{"id":"https://openalex.org/W4408355129","doi":"https://doi.org/10.1109/icassp49660.2025.10889122","title":"Learning Class Unique Features in Fine-Grained Visual Classification","display_name":"Learning Class Unique Features in Fine-Grained Visual Classification","publication_year":2025,"publication_date":"2025-03-12","ids":{"openalex":"https://openalex.org/W4408355129","doi":"https://doi.org/10.1109/icassp49660.2025.10889122"},"language":"en","primary_location":{"id":"doi:10.1109/icassp49660.2025.10889122","is_oa":false,"landing_page_url":"https://doi.org/10.1109/icassp49660.2025.10889122","pdf_url":null,"source":null,"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"ICASSP 2025 - 2025 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP)","raw_type":"proceedings-article"},"type":"conference-paper","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/A5021347960","display_name":"Runkai Zheng","orcid":"https://orcid.org/0000-0003-3120-5466"},"institutions":[{"id":"https://openalex.org/I74973139","display_name":"Carnegie Mellon University","ror":"https://ror.org/05x2bcf33","country_code":"US","type":"education","lineage":["https://openalex.org/I74973139"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Runkai Zheng","raw_affiliation_strings":["Carnegie Mellon University"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Carnegie Mellon University","institution_ids":["https://openalex.org/I74973139"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5100418818","display_name":"Li Liu","orcid":"https://orcid.org/0000-0002-4497-0135"},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Li Liu","raw_affiliation_strings":["The Hong Kong University of Science and Technology,Guangzhou"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"The Hong Kong University of Science and Technology,Guangzhou","institution_ids":[]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5113976460","display_name":"Zhijia Yu","orcid":null},"institutions":[{"id":"https://openalex.org/I4210095451","display_name":"Independent Dance","ror":"https://ror.org/00w7etd19","country_code":"GB","type":"other","lineage":["https://openalex.org/I4210095451"]}],"countries":["GB"],"is_corresponding":false,"raw_author_name":"Zhijia Yu","raw_affiliation_strings":["Byte Dance"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Byte Dance","institution_ids":["https://openalex.org/I4210095451"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5063649822","display_name":"Yinqi Zhang","orcid":"https://orcid.org/0000-0001-5839-2105"},"institutions":[{"id":"https://openalex.org/I66867065","display_name":"East China Normal University","ror":"https://ror.org/02n96ep67","country_code":"CN","type":"education","lineage":["https://openalex.org/I66867065"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Yinqi Zhang","raw_affiliation_strings":["East China Normal University"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"East China Normal University","institution_ids":["https://openalex.org/I66867065"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5081375588","display_name":"Hei Victor Cheng","orcid":"https://orcid.org/0000-0001-8432-3779"},"institutions":[{"id":"https://openalex.org/I204337017","display_name":"Aarhus University","ror":"https://ror.org/01aj84f44","country_code":"DK","type":"education","lineage":["https://openalex.org/I204337017"]}],"countries":["DK"],"is_corresponding":false,"raw_author_name":"Hei Victor Cheng","raw_affiliation_strings":["Aarhus University"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Aarhus University","institution_ids":["https://openalex.org/I204337017"]}]},{"author_position":"last","author":{"id":"https://openalex.org/A5102823200","display_name":"Chris Ding","orcid":"https://orcid.org/0009-0009-3374-1941"},"institutions":[{"id":"https://openalex.org/I4210116924","display_name":"Chinese University of Hong Kong, Shenzhen","ror":"https://ror.org/02d5ks197","country_code":"CN","type":"education","lineage":["https://openalex.org/I177725633","https://openalex.org/I180726961","https://openalex.org/I4210116924"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Chris Ding","raw_affiliation_strings":["The Chinese University of Hong Kong,Shenzhen"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"The Chinese University of Hong Kong,Shenzhen","institution_ids":["https://openalex.org/I4210116924"]}]}],"institutions":[],"countries_distinct_count":4,"institutions_distinct_count":5,"corresponding_author_ids":[],"corresponding_institution_ids":[],"apc_list":null,"apc_paid":null,"fwci":null,"has_fulltext":false,"cited_by_count":0,"citation_normalized_percentile":null,"cited_by_percentile_year":null,"biblio":{"volume":null,"issue":null,"first_page":"1","last_page":"5"},"is_retracted":false,"is_paratext":false,"is_xpac":false,"primary_topic":{"id":"https://openalex.org/T10057","display_name":"Face and Expression Recognition","score":0.8029000163078308,"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/T10057","display_name":"Face and Expression Recognition","score":0.8029000163078308,"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/T13114","display_name":"Image Processing Techniques and Applications","score":0.6850000023841858,"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/T10824","display_name":"Image Retrieval and Classification Techniques","score":0.6575999855995178,"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":0.7207471132278442},{"id":"https://openalex.org/keywords/class","display_name":"Class (philosophy)","score":0.611477792263031},{"id":"https://openalex.org/keywords/artificial-intelligence","display_name":"Artificial intelligence","score":0.5338175296783447},{"id":"https://openalex.org/keywords/pattern-recognition","display_name":"Pattern recognition (psychology)","score":0.336097776889801}],"concepts":[{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.7207471132278442},{"id":"https://openalex.org/C2777212361","wikidata":"https://www.wikidata.org/wiki/Q5127848","display_name":"Class (philosophy)","level":2,"score":0.611477792263031},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.5338175296783447},{"id":"https://openalex.org/C153180895","wikidata":"https://www.wikidata.org/wiki/Q7148389","display_name":"Pattern recognition (psychology)","level":2,"score":0.336097776889801}],"mesh":[],"locations_count":1,"locations":[{"id":"doi:10.1109/icassp49660.2025.10889122","is_oa":false,"landing_page_url":"https://doi.org/10.1109/icassp49660.2025.10889122","pdf_url":null,"source":null,"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"ICASSP 2025 - 2025 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP)","raw_type":"proceedings-article"}],"best_oa_location":null,"sustainable_development_goals":[{"display_name":"Quality Education","score":0.6100000143051147,"id":"https://metadata.un.org/sdg/4"}],"awards":[],"funders":[{"id":"https://openalex.org/F4320321001","display_name":"National Natural Science Foundation of China","ror":"https://ror.org/01h0zpd94"}],"has_content":{"grobid_xml":false,"pdf":false},"content_urls":null,"referenced_works_count":39,"referenced_works":["https://openalex.org/W56385144","https://openalex.org/W1984514441","https://openalex.org/W2099111195","https://openalex.org/W2104657103","https://openalex.org/W2138011018","https://openalex.org/W2183341477","https://openalex.org/W2194775991","https://openalex.org/W2737725206","https://openalex.org/W2740620254","https://openalex.org/W2773003563","https://openalex.org/W2798365843","https://openalex.org/W2807931652","https://openalex.org/W2883888092","https://openalex.org/W2891951760","https://openalex.org/W2961018736","https://openalex.org/W2962761264","https://openalex.org/W2963066927","https://openalex.org/W2963393555","https://openalex.org/W2963446712","https://openalex.org/W2981954115","https://openalex.org/W2990495699","https://openalex.org/W2997426000","https://openalex.org/W2997809650","https://openalex.org/W2998345525","https://openalex.org/W2998619563","https://openalex.org/W3009073662","https://openalex.org/W3009337595","https://openalex.org/W3035367622","https://openalex.org/W3130788242","https://openalex.org/W4387969322","https://openalex.org/W6633965997","https://openalex.org/W6637373629","https://openalex.org/W6638319203","https://openalex.org/W6638677478","https://openalex.org/W6677919164","https://openalex.org/W6732696085","https://openalex.org/W6755069125","https://openalex.org/W6766978945","https://openalex.org/W6787972765"],"related_works":["https://openalex.org/W4391375266","https://openalex.org/W2899084033","https://openalex.org/W2748952813","https://openalex.org/W2390279801","https://openalex.org/W4391913857","https://openalex.org/W2358668433","https://openalex.org/W4396701345","https://openalex.org/W2376932109","https://openalex.org/W2033914206","https://openalex.org/W2042327336"],"abstract_inverted_index":{"A":[0],"major":[1],"challenge":[2,33],"in":[3,43,148],"Fine-Grained":[4],"Visual":[5],"Classification":[6],"(FGVC)":[7],"is":[8,89,136],"distinguishing":[9],"various":[10],"categories":[11],"with":[12,94,159,170],"high":[13],"inter-class":[14,40],"similarity":[15],"by":[16,57],"learning":[17],"the":[18,22,53,59,62,117,129,149,152,180],"feature":[19,124],"that":[20,104,173],"differentiates":[21],"details.":[23],"Conventional":[24],"cross-entropy":[25],"trained":[26],"Convolutional":[27],"Neural":[28],"Network":[29],"(CNN)":[30],"fails":[31],"this":[32,46,73,95],"as":[34,138,143],"they":[35],"may":[36],"suffer":[37],"from":[38,67],"producing":[39],"invariant":[41],"features":[42,63,147],"FGVC.":[44],"In":[45],"work,":[47],"we":[48,75,108],"innovatively":[49],"propose":[50],"to":[51,91,99,127],"regularize":[52],"training":[54],"of":[55,61,64],"CNN":[56],"enforcing":[58],"uniqueness":[60],"each":[65,122],"category":[66],"an":[68],"information-theoretic":[69],"perspective.":[70],"To":[71],"achieve":[72],"goal,":[74],"formulate":[76],"a":[77,82,86,102,110,144],"minimax":[78,131],"loss":[79],"based":[80,115],"on":[81,116,165],"game-theoretic":[83],"framework,":[84],"where":[85],"Nash":[87],"equilibrium":[88],"proved":[90],"be":[92,141,157],"consistent":[93],"regularization":[96],"objective.":[97],"Besides,":[98],"avoid":[100],"getting":[101],"solution":[103],"produces":[105],"redundant":[106],"features,":[107],"present":[109],"Feature":[111],"Redundancy":[112],"Loss":[113],"(FRL)":[114],"normalized":[118],"inner":[119],"product":[120],"between":[121],"selected":[123],"map":[125],"pair":[126],"complement":[128],"proposed":[130,134],"loss.":[132],"The":[133],"method":[135,175],"versatile,":[137],"it":[139],"can":[140,156],"utilized":[142],"regularizer":[145],"for":[146],"mid-level":[150],"or":[151],"penultimate":[153],"layer,":[154],"and":[155,186],"combined":[158],"any":[160],"architectures.":[161],"Extensive":[162],"experimental":[163],"results":[164],"several":[166],"influential":[167],"benchmarks":[168],"along":[169],"visualization":[171],"show":[172],"our":[174],"obtains":[176],"significant":[177],"improvement":[178],"over":[179],"baseline":[181],"model":[182],"without":[183],"extra":[184],"cost":[185],"achieves":[187],"state-of-the-art":[188],"results.":[189]},"counts_by_year":[],"updated_date":"2026-07-14T23:27:15.235271","created_date":"2025-10-10T00:00:00"}