{"id":"https://openalex.org/W4392910527","doi":"https://doi.org/10.1109/icassp48485.2024.10447669","title":"Attention Decoupling for Query-Based Object Detection","display_name":"Attention Decoupling for Query-Based Object Detection","publication_year":2024,"publication_date":"2024-03-18","ids":{"openalex":"https://openalex.org/W4392910527","doi":"https://doi.org/10.1109/icassp48485.2024.10447669"},"language":"en","primary_location":{"id":"doi:10.1109/icassp48485.2024.10447669","is_oa":false,"landing_page_url":"https://doi.org/10.1109/icassp48485.2024.10447669","pdf_url":null,"source":null,"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"ICASSP 2024 - 2024 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP)","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/A5057651379","display_name":"Jia-Wei Ma","orcid":"https://orcid.org/0000-0002-7628-6047"},"institutions":[{"id":"https://openalex.org/I92403157","display_name":"University of Science and Technology Beijing","ror":"https://ror.org/02egmk993","country_code":"CN","type":"education","lineage":["https://openalex.org/I92403157"]}],"countries":["CN"],"is_corresponding":true,"raw_author_name":"Jia-Wei Ma","raw_affiliation_strings":["University of Science and Technology Beijing,School of Computer and Communication Engineering","School of Computer and Communication Engineering, University of Science and Technology Beijing"],"affiliations":[{"raw_affiliation_string":"University of Science and Technology Beijing,School of Computer and Communication Engineering","institution_ids":["https://openalex.org/I92403157"]},{"raw_affiliation_string":"School of Computer and Communication Engineering, University of Science and Technology Beijing","institution_ids":["https://openalex.org/I92403157"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5048076905","display_name":"Min Liang","orcid":"https://orcid.org/0000-0003-4749-7635"},"institutions":[{"id":"https://openalex.org/I92403157","display_name":"University of Science and Technology Beijing","ror":"https://ror.org/02egmk993","country_code":"CN","type":"education","lineage":["https://openalex.org/I92403157"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Min Liang","raw_affiliation_strings":["University of Science and Technology Beijing,School of Computer and Communication Engineering","School of Computer and Communication Engineering, University of Science and Technology Beijing"],"affiliations":[{"raw_affiliation_string":"University of Science and Technology Beijing,School of Computer and Communication Engineering","institution_ids":["https://openalex.org/I92403157"]},{"raw_affiliation_string":"School of Computer and Communication Engineering, University of Science and Technology Beijing","institution_ids":["https://openalex.org/I92403157"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5008416026","display_name":"Haixia Man","orcid":"https://orcid.org/0000-0001-7503-0085"},"institutions":[{"id":"https://openalex.org/I109173049","display_name":"Beijing Foreign Studies University","ror":"https://ror.org/00jdr0662","country_code":"CN","type":"education","lineage":["https://openalex.org/I109173049"]},{"id":"https://openalex.org/I92403157","display_name":"University of Science and Technology Beijing","ror":"https://ror.org/02egmk993","country_code":"CN","type":"education","lineage":["https://openalex.org/I92403157"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Haixia Man","raw_affiliation_strings":["University of Science and Technology Beijing,School of Foreign Studies","School of Foreign Studies, University of Science and Technology Beijing"],"affiliations":[{"raw_affiliation_string":"University of Science and Technology Beijing,School of Foreign Studies","institution_ids":["https://openalex.org/I109173049"]},{"raw_affiliation_string":"School of Foreign Studies, University of Science and Technology Beijing","institution_ids":["https://openalex.org/I92403157"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5012986058","display_name":"Shu Tian","orcid":"https://orcid.org/0000-0002-4910-6071"},"institutions":[{"id":"https://openalex.org/I92403157","display_name":"University of Science and Technology Beijing","ror":"https://ror.org/02egmk993","country_code":"CN","type":"education","lineage":["https://openalex.org/I92403157"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Shu Tian","raw_affiliation_strings":["University of Science and Technology Beijing,School of Computer and Communication Engineering","School of Computer and Communication Engineering, University of Science and Technology Beijing"],"affiliations":[{"raw_affiliation_string":"University of Science and Technology Beijing,School of Computer and Communication Engineering","institution_ids":["https://openalex.org/I92403157"]},{"raw_affiliation_string":"School of Computer and Communication Engineering, University of Science and Technology Beijing","institution_ids":["https://openalex.org/I92403157"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5072850755","display_name":"Jingyan Qin","orcid":"https://orcid.org/0000-0002-4101-4316"},"institutions":[{"id":"https://openalex.org/I92403157","display_name":"University of Science and Technology Beijing","ror":"https://ror.org/02egmk993","country_code":"CN","type":"education","lineage":["https://openalex.org/I92403157"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Jingyan Qin","raw_affiliation_strings":["University of Science and Technology Beijing,School of Computer and Communication Engineering","School of Computer and Communication Engineering, University of Science and Technology Beijing"],"affiliations":[{"raw_affiliation_string":"University of Science and Technology Beijing,School of Computer and Communication Engineering","institution_ids":["https://openalex.org/I92403157"]},{"raw_affiliation_string":"School of Computer and Communication Engineering, University of Science and Technology Beijing","institution_ids":["https://openalex.org/I92403157"]}]},{"author_position":"last","author":{"id":"https://openalex.org/A5074514262","display_name":"Xu-Cheng Yin","orcid":"https://orcid.org/0000-0003-0023-0220"},"institutions":[{"id":"https://openalex.org/I92403157","display_name":"University of Science and Technology Beijing","ror":"https://ror.org/02egmk993","country_code":"CN","type":"education","lineage":["https://openalex.org/I92403157"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Xu-Cheng Yin","raw_affiliation_strings":["University of Science and Technology Beijing,School of Computer and Communication Engineering","School of Computer and Communication Engineering, University of Science and Technology Beijing"],"affiliations":[{"raw_affiliation_string":"University of Science and Technology Beijing,School of Computer and Communication Engineering","institution_ids":["https://openalex.org/I92403157"]},{"raw_affiliation_string":"School of Computer and Communication Engineering, University of Science and Technology Beijing","institution_ids":["https://openalex.org/I92403157"]}]}],"institutions":[],"countries_distinct_count":1,"institutions_distinct_count":6,"corresponding_author_ids":["https://openalex.org/A5057651379"],"corresponding_institution_ids":["https://openalex.org/I92403157"],"apc_list":null,"apc_paid":null,"fwci":0.2015,"has_fulltext":false,"cited_by_count":1,"citation_normalized_percentile":{"value":0.45980431,"is_in_top_1_percent":false,"is_in_top_10_percent":false},"cited_by_percentile_year":{"min":91,"max":95},"biblio":{"volume":null,"issue":null,"first_page":"2850","last_page":"2854"},"is_retracted":false,"is_paratext":false,"is_xpac":false,"primary_topic":{"id":"https://openalex.org/T11992","display_name":"CCD and CMOS Imaging Sensors","score":0.9366000294685364,"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":0.9366000294685364,"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/T10036","display_name":"Advanced Neural Network Applications","score":0.9085000157356262,"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":0.7630351781845093},{"id":"https://openalex.org/keywords/decoupling","display_name":"Decoupling (probability)","score":0.645514726638794},{"id":"https://openalex.org/keywords/query-optimization","display_name":"Query optimization","score":0.5320205688476562},{"id":"https://openalex.org/keywords/sargable","display_name":"Sargable","score":0.47769811749458313},{"id":"https://openalex.org/keywords/query-language","display_name":"Query language","score":0.4449637234210968},{"id":"https://openalex.org/keywords/object","display_name":"Object (grammar)","score":0.4280136823654175},{"id":"https://openalex.org/keywords/query-expansion","display_name":"Query expansion","score":0.4214158058166504},{"id":"https://openalex.org/keywords/artificial-intelligence","display_name":"Artificial intelligence","score":0.38183122873306274},{"id":"https://openalex.org/keywords/information-retrieval","display_name":"Information retrieval","score":0.35074537992477417},{"id":"https://openalex.org/keywords/web-search-query","display_name":"Web search query","score":0.33405375480651855},{"id":"https://openalex.org/keywords/search-engine","display_name":"Search engine","score":0.1925903856754303}],"concepts":[{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.7630351781845093},{"id":"https://openalex.org/C205606062","wikidata":"https://www.wikidata.org/wiki/Q5249645","display_name":"Decoupling (probability)","level":2,"score":0.645514726638794},{"id":"https://openalex.org/C157692150","wikidata":"https://www.wikidata.org/wiki/Q2919848","display_name":"Query optimization","level":2,"score":0.5320205688476562},{"id":"https://openalex.org/C192939062","wikidata":"https://www.wikidata.org/wiki/Q104840822","display_name":"Sargable","level":4,"score":0.47769811749458313},{"id":"https://openalex.org/C192028432","wikidata":"https://www.wikidata.org/wiki/Q845739","display_name":"Query language","level":2,"score":0.4449637234210968},{"id":"https://openalex.org/C2781238097","wikidata":"https://www.wikidata.org/wiki/Q175026","display_name":"Object (grammar)","level":2,"score":0.4280136823654175},{"id":"https://openalex.org/C99016210","wikidata":"https://www.wikidata.org/wiki/Q5488129","display_name":"Query expansion","level":2,"score":0.4214158058166504},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.38183122873306274},{"id":"https://openalex.org/C23123220","wikidata":"https://www.wikidata.org/wiki/Q816826","display_name":"Information retrieval","level":1,"score":0.35074537992477417},{"id":"https://openalex.org/C164120249","wikidata":"https://www.wikidata.org/wiki/Q995982","display_name":"Web search query","level":3,"score":0.33405375480651855},{"id":"https://openalex.org/C97854310","wikidata":"https://www.wikidata.org/wiki/Q19541","display_name":"Search engine","level":2,"score":0.1925903856754303},{"id":"https://openalex.org/C133731056","wikidata":"https://www.wikidata.org/wiki/Q4917288","display_name":"Control engineering","level":1,"score":0.0},{"id":"https://openalex.org/C127413603","wikidata":"https://www.wikidata.org/wiki/Q11023","display_name":"Engineering","level":0,"score":0.0}],"mesh":[],"locations_count":1,"locations":[{"id":"doi:10.1109/icassp48485.2024.10447669","is_oa":false,"landing_page_url":"https://doi.org/10.1109/icassp48485.2024.10447669","pdf_url":null,"source":null,"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"ICASSP 2024 - 2024 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP)","raw_type":"proceedings-article"}],"best_oa_location":null,"sustainable_development_goals":[],"awards":[{"id":"https://openalex.org/G1614471940","display_name":null,"funder_award_id":"2020AAA0","funder_id":"https://openalex.org/F4320335777","funder_display_name":"National Key Research and Development Program of China"},{"id":"https://openalex.org/G1749324472","display_name":null,"funder_award_id":"21FZXB020","funder_id":"https://openalex.org/F4320335869","funder_display_name":"National Social Science Fund of China"},{"id":"https://openalex.org/G2141617675","display_name":null,"funder_award_id":"62076024","funder_id":"https://openalex.org/F4320321001","funder_display_name":"National Natural Science Foundation of China"},{"id":"https://openalex.org/G356227064","display_name":null,"funder_award_id":"62006018","funder_id":"https://openalex.org/F4320321001","funder_display_name":"National Natural Science Foundation of China"},{"id":"https://openalex.org/G8772010899","display_name":null,"funder_award_id":"62125601","funder_id":"https://openalex.org/F4320336125","funder_display_name":"National Science Fund for Distinguished Young Scholars"},{"id":"https://openalex.org/G8879444097","display_name":null,"funder_award_id":"2020AAA0109700","funder_id":"https://openalex.org/F4320335777","funder_display_name":"National Key Research and Development Program of China"}],"funders":[{"id":"https://openalex.org/F4320321001","display_name":"National Natural Science Foundation of China","ror":"https://ror.org/01h0zpd94"},{"id":"https://openalex.org/F4320335777","display_name":"National Key Research and Development Program of China","ror":null},{"id":"https://openalex.org/F4320335869","display_name":"National Social Science Fund of China","ror":null},{"id":"https://openalex.org/F4320336125","display_name":"National Science Fund for Distinguished Young Scholars","ror":null},{"id":"https://openalex.org/F4320337504","display_name":"Research and Development","ror":"https://ror.org/027s68j25"}],"has_content":{"grobid_xml":false,"pdf":false},"content_urls":null,"referenced_works_count":42,"referenced_works":["https://openalex.org/W1861492603","https://openalex.org/W2108598243","https://openalex.org/W2194775991","https://openalex.org/W2886904239","https://openalex.org/W2963150697","https://openalex.org/W2963351448","https://openalex.org/W2989676862","https://openalex.org/W3035396860","https://openalex.org/W3035473155","https://openalex.org/W3035694605","https://openalex.org/W3096609285","https://openalex.org/W3107331169","https://openalex.org/W3108849448","https://openalex.org/W3109381875","https://openalex.org/W3171660447","https://openalex.org/W3172087149","https://openalex.org/W3172752666","https://openalex.org/W3176187859","https://openalex.org/W3199093552","https://openalex.org/W3203974803","https://openalex.org/W3205483686","https://openalex.org/W4214489586","https://openalex.org/W4214507171","https://openalex.org/W4214627427","https://openalex.org/W4221138453","https://openalex.org/W4225745741","https://openalex.org/W4312312588","https://openalex.org/W4312412889","https://openalex.org/W4312692509","https://openalex.org/W4385245566","https://openalex.org/W4386065416","https://openalex.org/W4386083115","https://openalex.org/W4389403194","https://openalex.org/W4390872594","https://openalex.org/W4390873545","https://openalex.org/W4390873791","https://openalex.org/W6779586474","https://openalex.org/W6784094891","https://openalex.org/W6804794748","https://openalex.org/W6809665764","https://openalex.org/W6811230874","https://openalex.org/W6841543928"],"related_works":["https://openalex.org/W2096359267","https://openalex.org/W3125756434","https://openalex.org/W2572349046","https://openalex.org/W2392799717","https://openalex.org/W2146885082","https://openalex.org/W2026738364","https://openalex.org/W1981131819","https://openalex.org/W2017989738","https://openalex.org/W2017328855","https://openalex.org/W2186703450"],"abstract_inverted_index":{"Benefiting":[0],"from":[1],"attention":[2,45],"mechanisms,":[3],"query-based":[4,49,110],"detectors":[5,50],"have":[6],"a":[7,60,66,82,89,133],"strong":[8],"model":[9,37],"capacity.":[10],"They":[11],"predict":[12],"classification":[13,95],"and":[14,21,65,72,115],"regression":[15],"by":[16],"utilizing":[17],"their":[18],"shared":[19],"queries":[20],"features":[22],"in":[23,76],"the":[24,128,141],"decoder.":[25],"Inter-task":[26],"biases":[27],"cause":[28],"multi-directional":[29],"gradients":[30],"that":[31,127],"disturb":[32],"each":[33],"other":[34],"to":[35,51,94,97],"limit":[36],"optimization.":[38],"In":[39],"this":[40],"work,":[41],"we":[42,80],"introduce":[43],"an":[44],"decoupling":[46,74],"(AD)":[47],"for":[48],"explicitly":[52],"align":[53],"multi-task":[54,118],"features.":[55],"Specifically,":[56],"AD":[57,107,130],"consists":[58],"of":[59,135],"Dense-to-Sparse":[61],"Query":[62],"Generator":[63],"(DSQG)":[64],"Split":[67],"Cross-Attention":[68],"(SCA),":[69],"enabling":[70],"query":[71],"feature":[73],"respectively":[75],"decoding":[77,104],"phase.":[78],"Then,":[79],"propose":[81],"task":[83,91,100,112],"consistency":[84,101],"loss":[85,96],"(TCL)":[86],"which":[87],"integrates":[88],"novel":[90],"alignment":[92],"metric":[93],"further":[98],"improve":[99],"across":[102],"multiple":[103],"stages.":[105],"Thus,":[106],"effectively":[108],"mitigates":[109],"detectors\u2019":[111],"misalignment":[113],"problem":[114],"inspires":[116],"subsequent":[117],"paradigms.":[119],"Moreover,":[120],"extensive":[121],"experiments":[122],"on":[123],"COCO":[124],"dataset":[125],"demonstrate":[126],"proposed":[129],"can":[131],"enhance":[132],"variety":[134],"representative":[136],"detectors.":[137],"Remarkably,":[138],"AD-DINO":[139],"achieves":[140],"state-of-the-art":[142],"performance.":[143]},"counts_by_year":[{"year":2025,"cited_by_count":1}],"updated_date":"2026-04-21T08:09:41.155169","created_date":"2025-10-10T00:00:00"}