Cargando…

Time Coherent Full-Body Poses Estimated Using Only Five Inertial Sensors: Deep versus Shallow Learning

Full-body motion capture typically requires sensors/markers to be placed on each rigid body segment, which results in long setup times and is obtrusive. The number of sensors/markers can be reduced using deep learning or offline methods. However, this requires large training datasets and/or sufficie...

Descripción completa

Detalles Bibliográficos
Autores principales: Wouda, Frank J., Giuberti, Matteo, Rudigkeit, Nina, van Beijnum, Bert-Jan F., Poel, Mannes, Veltink, Peter H.
Formato: Online Artículo Texto
Lenguaje:English
Publicado: MDPI 2019
Materias:
Acceso en línea:https://www.ncbi.nlm.nih.gov/pmc/articles/PMC6749312/
https://www.ncbi.nlm.nih.gov/pubmed/31461958
http://dx.doi.org/10.3390/s19173716
_version_ 1783452248809930752
author Wouda, Frank J.
Giuberti, Matteo
Rudigkeit, Nina
van Beijnum, Bert-Jan F.
Poel, Mannes
Veltink, Peter H.
author_facet Wouda, Frank J.
Giuberti, Matteo
Rudigkeit, Nina
van Beijnum, Bert-Jan F.
Poel, Mannes
Veltink, Peter H.
author_sort Wouda, Frank J.
collection PubMed
description Full-body motion capture typically requires sensors/markers to be placed on each rigid body segment, which results in long setup times and is obtrusive. The number of sensors/markers can be reduced using deep learning or offline methods. However, this requires large training datasets and/or sufficient computational resources. Therefore, we investigate the following research question: “What is the performance of a shallow approach, compared to a deep learning one, for estimating time coherent full-body poses using only five inertial sensors?”. We propose to incorporate past/future inertial sensor information into a stacked input vector, which is fed to a shallow neural network for estimating full-body poses. Shallow and deep learning approaches are compared using the same input vector configurations. Additionally, the inclusion of acceleration input is evaluated. The results show that a shallow learning approach can estimate full-body poses with a similar accuracy (~6 cm) to that of a deep learning approach (~7 cm). However, the jerk errors are smaller using the deep learning approach, which can be the effect of explicit recurrent modelling. Furthermore, it is shown that the delay using a shallow learning approach (72 ms) is smaller than that of a deep learning approach (117 ms).
format Online
Article
Text
id pubmed-6749312
institution National Center for Biotechnology Information
language English
publishDate 2019
publisher MDPI
record_format MEDLINE/PubMed
spelling pubmed-67493122019-09-27 Time Coherent Full-Body Poses Estimated Using Only Five Inertial Sensors: Deep versus Shallow Learning Wouda, Frank J. Giuberti, Matteo Rudigkeit, Nina van Beijnum, Bert-Jan F. Poel, Mannes Veltink, Peter H. Sensors (Basel) Article Full-body motion capture typically requires sensors/markers to be placed on each rigid body segment, which results in long setup times and is obtrusive. The number of sensors/markers can be reduced using deep learning or offline methods. However, this requires large training datasets and/or sufficient computational resources. Therefore, we investigate the following research question: “What is the performance of a shallow approach, compared to a deep learning one, for estimating time coherent full-body poses using only five inertial sensors?”. We propose to incorporate past/future inertial sensor information into a stacked input vector, which is fed to a shallow neural network for estimating full-body poses. Shallow and deep learning approaches are compared using the same input vector configurations. Additionally, the inclusion of acceleration input is evaluated. The results show that a shallow learning approach can estimate full-body poses with a similar accuracy (~6 cm) to that of a deep learning approach (~7 cm). However, the jerk errors are smaller using the deep learning approach, which can be the effect of explicit recurrent modelling. Furthermore, it is shown that the delay using a shallow learning approach (72 ms) is smaller than that of a deep learning approach (117 ms). MDPI 2019-08-27 /pmc/articles/PMC6749312/ /pubmed/31461958 http://dx.doi.org/10.3390/s19173716 Text en © 2019 by the authors. 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 (http://creativecommons.org/licenses/by/4.0/).
spellingShingle Article
Wouda, Frank J.
Giuberti, Matteo
Rudigkeit, Nina
van Beijnum, Bert-Jan F.
Poel, Mannes
Veltink, Peter H.
Time Coherent Full-Body Poses Estimated Using Only Five Inertial Sensors: Deep versus Shallow Learning
title Time Coherent Full-Body Poses Estimated Using Only Five Inertial Sensors: Deep versus Shallow Learning
title_full Time Coherent Full-Body Poses Estimated Using Only Five Inertial Sensors: Deep versus Shallow Learning
title_fullStr Time Coherent Full-Body Poses Estimated Using Only Five Inertial Sensors: Deep versus Shallow Learning
title_full_unstemmed Time Coherent Full-Body Poses Estimated Using Only Five Inertial Sensors: Deep versus Shallow Learning
title_short Time Coherent Full-Body Poses Estimated Using Only Five Inertial Sensors: Deep versus Shallow Learning
title_sort time coherent full-body poses estimated using only five inertial sensors: deep versus shallow learning
topic Article
url https://www.ncbi.nlm.nih.gov/pmc/articles/PMC6749312/
https://www.ncbi.nlm.nih.gov/pubmed/31461958
http://dx.doi.org/10.3390/s19173716
work_keys_str_mv AT woudafrankj timecoherentfullbodyposesestimatedusingonlyfiveinertialsensorsdeepversusshallowlearning
AT giubertimatteo timecoherentfullbodyposesestimatedusingonlyfiveinertialsensorsdeepversusshallowlearning
AT rudigkeitnina timecoherentfullbodyposesestimatedusingonlyfiveinertialsensorsdeepversusshallowlearning
AT vanbeijnumbertjanf timecoherentfullbodyposesestimatedusingonlyfiveinertialsensorsdeepversusshallowlearning
AT poelmannes timecoherentfullbodyposesestimatedusingonlyfiveinertialsensorsdeepversusshallowlearning
AT veltinkpeterh timecoherentfullbodyposesestimatedusingonlyfiveinertialsensorsdeepversusshallowlearning