{"id":"https://openalex.org/W4402727538","doi":"https://doi.org/10.1109/cvpr52733.2024.00550","title":"SHViT: Single-Head Vision Transformer with Memory Efficient Macro Design","display_name":"SHViT: Single-Head Vision Transformer with Memory Efficient Macro Design","publication_year":2024,"publication_date":"2024-06-16","ids":{"openalex":"https://openalex.org/W4402727538","doi":"https://doi.org/10.1109/cvpr52733.2024.00550"},"language":"en","primary_location":{"id":"doi:10.1109/cvpr52733.2024.00550","is_oa":false,"landing_page_url":"https://doi.org/10.1109/cvpr52733.2024.00550","pdf_url":null,"source":null,"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"2024 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR)","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/A5114123474","display_name":"Seokju Yun","orcid":null},"institutions":[{"id":"https://openalex.org/I124633538","display_name":"University of Seoul","ror":"https://ror.org/05en5nh73","country_code":"KR","type":"education","lineage":["https://openalex.org/I124633538"]}],"countries":["KR"],"is_corresponding":false,"raw_author_name":"Seokju Yun","raw_affiliation_strings":["University of Seoul,Machine Intelligence Laboratory,Korea"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"University of Seoul,Machine Intelligence Laboratory,Korea","institution_ids":["https://openalex.org/I124633538"]}]},{"author_position":"last","author":{"id":"https://openalex.org/A5078628073","display_name":"Youngmin Ro","orcid":"https://orcid.org/0000-0002-0661-9192"},"institutions":[{"id":"https://openalex.org/I124633538","display_name":"University of Seoul","ror":"https://ror.org/05en5nh73","country_code":"KR","type":"education","lineage":["https://openalex.org/I124633538"]}],"countries":["KR"],"is_corresponding":false,"raw_author_name":"Youngmin Ro","raw_affiliation_strings":["University of Seoul,Machine Intelligence Laboratory,Korea"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"University of Seoul,Machine Intelligence Laboratory,Korea","institution_ids":["https://openalex.org/I124633538"]}]}],"institutions":[],"countries_distinct_count":1,"institutions_distinct_count":1,"corresponding_author_ids":[],"corresponding_institution_ids":["https://openalex.org/I124633538"],"apc_list":null,"apc_paid":null,"fwci":23.6083,"has_fulltext":false,"cited_by_count":130,"citation_normalized_percentile":{"value":0.99813192,"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":"5756","last_page":"5767"},"is_retracted":false,"is_paratext":false,"is_xpac":false,"primary_topic":{"id":"https://openalex.org/T11992","display_name":"CCD and CMOS Imaging Sensors","score":0.9998999834060669,"subfield":{"id":"https://openalex.org/subfields/2208","display_name":"Electrical and Electronic Engineering"},"field":{"id":"https://openalex.org/fields/22","display_name":"Engineering"},"domain":{"id":"https://openalex.org/domains/3","display_name":"Physical Sciences"}},"topics":[{"id":"https://openalex.org/T11992","display_name":"CCD and CMOS Imaging Sensors","score":0.9998999834060669,"subfield":{"id":"https://openalex.org/subfields/2208","display_name":"Electrical and Electronic Engineering"},"field":{"id":"https://openalex.org/fields/22","display_name":"Engineering"},"domain":{"id":"https://openalex.org/domains/3","display_name":"Physical Sciences"}},{"id":"https://openalex.org/T13114","display_name":"Image Processing Techniques and Applications","score":0.9969000220298767,"subfield":{"id":"https://openalex.org/subfields/2214","display_name":"Media Technology"},"field":{"id":"https://openalex.org/fields/22","display_name":"Engineering"},"domain":{"id":"https://openalex.org/domains/3","display_name":"Physical Sciences"}},{"id":"https://openalex.org/T10531","display_name":"Advanced Vision and Imaging","score":0.9961000084877014,"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/computer-science","display_name":"Computer science","score":0.6760217547416687},{"id":"https://openalex.org/keywords/transformer","display_name":"Transformer","score":0.5693308115005493},{"id":"https://openalex.org/keywords/macro","display_name":"Macro","score":0.5330036878585815},{"id":"https://openalex.org/keywords/computer-hardware","display_name":"Computer hardware","score":0.32262977957725525},{"id":"https://openalex.org/keywords/electrical-engineering","display_name":"Electrical engineering","score":0.19944360852241516},{"id":"https://openalex.org/keywords/engineering","display_name":"Engineering","score":0.16640257835388184},{"id":"https://openalex.org/keywords/voltage","display_name":"Voltage","score":0.1281220018863678},{"id":"https://openalex.org/keywords/programming-language","display_name":"Programming language","score":0.0729232132434845}],"concepts":[{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.6760217547416687},{"id":"https://openalex.org/C66322947","wikidata":"https://www.wikidata.org/wiki/Q11658","display_name":"Transformer","level":3,"score":0.5693308115005493},{"id":"https://openalex.org/C166955791","wikidata":"https://www.wikidata.org/wiki/Q629579","display_name":"Macro","level":2,"score":0.5330036878585815},{"id":"https://openalex.org/C9390403","wikidata":"https://www.wikidata.org/wiki/Q3966","display_name":"Computer hardware","level":1,"score":0.32262977957725525},{"id":"https://openalex.org/C119599485","wikidata":"https://www.wikidata.org/wiki/Q43035","display_name":"Electrical engineering","level":1,"score":0.19944360852241516},{"id":"https://openalex.org/C127413603","wikidata":"https://www.wikidata.org/wiki/Q11023","display_name":"Engineering","level":0,"score":0.16640257835388184},{"id":"https://openalex.org/C165801399","wikidata":"https://www.wikidata.org/wiki/Q25428","display_name":"Voltage","level":2,"score":0.1281220018863678},{"id":"https://openalex.org/C199360897","wikidata":"https://www.wikidata.org/wiki/Q9143","display_name":"Programming language","level":1,"score":0.0729232132434845}],"mesh":[],"locations_count":1,"locations":[{"id":"doi:10.1109/cvpr52733.2024.00550","is_oa":false,"landing_page_url":"https://doi.org/10.1109/cvpr52733.2024.00550","pdf_url":null,"source":null,"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"2024 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR)","raw_type":"proceedings-article"}],"best_oa_location":null,"sustainable_development_goals":[{"score":0.8500000238418579,"display_name":"Affordable and clean energy","id":"https://metadata.un.org/sdg/7"}],"awards":[{"id":"https://openalex.org/G2919836347","display_name":null,"funder_award_id":"RS-2022-00166109","funder_id":"https://openalex.org/F4320322120","funder_display_name":"National Research Foundation of Korea"}],"funders":[{"id":"https://openalex.org/F4320322120","display_name":"National Research Foundation of Korea","ror":"https://ror.org/013aysd81"}],"has_content":{"pdf":false,"grobid_xml":false},"content_urls":null,"referenced_works_count":88,"referenced_works":["https://openalex.org/W2108598243","https://openalex.org/W2194775991","https://openalex.org/W2560674852","https://openalex.org/W2765407302","https://openalex.org/W2899663614","https://openalex.org/W2908510526","https://openalex.org/W2946794439","https://openalex.org/W2949736877","https://openalex.org/W2963150697","https://openalex.org/W2963163009","https://openalex.org/W2963263347","https://openalex.org/W2963351448","https://openalex.org/W2990138404","https://openalex.org/W2998508940","https://openalex.org/W3012740297","https://openalex.org/W3033529678","https://openalex.org/W3035414587","https://openalex.org/W3094502228","https://openalex.org/W3096609285","https://openalex.org/W3131500599","https://openalex.org/W3137963805","https://openalex.org/W3138516171","https://openalex.org/W3139049060","https://openalex.org/W3157528469","https://openalex.org/W3167588376","https://openalex.org/W3167976421","https://openalex.org/W3169769781","https://openalex.org/W3175515048","https://openalex.org/W3175544090","https://openalex.org/W3190492058","https://openalex.org/W3207411217","https://openalex.org/W3213165621","https://openalex.org/W4214493665","https://openalex.org/W4214588794","https://openalex.org/W4221144813","https://openalex.org/W4226052632","https://openalex.org/W4226066935","https://openalex.org/W4236965008","https://openalex.org/W4281756776","https://openalex.org/W4281932953","https://openalex.org/W4287022992","https://openalex.org/W4288325606","https://openalex.org/W4297775537","https://openalex.org/W4308558335","https://openalex.org/W4312443924","https://openalex.org/W4312820606","https://openalex.org/W4313170858","https://openalex.org/W4320036918","https://openalex.org/W4320459518","https://openalex.org/W4361021286","https://openalex.org/W4385245566","https://openalex.org/W4386047745","https://openalex.org/W4386076083","https://openalex.org/W4386076539","https://openalex.org/W4390872447","https://openalex.org/W4390872550","https://openalex.org/W6637242042","https://openalex.org/W6638667902","https://openalex.org/W6726497184","https://openalex.org/W6755977528","https://openalex.org/W6757817989","https://openalex.org/W6762585180","https://openalex.org/W6762718338","https://openalex.org/W6764322716","https://openalex.org/W6771626834","https://openalex.org/W6779163297","https://openalex.org/W6780069040","https://openalex.org/W6783944145","https://openalex.org/W6784333009","https://openalex.org/W6788135285","https://openalex.org/W6791705549","https://openalex.org/W6792695861","https://openalex.org/W6794345597","https://openalex.org/W6796191787","https://openalex.org/W6797478244","https://openalex.org/W6797790494","https://openalex.org/W6798046796","https://openalex.org/W6798837711","https://openalex.org/W6800217721","https://openalex.org/W6802648153","https://openalex.org/W6810005255","https://openalex.org/W6810653034","https://openalex.org/W6838160470","https://openalex.org/W6838697126","https://openalex.org/W6839178539","https://openalex.org/W6846793137","https://openalex.org/W6850117044","https://openalex.org/W6851201142"],"related_works":["https://openalex.org/W4391375266","https://openalex.org/W2748952813","https://openalex.org/W2030816003","https://openalex.org/W4239992647","https://openalex.org/W2150013480","https://openalex.org/W1554458299","https://openalex.org/W81423522","https://openalex.org/W1509860481","https://openalex.org/W2488264085","https://openalex.org/W2076325756"],"abstract_inverted_index":{"Recently,":[0],"efficient":[1],"Vision":[2,149],"Transformers":[3],"have":[4],"shown":[5],"great":[6],"performance":[7,74,207],"with":[8,35,79,102],"low":[9],"latency":[10,219],"on":[11,159,179,198,220],"resource-constrained":[12],"devices.":[13],"Conventionally,":[14],"they":[15],"use":[16],"<tex":[17,164,169,175,213],"xmlns:mml=\"http://www.w3.org/1998/Math/MathML\"":[18,165,170,176,214],"xmlns:xlink=\"http://www.w3.org/1999/xlink\">$4\\times":[19],"4$</tex>":[20],"patch":[21],"embeddings":[22],"and":[23,104,129,137,168,182,195,222],"a":[24,54,120,147],"4-stage":[25],"structure":[26],"at":[27,38,49],"the":[28,39,84,96,109,153],"macro":[29],"level,":[30],"while":[31,187,211],"utilizing":[32],"sophisticated":[33],"attention":[34,93,106,122],"multi-head":[36],"configuration":[37],"micro":[40],"level.":[41],"This":[42],"paper":[43],"aims":[44],"to":[45,209],"address":[46],"computational":[47],"redundancy":[48,82,128],"all":[50],"design":[51],"levels":[52],"in":[53,95,108],"memory-efficient":[55],"manner.":[56],"We":[57],"discover":[58],"that":[59,92,124,151],"using":[60,201],"larger-stride":[61],"patchify":[62],"stem":[63],"not":[64],"only":[65],"reduces":[66],"memory":[67],"access":[68],"costs":[69],"but":[70],"also":[71],"achieves":[72,206],"competitive":[73],"by":[75,133],"leveraging":[76],"token":[77],"representations":[78],"reduced":[80],"spatial":[81],"from":[83],"early":[85,97],"stages.":[86],"Furthermore,":[87],"our":[88,142,161,204],"preliminary":[89],"analyses":[90],"suggest":[91],"layers":[94],"stages":[98,111],"can":[99],"be":[100],"substituted":[101],"convolutions,":[103],"several":[105],"heads":[107],"latter":[110],"are":[112],"computationally":[113],"redundant.":[114],"To":[115],"handle":[116],"this,":[117],"we":[118,144],"introduce":[119,145],"single-head":[121],"module":[123],"inherently":[125],"prevents":[126],"head":[127],"simultaneously":[130],"boosts":[131],"accuracy":[132],"parallelly":[134],"combining":[135],"global":[136],"local":[138],"information.":[139],"Building":[140],"upon":[141],"solutions,":[143],"SHViT,":[146],"Single-Head":[148],"Transformer":[150],"obtains":[152],"state-of-the-art":[154],"speed-accuracy":[155],"tradeoff.":[156],"For":[157,192],"example,":[158],"ImageNet-1k,":[160],"SHViT-S4":[162],"is":[163],"xmlns:xlink=\"http://www.w3.org/1999/xlink\">$3.3\\times,":[166],"8.l\\times$</tex>,":[167],"xmlns:xlink=\"http://www.w3.org/1999/xlink\">$2.4\\times":[171],"faster$</tex>":[172],"than":[173],"MobileViTv2":[174],"xmlns:xlink=\"http://www.w3.org/1999/xlink\">$\\times":[177],"l.0$</tex>":[178],"GPU,":[180],"CPU,":[181],"iPhone12":[183],"mobile":[184,223],"device,":[185,224],"respectively,":[186],"being":[188],"1.3%":[189],"more":[190],"accurate.":[191],"object":[193],"detection":[194],"instance":[196],"segmentation":[197],"MS":[199],"COCO":[200],"Mask-RCNN":[202],"head,":[203],"model":[205],"comparable":[208],"FastViT-SA12":[210],"exhibiting":[212],"xmlns:xlink=\"http://www.w3.org/1999/xlink\">$3.8\\times":[215],"and2.0\\times":[216],"lower$</tex>":[217],"backbone":[218],"GPU":[221],"respectively.":[225]},"counts_by_year":[{"year":2026,"cited_by_count":42},{"year":2025,"cited_by_count":80},{"year":2024,"cited_by_count":8}],"updated_date":"2026-06-26T08:34:08.712188","created_date":"2025-10-10T00:00:00"}