{"id":"https://openalex.org/W4415707832","doi":"https://doi.org/10.1109/icme59968.2025.11209327","title":"ESTJ: Efficient Semantic Segmentation via Token Joint Merging","display_name":"ESTJ: Efficient Semantic Segmentation via Token Joint Merging","publication_year":2025,"publication_date":"2025-06-30","ids":{"openalex":"https://openalex.org/W4415707832","doi":"https://doi.org/10.1109/icme59968.2025.11209327"},"language":null,"primary_location":{"id":"doi:10.1109/icme59968.2025.11209327","is_oa":false,"landing_page_url":"https://doi.org/10.1109/icme59968.2025.11209327","pdf_url":null,"source":null,"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"2025 IEEE International Conference on Multimedia and Expo (ICME)","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/A5074776274","display_name":"Ziniu Liu","orcid":"https://orcid.org/0000-0002-3835-4272"},"institutions":[{"id":"https://openalex.org/I116953780","display_name":"Tongji University","ror":"https://ror.org/03rc6as71","country_code":"CN","type":"education","lineage":["https://openalex.org/I116953780"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Ziniu Liu","raw_affiliation_strings":["Tongji University,School of Computer Science and Technology,Shanghai,China"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Tongji University,School of Computer Science and Technology,Shanghai,China","institution_ids":["https://openalex.org/I116953780"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5081962162","display_name":"Mingqing Liu","orcid":"https://orcid.org/0000-0002-5658-7811"},"institutions":[{"id":"https://openalex.org/I116953780","display_name":"Tongji University","ror":"https://ror.org/03rc6as71","country_code":"CN","type":"education","lineage":["https://openalex.org/I116953780"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Mingqing Liu","raw_affiliation_strings":["Tongji University,Department of Information and Communication Engineering,Shanghai,China"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Tongji University,Department of Information and Communication Engineering,Shanghai,China","institution_ids":["https://openalex.org/I116953780"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5020530995","display_name":"Fengxia Han","orcid":"https://orcid.org/0000-0001-5021-3686"},"institutions":[{"id":"https://openalex.org/I116953780","display_name":"Tongji University","ror":"https://ror.org/03rc6as71","country_code":"CN","type":"education","lineage":["https://openalex.org/I116953780"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Fengxia Han","raw_affiliation_strings":["Tongji University,School of Computer Science and Technology,Shanghai,China"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Tongji University,School of Computer Science and Technology,Shanghai,China","institution_ids":["https://openalex.org/I116953780"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5072774443","display_name":"Xingtong Liu","orcid":"https://orcid.org/0000-0002-8661-7460"},"institutions":[{"id":"https://openalex.org/I116953780","display_name":"Tongji University","ror":"https://ror.org/03rc6as71","country_code":"CN","type":"education","lineage":["https://openalex.org/I116953780"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Xingtong Liu","raw_affiliation_strings":["Tongji University,Guohao College,Shanghai,China"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Tongji University,Guohao College,Shanghai,China","institution_ids":["https://openalex.org/I116953780"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5088211940","display_name":"Chuan Liu","orcid":"https://orcid.org/0000-0002-0870-4382"},"institutions":[{"id":"https://openalex.org/I153473198","display_name":"North China Electric Power University","ror":"https://ror.org/04qr5t414","country_code":"CN","type":"education","lineage":["https://openalex.org/I153473198"]},{"id":"https://openalex.org/I4392738113","display_name":"China Electric Power Research Institute","ror":"https://ror.org/05ehpzy81","country_code":null,"type":"facility","lineage":["https://openalex.org/I17442442","https://openalex.org/I4392738113"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Chuan Liu","raw_affiliation_strings":["China Electric Power Research Institute,Department of Information and Communication,Beijing,China"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"China Electric Power Research Institute,Department of Information and Communication,Beijing,China","institution_ids":["https://openalex.org/I153473198","https://openalex.org/I4392738113"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5100430806","display_name":"Xi Zhang","orcid":"https://orcid.org/0000-0001-7331-3292"},"institutions":[{"id":"https://openalex.org/I153473198","display_name":"North China Electric Power University","ror":"https://ror.org/04qr5t414","country_code":"CN","type":"education","lineage":["https://openalex.org/I153473198"]},{"id":"https://openalex.org/I4392738113","display_name":"China Electric Power Research Institute","ror":"https://ror.org/05ehpzy81","country_code":null,"type":"facility","lineage":["https://openalex.org/I17442442","https://openalex.org/I4392738113"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Xi Zhang","raw_affiliation_strings":["China Electric Power Research Institute,Department of Information and Communication,Beijing,China"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"China Electric Power Research Institute,Department of Information and Communication,Beijing,China","institution_ids":["https://openalex.org/I153473198","https://openalex.org/I4392738113"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5049706398","display_name":"Hao Deng","orcid":"https://orcid.org/0000-0002-4627-9110"},"institutions":[{"id":"https://openalex.org/I116953780","display_name":"Tongji University","ror":"https://ror.org/03rc6as71","country_code":"CN","type":"education","lineage":["https://openalex.org/I116953780"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Hao Deng","raw_affiliation_strings":["Tongji University,School of Computer Science and Technology,Shanghai,China"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Tongji University,School of Computer Science and Technology,Shanghai,China","institution_ids":["https://openalex.org/I116953780"]}]},{"author_position":"last","author":{"id":"https://openalex.org/A5101612776","display_name":"Shengjie Zhao","orcid":"https://orcid.org/0000-0002-1908-1133"},"institutions":[{"id":"https://openalex.org/I116953780","display_name":"Tongji University","ror":"https://ror.org/03rc6as71","country_code":"CN","type":"education","lineage":["https://openalex.org/I116953780"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Shengjie Zhao","raw_affiliation_strings":["Tongji University,School of Computer Science and Technology,Shanghai,China"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Tongji University,School of Computer Science and Technology,Shanghai,China","institution_ids":["https://openalex.org/I116953780"]}]}],"institutions":[],"countries_distinct_count":1,"institutions_distinct_count":3,"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":"6"},"is_retracted":false,"is_paratext":false,"is_xpac":false,"primary_topic":{"id":"https://openalex.org/T10036","display_name":"Advanced Neural Network Applications","score":0.8863999843597412,"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":0.8863999843597412,"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":0.01899999938905239,"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/T11689","display_name":"Adversarial Robustness in Machine Learning","score":0.014100000262260437,"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/security-token","display_name":"Security token","score":0.6937000155448914},{"id":"https://openalex.org/keywords/segmentation","display_name":"Segmentation","score":0.6534000039100647},{"id":"https://openalex.org/keywords/inference","display_name":"Inference","score":0.6021999716758728},{"id":"https://openalex.org/keywords/leverage","display_name":"Leverage (statistics)","score":0.47119998931884766},{"id":"https://openalex.org/keywords/pooling","display_name":"Pooling","score":0.43140000104904175},{"id":"https://openalex.org/keywords/feature-extraction","display_name":"Feature extraction","score":0.40860000252723694},{"id":"https://openalex.org/keywords/pattern-recognition","display_name":"Pattern recognition (psychology)","score":0.4018000066280365},{"id":"https://openalex.org/keywords/bipartite-graph","display_name":"Bipartite graph","score":0.40119999647140503}],"concepts":[{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.7735999822616577},{"id":"https://openalex.org/C48145219","wikidata":"https://www.wikidata.org/wiki/Q1335365","display_name":"Security token","level":2,"score":0.6937000155448914},{"id":"https://openalex.org/C89600930","wikidata":"https://www.wikidata.org/wiki/Q1423946","display_name":"Segmentation","level":2,"score":0.6534000039100647},{"id":"https://openalex.org/C2776214188","wikidata":"https://www.wikidata.org/wiki/Q408386","display_name":"Inference","level":2,"score":0.6021999716758728},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.5493000149726868},{"id":"https://openalex.org/C153083717","wikidata":"https://www.wikidata.org/wiki/Q6535263","display_name":"Leverage (statistics)","level":2,"score":0.47119998931884766},{"id":"https://openalex.org/C70437156","wikidata":"https://www.wikidata.org/wiki/Q7228652","display_name":"Pooling","level":2,"score":0.43140000104904175},{"id":"https://openalex.org/C52622490","wikidata":"https://www.wikidata.org/wiki/Q1026626","display_name":"Feature extraction","level":2,"score":0.40860000252723694},{"id":"https://openalex.org/C153180895","wikidata":"https://www.wikidata.org/wiki/Q7148389","display_name":"Pattern recognition (psychology)","level":2,"score":0.4018000066280365},{"id":"https://openalex.org/C197657726","wikidata":"https://www.wikidata.org/wiki/Q174733","display_name":"Bipartite graph","level":3,"score":0.40119999647140503},{"id":"https://openalex.org/C2776401178","wikidata":"https://www.wikidata.org/wiki/Q12050496","display_name":"Feature (linguistics)","level":2,"score":0.3547999858856201},{"id":"https://openalex.org/C124101348","wikidata":"https://www.wikidata.org/wiki/Q172491","display_name":"Data mining","level":1,"score":0.3391999900341034},{"id":"https://openalex.org/C165064840","wikidata":"https://www.wikidata.org/wiki/Q1321061","display_name":"Matching (statistics)","level":2,"score":0.33469998836517334},{"id":"https://openalex.org/C124504099","wikidata":"https://www.wikidata.org/wiki/Q56933","display_name":"Image segmentation","level":3,"score":0.32899999618530273},{"id":"https://openalex.org/C179799912","wikidata":"https://www.wikidata.org/wiki/Q205084","display_name":"Computational complexity theory","level":2,"score":0.3237999975681305},{"id":"https://openalex.org/C63479239","wikidata":"https://www.wikidata.org/wiki/Q7353546","display_name":"Robustness (evolution)","level":3,"score":0.2971999943256378},{"id":"https://openalex.org/C119857082","wikidata":"https://www.wikidata.org/wiki/Q2539","display_name":"Machine learning","level":1,"score":0.2955000102519989},{"id":"https://openalex.org/C22367795","wikidata":"https://www.wikidata.org/wiki/Q7625208","display_name":"Structured prediction","level":2,"score":0.2939999997615814},{"id":"https://openalex.org/C2779304628","wikidata":"https://www.wikidata.org/wiki/Q3503480","display_name":"Face (sociological concept)","level":2,"score":0.2930000126361847},{"id":"https://openalex.org/C111335779","wikidata":"https://www.wikidata.org/wiki/Q3454686","display_name":"Reduction (mathematics)","level":2,"score":0.2915000021457672},{"id":"https://openalex.org/C103278499","wikidata":"https://www.wikidata.org/wiki/Q254465","display_name":"Similarity (geometry)","level":3,"score":0.2694999873638153},{"id":"https://openalex.org/C143271835","wikidata":"https://www.wikidata.org/wiki/Q254515","display_name":"Similitude","level":2,"score":0.2612999975681305},{"id":"https://openalex.org/C177774035","wikidata":"https://www.wikidata.org/wiki/Q1246948","display_name":"Granularity","level":2,"score":0.2605000138282776},{"id":"https://openalex.org/C18555067","wikidata":"https://www.wikidata.org/wiki/Q8375051","display_name":"Joint (building)","level":2,"score":0.2547999918460846}],"mesh":[],"locations_count":1,"locations":[{"id":"doi:10.1109/icme59968.2025.11209327","is_oa":false,"landing_page_url":"https://doi.org/10.1109/icme59968.2025.11209327","pdf_url":null,"source":null,"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"2025 IEEE International Conference on Multimedia and Expo (ICME)","raw_type":"proceedings-article"}],"best_oa_location":null,"sustainable_development_goals":[],"awards":[],"funders":[{"id":"https://openalex.org/F4320309612","display_name":"Natural Science Foundation of Shanghai","ror":null},{"id":"https://openalex.org/F4320321001","display_name":"National Natural Science Foundation of China","ror":"https://ror.org/01h0zpd94"},{"id":"https://openalex.org/F4320337504","display_name":"Research and Development","ror":"https://ror.org/027s68j25"}],"has_content":{"pdf":false,"grobid_xml":false},"content_urls":null,"referenced_works_count":13,"referenced_works":["https://openalex.org/W2125215748","https://openalex.org/W2194775991","https://openalex.org/W2340897893","https://openalex.org/W2737258237","https://openalex.org/W3170841864","https://openalex.org/W4214893857","https://openalex.org/W4312849330","https://openalex.org/W4386075588","https://openalex.org/W4390873017","https://openalex.org/W4390873032","https://openalex.org/W4394593136","https://openalex.org/W4394625744","https://openalex.org/W4402753804"],"related_works":[],"abstract_inverted_index":{"Vision":[0],"Transformers":[1],"(ViTs)":[2],"leverage":[3],"the":[4,144],"attention":[5],"mechanism":[6],"for":[7,74],"feature":[8],"extraction":[9],"but":[10],"often":[11],"suffer":[12],"from":[13],"high":[14],"computational":[15,168],"costs.":[16],"To":[17,60],"address":[18],"this":[19],"issue,":[20],"prior":[21],"works":[22],"have":[23],"introduced":[24],"token":[25,115],"reduction":[26],"methods":[27,38],"involving":[28],"fixed-window":[29],"local":[30,49,93],"merging":[31,45,53,145],"and":[32,51,100,116,133,170],"global":[33,58],"Bipartite":[34,102],"Matching.":[35],"However,":[36],"these":[37,62],"face":[39],"significant":[40],"challenges,":[41],"such":[42],"as":[43],"insufficient":[44],"due":[46],"to":[47,95,106,138,149,151,159],"fixed-size":[48],"windows":[50,94],"incorrect":[52],"of":[54],"informative":[55,108],"tokens":[56,82,109],"in":[57],"merging.":[59],"overcome":[61],"limitations,":[63],"we":[64],"propose":[65],"Efficient":[66],"Semantic":[67],"Segmentation":[68],"via":[69],"Token":[70],"Joint":[71],"Merging":[72],"(ESTJ)":[73],"ViT-based":[75],"semantic":[76],"segmentation":[77,171],"networks.":[78],"Specifically,":[79],"ESTJ":[80,124,141,162],"merges":[81],"using":[83,110],"two":[84],"strategies:":[85],"Hierarchical":[86],"Condition":[87],"Pooling":[88],"(HCP),":[89],"which":[90],"employs":[91],"hierarchical":[92],"effectively":[96],"select":[97],"sufficient":[98],"tokens,":[99],"Protected":[101],"Matching":[103],"(PBM),":[104],"designed":[105],"preserve":[107],"average":[111],"similarity":[112],"between":[113,167],"a":[114,164],"all":[117],"other":[118],"tokens.":[119],"Experimental":[120],"results":[121],"demonstrate":[122],"that":[123,153],"improves":[125],"throughput":[126],"by":[127,131,136],"75%,":[128],"reduces":[129],"GFLOPs":[130],"40%,":[132],"enhances":[134],"mIoU":[135],"up":[137],"1.1%.":[139],"Moreover,":[140],"can":[142],"adjust":[143],"threshold":[146],"during":[147],"inference":[148],"adapt":[150],"scenarios":[152],"prioritize":[154],"efficiency":[155,169],"or":[156],"accuracy.":[157,172],"Compared":[158],"existing":[160],"methods,":[161],"achieves":[163],"better":[165],"balance":[166]},"counts_by_year":[],"updated_date":"2026-07-14T23:27:15.235271","created_date":"2025-10-30T00:00:00"}