{"id":"https://openalex.org/W7160314573","doi":"https://doi.org/10.1109/wacv61042.2026.00600","title":"Distilling Offline Action Detection Models into Real-Time Streaming Models","display_name":"Distilling Offline Action Detection Models into Real-Time Streaming Models","publication_year":2026,"publication_date":"2026-03-06","ids":{"openalex":"https://openalex.org/W7160314573","doi":"https://doi.org/10.1109/wacv61042.2026.00600"},"language":null,"primary_location":{"id":"doi:10.1109/wacv61042.2026.00600","is_oa":false,"landing_page_url":"https://doi.org/10.1109/wacv61042.2026.00600","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/A5135406878","display_name":"Deep Patel","orcid":null},"institutions":[{"id":"https://openalex.org/I4210107353","display_name":"NEC (United States)","ror":"https://ror.org/01v791m31","country_code":"US","type":"company","lineage":["https://openalex.org/I118347220","https://openalex.org/I4210107353"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Deep Patel","raw_affiliation_strings":["NEC Laboratories America"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"NEC Laboratories America","institution_ids":["https://openalex.org/I4210107353"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5078140092","display_name":"Yasunori Babazaki","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Yasunori Babazaki","raw_affiliation_strings":["NEC Corporation"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"NEC Corporation","institution_ids":[]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5081773473","display_name":"Yasuto Nagase","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Yasuto Nagase","raw_affiliation_strings":["NEC Corporation"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"NEC Corporation","institution_ids":[]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5058301235","display_name":"Iain Melvin","orcid":null},"institutions":[{"id":"https://openalex.org/I4210107353","display_name":"NEC (United States)","ror":"https://ror.org/01v791m31","country_code":"US","type":"company","lineage":["https://openalex.org/I118347220","https://openalex.org/I4210107353"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Iain Melvin","raw_affiliation_strings":["NEC Laboratories America"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"NEC Laboratories America","institution_ids":["https://openalex.org/I4210107353"]}]},{"author_position":"last","author":{"id":"https://openalex.org/A5065576006","display_name":"Martin Renqiang Min","orcid":"https://orcid.org/0000-0002-8563-6133"},"institutions":[{"id":"https://openalex.org/I4210107353","display_name":"NEC (United States)","ror":"https://ror.org/01v791m31","country_code":"US","type":"company","lineage":["https://openalex.org/I118347220","https://openalex.org/I4210107353"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Martin Renqiang Min","raw_affiliation_strings":["NEC Laboratories America"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"NEC Laboratories America","institution_ids":["https://openalex.org/I4210107353"]}]}],"institutions":[],"countries_distinct_count":1,"institutions_distinct_count":1,"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":"6205","last_page":"6214"},"is_retracted":false,"is_paratext":false,"is_xpac":false,"primary_topic":{"id":"https://openalex.org/T10812","display_name":"Human Pose and Action Recognition","score":0.9452000260353088,"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/T10812","display_name":"Human Pose and Action Recognition","score":0.9452000260353088,"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/T10444","display_name":"Context-Aware Activity Recognition Systems","score":0.020400000736117363,"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/T10331","display_name":"Video Surveillance and Tracking Methods","score":0.005200000014156103,"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/action","display_name":"Action (physics)","score":0.5246000289916992},{"id":"https://openalex.org/keywords/key","display_name":"Key (lock)","score":0.337799996137619},{"id":"https://openalex.org/keywords/noise","display_name":"Noise (video)","score":0.2833999991416931},{"id":"https://openalex.org/keywords/computation","display_name":"Computation","score":0.22599999606609344},{"id":"https://openalex.org/keywords/online-and-offline","display_name":"Online and offline","score":0.2248000055551529}],"concepts":[{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.6225000023841858},{"id":"https://openalex.org/C2780791683","wikidata":"https://www.wikidata.org/wiki/Q846785","display_name":"Action (physics)","level":2,"score":0.5246000289916992},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.36059999465942383},{"id":"https://openalex.org/C26517878","wikidata":"https://www.wikidata.org/wiki/Q228039","display_name":"Key (lock)","level":2,"score":0.337799996137619},{"id":"https://openalex.org/C99498987","wikidata":"https://www.wikidata.org/wiki/Q2210247","display_name":"Noise (video)","level":3,"score":0.2833999991416931},{"id":"https://openalex.org/C79403827","wikidata":"https://www.wikidata.org/wiki/Q3988","display_name":"Real-time computing","level":1,"score":0.272599995136261},{"id":"https://openalex.org/C124101348","wikidata":"https://www.wikidata.org/wiki/Q172491","display_name":"Data mining","level":1,"score":0.24169999361038208},{"id":"https://openalex.org/C45374587","wikidata":"https://www.wikidata.org/wiki/Q12525525","display_name":"Computation","level":2,"score":0.22599999606609344},{"id":"https://openalex.org/C2780102126","wikidata":"https://www.wikidata.org/wiki/Q10928179","display_name":"Online and offline","level":2,"score":0.2248000055551529},{"id":"https://openalex.org/C9652623","wikidata":"https://www.wikidata.org/wiki/Q190109","display_name":"Field (mathematics)","level":2,"score":0.21629999577999115}],"mesh":[],"locations_count":1,"locations":[{"id":"doi:10.1109/wacv61042.2026.00600","is_oa":false,"landing_page_url":"https://doi.org/10.1109/wacv61042.2026.00600","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":[],"awards":[],"funders":[],"has_content":{"pdf":false,"grobid_xml":false},"content_urls":null,"referenced_works_count":32,"referenced_works":["https://openalex.org/W2126579184","https://openalex.org/W2618799552","https://openalex.org/W2963524571","https://openalex.org/W2963529931","https://openalex.org/W2963563276","https://openalex.org/W2963677766","https://openalex.org/W2987014658","https://openalex.org/W2990152177","https://openalex.org/W2990503944","https://openalex.org/W2991484432","https://openalex.org/W3096609285","https://openalex.org/W3126721948","https://openalex.org/W3173407577","https://openalex.org/W3204116406","https://openalex.org/W4214612132","https://openalex.org/W4214614183","https://openalex.org/W4312276384","https://openalex.org/W4312349930","https://openalex.org/W4312509322","https://openalex.org/W4312560592","https://openalex.org/W4312769131","https://openalex.org/W4312772544","https://openalex.org/W4313192411","https://openalex.org/W4319299890","https://openalex.org/W4386076295","https://openalex.org/W4388979610","https://openalex.org/W4390872366","https://openalex.org/W4390873033","https://openalex.org/W4394597406","https://openalex.org/W4394625738","https://openalex.org/W4406670962","https://openalex.org/W4409263255"],"related_works":[],"abstract_inverted_index":{"Vision":[0],"Transformers":[1],"(ViTs)":[2],"have":[3],"achieved":[4],"state-of-the-art":[5,157],"performance":[6,158],"in":[7,145,175],"offline":[8,48,153],"video":[9,178],"action":[10],"detection,":[11],"but":[12],"their":[13],"reliance":[14],"on":[15,109],"processing":[16],"fixed-size":[17],"clips":[18],"with":[19,81,95],"full":[20,68],"spatiotemporal":[21],"attention":[22,63,94],"makes":[23],"them":[24],"computationally":[25],"expensive":[26],"and":[27,84,147,167],"ill-suited":[28],"for":[29,98,159,170],"real-time":[30],"streaming":[31],"applications":[32],"due":[33],"to":[34,44,105,141,150],"massive":[35],"computational":[36],"redundancy.":[37],"This":[38],"paper":[39],"introduces":[40],"a":[41,72,86,110,128,142,165],"novel":[42],"framework":[43],"adapt":[45],"these":[46],"powerful":[47],"models":[49,54,134,174],"into":[50],"efficient,":[51],"online":[52,160],"student":[53],"through":[55],"knowledge":[56,115],"distillation.":[57],"We":[58],"propose":[59],"two":[60],"causal,":[61],"streaming-friendly":[62],"architectures":[64,102],"that":[65,77,90],"replace":[66],"the":[67,125,151],"self-attention":[69],"mechanism:":[70],"(1)":[71],"lightweight":[73],"Temporal":[74],"Shift":[75],"Attention":[76,89],"integrates":[78],"past":[79],"context":[80],"minimal":[82],"overhead,":[83],"(2)":[85],"Decomposed":[87],"Spatial-Temporal":[88],"combines":[91],"intraframe":[92],"spatial":[93],"an":[96,119],"LSTM":[97],"temporal":[99],"modeling.":[100],"Both":[101],"utilize":[103],"caching":[104],"eliminate":[106],"redundant":[107],"operations":[108],"frame-by-frame":[111],"basis.":[112],"To":[113],"maximize":[114],"transfer,":[116],"we":[117],"introduce":[118],"uncertainty-guided":[120],"distillation":[121],"process,":[122],"which":[123],"formulates":[124],"training":[126],"as":[127],"multi-task":[129],"learning":[130],"problem.":[131],"Our":[132,162],"resulting":[133],"demonstrate":[135],"significant":[136],"efficiency":[137],"gains,":[138],"achieving":[139],"up":[140],"4x":[143],"improvement":[144],"latency":[146],"throughput":[148],"compared":[149],"original":[152],"teacher":[154],"while":[155],"ensuring":[156],"methods.":[161],"work":[163],"provides":[164],"practical":[166],"effective":[168],"methodology":[169],"deploying":[171],"high-accuracy":[172],"transformer":[173],"latency-sensitive,":[176],"real-world":[177],"analysis":[179],"systems.":[180]},"counts_by_year":[],"updated_date":"2026-07-14T23:27:15.235271","created_date":"2026-05-06T00:00:00"}