{"id":"https://openalex.org/W3167086437","doi":"https://doi.org/10.1145/3476887.3476910","title":"Mixed Model OCR Training on Historical Latin Script for Out-of-the-Box Recognition and Finetuning","display_name":"Mixed Model OCR Training on Historical Latin Script for Out-of-the-Box Recognition and Finetuning","publication_year":2021,"publication_date":"2021-09-05","ids":{"openalex":"https://openalex.org/W3167086437","doi":"https://doi.org/10.1145/3476887.3476910","mag":"3167086437"},"language":"en","primary_location":{"id":"doi:10.1145/3476887.3476910","is_oa":false,"landing_page_url":"https://doi.org/10.1145/3476887.3476910","pdf_url":null,"source":null,"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"The 6th International Workshop on Historical Document Imaging and Processing","raw_type":"proceedings-article"},"type":"preprint","indexed_in":["arxiv","crossref","datacite"],"open_access":{"is_oa":true,"oa_status":"green","oa_url":"https://arxiv.org/pdf/2106.07881","any_repository_has_fulltext":true},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5063938010","display_name":"Christian Reul","orcid":"https://orcid.org/0000-0002-1776-1469"},"institutions":[{"id":"https://openalex.org/I1321014770","display_name":"Association for Computing Machinery","ror":"https://ror.org/03wsadn68","country_code":"US","type":"nonprofit","lineage":["https://openalex.org/I1321014770"]},{"id":"https://openalex.org/I25974101","display_name":"University of W\u00fcrzburg","ror":"https://ror.org/00fbnyb24","country_code":"DE","type":"education","lineage":["https://openalex.org/I25974101"]}],"countries":["DE","US"],"is_corresponding":false,"raw_author_name":"Christian Reul","raw_affiliation_strings":["University of Wuerzburg, Germany","Association for Computing Machinery"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"University of Wuerzburg, Germany","institution_ids":["https://openalex.org/I25974101"]},{"raw_affiliation_string":"Association for Computing Machinery","institution_ids":["https://openalex.org/I1321014770"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5014007811","display_name":"Christoph Wick","orcid":"https://orcid.org/0000-0003-3958-6240"},"institutions":[{"id":"https://openalex.org/I4210116325","display_name":"Planet","ror":"https://ror.org/02ekwm539","country_code":"DE","type":"company","lineage":["https://openalex.org/I4210116325"]}],"countries":["DE"],"is_corresponding":false,"raw_author_name":"Christoph Wick","raw_affiliation_strings":["Planet AI GmbH, Germany"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Planet AI GmbH, Germany","institution_ids":["https://openalex.org/I4210116325"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5050598951","display_name":"Maximilian N\u00f6th","orcid":"https://orcid.org/0000-0003-1235-057X"},"institutions":[{"id":"https://openalex.org/I25974101","display_name":"University of W\u00fcrzburg","ror":"https://ror.org/00fbnyb24","country_code":"DE","type":"education","lineage":["https://openalex.org/I25974101"]}],"countries":["DE"],"is_corresponding":false,"raw_author_name":"Maximilian Noeth","raw_affiliation_strings":["University of Wuerzburg, Germany"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"University of Wuerzburg, Germany","institution_ids":["https://openalex.org/I25974101"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5103161292","display_name":"Andreas B\u00fcttner","orcid":"https://orcid.org/0000-0001-6870-3798"},"institutions":[{"id":"https://openalex.org/I25974101","display_name":"University of W\u00fcrzburg","ror":"https://ror.org/00fbnyb24","country_code":"DE","type":"education","lineage":["https://openalex.org/I25974101"]}],"countries":["DE"],"is_corresponding":false,"raw_author_name":"Andreas Buettner","raw_affiliation_strings":["University of Wuerzburg, Germany"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"University of Wuerzburg, Germany","institution_ids":["https://openalex.org/I25974101"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5055761907","display_name":"Maximilian Wehner","orcid":null},"institutions":[{"id":"https://openalex.org/I25974101","display_name":"University of W\u00fcrzburg","ror":"https://ror.org/00fbnyb24","country_code":"DE","type":"education","lineage":["https://openalex.org/I25974101"]}],"countries":["DE"],"is_corresponding":false,"raw_author_name":"Maximilian Wehner","raw_affiliation_strings":["University of Wuerzburg, Germany"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"University of Wuerzburg, Germany","institution_ids":["https://openalex.org/I25974101"]}]},{"author_position":"last","author":{"id":"https://openalex.org/A5086884211","display_name":"Uwe Springmann","orcid":null},"institutions":[{"id":"https://openalex.org/I3018771216","display_name":"LMU Klinikum","ror":"https://ror.org/02jet3w32","country_code":"DE","type":"healthcare","lineage":["https://openalex.org/I3018771216","https://openalex.org/I8204097"]},{"id":"https://openalex.org/I8204097","display_name":"Ludwig-Maximilians-Universit\u00e4t M\u00fcnchen","ror":"https://ror.org/05591te55","country_code":"DE","type":"education","lineage":["https://openalex.org/I8204097"]}],"countries":["DE"],"is_corresponding":false,"raw_author_name":"Uwe Springmann","raw_affiliation_strings":["LMU Munich, Germany"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"LMU Munich, Germany","institution_ids":["https://openalex.org/I3018771216","https://openalex.org/I8204097"]}]}],"institutions":[],"countries_distinct_count":2,"institutions_distinct_count":6,"corresponding_author_ids":[],"corresponding_institution_ids":[],"apc_list":null,"apc_paid":null,"fwci":0.0,"has_fulltext":true,"cited_by_count":0,"citation_normalized_percentile":{"value":0.05620877,"is_in_top_1_percent":false,"is_in_top_10_percent":false},"cited_by_percentile_year":null,"biblio":{"volume":null,"issue":null,"first_page":"7","last_page":"12"},"is_retracted":false,"is_paratext":false,"is_xpac":false,"primary_topic":{"id":"https://openalex.org/T10601","display_name":"Handwritten Text Recognition Techniques","score":0.9997000098228455,"subfield":{"id":"https://openalex.org/subfields/1707","display_name":"Computer Vision and Pattern Recognition"},"field":{"id":"https://openalex.org/fields/17","display_name":"Computer Science"},"domain":{"id":"https://openalex.org/domains/3","display_name":"Physical Sciences"}},"topics":[{"id":"https://openalex.org/T10601","display_name":"Handwritten Text Recognition Techniques","score":0.9997000098228455,"subfield":{"id":"https://openalex.org/subfields/1707","display_name":"Computer Vision and Pattern Recognition"},"field":{"id":"https://openalex.org/fields/17","display_name":"Computer Science"},"domain":{"id":"https://openalex.org/domains/3","display_name":"Physical Sciences"}},{"id":"https://openalex.org/T14339","display_name":"Image Processing and 3D Reconstruction","score":0.9642000198364258,"subfield":{"id":"https://openalex.org/subfields/1707","display_name":"Computer Vision and Pattern Recognition"},"field":{"id":"https://openalex.org/fields/17","display_name":"Computer Science"},"domain":{"id":"https://openalex.org/domains/3","display_name":"Physical Sciences"}},{"id":"https://openalex.org/T10181","display_name":"Natural Language Processing Techniques","score":0.9580000042915344,"subfield":{"id":"https://openalex.org/subfields/1702","display_name":"Artificial Intelligence"},"field":{"id":"https://openalex.org/fields/17","display_name":"Computer Science"},"domain":{"id":"https://openalex.org/domains/3","display_name":"Physical Sciences"}}],"keywords":[{"id":"https://openalex.org/keywords/computer-science","display_name":"Computer science","score":0.8432366847991943},{"id":"https://openalex.org/keywords/optical-character-recognition","display_name":"Optical character recognition","score":0.6775925755500793},{"id":"https://openalex.org/keywords/artificial-intelligence","display_name":"Artificial intelligence","score":0.5855444073677063},{"id":"https://openalex.org/keywords/typeface","display_name":"Typeface","score":0.5853914022445679},{"id":"https://openalex.org/keywords/preprocessor","display_name":"Preprocessor","score":0.5589797496795654},{"id":"https://openalex.org/keywords/construct","display_name":"Construct (python library)","score":0.5502710938453674},{"id":"https://openalex.org/keywords/font","display_name":"Font","score":0.5140272378921509},{"id":"https://openalex.org/keywords/natural-language-processing","display_name":"Natural language processing","score":0.49327152967453003},{"id":"https://openalex.org/keywords/variety","display_name":"Variety (cybernetics)","score":0.48129335045814514},{"id":"https://openalex.org/keywords/language-model","display_name":"Language model","score":0.45733481645584106},{"id":"https://openalex.org/keywords/speech-recognition","display_name":"Speech recognition","score":0.433773934841156},{"id":"https://openalex.org/keywords/data-pre-processing","display_name":"Data pre-processing","score":0.42209863662719727},{"id":"https://openalex.org/keywords/machine-learning","display_name":"Machine learning","score":0.40242743492126465},{"id":"https://openalex.org/keywords/image","display_name":"Image (mathematics)","score":0.18513870239257812},{"id":"https://openalex.org/keywords/programming-language","display_name":"Programming language","score":0.11840283870697021}],"concepts":[{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.8432366847991943},{"id":"https://openalex.org/C546480517","wikidata":"https://www.wikidata.org/wiki/Q167555","display_name":"Optical character recognition","level":3,"score":0.6775925755500793},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.5855444073677063},{"id":"https://openalex.org/C80797182","wikidata":"https://www.wikidata.org/wiki/Q17451","display_name":"Typeface","level":2,"score":0.5853914022445679},{"id":"https://openalex.org/C34736171","wikidata":"https://www.wikidata.org/wiki/Q918333","display_name":"Preprocessor","level":2,"score":0.5589797496795654},{"id":"https://openalex.org/C2780801425","wikidata":"https://www.wikidata.org/wiki/Q5164392","display_name":"Construct (python library)","level":2,"score":0.5502710938453674},{"id":"https://openalex.org/C2777737414","wikidata":"https://www.wikidata.org/wiki/Q4868296","display_name":"Font","level":2,"score":0.5140272378921509},{"id":"https://openalex.org/C204321447","wikidata":"https://www.wikidata.org/wiki/Q30642","display_name":"Natural language processing","level":1,"score":0.49327152967453003},{"id":"https://openalex.org/C136197465","wikidata":"https://www.wikidata.org/wiki/Q1729295","display_name":"Variety (cybernetics)","level":2,"score":0.48129335045814514},{"id":"https://openalex.org/C137293760","wikidata":"https://www.wikidata.org/wiki/Q3621696","display_name":"Language model","level":2,"score":0.45733481645584106},{"id":"https://openalex.org/C28490314","wikidata":"https://www.wikidata.org/wiki/Q189436","display_name":"Speech recognition","level":1,"score":0.433773934841156},{"id":"https://openalex.org/C10551718","wikidata":"https://www.wikidata.org/wiki/Q5227332","display_name":"Data pre-processing","level":2,"score":0.42209863662719727},{"id":"https://openalex.org/C119857082","wikidata":"https://www.wikidata.org/wiki/Q2539","display_name":"Machine learning","level":1,"score":0.40242743492126465},{"id":"https://openalex.org/C115961682","wikidata":"https://www.wikidata.org/wiki/Q860623","display_name":"Image (mathematics)","level":2,"score":0.18513870239257812},{"id":"https://openalex.org/C199360897","wikidata":"https://www.wikidata.org/wiki/Q9143","display_name":"Programming language","level":1,"score":0.11840283870697021},{"id":"https://openalex.org/C111919701","wikidata":"https://www.wikidata.org/wiki/Q9135","display_name":"Operating system","level":1,"score":0.0}],"mesh":[],"locations_count":4,"locations":[{"id":"doi:10.1145/3476887.3476910","is_oa":false,"landing_page_url":"https://doi.org/10.1145/3476887.3476910","pdf_url":null,"source":null,"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"The 6th International Workshop on Historical Document Imaging and Processing","raw_type":"proceedings-article"},{"id":"pmh:oai:arXiv.org:2106.07881","is_oa":true,"landing_page_url":"http://arxiv.org/abs/2106.07881","pdf_url":"https://arxiv.org/pdf/2106.07881","source":{"id":"https://openalex.org/S4306400194","display_name":"arXiv (Cornell University)","issn_l":null,"issn":null,"is_oa":true,"is_in_doaj":false,"is_core":false,"host_organization":"https://openalex.org/I205783295","host_organization_name":"Cornell University","host_organization_lineage":["https://openalex.org/I205783295"],"host_organization_lineage_names":[],"type":"repository"},"license":null,"license_id":null,"version":"submittedVersion","is_accepted":false,"is_published":false,"raw_source_name":null,"raw_type":"text"},{"id":"mag:3167086437","is_oa":true,"landing_page_url":"http://arxiv.org/pdf/2106.07881.pdf","pdf_url":null,"source":{"id":"https://openalex.org/S4306400194","display_name":"arXiv (Cornell University)","issn_l":null,"issn":null,"is_oa":true,"is_in_doaj":false,"is_core":false,"host_organization":"https://openalex.org/I205783295","host_organization_name":"Cornell University","host_organization_lineage":["https://openalex.org/I205783295"],"host_organization_lineage_names":[],"type":"repository"},"license":null,"license_id":null,"version":"submittedVersion","is_accepted":false,"is_published":false,"raw_source_name":"arXiv (Cornell University)","raw_type":null},{"id":"doi:10.48550/arxiv.2106.07881","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2106.07881","pdf_url":null,"source":{"id":"https://openalex.org/S4306400194","display_name":"arXiv (Cornell University)","issn_l":null,"issn":null,"is_oa":true,"is_in_doaj":false,"is_core":false,"host_organization":"https://openalex.org/I205783295","host_organization_name":"Cornell University","host_organization_lineage":["https://openalex.org/I205783295"],"host_organization_lineage_names":[],"type":"repository"},"license":null,"license_id":null,"version":null,"is_accepted":false,"is_published":null,"raw_source_name":null,"raw_type":"article"}],"best_oa_location":{"id":"pmh:oai:arXiv.org:2106.07881","is_oa":true,"landing_page_url":"http://arxiv.org/abs/2106.07881","pdf_url":"https://arxiv.org/pdf/2106.07881","source":{"id":"https://openalex.org/S4306400194","display_name":"arXiv (Cornell University)","issn_l":null,"issn":null,"is_oa":true,"is_in_doaj":false,"is_core":false,"host_organization":"https://openalex.org/I205783295","host_organization_name":"Cornell University","host_organization_lineage":["https://openalex.org/I205783295"],"host_organization_lineage_names":[],"type":"repository"},"license":null,"license_id":null,"version":"submittedVersion","is_accepted":false,"is_published":false,"raw_source_name":null,"raw_type":"text"},"sustainable_development_goals":[{"id":"https://metadata.un.org/sdg/4","display_name":"Quality Education","score":0.8199999928474426}],"awards":[],"funders":[],"has_content":{"pdf":true,"grobid_xml":true},"content_urls":{"pdf":"https://content.openalex.org/works/W3167086437.pdf","grobid_xml":"https://content.openalex.org/works/W3167086437.grobid-xml"},"referenced_works_count":24,"referenced_works":["https://openalex.org/W1444003893","https://openalex.org/W2060580591","https://openalex.org/W2088483389","https://openalex.org/W2099106019","https://openalex.org/W2127141656","https://openalex.org/W2149128132","https://openalex.org/W2289181176","https://openalex.org/W2402677432","https://openalex.org/W2444136538","https://openalex.org/W2731900166","https://openalex.org/W2768817781","https://openalex.org/W2785875661","https://openalex.org/W2789003796","https://openalex.org/W2895456681","https://openalex.org/W2951815760","https://openalex.org/W2963414181","https://openalex.org/W2973047034","https://openalex.org/W2980728898","https://openalex.org/W2981597140","https://openalex.org/W3012971703","https://openalex.org/W3013242266","https://openalex.org/W3045327706","https://openalex.org/W3097176321","https://openalex.org/W3164447479"],"related_works":["https://openalex.org/W3211052011","https://openalex.org/W2962936306","https://openalex.org/W3181798399","https://openalex.org/W2783228166","https://openalex.org/W2186058360","https://openalex.org/W2824294645","https://openalex.org/W2986227781","https://openalex.org/W2011627645","https://openalex.org/W2755791930","https://openalex.org/W2159564332","https://openalex.org/W2915511574","https://openalex.org/W1520995466","https://openalex.org/W3105746227","https://openalex.org/W2581835087","https://openalex.org/W2294413629","https://openalex.org/W142320376","https://openalex.org/W3047738520","https://openalex.org/W2974932290","https://openalex.org/W2803238307","https://openalex.org/W3112069427"],"abstract_inverted_index":{"In":[0,118],"order":[1],"to":[2,8,21,51,82,125,207,216,219,225,228,243],"apply":[3],"Optical":[4],"Character":[5,32],"Recognition":[6],"(OCR)":[7],"historical":[9],"printings":[10,55],"of":[11,54,73,77,89,109,173,184,210,214],"Latin":[12,199],"script":[13],"fully":[14],"automatically,":[15],"we":[16,42,105,120],"report":[17],"on":[18,69,144,157,164],"our":[19,203],"efforts":[20],"construct":[22],"a":[23,31,70,138,158,171,176,182,190,208],"widely-applicable":[24],"polyfont":[25,65],"recognition":[26],"model":[27,46,66,180,193,205,238],"yielding":[28],"text":[29],"with":[30,56,181],"Error":[33],"Rate":[34],"(CER)":[35],"around":[36],"2%":[37],"when":[38],"applied":[39],"out-of-the-box.":[40],"Moreover,":[41],"show":[43],"how":[44],"this":[45],"can":[47],"be":[48],"further":[49],"finetuned":[50],"specific":[52],"classes":[53],"little":[57],"manual":[58],"and":[59,91,93,99,116,130,150,223],"computational":[60],"effort.":[61],"The":[62],"mixed":[63,204,237],"or":[64],"is":[67,239],"trained":[68],"wide":[71],"variety":[72],"materials,":[74],"in":[75,170],"terms":[76],"age":[78],"(from":[79],"the":[80,83,103,127,153,231,244],"15th":[81],"19th":[84],"century),":[85],"typography":[86],"(various":[87],"types":[88],"Fraktur":[90],"Antiqua),":[92],"languages":[94],"(among":[95],"others,":[96],"German,":[97],"Latin,":[98],"French).":[100],"To":[101],"optimize":[102],"results":[104],"combined":[106],"established":[107],"techniques":[108],"OCR":[110],"training":[111,128,156,220,229],"like":[112],"pretraining,":[113],"data":[114,129,149],"augmentation,":[115],"voting.":[117],"addition,":[119],"used":[121,178],"various":[122],"preprocessing":[123],"methods":[124],"enrich":[126],"obtain":[131],"more":[132,160,191],"robust":[133],"models.":[134],"We":[135],"also":[136],"implemented":[137],"two-stage":[139],"approach":[140],"which":[141],"first":[142],"trains":[143],"all":[145],"available,":[146],"considerably":[147],"unbalanced":[148],"then":[151],"refines":[152],"output":[154],"by":[155,186],"selected":[159],"balanced":[161],"subset.":[162],"Evaluations":[163],"29":[165],"previously":[166],"unseen":[167,196],"books":[168,200],"resulted":[169],"CER":[172,183,209],"1.73%,":[174],"outperforming":[175],"widely":[177],"standard":[179,233],"2.84%":[185],"almost":[187],"40%.":[188],"Training":[189],"specialized":[192],"for":[194],"some":[195],"Early":[197],"Modern":[198],"starting":[201],"from":[202,221,230],"led":[206],"1.47%,":[211],"an":[212],"improvement":[213],"up":[215,224],"50%":[217],"compared":[218,227],"scratch":[222],"30%":[226],"aforementioned":[232],"model.":[234],"Our":[235],"new":[236],"made":[240],"openly":[241],"available":[242],"community.":[245]},"counts_by_year":[],"updated_date":"2026-06-11T09:08:48.828518","created_date":"2025-10-10T00:00:00"}