{"id":"https://openalex.org/W4390872447","doi":"https://doi.org/10.1109/iccv51070.2023.01598","title":"SwiftFormer: Efficient Additive Attention for Transformer-based Real-time Mobile Vision Applications","display_name":"SwiftFormer: Efficient Additive Attention for Transformer-based Real-time Mobile Vision Applications","publication_year":2023,"publication_date":"2023-10-01","ids":{"openalex":"https://openalex.org/W4390872447","doi":"https://doi.org/10.1109/iccv51070.2023.01598"},"language":"en","primary_location":{"is_oa":false,"landing_page_url":"https://doi.org/10.1109/iccv51070.2023.01598","pdf_url":null,"source":{"id":"https://openalex.org/S4363607764","display_name":"2021 IEEE/CVF International Conference on Computer Vision (ICCV)","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":null,"is_accepted":false,"is_published":false},"type":"article","type_crossref":"proceedings-article","indexed_in":["crossref"],"open_access":{"is_oa":true,"oa_status":"green","oa_url":"https://arxiv.org/pdf/2303.15446","any_repository_has_fulltext":true},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5079602015","display_name":"Abdelrahman Shaker","orcid":"https://orcid.org/0000-0001-7651-4057"},"institutions":[{"id":"https://openalex.org/I4210113480","display_name":"Mohamed bin Zayed University of Artificial Intelligence","ror":"https://ror.org/0258gkt32","country_code":"AE","type":"education","lineage":["https://openalex.org/I4210113480"]}],"countries":["AE"],"is_corresponding":false,"raw_author_name":"Abdelrahman Shaker","raw_affiliation_strings":["Mohamed bin Zayed University of AI"],"affiliations":[{"raw_affiliation_string":"Mohamed bin Zayed University of AI","institution_ids":["https://openalex.org/I4210113480"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5034370385","display_name":"Muhammad Maaz","orcid":"https://orcid.org/0000-0002-3869-631X"},"institutions":[{"id":"https://openalex.org/I4210113480","display_name":"Mohamed bin Zayed University of Artificial Intelligence","ror":"https://ror.org/0258gkt32","country_code":"AE","type":"education","lineage":["https://openalex.org/I4210113480"]}],"countries":["AE"],"is_corresponding":false,"raw_author_name":"Muhammad Maaz","raw_affiliation_strings":["Mohamed bin Zayed University of AI"],"affiliations":[{"raw_affiliation_string":"Mohamed bin Zayed University of AI","institution_ids":["https://openalex.org/I4210113480"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5064948724","display_name":"Hanoona Rasheed","orcid":null},"institutions":[{"id":"https://openalex.org/I4210113480","display_name":"Mohamed bin Zayed University of Artificial Intelligence","ror":"https://ror.org/0258gkt32","country_code":"AE","type":"education","lineage":["https://openalex.org/I4210113480"]}],"countries":["AE"],"is_corresponding":false,"raw_author_name":"Hanoona Rasheed","raw_affiliation_strings":["Mohamed bin Zayed University of AI"],"affiliations":[{"raw_affiliation_string":"Mohamed bin Zayed University of AI","institution_ids":["https://openalex.org/I4210113480"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5101483978","display_name":"Salman Khan","orcid":"https://orcid.org/0000-0001-8732-3395"},"institutions":[{"id":"https://openalex.org/I4210113480","display_name":"Mohamed bin Zayed University of Artificial Intelligence","ror":"https://ror.org/0258gkt32","country_code":"AE","type":"education","lineage":["https://openalex.org/I4210113480"]}],"countries":["AE"],"is_corresponding":false,"raw_author_name":"Salman Khan","raw_affiliation_strings":["Mohamed bin Zayed University of AI"],"affiliations":[{"raw_affiliation_string":"Mohamed bin Zayed University of AI","institution_ids":["https://openalex.org/I4210113480"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5100418319","display_name":"Ming\u2013Hsuan Yang","orcid":"https://orcid.org/0000-0003-4848-2304"},"institutions":[{"id":"https://openalex.org/I156087764","display_name":"University of California, Merced","ror":"https://ror.org/00d9ah105","country_code":"US","type":"education","lineage":["https://openalex.org/I156087764"]},{"id":"https://openalex.org/I193775966","display_name":"Yonsei University","ror":"https://ror.org/01wjejq96","country_code":"KR","type":"education","lineage":["https://openalex.org/I193775966"]},{"id":"https://openalex.org/I1291425158","display_name":"Google (United States)","ror":"https://ror.org/00njsd438","country_code":"US","type":"company","lineage":["https://openalex.org/I1291425158","https://openalex.org/I4210128969"]}],"countries":["KR","US"],"is_corresponding":false,"raw_author_name":"Ming-Hsuan Yang","raw_affiliation_strings":["Google Research","University of California, Merced","Yonsei University"],"affiliations":[{"raw_affiliation_string":"University of California, Merced","institution_ids":["https://openalex.org/I156087764"]},{"raw_affiliation_string":"Yonsei University","institution_ids":["https://openalex.org/I193775966"]},{"raw_affiliation_string":"Google Research","institution_ids":["https://openalex.org/I1291425158"]}]},{"author_position":"last","author":{"id":"https://openalex.org/A5100760570","display_name":"Fahad Shahbaz Khan","orcid":"https://orcid.org/0000-0002-4263-3143"},"institutions":[{"id":"https://openalex.org/I4210113480","display_name":"Mohamed bin Zayed University of Artificial Intelligence","ror":"https://ror.org/0258gkt32","country_code":"AE","type":"education","lineage":["https://openalex.org/I4210113480"]}],"countries":["AE"],"is_corresponding":false,"raw_author_name":"Fahad Shahbaz Khan","raw_affiliation_strings":["Mohamed bin Zayed University of AI"],"affiliations":[{"raw_affiliation_string":"Mohamed bin Zayed University of AI","institution_ids":["https://openalex.org/I4210113480"]}]}],"institution_assertions":[],"countries_distinct_count":3,"institutions_distinct_count":4,"corresponding_author_ids":[],"corresponding_institution_ids":[],"apc_list":null,"apc_paid":null,"fwci":9.043,"has_fulltext":false,"cited_by_count":40,"citation_normalized_percentile":{"value":0.999956,"is_in_top_1_percent":true,"is_in_top_10_percent":true},"cited_by_percentile_year":{"min":99,"max":100},"biblio":{"volume":null,"issue":null,"first_page":null,"last_page":null},"is_retracted":false,"is_paratext":false,"primary_topic":{"id":"https://openalex.org/T10036","display_name":"Advanced Neural Network Applications","score":0.9997,"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/T10036","display_name":"Advanced Neural Network Applications","score":0.9997,"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/T10627","display_name":"Advanced Image and Video Retrieval Techniques","score":0.9985,"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/T12702","display_name":"Brain Tumor Detection and Classification","score":0.9967,"subfield":{"id":"https://openalex.org/subfields/2808","display_name":"Neurology"},"field":{"id":"https://openalex.org/fields/28","display_name":"Neuroscience"},"domain":{"id":"https://openalex.org/domains/1","display_name":"Life Sciences"}}],"keywords":[],"concepts":[{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.8197632},{"id":"https://openalex.org/C2780513914","wikidata":"https://www.wikidata.org/wiki/Q18210350","display_name":"Bottleneck","level":2,"score":0.57338333},{"id":"https://openalex.org/C17349429","wikidata":"https://www.wikidata.org/wiki/Q1049914","display_name":"Matrix multiplication","level":3,"score":0.51595837},{"id":"https://openalex.org/C82876162","wikidata":"https://www.wikidata.org/wiki/Q17096504","display_name":"Latency (audio)","level":2,"score":0.511648},{"id":"https://openalex.org/C186967261","wikidata":"https://www.wikidata.org/wiki/Q5082128","display_name":"Mobile device","level":2,"score":0.4682245},{"id":"https://openalex.org/C113775141","wikidata":"https://www.wikidata.org/wiki/Q428691","display_name":"Computer engineering","level":1,"score":0.4557135},{"id":"https://openalex.org/C2776214188","wikidata":"https://www.wikidata.org/wiki/Q408386","display_name":"Inference","level":2,"score":0.44677448},{"id":"https://openalex.org/C2780595030","wikidata":"https://www.wikidata.org/wiki/Q3860309","display_name":"Multiplication (music)","level":2,"score":0.41780657},{"id":"https://openalex.org/C120314980","wikidata":"https://www.wikidata.org/wiki/Q180634","display_name":"Distributed computing","level":1,"score":0.35433817},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.30069745},{"id":"https://openalex.org/C149635348","wikidata":"https://www.wikidata.org/wiki/Q193040","display_name":"Embedded system","level":1,"score":0.23929241},{"id":"https://openalex.org/C76155785","wikidata":"https://www.wikidata.org/wiki/Q418","display_name":"Telecommunications","level":1,"score":0.0},{"id":"https://openalex.org/C121332964","wikidata":"https://www.wikidata.org/wiki/Q413","display_name":"Physics","level":0,"score":0.0},{"id":"https://openalex.org/C62520636","wikidata":"https://www.wikidata.org/wiki/Q944","display_name":"Quantum mechanics","level":1,"score":0.0},{"id":"https://openalex.org/C24890656","wikidata":"https://www.wikidata.org/wiki/Q82811","display_name":"Acoustics","level":1,"score":0.0},{"id":"https://openalex.org/C84114770","wikidata":"https://www.wikidata.org/wiki/Q46344","display_name":"Quantum","level":2,"score":0.0},{"id":"https://openalex.org/C111919701","wikidata":"https://www.wikidata.org/wiki/Q9135","display_name":"Operating system","level":1,"score":0.0}],"mesh":[],"locations_count":2,"locations":[{"is_oa":false,"landing_page_url":"https://doi.org/10.1109/iccv51070.2023.01598","pdf_url":null,"source":{"id":"https://openalex.org/S4363607764","display_name":"2021 IEEE/CVF International Conference on Computer Vision (ICCV)","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":null,"is_accepted":false,"is_published":false},{"is_oa":true,"landing_page_url":"https://arxiv.org/abs/2303.15446","pdf_url":"https://arxiv.org/pdf/2303.15446","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":["Cornell University"],"type":"repository"},"license":null,"license_id":null,"version":"submittedVersion","is_accepted":false,"is_published":false}],"best_oa_location":{"is_oa":true,"landing_page_url":"https://arxiv.org/abs/2303.15446","pdf_url":"https://arxiv.org/pdf/2303.15446","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":["Cornell University"],"type":"repository"},"license":null,"license_id":null,"version":"submittedVersion","is_accepted":false,"is_published":false},"sustainable_development_goals":[{"score":0.51,"id":"https://metadata.un.org/sdg/8","display_name":"Decent work and economic growth"}],"grants":[],"datasets":[],"versions":[],"referenced_works_count":0,"referenced_works":[],"related_works":["https://openalex.org/W752783541","https://openalex.org/W4287593139","https://openalex.org/W4239424132","https://openalex.org/W4206811032","https://openalex.org/W3212757063","https://openalex.org/W3099313426","https://openalex.org/W2995605830","https://openalex.org/W2596457687","https://openalex.org/W2086123442","https://openalex.org/W1506547947"],"abstract_inverted_index":{"Self-attention":[0],"has":[1],"become":[2],"a":[3,52,64,71,100,133],"defacto":[4],"choice":[5],"for":[6,31,51],"capturing":[7],"global":[8],"context":[9],"in":[10,27,61,143],"various":[11],"vision":[12],"applications.":[13],"However,":[14],"its":[15,25,117],"quadratic":[16,81],"computational":[17],"complexity":[18],"with":[19,85,99,160],"respect":[20],"to":[21,43,176],"image":[22],"resolution":[23],"limits":[24],"use":[26],"real-time":[28],"applications,":[29],"especially":[30],"deployment":[32],"on":[33,165],"resource-constrained":[34],"mobile":[35,149],"devices.":[36],"Although":[37],"hybrid":[38],"approaches":[39],"have":[40],"been":[41],"proposed":[42,127],"combine":[44],"the":[45,56,80,93,123],"advantages":[46],"of":[47,114,122,135,145],"convolutions":[48],"and":[49,148,172,180],"self-attention":[50,62,115],"better":[53],"speed-accuracy":[54],"trade-off,":[55],"expensive":[57],"matrix":[58,82],"multiplication":[59,83],"operations":[60,84],"remain":[63],"bottleneck.":[65],"In":[66],"this":[67],"work,":[68],"we":[69,131],"introduce":[70],"novel":[72],"efficient":[73,112,128],"additive":[74,129],"attention":[75],"mechanism":[76],"that":[77,92],"effectively":[78],"replaces":[79],"linear":[86,101],"element-wise":[87],"multiplications.":[88],"Our":[89,152,178],"design":[90],"shows":[91],"key-value":[94],"interaction":[95],"can":[96],"be":[97],"replaced":[98],"layer":[102],"without":[103],"sacrificing":[104],"any":[105],"accuracy.":[106],"Unlike":[107],"previous":[108],"state-of-the-art":[109,141],"methods,":[110],"our":[111,126],"formulation":[113],"enables":[116],"usage":[118],"at":[119],"all":[120],"stages":[121],"network.":[124],"Using":[125],"attention,":[130],"build":[132],"series":[134],"models":[136],"called":[137],"\"SwiftFormer\"":[138],"which":[139,168],"achieves":[140,155],"performance":[142],"terms":[144],"both":[146],"accuracy":[147,159],"inference":[150],"speed.":[151],"small":[153],"variant":[154],"78.5%":[156],"top-1":[157],"ImageNet-1K":[158],"only":[161],"0.8":[162],"ms":[163],"latency":[164],"iPhone":[166],"14,":[167],"is":[169],"more":[170],"accurate":[171],"2\u00d7":[173],"faster":[174],"compared":[175],"MobileViT-v2.":[177],"code":[179],"models:":[181],"https://tinyurl.com/5ft8v46w":[182]},"cited_by_api_url":"https://api.openalex.org/works?filter=cites:W4390872447","counts_by_year":[{"year":2024,"cited_by_count":38},{"year":2023,"cited_by_count":2}],"updated_date":"2025-01-04T13:21:34.537625","created_date":"2024-01-16"}