{"id":"https://openalex.org/W7138985973","doi":"https://doi.org/10.48550/arxiv.2603.17069","title":"Edge-Efficient Two-Stream Multimodal Architecture for Non-Intrusive Bathroom Fall Detection","display_name":"Edge-Efficient Two-Stream Multimodal Architecture for Non-Intrusive Bathroom Fall Detection","publication_year":2026,"publication_date":"2026-03-17","ids":{"openalex":"https://openalex.org/W7138985973","doi":"https://doi.org/10.48550/arxiv.2603.17069"},"language":"en","primary_location":{"id":"pmh:oai:pure.atira.dk:publications/a9944180-0771-4009-827d-b3a57219425e","is_oa":true,"landing_page_url":"https://admin.research-repository.uwa.edu.au/en/publications/a9944180-0771-4009-827d-b3a57219425e","pdf_url":"https://api.research-repository.uwa.edu.au/ws/files/608162075/2603.17069v2.pdf","source":{"id":"https://openalex.org/S4306402523","display_name":"UWA Profiles and Research Repository (University of Western Australia)","issn_l":null,"issn":null,"is_oa":false,"is_in_doaj":false,"is_core":false,"host_organization":"https://openalex.org/I177877127","host_organization_name":"The University of Western Australia","host_organization_lineage":["https://openalex.org/I177877127"],"host_organization_lineage_names":[],"type":"repository"},"license":null,"license_id":null,"version":"submittedVersion","is_accepted":false,"is_published":false,"raw_source_name":"Wang, H, Wang, Y, Wang, X, Fung, S & Mansoor, A 2026 'Edge-Efficient Two-Stream Multimodal Architecture for Non-Intrusive Bathroom Fall Detection' arXiv.","raw_type":"info:eu-repo/semantics/preprint"},"type":"preprint","indexed_in":["datacite"],"open_access":{"is_oa":true,"oa_status":"green","oa_url":"https://api.research-repository.uwa.edu.au/ws/files/608162075/2603.17069v2.pdf","any_repository_has_fulltext":true},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5057583744","display_name":"Haitian Wang","orcid":"https://orcid.org/0009-0000-7544-4667"},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Wang, Haitian","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5121825573","display_name":"Yiren Wang","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Wang, Yiren","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5130174142","display_name":"Xinyu Wang","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Wang, Xinyu","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5108316589","display_name":"Sheldon Fung","orcid":"https://orcid.org/0000-0001-9125-7634"},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Fung, Sheldon","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"last","author":{"id":"https://openalex.org/A5121780712","display_name":"Atif Mansoor","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Mansoor, Atif","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]}],"institutions":[],"countries_distinct_count":0,"institutions_distinct_count":0,"corresponding_author_ids":[],"corresponding_institution_ids":[],"apc_list":null,"apc_paid":null,"fwci":null,"has_fulltext":true,"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/T10444","display_name":"Context-Aware Activity Recognition Systems","score":0.31790000200271606,"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/T10444","display_name":"Context-Aware Activity Recognition Systems","score":0.31790000200271606,"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/T12740","display_name":"Gait Recognition and Analysis","score":0.11429999768733978,"subfield":{"id":"https://openalex.org/subfields/2204","display_name":"Biomedical 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/T11196","display_name":"Non-Invasive Vital Sign Monitoring","score":0.08290000259876251,"subfield":{"id":"https://openalex.org/subfields/2204","display_name":"Biomedical 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/inference","display_name":"Inference","score":0.42809998989105225},{"id":"https://openalex.org/keywords/latency","display_name":"Latency (audio)","score":0.4251999855041504},{"id":"https://openalex.org/keywords/radar","display_name":"Radar","score":0.4174000024795532},{"id":"https://openalex.org/keywords/guard","display_name":"Guard (computer science)","score":0.3059000074863434},{"id":"https://openalex.org/keywords/feature","display_name":"Feature (linguistics)","score":0.2897999882698059},{"id":"https://openalex.org/keywords/benchmark","display_name":"Benchmark (surveying)","score":0.28929999470710754},{"id":"https://openalex.org/keywords/object-detection","display_name":"Object detection","score":0.2831999957561493}],"concepts":[{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.6639000177383423},{"id":"https://openalex.org/C79403827","wikidata":"https://www.wikidata.org/wiki/Q3988","display_name":"Real-time computing","level":1,"score":0.5376999974250793},{"id":"https://openalex.org/C2776214188","wikidata":"https://www.wikidata.org/wiki/Q408386","display_name":"Inference","level":2,"score":0.42809998989105225},{"id":"https://openalex.org/C82876162","wikidata":"https://www.wikidata.org/wiki/Q17096504","display_name":"Latency (audio)","level":2,"score":0.4251999855041504},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.42480000853538513},{"id":"https://openalex.org/C554190296","wikidata":"https://www.wikidata.org/wiki/Q47528","display_name":"Radar","level":2,"score":0.4174000024795532},{"id":"https://openalex.org/C44154836","wikidata":"https://www.wikidata.org/wiki/Q45045","display_name":"Simulation","level":1,"score":0.4171000123023987},{"id":"https://openalex.org/C31972630","wikidata":"https://www.wikidata.org/wiki/Q844240","display_name":"Computer vision","level":1,"score":0.31630000472068787},{"id":"https://openalex.org/C141141315","wikidata":"https://www.wikidata.org/wiki/Q2379942","display_name":"Guard (computer science)","level":2,"score":0.3059000074863434},{"id":"https://openalex.org/C2776401178","wikidata":"https://www.wikidata.org/wiki/Q12050496","display_name":"Feature (linguistics)","level":2,"score":0.2897999882698059},{"id":"https://openalex.org/C185798385","wikidata":"https://www.wikidata.org/wiki/Q1161707","display_name":"Benchmark (surveying)","level":2,"score":0.28929999470710754},{"id":"https://openalex.org/C2776151529","wikidata":"https://www.wikidata.org/wiki/Q3045304","display_name":"Object detection","level":3,"score":0.2831999957561493},{"id":"https://openalex.org/C162307627","wikidata":"https://www.wikidata.org/wiki/Q204833","display_name":"Enhanced Data Rates for GSM Evolution","level":2,"score":0.27730000019073486},{"id":"https://openalex.org/C104114177","wikidata":"https://www.wikidata.org/wiki/Q79782","display_name":"Motion (physics)","level":2,"score":0.26750001311302185},{"id":"https://openalex.org/C186370098","wikidata":"https://www.wikidata.org/wiki/Q442787","display_name":"Energy (signal processing)","level":2,"score":0.2621000111103058},{"id":"https://openalex.org/C33954974","wikidata":"https://www.wikidata.org/wiki/Q486494","display_name":"Sensor fusion","level":2,"score":0.2619999945163727},{"id":"https://openalex.org/C66746571","wikidata":"https://www.wikidata.org/wiki/Q1134833","display_name":"ENCODE","level":3,"score":0.26170000433921814},{"id":"https://openalex.org/C121704057","wikidata":"https://www.wikidata.org/wiki/Q352070","display_name":"Collision","level":2,"score":0.25999999046325684},{"id":"https://openalex.org/C2780624872","wikidata":"https://www.wikidata.org/wiki/Q852453","display_name":"Motion detection","level":3,"score":0.2565000057220459}],"mesh":[],"locations_count":2,"locations":[{"id":"pmh:oai:pure.atira.dk:publications/a9944180-0771-4009-827d-b3a57219425e","is_oa":true,"landing_page_url":"https://admin.research-repository.uwa.edu.au/en/publications/a9944180-0771-4009-827d-b3a57219425e","pdf_url":"https://api.research-repository.uwa.edu.au/ws/files/608162075/2603.17069v2.pdf","source":{"id":"https://openalex.org/S4306402523","display_name":"UWA Profiles and Research Repository (University of Western Australia)","issn_l":null,"issn":null,"is_oa":false,"is_in_doaj":false,"is_core":false,"host_organization":"https://openalex.org/I177877127","host_organization_name":"The University of Western Australia","host_organization_lineage":["https://openalex.org/I177877127"],"host_organization_lineage_names":[],"type":"repository"},"license":null,"license_id":null,"version":"submittedVersion","is_accepted":false,"is_published":false,"raw_source_name":"Wang, H, Wang, Y, Wang, X, Fung, S & Mansoor, A 2026 'Edge-Efficient Two-Stream Multimodal Architecture for Non-Intrusive Bathroom Fall Detection' arXiv.","raw_type":"info:eu-repo/semantics/preprint"},{"id":"doi:10.48550/arxiv.2603.17069","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2603.17069","pdf_url":null,"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":null,"is_accepted":false,"is_published":null,"raw_source_name":null,"raw_type":"Preprint"}],"best_oa_location":{"id":"pmh:oai:pure.atira.dk:publications/a9944180-0771-4009-827d-b3a57219425e","is_oa":true,"landing_page_url":"https://admin.research-repository.uwa.edu.au/en/publications/a9944180-0771-4009-827d-b3a57219425e","pdf_url":"https://api.research-repository.uwa.edu.au/ws/files/608162075/2603.17069v2.pdf","source":{"id":"https://openalex.org/S4306402523","display_name":"UWA Profiles and Research Repository (University of Western Australia)","issn_l":null,"issn":null,"is_oa":false,"is_in_doaj":false,"is_core":false,"host_organization":"https://openalex.org/I177877127","host_organization_name":"The University of Western Australia","host_organization_lineage":["https://openalex.org/I177877127"],"host_organization_lineage_names":[],"type":"repository"},"license":null,"license_id":null,"version":"submittedVersion","is_accepted":false,"is_published":false,"raw_source_name":"Wang, H, Wang, Y, Wang, X, Fung, S & Mansoor, A 2026 'Edge-Efficient Two-Stream Multimodal Architecture for Non-Intrusive Bathroom Fall Detection' arXiv.","raw_type":"info:eu-repo/semantics/preprint"},"sustainable_development_goals":[{"id":"https://metadata.un.org/sdg/11","display_name":"Sustainable cities and communities","score":0.5254178643226624}],"awards":[],"funders":[],"has_content":{"pdf":true,"grobid_xml":true},"content_urls":{"pdf":"https://content.openalex.org/works/W7138985973.pdf","grobid_xml":"https://content.openalex.org/works/W7138985973.grobid-xml"},"referenced_works_count":0,"referenced_works":[],"related_works":[],"abstract_inverted_index":{"Falls":[0],"in":[1],"wet":[2],"bathroom":[3,166],"environments":[4],"are":[5],"a":[6,98,106,135,158,165,213],"major":[7],"safety":[8],"risk":[9],"for":[10,109,154],"seniors":[11],"living":[12],"alone.":[13],"Recent":[14],"work":[15],"has":[16],"shown":[17],"that":[18,101,121],"mmWave-only,":[19],"vibration-only,":[20],"and":[21,33,49,60,63,74,84,86,113,125,134,141,144,182,197,218,235,251],"existing":[22],"multimodal":[23],"schemes,":[24],"such":[25],"as":[26,51],"vibration-triggered":[27],"radar":[28,103,181],"activation,":[29],"early":[30],"feature":[31],"concatenation,":[32],"decision-level":[34],"score":[35],"fusion,":[36],"can":[37],"support":[38],"privacy-preserving,":[39],"non-intrusive":[40],"fall":[41,167,236],"detection.":[42],"However,":[43],"these":[44],"designs":[45],"still":[46],"treat":[47],"motion":[48,111,140],"impact":[50,76,123,142],"loosely":[52],"coupled":[53],"streams,":[54],"depending":[55],"on":[56,89,157,264],"coarse":[57],"temporal":[58],"alignment":[59],"amplitude":[61],"thresholds,":[62],"do":[64],"not":[65],"explicitly":[66],"encode":[67],"the":[68,201,225,265],"causal":[69],"link":[70],"between":[71],"radar-observed":[72],"collapse":[73],"floor":[75,115],"or":[77],"address":[78],"timing":[79],"drift,":[80],"object":[81],"drop":[82],"confounders,":[83],"latency":[85,244],"energy":[87,253],"constraints":[88],"low-power":[90],"edge":[91],"devices.":[92],"To":[93],"this":[94],"end,":[95],"we":[96],"propose":[97],"two-stream":[99],"architecture":[100],"encodes":[102],"signals":[104],"with":[105,117,171,193,224],"Motion--Mamba":[107],"branch":[108,120],"long-range":[110],"patterns":[112],"processes":[114],"vibration":[116,184],"an":[118,219],"Impact--Griffin":[119],"emphasizes":[122],"transients":[124],"cross-axis":[126],"coupling.":[127],"Cross-conditioned":[128],"fusion":[129],"uses":[130],"low-rank":[131],"bilinear":[132],"interaction":[133],"Switch--MoE":[136],"head":[137],"to":[138,248,261],"align":[139],"tokens":[143],"suppress":[145],"object-drop":[146],"confounders.":[147],"The":[148],"model":[149,205],"keeps":[150],"inference":[151],"cost":[152],"suitable":[153],"real-time":[155],"execution":[156],"Raspberry":[159,266],"Pi":[160,267],"4B":[161,268],"gateway.":[162,269],"We":[163],"construct":[164],"detection":[168],"benchmark":[169],"dataset":[170],"frame-level":[172],"annotations,":[173],"comprising":[174],"more":[175],"than":[176],"3~h":[177],"of":[178,221],"synchronized":[179],"mmWave":[180],"triaxial":[183],"recordings":[185],"across":[186],"eight":[187],"scenarios":[188],"under":[189],"running":[190],"water,":[191],"together":[192],"subject-independent":[194],"training,":[195],"validation,":[196],"test":[198,202],"splits.":[199],"On":[200],"split,":[203],"our":[204],"attains":[206],"96.1%":[207],"accuracy,":[208],"94.8%":[209],"precision,":[210],"88.0%":[211],"recall,":[212],"91.1%":[214],"macro":[215],"F1":[216],"score,":[217],"AUC":[220],"0.968.":[222],"Compared":[223],"strongest":[226],"baseline,":[227],"it":[228],"improves":[229],"accuracy":[230],"by":[231,238],"2.0":[232],"percentage":[233,240],"points":[234],"recall":[237],"1.3":[239],"points,":[241],"while":[242],"reducing":[243],"from":[245,258],"35.9":[246],"ms":[247,250],"15.8":[249],"lowering":[252],"per":[254],"2.56":[255],"s":[256],"window":[257],"14200":[259],"mJ":[260,263],"10750":[262]},"counts_by_year":[],"updated_date":"2026-07-01T06:00:48.157686","created_date":"2026-03-20T00:00:00"}