{"id":"https://openalex.org/W3215758366","doi":null,"title":"Improved Multiscale Vision Transformers for Classification and Detection","display_name":"Improved Multiscale Vision Transformers for Classification and Detection","publication_year":2021,"publication_date":"2021-12-02","ids":{"openalex":"https://openalex.org/W3215758366","mag":"3215758366"},"language":"en","primary_location":{"id":"mag:3215758366","doi":null,"is_oa":true,"landing_page_url":"https://arxiv.org/pdf/2112.01526","pdf_url":null,"source":{"id":"https://openalex.org/S4306400194","display_name":"arXiv (Cornell University)","issn_l":"2331-8422","issn":["2331-8422"],"is_oa":true,"is_in_doaj":false,"is_core":false,"listed_in":[],"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},"volume":null,"type":"preprint","is_traditional":true,"indexed_in":[],"open_access":{"is_oa":true,"oa_status":"green","oa_url":"https://arxiv.org/pdf/2112.01526","any_repository_has_fulltext":true},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5029760000","display_name":"Yanghao Li","orcid":"https://orcid.org/0000-0002-5274-1367","observed_orcids":["https://orcid.org/0000-0002-5274-1367"]},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Yanghao Li","raw_affiliation_strings":[],"raw_orcid":"https://orcid.org/0000-0002-5274-1367","affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5023917342","display_name":"Chao-Yuan Wu","orcid":"https://orcid.org/0000-0002-5690-8865","observed_orcids":["https://orcid.org/0000-0002-5690-8865"]},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Chao-Yuan Wu","raw_affiliation_strings":[],"raw_orcid":"https://orcid.org/0000-0002-5690-8865","affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5022792966","display_name":"Haoqi Fan","orcid":null,"observed_orcids":[]},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Haoqi Fan","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5038469427","display_name":"Karttikeya Mangalam","orcid":"https://orcid.org/0000-0002-2169-1395","observed_orcids":["https://orcid.org/0000-0002-2169-1395"]},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Karttikeya Mangalam","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5102827408","display_name":"Bo Xiong","orcid":"https://orcid.org/0000-0001-7161-3144","observed_orcids":["https://orcid.org/0000-0001-7161-3144"]},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Bo Xiong","raw_affiliation_strings":[],"raw_orcid":"https://orcid.org/0000-0001-7161-3144","affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5001594573","display_name":"Jitendra Malik","orcid":"https://orcid.org/0000-0003-3695-1580","observed_orcids":["https://orcid.org/0000-0003-3695-1580"]},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Jitendra Malik","raw_affiliation_strings":[],"raw_orcid":"https://orcid.org/0000-0003-3695-1580","affiliations":[]},{"author_position":"last","author":{"id":"https://openalex.org/A5036069974","display_name":"Christoph Feichtenhofer","orcid":"https://orcid.org/0000-0001-9756-7238","observed_orcids":["https://orcid.org/0000-0001-9756-7238"]},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Christoph Feichtenhofer","raw_affiliation_strings":[],"raw_orcid":"https://orcid.org/0000-0001-9756-7238","affiliations":[]}],"institutions":[],"countries_distinct_count":0,"institutions_distinct_count":0,"corresponding_author_ids":[],"corresponding_institution_ids":[],"apc_list":null,"apc_paid":null,"fwci":null,"has_fulltext":false,"cited_by_count":23,"citation_normalized_percentile":null,"cited_by_percentile_year":null,"biblio":{"volume":null,"issue":null,"first_page":null,"last_page":null},"is_retracted":false,"is_paratext":false,"is_xpac":false,"primary_topic":{"id":"https://openalex.org/T10036","display_name":"Advanced Neural Network Applications","score":0.9951000213623047,"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":0.9951000213623047,"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":0.0017000000225380063,"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/T11605","display_name":"Visual Attention and Saliency Detection","score":0.0015999999595806003,"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/window-attention","display_name":"window attention","score":0.9210000038146973},{"id":"https://openalex.org/keywords/image-classification","display_name":"image classification","score":0.7960000038146973},{"id":"https://openalex.org/keywords/video-classification","display_name":"video classification","score":0.8849999904632568},{"id":"https://openalex.org/keywords/object-detection","display_name":"object detection","score":0.9290000200271606},{"id":"https://openalex.org/keywords/coco","display_name":"COCO","score":0.6620000004768372},{"id":"https://openalex.org/keywords/imagenet","display_name":"ImageNet","score":0.8550000190734863},{"id":"https://openalex.org/keywords/kinetics-400","display_name":"Kinetics-400","score":0.9070000052452087}],"concepts":[{"id":"https://openalex.org/C70437156","wikidata":"https://www.wikidata.org/wiki/Q7228652","display_name":"Pooling","level":2,"score":0.8707983493804932},{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.7679842710494995},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.7185425162315369},{"id":"https://openalex.org/C75294576","wikidata":"https://www.wikidata.org/wiki/Q5165192","display_name":"Contextual image classification","level":3,"score":0.616287887096405},{"id":"https://openalex.org/C153180895","wikidata":"https://www.wikidata.org/wiki/Q7148389","display_name":"Pattern recognition (psychology)","level":2,"score":0.6048628687858582},{"id":"https://openalex.org/C2776151529","wikidata":"https://www.wikidata.org/wiki/Q3045304","display_name":"Object detection","level":3,"score":0.5885235667228699},{"id":"https://openalex.org/C155512373","wikidata":"https://www.wikidata.org/wiki/Q287450","display_name":"Residual","level":2,"score":0.4593198299407959},{"id":"https://openalex.org/C66322947","wikidata":"https://www.wikidata.org/wiki/Q11658","display_name":"Transformer","level":3,"score":0.44899648427963257},{"id":"https://openalex.org/C123657996","wikidata":"https://www.wikidata.org/wiki/Q12271","display_name":"Architecture","level":2,"score":0.43752095103263855},{"id":"https://openalex.org/C119857082","wikidata":"https://www.wikidata.org/wiki/Q2539","display_name":"Machine learning","level":1,"score":0.41718801856040955},{"id":"https://openalex.org/C31972630","wikidata":"https://www.wikidata.org/wiki/Q844240","display_name":"Computer vision","level":1,"score":0.339822918176651},{"id":"https://openalex.org/C115961682","wikidata":"https://www.wikidata.org/wiki/Q860623","display_name":"Image (mathematics)","level":2,"score":0.24591422080993652},{"id":"https://openalex.org/C11413529","wikidata":"https://www.wikidata.org/wiki/Q8366","display_name":"Algorithm","level":1,"score":0.08989140391349792},{"id":"https://openalex.org/C127413603","wikidata":"https://www.wikidata.org/wiki/Q11023","display_name":"Engineering","level":0,"score":0.07585370540618896},{"id":"https://openalex.org/C153349607","wikidata":"https://www.wikidata.org/wiki/Q36649","display_name":"Visual arts","level":1,"score":0.0},{"id":"https://openalex.org/C165801399","wikidata":"https://www.wikidata.org/wiki/Q25428","display_name":"Voltage","level":2,"score":0.0},{"id":"https://openalex.org/C142362112","wikidata":"https://www.wikidata.org/wiki/Q735","display_name":"Art","level":0,"score":0.0},{"id":"https://openalex.org/C119599485","wikidata":"https://www.wikidata.org/wiki/Q43035","display_name":"Electrical engineering","level":1,"score":0.0}],"mesh":[],"locations_count":1,"locations":[{"id":"mag:3215758366","doi":null,"is_oa":true,"landing_page_url":"https://arxiv.org/pdf/2112.01526","pdf_url":null,"source":{"id":"https://openalex.org/S4306400194","display_name":"arXiv (Cornell University)","issn_l":"2331-8422","issn":["2331-8422"],"is_oa":true,"is_in_doaj":false,"is_core":false,"listed_in":[],"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}],"best_oa_location":{"id":"mag:3215758366","doi":null,"is_oa":true,"landing_page_url":"https://arxiv.org/pdf/2112.01526","pdf_url":null,"source":{"id":"https://openalex.org/S4306400194","display_name":"arXiv (Cornell University)","issn_l":"2331-8422","issn":["2331-8422"],"is_oa":true,"is_in_doaj":false,"is_core":false,"listed_in":[],"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},"sustainable_development_goals":[],"study_designs":[],"awards":[],"funders":[],"has_content":{"grobid_xml":false,"pdf":false},"content_urls":null,"referenced_works_count":86,"referenced_works":["https://openalex.org/W639708223","https://openalex.org/W1483870316","https://openalex.org/W1536680647","https://openalex.org/W1677182931","https://openalex.org/W1686810756","https://openalex.org/W1861492603","https://openalex.org/W2097117768","https://openalex.org/W2108598243","https://openalex.org/W2147800946","https://openalex.org/W2154579312","https://openalex.org/W2156303437","https://openalex.org/W2163605009","https://openalex.org/W2194775991","https://openalex.org/W2302255633","https://openalex.org/W2331143823","https://openalex.org/W2342662179","https://openalex.org/W2518108298","https://openalex.org/W2549139847","https://openalex.org/W2553594924","https://openalex.org/W2565639579","https://openalex.org/W2619082050","https://openalex.org/W2619947201","https://openalex.org/W2622263826","https://openalex.org/W2625366777","https://openalex.org/W2626778328","https://openalex.org/W2743706233","https://openalex.org/W2768282280","https://openalex.org/W2770804203","https://openalex.org/W2773514261","https://openalex.org/W2789541106","https://openalex.org/W2793904650","https://openalex.org/W2935837427","https://openalex.org/W2946948417","https://openalex.org/W2949605076","https://openalex.org/W2955874753","https://openalex.org/W2961193895","https://openalex.org/W2963150697","https://openalex.org/W2963246338","https://openalex.org/W2963399829","https://openalex.org/W2963563276","https://openalex.org/W2963849369","https://openalex.org/W2964241181","https://openalex.org/W2981385151","https://openalex.org/W2984287396","https://openalex.org/W2988396473","https://openalex.org/W2990152177","https://openalex.org/W2990503944","https://openalex.org/W2992308087","https://openalex.org/W2998508940","https://openalex.org/W3015538334","https://openalex.org/W3016265891","https://openalex.org/W3016719260","https://openalex.org/W3035303837","https://openalex.org/W3035452548","https://openalex.org/W3094502228","https://openalex.org/W3102631365","https://openalex.org/W3111156583","https://openalex.org/W3114896399","https://openalex.org/W3116489684","https://openalex.org/W3121523901","https://openalex.org/W3126337037","https://openalex.org/W3126721948","https://openalex.org/W3128633047","https://openalex.org/W3133696297","https://openalex.org/W3134206242","https://openalex.org/W3138516171","https://openalex.org/W3139633126","https://openalex.org/W3146097248","https://openalex.org/W3147387781","https://openalex.org/W3151130473","https://openalex.org/W3153842237","https://openalex.org/W3161838454","https://openalex.org/W3170841864","https://openalex.org/W3171087525","https://openalex.org/W3171206729","https://openalex.org/W3173407577","https://openalex.org/W3173621652","https://openalex.org/W3175515048","https://openalex.org/W3175859344","https://openalex.org/W3203606893","https://openalex.org/W3204076343","https://openalex.org/W3206471682","https://openalex.org/W3211432419","https://openalex.org/W3216272314","https://openalex.org/W4312560592","https://openalex.org/W4313007769"],"related_works":["https://openalex.org/W4312769131","https://openalex.org/W3156109214","https://openalex.org/W4312772544","https://openalex.org/W4312847199","https://openalex.org/W4367692538","https://openalex.org/W4312925417","https://openalex.org/W3206263120","https://openalex.org/W4390874319","https://openalex.org/W4293067824","https://openalex.org/W3147387781"],"abstract_inverted_index":{"In":[0],"this":[1,42],"paper,":[2],"we":[3],"study":[4],"Multiscale":[5],"Vision":[6],"Transformers":[7],"(MViT)":[8],"as":[9,18,20,102,104],"a":[10],"unified":[11],"architecture":[12,43],"for":[13,50],"image":[14],"and":[15,36,47,55,111],"video":[16,57,108],"classification,":[17,52,94],"well":[19,103],"object":[21,100],"detection.":[22],"We":[23,40,64],"present":[24],"an":[25],"improved":[26],"version":[27],"of":[28],"MViT":[29,83],"that":[30],"incorporates":[31],"decomposed":[32],"relative":[33],"positional":[34],"embeddings":[35],"residual":[37],"pooling":[38,68],"connections.":[39],"instantiate":[41],"in":[44,79,87],"five":[45],"sizes":[46],"evaluate":[48],"it":[49,60,75],"ImageNet":[51,93],"COCO":[53,99],"detection":[54,101],"Kinetics":[56],"recognition":[58],"where":[59,74],"outperforms":[61,76],"prior":[62],"work.":[63],"further":[65],"compare":[66],"MViTs'":[67],"attention":[69,72],"to":[70],"window":[71],"mechanisms":[73],"the":[77],"latter":[78],"accuracy/compute.":[80],"Without":[81],"bells-and-whistles,":[82],"has":[84],"state-of-the-art":[85],"performance":[86],"3":[88],"domains:":[89],"88.8%":[90],"accuracy":[91],"on":[92,98,106],"56.1":[95],"box":[96],"AP":[97],"86.1%":[105],"Kinetics-400":[107],"classification.":[109],"Code":[110],"models":[112],"will":[113],"be":[114],"made":[115],"publicly":[116],"available.":[117]},"counts_by_year":[{"year":2025,"cited_by_count":2},{"year":2024,"cited_by_count":5},{"year":2023,"cited_by_count":10},{"year":2022,"cited_by_count":6}],"updated_date":"2026-11-03T10:50:19.383937","created_date":"2021-12-06T00:00:00"}