{"id":"https://openalex.org/W4313172969","doi":"https://doi.org/10.1109/icpr56361.2022.9956616","title":"Augmenting Vision Language Pretraining by Learning Codebook with Visual Semantics","display_name":"Augmenting Vision Language Pretraining by Learning Codebook with Visual Semantics","publication_year":2022,"publication_date":"2022-08-21","ids":{"openalex":"https://openalex.org/W4313172969","doi":"https://doi.org/10.1109/icpr56361.2022.9956616"},"language":"en","primary_location":{"id":"doi:10.1109/icpr56361.2022.9956616","is_oa":false,"landing_page_url":"https://doi.org/10.1109/icpr56361.2022.9956616","pdf_url":null,"source":{"id":"https://openalex.org/S4363607731","display_name":"2022 26th International Conference on Pattern Recognition (ICPR)","issn_l":null,"issn":null,"is_oa":false,"is_in_doaj":false,"is_core":false,"host_organization":null,"host_organization_name":null,"host_organization_lineage":[],"host_organization_lineage_names":[],"type":"conference"},"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"2022 26th International Conference on Pattern Recognition (ICPR)","raw_type":"proceedings-article"},"type":"article","indexed_in":["crossref"],"open_access":{"is_oa":false,"oa_status":"closed","oa_url":null,"any_repository_has_fulltext":false},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5025487788","display_name":"Xiaoyuan Guo","orcid":"https://orcid.org/0000-0002-2904-0386"},"institutions":[{"id":"https://openalex.org/I150468666","display_name":"Emory University","ror":"https://ror.org/03czfpz43","country_code":"US","type":"education","lineage":["https://openalex.org/I150468666"]}],"countries":["US"],"is_corresponding":true,"raw_author_name":"Xiaoyuan Guo","raw_affiliation_strings":["Emory University,Department of Computer Science,GA,USA","Department of Computer Science, Emory University, GA, USA"],"affiliations":[{"raw_affiliation_string":"Emory University,Department of Computer Science,GA,USA","institution_ids":["https://openalex.org/I150468666"]},{"raw_affiliation_string":"Department of Computer Science, Emory University, GA, USA","institution_ids":["https://openalex.org/I150468666"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5101849949","display_name":"Jiali Duan","orcid":"https://orcid.org/0000-0002-9126-0531"},"institutions":[{"id":"https://openalex.org/I1174212","display_name":"University of Southern California","ror":"https://ror.org/03taz7m60","country_code":"US","type":"education","lineage":["https://openalex.org/I1174212"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Jiali Duan","raw_affiliation_strings":["University of Southern California,Ming Hsieh Department of Electrical and Computer Engineering,CA,USA","Ming Hsieh Department of Electrical and Computer Engineering, University of Southern California, CA, USA"],"affiliations":[{"raw_affiliation_string":"University of Southern California,Ming Hsieh Department of Electrical and Computer Engineering,CA,USA","institution_ids":["https://openalex.org/I1174212"]},{"raw_affiliation_string":"Ming Hsieh Department of Electrical and Computer Engineering, University of Southern California, CA, USA","institution_ids":["https://openalex.org/I1174212"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5001082656","display_name":"C.\u2010C. Jay Kuo","orcid":"https://orcid.org/0000-0001-9474-5035"},"institutions":[{"id":"https://openalex.org/I1174212","display_name":"University of Southern California","ror":"https://ror.org/03taz7m60","country_code":"US","type":"education","lineage":["https://openalex.org/I1174212"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"C.-C. Jay Kuo","raw_affiliation_strings":["University of Southern California,Ming Hsieh Department of Electrical and Computer Engineering,CA,USA","Ming Hsieh Department of Electrical and Computer Engineering, University of Southern California, CA, USA"],"affiliations":[{"raw_affiliation_string":"University of Southern California,Ming Hsieh Department of Electrical and Computer Engineering,CA,USA","institution_ids":["https://openalex.org/I1174212"]},{"raw_affiliation_string":"Ming Hsieh Department of Electrical and Computer Engineering, University of Southern California, CA, USA","institution_ids":["https://openalex.org/I1174212"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5076075666","display_name":"Judy Wawira Gichoya","orcid":"https://orcid.org/0000-0002-1097-316X"},"institutions":[{"id":"https://openalex.org/I150468666","display_name":"Emory University","ror":"https://ror.org/03czfpz43","country_code":"US","type":"education","lineage":["https://openalex.org/I150468666"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Judy Wawira Gichoya","raw_affiliation_strings":["Emory University,Department of Radiology and Imaging Sciences,GA,USA","Department of Radiology and Imaging Sciences, Emory University, GA, USA"],"affiliations":[{"raw_affiliation_string":"Emory University,Department of Radiology and Imaging Sciences,GA,USA","institution_ids":["https://openalex.org/I150468666"]},{"raw_affiliation_string":"Department of Radiology and Imaging Sciences, Emory University, GA, USA","institution_ids":["https://openalex.org/I150468666"]}]},{"author_position":"last","author":{"id":"https://openalex.org/A5024679224","display_name":"Imon Banerjee","orcid":"https://orcid.org/0000-0002-3327-8004"},"institutions":[{"id":"https://openalex.org/I2802423016","display_name":"WinnMed","ror":"https://ror.org/02s47w807","country_code":"US","type":"nonprofit","lineage":["https://openalex.org/I2802423016"]},{"id":"https://openalex.org/I4210146710","display_name":"Mayo Clinic in Florida","ror":"https://ror.org/03zzw1w08","country_code":"US","type":"healthcare","lineage":["https://openalex.org/I1330342723","https://openalex.org/I4210146710"]},{"id":"https://openalex.org/I55732556","display_name":"Arizona State University","ror":"https://ror.org/03efmqc40","country_code":"US","type":"education","lineage":["https://openalex.org/I55732556"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Imon Banerjee","raw_affiliation_strings":["Arizona State University,School of Computing, Informatics, and Decision Systems Engineering,AZ,USA","Department of Radiology, Mayo Clinic, AZ, USA","School of Computing, Informatics, and Decision Systems Engineering, Arizona State University, AZ, USA"],"affiliations":[{"raw_affiliation_string":"Arizona State University,School of Computing, Informatics, and Decision Systems Engineering,AZ,USA","institution_ids":["https://openalex.org/I55732556"]},{"raw_affiliation_string":"Department of Radiology, Mayo Clinic, AZ, USA","institution_ids":["https://openalex.org/I4210146710","https://openalex.org/I2802423016"]},{"raw_affiliation_string":"School of Computing, Informatics, and Decision Systems Engineering, Arizona State University, AZ, USA","institution_ids":["https://openalex.org/I55732556"]}]}],"institutions":[],"countries_distinct_count":1,"institutions_distinct_count":5,"corresponding_author_ids":["https://openalex.org/A5025487788"],"corresponding_institution_ids":["https://openalex.org/I150468666"],"apc_list":null,"apc_paid":null,"fwci":0.06,"has_fulltext":false,"cited_by_count":1,"citation_normalized_percentile":{"value":0.33352346,"is_in_top_1_percent":false,"is_in_top_10_percent":false},"cited_by_percentile_year":{"min":89,"max":94},"biblio":{"volume":null,"issue":null,"first_page":"4779","last_page":"4785"},"is_retracted":false,"is_paratext":false,"is_xpac":false,"primary_topic":{"id":"https://openalex.org/T11714","display_name":"Multimodal Machine Learning Applications","score":1.0,"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/T11714","display_name":"Multimodal Machine Learning Applications","score":1.0,"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/T11307","display_name":"Domain Adaptation and Few-Shot Learning","score":0.9969000220298767,"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"}},{"id":"https://openalex.org/T10812","display_name":"Human Pose and Action Recognition","score":0.9918000102043152,"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"}}],"keywords":[{"id":"https://openalex.org/keywords/computer-science","display_name":"Computer science","score":0.7902272343635559},{"id":"https://openalex.org/keywords/codebook","display_name":"Codebook","score":0.7812013626098633},{"id":"https://openalex.org/keywords/artificial-intelligence","display_name":"Artificial intelligence","score":0.6770712733268738},{"id":"https://openalex.org/keywords/natural-language-processing","display_name":"Natural language processing","score":0.651077389717102},{"id":"https://openalex.org/keywords/semantics","display_name":"Semantics (computer science)","score":0.5683547258377075},{"id":"https://openalex.org/keywords/vocabulary","display_name":"Vocabulary","score":0.42252302169799805},{"id":"https://openalex.org/keywords/linguistics","display_name":"Linguistics","score":0.1687069535255432},{"id":"https://openalex.org/keywords/programming-language","display_name":"Programming language","score":0.11788532137870789}],"concepts":[{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.7902272343635559},{"id":"https://openalex.org/C127759330","wikidata":"https://www.wikidata.org/wiki/Q637416","display_name":"Codebook","level":2,"score":0.7812013626098633},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.6770712733268738},{"id":"https://openalex.org/C204321447","wikidata":"https://www.wikidata.org/wiki/Q30642","display_name":"Natural language processing","level":1,"score":0.651077389717102},{"id":"https://openalex.org/C184337299","wikidata":"https://www.wikidata.org/wiki/Q1437428","display_name":"Semantics (computer science)","level":2,"score":0.5683547258377075},{"id":"https://openalex.org/C2777601683","wikidata":"https://www.wikidata.org/wiki/Q6499736","display_name":"Vocabulary","level":2,"score":0.42252302169799805},{"id":"https://openalex.org/C41895202","wikidata":"https://www.wikidata.org/wiki/Q8162","display_name":"Linguistics","level":1,"score":0.1687069535255432},{"id":"https://openalex.org/C199360897","wikidata":"https://www.wikidata.org/wiki/Q9143","display_name":"Programming language","level":1,"score":0.11788532137870789},{"id":"https://openalex.org/C138885662","wikidata":"https://www.wikidata.org/wiki/Q5891","display_name":"Philosophy","level":0,"score":0.0}],"mesh":[],"locations_count":1,"locations":[{"id":"doi:10.1109/icpr56361.2022.9956616","is_oa":false,"landing_page_url":"https://doi.org/10.1109/icpr56361.2022.9956616","pdf_url":null,"source":{"id":"https://openalex.org/S4363607731","display_name":"2022 26th International Conference on Pattern Recognition (ICPR)","issn_l":null,"issn":null,"is_oa":false,"is_in_doaj":false,"is_core":false,"host_organization":null,"host_organization_name":null,"host_organization_lineage":[],"host_organization_lineage_names":[],"type":"conference"},"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"2022 26th International Conference on Pattern Recognition (ICPR)","raw_type":"proceedings-article"}],"best_oa_location":null,"sustainable_development_goals":[{"display_name":"Quality Education","id":"https://metadata.un.org/sdg/4","score":0.8100000023841858}],"awards":[],"funders":[],"has_content":{"grobid_xml":false,"pdf":false},"content_urls":null,"referenced_works_count":72,"referenced_works":["https://openalex.org/W1773149199","https://openalex.org/W1861492603","https://openalex.org/W2109586012","https://openalex.org/W2277195237","https://openalex.org/W2547875792","https://openalex.org/W2560730294","https://openalex.org/W2752796333","https://openalex.org/W2886641317","https://openalex.org/W2896457183","https://openalex.org/W2908510526","https://openalex.org/W2912371042","https://openalex.org/W2962858109","https://openalex.org/W2963530300","https://openalex.org/W2963799213","https://openalex.org/W2966715458","https://openalex.org/W2968124245","https://openalex.org/W2969876226","https://openalex.org/W2970231061","https://openalex.org/W2981851019","https://openalex.org/W2998356391","https://openalex.org/W3001555892","https://openalex.org/W3034727271","https://openalex.org/W3035688398","https://openalex.org/W3090449556","https://openalex.org/W3091588028","https://openalex.org/W3094502228","https://openalex.org/W3126337491","https://openalex.org/W3126792443","https://openalex.org/W3129576130","https://openalex.org/W3166396011","https://openalex.org/W3167118264","https://openalex.org/W3170863103","https://openalex.org/W3173220247","https://openalex.org/W3177224328","https://openalex.org/W3179883066","https://openalex.org/W3184735396","https://openalex.org/W3184784418","https://openalex.org/W3215434919","https://openalex.org/W3217059257","https://openalex.org/W4292779060","https://openalex.org/W4297808394","https://openalex.org/W4312460555","https://openalex.org/W4312877428","https://openalex.org/W4313156423","https://openalex.org/W6639102338","https://openalex.org/W6676497082","https://openalex.org/W6729448088","https://openalex.org/W6755207826","https://openalex.org/W6755777632","https://openalex.org/W6757817989","https://openalex.org/W6758704467","https://openalex.org/W6766904570","https://openalex.org/W6767211374","https://openalex.org/W6767279747","https://openalex.org/W6773248631","https://openalex.org/W6775970589","https://openalex.org/W6776721752","https://openalex.org/W6778883912","https://openalex.org/W6779473860","https://openalex.org/W6784333009","https://openalex.org/W6787572754","https://openalex.org/W6789753369","https://openalex.org/W6789909235","https://openalex.org/W6790019176","https://openalex.org/W6790978476","https://openalex.org/W6791353385","https://openalex.org/W6796761347","https://openalex.org/W6798805250","https://openalex.org/W6803870738","https://openalex.org/W6804229649","https://openalex.org/W6805349323","https://openalex.org/W6844194202"],"related_works":["https://openalex.org/W2293149949","https://openalex.org/W2026099691","https://openalex.org/W4284672201","https://openalex.org/W2377486419","https://openalex.org/W2017956276","https://openalex.org/W2736714427","https://openalex.org/W2950156284","https://openalex.org/W2943202426","https://openalex.org/W2163679795","https://openalex.org/W3192589309"],"abstract_inverted_index":{"Language":[0,85],"modality":[1,24],"within":[2],"the":[3,15,33,49,95,99,110],"vision":[4,40],"language":[5,16,42,91],"pre-training":[6],"framework":[7],"is":[8,25],"innately":[9],"discretized,":[10],"endowing":[11],"each":[12,59],"word":[13],"in":[14],"vocabulary":[17],"a":[18,55,62,81,105],"semantic":[19],"meaning.":[20],"In":[21],"contrast,":[22],"visual":[23,50,60,69],"inherently":[26],"continuous":[27],"and":[28,41],"high-dimensional,":[29],"which":[30,87,103],"potentially":[31],"prohibits":[32],"alignment":[34],"as":[35,37,71],"well":[36],"fusion":[38],"between":[39],"modalities.":[43],"We":[44,64],"therefore":[45],"propose":[46],"to":[47],"\"discretize\"":[48],"representation":[51],"by":[52],"joint":[53],"learning":[54],"codebook":[56],"that":[57],"imbues":[58],"token":[61],"semantic.":[63],"then":[65],"utilize":[66],"these":[67],"discretized":[68],"semantics":[70],"self-supervised":[72],"ground-truths":[73],"for":[74,90],"building":[75],"our":[76,113],"Masked":[77,84],"Image":[78],"Modeling":[79,86],"objective,":[80],"counterpart":[82],"of":[83,101,112],"proves":[88],"successful":[89],"models.":[92],"To":[93],"optimize":[94],"codebook,":[96],"we":[97],"extend":[98],"formulation":[100],"VQ-VAE":[102],"gives":[104],"theoretic":[106],"guarantee.":[107],"Experiments":[108],"validate":[109],"effectiveness":[111],"approach":[114],"across":[115],"common":[116],"vision-language":[117],"benchmarks.":[118]},"counts_by_year":[{"year":2023,"cited_by_count":1}],"updated_date":"2025-11-06T03:46:38.306776","created_date":"2025-10-10T00:00:00"}