{"id":"https://openalex.org/W7160288605","doi":"https://doi.org/10.1109/wacv61042.2026.00204","title":"Evaluating Text-to-Image and Text-to-Video Synthesis with a Conditional Fr\u00e9chet Distance","display_name":"Evaluating Text-to-Image and Text-to-Video Synthesis with a Conditional Fr\u00e9chet Distance","publication_year":2026,"publication_date":"2026-03-06","ids":{"openalex":"https://openalex.org/W7160288605","doi":"https://doi.org/10.1109/wacv61042.2026.00204"},"language":null,"primary_location":{"id":"doi:10.1109/wacv61042.2026.00204","is_oa":false,"landing_page_url":"https://doi.org/10.1109/wacv61042.2026.00204","pdf_url":null,"source":null,"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"2026 IEEE/CVF Winter Conference on Applications of Computer Vision (WACV)","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/A5071208482","display_name":"Jaywon Koo","orcid":"https://orcid.org/0000-0002-5539-5244"},"institutions":[{"id":"https://openalex.org/I74775410","display_name":"Rice University","ror":"https://ror.org/008zs3103","country_code":"US","type":"education","lineage":["https://openalex.org/I74775410"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Jaywon Koo","raw_affiliation_strings":["Rice University"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Rice University","institution_ids":["https://openalex.org/I74775410"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5067752616","display_name":"Jefferson Hernandez","orcid":"https://orcid.org/0000-0002-7091-0478"},"institutions":[{"id":"https://openalex.org/I74775410","display_name":"Rice University","ror":"https://ror.org/008zs3103","country_code":"US","type":"education","lineage":["https://openalex.org/I74775410"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Jefferson Hernandez","raw_affiliation_strings":["Rice University"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Rice University","institution_ids":["https://openalex.org/I74775410"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5135341346","display_name":"Moayed Haji-Ali","orcid":null},"institutions":[{"id":"https://openalex.org/I74775410","display_name":"Rice University","ror":"https://ror.org/008zs3103","country_code":"US","type":"education","lineage":["https://openalex.org/I74775410"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Moayed Haji-Ali","raw_affiliation_strings":["Rice University"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Rice University","institution_ids":["https://openalex.org/I74775410"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5102577563","display_name":"Ziyan Yang","orcid":"https://orcid.org/0009-0006-0378-0462"},"institutions":[{"id":"https://openalex.org/I74775410","display_name":"Rice University","ror":"https://ror.org/008zs3103","country_code":"US","type":"education","lineage":["https://openalex.org/I74775410"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Ziyan Yang","raw_affiliation_strings":["Rice University"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Rice University","institution_ids":["https://openalex.org/I74775410"]}]},{"author_position":"last","author":{"id":"https://openalex.org/A5027328044","display_name":"Vicente Ord\u00f3\u00f1ez","orcid":null},"institutions":[{"id":"https://openalex.org/I74775410","display_name":"Rice University","ror":"https://ror.org/008zs3103","country_code":"US","type":"education","lineage":["https://openalex.org/I74775410"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Vicente Ordonez","raw_affiliation_strings":["Rice University"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Rice University","institution_ids":["https://openalex.org/I74775410"]}]}],"institutions":[],"countries_distinct_count":1,"institutions_distinct_count":5,"corresponding_author_ids":[],"corresponding_institution_ids":[],"apc_list":null,"apc_paid":null,"fwci":0.0,"has_fulltext":false,"cited_by_count":0,"citation_normalized_percentile":{"value":0.68811931,"is_in_top_1_percent":false,"is_in_top_10_percent":false},"cited_by_percentile_year":null,"biblio":{"volume":null,"issue":null,"first_page":"2052","last_page":"2062"},"is_retracted":false,"is_paratext":false,"is_xpac":false,"primary_topic":{"id":"https://openalex.org/T10775","display_name":"Generative Adversarial Networks and Image Synthesis","score":0.29899999499320984,"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/T10775","display_name":"Generative Adversarial Networks and Image Synthesis","score":0.29899999499320984,"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/T11439","display_name":"Video Analysis and Summarization","score":0.17489999532699585,"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/T12720","display_name":"Multimedia Communication and Technology","score":0.07810000330209732,"subfield":{"id":"https://openalex.org/subfields/3312","display_name":"Sociology and Political Science"},"field":{"id":"https://openalex.org/fields/33","display_name":"Social Sciences"},"domain":{"id":"https://openalex.org/domains/2","display_name":"Social Sciences"}}],"keywords":[{"id":"https://openalex.org/keywords/set","display_name":"Set (abstract data type)","score":0.2915000021457672},{"id":"https://openalex.org/keywords/identification","display_name":"Identification (biology)","score":0.2578999996185303},{"id":"https://openalex.org/keywords/feature","display_name":"Feature (linguistics)","score":0.2565999925136566},{"id":"https://openalex.org/keywords/noise","display_name":"Noise (video)","score":0.2535000145435333},{"id":"https://openalex.org/keywords/measure","display_name":"Measure (data warehouse)","score":0.2515999972820282}],"concepts":[{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.4413999915122986},{"id":"https://openalex.org/C33923547","wikidata":"https://www.wikidata.org/wiki/Q395","display_name":"Mathematics","level":0,"score":0.41909998655319214},{"id":"https://openalex.org/C11413529","wikidata":"https://www.wikidata.org/wiki/Q8366","display_name":"Algorithm","level":1,"score":0.40059998631477356},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.3068000078201294},{"id":"https://openalex.org/C177264268","wikidata":"https://www.wikidata.org/wiki/Q1514741","display_name":"Set (abstract data type)","level":2,"score":0.2915000021457672},{"id":"https://openalex.org/C116834253","wikidata":"https://www.wikidata.org/wiki/Q2039217","display_name":"Identification (biology)","level":2,"score":0.2578999996185303},{"id":"https://openalex.org/C2776401178","wikidata":"https://www.wikidata.org/wiki/Q12050496","display_name":"Feature (linguistics)","level":2,"score":0.2565999925136566},{"id":"https://openalex.org/C124101348","wikidata":"https://www.wikidata.org/wiki/Q172491","display_name":"Data mining","level":1,"score":0.2563999891281128},{"id":"https://openalex.org/C99498987","wikidata":"https://www.wikidata.org/wiki/Q2210247","display_name":"Noise (video)","level":3,"score":0.2535000145435333},{"id":"https://openalex.org/C2780009758","wikidata":"https://www.wikidata.org/wiki/Q6804172","display_name":"Measure (data warehouse)","level":2,"score":0.2515999972820282},{"id":"https://openalex.org/C98045186","wikidata":"https://www.wikidata.org/wiki/Q205663","display_name":"Process (computing)","level":2,"score":0.25130000710487366}],"mesh":[],"locations_count":1,"locations":[{"id":"doi:10.1109/wacv61042.2026.00204","is_oa":false,"landing_page_url":"https://doi.org/10.1109/wacv61042.2026.00204","pdf_url":null,"source":null,"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"2026 IEEE/CVF Winter Conference on Applications of Computer Vision (WACV)","raw_type":"proceedings-article"}],"best_oa_location":null,"sustainable_development_goals":[],"awards":[],"funders":[],"has_content":{"grobid_xml":false,"pdf":false},"content_urls":null,"referenced_works_count":28,"referenced_works":["https://openalex.org/W1861492603","https://openalex.org/W2183341477","https://openalex.org/W2962785568","https://openalex.org/W2963966654","https://openalex.org/W2964024144","https://openalex.org/W3153469116","https://openalex.org/W3176381160","https://openalex.org/W4312933868","https://openalex.org/W4386076215","https://openalex.org/W4390873054","https://openalex.org/W4390874002","https://openalex.org/W4402671992","https://openalex.org/W4402716201","https://openalex.org/W4402716307","https://openalex.org/W4402727293","https://openalex.org/W4402727496","https://openalex.org/W4402727647","https://openalex.org/W4403601051","https://openalex.org/W4403792127","https://openalex.org/W4404782905","https://openalex.org/W4410659598","https://openalex.org/W7125921921","https://openalex.org/W7133185348","https://openalex.org/W7133195915","https://openalex.org/W7133208264","https://openalex.org/W7133209815","https://openalex.org/W7133217041","https://openalex.org/W7133250524"],"related_works":[],"abstract_inverted_index":{"Evaluating":[0],"text-to-image":[1,115],"and":[2,20,57,97,117,168],"text-to-video":[3],"models":[4,93,116],"is":[5],"challenging":[6],"due":[7],"to":[8,15,26,87,100,102,130],"a":[9,27,41,46,61,123,144],"fundamental":[10],"disconnect:":[11],"established":[12],"metrics":[13,65,134],"fail":[14],"jointly":[16],"measure":[17],"visual":[18,55,88],"quality":[19,74],"semantic":[21],"alignment":[22,80],"with":[23,30,126,136],"text,":[24],"leading":[25],"poor":[28],"correlation":[29,125],"human":[31,127,137],"judgments.":[32],"To":[33],"address":[34],"this":[35,159],"critical":[36],"issue,":[37],"we":[38],"propose":[39],"cFreD,":[40],"general":[42],"metric":[43,147],"based":[44],"on":[45],"Conditional":[47],"Fr\u00e9chet":[48,68],"Distance":[49,70],"that":[50],"unifies":[51],"the":[52,149],"assessment":[53],"of":[54,152],"fidelity":[56],"text-prompt":[58],"consistency":[59],"into":[60],"single":[62],"score.":[63],"Existing":[64],"such":[66,82],"as":[67,83,143],"Inception":[69],"(FID)":[71],"capture":[72],"image":[73],"but":[75],"ignore":[76],"text":[77,153],"conditioning":[78],"while":[79],"scores":[81],"CLIPScore":[84],"are":[85,98],"insensitive":[86],"quality.":[89],"Furthermore,":[90],"learned":[91],"preference":[92],"require":[94],"constant":[95],"retraining":[96],"unlikely":[99],"generalize":[101],"novel":[103],"architectures":[104],"or":[105],"out-of-distribution":[106],"prompts.":[107],"Through":[108],"extensive":[109],"experiments":[110],"across":[111],"multiple":[112],"recently":[113],"proposed":[114],"diverse":[118],"prompt":[119],"datasets,":[120],"cFreD":[121,142],"exhibits":[122],"higher":[124],"judgments":[128],"compared":[129],"statistical":[131],"metrics,":[132],"including":[133],"trained":[135],"preferences.":[138],"Our":[139],"findings":[140],"validate":[141],"robust,":[145],"future-proof":[146],"for":[148],"systematic":[150],"evaluation":[151,166],"conditioned":[154],"models,":[155],"standardizing":[156],"benchmarking":[157],"in":[158],"rapidly":[160],"evolving":[161],"field.":[162],"We":[163],"release":[164],"our":[165],"toolkit":[167],"benchmark":[169],"here.":[170]},"counts_by_year":[],"updated_date":"2026-06-11T09:08:48.828518","created_date":"2026-05-06T00:00:00"}