{"id":"https://openalex.org/W4402727649","doi":"https://doi.org/10.1109/cvpr52733.2024.00393","title":"Learn to Rectify the Bias of CLIP for Unsupervised Semantic Segmentation","display_name":"Learn to Rectify the Bias of CLIP for Unsupervised Semantic Segmentation","publication_year":2024,"publication_date":"2024-06-16","ids":{"openalex":"https://openalex.org/W4402727649","doi":"https://doi.org/10.1109/cvpr52733.2024.00393"},"language":"en","primary_location":{"id":"doi:10.1109/cvpr52733.2024.00393","is_oa":false,"landing_page_url":"https://doi.org/10.1109/cvpr52733.2024.00393","pdf_url":null,"source":null,"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"2024 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR)","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/A5100779751","display_name":"Jingyun Wang","orcid":"https://orcid.org/0000-0002-9073-2349"},"institutions":[{"id":"https://openalex.org/I82880672","display_name":"Beihang University","ror":"https://ror.org/00wk2mp56","country_code":"CN","type":"education","lineage":["https://openalex.org/I82880672"]}],"countries":["CN"],"is_corresponding":true,"raw_author_name":"Jingyun Wang","raw_affiliation_strings":["Beihang University"],"affiliations":[{"raw_affiliation_string":"Beihang University","institution_ids":["https://openalex.org/I82880672"]}]},{"author_position":"last","author":{"id":"https://openalex.org/A5011488839","display_name":"Guoliang Kang","orcid":"https://orcid.org/0000-0003-1978-2025"},"institutions":[{"id":"https://openalex.org/I82880672","display_name":"Beihang University","ror":"https://ror.org/00wk2mp56","country_code":"CN","type":"education","lineage":["https://openalex.org/I82880672"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Guoliang Kang","raw_affiliation_strings":["Beihang University,Zhongguancun Laboratory"],"affiliations":[{"raw_affiliation_string":"Beihang University,Zhongguancun Laboratory","institution_ids":["https://openalex.org/I82880672"]}]}],"institutions":[],"countries_distinct_count":1,"institutions_distinct_count":2,"corresponding_author_ids":["https://openalex.org/A5100779751"],"corresponding_institution_ids":["https://openalex.org/I82880672"],"apc_list":null,"apc_paid":null,"fwci":1.6685,"has_fulltext":false,"cited_by_count":7,"citation_normalized_percentile":{"value":0.85464776,"is_in_top_1_percent":false,"is_in_top_10_percent":false},"cited_by_percentile_year":{"min":98,"max":99},"biblio":{"volume":null,"issue":null,"first_page":"4102","last_page":"4112"},"is_retracted":false,"is_paratext":false,"is_xpac":false,"primary_topic":{"id":"https://openalex.org/T10036","display_name":"Advanced Neural Network Applications","score":0.9995999932289124,"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":0.9995999932289124,"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":0.9994999766349792,"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/T11714","display_name":"Multimodal Machine Learning Applications","score":0.9990000128746033,"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":0.7347587943077087},{"id":"https://openalex.org/keywords/segmentation","display_name":"Segmentation","score":0.5784662961959839},{"id":"https://openalex.org/keywords/artificial-intelligence","display_name":"Artificial intelligence","score":0.5460072755813599},{"id":"https://openalex.org/keywords/natural-language-processing","display_name":"Natural language processing","score":0.5380610227584839},{"id":"https://openalex.org/keywords/unsupervised-learning","display_name":"Unsupervised learning","score":0.4136473834514618}],"concepts":[{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.7347587943077087},{"id":"https://openalex.org/C89600930","wikidata":"https://www.wikidata.org/wiki/Q1423946","display_name":"Segmentation","level":2,"score":0.5784662961959839},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.5460072755813599},{"id":"https://openalex.org/C204321447","wikidata":"https://www.wikidata.org/wiki/Q30642","display_name":"Natural language processing","level":1,"score":0.5380610227584839},{"id":"https://openalex.org/C8038995","wikidata":"https://www.wikidata.org/wiki/Q1152135","display_name":"Unsupervised learning","level":2,"score":0.4136473834514618}],"mesh":[],"locations_count":1,"locations":[{"id":"doi:10.1109/cvpr52733.2024.00393","is_oa":false,"landing_page_url":"https://doi.org/10.1109/cvpr52733.2024.00393","pdf_url":null,"source":null,"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"2024 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR)","raw_type":"proceedings-article"}],"best_oa_location":null,"sustainable_development_goals":[{"id":"https://metadata.un.org/sdg/10","display_name":"Reduced inequalities","score":0.5799999833106995}],"awards":[],"funders":[],"has_content":{"grobid_xml":false,"pdf":false},"content_urls":null,"referenced_works_count":67,"referenced_works":["https://openalex.org/W1903029394","https://openalex.org/W1923697677","https://openalex.org/W1945608308","https://openalex.org/W1994616650","https://openalex.org/W2031489346","https://openalex.org/W2125215748","https://openalex.org/W2222512263","https://openalex.org/W2412782625","https://openalex.org/W2560023338","https://openalex.org/W2737258237","https://openalex.org/W2765811365","https://openalex.org/W2924485953","https://openalex.org/W2947897713","https://openalex.org/W2962758679","https://openalex.org/W2971113548","https://openalex.org/W2982106100","https://openalex.org/W2998356391","https://openalex.org/W3034552520","https://openalex.org/W3090449556","https://openalex.org/W3092500685","https://openalex.org/W3105232955","https://openalex.org/W3107725261","https://openalex.org/W3110446398","https://openalex.org/W3129110783","https://openalex.org/W3159481202","https://openalex.org/W3172288085","https://openalex.org/W3176777339","https://openalex.org/W3182683290","https://openalex.org/W3184735396","https://openalex.org/W3198377975","https://openalex.org/W3213165621","https://openalex.org/W4282969702","https://openalex.org/W4286905173","https://openalex.org/W4288346431","https://openalex.org/W4312360228","https://openalex.org/W4312629998","https://openalex.org/W4312747482","https://openalex.org/W4312815172","https://openalex.org/W4312912313","https://openalex.org/W4312960937","https://openalex.org/W4312980231","https://openalex.org/W4313150877","https://openalex.org/W4321524523","https://openalex.org/W4385245566","https://openalex.org/W4386065432","https://openalex.org/W4386065556","https://openalex.org/W4386066015","https://openalex.org/W4386075561","https://openalex.org/W4386075766","https://openalex.org/W4386076397","https://openalex.org/W4390874346","https://openalex.org/W6729448088","https://openalex.org/W6739901393","https://openalex.org/W6763331980","https://openalex.org/W6763578275","https://openalex.org/W6763612067","https://openalex.org/W6770717842","https://openalex.org/W6784333009","https://openalex.org/W6790019176","https://openalex.org/W6791353385","https://openalex.org/W6798805250","https://openalex.org/W6798837711","https://openalex.org/W6811433417","https://openalex.org/W6839009181","https://openalex.org/W6846511317","https://openalex.org/W6850241730","https://openalex.org/W6850886806"],"related_works":["https://openalex.org/W4391375266","https://openalex.org/W2748952813","https://openalex.org/W2390279801","https://openalex.org/W2358668433","https://openalex.org/W4396701345","https://openalex.org/W2376932109","https://openalex.org/W2001405890","https://openalex.org/W4396696052","https://openalex.org/W2382290278","https://openalex.org/W3204019825"],"abstract_inverted_index":{"Recent":[0],"works":[1,36],"utilize":[2],"CLIP":[3,25,62,102,153],"to":[4,26,53,63,76,88,135,154],"perform":[5],"the":[6,45,58,65,82,99,106,127,137,144,148,151,155],"challenging":[7],"unsupervised":[8,66],"semantic":[9,67],"segmentation":[10,46,112,157],"task":[11],"where":[12],"only":[13],"images":[14],"without":[15],"annotations":[16],"are":[17],"available.":[18],"However,":[19],"we":[20,51,70,97,109,146],"observe":[21],"that":[22,175],"when":[23],"adopting":[24],"such":[27,40],"a":[28,72,93,111,115,119],"pixel-level":[29],"understanding":[30],"task,":[31],"unexpected":[32],"bias":[33,59,79,139],"occurs.":[34],"Previous":[35],"don't":[37],"explicitly":[38,54],"model":[39,55],"bias,":[41],"which":[42],"largely":[43],"constrains":[44],"performance.":[47],"In":[48],"this":[49],"paper,":[50],"propose":[52],"and":[56,80,126,165],"rectify":[57,98],"existing":[60],"in":[61],"facilitate":[64,136],"segmentation.":[68],"Specifically,":[69],"design":[71],"learnable":[73],"\u201cReference\u201d":[74],"prompt":[75],"encode":[77],"class-preference":[78],"project":[81],"positional":[83],"embedding":[84],"of":[85,101,130],"vision":[86],"transformer":[87],"represent":[89],"space-preference":[90],"bias.":[91],"Via":[92],"simple":[94],"element-wise":[95],"subtraction,":[96],"logits":[100],"classifier.":[103],"Based":[104],"on":[105,171],"rectified":[107,152],"logits,":[108],"generate":[110],"mask":[113],"via":[114,159],"Gumbel-Softmax":[116],"operation.":[117],"Then":[118],"contrastive":[120],"loss":[121,167],"between":[122],"masked":[123],"visual":[124],"feature":[125],"text":[128],"features":[129],"different":[131],"classes":[132],"is":[133,185],"imposed":[134],"effective":[138],"modeling.":[140],"To":[141],"further":[142],"improve":[143],"segmentation,":[145],"distill":[147],"knowledge":[149],"from":[150],"advanced":[156],"architecture":[158],"minimizing":[160],"our":[161,176],"designed":[162],"mask-guided,":[163],"feature-guided":[164],"text-guided":[166],"terms.":[168],"Extensive":[169],"experiments":[170],"standard":[172],"benchmarks":[173],"demonstrate":[174],"method":[177],"performs":[178],"favorably":[179],"against":[180],"previous":[181],"state-of-the-arts.":[182],"The":[183],"implementation":[184],"available":[186],"at":[187],"https://github.com/dogehhh/ReCLIP.":[188]},"counts_by_year":[{"year":2026,"cited_by_count":2},{"year":2025,"cited_by_count":5}],"updated_date":"2026-04-23T09:07:50.710637","created_date":"2025-10-10T00:00:00"}