EUVIMEDEUVIMED entdecken
Open Access · CC BY 4.0

Recurrent Network Solutions for Human Posture Recognition Based on Kinect Skeletal Data

Bruna Maria Vittoria Guerra iD, Stefano Ramat iD, Giorgio Beltrami, Micaela Schmid

Sensors · 2023

DOI: 10.3390/s23115260 ↗

Kurz gefasst

Fachartikel in der Zeitschrift Sensors, 2023. Verfasst von Bruna Maria Vittoria Guerra und 3 weiteren Autor:innen. Beteiligte Einrichtung: Laboratory of Bioengineering, Department of Electrical, Computer and Biomedical Engineering, University of Pavia, 27100 Pavia, Italy. Hauptthemen laut MEDLINE: Posture, Algorithms. Die Arbeit wurde bisher 16-mal zitiert (Crossref) und nennt 48 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0). Gefördert von: Regione Lombardia, Bando Smart Living 2016.

AbstractSchlagwörterBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Ambient Assisted Living (AAL) systems are designed to provide unobtrusive and user-friendly support in daily life and can be used for monitoring frail people based on various types of sensors, including wearables and cameras. Although cameras can be perceived as intrusive in terms of privacy, low-cost RGB-D devices (i.e., Kinect V2) that extract skeletal data can partially overcome these limits. In addition, deep learning-based algorithms, such as Recurrent Neural Networks (RNNs), can be trained on skeletal tracking data to automatically identify different human postures in the AAL domain. In this study, we investigate the performance of two RNN models (2BLSTM and 3BGRU) in identifying daily living postures and potentially dangerous situations in a home monitoring system, based on 3D skeletal data acquired with Kinect V2. We tested the RNN models with two different feature sets: one consisting of eight human-crafted kinematic features selected by a genetic algorithm, and another consisting of 52 ego-centric 3D coordinates of each considered skeleton joint, plus the subject’s distance from the Kinect V2. To improve the generalization ability of the 3BGRU model, we also applied a data augmentation method to balance the training dataset. With this last solution we reached an accuracy of 88%, the best we achieved so far.

Schlagwörter

MeSH-Schlagwörter (MEDLINE)

PostureAlgorithmsHumansNeural Networks, ComputerSkeletonMonitoring, Physiologic

Stichwörter der Autor:innen

ambient assisted livingdeep learninghuman action recognitionrecurrent neural networkskeletal data

Autor:innen und Institutionen

  • Laboratory of Bioengineering, Department of Electrical, Computer and Biomedical Engineering, University of Pavia, 27100 Pavia, Italy.

ORCID-Profile: Bruna Maria Vittoria Guerra, Stefano Ramat

Förderung

  • Regione Lombardia, Bando Smart Living 2016

Zitieren

APA

Guerra, B. M. V., Ramat, S., Beltrami, G., & Schmid, M. (2023). Recurrent Network Solutions for Human Posture Recognition Based on Kinect Skeletal Data. Sensors. https://doi.org/10.3390/s23115260

Vancouver

Guerra BMV, Ramat S, Beltrami G, Schmid M. Recurrent Network Solutions for Human Posture Recognition Based on Kinect Skeletal Data. Sensors. 2023. doi:10.3390/s23115260

BibTeX
@article{Guerra2023,
  author = {Guerra, Bruna Maria Vittoria and Ramat, Stefano and Beltrami, Giorgio and Schmid, Micaela},
  title = {Recurrent Network Solutions for Human Posture Recognition Based on Kinect Skeletal Data},
  journal = {Sensors},
  year = {2023},
  doi = {10.3390/s23115260}
}
Kostenlos starten →