{"id":"https://openalex.org/W4387968307","doi":"https://doi.org/10.1145/3581783.3612407","title":"Open-Vocabulary Object Detection via Scene Graph Discovery","display_name":"Open-Vocabulary Object Detection via Scene Graph Discovery","publication_year":2023,"publication_date":"2023-10-26","ids":{"openalex":"https://openalex.org/W4387968307","doi":"https://doi.org/10.1145/3581783.3612407"},"language":"en","primary_location":{"id":"doi:10.1145/3581783.3612407","is_oa":false,"landing_page_url":"https://doi.org/10.1145/3581783.3612407","pdf_url":null,"source":null,"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"Proceedings of the 31st ACM International Conference on Multimedia","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/A5077320973","display_name":"Hengcan Shi","orcid":"https://orcid.org/0000-0002-1340-0009"},"institutions":[{"id":"https://openalex.org/I56590836","display_name":"Monash University","ror":"https://ror.org/02bfwt286","country_code":"AU","type":"education","lineage":["https://openalex.org/I56590836"]}],"countries":["AU"],"is_corresponding":true,"raw_author_name":"Hengcan Shi","raw_affiliation_strings":["Monash University, Melbourne, VIC, Australia"],"affiliations":[{"raw_affiliation_string":"Monash University, Melbourne, VIC, Australia","institution_ids":["https://openalex.org/I56590836"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5041437150","display_name":"Munawar Hayat","orcid":"https://orcid.org/0000-0002-2706-5985"},"institutions":[{"id":"https://openalex.org/I56590836","display_name":"Monash University","ror":"https://ror.org/02bfwt286","country_code":"AU","type":"education","lineage":["https://openalex.org/I56590836"]}],"countries":["AU"],"is_corresponding":false,"raw_author_name":"Munawar Hayat","raw_affiliation_strings":["Monash University, Melbourne, VIC, Australia"],"affiliations":[{"raw_affiliation_string":"Monash University, Melbourne, VIC, Australia","institution_ids":["https://openalex.org/I56590836"]}]},{"author_position":"last","author":{"id":"https://openalex.org/A5100635804","display_name":"Jianfei Cai","orcid":"https://orcid.org/0000-0002-9444-3763"},"institutions":[{"id":"https://openalex.org/I56590836","display_name":"Monash University","ror":"https://ror.org/02bfwt286","country_code":"AU","type":"education","lineage":["https://openalex.org/I56590836"]}],"countries":["AU"],"is_corresponding":false,"raw_author_name":"Jianfei Cai","raw_affiliation_strings":["Monash University, Melbourne, VIC, Australia"],"affiliations":[{"raw_affiliation_string":"Monash University, Melbourne, VIC, Australia","institution_ids":["https://openalex.org/I56590836"]}]}],"institutions":[],"countries_distinct_count":1,"institutions_distinct_count":3,"corresponding_author_ids":["https://openalex.org/A5077320973"],"corresponding_institution_ids":["https://openalex.org/I56590836"],"apc_list":null,"apc_paid":null,"fwci":2.3289,"has_fulltext":false,"cited_by_count":19,"citation_normalized_percentile":{"value":0.90597622,"is_in_top_1_percent":false,"is_in_top_10_percent":true},"cited_by_percentile_year":{"min":98,"max":99},"biblio":{"volume":null,"issue":null,"first_page":"4012","last_page":"4021"},"is_retracted":false,"is_paratext":false,"is_xpac":false,"primary_topic":{"id":"https://openalex.org/T11714","display_name":"Multimodal Machine Learning Applications","score":0.9998999834060669,"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":0.9998999834060669,"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/T11273","display_name":"Advanced Graph Neural Networks","score":0.9934999942779541,"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/T10036","display_name":"Advanced Neural Network Applications","score":0.9915000200271606,"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":0.8013983964920044},{"id":"https://openalex.org/keywords/artificial-intelligence","display_name":"Artificial intelligence","score":0.6182977557182312},{"id":"https://openalex.org/keywords/scene-graph","display_name":"Scene graph","score":0.5793759822845459},{"id":"https://openalex.org/keywords/object-detection","display_name":"Object detection","score":0.5264891982078552},{"id":"https://openalex.org/keywords/graph","display_name":"Graph","score":0.4936996400356293},{"id":"https://openalex.org/keywords/leverage","display_name":"Leverage (statistics)","score":0.4685058891773224},{"id":"https://openalex.org/keywords/computer-vision","display_name":"Computer vision","score":0.41995108127593994},{"id":"https://openalex.org/keywords/vocabulary","display_name":"Vocabulary","score":0.4170699715614319},{"id":"https://openalex.org/keywords/pattern-recognition","display_name":"Pattern recognition (psychology)","score":0.3844667673110962},{"id":"https://openalex.org/keywords/theoretical-computer-science","display_name":"Theoretical computer science","score":0.169129878282547}],"concepts":[{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.8013983964920044},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.6182977557182312},{"id":"https://openalex.org/C179372163","wikidata":"https://www.wikidata.org/wiki/Q1406181","display_name":"Scene graph","level":3,"score":0.5793759822845459},{"id":"https://openalex.org/C2776151529","wikidata":"https://www.wikidata.org/wiki/Q3045304","display_name":"Object detection","level":3,"score":0.5264891982078552},{"id":"https://openalex.org/C132525143","wikidata":"https://www.wikidata.org/wiki/Q141488","display_name":"Graph","level":2,"score":0.4936996400356293},{"id":"https://openalex.org/C153083717","wikidata":"https://www.wikidata.org/wiki/Q6535263","display_name":"Leverage (statistics)","level":2,"score":0.4685058891773224},{"id":"https://openalex.org/C31972630","wikidata":"https://www.wikidata.org/wiki/Q844240","display_name":"Computer vision","level":1,"score":0.41995108127593994},{"id":"https://openalex.org/C2777601683","wikidata":"https://www.wikidata.org/wiki/Q6499736","display_name":"Vocabulary","level":2,"score":0.4170699715614319},{"id":"https://openalex.org/C153180895","wikidata":"https://www.wikidata.org/wiki/Q7148389","display_name":"Pattern recognition (psychology)","level":2,"score":0.3844667673110962},{"id":"https://openalex.org/C80444323","wikidata":"https://www.wikidata.org/wiki/Q2878974","display_name":"Theoretical computer science","level":1,"score":0.169129878282547},{"id":"https://openalex.org/C41895202","wikidata":"https://www.wikidata.org/wiki/Q8162","display_name":"Linguistics","level":1,"score":0.0},{"id":"https://openalex.org/C205711294","wikidata":"https://www.wikidata.org/wiki/Q176953","display_name":"Rendering (computer graphics)","level":2,"score":0.0},{"id":"https://openalex.org/C138885662","wikidata":"https://www.wikidata.org/wiki/Q5891","display_name":"Philosophy","level":0,"score":0.0}],"mesh":[],"locations_count":1,"locations":[{"id":"doi:10.1145/3581783.3612407","is_oa":false,"landing_page_url":"https://doi.org/10.1145/3581783.3612407","pdf_url":null,"source":null,"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"Proceedings of the 31st ACM International Conference on Multimedia","raw_type":"proceedings-article"}],"best_oa_location":null,"sustainable_development_goals":[{"score":0.5199999809265137,"id":"https://metadata.un.org/sdg/4","display_name":"Quality Education"}],"awards":[],"funders":[{"id":"https://openalex.org/F4320334704","display_name":"Australian Research Council","ror":"https://ror.org/05mmh0f86"}],"has_content":{"pdf":false,"grobid_xml":false},"content_urls":null,"referenced_works_count":41,"referenced_works":["https://openalex.org/W639708223","https://openalex.org/W1773149199","https://openalex.org/W1861492603","https://openalex.org/W2194775991","https://openalex.org/W2250378130","https://openalex.org/W2277195237","https://openalex.org/W2579549467","https://openalex.org/W2799215407","https://openalex.org/W2898024267","https://openalex.org/W2948672349","https://openalex.org/W2963649796","https://openalex.org/W2963936013","https://openalex.org/W2991044292","https://openalex.org/W3035268124","https://openalex.org/W3040708944","https://openalex.org/W3093314701","https://openalex.org/W3173859428","https://openalex.org/W3181556077","https://openalex.org/W3198299822","https://openalex.org/W3201890160","https://openalex.org/W3205021045","https://openalex.org/W3205869909","https://openalex.org/W3207759943","https://openalex.org/W3207849145","https://openalex.org/W3216939881","https://openalex.org/W4226104391","https://openalex.org/W4241886172","https://openalex.org/W4304091789","https://openalex.org/W4304092470","https://openalex.org/W4304098330","https://openalex.org/W4312424618","https://openalex.org/W4312559104","https://openalex.org/W4312563197","https://openalex.org/W4312563428","https://openalex.org/W4312599396","https://openalex.org/W4312747482","https://openalex.org/W4312773012","https://openalex.org/W4312873085","https://openalex.org/W4312956471","https://openalex.org/W4377235251","https://openalex.org/W4386076571"],"related_works":["https://openalex.org/W2349784553","https://openalex.org/W3022596247","https://openalex.org/W2601444686","https://openalex.org/W4292238148","https://openalex.org/W2137671468","https://openalex.org/W4323660495","https://openalex.org/W1979153711","https://openalex.org/W2385319785","https://openalex.org/W766546768","https://openalex.org/W4387129494"],"abstract_inverted_index":{"In":[0,79],"recent":[1],"years,":[2],"open-vocabulary":[3],"(OV)":[4],"object":[5,145,171],"detection":[6,21],"has":[7],"attracted":[8],"increasing":[9],"research":[10],"attention.":[11],"Unlike":[12],"traditional":[13],"detection,":[14,196],"which":[15,72],"only":[16,49],"recognizes":[17],"fixed-category":[18],"objects,":[19],"OV":[20,45,77,96,119,170,193,199],"aims":[22],"to":[23,43,117,136,162],"detect":[24],"objects":[25,56],"in":[26,57,154],"an":[27],"open":[28],"category":[29],"set.":[30],"Previous":[31],"works":[32],"often":[33],"leverage":[34],"vision-language":[35],"(VL)":[36],"training":[37],"data":[38,62],"(e.g.,":[39],"referring":[40],"grounding":[41],"data)":[42],"recognize":[44],"objects.":[46,120],"However,":[47],"they":[48],"use":[50],"pairs":[51],"of":[52,181,189],"nouns":[53],"and":[54,114,144,176],"individual":[55],"VL":[58],"data,":[59],"while":[60,197],"these":[61],"usually":[63],"contain":[64],"much":[65],"more":[66],"information,":[67],"such":[68],"as":[69,160],"scene":[70,92,112,141,158,194,200],"graphs,":[71],"are":[73],"also":[74],"crucial":[75],"for":[76,95,169,192],"detection.":[78,97],"this":[80,206],"paper,":[81],"we":[82,122,128,148,185],"propose":[83,123],"a":[84,99,130,150],"novel":[85],"Scene-Graph-Based":[86],"Discovery":[87],"Network":[88],"(SGDN)":[89],"that":[90],"exploits":[91],"graph":[93,142,195,201],"cues":[94],"Firstly,":[98],"scene-graph-based":[100,124,131],"decoder":[101],"(SGDecoder)":[102],"including":[103],"sparse":[104],"scene-graph-guided":[105],"attention":[106],"(SSGA)":[107],"is":[108],"presented.":[109],"It":[110,156],"captures":[111],"graphs":[113,159],"leverages":[115],"them":[116],"discover":[118],"Secondly,":[121],"prediction":[125],"(SGPred),":[126],"where":[127],"build":[129],"offset":[132],"regression":[133],"(SGOR)":[134],"mechanism":[135,153],"enable":[137],"mutual":[138],"enhancement":[139],"between":[140,166],"extraction":[143],"localization.":[146],"Thirdly,":[147],"design":[149],"cross-modal":[151,167],"learning":[152],"SGPred.":[155],"takes":[157],"bridges":[161],"improve":[163],"the":[164,179,187],"consistency":[165],"embeddings":[168],"classification.":[172],"Experiments":[173],"on":[174],"COCO":[175],"LVIS":[177],"demonstrate":[178],"effectiveness":[180],"our":[182,190],"approach.":[183],"Moreover,":[184],"show":[186],"ability":[188],"model":[191],"previous":[198],"generation":[202],"methods":[203],"cannot":[204],"tackle":[205],"task.":[207]},"counts_by_year":[{"year":2025,"cited_by_count":10},{"year":2024,"cited_by_count":9}],"updated_date":"2026-02-27T16:54:17.756197","created_date":"2025-10-10T00:00:00"}