An Enhanced Joint Hilbert Embedding-Based Metric to Support Mocap Data Classification with Preserved Interpretability
Motion capture (Mocap) data are widely used as time series to study human movement. Indeed, animation movies, video games, and biomechanical systems for rehabilitation are significant applications related to Mocap data. However, classifying multi-channel time series from Mocap requires coding the in...
Main Authors: | , , , , |
---|---|
Format: | Article |
Language: | English |
Published: |
MDPI AG
2021-06-01
|
Series: | Sensors |
Subjects: | |
Online Access: | https://www.mdpi.com/1424-8220/21/13/4443 |
_version_ | 1797528384923762688 |
---|---|
author | Cristian Kaori Valencia-Marin Juan Diego Pulgarin-Giraldo Luisa Fernanda Velasquez-Martinez Andres Marino Alvarez-Meza German Castellanos-Dominguez |
author_facet | Cristian Kaori Valencia-Marin Juan Diego Pulgarin-Giraldo Luisa Fernanda Velasquez-Martinez Andres Marino Alvarez-Meza German Castellanos-Dominguez |
author_sort | Cristian Kaori Valencia-Marin |
collection | DOAJ |
description | Motion capture (Mocap) data are widely used as time series to study human movement. Indeed, animation movies, video games, and biomechanical systems for rehabilitation are significant applications related to Mocap data. However, classifying multi-channel time series from Mocap requires coding the intrinsic dependencies (even nonlinear relationships) between human body joints. Furthermore, the same human action may have variations because the individual alters their movement and therefore the inter/intraclass variability. Here, we introduce an enhanced Hilbert embedding-based approach from a cross-covariance operator, termed EHECCO, to map the input Mocap time series to a tensor space built from both 3D skeletal joints and a principal component analysis-based projection. Obtained results demonstrate how EHECCO represents and discriminates joint probability distributions as kernel-based evaluation of input time series within a tensor reproducing kernel Hilbert space (RKHS). Our approach achieves competitive classification results for style/subject and action recognition tasks on well-known publicly available databases. Moreover, EHECCO favors the interpretation of relevant anthropometric variables correlated with players’ expertise and acted movement on a Tennis-Mocap database (also publicly available with this work). Thereby, our EHECCO-based framework provides a unified representation (through the tensor RKHS) of the Mocap time series to compute linear correlations between a coded metric from joint distributions and player properties, i.e., age, body measurements, and sport movement (action class). |
first_indexed | 2024-03-10T09:58:30Z |
format | Article |
id | doaj.art-3b6172aa4fbd454f865ceaa7f741d120 |
institution | Directory Open Access Journal |
issn | 1424-8220 |
language | English |
last_indexed | 2024-03-10T09:58:30Z |
publishDate | 2021-06-01 |
publisher | MDPI AG |
record_format | Article |
series | Sensors |
spelling | doaj.art-3b6172aa4fbd454f865ceaa7f741d1202023-11-22T02:08:37ZengMDPI AGSensors1424-82202021-06-012113444310.3390/s21134443An Enhanced Joint Hilbert Embedding-Based Metric to Support Mocap Data Classification with Preserved InterpretabilityCristian Kaori Valencia-Marin0Juan Diego Pulgarin-Giraldo1Luisa Fernanda Velasquez-Martinez2Andres Marino Alvarez-Meza3German Castellanos-Dominguez4Faculty of Engineering, Universidad Tecnológica de Pereira, Pereira 660003, ColombiaG-Bio Research Group, Automatic and Electronic Department, Universidad Autónoma de Occidente, Cali 760030, ColombiaSignal Processing and Recognition Group, Universidad Nacional de Colombia sede Manizales, Manizales 170001, ColombiaSignal Processing and Recognition Group, Universidad Nacional de Colombia sede Manizales, Manizales 170001, ColombiaSignal Processing and Recognition Group, Universidad Nacional de Colombia sede Manizales, Manizales 170001, ColombiaMotion capture (Mocap) data are widely used as time series to study human movement. Indeed, animation movies, video games, and biomechanical systems for rehabilitation are significant applications related to Mocap data. However, classifying multi-channel time series from Mocap requires coding the intrinsic dependencies (even nonlinear relationships) between human body joints. Furthermore, the same human action may have variations because the individual alters their movement and therefore the inter/intraclass variability. Here, we introduce an enhanced Hilbert embedding-based approach from a cross-covariance operator, termed EHECCO, to map the input Mocap time series to a tensor space built from both 3D skeletal joints and a principal component analysis-based projection. Obtained results demonstrate how EHECCO represents and discriminates joint probability distributions as kernel-based evaluation of input time series within a tensor reproducing kernel Hilbert space (RKHS). Our approach achieves competitive classification results for style/subject and action recognition tasks on well-known publicly available databases. Moreover, EHECCO favors the interpretation of relevant anthropometric variables correlated with players’ expertise and acted movement on a Tennis-Mocap database (also publicly available with this work). Thereby, our EHECCO-based framework provides a unified representation (through the tensor RKHS) of the Mocap time series to compute linear correlations between a coded metric from joint distributions and player properties, i.e., age, body measurements, and sport movement (action class).https://www.mdpi.com/1424-8220/21/13/4443Hilbert embeddingjoint distributiontime seriesclassificationMocap data |
spellingShingle | Cristian Kaori Valencia-Marin Juan Diego Pulgarin-Giraldo Luisa Fernanda Velasquez-Martinez Andres Marino Alvarez-Meza German Castellanos-Dominguez An Enhanced Joint Hilbert Embedding-Based Metric to Support Mocap Data Classification with Preserved Interpretability Sensors Hilbert embedding joint distribution time series classification Mocap data |
title | An Enhanced Joint Hilbert Embedding-Based Metric to Support Mocap Data Classification with Preserved Interpretability |
title_full | An Enhanced Joint Hilbert Embedding-Based Metric to Support Mocap Data Classification with Preserved Interpretability |
title_fullStr | An Enhanced Joint Hilbert Embedding-Based Metric to Support Mocap Data Classification with Preserved Interpretability |
title_full_unstemmed | An Enhanced Joint Hilbert Embedding-Based Metric to Support Mocap Data Classification with Preserved Interpretability |
title_short | An Enhanced Joint Hilbert Embedding-Based Metric to Support Mocap Data Classification with Preserved Interpretability |
title_sort | enhanced joint hilbert embedding based metric to support mocap data classification with preserved interpretability |
topic | Hilbert embedding joint distribution time series classification Mocap data |
url | https://www.mdpi.com/1424-8220/21/13/4443 |
work_keys_str_mv | AT cristiankaorivalenciamarin anenhancedjointhilbertembeddingbasedmetrictosupportmocapdataclassificationwithpreservedinterpretability AT juandiegopulgaringiraldo anenhancedjointhilbertembeddingbasedmetrictosupportmocapdataclassificationwithpreservedinterpretability AT luisafernandavelasquezmartinez anenhancedjointhilbertembeddingbasedmetrictosupportmocapdataclassificationwithpreservedinterpretability AT andresmarinoalvarezmeza anenhancedjointhilbertembeddingbasedmetrictosupportmocapdataclassificationwithpreservedinterpretability AT germancastellanosdominguez anenhancedjointhilbertembeddingbasedmetrictosupportmocapdataclassificationwithpreservedinterpretability AT cristiankaorivalenciamarin enhancedjointhilbertembeddingbasedmetrictosupportmocapdataclassificationwithpreservedinterpretability AT juandiegopulgaringiraldo enhancedjointhilbertembeddingbasedmetrictosupportmocapdataclassificationwithpreservedinterpretability AT luisafernandavelasquezmartinez enhancedjointhilbertembeddingbasedmetrictosupportmocapdataclassificationwithpreservedinterpretability AT andresmarinoalvarezmeza enhancedjointhilbertembeddingbasedmetrictosupportmocapdataclassificationwithpreservedinterpretability AT germancastellanosdominguez enhancedjointhilbertembeddingbasedmetrictosupportmocapdataclassificationwithpreservedinterpretability |