{"id":"https://openalex.org/W4413925097","doi":"https://doi.org/10.1109/icra55743.2025.11128104","title":"3D Dense Captioning via Prototypical Momentum Distillation","display_name":"3D Dense Captioning via Prototypical Momentum Distillation","publication_year":2025,"publication_date":"2025-05-19","ids":{"openalex":"https://openalex.org/W4413925097","doi":"https://doi.org/10.1109/icra55743.2025.11128104"},"language":"en","primary_location":{"id":"doi:10.1109/icra55743.2025.11128104","is_oa":false,"landing_page_url":"https://doi.org/10.1109/icra55743.2025.11128104","pdf_url":null,"source":null,"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"2025 IEEE International Conference on Robotics and Automation (ICRA)","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/A5030282812","display_name":"Jinpeng Mi","orcid":"https://orcid.org/0000-0002-0506-9707"},"institutions":[{"id":"https://openalex.org/I283854653","display_name":"Machine Intelligence Research Institute","ror":"https://ror.org/01h4ass90","country_code":"US","type":"nonprofit","lineage":["https://openalex.org/I283854653"]}],"countries":["US"],"is_corresponding":true,"raw_author_name":"Jinpeng Mi","raw_affiliation_strings":["Institute of Machine Intelligence, University of Shanghai for Science and Technology,China"],"affiliations":[{"raw_affiliation_string":"Institute of Machine Intelligence, University of Shanghai for Science and Technology,China","institution_ids":["https://openalex.org/I283854653"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5100347212","display_name":"Ying Wang","orcid":"https://orcid.org/0000-0003-2121-7025"},"institutions":[{"id":"https://openalex.org/I283854653","display_name":"Machine Intelligence Research Institute","ror":"https://ror.org/01h4ass90","country_code":"US","type":"nonprofit","lineage":["https://openalex.org/I283854653"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Ying Wang","raw_affiliation_strings":["Institute of Machine Intelligence, University of Shanghai for Science and Technology,China"],"affiliations":[{"raw_affiliation_string":"Institute of Machine Intelligence, University of Shanghai for Science and Technology,China","institution_ids":["https://openalex.org/I283854653"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5083029783","display_name":"Shaofei Jin","orcid":"https://orcid.org/0000-0002-1722-4559"},"institutions":[{"id":"https://openalex.org/I283854653","display_name":"Machine Intelligence Research Institute","ror":"https://ror.org/01h4ass90","country_code":"US","type":"nonprofit","lineage":["https://openalex.org/I283854653"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Shaofei Jin","raw_affiliation_strings":["Institute of Machine Intelligence, University of Shanghai for Science and Technology,China"],"affiliations":[{"raw_affiliation_string":"Institute of Machine Intelligence, University of Shanghai for Science and Technology,China","institution_ids":["https://openalex.org/I283854653"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5101596415","display_name":"Shiming Zhang","orcid":"https://orcid.org/0000-0002-4961-819X"},"institutions":[{"id":"https://openalex.org/I283854653","display_name":"Machine Intelligence Research Institute","ror":"https://ror.org/01h4ass90","country_code":"US","type":"nonprofit","lineage":["https://openalex.org/I283854653"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Shiming Zhang","raw_affiliation_strings":["Institute of Machine Intelligence, University of Shanghai for Science and Technology,China"],"affiliations":[{"raw_affiliation_string":"Institute of Machine Intelligence, University of Shanghai for Science and Technology,China","institution_ids":["https://openalex.org/I283854653"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5030210439","display_name":"Xian Wei","orcid":"https://orcid.org/0000-0002-9566-7814"},"institutions":[{"id":"https://openalex.org/I66867065","display_name":"East China Normal University","ror":"https://ror.org/02n96ep67","country_code":"CN","type":"education","lineage":["https://openalex.org/I66867065"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Xian Wei","raw_affiliation_strings":["Software Engineering Institute, East China Normal University,China"],"affiliations":[{"raw_affiliation_string":"Software Engineering Institute, East China Normal University,China","institution_ids":["https://openalex.org/I66867065"]}]},{"author_position":"last","author":{"id":"https://openalex.org/A5100740036","display_name":"Jianwei Zhang","orcid":"https://orcid.org/0000-0002-9943-9966"},"institutions":[{"id":"https://openalex.org/I884043246","display_name":"Hamburg University of Technology","ror":"https://ror.org/04bs1pb34","country_code":"DE","type":"education","lineage":["https://openalex.org/I884043246"]}],"countries":["DE"],"is_corresponding":false,"raw_author_name":"Jianwei Zhang","raw_affiliation_strings":["Universität Hamburg,Technical Aspects of Multimodal Systems,Department of Informatics,Germany"],"affiliations":[{"raw_affiliation_string":"Universität Hamburg,Technical Aspects of Multimodal Systems,Department of Informatics,Germany","institution_ids":["https://openalex.org/I884043246"]}]}],"institutions":[],"countries_distinct_count":3,"institutions_distinct_count":6,"corresponding_author_ids":["https://openalex.org/A5030282812"],"corresponding_institution_ids":["https://openalex.org/I283854653"],"apc_list":null,"apc_paid":null,"fwci":0.0,"has_fulltext":false,"cited_by_count":0,"citation_normalized_percentile":{"value":0.21169079,"is_in_top_1_percent":false,"is_in_top_10_percent":false},"cited_by_percentile_year":null,"biblio":{"volume":null,"issue":null,"first_page":"1444","last_page":"1450"},"is_retracted":false,"is_paratext":false,"is_xpac":false,"primary_topic":{"id":"https://openalex.org/T10531","display_name":"Advanced Vision and Imaging","score":0.9983000159263611,"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":0.9983000159263611,"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/T11439","display_name":"Video Analysis and Summarization","score":0.9940000176429749,"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/T11019","display_name":"Image Enhancement Techniques","score":0.9919000267982483,"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/closed-captioning","display_name":"Closed captioning","score":0.8681693077087402},{"id":"https://openalex.org/keywords/distillation","display_name":"Distillation","score":0.7171627879142761},{"id":"https://openalex.org/keywords/computer-science","display_name":"Computer science","score":0.6313128471374512},{"id":"https://openalex.org/keywords/momentum","display_name":"Momentum (technical analysis)","score":0.5181742906570435},{"id":"https://openalex.org/keywords/artificial-intelligence","display_name":"Artificial intelligence","score":0.26329392194747925},{"id":"https://openalex.org/keywords/chemistry","display_name":"Chemistry","score":0.13794055581092834},{"id":"https://openalex.org/keywords/image","display_name":"Image (mathematics)","score":0.06960546970367432},{"id":"https://openalex.org/keywords/chromatography","display_name":"Chromatography","score":0.06378722190856934}],"concepts":[{"id":"https://openalex.org/C157657479","wikidata":"https://www.wikidata.org/wiki/Q2367247","display_name":"Closed captioning","level":3,"score":0.8681693077087402},{"id":"https://openalex.org/C204030448","wikidata":"https://www.wikidata.org/wiki/Q101017","display_name":"Distillation","level":2,"score":0.7171627879142761},{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.6313128471374512},{"id":"https://openalex.org/C60718061","wikidata":"https://www.wikidata.org/wiki/Q1414747","display_name":"Momentum (technical analysis)","level":2,"score":0.5181742906570435},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.26329392194747925},{"id":"https://openalex.org/C185592680","wikidata":"https://www.wikidata.org/wiki/Q2329","display_name":"Chemistry","level":0,"score":0.13794055581092834},{"id":"https://openalex.org/C115961682","wikidata":"https://www.wikidata.org/wiki/Q860623","display_name":"Image (mathematics)","level":2,"score":0.06960546970367432},{"id":"https://openalex.org/C43617362","wikidata":"https://www.wikidata.org/wiki/Q170050","display_name":"Chromatography","level":1,"score":0.06378722190856934},{"id":"https://openalex.org/C10138342","wikidata":"https://www.wikidata.org/wiki/Q43015","display_name":"Finance","level":1,"score":0.0},{"id":"https://openalex.org/C162324750","wikidata":"https://www.wikidata.org/wiki/Q8134","display_name":"Economics","level":0,"score":0.0}],"mesh":[],"locations_count":1,"locations":[{"id":"doi:10.1109/icra55743.2025.11128104","is_oa":false,"landing_page_url":"https://doi.org/10.1109/icra55743.2025.11128104","pdf_url":null,"source":null,"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"2025 IEEE International Conference on Robotics and Automation (ICRA)","raw_type":"proceedings-article"}],"best_oa_location":null,"sustainable_development_goals":[],"awards":[{"id":"https://openalex.org/G2660893481","display_name":null,"funder_award_id":"62106026,62272170,42130112","funder_id":"https://openalex.org/F4320321001","funder_display_name":"National Natural Science Foundation of China"}],"funders":[{"id":"https://openalex.org/F4320320879","display_name":"Deutsche Forschungsgemeinschaft","ror":"https://ror.org/018mejw64"},{"id":"https://openalex.org/F4320321001","display_name":"National Natural Science Foundation of China","ror":"https://ror.org/01h0zpd94"}],"has_content":{"grobid_xml":false,"pdf":false},"content_urls":null,"referenced_works_count":34,"referenced_works":["https://openalex.org/W1895577753","https://openalex.org/W1956340063","https://openalex.org/W2101105183","https://openalex.org/W2294370754","https://openalex.org/W2489434015","https://openalex.org/W2594519801","https://openalex.org/W2963084599","https://openalex.org/W2963758027","https://openalex.org/W2988715931","https://openalex.org/W3095974555","https://openalex.org/W3107521863","https://openalex.org/W3182910454","https://openalex.org/W3207626483","https://openalex.org/W4214624153","https://openalex.org/W4221167937","https://openalex.org/W4285606658","https://openalex.org/W4312385518","https://openalex.org/W4312650216","https://openalex.org/W4319300501","https://openalex.org/W4377707861","https://openalex.org/W4382460199","https://openalex.org/W4383108922","https://openalex.org/W4386066792","https://openalex.org/W4386076628","https://openalex.org/W4389667156","https://openalex.org/W4390872240","https://openalex.org/W4390872900","https://openalex.org/W4390873695","https://openalex.org/W4391147771","https://openalex.org/W4393150012","https://openalex.org/W4394698791","https://openalex.org/W4394773594","https://openalex.org/W4402727514","https://openalex.org/W4403792088"],"related_works":["https://openalex.org/W4391375266","https://openalex.org/W2899084033","https://openalex.org/W2748952813","https://openalex.org/W4210416330","https://openalex.org/W2775506363","https://openalex.org/W3088136942","https://openalex.org/W2963177403","https://openalex.org/W4290852288","https://openalex.org/W2949362007","https://openalex.org/W4283207562"],"abstract_inverted_index":{"3D":[0,10,49],"dense":[1,50],"captioning":[2],"aims":[3],"to":[4,55,81,85,102,138,164,170,182],"describe":[5],"the":[6,14,65,83,107,110,116,120,132,135,142,146,150,166,176,180,184,188,216],"crucial":[7],"regions":[8],"in":[9,13,68],"visual":[11],"scenes":[12],"form":[15],"of":[16,109,131,145,187],"natural":[17],"language.":[18],"Recent":[19],"prevailing":[20],"approaches":[21,214],"achieve":[22],"promising":[23],"results":[24,207],"by":[25,105,153],"leveraging":[26],"complicated":[27],"structures":[28],"incorporated":[29],"with":[30,45,95,218],"large-scale":[31],"models,":[32],"which":[33],"necessitate":[34],"abundant":[35],"parameters":[36],"and":[37,123,194,204],"pose":[38],"challenges":[39],"regarding":[40],"its":[41,171,222],"practical":[42],"applications.":[43],"Besides,":[44],"limited":[46],"training":[47,185],"data,":[48],"captioners":[51],"are":[52],"often":[53],"susceptible":[54],"overfitting,":[56],"directly":[57],"degrading":[58],"caption":[59],"generation":[60],"performance.":[61],"Drawing":[62],"inspiration":[63],"from":[64,179],"recent":[66],"advancements":[67],"knowledge":[69,104,140,168,178],"distillation,":[70],"we":[71,114,156],"propose":[72],"a":[73],"novel":[74],"approach":[75],"termed":[76],"Prototypical":[77],"Momentum":[78,92],"Distillation":[79,93],"(PMD)":[80],"prompt":[82],"model":[84,122,137],"generate":[86],"more":[87],"detailed":[88],"captions.":[89],"PMD":[90,210],"incorporates":[91],"(MD)":[94],"an":[96,125,158],"Uncertainty-aware":[97,159],"Prototype-anchored":[98,160],"Clustering":[99,161],"(UPC)":[100,162],"strategy":[101,163],"transfer":[103,175],"considering":[106],"uncertainty":[108],"teacher":[111,136,181],"knowledge.":[112],"Specifically,":[113],"employ":[115],"original":[117],"captioner":[118,133],"as":[119,134,141],"student":[121],"maintain":[124],"Exponential":[126],"Moving":[127],"Average":[128],"(EMA)":[129],"copy":[130],"impart":[139],"auxiliary":[143],"supervision":[144],"student.":[147,189],"To":[148],"abate":[149],"misleading":[151],"caused":[152],"uncertain":[154],"knowledge,":[155],"present":[157],"cluster":[165],"distilled":[167],"according":[169],"confidence.":[172],"We":[173,190],"then":[174],"rearranged":[177],"guide":[183],"route":[186],"conduct":[191],"extensive":[192],"experiments":[193],"ablation":[195],"studies":[196],"on":[197,215],"two":[198],"widely":[199],"used":[200],"benchmark":[201],"datasets,":[202],"ScanRefer":[203],"Nr3D.":[205],"Experimental":[206],"demonstrate":[208],"that":[209],"outperforms":[211],"all":[212],"state-of-the-art":[213],"benchmarks":[217],"MLE":[219],"training,":[220],"highlighting":[221],"effectiveness.":[223]},"counts_by_year":[],"updated_date":"2026-04-21T08:09:41.155169","created_date":"2025-10-10T00:00:00"}