{"id":"https://openalex.org/W7139041021","doi":"https://doi.org/10.5220/0014327700004084","title":"Yolo-Key-6D: Single Stage Monocular 6D Pose Estimation with Keypoint Enhancements","display_name":"Yolo-Key-6D: Single Stage Monocular 6D Pose Estimation with Keypoint Enhancements","publication_year":2026,"publication_date":"2026-01-01","ids":{"openalex":"https://openalex.org/W7139041021","doi":"https://doi.org/10.5220/0014327700004084"},"language":null,"primary_location":{"id":"doi:10.5220/0014327700004084","is_oa":false,"landing_page_url":"https://doi.org/10.5220/0014327700004084","pdf_url":null,"source":null,"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"Proceedings of the 21st International Conference on Computer Vision Theory and Applications","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":null,"display_name":"Kemal \u00c7etiner","orcid":null},"institutions":[{"id":"https://openalex.org/I48912391","display_name":"Istanbul Technical University","ror":"https://ror.org/059636586","country_code":"TR","type":"education","lineage":["https://openalex.org/I48912391"]},{"id":"https://openalex.org/I56303344","display_name":"Aselsan (Turkey)","ror":"https://ror.org/04knh8e66","country_code":"TR","type":"company","lineage":["https://openalex.org/I56303344"]}],"countries":["TR"],"is_corresponding":false,"raw_author_name":"Kemal \u00c7etiner","raw_affiliation_strings":["ASELSAN, Aerospace Technologies, Turkey","Istanbul Technical University, Department of Computer Engineering, Turkey"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"ASELSAN, Aerospace Technologies, Turkey","institution_ids":["https://openalex.org/I56303344"]},{"raw_affiliation_string":"Istanbul Technical University, Department of Computer Engineering, Turkey","institution_ids":["https://openalex.org/I48912391"]}]},{"author_position":"last","author":{"id":null,"display_name":"Haz\u0131m Ekenel","orcid":null},"institutions":[{"id":"https://openalex.org/I120250893","display_name":"New York University Abu Dhabi","ror":"https://ror.org/00e5k0821","country_code":"AE","type":"education","lineage":["https://openalex.org/I120250893","https://openalex.org/I57206974"]},{"id":"https://openalex.org/I48912391","display_name":"Istanbul Technical University","ror":"https://ror.org/059636586","country_code":"TR","type":"education","lineage":["https://openalex.org/I48912391"]}],"countries":["AE","TR"],"is_corresponding":false,"raw_author_name":"Haz\u0131m Ekenel","raw_affiliation_strings":["Istanbul Technical University, Department of Computer Engineering, Turkey","New York University Abu Dhabi, Division of Engineering, U.A.E"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Istanbul Technical University, Department of Computer Engineering, Turkey","institution_ids":["https://openalex.org/I48912391"]},{"raw_affiliation_string":"New York University Abu Dhabi, Division of Engineering, U.A.E","institution_ids":["https://openalex.org/I120250893"]}]}],"institutions":[],"countries_distinct_count":2,"institutions_distinct_count":3,"corresponding_author_ids":[],"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":"424","last_page":"431"},"is_retracted":false,"is_paratext":false,"is_xpac":false,"primary_topic":{"id":"https://openalex.org/T10571","display_name":"Robotic Mechanisms and Dynamics","score":0.15929999947547913,"subfield":{"id":"https://openalex.org/subfields/2207","display_name":"Control and Systems Engineering"},"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/T10571","display_name":"Robotic Mechanisms and Dynamics","score":0.15929999947547913,"subfield":{"id":"https://openalex.org/subfields/2207","display_name":"Control and Systems Engineering"},"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/T10653","display_name":"Robot Manipulation and Learning","score":0.1387999951839447,"subfield":{"id":"https://openalex.org/subfields/2207","display_name":"Control and Systems Engineering"},"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/T11687","display_name":"Teleoperation and Haptic Systems","score":0.0778999999165535,"subfield":{"id":"https://openalex.org/subfields/2210","display_name":"Mechanical 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/pose","display_name":"Pose","score":0.6349999904632568},{"id":"https://openalex.org/keywords/monocular","display_name":"Monocular","score":0.5494999885559082},{"id":"https://openalex.org/keywords/single-stage","display_name":"Single stage","score":0.43689998984336853},{"id":"https://openalex.org/keywords/3d-pose-estimation","display_name":"3D pose estimation","score":0.37310001254081726},{"id":"https://openalex.org/keywords/stage","display_name":"Stage (stratigraphy)","score":0.3637999892234802},{"id":"https://openalex.org/keywords/monocular-vision","display_name":"Monocular vision","score":0.35040000081062317}],"concepts":[{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.7713000178337097},{"id":"https://openalex.org/C31972630","wikidata":"https://www.wikidata.org/wiki/Q844240","display_name":"Computer vision","level":1,"score":0.7383000254631042},{"id":"https://openalex.org/C52102323","wikidata":"https://www.wikidata.org/wiki/Q1671968","display_name":"Pose","level":2,"score":0.6349999904632568},{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.6194000244140625},{"id":"https://openalex.org/C65909025","wikidata":"https://www.wikidata.org/wiki/Q1945033","display_name":"Monocular","level":2,"score":0.5494999885559082},{"id":"https://openalex.org/C3020376581","wikidata":"https://www.wikidata.org/wiki/Q16866784","display_name":"Single stage","level":2,"score":0.43689998984336853},{"id":"https://openalex.org/C36613465","wikidata":"https://www.wikidata.org/wiki/Q4636322","display_name":"3D pose estimation","level":3,"score":0.37310001254081726},{"id":"https://openalex.org/C146357865","wikidata":"https://www.wikidata.org/wiki/Q1123245","display_name":"Stage (stratigraphy)","level":2,"score":0.3637999892234802},{"id":"https://openalex.org/C158829959","wikidata":"https://www.wikidata.org/wiki/Q1640606","display_name":"Monocular vision","level":2,"score":0.35040000081062317},{"id":"https://openalex.org/C153180895","wikidata":"https://www.wikidata.org/wiki/Q7148389","display_name":"Pattern recognition (psychology)","level":2,"score":0.34450000524520874},{"id":"https://openalex.org/C3018868555","wikidata":"https://www.wikidata.org/wiki/Q2918907","display_name":"Single camera","level":2,"score":0.3037000000476837},{"id":"https://openalex.org/C2776401178","wikidata":"https://www.wikidata.org/wiki/Q12050496","display_name":"Feature (linguistics)","level":2,"score":0.26440000534057617},{"id":"https://openalex.org/C22100474","wikidata":"https://www.wikidata.org/wiki/Q4800952","display_name":"Articulated body pose estimation","level":4,"score":0.2581999897956848}],"mesh":[],"locations_count":2,"locations":[{"id":"doi:10.5220/0014327700004084","is_oa":false,"landing_page_url":"https://doi.org/10.5220/0014327700004084","pdf_url":null,"source":null,"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"Proceedings of the 21st International Conference on Computer Vision Theory and Applications","raw_type":"proceedings-article"},{"id":"pmh:oai:polen.itu.edu.tr:11527/75049","is_oa":false,"landing_page_url":"https://hdl.handle.net/11527/75049","pdf_url":null,"source":{"id":"https://openalex.org/S4306400460","display_name":"Istanbul Technical University Academic Open Archive (Istanbul Technical University)","issn_l":null,"issn":null,"is_oa":false,"is_in_doaj":false,"is_core":false,"host_organization":"https://openalex.org/I48912391","host_organization_name":"Istanbul Technical University","host_organization_lineage":["https://openalex.org/I48912391"],"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":"Article"}],"best_oa_location":null,"sustainable_development_goals":[],"awards":[],"funders":[],"has_content":{"grobid_xml":false,"pdf":false},"content_urls":null,"referenced_works_count":0,"referenced_works":[],"related_works":[],"abstract_inverted_index":{"Estimating":[0],"the":[1,75,91,118,134],"6D":[2,52],"pose":[3,53],"of":[4,78,94,128,163],"objects":[5],"from":[6,28],"a":[7,12,44,64,106,150,158],"single":[8,46,153],"RGB":[9],"image":[10],"is":[11],"critical":[13],"task":[14,88],"for":[15,34,50,56,167],"robotics":[16],"and":[17,59,120,130,160,165],"extended":[18],"reality":[19],"applications.":[20],"However,":[21],"state-of-the-art":[22],"multi":[23],"stage":[24,154],"methods":[25],"often":[26],"suffer":[27],"high":[29],"latency,":[30],"making":[31],"them":[32],"unsuitable":[33],"real":[35,144,168],"time":[36],"use.":[37],"In":[38],"this":[39],"paper,":[40],"we":[41,101],"present":[42],"Yolo-Key-6D,":[43],"novel":[45],"stage,":[47],"end-to-end":[48,99],"framework":[49],"monocular":[51],"estimation":[54],"designed":[55,152],"both":[57],"speed":[58],"accuracy.":[60],"Our":[61,146],"approach":[62],"enhances":[63],"YOLO":[65],"based":[66],"architecture":[67],"by":[68],"integrating":[69],"an":[70,79],"auxiliary":[71],"head":[72],"that":[73,149],"regresses":[74],"2D":[76],"projections":[77],"object's":[80],"3D":[81,95],"bounding":[82],"box":[83],"corners.":[84],"This":[85],"keypoint":[86],"detection":[87],"significantly":[89],"improves":[90],"network's":[92],"understanding":[93],"geometry.":[96],"For":[97],"stable":[98],"training,":[100],"directly":[102],"regress":[103],"rotation":[104],"using":[105],"continuous":[107],"9D":[108],"representation":[109],"projected":[110],"to":[111,141],"SO(3)":[112],"via":[113],"singular":[114],"value":[115],"decomposition.":[116],"On":[117],"LINEMOD":[119],"LINEMOD-Occluded":[121],"benchmarks,":[122],"YOLO-Key-6D":[123],"achieves":[124],"competitive":[125],"accuracy":[126],"scores":[127],"96.24%":[129],"69.41%,":[131],"respectively,":[132],"with":[133],"ADD(-S)":[135],"0.1d":[136],"metric,":[137],"while":[138],"proving":[139],"itself":[140],"operate":[142],"in":[143],"time.":[145],"results":[147],"demonstrate":[148],"carefully":[151],"method":[155],"can":[156],"provide":[157],"practical":[159],"effective":[161],"balance":[162],"performance":[164],"efficiency":[166],"world":[169],"deployment.":[170]},"counts_by_year":[],"updated_date":"2026-07-14T23:27:15.235271","created_date":"2026-03-10T00:00:00"}