371-76. Kyoto: CPS, 2017. <a href="https://doi.org/10.1109/ICDAR.2017.224">https://doi.org/10.1109/ICDAR.2017.224</a></p> <p>We proposed four independent classification tasks which attracted 10 registered teams, with 6 submitted classifiers from 4 participants. Those classifiers are trained on a set of 3540 images with their ground<br> truths. In task 1 (Script classification) and task 3 (Date classification), the classifiers have been evaluated by a test set of 2000 greyscale, tiff, 300 dpi images. In task 2 (Script classification) and task 4 (Date classification), the test set consists of 1000 images in different formats, resolutions and color<br> representation.</p> <p>The present dataset contains the training dataset, both test datasets (tasks 1 and 3, and tasks 2 and 4) and the matrices provided by the competitors. It was first published on <a href="https://clamm.irht.cnrs.fr/icdar-2017/">https://clamm.irht.cnrs.fr/icdar-2017/</a> in Nov. 2017.</p>
ScienceDex guides
Understand access before you commit
These curated guides explain access requirements, typical timelines, costs, and reuse considerations for widely used research datasets.
Allen Brain Atlas
Allen Brain Atlas is an Allen Institute collection of brain map atlases, datasets, APIs, and analysis tools covering mouse, human, and non-human primate brain resources.
Annotated Behaviour and Observability Dataset (ABODe)
ABODe is a University of Edinburgh DataShare dataset for behavior classification in group-housed mice using home-cage video, identities, bounding boxes, ground-plate positions, and annotator labels.
DANDI Archive for NWB datasets
DANDI is a BRAIN Initiative archive for publishing and sharing neurophysiology data, including electrophysiology, optophysiology, and behavioral data packaged as NWB and related standards.
International Brain Laboratory public data
The International Brain Laboratory public data releases expose standardized mouse decision-making experiments, including Neuropixels recordings, widefield calcium imaging, behavior, and session metadata accessed through the ONE API.
OpenNeuro
OpenNeuro is a free, open platform for sharing neuroimaging datasets, with public search, dataset pages, and download paths for web, S3, DataLad, and the OpenNeuro CLI.