{"id":"https://openalex.org/W4413156641","doi":"https://doi.org/10.1109/cvpr52734.2025.02197","title":"Devil is in the Detail: Towards Injecting Fine Details of Image Prompt in Image Generation via Conflict-free Guidance and Stratified Attention","display_name":"Devil is in the Detail: Towards Injecting Fine Details of Image Prompt in Image Generation via Conflict-free Guidance and Stratified Attention","publication_year":2025,"publication_date":"2025-06-10","ids":{"openalex":"https://openalex.org/W4413156641","doi":"https://doi.org/10.1109/cvpr52734.2025.02197"},"language":"en","primary_location":{"id":"doi:10.1109/cvpr52734.2025.02197","is_oa":false,"landing_page_url":"https://doi.org/10.1109/cvpr52734.2025.02197","pdf_url":null,"source":null,"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"2025 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR)","raw_type":"proceedings-article"},"type":"conference-paper","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/A5113944043","display_name":"Kyungmin Jo","orcid":"https://orcid.org/0000-0002-6291-9800"},"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":"Kyungmin Jo","raw_affiliation_strings":["Korea Advanced Institute of Science and Technology (KAIST),Daejeon,Korea"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Korea Advanced Institute of Science and Technology (KAIST),Daejeon,Korea","institution_ids":["https://openalex.org/I157485424"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5041375258","display_name":"Jooyeol Yun","orcid":"https://orcid.org/0000-0001-7853-7703"},"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":"Jooyeol Yun","raw_affiliation_strings":["Korea Advanced Institute of Science and Technology (KAIST),Daejeon,Korea"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Korea Advanced Institute of Science and Technology (KAIST),Daejeon,Korea","institution_ids":["https://openalex.org/I157485424"]}]},{"author_position":"last","author":{"id":"https://openalex.org/A5047912015","display_name":"Jaegul Choo","orcid":"https://orcid.org/0000-0003-1071-4835"},"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":"Jaegul Choo","raw_affiliation_strings":["Korea Advanced Institute of Science and Technology (KAIST),Daejeon,Korea"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Korea Advanced Institute of Science and Technology (KAIST),Daejeon,Korea","institution_ids":["https://openalex.org/I157485424"]}]}],"institutions":[],"countries_distinct_count":1,"institutions_distinct_count":1,"corresponding_author_ids":[],"corresponding_institution_ids":["https://openalex.org/I157485424"],"apc_list":null,"apc_paid":null,"fwci":0.0,"has_fulltext":false,"cited_by_count":0,"citation_normalized_percentile":{"value":0.19863364,"is_in_top_1_percent":false,"is_in_top_10_percent":false},"cited_by_percentile_year":null,"biblio":{"volume":null,"issue":null,"first_page":"23595","last_page":"23603"},"is_retracted":false,"is_paratext":false,"is_xpac":false,"primary_topic":{"id":"https://openalex.org/T10531","display_name":"Advanced Vision and Imaging","score":0.992900013923645,"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/T10531","display_name":"Advanced Vision and Imaging","score":0.992900013923645,"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/T10627","display_name":"Advanced Image and Video Retrieval Techniques","score":0.9821000099182129,"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/T10191","display_name":"Robotics and Sensor-Based Localization","score":0.9779000282287598,"subfield":{"id":"https://openalex.org/subfields/2202","display_name":"Aerospace Engineering"},"field":{"id":"https://openalex.org/fields/22","display_name":"Engineering"},"domain":{"id":"https://openalex.org/domains/3","display_name":"Physical Sciences"}}],"keywords":[{"id":"https://openalex.org/keywords/image","display_name":"Image (mathematics)","score":0.6584359407424927},{"id":"https://openalex.org/keywords/computer-science","display_name":"Computer science","score":0.5999077558517456},{"id":"https://openalex.org/keywords/artificial-intelligence","display_name":"Artificial intelligence","score":0.4055016338825226},{"id":"https://openalex.org/keywords/computer-vision","display_name":"Computer vision","score":0.38806214928627014}],"concepts":[{"id":"https://openalex.org/C115961682","wikidata":"https://www.wikidata.org/wiki/Q860623","display_name":"Image (mathematics)","level":2,"score":0.6584359407424927},{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.5999077558517456},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.4055016338825226},{"id":"https://openalex.org/C31972630","wikidata":"https://www.wikidata.org/wiki/Q844240","display_name":"Computer vision","level":1,"score":0.38806214928627014}],"mesh":[],"locations_count":1,"locations":[{"id":"doi:10.1109/cvpr52734.2025.02197","is_oa":false,"landing_page_url":"https://doi.org/10.1109/cvpr52734.2025.02197","pdf_url":null,"source":null,"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"2025 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR)","raw_type":"proceedings-article"}],"best_oa_location":null,"sustainable_development_goals":[],"awards":[],"funders":[{"id":"https://openalex.org/F4320322120","display_name":"National Research Foundation of Korea","ror":"https://ror.org/013aysd81"}],"has_content":{"grobid_xml":false,"pdf":false},"content_urls":null,"referenced_works_count":12,"referenced_works":["https://openalex.org/W1861492603","https://openalex.org/W3159481202","https://openalex.org/W4312933868","https://openalex.org/W4386072096","https://openalex.org/W4386113271","https://openalex.org/W4390873054","https://openalex.org/W4390873195","https://openalex.org/W4402667886","https://openalex.org/W4402704604","https://openalex.org/W4402716004","https://openalex.org/W4402727926","https://openalex.org/W4402775818"],"related_works":["https://openalex.org/W2772917594","https://openalex.org/W2036807459","https://openalex.org/W2058170566","https://openalex.org/W2755342338","https://openalex.org/W2166024367","https://openalex.org/W3116076068","https://openalex.org/W2229312674","https://openalex.org/W2951359407","https://openalex.org/W2079911747","https://openalex.org/W1969923398"],"abstract_inverted_index":{"While":[0],"large-scale":[1],"text-to-image":[2],"diffusion":[3,108],"models":[4,109,303],"enable":[5],"the":[6,26,53,67,72,77,112,130,140,150,153,193,198,206,216,219,225,245,254,297,307],"generation":[7,292],"of":[8,103,132,142,218],"high-quality,":[9],"diverse":[10],"images":[11,40,62,126,279],"from":[12,29,71,110,158,235,244,253,277],"text":[13,90],"prompts,":[14],"these":[15,117],"prompts":[16,144,166,186],"struggle":[17],"to":[18,36,38,46,56,80,88],"capture":[19],"intricate":[20],"details,":[21],"such":[22],"as":[23,47,167,188],"textures,":[24],"preventing":[25],"user":[27],"intent":[28],"being":[30],"reflected.":[31],"This":[32,75],"limitation":[33],"has":[34],"led":[35],"efforts":[37],"generate":[39],"conditioned":[41],"on":[42],"user-provided":[43],"images,":[44],"referred":[45],"image":[48,58,73,113,133,143,165,185,195,199,221,226,246,256,291,308],"prompts.":[49,91,134],"Recent":[50],"work":[51,81],"modifies":[52],"self-attention":[54,78,105,210,266],"mechanism":[55],"impose":[57],"conditions":[59,155],"in":[60,100,145,304],"generated":[61,194,220,255],"by":[63,183,229],"replacing":[64],"or":[65],"concatenating":[66],"keys":[68,232,241,274],"and":[69,156,170,222,233,242,275],"values":[70,234,243,276],"prompt.":[74,114,200,309],"enables":[76],"layer":[79],"like":[82],"a":[83,121,213],"cross-attention":[84],"layer,":[85],"generally":[86],"used":[87],"incorporate":[89],"In":[92,201],"this":[93],"paper,":[94],"we":[95,119,179,203,262,294],"identify":[96],"two":[97,207],"common":[98,209],"issues":[99],"existing":[101,136,301],"methods":[102,163],"modifying":[104],"that":[106,124,127,192,205,296],"hinder":[107],"reflecting":[111,306],"By":[115],"addressing":[116],"issues,":[118],"propose":[120,180,263],"novel":[122],"method":[123,299],"generates":[125],"properly":[128],"reflect":[129],"details":[131],"First,":[135],"approaches":[137],"often":[138],"neglect":[139],"importance":[141],"classifier-free":[146],"guidance,":[147],"which":[148,271],"directs":[149],"model":[151],"towards":[152],"intended":[154],"away":[157],"those":[159],"undesirable.":[160],"Specifically,":[161,238],"current":[162],"use":[164],"both":[168,236,278],"desired":[169,189],"undesired":[171],"conditions,":[172,190],"causing":[173],"conflicting":[174],"signals.":[175],"To":[176,259],"resolve":[177],"this,":[178],"conflict-free":[181],"guidance":[182],"using":[184,231],"only":[187],"ensuring":[191],"faithfully":[196,305],"reflects":[197],"addition,":[202],"observe":[204],"most":[208],"modifications":[211],"involve":[212],"trade-off":[214],"between":[215,283],"realism":[217],"alignment":[223],"with":[224],"prompt,":[227],"achieved":[228],"selectively":[230],"images.":[237],"selecting":[239,251,282],"more":[240,252],"prompt":[247],"improves":[248],"alignment,":[249],"while":[250],"enhances":[257],"realism.":[258],"balance":[260],"both,":[261],"an":[264],"alternative":[265],"modification":[267],"method,":[268],"Stratified":[269],"Attention,":[270],"jointly":[272],"uses":[273],"rather":[280],"than":[281],"them.":[284],"Through":[285],"extensive":[286],"experiments":[287],"across":[288],"three":[289],"distinct":[290],"tasks,":[293],"demonstrate":[295],"proposed":[298],"outperforms":[300],"image-prompting":[302]},"counts_by_year":[],"updated_date":"2026-07-29T14:22:42.915294","created_date":"2025-10-10T00:00:00"}