Avoiding face-touches has been one of the most common medical recommendations since the beginning of the COVID-19 pandemic. This work aims at providing people with help in contrasting this widespread, yet noxious habit. The solution we present exploits wearable devices to detect hand motions ending up into a face-touch and promptly notify the user exploiting haptic feedback. To this aim, we propose a recurrent neural network taking as input temporal sequences of accelerometer data acquired by a smartwatch worn by the user. The trained RNN (NFT RNN) achieves good generalization capabilities to data coming from different users, besides a lower false detections rate with respect to a rule-based detection algorithm. The suggested solution is ready-to-use and large-scale deployable, being portable on smartwatches, fitness bands and DIY devices.
Marullo, S., Lisini Baldi, T., Paolocci, G., D'Aurizio, N., Prattichizzo, D. (2021). No Face-Touch: Exploiting Wearable Devices and Machine Learning for Gesture Detection. In 2021 IEEE International Conference on Robotics and Automation (ICRA) (pp.4187-4193). New York : Institute of Electrical and Electronics Engineers Inc. [10.1109/ICRA48506.2021.9561178].
No Face-Touch: Exploiting Wearable Devices and Machine Learning for Gesture Detection
Marullo S.
;Lisini Baldi T.;Paolocci G.;D'Aurizio N.;Prattichizzo D.
2021-01-01
Abstract
Avoiding face-touches has been one of the most common medical recommendations since the beginning of the COVID-19 pandemic. This work aims at providing people with help in contrasting this widespread, yet noxious habit. The solution we present exploits wearable devices to detect hand motions ending up into a face-touch and promptly notify the user exploiting haptic feedback. To this aim, we propose a recurrent neural network taking as input temporal sequences of accelerometer data acquired by a smartwatch worn by the user. The trained RNN (NFT RNN) achieves good generalization capabilities to data coming from different users, besides a lower false detections rate with respect to a rule-based detection algorithm. The suggested solution is ready-to-use and large-scale deployable, being portable on smartwatches, fitness bands and DIY devices.File | Dimensione | Formato | |
---|---|---|---|
No_Face-Touch_Exploiting_Wearable_Devices_and_Machine_Learning_for_Gesture_Detection.pdf
non disponibili
Tipologia:
PDF editoriale
Licenza:
NON PUBBLICO - Accesso privato/ristretto
Dimensione
3.52 MB
Formato
Adobe PDF
|
3.52 MB | Adobe PDF | Visualizza/Apri Richiedi una copia |
I documenti in IRIS sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.
https://hdl.handle.net/11365/1213936