{"id":"https://openalex.org/W4386597555","doi":"https://doi.org/10.1109/icip49359.2023.10222502","title":"Mitigating Dataset Bias in Image Captioning Through Clip Confounder-Free Captioning Network","display_name":"Mitigating Dataset Bias in Image Captioning Through Clip Confounder-Free Captioning Network","publication_year":2023,"publication_date":"2023-09-11","ids":{"openalex":"https://openalex.org/W4386597555","doi":"https://doi.org/10.1109/icip49359.2023.10222502"},"language":"en","primary_location":{"id":"doi:10.1109/icip49359.2023.10222502","is_oa":false,"landing_page_url":"https://doi.org/10.1109/icip49359.2023.10222502","pdf_url":null,"source":null,"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"2023 IEEE International Conference on Image Processing (ICIP)","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/A5043898799","display_name":"Yeonju Kim","orcid":null},"institutions":[{"id":"https://openalex.org/I157485424","display_name":"Korea Advanced Institute of Science and Technology","ror":"https://ror.org/05apxxy63","country_code":"KR","type":"education","lineage":["https://openalex.org/I157485424"]}],"countries":["KR"],"is_corresponding":false,"raw_author_name":"Yeonju Kim","raw_affiliation_strings":["KAIST,Image and Video Systems Lab, School of Electrical Engineering,South Korea","Image and Video Systems Lab, School of Electrical Engineering, KAIST, South Korea"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"KAIST,Image and Video Systems Lab, School of Electrical Engineering,South Korea","institution_ids":["https://openalex.org/I157485424"]},{"raw_affiliation_string":"Image and Video Systems Lab, School of Electrical Engineering, KAIST, South Korea","institution_ids":["https://openalex.org/I157485424"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5100659571","display_name":"Junho Kim","orcid":"https://orcid.org/0000-0002-7769-910X"},"institutions":[{"id":"https://openalex.org/I157485424","display_name":"Korea Advanced Institute of Science and Technology","ror":"https://ror.org/05apxxy63","country_code":"KR","type":"education","lineage":["https://openalex.org/I157485424"]}],"countries":["KR"],"is_corresponding":false,"raw_author_name":"Junho Kim","raw_affiliation_strings":["KAIST,Image and Video Systems Lab, School of Electrical Engineering,South Korea","Image and Video Systems Lab, School of Electrical Engineering, KAIST, South Korea"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"KAIST,Image and Video Systems Lab, School of Electrical Engineering,South Korea","institution_ids":["https://openalex.org/I157485424"]},{"raw_affiliation_string":"Image and Video Systems Lab, School of Electrical Engineering, KAIST, South Korea","institution_ids":["https://openalex.org/I157485424"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5063249657","display_name":"Byung\u2010Kwan Lee","orcid":"https://orcid.org/0000-0002-5624-7518"},"institutions":[{"id":"https://openalex.org/I157485424","display_name":"Korea Advanced Institute of Science and Technology","ror":"https://ror.org/05apxxy63","country_code":"KR","type":"education","lineage":["https://openalex.org/I157485424"]}],"countries":["KR"],"is_corresponding":false,"raw_author_name":"Byung-Kwan Lee","raw_affiliation_strings":["KAIST,Image and Video Systems Lab, School of Electrical Engineering,South Korea","Image and Video Systems Lab, School of Electrical Engineering, KAIST, South Korea"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"KAIST,Image and Video Systems Lab, School of Electrical Engineering,South Korea","institution_ids":["https://openalex.org/I157485424"]},{"raw_affiliation_string":"Image and Video Systems Lab, School of Electrical Engineering, KAIST, South Korea","institution_ids":["https://openalex.org/I157485424"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5112998100","display_name":"Sebin Shin","orcid":null},"institutions":[{"id":"https://openalex.org/I157485424","display_name":"Korea Advanced Institute of Science and Technology","ror":"https://ror.org/05apxxy63","country_code":"KR","type":"education","lineage":["https://openalex.org/I157485424"]}],"countries":["KR"],"is_corresponding":false,"raw_author_name":"Sebin Shin","raw_affiliation_strings":["KAIST,Image and Video Systems Lab, School of Electrical Engineering,South Korea","Image and Video Systems Lab, School of Electrical Engineering, KAIST, South Korea"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"KAIST,Image and Video Systems Lab, School of Electrical Engineering,South Korea","institution_ids":["https://openalex.org/I157485424"]},{"raw_affiliation_string":"Image and Video Systems Lab, School of Electrical Engineering, KAIST, South Korea","institution_ids":["https://openalex.org/I157485424"]}]},{"author_position":"last","author":{"id":"https://openalex.org/A5038798134","display_name":"Yong Man Ro","orcid":"https://orcid.org/0000-0001-5306-6853"},"institutions":[{"id":"https://openalex.org/I157485424","display_name":"Korea Advanced Institute of Science and Technology","ror":"https://ror.org/05apxxy63","country_code":"KR","type":"education","lineage":["https://openalex.org/I157485424"]}],"countries":["KR"],"is_corresponding":false,"raw_author_name":"Yong Man Ro","raw_affiliation_strings":["KAIST,Image and Video Systems Lab, School of Electrical Engineering,South Korea","Image and Video Systems Lab, School of Electrical Engineering, KAIST, South Korea"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"KAIST,Image and Video Systems Lab, School of Electrical Engineering,South Korea","institution_ids":["https://openalex.org/I157485424"]},{"raw_affiliation_string":"Image and Video Systems Lab, School of Electrical Engineering, KAIST, South Korea","institution_ids":["https://openalex.org/I157485424"]}]}],"institutions":[],"countries_distinct_count":1,"institutions_distinct_count":5,"corresponding_author_ids":[],"corresponding_institution_ids":[],"apc_list":null,"apc_paid":null,"fwci":0.6737,"has_fulltext":false,"cited_by_count":6,"citation_normalized_percentile":{"value":0.71527968,"is_in_top_1_percent":false,"is_in_top_10_percent":false},"cited_by_percentile_year":{"min":94,"max":98},"biblio":{"volume":null,"issue":null,"first_page":"1720","last_page":"1724"},"is_retracted":false,"is_paratext":false,"is_xpac":false,"primary_topic":{"id":"https://openalex.org/T11714","display_name":"Multimodal Machine Learning Applications","score":1.0,"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/T11714","display_name":"Multimodal Machine Learning Applications","score":1.0,"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":0.9904000163078308,"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/T10627","display_name":"Advanced Image and Video Retrieval Techniques","score":0.9825999736785889,"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/closed-captioning","display_name":"Closed captioning","score":0.940305233001709},{"id":"https://openalex.org/keywords/computer-science","display_name":"Computer science","score":0.7206059694290161},{"id":"https://openalex.org/keywords/inference","display_name":"Inference","score":0.7141047716140747},{"id":"https://openalex.org/keywords/causal-inference","display_name":"Causal inference","score":0.6394364237785339},{"id":"https://openalex.org/keywords/word","display_name":"Word (group theory)","score":0.6183569431304932},{"id":"https://openalex.org/keywords/benchmark","display_name":"Benchmark (surveying)","score":0.5738794207572937},{"id":"https://openalex.org/keywords/artificial-intelligence","display_name":"Artificial intelligence","score":0.553303062915802},{"id":"https://openalex.org/keywords/graph","display_name":"Graph","score":0.5352957844734192},{"id":"https://openalex.org/keywords/confounding","display_name":"Confounding","score":0.5262224674224854},{"id":"https://openalex.org/keywords/natural-language-processing","display_name":"Natural language processing","score":0.44666558504104614},{"id":"https://openalex.org/keywords/perspective","display_name":"Perspective (graphical)","score":0.4113262891769409},{"id":"https://openalex.org/keywords/image","display_name":"Image (mathematics)","score":0.3959365487098694},{"id":"https://openalex.org/keywords/machine-learning","display_name":"Machine learning","score":0.38023629784584045},{"id":"https://openalex.org/keywords/theoretical-computer-science","display_name":"Theoretical computer science","score":0.12452059984207153},{"id":"https://openalex.org/keywords/statistics","display_name":"Statistics","score":0.12143656611442566},{"id":"https://openalex.org/keywords/mathematics","display_name":"Mathematics","score":0.10005307197570801}],"concepts":[{"id":"https://openalex.org/C157657479","wikidata":"https://www.wikidata.org/wiki/Q2367247","display_name":"Closed captioning","level":3,"score":0.940305233001709},{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.7206059694290161},{"id":"https://openalex.org/C2776214188","wikidata":"https://www.wikidata.org/wiki/Q408386","display_name":"Inference","level":2,"score":0.7141047716140747},{"id":"https://openalex.org/C158600405","wikidata":"https://www.wikidata.org/wiki/Q5054566","display_name":"Causal inference","level":2,"score":0.6394364237785339},{"id":"https://openalex.org/C90805587","wikidata":"https://www.wikidata.org/wiki/Q10944557","display_name":"Word (group theory)","level":2,"score":0.6183569431304932},{"id":"https://openalex.org/C185798385","wikidata":"https://www.wikidata.org/wiki/Q1161707","display_name":"Benchmark (surveying)","level":2,"score":0.5738794207572937},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.553303062915802},{"id":"https://openalex.org/C132525143","wikidata":"https://www.wikidata.org/wiki/Q141488","display_name":"Graph","level":2,"score":0.5352957844734192},{"id":"https://openalex.org/C77350462","wikidata":"https://www.wikidata.org/wiki/Q1125472","display_name":"Confounding","level":2,"score":0.5262224674224854},{"id":"https://openalex.org/C204321447","wikidata":"https://www.wikidata.org/wiki/Q30642","display_name":"Natural language processing","level":1,"score":0.44666558504104614},{"id":"https://openalex.org/C12713177","wikidata":"https://www.wikidata.org/wiki/Q1900281","display_name":"Perspective (graphical)","level":2,"score":0.4113262891769409},{"id":"https://openalex.org/C115961682","wikidata":"https://www.wikidata.org/wiki/Q860623","display_name":"Image (mathematics)","level":2,"score":0.3959365487098694},{"id":"https://openalex.org/C119857082","wikidata":"https://www.wikidata.org/wiki/Q2539","display_name":"Machine learning","level":1,"score":0.38023629784584045},{"id":"https://openalex.org/C80444323","wikidata":"https://www.wikidata.org/wiki/Q2878974","display_name":"Theoretical computer science","level":1,"score":0.12452059984207153},{"id":"https://openalex.org/C105795698","wikidata":"https://www.wikidata.org/wiki/Q12483","display_name":"Statistics","level":1,"score":0.12143656611442566},{"id":"https://openalex.org/C33923547","wikidata":"https://www.wikidata.org/wiki/Q395","display_name":"Mathematics","level":0,"score":0.10005307197570801},{"id":"https://openalex.org/C205649164","wikidata":"https://www.wikidata.org/wiki/Q1071","display_name":"Geography","level":0,"score":0.0},{"id":"https://openalex.org/C13280743","wikidata":"https://www.wikidata.org/wiki/Q131089","display_name":"Geodesy","level":1,"score":0.0},{"id":"https://openalex.org/C2524010","wikidata":"https://www.wikidata.org/wiki/Q8087","display_name":"Geometry","level":1,"score":0.0}],"mesh":[],"locations_count":1,"locations":[{"id":"doi:10.1109/icip49359.2023.10222502","is_oa":false,"landing_page_url":"https://doi.org/10.1109/icip49359.2023.10222502","pdf_url":null,"source":null,"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"2023 IEEE International Conference on Image Processing (ICIP)","raw_type":"proceedings-article"}],"best_oa_location":null,"sustainable_development_goals":[],"awards":[],"funders":[],"has_content":{"pdf":false,"grobid_xml":false},"content_urls":null,"referenced_works_count":33,"referenced_works":["https://openalex.org/W639708223","https://openalex.org/W1861492603","https://openalex.org/W1956340063","https://openalex.org/W2016089260","https://openalex.org/W2101105183","https://openalex.org/W2112912048","https://openalex.org/W2130942839","https://openalex.org/W2154652894","https://openalex.org/W2250539671","https://openalex.org/W2486285194","https://openalex.org/W2506483933","https://openalex.org/W2600463316","https://openalex.org/W2745461083","https://openalex.org/W2885013662","https://openalex.org/W2890531016","https://openalex.org/W2963084599","https://openalex.org/W2963101956","https://openalex.org/W2963349562","https://openalex.org/W2986670728","https://openalex.org/W3034655362","https://openalex.org/W3035651653","https://openalex.org/W3035682985","https://openalex.org/W3202965415","https://openalex.org/W3206022579","https://openalex.org/W4288329833","https://openalex.org/W4312232840","https://openalex.org/W6620707391","https://openalex.org/W6676647902","https://openalex.org/W6678262379","https://openalex.org/W6679436768","https://openalex.org/W6682631176","https://openalex.org/W6763643401","https://openalex.org/W6791353385"],"related_works":["https://openalex.org/W3217195652","https://openalex.org/W3023719900","https://openalex.org/W4287798354","https://openalex.org/W3035083705","https://openalex.org/W2030287811","https://openalex.org/W3002087755","https://openalex.org/W4386534229","https://openalex.org/W4386150491","https://openalex.org/W2806152055","https://openalex.org/W4301105698"],"abstract_inverted_index":{"The":[0],"dataset":[1,41,51],"bias":[2,42],"has":[3],"been":[4],"identified":[5],"as":[6],"a":[7,19,71,81],"major":[8],"challenge":[9],"in":[10,45,107],"image":[11,15,109],"captioning.":[12],"When":[13],"the":[14,24,29,32,40,50,65,76,98,104,108,112,128],"captioning":[16,94],"model":[17,33,113],"predicts":[18],"word,":[20,30],"it":[21],"should":[22],"consider":[23],"visual":[25,100],"evidence":[26,38],"associated":[27],"with":[28],"but":[31],"tends":[34],"to":[35,102,114],"use":[36,97],"contextual":[37],"from":[39,64],"and":[43,69,110,126],"results":[44],"biased":[46,53],"captions,":[47],"especially":[48],"when":[49],"is":[52,91],"toward":[54],"some":[55],"specific":[56],"situations.":[57],"To":[58],"solve":[59],"this":[60],"problem,":[61],"we":[62,79],"approach":[63],"causal":[66,72,77],"inference":[67],"perspective":[68],"design":[70],"graph.":[73],"Based":[74],"on":[75,123],"graph,":[78],"propose":[80],"novel":[82],"method":[83,122],"named":[84],"C":[85],"<sup":[86],"xmlns:mml=\"http://www.w3.org/1998/Math/MathML\"":[87],"xmlns:xlink=\"http://www.w3.org/1999/xlink\">2</sup>":[88],"Cap":[89],"which":[90],"CLIP":[92],"confounder-free":[93],"network.":[95],"We":[96,118],"global":[99],"confounder":[101],"control":[103],"confounding":[105],"factors":[106],"train":[111],"produce":[115],"debiased":[116],"captions.":[117],"validate":[119],"our":[120,131],"proposed":[121],"MSCOCO":[124],"benchmark":[125],"demonstrate":[127],"effectiveness":[129],"of":[130],"method.":[132],"https://github.com/yeonju7kim/C2Cap":[133]},"counts_by_year":[{"year":2025,"cited_by_count":4},{"year":2024,"cited_by_count":2}],"updated_date":"2026-06-11T09:08:48.828518","created_date":"2025-10-10T00:00:00"}