Cargando…

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

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-co...

Descripción completa

Detalles Bibliográficos
Autores principales: Guerra, Bruna Maria Vittoria, Ramat, Stefano, Beltrami, Giorgio, Schmid, Micaela
Formato: Online Artículo Texto
Lenguaje:English
Publicado: MDPI 2023
Materias:
Acceso en línea:https://www.ncbi.nlm.nih.gov/pmc/articles/PMC10255961/
https://www.ncbi.nlm.nih.gov/pubmed/37299986
http://dx.doi.org/10.3390/s23115260
_version_ 1785056999133675520
author Guerra, Bruna Maria Vittoria
Ramat, Stefano
Beltrami, Giorgio
Schmid, Micaela
author_facet Guerra, Bruna Maria Vittoria
Ramat, Stefano
Beltrami, Giorgio
Schmid, Micaela
author_sort Guerra, Bruna Maria Vittoria
collection PubMed
description 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.
format Online
Article
Text
id pubmed-10255961
institution National Center for Biotechnology Information
language English
publishDate 2023
publisher MDPI
record_format MEDLINE/PubMed
spelling pubmed-102559612023-06-10 Recurrent Network Solutions for Human Posture Recognition Based on Kinect Skeletal Data Guerra, Bruna Maria Vittoria Ramat, Stefano Beltrami, Giorgio Schmid, Micaela Sensors (Basel) Article 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. MDPI 2023-06-01 /pmc/articles/PMC10255961/ /pubmed/37299986 http://dx.doi.org/10.3390/s23115260 Text en © 2023 by the authors. https://creativecommons.org/licenses/by/4.0/Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (https://creativecommons.org/licenses/by/4.0/).
spellingShingle Article
Guerra, Bruna Maria Vittoria
Ramat, Stefano
Beltrami, Giorgio
Schmid, Micaela
Recurrent Network Solutions for Human Posture Recognition Based on Kinect Skeletal Data
title Recurrent Network Solutions for Human Posture Recognition Based on Kinect Skeletal Data
title_full Recurrent Network Solutions for Human Posture Recognition Based on Kinect Skeletal Data
title_fullStr Recurrent Network Solutions for Human Posture Recognition Based on Kinect Skeletal Data
title_full_unstemmed Recurrent Network Solutions for Human Posture Recognition Based on Kinect Skeletal Data
title_short Recurrent Network Solutions for Human Posture Recognition Based on Kinect Skeletal Data
title_sort recurrent network solutions for human posture recognition based on kinect skeletal data
topic Article
url https://www.ncbi.nlm.nih.gov/pmc/articles/PMC10255961/
https://www.ncbi.nlm.nih.gov/pubmed/37299986
http://dx.doi.org/10.3390/s23115260
work_keys_str_mv AT guerrabrunamariavittoria recurrentnetworksolutionsforhumanposturerecognitionbasedonkinectskeletaldata
AT ramatstefano recurrentnetworksolutionsforhumanposturerecognitionbasedonkinectskeletaldata
AT beltramigiorgio recurrentnetworksolutionsforhumanposturerecognitionbasedonkinectskeletaldata
AT schmidmicaela recurrentnetworksolutionsforhumanposturerecognitionbasedonkinectskeletaldata