{"id":"https://openalex.org/W4312793544","doi":"https://doi.org/10.1109/icpr56361.2022.9956241","title":"CAVAN: Commonsense Knowledge Anchored Video Captioning","display_name":"CAVAN: Commonsense Knowledge Anchored Video Captioning","publication_year":2022,"publication_date":"2022-08-21","ids":{"openalex":"https://openalex.org/W4312793544","doi":"https://doi.org/10.1109/icpr56361.2022.9956241"},"language":"en","primary_location":{"id":"doi:10.1109/icpr56361.2022.9956241","is_oa":false,"landing_page_url":"https://doi.org/10.1109/icpr56361.2022.9956241","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/A5070723020","display_name":"Huiliang Shao","orcid":null},"institutions":[{"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":true,"raw_author_name":"Huiliang Shao","raw_affiliation_strings":["Arizona State University,School of Electrical, Computer and Energy Engineering"],"affiliations":[{"raw_affiliation_string":"Arizona State University,School of Electrical, Computer and Energy Engineering","institution_ids":["https://openalex.org/I55732556"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5065602294","display_name":"Zhiyuan Fang","orcid":"https://orcid.org/0000-0003-1186-5221"},"institutions":[{"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":"Zhiyuan Fang","raw_affiliation_strings":["Arizona State University,School of Computing and Augmented Intelligence"],"affiliations":[{"raw_affiliation_string":"Arizona State University,School of Computing and Augmented Intelligence","institution_ids":["https://openalex.org/I55732556"]}]},{"author_position":"last","author":{"id":"https://openalex.org/A5002278578","display_name":"Yezhou Yang","orcid":"https://orcid.org/0000-0003-0126-8976"},"institutions":[{"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":"Yezhou Yang","raw_affiliation_strings":["Arizona State University,School of Computing and Augmented Intelligence"],"affiliations":[{"raw_affiliation_string":"Arizona State University,School of Computing and Augmented Intelligence","institution_ids":["https://openalex.org/I55732556"]}]}],"institutions":[],"countries_distinct_count":1,"institutions_distinct_count":3,"corresponding_author_ids":["https://openalex.org/A5070723020"],"corresponding_institution_ids":["https://openalex.org/I55732556"],"apc_list":null,"apc_paid":null,"fwci":0.1799,"has_fulltext":false,"cited_by_count":3,"citation_normalized_percentile":{"value":0.53122772,"is_in_top_1_percent":false,"is_in_top_10_percent":false},"cited_by_percentile_year":{"min":89,"max":96},"biblio":{"volume":null,"issue":null,"first_page":"4095","last_page":"4102"},"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/T10812","display_name":"Human Pose and Action Recognition","score":0.9965000152587891,"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/T11439","display_name":"Video Analysis and Summarization","score":0.9915000200271606,"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":0.9517967104911804},{"id":"https://openalex.org/keywords/computer-science","display_name":"Computer science","score":0.8416991233825684},{"id":"https://openalex.org/keywords/commonsense-knowledge","display_name":"Commonsense knowledge","score":0.7504638433456421},{"id":"https://openalex.org/keywords/sentence","display_name":"Sentence","score":0.6335654854774475},{"id":"https://openalex.org/keywords/natural-language-processing","display_name":"Natural language processing","score":0.6094240546226501},{"id":"https://openalex.org/keywords/artificial-intelligence","display_name":"Artificial intelligence","score":0.5757387280464172},{"id":"https://openalex.org/keywords/commonsense-reasoning","display_name":"Commonsense reasoning","score":0.5208142995834351},{"id":"https://openalex.org/keywords/logical-consequence","display_name":"Logical consequence","score":0.4656779170036316},{"id":"https://openalex.org/keywords/knowledge-representation-and-reasoning","display_name":"Knowledge representation and reasoning","score":0.18253421783447266},{"id":"https://openalex.org/keywords/image","display_name":"Image (mathematics)","score":0.15720659494400024}],"concepts":[{"id":"https://openalex.org/C157657479","wikidata":"https://www.wikidata.org/wiki/Q2367247","display_name":"Closed captioning","level":3,"score":0.9517967104911804},{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.8416991233825684},{"id":"https://openalex.org/C30542707","wikidata":"https://www.wikidata.org/wiki/Q1603203","display_name":"Commonsense knowledge","level":3,"score":0.7504638433456421},{"id":"https://openalex.org/C2777530160","wikidata":"https://www.wikidata.org/wiki/Q41796","display_name":"Sentence","level":2,"score":0.6335654854774475},{"id":"https://openalex.org/C204321447","wikidata":"https://www.wikidata.org/wiki/Q30642","display_name":"Natural language processing","level":1,"score":0.6094240546226501},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.5757387280464172},{"id":"https://openalex.org/C193221554","wikidata":"https://www.wikidata.org/wiki/Q5153664","display_name":"Commonsense reasoning","level":2,"score":0.5208142995834351},{"id":"https://openalex.org/C134752490","wikidata":"https://www.wikidata.org/wiki/Q374182","display_name":"Logical consequence","level":2,"score":0.4656779170036316},{"id":"https://openalex.org/C161301231","wikidata":"https://www.wikidata.org/wiki/Q3478658","display_name":"Knowledge representation and reasoning","level":2,"score":0.18253421783447266},{"id":"https://openalex.org/C115961682","wikidata":"https://www.wikidata.org/wiki/Q860623","display_name":"Image (mathematics)","level":2,"score":0.15720659494400024}],"mesh":[],"locations_count":1,"locations":[{"id":"doi:10.1109/icpr56361.2022.9956241","is_oa":false,"landing_page_url":"https://doi.org/10.1109/icpr56361.2022.9956241","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":[{"score":0.6600000262260437,"id":"https://metadata.un.org/sdg/10","display_name":"Reduced inequalities"}],"awards":[],"funders":[{"id":"https://openalex.org/F4320306076","display_name":"National Science Foundation","ror":"https://ror.org/021nxhr62"}],"has_content":{"pdf":false,"grobid_xml":false},"content_urls":null,"referenced_works_count":64,"referenced_works":["https://openalex.org/W648786980","https://openalex.org/W1514535095","https://openalex.org/W1586939924","https://openalex.org/W1614298861","https://openalex.org/W1905882502","https://openalex.org/W1956340063","https://openalex.org/W2108598243","https://openalex.org/W2123301721","https://openalex.org/W2131494463","https://openalex.org/W2139501017","https://openalex.org/W2154652894","https://openalex.org/W2176263492","https://openalex.org/W2277195237","https://openalex.org/W2425121537","https://openalex.org/W2506483933","https://openalex.org/W2558533273","https://openalex.org/W2607151106","https://openalex.org/W2619947201","https://openalex.org/W2729842244","https://openalex.org/W2775221064","https://openalex.org/W2889418103","https://openalex.org/W2896457183","https://openalex.org/W2905145027","https://openalex.org/W2948358897","https://openalex.org/W2949376505","https://openalex.org/W2951390634","https://openalex.org/W2953106684","https://openalex.org/W2962681491","https://openalex.org/W2962918138","https://openalex.org/W2962937869","https://openalex.org/W2962990649","https://openalex.org/W2963084599","https://openalex.org/W2963101081","https://openalex.org/W2963177403","https://openalex.org/W2963524571","https://openalex.org/W2963971014","https://openalex.org/W2964089981","https://openalex.org/W2964350391","https://openalex.org/W2964352247","https://openalex.org/W2989322838","https://openalex.org/W3035365026","https://openalex.org/W3035392611","https://openalex.org/W3101767943","https://openalex.org/W3103022576","https://openalex.org/W4385245566","https://openalex.org/W6620707391","https://openalex.org/W6621543089","https://openalex.org/W6630875275","https://openalex.org/W6636510571","https://openalex.org/W6641064462","https://openalex.org/W6678262379","https://openalex.org/W6679844565","https://openalex.org/W6682631176","https://openalex.org/W6685322675","https://openalex.org/W6694260854","https://openalex.org/W6694395031","https://openalex.org/W6722524744","https://openalex.org/W6730719754","https://openalex.org/W6739901393","https://openalex.org/W6741068176","https://openalex.org/W6746901506","https://openalex.org/W6755207826","https://openalex.org/W6898505805","https://openalex.org/W6955071965"],"related_works":["https://openalex.org/W3035583586","https://openalex.org/W4320165839","https://openalex.org/W2151799802","https://openalex.org/W4386607580","https://openalex.org/W4385488510","https://openalex.org/W2196562041","https://openalex.org/W2073302931","https://openalex.org/W4378501473","https://openalex.org/W3082691151","https://openalex.org/W4287633646"],"abstract_inverted_index":{"It":[0],"is":[1],"not":[2],"merely":[3],"an":[4],"aggregation":[5],"of":[6,18,72,127,156,163],"static":[7],"entities":[8],"that":[9,160],"a":[10,16,29,54,77,95,121],"video":[11,30,73,128,167],"clip":[12],"carries,":[13],"but":[14],"also":[15],"variety":[17],"interactions":[19],"and":[20,41,101,131,149,158],"relations":[21],"among":[22],"these":[23],"entities.":[24],"Challenges":[25],"still":[26],"remain":[27],"for":[28,80,124],"captioning":[31,74,129],"system":[32],"to":[33,68],"generate":[34],"descriptions":[35],"focusing":[36],"on":[37,144],"the":[38,44,70,103,125,133,154,161],"prominent":[39],"interest":[40],"aligning":[42],"with":[43,76,142],"latent":[45],"aspects":[46],"beyond":[47],"observations.":[48],"In":[49],"this":[50,116],"work,":[51],"we":[52,85],"present":[53],"Commonsense":[55],"knowledge":[56,67,88,97,165],"Anchored":[57],"Video":[58],"cAptioNing(dubbed":[59],"as":[60,120],"CAVAN)":[61],"approach.":[62],"CAVAN":[63,157],"exploits":[64],"inferential":[65],"commonsense":[66,87,122,164],"assist":[69],"training":[71,91,126],"model":[75,113,134],"novel":[78],"paradigm":[79],"sentence-level":[81],"semantic":[82],"alignment.":[83],"Specifically,":[84],"acquire":[86],"complementing":[89],"per":[90],"caption":[92,168],"by":[93],"querying":[94],"generic":[96],"atlas":[98],"(ATOMIC":[99],"[1]),":[100],"form":[102],"commonsense-caption":[104],"entailment":[105,112],"corpus.":[106],"A":[107],"BERT":[108],"[2]":[109],"based":[110],"language":[111],"trained":[114],"from":[115,135],"corpus":[117],"then":[118],"serves":[119],"discriminator":[123],"model,":[130],"penalizes":[132],"generating":[136],"semantically":[137],"misaligned":[138],"captions.":[139],"Experimental":[140],"results":[141],"ablations":[143],"MSRVTT":[145],"[3],":[146],"V2C":[147],"[4]":[148],"VATEX":[150],"[5]":[151],"datasets":[152],"validate":[153],"effectiveness":[155],"reveal":[159],"use":[162],"benefits":[166],"generation.":[169]},"counts_by_year":[{"year":2025,"cited_by_count":2},{"year":2023,"cited_by_count":1}],"updated_date":"2025-11-06T03:46:38.306776","created_date":"2025-10-10T00:00:00"}