Several characteristics of the human body turn into postural behavior, recognizable also during sport activities. The presence of differences between body types could lead to different behavior of wearable and fitness-devote products. A new wearable based on machine learning techniques for the exercise detection and repetitions count is described in this work. A proper dataset has been obtained in order to offline train the network. Eventually, the machine learning algorithm has been implemented inside an edge device for real-time test e verification.

A Novel Fitness Tracker Using Edge Machine Learning

Merenda M.
;
Della Corte F
2020-01-01

Abstract

Several characteristics of the human body turn into postural behavior, recognizable also during sport activities. The presence of differences between body types could lead to different behavior of wearable and fitness-devote products. A new wearable based on machine learning techniques for the exercise detection and repetitions count is described in this work. A proper dataset has been obtained in order to offline train the network. Eventually, the machine learning algorithm has been implemented inside an edge device for real-time test e verification.
2020
978-1-7281-5200-4
Artificial neural networks, Machine learning, Random access memory, Sensors, Accelerometers, Hardware, Testing
File in questo prodotto:
File Dimensione Formato  
Merenda_2020_MELECON_Novel_Post.pdf

accesso aperto

Tipologia: Documento in Post-print
Licenza: Tutti i diritti riservati (All rights reserved)
Dimensione 548.74 kB
Formato Adobe PDF
548.74 kB Adobe PDF Visualizza/Apri

I documenti in IRIS sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.

Utilizza questo identificativo per citare o creare un link a questo documento: https://hdl.handle.net/20.500.12318/66421
Citazioni
  • ???jsp.display-item.citation.pmc??? ND
  • Scopus 11
  • ???jsp.display-item.citation.isi??? ND
social impact