{"id":"https://openalex.org/W4417281277","doi":"https://doi.org/10.1109/iccv51701.2025.02265","title":"Large-Scale Pre-Training for Grounded Video Caption Generation","display_name":"Large-Scale Pre-Training for Grounded Video Caption Generation","publication_year":2025,"publication_date":"2025-10-19","ids":{"openalex":"https://openalex.org/W4417281277","doi":"https://doi.org/10.1109/iccv51701.2025.02265"},"language":"en","primary_location":{"id":"doi:10.1109/iccv51701.2025.02265","is_oa":false,"landing_page_url":"https://doi.org/10.1109/iccv51701.2025.02265","pdf_url":null,"source":null,"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"2025 IEEE/CVF International Conference on Computer Vision (ICCV)","raw_type":"proceedings-article"},"type":"conference-paper","indexed_in":["arxiv","crossref","datacite"],"open_access":{"is_oa":true,"oa_status":"green","oa_url":"https://arxiv.org/pdf/2503.10781","any_repository_has_fulltext":true},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5003465914","display_name":"Evangelos Kazakos","orcid":"https://orcid.org/0000-0002-3045-720X"},"institutions":[{"id":"https://openalex.org/I4210152232","display_name":"Institute of Informatics of the Slovak Academy of Sciences","ror":"https://ror.org/04jgqpc26","country_code":"SK","type":"facility","lineage":["https://openalex.org/I207624831","https://openalex.org/I4210152232"]},{"id":"https://openalex.org/I44504214","display_name":"Czech Technical University in Prague","ror":"https://ror.org/03kqpb082","country_code":"CZ","type":"education","lineage":["https://openalex.org/I44504214"]}],"countries":["CZ","SK"],"is_corresponding":false,"raw_author_name":"Evangelos Kazakos","raw_affiliation_strings":["Czech Institute of Informatics, Robotics and Cybernetics at the Czech Technical University in Prague"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Czech Institute of Informatics, Robotics and Cybernetics at the Czech Technical University in Prague","institution_ids":["https://openalex.org/I4210152232","https://openalex.org/I44504214"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5109890544","display_name":"Cordelia Schmid","orcid":null},"institutions":[{"id":"https://openalex.org/I4210103330","display_name":"\u00c9cole Normale Sup\u00e9rieure","ror":"https://ror.org/01dp7jr64","country_code":"BI","type":"education","lineage":["https://openalex.org/I4210103330"]}],"countries":["BI"],"is_corresponding":false,"raw_author_name":"Cordelia Schmid","raw_affiliation_strings":["PSL Research University,Inria, © Ecole normale sup©erieure, CNRS"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"PSL Research University,Inria, © Ecole normale sup©erieure, CNRS","institution_ids":["https://openalex.org/I4210103330"]}]},{"author_position":"last","author":{"id":"https://openalex.org/A5025398783","display_name":"Josef \u0160ivic","orcid":"https://orcid.org/0000-0002-2554-5301"},"institutions":[{"id":"https://openalex.org/I4210152232","display_name":"Institute of Informatics of the Slovak Academy of Sciences","ror":"https://ror.org/04jgqpc26","country_code":"SK","type":"facility","lineage":["https://openalex.org/I207624831","https://openalex.org/I4210152232"]},{"id":"https://openalex.org/I44504214","display_name":"Czech Technical University in Prague","ror":"https://ror.org/03kqpb082","country_code":"CZ","type":"education","lineage":["https://openalex.org/I44504214"]}],"countries":["CZ","SK"],"is_corresponding":false,"raw_author_name":"Josef Sivic","raw_affiliation_strings":["Czech Institute of Informatics, Robotics and Cybernetics at the Czech Technical University in Prague"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Czech Institute of Informatics, Robotics and Cybernetics at the Czech Technical University in Prague","institution_ids":["https://openalex.org/I4210152232","https://openalex.org/I44504214"]}]}],"institutions":[],"countries_distinct_count":3,"institutions_distinct_count":3,"corresponding_author_ids":[],"corresponding_institution_ids":[],"apc_list":null,"apc_paid":null,"fwci":null,"has_fulltext":false,"cited_by_count":0,"citation_normalized_percentile":null,"cited_by_percentile_year":null,"biblio":{"volume":null,"issue":null,"first_page":"24434","last_page":"24444"},"is_retracted":false,"is_paratext":false,"is_xpac":false,"primary_topic":{"id":"https://openalex.org/T11714","display_name":"Multimodal Machine Learning Applications","score":0.98089998960495,"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":0.98089998960495,"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/T10775","display_name":"Generative Adversarial Networks and Image Synthesis","score":0.009200000204145908,"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/T10812","display_name":"Human Pose and Action Recognition","score":0.0015999999595806003,"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/closed-captioning","display_name":"Closed captioning","score":0.8267999887466431},{"id":"https://openalex.org/keywords/annotation","display_name":"Annotation","score":0.6876999735832214},{"id":"https://openalex.org/keywords/bounding-overwatch","display_name":"Bounding overwatch","score":0.6460999846458435},{"id":"https://openalex.org/keywords/key","display_name":"Key (lock)","score":0.5013999938964844},{"id":"https://openalex.org/keywords/construct","display_name":"Construct (python library)","score":0.451200008392334},{"id":"https://openalex.org/keywords/object","display_name":"Object (grammar)","score":0.4375},{"id":"https://openalex.org/keywords/minimum-bounding-box","display_name":"Minimum bounding box","score":0.42899999022483826}],"concepts":[{"id":"https://openalex.org/C157657479","wikidata":"https://www.wikidata.org/wiki/Q2367247","display_name":"Closed captioning","level":3,"score":0.8267999887466431},{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.8014000058174133},{"id":"https://openalex.org/C2776321320","wikidata":"https://www.wikidata.org/wiki/Q857525","display_name":"Annotation","level":2,"score":0.6876999735832214},{"id":"https://openalex.org/C63584917","wikidata":"https://www.wikidata.org/wiki/Q333286","display_name":"Bounding overwatch","level":2,"score":0.6460999846458435},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.5929999947547913},{"id":"https://openalex.org/C26517878","wikidata":"https://www.wikidata.org/wiki/Q228039","display_name":"Key (lock)","level":2,"score":0.5013999938964844},{"id":"https://openalex.org/C2780801425","wikidata":"https://www.wikidata.org/wiki/Q5164392","display_name":"Construct (python library)","level":2,"score":0.451200008392334},{"id":"https://openalex.org/C2781238097","wikidata":"https://www.wikidata.org/wiki/Q175026","display_name":"Object (grammar)","level":2,"score":0.4375},{"id":"https://openalex.org/C147037132","wikidata":"https://www.wikidata.org/wiki/Q6865426","display_name":"Minimum bounding box","level":3,"score":0.42899999022483826},{"id":"https://openalex.org/C204321447","wikidata":"https://www.wikidata.org/wiki/Q30642","display_name":"Natural language processing","level":1,"score":0.3961000144481659},{"id":"https://openalex.org/C2780009758","wikidata":"https://www.wikidata.org/wiki/Q6804172","display_name":"Measure (data warehouse)","level":2,"score":0.38929998874664307},{"id":"https://openalex.org/C23123220","wikidata":"https://www.wikidata.org/wiki/Q816826","display_name":"Information retrieval","level":1,"score":0.3725999891757965},{"id":"https://openalex.org/C2776760102","wikidata":"https://www.wikidata.org/wiki/Q5139990","display_name":"Code (set theory)","level":3,"score":0.3384000062942505},{"id":"https://openalex.org/C156325361","wikidata":"https://www.wikidata.org/wiki/Q1152864","display_name":"Grounded theory","level":3,"score":0.3312000036239624},{"id":"https://openalex.org/C177212765","wikidata":"https://www.wikidata.org/wiki/Q627335","display_name":"Workflow","level":2,"score":0.30649998784065247},{"id":"https://openalex.org/C119857082","wikidata":"https://www.wikidata.org/wiki/Q2539","display_name":"Machine learning","level":1,"score":0.29190000891685486},{"id":"https://openalex.org/C157486923","wikidata":"https://www.wikidata.org/wiki/Q1376436","display_name":"String (physics)","level":2,"score":0.290800005197525},{"id":"https://openalex.org/C137293760","wikidata":"https://www.wikidata.org/wiki/Q3621696","display_name":"Language model","level":2,"score":0.2890999913215637},{"id":"https://openalex.org/C31972630","wikidata":"https://www.wikidata.org/wiki/Q844240","display_name":"Computer vision","level":1,"score":0.2793999910354614},{"id":"https://openalex.org/C2776151529","wikidata":"https://www.wikidata.org/wiki/Q3045304","display_name":"Object detection","level":3,"score":0.26269999146461487},{"id":"https://openalex.org/C98045186","wikidata":"https://www.wikidata.org/wiki/Q205663","display_name":"Process (computing)","level":2,"score":0.2524999976158142}],"mesh":[],"locations_count":3,"locations":[{"id":"doi:10.1109/iccv51701.2025.02265","is_oa":false,"landing_page_url":"https://doi.org/10.1109/iccv51701.2025.02265","pdf_url":null,"source":null,"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"2025 IEEE/CVF International Conference on Computer Vision (ICCV)","raw_type":"proceedings-article"},{"id":"pmh:oai:arXiv.org:2503.10781","is_oa":true,"landing_page_url":"http://arxiv.org/abs/2503.10781","pdf_url":"https://arxiv.org/pdf/2503.10781","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":"doi:10.48550/arxiv.2503.10781","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2503.10781","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":"Preprint"}],"best_oa_location":{"id":"pmh:oai:arXiv.org:2503.10781","is_oa":true,"landing_page_url":"http://arxiv.org/abs/2503.10781","pdf_url":"https://arxiv.org/pdf/2503.10781","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":[],"awards":[{"id":"https://openalex.org/G1908239485","display_name":null,"funder_award_id":"ANR-21-FAI1-0002-01","funder_id":"https://openalex.org/F7400742617","funder_display_name":"Vermont Agency of Natural Resources"}],"funders":[{"id":"https://openalex.org/F4320311039","display_name":"Ministry of Higher Education and Scientific Research","ror":"https://ror.org/00kab6t91"},{"id":"https://openalex.org/F4320320300","display_name":"European Commission","ror":"https://ror.org/00k4n6c32"},{"id":"https://openalex.org/F4320338335","display_name":"H2020 European Research Council","ror":"https://ror.org/0472cxd90"},{"id":"https://openalex.org/F7400742617","display_name":"Vermont Agency of Natural Resources","ror":null}],"has_content":{"pdf":false,"grobid_xml":false},"content_urls":null,"referenced_works_count":0,"referenced_works":[],"related_works":[],"abstract_inverted_index":{"We":[0,28,55,71],"propose":[1],"a":[2,36,65,74],"novel":[3],"approach":[4,58,134],"for":[5],"captioning":[6],"and":[7,52,82,100,151,176,187],"object":[8],"grounding":[9],"in":[10,15,20],"video,":[11],"where":[12],"the":[13,16,21,30,60,84,139,147,157,171,178],"objects":[14],"caption":[17],"are":[18,189],"grounded":[19,45,103],"video":[22],"via":[23],"temporally":[24,50],"dense":[25,51,101],"bounding":[26,47,104],"boxes.":[27,105],"introduce":[29,73,90],"following":[31],"contributions.":[32],"First,":[33],"we":[34,89,130],"present":[35],"large-scale":[37,66],"automatic":[38],"annotation":[39],"method":[40],"that":[41,132],"aggregates":[42],"frame-level":[43],"captions":[44,99],"with":[46,96],"boxes":[48],"into":[49],"consistent":[53],"annotations.":[54],"apply":[56],"this":[57,113,124],"on":[59,86,112,123,138,146,161,170],"HowTo100M":[61],"dataset":[62,92,166,175,186],"to":[63,109,119],"construct":[64],"pre-training":[67,160],"dataset,":[68,142],"named":[69],"HowToGround1M.":[70,87],"also":[72],"Grounded":[75],"Video":[76],"Caption":[77],"Generation":[78],"model,":[79],"dubbed":[80],"GROVE,":[81],"pre-train":[83],"model":[85,122],"Second,":[88],"iGround--a":[91],"of":[93,159,182],"3513":[94],"videos":[95],"manually":[97,172],"annotated":[98,164,173],"spatio-temporally":[102],"This":[106],"allows":[107],"us":[108],"measure":[110],"progress":[111],"challenging":[114],"problem,":[115],"as":[116,118,143,145],"well":[117,144],"fine-tune":[120],"our":[121,133,162,183],"small-scale":[125],"but":[126],"high-quality":[127],"data.":[128],"Third,":[129],"demonstrate":[131,156],"achieves":[135],"state-of-the-art":[136],"results":[137],"proposed":[140],"iGround":[141,174],"VidSTG,":[148],"ActivityNet-Entities,":[149],"GroundingYouTube,":[150],"YouCook-Interactions":[152],"datasets.":[153],"Our":[154],"ablations":[155],"importance":[158],"automatically":[163],"HowToGround1M":[165],"followed":[167],"by":[168],"fine-tuning":[169],"validate":[177],"key":[179],"technical":[180],"contributions":[181],"model.":[184],"The":[185],"code":[188],"available":[190],"at":[191],"https://ekazakos.github.io/grounded_video_caption_generation/.":[192]},"counts_by_year":[],"updated_date":"2026-07-14T23:27:15.235271","created_date":"2025-10-10T00:00:00"}