\n \n \n
\n
\n \n Antonacopoulos, A.\n \n \n (2)\n \n \n
\n
\n \n \n
\n\n\n
\n
\n\n \n \n \n \n \n \n The PAGE (Page Analysis and Ground-Truth Elements) Format Framework.\n \n \n \n \n\n\n \n Pletschacher, S.; and Antonacopoulos, A.\n\n\n \n\n\n\n In pages 257–260, August 2010. IEEE\n
\n\n
\n\n
\n\n
\n\n \n \n Paper\n \n \n\n \n \n doi\n \n \n\n \n link\n \n \n\n bibtex\n \n\n \n\n \n \n \n 1 download\n \n \n\n \n \n \n \n \n \n \n\n \n \n \n\n\n\n
\n
@inproceedings{pletschacher_page_2010,\n\ttitle = {The {PAGE} ({Page} {Analysis} and {Ground}-{Truth} {Elements}) {Format} {Framework}},\n\tisbn = {978-1-4244-7542-1},\n\turl = {http://ieeexplore.ieee.org/document/5597587/},\n\tdoi = {10.1109/ICPR.2010.72},\n\turldate = {2018-06-22},\n\tpublisher = {IEEE},\n\tauthor = {Pletschacher, Stefan and Antonacopoulos, Apostolos},\n\tmonth = aug,\n\tyear = {2010},\n\tpages = {257--260},\n}\n
\n
\n\n\n\n
\n\n\n\n\n\n
\n
\n\n
\n
\n \n Baierer, K.\n \n \n (1)\n \n \n
\n
\n
\n\n
\n
\n \n Clausner, C.\n \n \n (1)\n \n \n
\n
\n
\n\n
\n
\n \n Diem, M.\n \n \n (1)\n \n \n
\n
\n \n \n
\n
\n\n \n \n \n \n \n \n READ-BAD: A New Dataset and Evaluation Scheme for Baseline Detection in Archival Documents.\n \n \n \n \n\n\n \n Gruning, T.; Labahn, R.; Diem, M.; Kleber, F.; and Fiel, S.\n\n\n \n\n\n\n In
13th IAPR International Workshop on Document Analysis Systems, DAS 2018, Vienna, Austria, April 24-27, 2018, pages 351–356, 2018. IEEE Computer Society\n
\n\n
\n\n
\n\n
\n\n \n \n Paper\n \n \n\n \n \n doi\n \n \n\n \n link\n \n \n\n bibtex\n \n\n \n\n \n \n \n 1 download\n \n \n\n \n \n \n \n \n \n \n\n \n \n \n\n\n\n
\n
@inproceedings{gruning_read-bad:_2018,\n\ttitle = {{READ}-{BAD}: {A} {New} {Dataset} and {Evaluation} {Scheme} for {Baseline} {Detection} in {Archival} {Documents}},\n\tisbn = {978-1-5386-3346-5},\n\tshorttitle = {{READ}-{BAD}},\n\turl = {http://doi.ieeecomputersociety.org/10.1109/DAS.2018.38},\n\tdoi = {10.1109/DAS.2018.38},\n\turldate = {2018-06-29},\n\tbooktitle = {13th {IAPR} {International} {Workshop} on {Document} {Analysis} {Systems}, {DAS} 2018, {Vienna}, {Austria}, {April} 24-27, 2018},\n\tpublisher = {IEEE Computer Society},\n\tauthor = {Gruning, Tobias and Labahn, Roger and Diem, Markus and Kleber, Florian and Fiel, Stefan},\n\tyear = {2018},\n\tpages = {351--356},\n}\n\n
\n
\n\n\n\n
\n\n\n\n\n\n
\n
\n\n
\n
\n \n Dixon, S.\n \n \n (1)\n \n \n
\n
\n \n \n
\n
\n\n \n \n \n \n \n Digital performance : a history of new media in theater, dance, performance art, and installation.\n \n \n \n\n\n \n Dixon, S.; and Smith, B.\n\n\n \n\n\n\n of LeonardoThe MIT Press, Cambridge, Massachusetts, London, England, [Paperback edition] edition, 2015.\n
\n\n
\n\n
\n\n
\n\n \n\n \n\n \n link\n \n \n\n bibtex\n \n\n \n\n \n\n \n \n \n \n \n \n \n\n \n \n \n\n\n\n
\n
@book{dixon_digital_2015,\n\taddress = {Cambridge, Massachusetts, London, England},\n\tedition = {[Paperback edition]},\n\tseries = {Leonardo},\n\ttitle = {Digital performance : a history of new media in theater, dance, performance art, and installation},\n\tisbn = {978-0-262-52752-1},\n\tshorttitle = {Digital performance},\n\tpublisher = {The MIT Press},\n\tauthor = {Dixon, Steve and Smith, Barry},\n\tyear = {2015},\n}\n\n
\n
\n\n\n\n
\n\n\n\n\n\n
\n
\n\n
\n
\n \n Déjean, H.\n \n \n (1)\n \n \n
\n
\n \n \n
\n
\n\n \n \n \n \n \n \n ICDAR 2019 Competition on Table Detection and Recognition (cTDaR).\n \n \n \n \n\n\n \n Déjean, H.; Meunier, J.; Gao, L.; Huang, Y.; Fang, Y.; Kleber, F.; and Lang, E.\n\n\n \n\n\n\n April 2019.\n
\n\n
\n\n
\n\n
\n\n \n \n Paper\n \n \n\n \n \n doi\n \n \n\n \n link\n \n \n\n bibtex\n \n\n \n \n \n abstract \n \n\n \n\n \n \n \n \n \n \n \n\n \n \n \n\n\n\n
\n
@misc{dejean_icdar_2019,\n\ttitle = {{ICDAR} 2019 {Competition} on {Table} {Detection} and {Recognition} ({cTDaR})},\n\turl = {https://zenodo.org/record/3239032#.X1IyZdbgqrI},\n\tdoi = {10.5281/zenodo.3239032},\n\tabstract = {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]): {\\textless}?xml version="1.0" encoding="UTF-8"?{\\textgreater} {\\textless}document filename='filename.jpg'{\\textgreater} {\\textless}table id='Table\\_1540517170416\\_3'{\\textgreater} {\\textless}Coords points="180,160 4354,160 4354,3287 180,3287"/{\\textgreater} {\\textless}cell id='TableCell\\_1540517477147\\_58' start-row='0' start-col='0' end-row='1' end-col='2'{\\textgreater} {\\textless}Coords points="180,160 177,456 614,456 615,163"/{\\textgreater} {\\textless}/cell{\\textgreater} ... {\\textless}/table{\\textgreater} ... {\\textless}/document{\\textgreater} 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. Important Note: 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. See also: http://sac.founderit.com/tasks.html The evaluation tool is available at github: https://github.com/cndplab-founder/ctdar\\_measurement\\_tool},\n\turldate = {2020-09-04},\n\tpublisher = {Zenodo},\n\tauthor = {Déjean, Hervé and Meunier, Jean-Luc and Gao, Liangcai and Huang, Yilun and Fang, Yu and Kleber, Florian and Lang, Eva-Maria},\n\tmonth = apr,\n\tyear = {2019},\n}\n\n
\n
\n\n\n
\n 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]): \\textless?xml version=\"1.0\" encoding=\"UTF-8\"?\\textgreater \\textlessdocument filename='filename.jpg'\\textgreater \\textlesstable id='Table_1540517170416_3'\\textgreater \\textlessCoords points=\"180,160 4354,160 4354,3287 180,3287\"/\\textgreater \\textlesscell id='TableCell_1540517477147_58' start-row='0' start-col='0' end-row='1' end-col='2'\\textgreater \\textlessCoords points=\"180,160 177,456 614,456 615,163\"/\\textgreater \\textless/cell\\textgreater ... \\textless/table\\textgreater ... \\textless/document\\textgreater 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. Important Note: 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. See also: http://sac.founderit.com/tasks.html The evaluation tool is available at github: https://github.com/cndplab-founder/ctdar_measurement_tool\n
\n\n\n
\n\n\n\n\n\n
\n
\n\n
\n
\n \n Fang, Y.\n \n \n (1)\n \n \n
\n
\n \n \n
\n
\n\n \n \n \n \n \n \n ICDAR 2019 Competition on Table Detection and Recognition (cTDaR).\n \n \n \n \n\n\n \n Déjean, H.; Meunier, J.; Gao, L.; Huang, Y.; Fang, Y.; Kleber, F.; and Lang, E.\n\n\n \n\n\n\n April 2019.\n
\n\n
\n\n
\n\n
\n\n \n \n Paper\n \n \n\n \n \n doi\n \n \n\n \n link\n \n \n\n bibtex\n \n\n \n \n \n abstract \n \n\n \n\n \n \n \n \n \n \n \n\n \n \n \n\n\n\n
\n
@misc{dejean_icdar_2019,\n\ttitle = {{ICDAR} 2019 {Competition} on {Table} {Detection} and {Recognition} ({cTDaR})},\n\turl = {https://zenodo.org/record/3239032#.X1IyZdbgqrI},\n\tdoi = {10.5281/zenodo.3239032},\n\tabstract = {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]): {\\textless}?xml version="1.0" encoding="UTF-8"?{\\textgreater} {\\textless}document filename='filename.jpg'{\\textgreater} {\\textless}table id='Table\\_1540517170416\\_3'{\\textgreater} {\\textless}Coords points="180,160 4354,160 4354,3287 180,3287"/{\\textgreater} {\\textless}cell id='TableCell\\_1540517477147\\_58' start-row='0' start-col='0' end-row='1' end-col='2'{\\textgreater} {\\textless}Coords points="180,160 177,456 614,456 615,163"/{\\textgreater} {\\textless}/cell{\\textgreater} ... {\\textless}/table{\\textgreater} ... {\\textless}/document{\\textgreater} 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. Important Note: 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. See also: http://sac.founderit.com/tasks.html The evaluation tool is available at github: https://github.com/cndplab-founder/ctdar\\_measurement\\_tool},\n\turldate = {2020-09-04},\n\tpublisher = {Zenodo},\n\tauthor = {Déjean, Hervé and Meunier, Jean-Luc and Gao, Liangcai and Huang, Yilun and Fang, Yu and Kleber, Florian and Lang, Eva-Maria},\n\tmonth = apr,\n\tyear = {2019},\n}\n\n
\n
\n\n\n
\n 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]): \\textless?xml version=\"1.0\" encoding=\"UTF-8\"?\\textgreater \\textlessdocument filename='filename.jpg'\\textgreater \\textlesstable id='Table_1540517170416_3'\\textgreater \\textlessCoords points=\"180,160 4354,160 4354,3287 180,3287\"/\\textgreater \\textlesscell id='TableCell_1540517477147_58' start-row='0' start-col='0' end-row='1' end-col='2'\\textgreater \\textlessCoords points=\"180,160 177,456 614,456 615,163\"/\\textgreater \\textless/cell\\textgreater ... \\textless/table\\textgreater ... \\textless/document\\textgreater 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. Important Note: 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. See also: http://sac.founderit.com/tasks.html The evaluation tool is available at github: https://github.com/cndplab-founder/ctdar_measurement_tool\n
\n\n\n
\n\n\n\n\n\n
\n
\n\n
\n
\n \n Fiel, S.\n \n \n (1)\n \n \n
\n
\n \n \n
\n
\n\n \n \n \n \n \n \n READ-BAD: A New Dataset and Evaluation Scheme for Baseline Detection in Archival Documents.\n \n \n \n \n\n\n \n Gruning, T.; Labahn, R.; Diem, M.; Kleber, F.; and Fiel, S.\n\n\n \n\n\n\n In
13th IAPR International Workshop on Document Analysis Systems, DAS 2018, Vienna, Austria, April 24-27, 2018, pages 351–356, 2018. IEEE Computer Society\n
\n\n
\n\n
\n\n
\n\n \n \n Paper\n \n \n\n \n \n doi\n \n \n\n \n link\n \n \n\n bibtex\n \n\n \n\n \n \n \n 1 download\n \n \n\n \n \n \n \n \n \n \n\n \n \n \n\n\n\n
\n
@inproceedings{gruning_read-bad:_2018,\n\ttitle = {{READ}-{BAD}: {A} {New} {Dataset} and {Evaluation} {Scheme} for {Baseline} {Detection} in {Archival} {Documents}},\n\tisbn = {978-1-5386-3346-5},\n\tshorttitle = {{READ}-{BAD}},\n\turl = {http://doi.ieeecomputersociety.org/10.1109/DAS.2018.38},\n\tdoi = {10.1109/DAS.2018.38},\n\turldate = {2018-06-29},\n\tbooktitle = {13th {IAPR} {International} {Workshop} on {Document} {Analysis} {Systems}, {DAS} 2018, {Vienna}, {Austria}, {April} 24-27, 2018},\n\tpublisher = {IEEE Computer Society},\n\tauthor = {Gruning, Tobias and Labahn, Roger and Diem, Markus and Kleber, Florian and Fiel, Stefan},\n\tyear = {2018},\n\tpages = {351--356},\n}\n\n
\n
\n\n\n\n
\n\n\n\n\n\n
\n
\n\n
\n
\n \n Gao, L.\n \n \n (1)\n \n \n
\n
\n \n \n
\n
\n\n \n \n \n \n \n \n ICDAR 2019 Competition on Table Detection and Recognition (cTDaR).\n \n \n \n \n\n\n \n Déjean, H.; Meunier, J.; Gao, L.; Huang, Y.; Fang, Y.; Kleber, F.; and Lang, E.\n\n\n \n\n\n\n April 2019.\n
\n\n
\n\n
\n\n
\n\n \n \n Paper\n \n \n\n \n \n doi\n \n \n\n \n link\n \n \n\n bibtex\n \n\n \n \n \n abstract \n \n\n \n\n \n \n \n \n \n \n \n\n \n \n \n\n\n\n
\n
@misc{dejean_icdar_2019,\n\ttitle = {{ICDAR} 2019 {Competition} on {Table} {Detection} and {Recognition} ({cTDaR})},\n\turl = {https://zenodo.org/record/3239032#.X1IyZdbgqrI},\n\tdoi = {10.5281/zenodo.3239032},\n\tabstract = {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]): {\\textless}?xml version="1.0" encoding="UTF-8"?{\\textgreater} {\\textless}document filename='filename.jpg'{\\textgreater} {\\textless}table id='Table\\_1540517170416\\_3'{\\textgreater} {\\textless}Coords points="180,160 4354,160 4354,3287 180,3287"/{\\textgreater} {\\textless}cell id='TableCell\\_1540517477147\\_58' start-row='0' start-col='0' end-row='1' end-col='2'{\\textgreater} {\\textless}Coords points="180,160 177,456 614,456 615,163"/{\\textgreater} {\\textless}/cell{\\textgreater} ... {\\textless}/table{\\textgreater} ... {\\textless}/document{\\textgreater} 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. Important Note: 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. See also: http://sac.founderit.com/tasks.html The evaluation tool is available at github: https://github.com/cndplab-founder/ctdar\\_measurement\\_tool},\n\turldate = {2020-09-04},\n\tpublisher = {Zenodo},\n\tauthor = {Déjean, Hervé and Meunier, Jean-Luc and Gao, Liangcai and Huang, Yilun and Fang, Yu and Kleber, Florian and Lang, Eva-Maria},\n\tmonth = apr,\n\tyear = {2019},\n}\n\n
\n
\n\n\n
\n 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]): \\textless?xml version=\"1.0\" encoding=\"UTF-8\"?\\textgreater \\textlessdocument filename='filename.jpg'\\textgreater \\textlesstable id='Table_1540517170416_3'\\textgreater \\textlessCoords points=\"180,160 4354,160 4354,3287 180,3287\"/\\textgreater \\textlesscell id='TableCell_1540517477147_58' start-row='0' start-col='0' end-row='1' end-col='2'\\textgreater \\textlessCoords points=\"180,160 177,456 614,456 615,163\"/\\textgreater \\textless/cell\\textgreater ... \\textless/table\\textgreater ... \\textless/document\\textgreater 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. Important Note: 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. See also: http://sac.founderit.com/tasks.html The evaluation tool is available at github: https://github.com/cndplab-founder/ctdar_measurement_tool\n
\n\n\n
\n\n\n\n\n\n
\n
\n\n
\n
\n \n Gatos, Ba\n \n \n (1)\n \n \n
\n
\n \n \n
\n
\n\n \n \n \n \n \n Goal-oriented performance evaluation methodology for page segmentation techniques.\n \n \n \n\n\n \n Stamatopoulos, Nikolaos; and Gatos, Basilis\n\n\n \n\n\n\n In
Proceedings of the 13th international conference on document analysis and recognition (ICDAR), pages 281–285. 2015.\n
\n\n
\n\n
\n\n
\n\n \n\n \n\n \n link\n \n \n\n bibtex\n \n\n \n\n \n\n \n \n \n \n \n \n \n\n \n \n \n\n\n\n
\n
@incollection{stamatopoulos_nikolaos_goal-oriented_2015,\n\ttitle = {Goal-oriented performance evaluation methodology for page segmentation techniques},\n\tbooktitle = {Proceedings of the 13th international conference on document analysis and recognition ({ICDAR})},\n\tauthor = {{Stamatopoulos, Nikolaos} and {Gatos, Basilis}},\n\tyear = {2015},\n\tpages = {281--285},\n}\n\n
\n
\n\n\n\n
\n\n\n\n\n\n
\n
\n\n
\n
\n \n Gerber, M.\n \n \n (1)\n \n \n
\n
\n
\n\n
\n
\n \n Gruning, T.\n \n \n (1)\n \n \n
\n
\n \n \n
\n
\n\n \n \n \n \n \n \n READ-BAD: A New Dataset and Evaluation Scheme for Baseline Detection in Archival Documents.\n \n \n \n \n\n\n \n Gruning, T.; Labahn, R.; Diem, M.; Kleber, F.; and Fiel, S.\n\n\n \n\n\n\n In
13th IAPR International Workshop on Document Analysis Systems, DAS 2018, Vienna, Austria, April 24-27, 2018, pages 351–356, 2018. IEEE Computer Society\n
\n\n
\n\n
\n\n
\n\n \n \n Paper\n \n \n\n \n \n doi\n \n \n\n \n link\n \n \n\n bibtex\n \n\n \n\n \n \n \n 1 download\n \n \n\n \n \n \n \n \n \n \n\n \n \n \n\n\n\n
\n
@inproceedings{gruning_read-bad:_2018,\n\ttitle = {{READ}-{BAD}: {A} {New} {Dataset} and {Evaluation} {Scheme} for {Baseline} {Detection} in {Archival} {Documents}},\n\tisbn = {978-1-5386-3346-5},\n\tshorttitle = {{READ}-{BAD}},\n\turl = {http://doi.ieeecomputersociety.org/10.1109/DAS.2018.38},\n\tdoi = {10.1109/DAS.2018.38},\n\turldate = {2018-06-29},\n\tbooktitle = {13th {IAPR} {International} {Workshop} on {Document} {Analysis} {Systems}, {DAS} 2018, {Vienna}, {Austria}, {April} 24-27, 2018},\n\tpublisher = {IEEE Computer Society},\n\tauthor = {Gruning, Tobias and Labahn, Roger and Diem, Markus and Kleber, Florian and Fiel, Stefan},\n\tyear = {2018},\n\tpages = {351--356},\n}\n\n
\n
\n\n\n\n
\n\n\n\n\n\n
\n
\n\n
\n
\n \n Grüning, T.\n \n \n (1)\n \n \n
\n
\n
\n\n
\n
\n \n Hodel, T.\n \n \n (1)\n \n \n
\n
\n \n \n
\n
\n\n \n \n \n \n \n ICFHR2018 Competition on Automated Text Recognition on a READ Dataset.\n \n \n \n\n\n \n Strauß, T.; Leifert, G.; Labahn, R.; Hodel, T.; and Mühlberger, G.\n\n\n \n\n\n\n In
2018 16th International Conference on Frontiers in Handwriting Recognition (ICFHR), pages 477–482, August 2018. \n
\n\n
\n\n
\n\n
\n\n \n\n \n \n doi\n \n \n\n \n link\n \n \n\n bibtex\n \n\n \n \n \n abstract \n \n\n \n\n \n \n \n \n \n \n \n\n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n\n\n\n
\n
@inproceedings{straus_icfhr2018_2018,\n\ttitle = {{ICFHR2018} {Competition} on {Automated} {Text} {Recognition} on a {READ} {Dataset}},\n\tdoi = {10.1109/ICFHR-2018.2018.00089},\n\tabstract = {We summarize the results of a competition on Automated Text Recognition targeting the effective adaptation of recognition engines to essentially new data. The task consists in achieving a minimum character error rate on a previously unknown text corpus from which only a few pages are available for adjusting an already pre-trained recognition engine. This issue addresses a frequent application scenario where only a small amount of task-specific training data is available, because producing this data usually requires much effort. We present the results of five submission. They show that the task is a challenging issue but for certain documents 16 pages of transcription are sufficient to adapt a pre-trained recognition system.},\n\tbooktitle = {2018 16th {International} {Conference} on {Frontiers} in {Handwriting} {Recognition} ({ICFHR})},\n\tauthor = {Strauß, Tobias and Leifert, Gundram and Labahn, Roger and Hodel, Tobias and Mühlberger, Günter},\n\tmonth = aug,\n\tyear = {2018},\n\tkeywords = {Computational modeling, Data models, Optical imaging, Task analysis, Text recognition, Training, Training data, automated text recognition, fast adaptation, few shot learning, historical documents},\n\tpages = {477--482},\n}\n\n
\n
\n\n\n
\n We summarize the results of a competition on Automated Text Recognition targeting the effective adaptation of recognition engines to essentially new data. The task consists in achieving a minimum character error rate on a previously unknown text corpus from which only a few pages are available for adjusting an already pre-trained recognition engine. This issue addresses a frequent application scenario where only a small amount of task-specific training data is available, because producing this data usually requires much effort. We present the results of five submission. They show that the task is a challenging issue but for certain documents 16 pages of transcription are sufficient to adapt a pre-trained recognition system.\n
\n\n\n
\n\n\n\n\n\n
\n
\n\n
\n
\n \n Huang, Y.\n \n \n (1)\n \n \n
\n
\n \n \n
\n
\n\n \n \n \n \n \n \n ICDAR 2019 Competition on Table Detection and Recognition (cTDaR).\n \n \n \n \n\n\n \n Déjean, H.; Meunier, J.; Gao, L.; Huang, Y.; Fang, Y.; Kleber, F.; and Lang, E.\n\n\n \n\n\n\n April 2019.\n
\n\n
\n\n
\n\n
\n\n \n \n Paper\n \n \n\n \n \n doi\n \n \n\n \n link\n \n \n\n bibtex\n \n\n \n \n \n abstract \n \n\n \n\n \n \n \n \n \n \n \n\n \n \n \n\n\n\n
\n
@misc{dejean_icdar_2019,\n\ttitle = {{ICDAR} 2019 {Competition} on {Table} {Detection} and {Recognition} ({cTDaR})},\n\turl = {https://zenodo.org/record/3239032#.X1IyZdbgqrI},\n\tdoi = {10.5281/zenodo.3239032},\n\tabstract = {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]): {\\textless}?xml version="1.0" encoding="UTF-8"?{\\textgreater} {\\textless}document filename='filename.jpg'{\\textgreater} {\\textless}table id='Table\\_1540517170416\\_3'{\\textgreater} {\\textless}Coords points="180,160 4354,160 4354,3287 180,3287"/{\\textgreater} {\\textless}cell id='TableCell\\_1540517477147\\_58' start-row='0' start-col='0' end-row='1' end-col='2'{\\textgreater} {\\textless}Coords points="180,160 177,456 614,456 615,163"/{\\textgreater} {\\textless}/cell{\\textgreater} ... {\\textless}/table{\\textgreater} ... {\\textless}/document{\\textgreater} 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. Important Note: 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. See also: http://sac.founderit.com/tasks.html The evaluation tool is available at github: https://github.com/cndplab-founder/ctdar\\_measurement\\_tool},\n\turldate = {2020-09-04},\n\tpublisher = {Zenodo},\n\tauthor = {Déjean, Hervé and Meunier, Jean-Luc and Gao, Liangcai and Huang, Yilun and Fang, Yu and Kleber, Florian and Lang, Eva-Maria},\n\tmonth = apr,\n\tyear = {2019},\n}\n\n
\n
\n\n\n
\n 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]): \\textless?xml version=\"1.0\" encoding=\"UTF-8\"?\\textgreater \\textlessdocument filename='filename.jpg'\\textgreater \\textlesstable id='Table_1540517170416_3'\\textgreater \\textlessCoords points=\"180,160 4354,160 4354,3287 180,3287\"/\\textgreater \\textlesscell id='TableCell_1540517477147_58' start-row='0' start-col='0' end-row='1' end-col='2'\\textgreater \\textlessCoords points=\"180,160 177,456 614,456 615,163\"/\\textgreater \\textless/cell\\textgreater ... \\textless/table\\textgreater ... \\textless/document\\textgreater 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. Important Note: 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. See also: http://sac.founderit.com/tasks.html The evaluation tool is available at github: https://github.com/cndplab-founder/ctdar_measurement_tool\n
\n\n\n
\n\n\n\n\n\n
\n
\n\n
\n
\n \n Kleber, F.\n \n \n (2)\n \n \n
\n
\n \n \n
\n
\n\n \n \n \n \n \n \n ICDAR 2019 Competition on Table Detection and Recognition (cTDaR).\n \n \n \n \n\n\n \n Déjean, H.; Meunier, J.; Gao, L.; Huang, Y.; Fang, Y.; Kleber, F.; and Lang, E.\n\n\n \n\n\n\n April 2019.\n
\n\n
\n\n
\n\n
\n\n \n \n Paper\n \n \n\n \n \n doi\n \n \n\n \n link\n \n \n\n bibtex\n \n\n \n \n \n abstract \n \n\n \n\n \n \n \n \n \n \n \n\n \n \n \n\n\n\n
\n
@misc{dejean_icdar_2019,\n\ttitle = {{ICDAR} 2019 {Competition} on {Table} {Detection} and {Recognition} ({cTDaR})},\n\turl = {https://zenodo.org/record/3239032#.X1IyZdbgqrI},\n\tdoi = {10.5281/zenodo.3239032},\n\tabstract = {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]): {\\textless}?xml version="1.0" encoding="UTF-8"?{\\textgreater} {\\textless}document filename='filename.jpg'{\\textgreater} {\\textless}table id='Table\\_1540517170416\\_3'{\\textgreater} {\\textless}Coords points="180,160 4354,160 4354,3287 180,3287"/{\\textgreater} {\\textless}cell id='TableCell\\_1540517477147\\_58' start-row='0' start-col='0' end-row='1' end-col='2'{\\textgreater} {\\textless}Coords points="180,160 177,456 614,456 615,163"/{\\textgreater} {\\textless}/cell{\\textgreater} ... {\\textless}/table{\\textgreater} ... {\\textless}/document{\\textgreater} 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. Important Note: 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. See also: http://sac.founderit.com/tasks.html The evaluation tool is available at github: https://github.com/cndplab-founder/ctdar\\_measurement\\_tool},\n\turldate = {2020-09-04},\n\tpublisher = {Zenodo},\n\tauthor = {Déjean, Hervé and Meunier, Jean-Luc and Gao, Liangcai and Huang, Yilun and Fang, Yu and Kleber, Florian and Lang, Eva-Maria},\n\tmonth = apr,\n\tyear = {2019},\n}\n\n
\n
\n\n\n
\n 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]): \\textless?xml version=\"1.0\" encoding=\"UTF-8\"?\\textgreater \\textlessdocument filename='filename.jpg'\\textgreater \\textlesstable id='Table_1540517170416_3'\\textgreater \\textlessCoords points=\"180,160 4354,160 4354,3287 180,3287\"/\\textgreater \\textlesscell id='TableCell_1540517477147_58' start-row='0' start-col='0' end-row='1' end-col='2'\\textgreater \\textlessCoords points=\"180,160 177,456 614,456 615,163\"/\\textgreater \\textless/cell\\textgreater ... \\textless/table\\textgreater ... \\textless/document\\textgreater 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. Important Note: 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. See also: http://sac.founderit.com/tasks.html The evaluation tool is available at github: https://github.com/cndplab-founder/ctdar_measurement_tool\n
\n\n\n
\n\n\n
\n
\n\n \n \n \n \n \n \n READ-BAD: A New Dataset and Evaluation Scheme for Baseline Detection in Archival Documents.\n \n \n \n \n\n\n \n Gruning, T.; Labahn, R.; Diem, M.; Kleber, F.; and Fiel, S.\n\n\n \n\n\n\n In
13th IAPR International Workshop on Document Analysis Systems, DAS 2018, Vienna, Austria, April 24-27, 2018, pages 351–356, 2018. IEEE Computer Society\n
\n\n
\n\n
\n\n
\n\n \n \n Paper\n \n \n\n \n \n doi\n \n \n\n \n link\n \n \n\n bibtex\n \n\n \n\n \n \n \n 1 download\n \n \n\n \n \n \n \n \n \n \n\n \n \n \n\n\n\n
\n
@inproceedings{gruning_read-bad:_2018,\n\ttitle = {{READ}-{BAD}: {A} {New} {Dataset} and {Evaluation} {Scheme} for {Baseline} {Detection} in {Archival} {Documents}},\n\tisbn = {978-1-5386-3346-5},\n\tshorttitle = {{READ}-{BAD}},\n\turl = {http://doi.ieeecomputersociety.org/10.1109/DAS.2018.38},\n\tdoi = {10.1109/DAS.2018.38},\n\turldate = {2018-06-29},\n\tbooktitle = {13th {IAPR} {International} {Workshop} on {Document} {Analysis} {Systems}, {DAS} 2018, {Vienna}, {Austria}, {April} 24-27, 2018},\n\tpublisher = {IEEE Computer Society},\n\tauthor = {Gruning, Tobias and Labahn, Roger and Diem, Markus and Kleber, Florian and Fiel, Stefan},\n\tyear = {2018},\n\tpages = {351--356},\n}\n\n
\n
\n\n\n\n
\n\n\n\n\n\n
\n
\n\n
\n
\n \n Labahn, R.\n \n \n (3)\n \n \n
\n
\n \n \n
\n
\n\n \n \n \n \n \n ICFHR2018 Competition on Automated Text Recognition on a READ Dataset.\n \n \n \n\n\n \n Strauß, T.; Leifert, G.; Labahn, R.; Hodel, T.; and Mühlberger, G.\n\n\n \n\n\n\n In
2018 16th International Conference on Frontiers in Handwriting Recognition (ICFHR), pages 477–482, August 2018. \n
\n\n
\n\n
\n\n
\n\n \n\n \n \n doi\n \n \n\n \n link\n \n \n\n bibtex\n \n\n \n \n \n abstract \n \n\n \n\n \n \n \n \n \n \n \n\n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n\n\n\n
\n
@inproceedings{straus_icfhr2018_2018,\n\ttitle = {{ICFHR2018} {Competition} on {Automated} {Text} {Recognition} on a {READ} {Dataset}},\n\tdoi = {10.1109/ICFHR-2018.2018.00089},\n\tabstract = {We summarize the results of a competition on Automated Text Recognition targeting the effective adaptation of recognition engines to essentially new data. The task consists in achieving a minimum character error rate on a previously unknown text corpus from which only a few pages are available for adjusting an already pre-trained recognition engine. This issue addresses a frequent application scenario where only a small amount of task-specific training data is available, because producing this data usually requires much effort. We present the results of five submission. They show that the task is a challenging issue but for certain documents 16 pages of transcription are sufficient to adapt a pre-trained recognition system.},\n\tbooktitle = {2018 16th {International} {Conference} on {Frontiers} in {Handwriting} {Recognition} ({ICFHR})},\n\tauthor = {Strauß, Tobias and Leifert, Gundram and Labahn, Roger and Hodel, Tobias and Mühlberger, Günter},\n\tmonth = aug,\n\tyear = {2018},\n\tkeywords = {Computational modeling, Data models, Optical imaging, Task analysis, Text recognition, Training, Training data, automated text recognition, fast adaptation, few shot learning, historical documents},\n\tpages = {477--482},\n}\n\n
\n
\n\n\n
\n We summarize the results of a competition on Automated Text Recognition targeting the effective adaptation of recognition engines to essentially new data. The task consists in achieving a minimum character error rate on a previously unknown text corpus from which only a few pages are available for adjusting an already pre-trained recognition engine. This issue addresses a frequent application scenario where only a small amount of task-specific training data is available, because producing this data usually requires much effort. We present the results of five submission. They show that the task is a challenging issue but for certain documents 16 pages of transcription are sufficient to adapt a pre-trained recognition system.\n
\n\n\n
\n\n\n
\n\n\n
\n
\n\n \n \n \n \n \n \n READ-BAD: A New Dataset and Evaluation Scheme for Baseline Detection in Archival Documents.\n \n \n \n \n\n\n \n Gruning, T.; Labahn, R.; Diem, M.; Kleber, F.; and Fiel, S.\n\n\n \n\n\n\n In
13th IAPR International Workshop on Document Analysis Systems, DAS 2018, Vienna, Austria, April 24-27, 2018, pages 351–356, 2018. IEEE Computer Society\n
\n\n
\n\n
\n\n
\n\n \n \n Paper\n \n \n\n \n \n doi\n \n \n\n \n link\n \n \n\n bibtex\n \n\n \n\n \n \n \n 1 download\n \n \n\n \n \n \n \n \n \n \n\n \n \n \n\n\n\n
\n
@inproceedings{gruning_read-bad:_2018,\n\ttitle = {{READ}-{BAD}: {A} {New} {Dataset} and {Evaluation} {Scheme} for {Baseline} {Detection} in {Archival} {Documents}},\n\tisbn = {978-1-5386-3346-5},\n\tshorttitle = {{READ}-{BAD}},\n\turl = {http://doi.ieeecomputersociety.org/10.1109/DAS.2018.38},\n\tdoi = {10.1109/DAS.2018.38},\n\turldate = {2018-06-29},\n\tbooktitle = {13th {IAPR} {International} {Workshop} on {Document} {Analysis} {Systems}, {DAS} 2018, {Vienna}, {Austria}, {April} 24-27, 2018},\n\tpublisher = {IEEE Computer Society},\n\tauthor = {Gruning, Tobias and Labahn, Roger and Diem, Markus and Kleber, Florian and Fiel, Stefan},\n\tyear = {2018},\n\tpages = {351--356},\n}\n\n
\n
\n\n\n\n
\n\n\n\n\n\n
\n
\n\n
\n
\n \n Lang, E.\n \n \n (1)\n \n \n
\n
\n \n \n
\n
\n\n \n \n \n \n \n \n ICDAR 2019 Competition on Table Detection and Recognition (cTDaR).\n \n \n \n \n\n\n \n Déjean, H.; Meunier, J.; Gao, L.; Huang, Y.; Fang, Y.; Kleber, F.; and Lang, E.\n\n\n \n\n\n\n April 2019.\n
\n\n
\n\n
\n\n
\n\n \n \n Paper\n \n \n\n \n \n doi\n \n \n\n \n link\n \n \n\n bibtex\n \n\n \n \n \n abstract \n \n\n \n\n \n \n \n \n \n \n \n\n \n \n \n\n\n\n
\n
@misc{dejean_icdar_2019,\n\ttitle = {{ICDAR} 2019 {Competition} on {Table} {Detection} and {Recognition} ({cTDaR})},\n\turl = {https://zenodo.org/record/3239032#.X1IyZdbgqrI},\n\tdoi = {10.5281/zenodo.3239032},\n\tabstract = {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]): {\\textless}?xml version="1.0" encoding="UTF-8"?{\\textgreater} {\\textless}document filename='filename.jpg'{\\textgreater} {\\textless}table id='Table\\_1540517170416\\_3'{\\textgreater} {\\textless}Coords points="180,160 4354,160 4354,3287 180,3287"/{\\textgreater} {\\textless}cell id='TableCell\\_1540517477147\\_58' start-row='0' start-col='0' end-row='1' end-col='2'{\\textgreater} {\\textless}Coords points="180,160 177,456 614,456 615,163"/{\\textgreater} {\\textless}/cell{\\textgreater} ... {\\textless}/table{\\textgreater} ... {\\textless}/document{\\textgreater} 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. Important Note: 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. See also: http://sac.founderit.com/tasks.html The evaluation tool is available at github: https://github.com/cndplab-founder/ctdar\\_measurement\\_tool},\n\turldate = {2020-09-04},\n\tpublisher = {Zenodo},\n\tauthor = {Déjean, Hervé and Meunier, Jean-Luc and Gao, Liangcai and Huang, Yilun and Fang, Yu and Kleber, Florian and Lang, Eva-Maria},\n\tmonth = apr,\n\tyear = {2019},\n}\n\n
\n
\n\n\n
\n 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]): \\textless?xml version=\"1.0\" encoding=\"UTF-8\"?\\textgreater \\textlessdocument filename='filename.jpg'\\textgreater \\textlesstable id='Table_1540517170416_3'\\textgreater \\textlessCoords points=\"180,160 4354,160 4354,3287 180,3287\"/\\textgreater \\textlesscell id='TableCell_1540517477147_58' start-row='0' start-col='0' end-row='1' end-col='2'\\textgreater \\textlessCoords points=\"180,160 177,456 614,456 615,163\"/\\textgreater \\textless/cell\\textgreater ... \\textless/table\\textgreater ... \\textless/document\\textgreater 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. Important Note: 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. See also: http://sac.founderit.com/tasks.html The evaluation tool is available at github: https://github.com/cndplab-founder/ctdar_measurement_tool\n
\n\n\n
\n\n\n\n\n\n
\n
\n\n
\n
\n \n Leifert, G.\n \n \n (2)\n \n \n
\n
\n \n \n
\n
\n\n \n \n \n \n \n ICFHR2018 Competition on Automated Text Recognition on a READ Dataset.\n \n \n \n\n\n \n Strauß, T.; Leifert, G.; Labahn, R.; Hodel, T.; and Mühlberger, G.\n\n\n \n\n\n\n In
2018 16th International Conference on Frontiers in Handwriting Recognition (ICFHR), pages 477–482, August 2018. \n
\n\n
\n\n
\n\n
\n\n \n\n \n \n doi\n \n \n\n \n link\n \n \n\n bibtex\n \n\n \n \n \n abstract \n \n\n \n\n \n \n \n \n \n \n \n\n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n\n\n\n
\n
@inproceedings{straus_icfhr2018_2018,\n\ttitle = {{ICFHR2018} {Competition} on {Automated} {Text} {Recognition} on a {READ} {Dataset}},\n\tdoi = {10.1109/ICFHR-2018.2018.00089},\n\tabstract = {We summarize the results of a competition on Automated Text Recognition targeting the effective adaptation of recognition engines to essentially new data. The task consists in achieving a minimum character error rate on a previously unknown text corpus from which only a few pages are available for adjusting an already pre-trained recognition engine. This issue addresses a frequent application scenario where only a small amount of task-specific training data is available, because producing this data usually requires much effort. We present the results of five submission. They show that the task is a challenging issue but for certain documents 16 pages of transcription are sufficient to adapt a pre-trained recognition system.},\n\tbooktitle = {2018 16th {International} {Conference} on {Frontiers} in {Handwriting} {Recognition} ({ICFHR})},\n\tauthor = {Strauß, Tobias and Leifert, Gundram and Labahn, Roger and Hodel, Tobias and Mühlberger, Günter},\n\tmonth = aug,\n\tyear = {2018},\n\tkeywords = {Computational modeling, Data models, Optical imaging, Task analysis, Text recognition, Training, Training data, automated text recognition, fast adaptation, few shot learning, historical documents},\n\tpages = {477--482},\n}\n\n
\n
\n\n\n
\n We summarize the results of a competition on Automated Text Recognition targeting the effective adaptation of recognition engines to essentially new data. The task consists in achieving a minimum character error rate on a previously unknown text corpus from which only a few pages are available for adjusting an already pre-trained recognition engine. This issue addresses a frequent application scenario where only a small amount of task-specific training data is available, because producing this data usually requires much effort. We present the results of five submission. They show that the task is a challenging issue but for certain documents 16 pages of transcription are sufficient to adapt a pre-trained recognition system.\n
\n\n\n
\n\n\n
\n\n\n\n\n\n
\n
\n\n
\n
\n \n Meunier, J.\n \n \n (1)\n \n \n
\n
\n \n \n
\n
\n\n \n \n \n \n \n \n ICDAR 2019 Competition on Table Detection and Recognition (cTDaR).\n \n \n \n \n\n\n \n Déjean, H.; Meunier, J.; Gao, L.; Huang, Y.; Fang, Y.; Kleber, F.; and Lang, E.\n\n\n \n\n\n\n April 2019.\n
\n\n
\n\n
\n\n
\n\n \n \n Paper\n \n \n\n \n \n doi\n \n \n\n \n link\n \n \n\n bibtex\n \n\n \n \n \n abstract \n \n\n \n\n \n \n \n \n \n \n \n\n \n \n \n\n\n\n
\n
@misc{dejean_icdar_2019,\n\ttitle = {{ICDAR} 2019 {Competition} on {Table} {Detection} and {Recognition} ({cTDaR})},\n\turl = {https://zenodo.org/record/3239032#.X1IyZdbgqrI},\n\tdoi = {10.5281/zenodo.3239032},\n\tabstract = {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]): {\\textless}?xml version="1.0" encoding="UTF-8"?{\\textgreater} {\\textless}document filename='filename.jpg'{\\textgreater} {\\textless}table id='Table\\_1540517170416\\_3'{\\textgreater} {\\textless}Coords points="180,160 4354,160 4354,3287 180,3287"/{\\textgreater} {\\textless}cell id='TableCell\\_1540517477147\\_58' start-row='0' start-col='0' end-row='1' end-col='2'{\\textgreater} {\\textless}Coords points="180,160 177,456 614,456 615,163"/{\\textgreater} {\\textless}/cell{\\textgreater} ... {\\textless}/table{\\textgreater} ... {\\textless}/document{\\textgreater} 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. Important Note: 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. See also: http://sac.founderit.com/tasks.html The evaluation tool is available at github: https://github.com/cndplab-founder/ctdar\\_measurement\\_tool},\n\turldate = {2020-09-04},\n\tpublisher = {Zenodo},\n\tauthor = {Déjean, Hervé and Meunier, Jean-Luc and Gao, Liangcai and Huang, Yilun and Fang, Yu and Kleber, Florian and Lang, Eva-Maria},\n\tmonth = apr,\n\tyear = {2019},\n}\n\n
\n
\n\n\n
\n 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]): \\textless?xml version=\"1.0\" encoding=\"UTF-8\"?\\textgreater \\textlessdocument filename='filename.jpg'\\textgreater \\textlesstable id='Table_1540517170416_3'\\textgreater \\textlessCoords points=\"180,160 4354,160 4354,3287 180,3287\"/\\textgreater \\textlesscell id='TableCell_1540517477147_58' start-row='0' start-col='0' end-row='1' end-col='2'\\textgreater \\textlessCoords points=\"180,160 177,456 614,456 615,163\"/\\textgreater \\textless/cell\\textgreater ... \\textless/table\\textgreater ... \\textless/document\\textgreater 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. Important Note: 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. See also: http://sac.founderit.com/tasks.html The evaluation tool is available at github: https://github.com/cndplab-founder/ctdar_measurement_tool\n
\n\n\n
\n\n\n\n\n\n
\n
\n\n
\n
\n \n Michael, J.\n \n \n (1)\n \n \n
\n
\n
\n\n
\n
\n \n Mühlberger, G.\n \n \n (1)\n \n \n
\n
\n \n \n
\n
\n\n \n \n \n \n \n ICFHR2018 Competition on Automated Text Recognition on a READ Dataset.\n \n \n \n\n\n \n Strauß, T.; Leifert, G.; Labahn, R.; Hodel, T.; and Mühlberger, G.\n\n\n \n\n\n\n In
2018 16th International Conference on Frontiers in Handwriting Recognition (ICFHR), pages 477–482, August 2018. \n
\n\n
\n\n
\n\n
\n\n \n\n \n \n doi\n \n \n\n \n link\n \n \n\n bibtex\n \n\n \n \n \n abstract \n \n\n \n\n \n \n \n \n \n \n \n\n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n\n\n\n
\n
@inproceedings{straus_icfhr2018_2018,\n\ttitle = {{ICFHR2018} {Competition} on {Automated} {Text} {Recognition} on a {READ} {Dataset}},\n\tdoi = {10.1109/ICFHR-2018.2018.00089},\n\tabstract = {We summarize the results of a competition on Automated Text Recognition targeting the effective adaptation of recognition engines to essentially new data. The task consists in achieving a minimum character error rate on a previously unknown text corpus from which only a few pages are available for adjusting an already pre-trained recognition engine. This issue addresses a frequent application scenario where only a small amount of task-specific training data is available, because producing this data usually requires much effort. We present the results of five submission. They show that the task is a challenging issue but for certain documents 16 pages of transcription are sufficient to adapt a pre-trained recognition system.},\n\tbooktitle = {2018 16th {International} {Conference} on {Frontiers} in {Handwriting} {Recognition} ({ICFHR})},\n\tauthor = {Strauß, Tobias and Leifert, Gundram and Labahn, Roger and Hodel, Tobias and Mühlberger, Günter},\n\tmonth = aug,\n\tyear = {2018},\n\tkeywords = {Computational modeling, Data models, Optical imaging, Task analysis, Text recognition, Training, Training data, automated text recognition, fast adaptation, few shot learning, historical documents},\n\tpages = {477--482},\n}\n\n
\n
\n\n\n
\n We summarize the results of a competition on Automated Text Recognition targeting the effective adaptation of recognition engines to essentially new data. The task consists in achieving a minimum character error rate on a previously unknown text corpus from which only a few pages are available for adjusting an already pre-trained recognition engine. This issue addresses a frequent application scenario where only a small amount of task-specific training data is available, because producing this data usually requires much effort. We present the results of five submission. They show that the task is a challenging issue but for certain documents 16 pages of transcription are sufficient to adapt a pre-trained recognition system.\n
\n\n\n
\n\n\n\n\n\n
\n
\n\n
\n
\n \n Neudecker, C.\n \n \n (1)\n \n \n
\n
\n
\n\n
\n
\n \n Pletschacher, S.\n \n \n (2)\n \n \n
\n
\n \n \n
\n\n\n
\n
\n\n \n \n \n \n \n \n The PAGE (Page Analysis and Ground-Truth Elements) Format Framework.\n \n \n \n \n\n\n \n Pletschacher, S.; and Antonacopoulos, A.\n\n\n \n\n\n\n In pages 257–260, August 2010. IEEE\n
\n\n
\n\n
\n\n
\n\n \n \n Paper\n \n \n\n \n \n doi\n \n \n\n \n link\n \n \n\n bibtex\n \n\n \n\n \n \n \n 1 download\n \n \n\n \n \n \n \n \n \n \n\n \n \n \n\n\n\n
\n
@inproceedings{pletschacher_page_2010,\n\ttitle = {The {PAGE} ({Page} {Analysis} and {Ground}-{Truth} {Elements}) {Format} {Framework}},\n\tisbn = {978-1-4244-7542-1},\n\turl = {http://ieeexplore.ieee.org/document/5597587/},\n\tdoi = {10.1109/ICPR.2010.72},\n\turldate = {2018-06-22},\n\tpublisher = {IEEE},\n\tauthor = {Pletschacher, Stefan and Antonacopoulos, Apostolos},\n\tmonth = aug,\n\tyear = {2010},\n\tpages = {257--260},\n}\n
\n
\n\n\n\n
\n\n\n\n\n\n
\n
\n\n
\n
\n \n Smith, B.\n \n \n (1)\n \n \n
\n
\n \n \n
\n
\n\n \n \n \n \n \n Digital performance : a history of new media in theater, dance, performance art, and installation.\n \n \n \n\n\n \n Dixon, S.; and Smith, B.\n\n\n \n\n\n\n of LeonardoThe MIT Press, Cambridge, Massachusetts, London, England, [Paperback edition] edition, 2015.\n
\n\n
\n\n
\n\n
\n\n \n\n \n\n \n link\n \n \n\n bibtex\n \n\n \n\n \n\n \n \n \n \n \n \n \n\n \n \n \n\n\n\n
\n
@book{dixon_digital_2015,\n\taddress = {Cambridge, Massachusetts, London, England},\n\tedition = {[Paperback edition]},\n\tseries = {Leonardo},\n\ttitle = {Digital performance : a history of new media in theater, dance, performance art, and installation},\n\tisbn = {978-0-262-52752-1},\n\tshorttitle = {Digital performance},\n\tpublisher = {The MIT Press},\n\tauthor = {Dixon, Steve and Smith, Barry},\n\tyear = {2015},\n}\n\n
\n
\n\n\n\n
\n\n\n\n\n\n
\n
\n\n
\n
\n \n Stamatopoulos, Ni\n \n \n (1)\n \n \n
\n
\n \n \n
\n
\n\n \n \n \n \n \n Goal-oriented performance evaluation methodology for page segmentation techniques.\n \n \n \n\n\n \n Stamatopoulos, Nikolaos; and Gatos, Basilis\n\n\n \n\n\n\n In
Proceedings of the 13th international conference on document analysis and recognition (ICDAR), pages 281–285. 2015.\n
\n\n
\n\n
\n\n
\n\n \n\n \n\n \n link\n \n \n\n bibtex\n \n\n \n\n \n\n \n \n \n \n \n \n \n\n \n \n \n\n\n\n
\n
@incollection{stamatopoulos_nikolaos_goal-oriented_2015,\n\ttitle = {Goal-oriented performance evaluation methodology for page segmentation techniques},\n\tbooktitle = {Proceedings of the 13th international conference on document analysis and recognition ({ICDAR})},\n\tauthor = {{Stamatopoulos, Nikolaos} and {Gatos, Basilis}},\n\tyear = {2015},\n\tpages = {281--285},\n}\n\n
\n
\n\n\n\n
\n\n\n\n\n\n
\n
\n\n
\n
\n \n Strauss, T.\n \n \n (1)\n \n \n
\n
\n
\n\n
\n
\n \n Strauß, T.\n \n \n (1)\n \n \n
\n
\n \n \n
\n
\n\n \n \n \n \n \n ICFHR2018 Competition on Automated Text Recognition on a READ Dataset.\n \n \n \n\n\n \n Strauß, T.; Leifert, G.; Labahn, R.; Hodel, T.; and Mühlberger, G.\n\n\n \n\n\n\n In
2018 16th International Conference on Frontiers in Handwriting Recognition (ICFHR), pages 477–482, August 2018. \n
\n\n
\n\n
\n\n
\n\n \n\n \n \n doi\n \n \n\n \n link\n \n \n\n bibtex\n \n\n \n \n \n abstract \n \n\n \n\n \n \n \n \n \n \n \n\n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n\n\n\n
\n
@inproceedings{straus_icfhr2018_2018,\n\ttitle = {{ICFHR2018} {Competition} on {Automated} {Text} {Recognition} on a {READ} {Dataset}},\n\tdoi = {10.1109/ICFHR-2018.2018.00089},\n\tabstract = {We summarize the results of a competition on Automated Text Recognition targeting the effective adaptation of recognition engines to essentially new data. The task consists in achieving a minimum character error rate on a previously unknown text corpus from which only a few pages are available for adjusting an already pre-trained recognition engine. This issue addresses a frequent application scenario where only a small amount of task-specific training data is available, because producing this data usually requires much effort. We present the results of five submission. They show that the task is a challenging issue but for certain documents 16 pages of transcription are sufficient to adapt a pre-trained recognition system.},\n\tbooktitle = {2018 16th {International} {Conference} on {Frontiers} in {Handwriting} {Recognition} ({ICFHR})},\n\tauthor = {Strauß, Tobias and Leifert, Gundram and Labahn, Roger and Hodel, Tobias and Mühlberger, Günter},\n\tmonth = aug,\n\tyear = {2018},\n\tkeywords = {Computational modeling, Data models, Optical imaging, Task analysis, Text recognition, Training, Training data, automated text recognition, fast adaptation, few shot learning, historical documents},\n\tpages = {477--482},\n}\n\n
\n
\n\n\n
\n We summarize the results of a competition on Automated Text Recognition targeting the effective adaptation of recognition engines to essentially new data. The task consists in achieving a minimum character error rate on a previously unknown text corpus from which only a few pages are available for adjusting an already pre-trained recognition engine. This issue addresses a frequent application scenario where only a small amount of task-specific training data is available, because producing this data usually requires much effort. We present the results of five submission. They show that the task is a challenging issue but for certain documents 16 pages of transcription are sufficient to adapt a pre-trained recognition system.\n
\n\n\n
\n\n\n\n\n\n
\n
\n\n
\n
\n \n Tomarchio, L.\n \n \n (1)\n \n \n
\n
\n
\n\n
\n
\n \n Tompkins, J.\n \n \n (1)\n \n \n
\n
\n
\n\n
\n
\n \n Weidemann, M.\n \n \n (1)\n \n \n
\n
\n
\n\n
\n
\n \n undefined\n \n \n (2)\n \n \n
\n
\n \n \n
\n
\n\n \n \n \n \n \n \n The Shape of Data in the Digital Humanities: Modeling Texts and Text-based Resources.\n \n \n \n \n\n\n \n Flanders, J.; and Jannidis, F.,\n editors.\n \n\n\n \n\n\n\n Routledge, Abingdon, Oxon ; New York, NY : Routledge, 2019. \\textbar Series: Digital research in the arts and humanities, 1 edition, November 2018.\n
\n\n
\n\n
\n\n
\n\n \n \n Paper\n \n \n\n \n \n doi\n \n \n\n \n link\n \n \n\n bibtex\n \n\n \n\n \n\n \n \n \n \n \n \n \n\n \n \n \n\n\n\n
\n
@book{flanders_shape_2018,\n\taddress = {Abingdon, Oxon ; New York, NY : Routledge, 2019. {\\textbar} Series: Digital research in the arts and humanities},\n\tedition = {1},\n\ttitle = {The {Shape} of {Data} in the {Digital} {Humanities}: {Modeling} {Texts} and {Text}-based {Resources}},\n\tisbn = {978-1-315-55294-1},\n\tshorttitle = {The {Shape} of {Data} in the {Digital} {Humanities}},\n\turl = {https://www.taylorfrancis.com/books/9781317016151},\n\tlanguage = {en},\n\turldate = {2020-01-14},\n\tpublisher = {Routledge},\n\teditor = {Flanders, Julia and Jannidis, Fotis},\n\tmonth = nov,\n\tyear = {2018},\n\tdoi = {10.4324/9781315552941},\n}\n\n
\n
\n\n\n\n
\n\n\n
\n
\n\n \n \n \n \n \n \n STTS Tag Table. Institut für Maschinelle Sprachverarbeitung. Universität Stuttgart.\n \n \n \n \n\n\n \n \n\n\n \n\n\n\n 1999.\n
\n\n
\n\n
\n\n
\n\n \n \n Paper\n \n \n\n \n\n \n link\n \n \n\n bibtex\n \n\n \n\n \n\n \n \n \n \n \n \n \n\n \n \n \n\n\n\n
\n
@misc{noauthor_stts_1999,\n\ttitle = {{STTS} {Tag} {Table}. {Institut} für {Maschinelle} {Sprachverarbeitung}. {Universität} {Stuttgart}},\n\tshorttitle = {{STTS}},\n\turl = {http://www.ims.uni-stuttgart.de/forschung/ressourcen/lexika/TagSets/stts-table.html},\n\turldate = {2014-07-29},\n\tjournal = {STTS Tag Table (1995/1999)},\n\tyear = {1999},\n}\n\n
\n
\n\n\n\n
\n\n\n\n\n\n
\n
\n\n\n\n\n