{"id":"https://openalex.org/W4297632039","doi":"https://doi.org/10.48550/arxiv.2209.01620","title":"MAFormer: A Transformer Network with Multi-scale Attention Fusion for Visual Recognition","display_name":"MAFormer: A Transformer Network with Multi-scale Attention Fusion for Visual Recognition","publication_year":2022,"publication_date":"2022-08-31","ids":{"openalex":"https://openalex.org/W4297632039","doi":"https://doi.org/10.48550/arxiv.2209.01620"},"language":"en","primary_location":{"id":"pmh:oai:arXiv.org:2209.01620","is_oa":true,"landing_page_url":"http://arxiv.org/abs/2209.01620","pdf_url":"https://arxiv.org/pdf/2209.01620","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":[],"type":"repository"},"license":"cc-by","license_id":"https://openalex.org/licenses/cc-by","version":"submittedVersion","is_accepted":false,"is_published":false,"raw_source_name":"","raw_type":"text"},"type":"preprint","indexed_in":["arxiv","datacite"],"open_access":{"is_oa":true,"oa_status":"green","oa_url":"https://arxiv.org/pdf/2209.01620","any_repository_has_fulltext":true},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5101775309","display_name":"Yunhao Wang","orcid":"https://orcid.org/0000-0002-4308-1555"},"institutions":[],"countries":[],"is_corresponding":true,"raw_author_name":"Wang, Yunhao","raw_affiliation_strings":[],"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5100567878","display_name":"Huixin Sun","orcid":"https://orcid.org/0009-0005-3950-4163"},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Sun, Huixin","raw_affiliation_strings":[],"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5100760481","display_name":"Xiaodi Wang","orcid":"https://orcid.org/0000-0001-5854-2003"},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Wang, Xiaodi","raw_affiliation_strings":[],"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5100392769","display_name":"Bin Zhang","orcid":"https://orcid.org/0000-0001-5305-2494"},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Zhang, Bin","raw_affiliation_strings":[],"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5100323094","display_name":"Chao Li","orcid":"https://orcid.org/0000-0001-6110-6210"},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Li, Chao","raw_affiliation_strings":[],"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5038696088","display_name":"Ying Xin","orcid":"https://orcid.org/0000-0001-7591-9423"},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Xin, Ying","raw_affiliation_strings":[],"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5015525872","display_name":"Baochang Zhang","orcid":"https://orcid.org/0000-0001-7396-6218"},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Zhang, Baochang","raw_affiliation_strings":[],"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5050031109","display_name":"Errui Ding","orcid":"https://orcid.org/0000-0002-1867-5378"},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Ding, Errui","raw_affiliation_strings":[],"affiliations":[]},{"author_position":"last","author":{"id":"https://openalex.org/A5101869826","display_name":"Shumin Han","orcid":"https://orcid.org/0000-0001-5390-0182"},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Han, Shumin","raw_affiliation_strings":[],"affiliations":[]}],"institutions":[],"countries_distinct_count":0,"institutions_distinct_count":9,"corresponding_author_ids":["https://openalex.org/A5101775309"],"corresponding_institution_ids":[],"apc_list":null,"apc_paid":null,"fwci":null,"has_fulltext":true,"cited_by_count":4,"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":0.9994000196456909,"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":0.9994000196456909,"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":0.9944999814033508,"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":0.9939000010490417,"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"}}],"keywords":[{"id":"https://openalex.org/keywords/computer-science","display_name":"Computer science","score":0.7922537326812744},{"id":"https://openalex.org/keywords/transformer","display_name":"Transformer","score":0.6739199161529541},{"id":"https://openalex.org/keywords/feature-learning","display_name":"Feature learning","score":0.6427141427993774},{"id":"https://openalex.org/keywords/artificial-intelligence","display_name":"Artificial intelligence","score":0.6385718584060669},{"id":"https://openalex.org/keywords/security-token","display_name":"Security token","score":0.6295715570449829},{"id":"https://openalex.org/keywords/feature-extraction","display_name":"Feature extraction","score":0.5298623442649841},{"id":"https://openalex.org/keywords/deep-learning","display_name":"Deep learning","score":0.4334186315536499},{"id":"https://openalex.org/keywords/pattern-recognition","display_name":"Pattern recognition (psychology)","score":0.42748144268989563},{"id":"https://openalex.org/keywords/segmentation","display_name":"Segmentation","score":0.4222601056098938},{"id":"https://openalex.org/keywords/computer-vision","display_name":"Computer vision","score":0.3338030278682709},{"id":"https://openalex.org/keywords/engineering","display_name":"Engineering","score":0.1006627082824707}],"concepts":[{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.7922537326812744},{"id":"https://openalex.org/C66322947","wikidata":"https://www.wikidata.org/wiki/Q11658","display_name":"Transformer","level":3,"score":0.6739199161529541},{"id":"https://openalex.org/C59404180","wikidata":"https://www.wikidata.org/wiki/Q17013334","display_name":"Feature learning","level":2,"score":0.6427141427993774},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.6385718584060669},{"id":"https://openalex.org/C48145219","wikidata":"https://www.wikidata.org/wiki/Q1335365","display_name":"Security token","level":2,"score":0.6295715570449829},{"id":"https://openalex.org/C52622490","wikidata":"https://www.wikidata.org/wiki/Q1026626","display_name":"Feature extraction","level":2,"score":0.5298623442649841},{"id":"https://openalex.org/C108583219","wikidata":"https://www.wikidata.org/wiki/Q197536","display_name":"Deep learning","level":2,"score":0.4334186315536499},{"id":"https://openalex.org/C153180895","wikidata":"https://www.wikidata.org/wiki/Q7148389","display_name":"Pattern recognition (psychology)","level":2,"score":0.42748144268989563},{"id":"https://openalex.org/C89600930","wikidata":"https://www.wikidata.org/wiki/Q1423946","display_name":"Segmentation","level":2,"score":0.4222601056098938},{"id":"https://openalex.org/C31972630","wikidata":"https://www.wikidata.org/wiki/Q844240","display_name":"Computer vision","level":1,"score":0.3338030278682709},{"id":"https://openalex.org/C127413603","wikidata":"https://www.wikidata.org/wiki/Q11023","display_name":"Engineering","level":0,"score":0.1006627082824707},{"id":"https://openalex.org/C38652104","wikidata":"https://www.wikidata.org/wiki/Q3510521","display_name":"Computer security","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/C119599485","wikidata":"https://www.wikidata.org/wiki/Q43035","display_name":"Electrical engineering","level":1,"score":0.0}],"mesh":[],"locations_count":2,"locations":[{"id":"pmh:oai:arXiv.org:2209.01620","is_oa":true,"landing_page_url":"http://arxiv.org/abs/2209.01620","pdf_url":"https://arxiv.org/pdf/2209.01620","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":[],"type":"repository"},"license":"cc-by","license_id":"https://openalex.org/licenses/cc-by","version":"submittedVersion","is_accepted":false,"is_published":false,"raw_source_name":"","raw_type":"text"},{"id":"doi:10.48550/arxiv.2209.01620","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2209.01620","pdf_url":null,"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":[],"type":"repository"},"license":"cc-by","license_id":"https://openalex.org/licenses/cc-by","version":null,"is_accepted":false,"is_published":null,"raw_source_name":null,"raw_type":"article"}],"best_oa_location":{"id":"pmh:oai:arXiv.org:2209.01620","is_oa":true,"landing_page_url":"http://arxiv.org/abs/2209.01620","pdf_url":"https://arxiv.org/pdf/2209.01620","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":[],"type":"repository"},"license":"cc-by","license_id":"https://openalex.org/licenses/cc-by","version":"submittedVersion","is_accepted":false,"is_published":false,"raw_source_name":"","raw_type":"text"},"sustainable_development_goals":[],"awards":[],"funders":[],"has_content":{"pdf":true,"grobid_xml":true},"content_urls":{"pdf":"https://content.openalex.org/works/W4297632039.pdf","grobid_xml":"https://content.openalex.org/works/W4297632039.grobid-xml"},"referenced_works_count":0,"referenced_works":[],"related_works":["https://openalex.org/W4388335561","https://openalex.org/W2970530566","https://openalex.org/W4288261899","https://openalex.org/W4307309205","https://openalex.org/W2967478618","https://openalex.org/W4385009901","https://openalex.org/W3016124757","https://openalex.org/W3048601286","https://openalex.org/W2965925734","https://openalex.org/W4309346246"],"abstract_inverted_index":{"Vision":[0],"Transformer":[1],"and":[2,36,58,89,96,180,184,201],"its":[3],"variants":[4],"have":[5],"demonstrated":[6],"great":[7],"potential":[8,80,211],"in":[9,62],"various":[10],"computer":[11],"vision":[12,16,167],"tasks.":[13,168],"But":[14],"conventional":[15],"transformers":[17,82],"often":[18],"focus":[19],"on":[20,33,165,176,198,203],"global":[21,34,59,125],"dependency":[22],"at":[23,39,92],"a":[24,30,40,63,71,93,109,129,148,214],"coarse":[25],"level,":[26],"which":[27,54],"suffer":[28],"from":[29],"learning":[31,87],"challenge":[32],"relationships":[35],"fine-grained":[37,88,121],"representation":[38,85],"token":[41,94],"level.":[42],"In":[43,169],"this":[44],"paper,":[45],"we":[46],"introduce":[47],"Multi-scale":[48,101],"Attention":[49,102],"Fusion":[50,103],"into":[51],"transformer":[52],"(MAFormer),":[53],"explores":[55],"local":[56,110,122],"aggregation":[57],"feature":[60,126],"extraction":[61,127],"dual-stream":[64],"framework":[65],"for":[66,83],"visual":[67,84],"recognition.":[68],"We":[69],"develop":[70],"simple":[72],"but":[73],"effective":[74],"module":[75,150],"to":[76,137,212],"explore":[77],"the":[78,144,153,191,210],"full":[79],"of":[81,155],"by":[86,182,195],"coarse-grained":[90],"features":[91,157],"level":[95],"dynamically":[97],"fusing":[98],"them.":[99],"Our":[100,160],"(MAF)":[104],"block":[105],"consists":[106],"of:":[107],"i)":[108],"window":[111],"attention":[112],"branch":[113],"that":[114,151],"learns":[115],"short-range":[116],"interactions":[117],"within":[118,143],"windows,":[119],"aggregating":[120],"features;":[123],"ii)":[124],"through":[128],"novel":[130],"Global":[131],"Learning":[132],"with":[133,206],"Down-sampling":[134],"(GLD)":[135],"operation":[136],"efficiently":[138],"capture":[139],"long-range":[140],"context":[141],"information":[142],"whole":[145],"image;":[146],"iii)":[147],"fusion":[149],"self-explores":[152],"integration":[154],"both":[156],"via":[158],"attention.":[159],"MAFormer":[161,189],"achieves":[162,172],"state-of-the-art":[163],"performance":[164],"common":[166],"particular,":[170],"MAFormer-L":[171],"85.9$\\%$":[173],"Top-1":[174],"accuracy":[175],"ImageNet,":[177],"surpassing":[178],"CSWin-B":[179],"LV-ViT-L":[181],"1.7$\\%$":[183,196],"0.6$\\%$":[185],"respectively.":[186],"On":[187],"MSCOCO,":[188],"outperforms":[190],"prior":[192],"art":[193],"CSWin":[194],"mAPs":[197],"object":[199],"detection":[200],"1.4$\\%$":[202],"instance":[204],"segmentation":[205],"similar-sized":[207],"parameters,":[208],"demonstrating":[209],"be":[213],"general":[215],"backbone":[216],"network.":[217]},"counts_by_year":[{"year":2024,"cited_by_count":2},{"year":2023,"cited_by_count":2}],"updated_date":"2026-03-10T16:38:18.471706","created_date":"2025-10-10T00:00:00"}