{"id":"https://openalex.org/W3156423610","doi":"https://doi.org/10.1109/ijcnn52387.2021.9533961","title":"Object-Centric Representation Learning for Video Question Answering","display_name":"Object-Centric Representation Learning for Video Question Answering","publication_year":2021,"publication_date":"2021-07-18","ids":{"openalex":"https://openalex.org/W3156423610","doi":"https://doi.org/10.1109/ijcnn52387.2021.9533961","mag":"3156423610"},"language":"en","primary_location":{"id":"doi:10.1109/ijcnn52387.2021.9533961","is_oa":false,"landing_page_url":"https://doi.org/10.1109/ijcnn52387.2021.9533961","pdf_url":null,"source":null,"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"2021 International Joint Conference on Neural Networks (IJCNN)","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/2104.05166","any_repository_has_fulltext":true},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5008734340","display_name":"Long Hoang Dang","orcid":"https://orcid.org/0000-0002-2701-2365"},"institutions":[{"id":"https://openalex.org/I149704539","display_name":"Deakin University","ror":"https://ror.org/02czsnj07","country_code":"AU","type":"education","lineage":["https://openalex.org/I149704539"]}],"countries":["AU"],"is_corresponding":false,"raw_author_name":"Long Hoang Dang","raw_affiliation_strings":["Applied Artificial Intelligence Institute, Deakin University, Australia","Deakin University"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Applied Artificial Intelligence Institute, Deakin University, Australia","institution_ids":["https://openalex.org/I149704539"]},{"raw_affiliation_string":"Deakin University","institution_ids":["https://openalex.org/I149704539"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5079045166","display_name":"Thao Minh Le","orcid":"https://orcid.org/0000-0002-8089-9962"},"institutions":[{"id":"https://openalex.org/I149704539","display_name":"Deakin University","ror":"https://ror.org/02czsnj07","country_code":"AU","type":"education","lineage":["https://openalex.org/I149704539"]}],"countries":["AU"],"is_corresponding":false,"raw_author_name":"Thao Minh Le","raw_affiliation_strings":["Applied Artificial Intelligence Institute, Deakin University, Australia","Deakin University"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Applied Artificial Intelligence Institute, Deakin University, Australia","institution_ids":["https://openalex.org/I149704539"]},{"raw_affiliation_string":"Deakin University","institution_ids":["https://openalex.org/I149704539"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5029157904","display_name":"Vuong Le","orcid":"https://orcid.org/0000-0003-1582-1269"},"institutions":[{"id":"https://openalex.org/I149704539","display_name":"Deakin University","ror":"https://ror.org/02czsnj07","country_code":"AU","type":"education","lineage":["https://openalex.org/I149704539"]}],"countries":["AU"],"is_corresponding":false,"raw_author_name":"Vuong Le","raw_affiliation_strings":["Applied Artificial Intelligence Institute, Deakin University, Australia","Deakin University"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Applied Artificial Intelligence Institute, Deakin University, Australia","institution_ids":["https://openalex.org/I149704539"]},{"raw_affiliation_string":"Deakin University","institution_ids":["https://openalex.org/I149704539"]}]},{"author_position":"last","author":{"id":"https://openalex.org/A5085471517","display_name":"Truyen Tran","orcid":"https://orcid.org/0000-0001-6531-8907"},"institutions":[{"id":"https://openalex.org/I149704539","display_name":"Deakin University","ror":"https://ror.org/02czsnj07","country_code":"AU","type":"education","lineage":["https://openalex.org/I149704539"]}],"countries":["AU"],"is_corresponding":false,"raw_author_name":"Truyen Tran","raw_affiliation_strings":["Applied Artificial Intelligence Institute, Deakin University, Australia","Deakin University"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Applied Artificial Intelligence Institute, Deakin University, Australia","institution_ids":["https://openalex.org/I149704539"]},{"raw_affiliation_string":"Deakin University","institution_ids":["https://openalex.org/I149704539"]}]}],"institutions":[],"countries_distinct_count":1,"institutions_distinct_count":1,"corresponding_author_ids":[],"corresponding_institution_ids":["https://openalex.org/I149704539"],"apc_list":null,"apc_paid":null,"fwci":null,"has_fulltext":true,"cited_by_count":0,"citation_normalized_percentile":null,"cited_by_percentile_year":null,"biblio":{"volume":null,"issue":null,"first_page":"1","last_page":"8"},"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":0.9959999918937683,"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":0.994700014591217,"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":0.7912455797195435},{"id":"https://openalex.org/keywords/question-answering","display_name":"Question answering","score":0.5806946158409119},{"id":"https://openalex.org/keywords/feature-learning","display_name":"Feature learning","score":0.5496366024017334},{"id":"https://openalex.org/keywords/artificial-intelligence","display_name":"Artificial intelligence","score":0.5420771837234497},{"id":"https://openalex.org/keywords/representation","display_name":"Representation (politics)","score":0.5139355063438416},{"id":"https://openalex.org/keywords/semantics","display_name":"Semantics (computer science)","score":0.470704048871994},{"id":"https://openalex.org/keywords/knowledge-representation-and-reasoning","display_name":"Knowledge representation and reasoning","score":0.46064358949661255},{"id":"https://openalex.org/keywords/object","display_name":"Object (grammar)","score":0.4401482343673706},{"id":"https://openalex.org/keywords/natural-language-processing","display_name":"Natural language processing","score":0.4128655791282654},{"id":"https://openalex.org/keywords/programming-language","display_name":"Programming language","score":0.08921033143997192}],"concepts":[{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.7912455797195435},{"id":"https://openalex.org/C44291984","wikidata":"https://www.wikidata.org/wiki/Q1074173","display_name":"Question answering","level":2,"score":0.5806946158409119},{"id":"https://openalex.org/C59404180","wikidata":"https://www.wikidata.org/wiki/Q17013334","display_name":"Feature learning","level":2,"score":0.5496366024017334},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.5420771837234497},{"id":"https://openalex.org/C2776359362","wikidata":"https://www.wikidata.org/wiki/Q2145286","display_name":"Representation (politics)","level":3,"score":0.5139355063438416},{"id":"https://openalex.org/C184337299","wikidata":"https://www.wikidata.org/wiki/Q1437428","display_name":"Semantics (computer science)","level":2,"score":0.470704048871994},{"id":"https://openalex.org/C161301231","wikidata":"https://www.wikidata.org/wiki/Q3478658","display_name":"Knowledge representation and reasoning","level":2,"score":0.46064358949661255},{"id":"https://openalex.org/C2781238097","wikidata":"https://www.wikidata.org/wiki/Q175026","display_name":"Object (grammar)","level":2,"score":0.4401482343673706},{"id":"https://openalex.org/C204321447","wikidata":"https://www.wikidata.org/wiki/Q30642","display_name":"Natural language processing","level":1,"score":0.4128655791282654},{"id":"https://openalex.org/C199360897","wikidata":"https://www.wikidata.org/wiki/Q9143","display_name":"Programming language","level":1,"score":0.08921033143997192},{"id":"https://openalex.org/C94625758","wikidata":"https://www.wikidata.org/wiki/Q7163","display_name":"Politics","level":2,"score":0.0},{"id":"https://openalex.org/C199539241","wikidata":"https://www.wikidata.org/wiki/Q7748","display_name":"Law","level":1,"score":0.0},{"id":"https://openalex.org/C17744445","wikidata":"https://www.wikidata.org/wiki/Q36442","display_name":"Political science","level":0,"score":0.0}],"mesh":[],"locations_count":5,"locations":[{"id":"doi:10.1109/ijcnn52387.2021.9533961","is_oa":false,"landing_page_url":"https://doi.org/10.1109/ijcnn52387.2021.9533961","pdf_url":null,"source":null,"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"2021 International Joint Conference on Neural Networks (IJCNN)","raw_type":"proceedings-article"},{"id":"pmh:oai:arXiv.org:2104.05166","is_oa":true,"landing_page_url":"http://arxiv.org/abs/2104.05166","pdf_url":"https://arxiv.org/pdf/2104.05166","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":"","raw_type":"text"},{"id":"mag:3156423610","is_oa":true,"landing_page_url":"http://export.arxiv.org/pdf/2104.05166","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":"pmh:oai:figshare.com:article/21064465","is_oa":true,"landing_page_url":"https://figshare.com/articles/conference_contribution/Object-Centric_Representation_Learning_for_Video_Question_Answering/21064465","pdf_url":null,"source":{"id":"https://openalex.org/S4377196282","display_name":"Figshare","issn_l":null,"issn":null,"is_oa":false,"is_in_doaj":false,"is_core":false,"host_organization":"https://openalex.org/I4210132348","host_organization_name":"Figshare (United Kingdom)","host_organization_lineage":["https://openalex.org/I4210132348"],"host_organization_lineage_names":[],"type":"repository"},"license":"cc-by","license_id":"https://openalex.org/licenses/cc-by","version":"submittedVersion","is_accepted":false,"is_published":false,"raw_source_name":null,"raw_type":"Conference contribution"},{"id":"doi:10.48550/arxiv.2104.05166","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2104.05166","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:2104.05166","is_oa":true,"landing_page_url":"http://arxiv.org/abs/2104.05166","pdf_url":"https://arxiv.org/pdf/2104.05166","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":"","raw_type":"text"},"sustainable_development_goals":[{"id":"https://metadata.un.org/sdg/4","display_name":"Quality Education","score":0.6000000238418579}],"awards":[],"funders":[],"has_content":{"grobid_xml":true,"pdf":true},"content_urls":{"pdf":"https://content.openalex.org/works/W3156423610.pdf","grobid_xml":"https://content.openalex.org/works/W3156423610.grobid-xml"},"referenced_works_count":40,"referenced_works":["https://openalex.org/W639708223","https://openalex.org/W1699946128","https://openalex.org/W2095739681","https://openalex.org/W2194775991","https://openalex.org/W2250539671","https://openalex.org/W2425121537","https://openalex.org/W2561715562","https://openalex.org/W2603203130","https://openalex.org/W2606982687","https://openalex.org/W2611596598","https://openalex.org/W2613718673","https://openalex.org/W2735159761","https://openalex.org/W2737435850","https://openalex.org/W2765716052","https://openalex.org/W2777602943","https://openalex.org/W2890598214","https://openalex.org/W2897857500","https://openalex.org/W2904910963","https://openalex.org/W2930003055","https://openalex.org/W2954199749","https://openalex.org/W2962910007","https://openalex.org/W2962949233","https://openalex.org/W2963223524","https://openalex.org/W2963224792","https://openalex.org/W2963820951","https://openalex.org/W2964146787","https://openalex.org/W2978723076","https://openalex.org/W2996131066","https://openalex.org/W3023700693","https://openalex.org/W3034513622","https://openalex.org/W3034730770","https://openalex.org/W3089358914","https://openalex.org/W3113055895","https://openalex.org/W4251009746","https://openalex.org/W6726873649","https://openalex.org/W6748270630","https://openalex.org/W6764326188","https://openalex.org/W6777188697","https://openalex.org/W6777321055","https://openalex.org/W6787728148"],"related_works":["https://openalex.org/W3198966763","https://openalex.org/W3175625489","https://openalex.org/W3034513622","https://openalex.org/W2981774933","https://openalex.org/W3121327184","https://openalex.org/W2966369713","https://openalex.org/W3173582687","https://openalex.org/W2808675313","https://openalex.org/W2998631105","https://openalex.org/W3124461501","https://openalex.org/W1513491352","https://openalex.org/W2963760481","https://openalex.org/W2916723116","https://openalex.org/W3034636873","https://openalex.org/W3157224510","https://openalex.org/W3215632564","https://openalex.org/W3108249959","https://openalex.org/W3136759792","https://openalex.org/W3170410460","https://openalex.org/W3186088969"],"abstract_inverted_index":{"Video":[0,181],"question":[1],"answering":[2],"(Video":[3],"QA)":[4],"presents":[5],"a":[6,40,103,129,136],"powerful":[7],"testbed":[8],"for":[9,99,105,164],"human-like":[10],"intelligent":[11],"behaviors.":[12],"The":[13,154,175],"task":[14],"demands":[15],"new":[16,130],"capabilities":[17],"to":[18,28,43,67,134,172,191],"integrate":[19],"video":[20,137,192],"processing,":[21],"language":[22],"understanding,":[23],"binding":[24],"abstract":[25],"linguistic":[26],"concepts":[27],"concrete":[29],"visual":[30],"artifacts,":[31],"and":[32,55,71,120,147,151],"deliberative":[33,165],"reasoning":[34,88,167],"over":[35],"spacetime.":[36],"Neural":[37],"networks":[38,59],"offer":[39],"promising":[41],"approach":[42,190],"reach":[44],"this":[45,95,125],"potential":[46],"through":[47,81],"learning":[48],"from":[49,109],"examples":[50],"rather":[51],"than":[52],"handcrafting":[53],"features":[54,146],"rules.":[56],"However,":[57],"neural":[58],"are":[60,149,157],"predominantly":[61],"feature-based":[62],"-":[63],"they":[64],"map":[65],"data":[66],"unstructured":[68],"vectorial":[69],"representation":[70,101,132],"thus":[72],"can":[73],"fall":[74],"into":[75,138,160],"the":[76,113,173,188],"trap":[77],"of":[78,85,143,187],"exploiting":[79],"shortcuts":[80],"surface":[82],"statistics":[83],"instead":[84],"true":[86],"systematic":[87],"seen":[89],"in":[90],"symbolic":[91,122],"systems.":[92],"To":[93,124],"tackle":[94],"issue,":[96],"we":[97,127],"advocate":[98],"object-centric":[100,189],"as":[102],"basis":[104],"constructing":[106],"spatio-temporal":[107],"structures":[108],"videos,":[110],"essentially":[111],"bridging":[112],"semantic":[114],"gap":[115],"between":[116],"low-level":[117],"pattern":[118],"recognition":[119],"high-level":[121],"algebra.":[123],"end,":[126],"propose":[128],"query-guided":[131],"framework":[133,176],"turn":[135],"an":[139,170],"evolving":[140],"relational":[141,166],"graph":[142],"objects,":[144],"whose":[145],"interactions":[148],"dynamically":[150],"conditionally":[152],"inferred.":[153],"object":[155],"lives":[156],"then":[158],"summarized":[159],"r\u00e9sum\u00e9s,":[161],"lending":[162],"naturally":[163],"that":[168],"produces":[169],"answer":[171],"query.":[174],"is":[177],"evaluated":[178],"on":[179],"major":[180],"QA":[182],"datasets,":[183],"demonstrating":[184],"clear":[185],"benefits":[186],"reasoning.":[193]},"counts_by_year":[],"updated_date":"2026-07-14T23:27:15.235271","created_date":"2025-10-10T00:00:00"}