{"id":"https://openalex.org/W3193423183","doi":"https://doi.org/10.13016/1ask-0snf","title":"LEARNING TO DETECT, PREDICT AND SYNTHESIZE OBJECTS","display_name":"LEARNING TO DETECT, PREDICT AND SYNTHESIZE OBJECTS","publication_year":2020,"publication_date":"2020-01-01","ids":{"openalex":"https://openalex.org/W3193423183","doi":"https://doi.org/10.13016/1ask-0snf","mag":"3193423183"},"language":"en","primary_location":{"id":"pmh:oai:drum.lib.umd.edu:1903/27225","is_oa":false,"landing_page_url":"http://hdl.handle.net/1903/27225","pdf_url":null,"source":{"id":"https://openalex.org/S4306401518","display_name":"University Libraries (University of Maryland)","issn_l":null,"issn":null,"is_oa":false,"is_in_doaj":false,"is_core":false,"host_organization":"https://openalex.org/I66946132","host_organization_name":"University of Maryland, College Park","host_organization_lineage":["https://openalex.org/I66946132"],"host_organization_lineage_names":[],"type":"repository"},"license":null,"license_id":null,"version":"submittedVersion","is_accepted":false,"is_published":false,"raw_source_name":null,"raw_type":"Dissertation"},"type":"dissertation","indexed_in":["datacite"],"open_access":{"is_oa":true,"oa_status":"green","oa_url":"https://doi.org/10.13016/1ask-0snf","any_repository_has_fulltext":true},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5057903960","display_name":"Navaneeth Bodla","orcid":null},"institutions":[],"countries":[],"is_corresponding":true,"raw_author_name":"Bodla, Navaneeth","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]}],"institutions":[],"countries_distinct_count":0,"institutions_distinct_count":0,"corresponding_author_ids":["https://openalex.org/A5057903960"],"corresponding_institution_ids":[],"apc_list":null,"apc_paid":null,"fwci":null,"has_fulltext":false,"cited_by_count":0,"citation_normalized_percentile":null,"cited_by_percentile_year":null,"biblio":{"volume":null,"issue":null,"first_page":null,"last_page":null},"is_retracted":false,"is_paratext":false,"is_xpac":false,"primary_topic":{"id":"https://openalex.org/T11122","display_name":"Online Learning and Analytics","score":0.05689999833703041,"subfield":{"id":"https://openalex.org/subfields/1706","display_name":"Computer Science Applications"},"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/T11122","display_name":"Online Learning and Analytics","score":0.05689999833703041,"subfield":{"id":"https://openalex.org/subfields/1706","display_name":"Computer Science Applications"},"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/T11902","display_name":"Intelligent Tutoring Systems and Adaptive Learning","score":0.05000000074505806,"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"}}],"keywords":[{"id":"https://openalex.org/keywords/computer-science","display_name":"Computer science","score":0.4429393410682678},{"id":"https://openalex.org/keywords/artificial-intelligence","display_name":"Artificial intelligence","score":0.4082260727882385},{"id":"https://openalex.org/keywords/data-science","display_name":"Data science","score":0.35294675827026367},{"id":"https://openalex.org/keywords/machine-learning","display_name":"Machine learning","score":0.32231616973876953}],"concepts":[{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.4429393410682678},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.4082260727882385},{"id":"https://openalex.org/C2522767166","wikidata":"https://www.wikidata.org/wiki/Q2374463","display_name":"Data science","level":1,"score":0.35294675827026367},{"id":"https://openalex.org/C119857082","wikidata":"https://www.wikidata.org/wiki/Q2539","display_name":"Machine learning","level":1,"score":0.32231616973876953}],"mesh":[],"locations_count":3,"locations":[{"id":"pmh:oai:drum.lib.umd.edu:1903/27225","is_oa":false,"landing_page_url":"http://hdl.handle.net/1903/27225","pdf_url":null,"source":{"id":"https://openalex.org/S4306401518","display_name":"University Libraries (University of Maryland)","issn_l":null,"issn":null,"is_oa":false,"is_in_doaj":false,"is_core":false,"host_organization":"https://openalex.org/I66946132","host_organization_name":"University of Maryland, College Park","host_organization_lineage":["https://openalex.org/I66946132"],"host_organization_lineage_names":[],"type":"repository"},"license":null,"license_id":null,"version":"submittedVersion","is_accepted":false,"is_published":false,"raw_source_name":null,"raw_type":"Dissertation"},{"id":"mag:3193423183","is_oa":false,"landing_page_url":"https://drum.lib.umd.edu/handle/1903/27225","pdf_url":null,"source":null,"license":null,"license_id":null,"version":null,"is_accepted":false,"is_published":null,"raw_source_name":null,"raw_type":null},{"id":"doi:10.13016/1ask-0snf","is_oa":true,"landing_page_url":"https://doi.org/10.13016/1ask-0snf","pdf_url":null,"source":{"id":"https://openalex.org/S4306402644","display_name":"Digital Repository at the University of Maryland (University of Maryland College Park)","issn_l":null,"issn":null,"is_oa":false,"is_in_doaj":false,"is_core":false,"host_organization":"https://openalex.org/I66946132","host_organization_name":"University of Maryland, College Park","host_organization_lineage":["https://openalex.org/I66946132"],"host_organization_lineage_names":[],"type":"repository"},"license":null,"license_id":null,"version":null,"is_accepted":false,"is_published":null,"raw_source_name":null,"raw_type":"Collection"}],"best_oa_location":{"id":"doi:10.13016/1ask-0snf","is_oa":true,"landing_page_url":"https://doi.org/10.13016/1ask-0snf","pdf_url":null,"source":{"id":"https://openalex.org/S4306402644","display_name":"Digital Repository at the University of Maryland (University of Maryland College Park)","issn_l":null,"issn":null,"is_oa":false,"is_in_doaj":false,"is_core":false,"host_organization":"https://openalex.org/I66946132","host_organization_name":"University of Maryland, College Park","host_organization_lineage":["https://openalex.org/I66946132"],"host_organization_lineage_names":[],"type":"repository"},"license":null,"license_id":null,"version":null,"is_accepted":false,"is_published":false,"raw_source_name":null,"raw_type":"Collection"},"sustainable_development_goals":[{"display_name":"Reduced inequalities","score":0.4399999976158142,"id":"https://metadata.un.org/sdg/10"}],"awards":[],"funders":[],"has_content":{"pdf":false,"grobid_xml":false},"content_urls":null,"referenced_works_count":0,"referenced_works":[],"related_works":["https://openalex.org/W2925476370","https://openalex.org/W2891885665"],"abstract_inverted_index":{"The":[0,21,72,620,640],"performance":[1,295],"of":[2,52,64,108,177,276,282,307,312,346,362,413,484,565,600,608,693],"object":[3,146,184,223,277,569,572,615],"detection":[4,101,117,129,171,278],"has":[5,405],"steadily":[6],"improved":[7,15],"over":[8,538,574],"the":[9,61,76,99,106,113,116,121,149,195,239,245,249,253,263,274,280,294,305,320,331,344,347,358,373,392,427,461,468,478,482,485,525,563,591,609,627,650,657],"past":[10],"decade,":[11],"primarily":[12],"due":[13,81,298],"to":[14,37,60,68,75,82,87,97,155,244,256,292,299,330,440,500,596,655,696],"CNN":[16],"architectures":[17],"and":[18,34,103,126,210,212,217,241,285,339,354,381,398,425,460,496,502,540,571,614,631,636,670,684],"well-annotated":[19],"datasets.":[20],"state-of-the-art":[22,222],"detectors":[23,224],"learn":[24,595],"category":[25],"independent":[26],"region":[27,63],"proposals":[28,56,246,258],"using":[29,586,633],"a":[30,132,136,156,174,287,310,326,363,400,406,410,472,532,543,554,581,587,598,605,645,689],"convolutional":[31],"neural":[32,638],"network":[33],"refine":[35],"them":[36],"improve":[38],"objects\u2019":[39,170],"localization.":[40],"There":[41],"are":[42,141,324,357],"several":[43],"open-ended":[44],"problems":[45],"with":[46,120,131,139,180,314,409,422,445,507],"these":[47,257],"approaches.":[48],"We":[49,383,448,480,686],"address":[50],"two":[51,453,469,529,667],"them:":[53],"1)":[54],"multiple":[55,232],"often":[57],"get":[58],"regressed":[59],"same":[62],"interest":[65],"(RoI),":[66],"leading":[67],"cluttered":[69,100],"detections.":[70],"2)":[71],"foreground/background":[73],"assignment":[74],"proposal":[77],"could":[78,259],"be":[79,260,388],"incorrect":[80,88,261],"missing":[83,283,300],"annotations":[84,228,284],"that":[85,166,319,385,404,653],"leads":[86,154],"signals":[89],"during":[90],"backpropagation.":[91],"Firstly,":[92],"non-maximum":[93],"suppression":[94],"is":[95,124,185,265,513,558,604,623],"used":[96,247],"mitigate":[98,159],"problem":[102],"significantly":[104],"reduce":[105],"number":[107],"false":[109],"positives.":[110],"As":[111],"per":[112],"algorithm\u2019s":[114],"design,":[115],"box":[118],"M":[119],"maximum":[122],"score":[123],"selected,":[125],"all":[127,168],"other":[128,169,462],"boxes":[130],"significant":[133,673],"overlap":[134,151,179,313],"(using":[135],"pre-defined":[137],"threshold)":[138],"$M$":[140],"suppressed.":[142],"Therefore,":[143],"if":[144],"an":[145,164,511],"lies":[147],"within":[148],"predefined":[150],"threshold,":[152],"it":[153],"miss.":[157],"To":[158,269,576],"this,":[160],"we":[161,272,579,594],"propose":[162,286,399,580],"Soft-NMS,":[163],"algorithm":[165],"decays":[167],"scores":[172],"as":[173,309,494,536,644,681],"continuous":[175],"function":[176,311],"their":[178],"$M$.":[181],"Hence,":[182],"no":[183],"eliminated":[186],"in":[187,236,248,438,487,675],"this":[188,270,450,577],"process.":[189],"Soft-NMS":[190],"obtains":[191],"consistent":[192],"improvements":[193,674],"for":[194,207,215,219,229,456,463,547,553,617,649],"coco-style":[196],"mAP":[197],"metric":[198],"on":[199,335,371,518,666],"standard":[200,676],"datasets":[201,342],"like":[202],"PASCAL":[203,337],"VOC":[204,338],"2007":[205],"($1.7\\%$":[206],"both":[208,612],"R-FCN":[209,216],"Faster-RCNN)":[211],"MS-COCO":[213],"($1.3\\%$":[214],"$1.1\\%$":[218],"Faster-RCNN).":[220],"Secondly,":[221],"require":[225,419],"detailed":[226,690],"bounding-box":[227],"images":[230,434],"containing":[231,611],"objects.":[233],"Annotations":[234],"help":[235],"accurately":[237],"assigning":[238],"foreground":[240],"background":[242,322],"labels":[243,254],"classification":[250],"stage.":[251],"However,":[252],"assigned":[255],"when":[262],"dataset":[264],"not":[266],"fully":[267],"annotated.":[268],"end,":[271,578],"investigate":[273],"robustness":[275],"under":[279],"presence":[281],"method":[288,401],"called":[289,402],"Soft":[290,302,349],"Sampling":[291,303,350],"bridge":[293],"gap,":[296],"caused":[297],"annotation.":[301],"re-weights":[304],"gradients":[306],"RoIs":[308],"positive":[315],"instances.":[316],"This":[317],"ensures":[318],"uncertain":[321],"regions":[323],"given":[325],"smaller":[327],"weight":[328],"compared":[329],"hard-negatives.":[332],"Extensive":[333],"experiments":[334],"curated":[336],"OpenImages":[340],"V3":[341],"demonstrate":[343,481,697],"effectiveness":[345],"proposed":[348],"method.":[351],"Fidelity,":[352],"diversity,":[353],"controllable":[355,374],"sampling":[356,375],"main":[359],"quality":[360],"measures":[361],"good":[364],"image":[365,458,465,490],"generation":[366,393,491,527],"model.":[367],"Our":[368,515,662],"work":[369,516],"focuses":[370],"improving":[372],"while":[376,509],"having":[377],"very":[378,560],"high":[379,446],"fidelity":[380],"diversity.":[382],"argue":[384],"controllability":[386],"can":[387,429],"achieved":[389],"by":[390,451,625],"disentangling":[391,477],"process":[394],"into":[395,528,659],"various":[396,698],"stages":[397],"FusedGAN":[403,428,486],"single-stage":[407],"pipeline":[408],"built-in":[411],"stacking":[412],"GANs.":[414],"Unlike":[415],"existing":[416],"methods,":[417],"which":[418,523],"full":[420],"supervision":[421],"paired":[423],"conditions":[424,437],"images,":[426],"effectively":[430],"leverage":[431],"more":[432,442],"abundant":[433],"without":[435],"corresponding":[436],"training,":[439],"produce":[441],"diverse":[443],"samples":[444],"fidelity.":[447],"achieve":[449],"fusing":[452],"generators:":[454],"one":[455],"unconditional":[457],"generation,":[459,466],"conditional":[464],"where":[467],"partly":[470],"share":[471],"common":[473],"latent":[474],"space":[475],"thereby":[476],"generation.":[479,498,549],"efficacy":[483],"fine":[488],"grained":[489],"tasks":[492],"such":[493,535,680],"text-to-image,":[495],"attribute-to-face":[497],"Learning":[499],"model":[501,546,585,695],"predict":[503,597],"how":[504],"humans":[505],"interact":[506],"objects":[508,632],"performing":[510],"action":[512,551],"challenging.":[514],"builds":[517],"hierarchical":[519],"video":[520,526,610,677],"prediction":[521],"models,":[522],"disentangle":[524],"stages:":[530],"predicting":[531],"high-level":[533,606],"representation,":[534],"pose,":[537,566],"time,":[539],"then":[541],"learning":[542],"pose-to-pixels":[544],"translation":[545],"pixel":[548,660],"An":[550],"sequence":[552,599,622,642],"human-object":[555],"interaction":[556],"task":[557],"typically":[559],"complicated,":[561],"involving":[562],"evolution":[564],"person's":[567],"appearance,":[568],"locations,":[570],"appearances":[573],"time.":[575],"Hierarchical":[582],"Video":[583],"Prediction":[584],"Relational":[588],"Layout.":[589],"In":[590],"first":[592],"stage,":[593],"layouts.":[601],"A":[602],"layout":[603,621,641],"representation":[607],"pose":[613,630],"information":[616],"every":[618],"frame.":[619],"learned":[624],"modeling":[626],"relationships":[628],"between":[629],"relational":[634],"reasoning":[635],"recurrent":[637],"networks.":[639],"acts":[643],"strong":[646],"structure":[647],"prior":[648],"second":[651],"stage":[652],"learns":[654],"map":[656],"layouts":[658],"space.":[661],"method\u2019s":[663],"experimental":[664],"evaluation":[665,678],"datasets,":[668],"UMD-HOI":[669],"Bimanual,":[671],"shows":[672],"metrics":[679],"LPIPS,":[682],"PSNR,":[683],"SSIM.":[685],"also":[687],"perform":[688],"qualitative":[691],"analysis":[692],"our":[694],"generalizations.":[699]},"counts_by_year":[],"updated_date":"2026-07-14T23:27:15.235271","created_date":"2025-10-10T00:00:00"}