{"id":"https://openalex.org/W4394593020","doi":"https://doi.org/10.1109/wacv57701.2024.00349","title":"ScanEnts3D: Exploiting Phrase-to-3D-Object Correspondences for Improved Visio-Linguistic Models in 3D Scenes","display_name":"ScanEnts3D: Exploiting Phrase-to-3D-Object Correspondences for Improved Visio-Linguistic Models in 3D Scenes","publication_year":2024,"publication_date":"2024-01-03","ids":{"openalex":"https://openalex.org/W4394593020","doi":"https://doi.org/10.1109/wacv57701.2024.00349"},"language":"en","primary_location":{"id":"doi:10.1109/wacv57701.2024.00349","doi":"https://doi.org/10.1109/wacv57701.2024.00349","is_oa":false,"landing_page_url":"https://doi.org/10.1109/wacv57701.2024.00349","pdf_url":null,"source":{"id":"https://openalex.org/S4363607736","display_name":"IEEE/CVF Winter Conference on Applications of Computer Vision (WACV)","issn_l":null,"issn":null,"is_oa":false,"is_in_doaj":false,"is_core":false,"listed_in":[],"host_organization":null,"host_organization_name":null,"host_organization_lineage":[],"host_organization_lineage_names":[],"type":"conference"},"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"2024 IEEE/CVF Winter Conference on Applications of Computer Vision (WACV)","raw_type":"proceedings-article"},"volume":null,"type":"conference-paper","indexed_in":["arxiv","crossref","datacite"],"open_access":{"is_oa":true,"oa_status":"green","oa_url":"https://arxiv.org/pdf/2212.06250","any_repository_has_fulltext":true},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5010060310","display_name":"Ahmed Abdelreheem","orcid":"https://orcid.org/0000-0003-0800-7576","observed_orcids":["https://orcid.org/0000-0003-0800-7576"]},"institutions":[{"id":"https://openalex.org/I4210142583","display_name":"Snap (United States)","ror":"https://ror.org/04dgkhg68","country_code":"US","type":"company","lineage":["https://openalex.org/I4210142583"]},{"id":"https://openalex.org/I71920554","display_name":"King Abdullah University of Science and Technology","ror":"https://ror.org/01q3tbs38","country_code":"SA","type":"education","lineage":["https://openalex.org/I71920554"]}],"countries":["SA","US"],"is_corresponding":false,"raw_author_name":"Ahmed Abdelreheem","raw_affiliation_strings":["King Abdullah University of Science and Technology (KAUST)","Snap Inc"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"King Abdullah University of Science and Technology (KAUST)","institution_ids":["https://openalex.org/I71920554"]},{"raw_affiliation_string":"Snap Inc","institution_ids":["https://openalex.org/I4210142583"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5108705955","display_name":"Kyle Olszewski","orcid":"https://orcid.org/0000-0001-8775-6879","observed_orcids":["https://orcid.org/0000-0001-8775-6879"]},"institutions":[{"id":"https://openalex.org/I4210142583","display_name":"Snap (United States)","ror":"https://ror.org/04dgkhg68","country_code":"US","type":"company","lineage":["https://openalex.org/I4210142583"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Kyle Olszewski","raw_affiliation_strings":["Snap Inc"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Snap Inc","institution_ids":["https://openalex.org/I4210142583"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5114488729","display_name":"Hsin-Ying Lee","orcid":"https://orcid.org/0009-0001-5252-1137","observed_orcids":["https://orcid.org/0009-0001-5252-1137","https://orcid.org/0000-0003-2442-0117"]},"institutions":[{"id":"https://openalex.org/I4210142583","display_name":"Snap (United States)","ror":"https://ror.org/04dgkhg68","country_code":"US","type":"company","lineage":["https://openalex.org/I4210142583"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Hsin-Ying Lee","raw_affiliation_strings":["Snap Inc"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Snap Inc","institution_ids":["https://openalex.org/I4210142583"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5076768552","display_name":"Peter Wonka","orcid":"https://orcid.org/0000-0003-0627-9746","observed_orcids":["https://orcid.org/0000-0003-0627-9746"]},"institutions":[{"id":"https://openalex.org/I71920554","display_name":"King Abdullah University of Science and Technology","ror":"https://ror.org/01q3tbs38","country_code":"SA","type":"education","lineage":["https://openalex.org/I71920554"]}],"countries":["SA"],"is_corresponding":false,"raw_author_name":"Peter Wonka","raw_affiliation_strings":["King Abdullah University of Science and Technology (KAUST)"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"King Abdullah University of Science and Technology (KAUST)","institution_ids":["https://openalex.org/I71920554"]}]},{"author_position":"last","author":{"id":"https://openalex.org/A5038007540","display_name":"Panos Achlioptas","orcid":"https://orcid.org/0000-0001-6427-6055","observed_orcids":["https://orcid.org/0000-0001-6427-6055"]},"institutions":[{"id":"https://openalex.org/I4210142583","display_name":"Snap (United States)","ror":"https://ror.org/04dgkhg68","country_code":"US","type":"company","lineage":["https://openalex.org/I4210142583"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Panos Achlioptas","raw_affiliation_strings":["Snap Inc"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Snap Inc","institution_ids":["https://openalex.org/I4210142583"]}]}],"institutions":[],"countries_distinct_count":2,"institutions_distinct_count":2,"corresponding_author_ids":[],"corresponding_institution_ids":[],"apc_list":null,"apc_paid":null,"fwci":2.2158,"has_fulltext":true,"cited_by_count":15,"citation_normalized_percentile":{"value":0.91305229,"is_in_top_1_percent":false,"is_in_top_10_percent":true},"cited_by_percentile_year":{"min":96,"max":98},"biblio":{"volume":null,"issue":null,"first_page":"3512","last_page":"3522"},"is_retracted":false,"is_paratext":false,"is_xpac":false,"primary_topic":{"id":"https://openalex.org/T11714","display_name":"Multimodal Machine Learning Applications","score":0.9984999895095825,"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":0.9984999895095825,"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/T10719","display_name":"3D Shape Modeling and Analysis","score":0.00019999999494757503,"subfield":{"id":"https://openalex.org/subfields/2206","display_name":"Computational Mechanics"},"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":0.00019999999494757503,"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"}}],"keywords":[{"id":"https://openalex.org/keywords/3d-scene-understanding","display_name":"3D scene understanding","score":0.800000011920929},{"id":"https://openalex.org/keywords/scanrefer","display_name":"ScanRefer","score":0.765999972820282},{"id":"https://openalex.org/keywords/language-generation","display_name":"language generation","score":0.7720000147819519},{"id":"https://openalex.org/keywords/data-curation","display_name":"data curation","score":0.7059999704360962},{"id":"https://openalex.org/keywords/3d-shape-retrieval","display_name":"3D model retrieval","score":0.4390000104904175}],"concepts":[{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.7095911502838135},{"id":"https://openalex.org/C2781238097","wikidata":"https://www.wikidata.org/wiki/Q175026","display_name":"Object (grammar)","level":2,"score":0.6894806027412415},{"id":"https://openalex.org/C2776224158","wikidata":"https://www.wikidata.org/wiki/Q187931","display_name":"Phrase","level":2,"score":0.6372444033622742},{"id":"https://openalex.org/C204321447","wikidata":"https://www.wikidata.org/wiki/Q30642","display_name":"Natural language processing","level":1,"score":0.5501294732093811},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.5466816425323486},{"id":"https://openalex.org/C41895202","wikidata":"https://www.wikidata.org/wiki/Q8162","display_name":"Linguistics","level":1,"score":0.40594977140426636},{"id":"https://openalex.org/C31972630","wikidata":"https://www.wikidata.org/wiki/Q844240","display_name":"Computer vision","level":1,"score":0.3290012776851654},{"id":"https://openalex.org/C138885662","wikidata":"https://www.wikidata.org/wiki/Q5891","display_name":"Philosophy","level":0,"score":0.0}],"mesh":[],"locations_count":4,"locations":[{"id":"doi:10.1109/wacv57701.2024.00349","doi":"https://doi.org/10.1109/wacv57701.2024.00349","is_oa":false,"landing_page_url":"https://doi.org/10.1109/wacv57701.2024.00349","pdf_url":null,"source":{"id":"https://openalex.org/S4363607736","display_name":"IEEE/CVF Winter Conference on Applications of Computer Vision (WACV)","issn_l":null,"issn":null,"is_oa":false,"is_in_doaj":false,"is_core":false,"listed_in":[],"host_organization":null,"host_organization_name":null,"host_organization_lineage":[],"host_organization_lineage_names":[],"type":"conference"},"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"2024 IEEE/CVF Winter Conference on Applications of Computer Vision (WACV)","raw_type":"proceedings-article"},{"id":"pmh:oai:arXiv.org:2212.06250","doi":null,"is_oa":true,"landing_page_url":"http://arxiv.org/abs/2212.06250","pdf_url":"https://arxiv.org/pdf/2212.06250","source":{"id":"https://openalex.org/S4306400194","display_name":"arXiv (Cornell University)","issn_l":"2331-8422","issn":["2331-8422"],"is_oa":true,"is_in_doaj":false,"is_core":false,"listed_in":[],"host_organization":"https://openalex.org/I205783295","host_organization_name":"Cornell University","host_organization_lineage":["https://openalex.org/I205783295"],"host_organization_lineage_names":[],"type":"repository"},"license":"cc-by-nc-sa","license_id":"https://openalex.org/licenses/cc-by-nc-sa","version":"submittedVersion","is_accepted":false,"is_published":false,"raw_source_name":null,"raw_type":"text"},{"id":"pmh:oai:repository.kaust.edu.sa:10754/686471","doi":null,"is_oa":false,"landing_page_url":"http://hdl.handle.net/10754/686471","pdf_url":null,"source":{"id":"https://openalex.org/S4306401596","display_name":"King Abdullah University of Science and Technology Repository (King Abdullah University of Science and Technology)","issn_l":null,"issn":null,"is_oa":false,"is_in_doaj":false,"is_core":false,"listed_in":[],"host_organization":"https://openalex.org/I71920554","host_organization_name":"King Abdullah University of Science and Technology","host_organization_lineage":["https://openalex.org/I71920554"],"host_organization_lineage_names":[],"type":"repository"},"license":null,"license_id":null,"version":"submittedVersion","is_accepted":false,"is_published":false,"raw_source_name":"","raw_type":"Preprint"},{"id":"doi:10.48550/arxiv.2212.06250","doi":"https://doi.org/10.48550/arxiv.2212.06250","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2212.06250","pdf_url":null,"source":{"id":"https://openalex.org/S4306400194","display_name":"arXiv (Cornell University)","issn_l":"2331-8422","issn":["2331-8422"],"is_oa":true,"is_in_doaj":false,"is_core":false,"listed_in":[],"host_organization":"https://openalex.org/I205783295","host_organization_name":"Cornell University","host_organization_lineage":["https://openalex.org/I205783295"],"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":"Preprint"}],"best_oa_location":{"id":"pmh:oai:arXiv.org:2212.06250","doi":null,"is_oa":true,"landing_page_url":"http://arxiv.org/abs/2212.06250","pdf_url":"https://arxiv.org/pdf/2212.06250","source":{"id":"https://openalex.org/S4306400194","display_name":"arXiv (Cornell University)","issn_l":"2331-8422","issn":["2331-8422"],"is_oa":true,"is_in_doaj":false,"is_core":false,"listed_in":[],"host_organization":"https://openalex.org/I205783295","host_organization_name":"Cornell University","host_organization_lineage":["https://openalex.org/I205783295"],"host_organization_lineage_names":[],"type":"repository"},"license":"cc-by-nc-sa","license_id":"https://openalex.org/licenses/cc-by-nc-sa","version":"submittedVersion","is_accepted":false,"is_published":false,"raw_source_name":null,"raw_type":"text"},"sustainable_development_goals":[],"sustainable_development_goals_aurora":[],"study_designs":[],"awards":[],"funders":[],"has_content":{"pdf":true,"grobid_xml":true},"content_urls":{"pdf":"https://content.openalex.org/works/W4394593020.pdf","grobid_xml":"https://content.openalex.org/works/W4394593020.grobid-xml"},"referenced_works_count":90,"referenced_works":["https://openalex.org/W125693051","https://openalex.org/W1689711448","https://openalex.org/W1773149199","https://openalex.org/W1895577753","https://openalex.org/W1956340063","https://openalex.org/W1985238052","https://openalex.org/W2016589492","https://openalex.org/W2101105183","https://openalex.org/W2169203263","https://openalex.org/W2251286718","https://openalex.org/W2251512949","https://openalex.org/W2460657278","https://openalex.org/W2489434015","https://openalex.org/W2563685048","https://openalex.org/W2594519801","https://openalex.org/W2772262724","https://openalex.org/W2930283066","https://openalex.org/W2944663535","https://openalex.org/W2962684798","https://openalex.org/W2963109634","https://openalex.org/W2964078384","https://openalex.org/W2964345792","https://openalex.org/W2964487155","https://openalex.org/W2969056421","https://openalex.org/W2987068636","https://openalex.org/W2987734933","https://openalex.org/W2995439012","https://openalex.org/W3034862692","https://openalex.org/W3093319304","https://openalex.org/W3095974555","https://openalex.org/W3107521863","https://openalex.org/W3110435696","https://openalex.org/W3111353235","https://openalex.org/W3133833192","https://openalex.org/W3139332675","https://openalex.org/W3140398265","https://openalex.org/W3166573884","https://openalex.org/W3178489527","https://openalex.org/W3182910454","https://openalex.org/W3184584813","https://openalex.org/W3192668118","https://openalex.org/W3195833508","https://openalex.org/W3196466825","https://openalex.org/W3201922548","https://openalex.org/W3202021185","https://openalex.org/W3203949114","https://openalex.org/W3204757729","https://openalex.org/W3206171352","https://openalex.org/W4200629315","https://openalex.org/W4212872366","https://openalex.org/W4214625308","https://openalex.org/W4221167937","https://openalex.org/W4226376247","https://openalex.org/W4284675794","https://openalex.org/W4285606658","https://openalex.org/W4287643567","https://openalex.org/W4293566037","https://openalex.org/W4304699901","https://openalex.org/W4306704712","https://openalex.org/W4310286406","https://openalex.org/W4312234819","https://openalex.org/W4312385518","https://openalex.org/W4312649925","https://openalex.org/W4312852845","https://openalex.org/W4313118252","https://openalex.org/W4365441113","https://openalex.org/W4382451063","https://openalex.org/W4385573471","https://openalex.org/W4386072322","https://openalex.org/W4386075583","https://openalex.org/W4388092550","https://openalex.org/W4389574979","https://openalex.org/W4390872240","https://openalex.org/W4394671432","https://openalex.org/W6602002561","https://openalex.org/W6678262379","https://openalex.org/W6682631176","https://openalex.org/W6687484953","https://openalex.org/W6739778489","https://openalex.org/W6743188669","https://openalex.org/W6747106673","https://openalex.org/W6795921721","https://openalex.org/W6798370836","https://openalex.org/W6798634944","https://openalex.org/W6805700901","https://openalex.org/W6839590817","https://openalex.org/W6846068340","https://openalex.org/W6846381433","https://openalex.org/W6847071135","https://openalex.org/W6857214515"],"related_works":["https://openalex.org/W2039546652","https://openalex.org/W2012262991","https://openalex.org/W2373794620","https://openalex.org/W2060629350","https://openalex.org/W2357294589","https://openalex.org/W2386861027","https://openalex.org/W2349302580","https://openalex.org/W2390154576","https://openalex.org/W2916983164","https://openalex.org/W2356613672"],"abstract_inverted_index":{"The":[0],"two":[1],"popular":[2],"datasets":[3],"ScanRefer":[4,119],"[20]":[5],"and":[6,85,96,104,118,123],"ReferIt3D":[7],"[5]":[8],"connect":[9],"natural":[10],"language":[11,105,127],"to":[12,154],"real-world":[13,78],"3D":[14,44,63],"scenes.":[15,79],"In":[16,46],"this":[17,51,91],"paper,":[18],"we":[19,110,129],"curate":[20],"a":[21,35,43,54],"complementary":[22],"dataset":[23,65,168],"extending":[24],"both":[25,101,115,142],"the":[26,98,112,116,131,138,146,172],"aforementioned":[27],"ones.":[28],"We":[29,80],"associate":[30],"all":[31],"objects":[32,71],"mentioned":[33],"in":[34,62,114],"referential":[36,74],"sentence":[37],"with":[38],"their":[39],"underlying":[40],"instances":[41],"inside":[42],"scene.":[45],"contrast,":[47],"previous":[48],"work":[49],"did":[50],"only":[52,152],"for":[53,100],"single":[55],"object":[56],"per":[57],"sentence.":[58],"Our":[59,166],"Scan":[60],"Entities":[61],"(ScanEnts3D)":[64],"provides":[66],"explicit":[67],"correspondences":[68],"between":[69],"369k":[70],"across":[72],"84k":[73],"sentences,":[75],"covering":[76],"705":[77],"propose":[81],"novel":[82],"architecture":[83],"modifications":[84],"losses":[86],"that":[87],"enable":[88],"learning":[89],"from":[90],"new":[92,147],"type":[93,148],"of":[94,143,149],"data":[95,150],"improve":[97,111,130,155],"performance":[99],"neural":[102,108],"listening":[103],"generation.":[106],"For":[107,126,141],"listening,":[109],"SoTA":[113,132],"Nr3D":[117,139],"benchmarks":[120],"by":[121,133],"4.3%":[122],"5.0%,":[124],"respectively.":[125],"generation,":[128],"13.2":[134],"CIDEr":[135],"points":[136],"on":[137,171],"benchmark.":[140],"these":[144],"tasks,":[145],"is":[151,169],"used":[153],"training,":[156],"but":[157],"no":[158],"additional":[159],"annotations":[160],"are":[161],"required":[162],"at":[163,175],"inference":[164],"time.":[165],"introduced":[167],"available":[170],"project\u2019s":[173],"webpage":[174],"https://scanents3d.github.io/.":[176]},"counts_by_year":[{"year":2026,"cited_by_count":2},{"year":2025,"cited_by_count":5},{"year":2024,"cited_by_count":3},{"year":2023,"cited_by_count":5}],"updated_date":"2026-10-12T09:40:04.860316","created_date":"2025-10-10T00:00:00"}