{"id":"https://openalex.org/W4392903436","doi":"https://doi.org/10.1109/icassp48485.2024.10447874","title":"GBSD: Generative Bokeh with Stage Diffusion","display_name":"GBSD: Generative Bokeh with Stage Diffusion","publication_year":2024,"publication_date":"2024-03-18","ids":{"openalex":"https://openalex.org/W4392903436","doi":"https://doi.org/10.1109/icassp48485.2024.10447874"},"language":"en","primary_location":{"id":"doi:10.1109/icassp48485.2024.10447874","is_oa":false,"landing_page_url":"https://doi.org/10.1109/icassp48485.2024.10447874","pdf_url":null,"source":null,"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"ICASSP 2024 - 2024 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/A5080327156","display_name":"Jieren Deng","orcid":"https://orcid.org/0000-0002-5738-0927"},"institutions":[{"id":"https://openalex.org/I98301712","display_name":"Baidu (China)","ror":"https://ror.org/03vs3wt56","country_code":"CN","type":"company","lineage":["https://openalex.org/I98301712"]},{"id":"https://openalex.org/I140172145","display_name":"University of Connecticut","ror":"https://ror.org/02der9h97","country_code":"US","type":"education","lineage":["https://openalex.org/I140172145"]}],"countries":["CN","US"],"is_corresponding":true,"raw_author_name":"Jieren Deng","raw_affiliation_strings":["Baidu Research USA,Sunnyvale,CA,USA","University of Connecticut, CT, USA","Baidu Research USA, Sunnyvale, CA, USA"],"affiliations":[{"raw_affiliation_string":"Baidu Research USA,Sunnyvale,CA,USA","institution_ids":["https://openalex.org/I98301712"]},{"raw_affiliation_string":"University of Connecticut, CT, USA","institution_ids":["https://openalex.org/I140172145"]},{"raw_affiliation_string":"Baidu Research USA, Sunnyvale, CA, USA","institution_ids":[]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5100424243","display_name":"Xin Zhou","orcid":"https://orcid.org/0009-0004-3288-5306"},"institutions":[{"id":"https://openalex.org/I98301712","display_name":"Baidu (China)","ror":"https://ror.org/03vs3wt56","country_code":"CN","type":"company","lineage":["https://openalex.org/I98301712"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Xin Zhou","raw_affiliation_strings":["Baidu Research USA,Sunnyvale,CA,USA","Baidu Research USA, Sunnyvale, CA, USA"],"affiliations":[{"raw_affiliation_string":"Baidu Research USA,Sunnyvale,CA,USA","institution_ids":["https://openalex.org/I98301712"]},{"raw_affiliation_string":"Baidu Research USA, Sunnyvale, CA, USA","institution_ids":[]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5100771541","display_name":"Hao Tian","orcid":"https://orcid.org/0000-0001-8219-9743"},"institutions":[{"id":"https://openalex.org/I98301712","display_name":"Baidu (China)","ror":"https://ror.org/03vs3wt56","country_code":"CN","type":"company","lineage":["https://openalex.org/I98301712"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Hao Tian","raw_affiliation_strings":["Baidu Research USA,Sunnyvale,CA,USA","Baidu Research USA, Sunnyvale, CA, USA"],"affiliations":[{"raw_affiliation_string":"Baidu Research USA,Sunnyvale,CA,USA","institution_ids":["https://openalex.org/I98301712"]},{"raw_affiliation_string":"Baidu Research USA, Sunnyvale, CA, USA","institution_ids":[]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5014233468","display_name":"Zhihong Pan","orcid":"https://orcid.org/0000-0003-0866-762X"},"institutions":[{"id":"https://openalex.org/I98301712","display_name":"Baidu (China)","ror":"https://ror.org/03vs3wt56","country_code":"CN","type":"company","lineage":["https://openalex.org/I98301712"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Zhihong Pan","raw_affiliation_strings":["Baidu Research USA,Sunnyvale,CA,USA","Baidu Research USA, Sunnyvale, CA, USA"],"affiliations":[{"raw_affiliation_string":"Baidu Research USA,Sunnyvale,CA,USA","institution_ids":["https://openalex.org/I98301712"]},{"raw_affiliation_string":"Baidu Research USA, Sunnyvale, CA, USA","institution_ids":[]}]},{"author_position":"last","author":{"id":"https://openalex.org/A5083334761","display_name":"Derek Aguiar","orcid":"https://orcid.org/0000-0001-9166-8783"},"institutions":[{"id":"https://openalex.org/I140172145","display_name":"University of Connecticut","ror":"https://ror.org/02der9h97","country_code":"US","type":"education","lineage":["https://openalex.org/I140172145"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Derek Aguiar","raw_affiliation_strings":["University of Connecticut,CT,USA","University of Connecticut, CT, USA"],"affiliations":[{"raw_affiliation_string":"University of Connecticut,CT,USA","institution_ids":["https://openalex.org/I140172145"]},{"raw_affiliation_string":"University of Connecticut, CT, USA","institution_ids":["https://openalex.org/I140172145"]}]}],"institutions":[],"countries_distinct_count":2,"institutions_distinct_count":5,"corresponding_author_ids":["https://openalex.org/A5080327156"],"corresponding_institution_ids":["https://openalex.org/I140172145","https://openalex.org/I98301712"],"apc_list":null,"apc_paid":null,"fwci":0.3374,"has_fulltext":false,"cited_by_count":1,"citation_normalized_percentile":{"value":0.57566086,"is_in_top_1_percent":false,"is_in_top_10_percent":false},"cited_by_percentile_year":{"min":91,"max":95},"biblio":{"volume":null,"issue":null,"first_page":"7070","last_page":"7074"},"is_retracted":false,"is_paratext":false,"is_xpac":false,"primary_topic":{"id":"https://openalex.org/T13114","display_name":"Image Processing Techniques and Applications","score":0.9986000061035156,"subfield":{"id":"https://openalex.org/subfields/2214","display_name":"Media Technology"},"field":{"id":"https://openalex.org/fields/22","display_name":"Engineering"},"domain":{"id":"https://openalex.org/domains/3","display_name":"Physical Sciences"}},"topics":[{"id":"https://openalex.org/T13114","display_name":"Image Processing Techniques and Applications","score":0.9986000061035156,"subfield":{"id":"https://openalex.org/subfields/2214","display_name":"Media Technology"},"field":{"id":"https://openalex.org/fields/22","display_name":"Engineering"},"domain":{"id":"https://openalex.org/domains/3","display_name":"Physical Sciences"}},{"id":"https://openalex.org/T12859","display_name":"Cell Image Analysis Techniques","score":0.9929999709129333,"subfield":{"id":"https://openalex.org/subfields/1304","display_name":"Biophysics"},"field":{"id":"https://openalex.org/fields/13","display_name":"Biochemistry, Genetics and Molecular Biology"},"domain":{"id":"https://openalex.org/domains/1","display_name":"Life Sciences"}},{"id":"https://openalex.org/T11105","display_name":"Advanced Image Processing Techniques","score":0.9889000058174133,"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/rendering","display_name":"Rendering (computer graphics)","score":0.8554463386535645},{"id":"https://openalex.org/keywords/computer-science","display_name":"Computer science","score":0.7882481813430786},{"id":"https://openalex.org/keywords/generative-grammar","display_name":"Generative grammar","score":0.7360479831695557},{"id":"https://openalex.org/keywords/artificial-intelligence","display_name":"Artificial intelligence","score":0.5840730667114258},{"id":"https://openalex.org/keywords/generative-model","display_name":"Generative model","score":0.5487202405929565},{"id":"https://openalex.org/keywords/image-synthesis","display_name":"Image synthesis","score":0.5447521805763245},{"id":"https://openalex.org/keywords/computer-graphics","display_name":"Computer graphics","score":0.5078323483467102},{"id":"https://openalex.org/keywords/computer-vision","display_name":"Computer vision","score":0.49479907751083374},{"id":"https://openalex.org/keywords/image-based-modeling-and-rendering","display_name":"Image-based modeling and rendering","score":0.4480040669441223},{"id":"https://openalex.org/keywords/computer-graphics","display_name":"Computer graphics (images)","score":0.44650745391845703},{"id":"https://openalex.org/keywords/focus","display_name":"Focus (optics)","score":0.42889320850372314},{"id":"https://openalex.org/keywords/graphics","display_name":"Graphics","score":0.4173937141895294},{"id":"https://openalex.org/keywords/image","display_name":"Image (mathematics)","score":0.3624194264411926}],"concepts":[{"id":"https://openalex.org/C205711294","wikidata":"https://www.wikidata.org/wiki/Q176953","display_name":"Rendering (computer graphics)","level":2,"score":0.8554463386535645},{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.7882481813430786},{"id":"https://openalex.org/C39890363","wikidata":"https://www.wikidata.org/wiki/Q36108","display_name":"Generative grammar","level":2,"score":0.7360479831695557},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.5840730667114258},{"id":"https://openalex.org/C167966045","wikidata":"https://www.wikidata.org/wiki/Q5532625","display_name":"Generative model","level":3,"score":0.5487202405929565},{"id":"https://openalex.org/C2989087649","wikidata":"https://www.wikidata.org/wiki/Q176953","display_name":"Image synthesis","level":3,"score":0.5447521805763245},{"id":"https://openalex.org/C77660652","wikidata":"https://www.wikidata.org/wiki/Q150971","display_name":"Computer graphics","level":2,"score":0.5078323483467102},{"id":"https://openalex.org/C31972630","wikidata":"https://www.wikidata.org/wiki/Q844240","display_name":"Computer vision","level":1,"score":0.49479907751083374},{"id":"https://openalex.org/C44185422","wikidata":"https://www.wikidata.org/wiki/Q6002064","display_name":"Image-based modeling and rendering","level":3,"score":0.4480040669441223},{"id":"https://openalex.org/C121684516","wikidata":"https://www.wikidata.org/wiki/Q7600677","display_name":"Computer graphics (images)","level":1,"score":0.44650745391845703},{"id":"https://openalex.org/C192209626","wikidata":"https://www.wikidata.org/wiki/Q190909","display_name":"Focus (optics)","level":2,"score":0.42889320850372314},{"id":"https://openalex.org/C21442007","wikidata":"https://www.wikidata.org/wiki/Q1027879","display_name":"Graphics","level":2,"score":0.4173937141895294},{"id":"https://openalex.org/C115961682","wikidata":"https://www.wikidata.org/wiki/Q860623","display_name":"Image (mathematics)","level":2,"score":0.3624194264411926},{"id":"https://openalex.org/C121332964","wikidata":"https://www.wikidata.org/wiki/Q413","display_name":"Physics","level":0,"score":0.0},{"id":"https://openalex.org/C120665830","wikidata":"https://www.wikidata.org/wiki/Q14620","display_name":"Optics","level":1,"score":0.0}],"mesh":[],"locations_count":1,"locations":[{"id":"doi:10.1109/icassp48485.2024.10447874","is_oa":false,"landing_page_url":"https://doi.org/10.1109/icassp48485.2024.10447874","pdf_url":null,"source":null,"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"ICASSP 2024 - 2024 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP)","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":31,"referenced_works":["https://openalex.org/W1583837637","https://openalex.org/W1965475047","https://openalex.org/W2116196132","https://openalex.org/W2155113687","https://openalex.org/W2156699058","https://openalex.org/W2311236754","https://openalex.org/W2885098754","https://openalex.org/W2894627214","https://openalex.org/W2896512449","https://openalex.org/W2914304175","https://openalex.org/W3034482833","https://openalex.org/W3034960835","https://openalex.org/W3106070046","https://openalex.org/W3155072588","https://openalex.org/W3166762869","https://openalex.org/W3174957407","https://openalex.org/W4212774754","https://openalex.org/W4224035735","https://openalex.org/W4281485151","https://openalex.org/W4283388932","https://openalex.org/W4312497550","https://openalex.org/W4312933868","https://openalex.org/W4313177683","https://openalex.org/W4386057725","https://openalex.org/W6779823529","https://openalex.org/W6783713337","https://openalex.org/W6791353385","https://openalex.org/W6795288823","https://openalex.org/W6809885388","https://openalex.org/W6838639034","https://openalex.org/W6839643428"],"related_works":["https://openalex.org/W4365211920","https://openalex.org/W3014948380","https://openalex.org/W4380551139","https://openalex.org/W4317695495","https://openalex.org/W4395044357","https://openalex.org/W4287117424","https://openalex.org/W4387506531","https://openalex.org/W2087346071","https://openalex.org/W2967848559","https://openalex.org/W4299831724"],"abstract_inverted_index":{"The":[0],"bokeh":[1,39,65,80,96,121,137],"effect":[2,132,138],"is":[3,139],"an":[4],"artistic":[5],"technique":[6],"that":[7,67,90],"blurs":[8],"out-of-focus":[9],"areas":[10],"in":[11,22,70,105,160],"a":[12,95,115],"photograph":[13],"and":[14,25,31,151,153,163],"has":[15],"gained":[16],"interest":[17],"due":[18],"to":[19,63,119,157],"recent":[20],"developments":[21],"text-to-image":[23,88,162],"synthesis":[24,102],"the":[26,71,85,130],"ubiquity":[27],"of":[28],"smartphone":[29],"cameras":[30],"photo":[32],"sharing":[33],"apps.":[34],"Prior":[35],"work":[36],"on":[37,43,123,133],"rendering":[38,52,144],"effects":[40,66,122],"have":[41,55],"focused":[42],"manipulating":[44],"photographs":[45],"using":[46],"classical":[47,143],"computer":[48],"graphics":[49],"or":[50,60],"neural":[51],"techniques,":[53],"but":[54],"either":[56],"depth":[57],"discontinuity":[58],"artifacts":[59],"are":[61,68],"restricted":[62],"reproducing":[64],"present":[69,78],"training":[72],"data.":[73],"In":[74],"this":[75,135],"paper,":[76],"we":[77],"generative":[79,87],"with":[81,94,114],"stage":[82],"diffusion":[83,106,112],"(GBSD),":[84],"first":[86],"model":[89],"synthesizes":[91],"photorealistic":[92],"images":[93],"style.":[97],"Motivated":[98],"by":[99],"how":[100],"image":[101],"occurs":[103],"progressively":[104],"models,":[107],"our":[108],"approach":[109],"combines":[110],"latent":[111],"models":[113],"2-stage":[116],"conditioning":[117],"algorithm":[118],"render":[120],"semantically":[124],"defined":[125],"objects.":[126],"Since":[127],"GBSD":[128,148],"focuses":[129],"blurring":[131],"objects,":[134],"semantic":[136],"more":[140],"versatile":[141],"than":[142],"techniques.":[145],"We":[146],"evaluate":[147],"both":[149,161],"quantitatively":[150],"qualitatively":[152],"demonstrate":[154],"its":[155],"ability":[156],"be":[158],"applied":[159],"image-to-image":[164],"settings.":[165]},"counts_by_year":[{"year":2025,"cited_by_count":1}],"updated_date":"2025-11-06T03:46:38.306776","created_date":"2025-10-10T00:00:00"}