{"id":"https://openalex.org/W7160299547","doi":"https://doi.org/10.1109/wacv61042.2026.00629","title":"Hybrid State Representation for Video Procedure Planning","display_name":"Hybrid State Representation for Video Procedure Planning","publication_year":2026,"publication_date":"2026-03-06","ids":{"openalex":"https://openalex.org/W7160299547","doi":"https://doi.org/10.1109/wacv61042.2026.00629"},"language":null,"primary_location":{"id":"doi:10.1109/wacv61042.2026.00629","is_oa":false,"landing_page_url":"https://doi.org/10.1109/wacv61042.2026.00629","pdf_url":null,"source":null,"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"2026 IEEE/CVF Winter Conference on Applications of Computer Vision (WACV)","raw_type":"proceedings-article"},"type":"conference-paper","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/A5135304764","display_name":"Woo Suk Choi","orcid":null},"institutions":[{"id":"https://openalex.org/I139264467","display_name":"Seoul National University","ror":"https://ror.org/04h9pn542","country_code":"KR","type":"education","lineage":["https://openalex.org/I139264467"]}],"countries":["KR"],"is_corresponding":false,"raw_author_name":"Woo Suk Choi","raw_affiliation_strings":["Seoul National University"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Seoul National University","institution_ids":["https://openalex.org/I139264467"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5084585820","display_name":"Youwon Jang","orcid":"https://orcid.org/0000-0001-6714-3717"},"institutions":[{"id":"https://openalex.org/I139264467","display_name":"Seoul National University","ror":"https://ror.org/04h9pn542","country_code":"KR","type":"education","lineage":["https://openalex.org/I139264467"]}],"countries":["KR"],"is_corresponding":false,"raw_author_name":"Youwon Jang","raw_affiliation_strings":["Seoul National University"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Seoul National University","institution_ids":["https://openalex.org/I139264467"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5135313123","display_name":"Minsu Lee","orcid":"https://orcid.org/0000-0002-9601-3863"},"institutions":[{"id":"https://openalex.org/I165677929","display_name":"Sungshin Women's University","ror":"https://ror.org/0500xzf72","country_code":"KR","type":"education","lineage":["https://openalex.org/I165677929"]}],"countries":["KR"],"is_corresponding":false,"raw_author_name":"Minsu Lee","raw_affiliation_strings":["Sungshin Women’s University"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Sungshin Women’s University","institution_ids":["https://openalex.org/I165677929"]}]},{"author_position":"last","author":{"id":"https://openalex.org/A5135361855","display_name":"Byoung-Tak Zhang","orcid":"https://orcid.org/0000-0001-9890-0389"},"institutions":[{"id":"https://openalex.org/I139264467","display_name":"Seoul National University","ror":"https://ror.org/04h9pn542","country_code":"KR","type":"education","lineage":["https://openalex.org/I139264467"]}],"countries":["KR"],"is_corresponding":false,"raw_author_name":"Byoung-Tak Zhang","raw_affiliation_strings":["Seoul National University"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Seoul National University","institution_ids":["https://openalex.org/I139264467"]}]}],"institutions":[],"countries_distinct_count":1,"institutions_distinct_count":2,"corresponding_author_ids":[],"corresponding_institution_ids":[],"apc_list":null,"apc_paid":null,"fwci":0.0,"has_fulltext":false,"cited_by_count":0,"citation_normalized_percentile":{"value":0.51987732,"is_in_top_1_percent":false,"is_in_top_10_percent":false},"cited_by_percentile_year":null,"biblio":{"volume":null,"issue":null,"first_page":"6507","last_page":"6516"},"is_retracted":false,"is_paratext":false,"is_xpac":false,"primary_topic":{"id":"https://openalex.org/T11439","display_name":"Video Analysis and Summarization","score":0.22709999978542328,"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/T11439","display_name":"Video Analysis and Summarization","score":0.22709999978542328,"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/T10906","display_name":"AI-based Problem Solving and Planning","score":0.14630000293254852,"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":0.12280000001192093,"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/representation","display_name":"Representation (politics)","score":0.49459999799728394},{"id":"https://openalex.org/keywords/state","display_name":"State (computer science)","score":0.45730000734329224},{"id":"https://openalex.org/keywords/key","display_name":"Key (lock)","score":0.3231000006198883},{"id":"https://openalex.org/keywords/context","display_name":"Context (archaeology)","score":0.2831000089645386},{"id":"https://openalex.org/keywords/feature","display_name":"Feature (linguistics)","score":0.28130000829696655}],"concepts":[{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.5968000292778015},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.5067999958992004},{"id":"https://openalex.org/C2776359362","wikidata":"https://www.wikidata.org/wiki/Q2145286","display_name":"Representation (politics)","level":3,"score":0.49459999799728394},{"id":"https://openalex.org/C48103436","wikidata":"https://www.wikidata.org/wiki/Q599031","display_name":"State (computer science)","level":2,"score":0.45730000734329224},{"id":"https://openalex.org/C31972630","wikidata":"https://www.wikidata.org/wiki/Q844240","display_name":"Computer vision","level":1,"score":0.43549999594688416},{"id":"https://openalex.org/C11413529","wikidata":"https://www.wikidata.org/wiki/Q8366","display_name":"Algorithm","level":1,"score":0.3312999904155731},{"id":"https://openalex.org/C26517878","wikidata":"https://www.wikidata.org/wiki/Q228039","display_name":"Key (lock)","level":2,"score":0.3231000006198883},{"id":"https://openalex.org/C2779343474","wikidata":"https://www.wikidata.org/wiki/Q3109175","display_name":"Context (archaeology)","level":2,"score":0.2831000089645386},{"id":"https://openalex.org/C2776401178","wikidata":"https://www.wikidata.org/wiki/Q12050496","display_name":"Feature (linguistics)","level":2,"score":0.28130000829696655},{"id":"https://openalex.org/C126042441","wikidata":"https://www.wikidata.org/wiki/Q1324888","display_name":"Frame (networking)","level":2,"score":0.26589998602867126},{"id":"https://openalex.org/C33923547","wikidata":"https://www.wikidata.org/wiki/Q395","display_name":"Mathematics","level":0,"score":0.2621999979019165},{"id":"https://openalex.org/C2779304628","wikidata":"https://www.wikidata.org/wiki/Q3503480","display_name":"Face (sociological concept)","level":2,"score":0.257099986076355}],"mesh":[],"locations_count":1,"locations":[{"id":"doi:10.1109/wacv61042.2026.00629","is_oa":false,"landing_page_url":"https://doi.org/10.1109/wacv61042.2026.00629","pdf_url":null,"source":null,"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"2026 IEEE/CVF Winter Conference on Applications of Computer Vision (WACV)","raw_type":"proceedings-article"}],"best_oa_location":null,"sustainable_development_goals":[{"display_name":"Peace, Justice and strong institutions","score":0.4409756362438202,"id":"https://metadata.un.org/sdg/16"}],"awards":[],"funders":[],"has_content":{"grobid_xml":false,"pdf":false},"content_urls":null,"referenced_works_count":26,"referenced_works":["https://openalex.org/W2049799298","https://openalex.org/W2077069816","https://openalex.org/W2128700566","https://openalex.org/W2957775769","https://openalex.org/W2962795934","https://openalex.org/W2964094654","https://openalex.org/W3035392611","https://openalex.org/W3106768499","https://openalex.org/W3199332348","https://openalex.org/W3203256294","https://openalex.org/W4214482673","https://openalex.org/W4224440661","https://openalex.org/W4287854977","https://openalex.org/W4301900652","https://openalex.org/W4312478760","https://openalex.org/W4385570410","https://openalex.org/W4386075710","https://openalex.org/W4388593766","https://openalex.org/W4390873421","https://openalex.org/W4390873991","https://openalex.org/W4390874412","https://openalex.org/W4402770916","https://openalex.org/W4405078884","https://openalex.org/W4409262267","https://openalex.org/W4409366039","https://openalex.org/W7133193597"],"related_works":[],"abstract_inverted_index":{"Accurate":[0],"state":[1,101],"representation":[2,123],"is":[3],"critical":[4,100],"for":[5,171],"effective":[6],"procedure":[7],"planning":[8],"from":[9],"visual":[10,126],"inputs.":[11],"Existing":[12],"methods":[13],"typically":[14],"rely":[15],"on":[16,130,148],"sentence-level":[17],"natural":[18],"language":[19],"descriptions":[20],"to":[21,32,38,98,109,119],"represent":[22],"states.":[23],"However,":[24],"such":[25],"representations":[26],"are":[27,169],"often":[28],"ambiguous":[29],"and":[30,66,84,87,113,134,163],"fail":[31],"capture":[33],"fine-grained":[34],"object":[35],"interactions,":[36],"leading":[37],"errors":[39],"in":[40,124],"complex":[41],"scenarios.":[42],"To":[43],"overcome":[44],"these":[45,111],"limitations,":[46],"we":[47],"propose":[48],"a":[49,106,115,142],"Hybrid":[50],"State":[51,76],"Representation":[52],"(HSR),":[53],"inspired":[54],"by":[55],"human-like":[56],"reasoning,":[57],"that":[58,138,158],"models":[59],"procedural":[60],"states":[61],"with":[62],"both":[63,159],"structural":[64,161],"precision":[65],"contextual":[67,89,164],"clarity.":[68],"HSR":[69,168],"integrates":[70],"two":[71],"complementary":[72],"modalities:":[73],"(1)":[74],"Semantic":[75],"Graphs":[77],"(SSGs),":[78],"which":[79,93],"explicitly":[80],"encode":[81],"objects,":[82],"attributes,":[83],"their":[85],"relations,":[86],"(2)":[88],"Question-Answer":[90],"(QA)":[91,165],"pairs,":[92],"act":[94],"as":[95],"semantic":[96],"probes":[97],"disambiguate":[99],"transitions.":[102],"We":[103],"further":[104],"design":[105],"heterogeneous":[107],"encoder":[108],"fuse":[110],"components":[112,166],"intro-duce":[114],"visual-state":[116],"alignment":[117],"objective":[118],"ground":[120],"the":[121,125,131,149,160,172],"hybrid":[122],"context.":[127],"Extensive":[128],"experiments":[129],"CrossTask,":[132],"COIN,":[133],"NIV":[135],"benchmarks":[136],"demonstrate":[137],"our":[139],"method":[140],"establishes":[141],"new":[143],"state-of-the-art,":[144],"achieving":[145],"significant":[146],"gains":[147],"strict":[150],"Success":[151],"Rate":[152],"(SR)":[153],"metric.":[154],"Ablation":[155],"studies":[156],"confirm":[157],"(SSG)":[162],"of":[167],"essential":[170],"observed":[173],"performance":[174],"gains.":[175]},"counts_by_year":[],"updated_date":"2026-07-29T14:22:42.915294","created_date":"2026-05-06T00:00:00"}