{"id":"https://openalex.org/W4408345821","doi":"https://doi.org/10.1109/icassp49660.2025.10890804","title":"Multi-Layer Knowledge Distillation for Continual Semantic Segmentation","display_name":"Multi-Layer Knowledge Distillation for Continual Semantic Segmentation","publication_year":2025,"publication_date":"2025-03-12","ids":{"openalex":"https://openalex.org/W4408345821","doi":"https://doi.org/10.1109/icassp49660.2025.10890804"},"language":"en","primary_location":{"id":"doi:10.1109/icassp49660.2025.10890804","is_oa":false,"landing_page_url":"https://doi.org/10.1109/icassp49660.2025.10890804","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":"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/A5100322854","display_name":"Yan Wang","orcid":"https://orcid.org/0000-0002-9876-5823"},"institutions":[{"id":"https://openalex.org/I139759216","display_name":"Beijing University of Posts and Telecommunications","ror":"https://ror.org/04w9fbh59","country_code":"CN","type":"education","lineage":["https://openalex.org/I139759216"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Yan Wang","raw_affiliation_strings":["Beijing University of Posts and Telecommunications,School of Artificial Intelligence,Beijing,China"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Beijing University of Posts and Telecommunications,School of Artificial Intelligence,Beijing,China","institution_ids":["https://openalex.org/I139759216"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5100554568","display_name":"Pengju Xu","orcid":"https://orcid.org/0009-0007-0465-3622"},"institutions":[{"id":"https://openalex.org/I139759216","display_name":"Beijing University of Posts and Telecommunications","ror":"https://ror.org/04w9fbh59","country_code":"CN","type":"education","lineage":["https://openalex.org/I139759216"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Pengju Xu","raw_affiliation_strings":["Beijing University of Posts and Telecommunications,School of Artificial Intelligence,Beijing,China"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Beijing University of Posts and Telecommunications,School of Artificial Intelligence,Beijing,China","institution_ids":["https://openalex.org/I139759216"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5100408153","display_name":"Bo Wang","orcid":"https://orcid.org/0000-0002-9041-878X"},"institutions":[{"id":"https://openalex.org/I139759216","display_name":"Beijing University of Posts and Telecommunications","ror":"https://ror.org/04w9fbh59","country_code":"CN","type":"education","lineage":["https://openalex.org/I139759216"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Bingye Wang","raw_affiliation_strings":["Beijing University of Posts and Telecommunications,School of Artificial Intelligence,Beijing,China"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Beijing University of Posts and Telecommunications,School of Artificial Intelligence,Beijing,China","institution_ids":["https://openalex.org/I139759216"]}]},{"author_position":"last","author":{"id":"https://openalex.org/A5111190383","display_name":"Haiying Zhao","orcid":null},"institutions":[{"id":"https://openalex.org/I139759216","display_name":"Beijing University of Posts and Telecommunications","ror":"https://ror.org/04w9fbh59","country_code":"CN","type":"education","lineage":["https://openalex.org/I139759216"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Haiying Zhao","raw_affiliation_strings":["Beijing University of Posts and Telecommunications,School of Artificial Intelligence,Beijing,China"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Beijing University of Posts and Telecommunications,School of Artificial Intelligence,Beijing,China","institution_ids":["https://openalex.org/I139759216"]}]}],"institutions":[],"countries_distinct_count":1,"institutions_distinct_count":1,"corresponding_author_ids":[],"corresponding_institution_ids":["https://openalex.org/I139759216"],"apc_list":null,"apc_paid":null,"fwci":0.0,"has_fulltext":false,"cited_by_count":0,"citation_normalized_percentile":{"value":0.02962843,"is_in_top_1_percent":false,"is_in_top_10_percent":false},"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/T14339","display_name":"Image Processing and 3D Reconstruction","score":0.4846999943256378,"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/T14339","display_name":"Image Processing and 3D Reconstruction","score":0.4846999943256378,"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/T10320","display_name":"Neural Networks and Applications","score":0.47940000891685486,"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/T12016","display_name":"Web Data Mining and Analysis","score":0.42739999294281006,"subfield":{"id":"https://openalex.org/subfields/1710","display_name":"Information Systems"},"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":0.7684305906295776},{"id":"https://openalex.org/keywords/segmentation","display_name":"Segmentation","score":0.6161627173423767},{"id":"https://openalex.org/keywords/layer","display_name":"Layer (electronics)","score":0.5875613689422607},{"id":"https://openalex.org/keywords/distillation","display_name":"Distillation","score":0.5440579652786255},{"id":"https://openalex.org/keywords/artificial-intelligence","display_name":"Artificial intelligence","score":0.5285277366638184},{"id":"https://openalex.org/keywords/natural-language-processing","display_name":"Natural language processing","score":0.4658874571323395},{"id":"https://openalex.org/keywords/materials-science","display_name":"Materials science","score":0.09233224391937256},{"id":"https://openalex.org/keywords/chemistry","display_name":"Chemistry","score":0.08604210615158081}],"concepts":[{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.7684305906295776},{"id":"https://openalex.org/C89600930","wikidata":"https://www.wikidata.org/wiki/Q1423946","display_name":"Segmentation","level":2,"score":0.6161627173423767},{"id":"https://openalex.org/C2779227376","wikidata":"https://www.wikidata.org/wiki/Q6505497","display_name":"Layer (electronics)","level":2,"score":0.5875613689422607},{"id":"https://openalex.org/C204030448","wikidata":"https://www.wikidata.org/wiki/Q101017","display_name":"Distillation","level":2,"score":0.5440579652786255},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.5285277366638184},{"id":"https://openalex.org/C204321447","wikidata":"https://www.wikidata.org/wiki/Q30642","display_name":"Natural language processing","level":1,"score":0.4658874571323395},{"id":"https://openalex.org/C192562407","wikidata":"https://www.wikidata.org/wiki/Q228736","display_name":"Materials science","level":0,"score":0.09233224391937256},{"id":"https://openalex.org/C185592680","wikidata":"https://www.wikidata.org/wiki/Q2329","display_name":"Chemistry","level":0,"score":0.08604210615158081},{"id":"https://openalex.org/C178790620","wikidata":"https://www.wikidata.org/wiki/Q11351","display_name":"Organic chemistry","level":1,"score":0.0},{"id":"https://openalex.org/C159985019","wikidata":"https://www.wikidata.org/wiki/Q181790","display_name":"Composite material","level":1,"score":0.0}],"mesh":[],"locations_count":1,"locations":[{"id":"doi:10.1109/icassp49660.2025.10890804","is_oa":false,"landing_page_url":"https://doi.org/10.1109/icassp49660.2025.10890804","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":[],"awards":[],"funders":[{"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":18,"referenced_works":["https://openalex.org/W2035751167","https://openalex.org/W2037227137","https://openalex.org/W2194775991","https://openalex.org/W2412782625","https://openalex.org/W2473930607","https://openalex.org/W2560647685","https://openalex.org/W2737258237","https://openalex.org/W2786446225","https://openalex.org/W2964189064","https://openalex.org/W3011986058","https://openalex.org/W3034435444","https://openalex.org/W3166525903","https://openalex.org/W3171888599","https://openalex.org/W4225922183","https://openalex.org/W4312447529","https://openalex.org/W4385430461","https://openalex.org/W4402915797","https://openalex.org/W6741217325"],"related_works":["https://openalex.org/W3026162553","https://openalex.org/W2344382886","https://openalex.org/W19111321","https://openalex.org/W4379231730","https://openalex.org/W2412887479","https://openalex.org/W32245304","https://openalex.org/W2953684491","https://openalex.org/W4285338581","https://openalex.org/W2768175398","https://openalex.org/W3204019825"],"abstract_inverted_index":{"Recently,":[0],"knowledge":[1,118],"distillation":[2,119],"has":[3],"shown":[4],"promising":[5],"results":[6],"for":[7,65,69],"continual":[8],"semantic":[9,124],"segmentation":[10],"(CSS).":[11],"Despite":[12],"their":[13],"success,":[14],"several":[15],"key":[16],"issues":[17],"have":[18],"not":[19,44],"been":[20],"well":[21],"addressed":[22],"in":[23,41,49],"existing":[24],"studies:":[25],"1)":[26],"Difficulty":[27],"to":[28,81,108,121,154],"balance":[29],"new":[30,98],"and":[31,47,95,132],"old":[32,92],"classes,":[33],"challenged":[34],"by":[35,85],"catastrophic":[36,83],"forgetting.":[37],"2)":[38],"Complex":[39],"information":[40],"images":[42],"is":[43],"fully":[45],"learned":[46],"accuracy":[48],"CSS":[50],"needs":[51],"improvement.":[52],"To":[53],"tackle":[54],"these":[55],"challenges,":[56],"we":[57,74,103,114],"propose":[58],"a":[59,116],"novel":[60],"Multi-Layer":[61],"Knowledge":[62],"Distillation":[63],"(MLKD)":[64],"Continual":[66,70],"Semantic":[67,71],"Segmentation":[68],"Segmentation.":[72],"Specifically,":[73],"utilize":[75],"the":[76,91,97,105,110,129,133,144,149],"branch":[77],"incremental":[78],"learning":[79],"module":[80,120],"mitigate":[82],"forgetting":[84],"having":[86],"two":[87,140],"branches":[88],"that":[89],"freeze":[90],"class":[93],"parameters":[94],"train":[96],"classes.":[99],"For":[100],"each":[101],"branch,":[102],"execute":[104],"contextual-awareness":[106],"component":[107],"learn":[109],"contextual":[111],"information.":[112],"Then,":[113],"employ":[115],"multi-layer":[117],"comprehensively":[122],"capture":[123],"correlations":[125],"between":[126],"classes":[127],"from":[128],"intermediate":[130],"layer":[131],"final":[134],"output":[135],"layer.":[136],"Extensive":[137],"experiments":[138],"on":[139],"real-world":[141],"datasets":[142],"demonstrate":[143],"effectiveness":[145],"of":[146],"MLKD,":[147],"where":[148],"maximum":[150],"improvement":[151],"can":[152],"reach":[153],"2.7%.":[155]},"counts_by_year":[],"updated_date":"2025-12-28T23:10:05.387466","created_date":"2025-10-10T00:00:00"}