{"id":"https://openalex.org/W4403792164","doi":"https://doi.org/10.1145/3664647.3681618","title":"EAGLE: Egocentric AGgregated Language-video Engine","display_name":"EAGLE: Egocentric AGgregated Language-video Engine","publication_year":2024,"publication_date":"2024-10-26","ids":{"openalex":"https://openalex.org/W4403792164","doi":"https://doi.org/10.1145/3664647.3681618"},"language":"en","primary_location":{"id":"doi:10.1145/3664647.3681618","is_oa":true,"landing_page_url":"https://doi.org/10.1145/3664647.3681618","pdf_url":null,"source":null,"license":"cc-by","license_id":"https://openalex.org/licenses/cc-by","version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"Proceedings of the 32nd ACM International Conference on Multimedia","raw_type":"proceedings-article"},"type":"preprint","indexed_in":["arxiv","crossref"],"open_access":{"is_oa":true,"oa_status":"gold","oa_url":"https://doi.org/10.1145/3664647.3681618","any_repository_has_fulltext":true},"authorships":[{"author_position":"first","author":{"id":null,"display_name":"Jing Bi","orcid":"https://orcid.org/0009-0006-8235-2158"},"institutions":[{"id":"https://openalex.org/I5388228","display_name":"University of Rochester","ror":"https://ror.org/022kthw22","country_code":"US","type":"education","lineage":["https://openalex.org/I5388228"]}],"countries":["US"],"is_corresponding":true,"raw_author_name":"Jing Bi","raw_affiliation_strings":["University of Rochester, Rochester, NY, USA"],"raw_orcid":"https://orcid.org/0009-0006-8235-2158","affiliations":[{"raw_affiliation_string":"University of Rochester, Rochester, NY, USA","institution_ids":["https://openalex.org/I5388228"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5100669165","display_name":"Yunlong Tang","orcid":"https://orcid.org/0000-0003-2796-1787"},"institutions":[{"id":"https://openalex.org/I5388228","display_name":"University of Rochester","ror":"https://ror.org/022kthw22","country_code":"US","type":"education","lineage":["https://openalex.org/I5388228"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Yunlong Tang","raw_affiliation_strings":["University of Rochester, Rochester, NY, USA"],"raw_orcid":"https://orcid.org/0000-0003-2796-1787","affiliations":[{"raw_affiliation_string":"University of Rochester, Rochester, NY, USA","institution_ids":["https://openalex.org/I5388228"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5024413169","display_name":"Luchuan Song","orcid":"https://orcid.org/0000-0002-0126-1259"},"institutions":[{"id":"https://openalex.org/I5388228","display_name":"University of Rochester","ror":"https://ror.org/022kthw22","country_code":"US","type":"education","lineage":["https://openalex.org/I5388228"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Luchuan Song","raw_affiliation_strings":["University of Rochester, Rochester, NY, USA"],"raw_orcid":"https://orcid.org/0000-0002-0126-1259","affiliations":[{"raw_affiliation_string":"University of Rochester, Rochester, NY, USA","institution_ids":["https://openalex.org/I5388228"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5058314486","display_name":"Ali Vosoughi","orcid":"https://orcid.org/0000-0003-1014-2937"},"institutions":[{"id":"https://openalex.org/I5388228","display_name":"University of Rochester","ror":"https://ror.org/022kthw22","country_code":"US","type":"education","lineage":["https://openalex.org/I5388228"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Ali Vosoughi","raw_affiliation_strings":["University of Rochester, Rochester, NY, USA"],"raw_orcid":"https://orcid.org/0000-0003-1014-2937","affiliations":[{"raw_affiliation_string":"University of Rochester, Rochester, NY, USA","institution_ids":["https://openalex.org/I5388228"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5107446421","display_name":"Nguyen Nguyen","orcid":"https://orcid.org/0000-0003-1553-3224"},"institutions":[{"id":"https://openalex.org/I5388228","display_name":"University of Rochester","ror":"https://ror.org/022kthw22","country_code":"US","type":"education","lineage":["https://openalex.org/I5388228"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Nguyen Nguyen","raw_affiliation_strings":["University of Rochester, Rochester, NY, USA"],"raw_orcid":"https://orcid.org/0000-0003-1553-3224","affiliations":[{"raw_affiliation_string":"University of Rochester, Rochester, NY, USA","institution_ids":["https://openalex.org/I5388228"]}]},{"author_position":"last","author":{"id":"https://openalex.org/A5064805926","display_name":"Chenliang Xu","orcid":"https://orcid.org/0000-0002-2183-822X"},"institutions":[{"id":"https://openalex.org/I5388228","display_name":"University of Rochester","ror":"https://ror.org/022kthw22","country_code":"US","type":"education","lineage":["https://openalex.org/I5388228"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Chenliang Xu","raw_affiliation_strings":["University of Rochester, Rochester, NY, USA"],"raw_orcid":"https://orcid.org/0000-0002-2183-822X","affiliations":[{"raw_affiliation_string":"University of Rochester, Rochester, NY, USA","institution_ids":["https://openalex.org/I5388228"]}]}],"institutions":[],"countries_distinct_count":1,"institutions_distinct_count":6,"corresponding_author_ids":[],"corresponding_institution_ids":["https://openalex.org/I5388228"],"apc_list":null,"apc_paid":null,"fwci":0.4762,"has_fulltext":false,"cited_by_count":2,"citation_normalized_percentile":{"value":0.6517334,"is_in_top_1_percent":false,"is_in_top_10_percent":false},"cited_by_percentile_year":{"min":95,"max":96},"biblio":{"volume":null,"issue":null,"first_page":"1682","last_page":"1691"},"is_retracted":false,"is_paratext":false,"is_xpac":false,"primary_topic":{"id":"https://openalex.org/T11714","display_name":"Multimodal Machine Learning Applications","score":0.9998000264167786,"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":0.9998000264167786,"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/T10812","display_name":"Human Pose and Action Recognition","score":0.9997000098228455,"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":0.9972000122070312,"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":0.6928915977478027},{"id":"https://openalex.org/keywords/eagle","display_name":"Eagle","score":0.6494576930999756},{"id":"https://openalex.org/keywords/computer-graphics","display_name":"Computer graphics (images)","score":0.34183672070503235},{"id":"https://openalex.org/keywords/geology","display_name":"Geology","score":0.0934324860572815}],"concepts":[{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.6928915977478027},{"id":"https://openalex.org/C2780985876","wikidata":"https://www.wikidata.org/wiki/Q2092297","display_name":"Eagle","level":2,"score":0.6494576930999756},{"id":"https://openalex.org/C121684516","wikidata":"https://www.wikidata.org/wiki/Q7600677","display_name":"Computer graphics (images)","level":1,"score":0.34183672070503235},{"id":"https://openalex.org/C127313418","wikidata":"https://www.wikidata.org/wiki/Q1069","display_name":"Geology","level":0,"score":0.0934324860572815},{"id":"https://openalex.org/C151730666","wikidata":"https://www.wikidata.org/wiki/Q7205","display_name":"Paleontology","level":1,"score":0.0}],"mesh":[],"locations_count":2,"locations":[{"id":"doi:10.1145/3664647.3681618","is_oa":true,"landing_page_url":"https://doi.org/10.1145/3664647.3681618","pdf_url":null,"source":null,"license":"cc-by","license_id":"https://openalex.org/licenses/cc-by","version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"Proceedings of the 32nd ACM International Conference on Multimedia","raw_type":"proceedings-article"},{"id":"pmh:oai:arXiv.org:2409.17523","is_oa":true,"landing_page_url":"http://arxiv.org/abs/2409.17523","pdf_url":"https://arxiv.org/pdf/2409.17523","source":{"id":"https://openalex.org/S4306400194","display_name":"arXiv (Cornell University)","issn_l":null,"issn":null,"is_oa":true,"is_in_doaj":false,"is_core":false,"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","license_id":"https://openalex.org/licenses/cc-by","version":"submittedVersion","is_accepted":false,"is_published":false,"raw_source_name":null,"raw_type":"text"}],"best_oa_location":{"id":"doi:10.1145/3664647.3681618","is_oa":true,"landing_page_url":"https://doi.org/10.1145/3664647.3681618","pdf_url":null,"source":null,"license":"cc-by","license_id":"https://openalex.org/licenses/cc-by","version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"Proceedings of the 32nd ACM International Conference on Multimedia","raw_type":"proceedings-article"},"sustainable_development_goals":[{"id":"https://metadata.un.org/sdg/10","display_name":"Reduced inequalities","score":0.5400000214576721}],"awards":[],"funders":[],"has_content":{"grobid_xml":false,"pdf":false},"content_urls":null,"referenced_works_count":25,"referenced_works":["https://openalex.org/W1927052826","https://openalex.org/W1947050545","https://openalex.org/W2165605600","https://openalex.org/W2558333741","https://openalex.org/W2777985721","https://openalex.org/W2886641317","https://openalex.org/W2914911817","https://openalex.org/W2930957955","https://openalex.org/W2981852735","https://openalex.org/W2994857106","https://openalex.org/W2995264981","https://openalex.org/W3043142510","https://openalex.org/W3126721948","https://openalex.org/W3203256294","https://openalex.org/W3204588463","https://openalex.org/W4200367987","https://openalex.org/W4234552385","https://openalex.org/W4288089799","https://openalex.org/W4312422429","https://openalex.org/W4313071313","https://openalex.org/W4389519587","https://openalex.org/W4390873373","https://openalex.org/W4390874670","https://openalex.org/W4399070666","https://openalex.org/W4402754238"],"related_works":["https://openalex.org/W4391375266","https://openalex.org/W2899084033","https://openalex.org/W2748952813","https://openalex.org/W647060948","https://openalex.org/W2380223122","https://openalex.org/W1491901162","https://openalex.org/W2371861705","https://openalex.org/W2374010249","https://openalex.org/W4390918132","https://openalex.org/W4229005970"],"abstract_inverted_index":{"The":[0],"rapid":[1],"evolution":[2],"of":[3,48,97,132,141],"egocentric":[4,74,86,144],"video":[5,49,75,110,145,166],"analysis":[6],"brings":[7],"new":[8],"insights":[9],"into":[10],"understanding":[11,76,163],"human":[12],"activities":[13],"and":[14,32,40,62,123,179],"intentions":[15],"from":[16,99],"a":[17,45,68,94,108,130,138],"first-person":[18],"perspective.":[19],"Despite":[20],"this":[21],"progress,":[22],"the":[23,55,63,79,174],"fragmentation":[24],"in":[25,182],"tasks":[26,98],"like":[27],"action":[28],"recognition,":[29],"procedure":[30,103],"learning,":[31],"moment":[33],"retrieval,":[34],"\\etc,":[35],"coupled":[36],"with":[37,164],"inconsistent":[38],"annotations":[39],"isolated":[41],"model":[42,61,114],"development,":[43],"hinders":[44],"holistic":[46,165],"interpretation":[47],"content.":[50],"In":[51,126],"response,":[52],"we":[53,128,170],"introduce":[54],"EAGLE":[56],"(Egocentric":[57],"AGgregated":[58],"Language-video":[59],"Engine)":[60],"EAGLE-400K":[64],"dataset":[65,83],"to":[66,92,102,118,136,160,172],"provide":[67],"unified":[69],"framework":[70],"that":[71],"integrates":[72],"various":[73],"tasks.":[77],"EAGLE-400K,":[78],"\\textit{first}":[80],"large-scale":[81],"instruction-tuning":[82],"tailored":[84],"for":[85,143,176],"video,":[87],"features":[88],"400K":[89],"diverse":[90],"samples":[91],"enhance":[93],"broad":[95],"spectrum":[96],"activity":[100],"recognition":[101],"knowledge":[104],"learning.":[105],"Moreover,":[106],"EAGLE,":[107,169],"strong":[109],"multimodal":[111],"large":[112],"language":[113],"(MLLM),":[115],"is":[116],"designed":[117,135],"effectively":[119],"capture":[120],"both":[121],"spatial":[122],"temporal":[124],"information.":[125],"addition,":[127],"propose":[129],"set":[131],"evaluation":[133],"metrics":[134],"facilitate":[137],"thorough":[139],"assessment":[140],"MLLM":[142],"understanding.":[146],"Our":[147],"extensive":[148],"experiments":[149],"demonstrate":[150],"EAGLE's":[151],"superior":[152],"performance":[153],"over":[154],"existing":[155],"models,":[156],"highlighting":[157],"its":[158],"ability":[159],"balance":[161],"task-specific":[162],"interpretation.":[167],"With":[168],"aim":[171],"pave":[173],"way":[175],"research":[177],"opportunities":[178],"practical":[180],"applications":[181],"real-world":[183],"scenarios.":[184]},"counts_by_year":[{"year":2025,"cited_by_count":2}],"updated_date":"2026-03-27T05:58:40.876381","created_date":"2025-10-10T00:00:00"}