Counting Activities Using Weakly Labeled Raw Acceleration Data: A Variable-Length Sequence Approach with Deep Learning to Maintain Event Duration Flexibility

Georgios Sopidis; Michael Haslgrübler-Huemer; Alois Ferscha

doi:10.3390/s23115057

Counting Activities Using Weakly Labeled Raw Acceleration Data: A Variable-Length Sequence Approach with Deep Learning to Maintain Event Duration Flexibility

Georgios Sopidis
, Michael Haslgrübler-Huemer
, Alois Ferscha

Institute of Pervasive Computing

Research output: Contribution to journal › Article › peer-review

Original language	English
Article number	5057
Number of pages	16
Journal	Sensors
Volume	23
Issue number	11
DOIs	https://doi.org/10.3390/s23115057
Publication status	Published - 25 May 2023

UN SDGs

This output contributes to the following UN Sustainable Development Goals (SDGs)

SDG 9 Industry, Innovation, and Infrastructure

Fields of science

202017 Embedded systems
102 Computer Sciences
102009 Computer simulation
102013 Human-computer interaction
102019 Machine learning
102020 Medical informatics
102021 Pervasive computing
102022 Software development
102025 Distributed systems
211902 Assistive technologies
211912 Product design

JKU Focus areas

Digital Transformation

Access to Document

10.3390/s23115057

Cite this

@article{9a86999f956a418d9f0f7ea9b74558a4,

title = "Counting Activities Using Weakly Labeled Raw Acceleration Data: A Variable-Length Sequence Approach with Deep Learning to Maintain Event Duration Flexibility",

author = "Georgios Sopidis and Michael Haslgr{\"u}bler-Huemer and Alois Ferscha",

year = "2023",

month = may,

day = "25",

doi = "10.3390/s23115057",

language = "English",

volume = "23",

journal = "Sensors",

issn = "1424-8220",

publisher = "MDPI",

number = "11",

}

TY - JOUR

T1 - Counting Activities Using Weakly Labeled Raw Acceleration Data: A Variable-Length Sequence Approach with Deep Learning to Maintain Event Duration Flexibility

AU - Sopidis, Georgios

AU - Haslgrübler-Huemer, Michael

AU - Ferscha, Alois

PY - 2023/5/25

Y1 - 2023/5/25

UR - https://www.scopus.com/pages/publications/85161621345

U2 - 10.3390/s23115057

DO - 10.3390/s23115057

M3 - Article

C2 - 37299784

SN - 1424-8220

VL - 23

JO - Sensors

JF - Sensors

IS - 11

M1 - 5057

ER -

Counting Activities Using Weakly Labeled Raw Acceleration Data: A Variable-Length Sequence Approach with Deep Learning to Maintain Event Duration Flexibility

UN SDGs

Fields of science

JKU Focus areas

Access to Document

Other files and links

Pro2Future - Products and Production Systems of the Future

Cite this

Counting Activities Using Weakly Labeled Raw Acceleration Data: A Variable-Length Sequence Approach with Deep Learning to Maintain Event Duration Flexibility

UN SDGs

Fields of science

JKU Focus areas

Access to Document

Other files and links

Projects

Pro2Future - Products and Production Systems of the Future

Cite this