Dance Pose Identification from Motion Capture Data: A Comparison of Classifiers

Dance Pose Identification from Motion Capture Data: A Comparison of Classifiers

In this paper, we scrutinize the effectiveness of classification techniques in recognizing dance types based on motion-captured human skeleton data. In particular, the goal is to identify poses which are characteristic for each dance performed, based on information on body joints, acquired by a Kine...

Full description

Bibliographic Details
Main Authors: Eftychios Protopapadakis, Athanasios Voulodimos, Anastasios Doulamis, Stephanos Camarinopoulos, Nikolaos Doulamis, Georgios Miaoulis
Format: Article
Language:English
Published: MDPI AG 2018-03-01
Series:Technologies
Subjects:
pose identification
Kinect sensor
motion capture device
classification
dance analysis
Online Access:http://www.mdpi.com/2227-7080/6/1/31
id doaj-8676e412e30d4b10ab288a4933bc25f6
record_format Article
spelling doaj-8676e412e30d4b10ab288a4933bc25f62020-11-25T00:23:40ZengMDPI AGTechnologies2227-70802018-03-01613110.3390/technologies6010031technologies6010031Dance Pose Identification from Motion Capture Data: A Comparison of ClassifiersEftychios Protopapadakis0Athanasios Voulodimos1Anastasios Doulamis2Stephanos Camarinopoulos3Nikolaos Doulamis4Georgios Miaoulis5School of Rural and Surveying Engineering, National Technical University of Athens, 15780 Zografou, GreeceDepartment of Informatics, Technological Educational Institute of Athens, 12243 Egaleo, GreeceSchool of Rural and Surveying Engineering, National Technical University of Athens, 15780 Zografou, GreeceRISA Sicherheitsanalysen GmbH, 10707 Berlin, GermanySchool of Rural and Surveying Engineering, National Technical University of Athens, 15780 Zografou, GreeceDepartment of Informatics, Technological Educational Institute of Athens, 12243 Egaleo, GreeceIn this paper, we scrutinize the effectiveness of classification techniques in recognizing dance types based on motion-captured human skeleton data. In particular, the goal is to identify poses which are characteristic for each dance performed, based on information on body joints, acquired by a Kinect sensor. The datasets used include sequences from six folk dances and their variations. Multiple pose identification schemes are applied using temporal constraints, spatial information, and feature space distributions for the creation of an adequate training dataset. The obtained results are evaluated and discussed.http://www.mdpi.com/2227-7080/6/1/31pose identificationKinect sensormotion capture deviceclassificationdance analysis
collection DOAJ
language English
format Article
sources DOAJ
author Eftychios Protopapadakis
Athanasios Voulodimos
Anastasios Doulamis
Stephanos Camarinopoulos
Nikolaos Doulamis
Georgios Miaoulis
spellingShingle Eftychios Protopapadakis
Athanasios Voulodimos
Anastasios Doulamis
Stephanos Camarinopoulos
Nikolaos Doulamis
Georgios Miaoulis
Dance Pose Identification from Motion Capture Data: A Comparison of Classifiers
Technologies
pose identification
Kinect sensor
motion capture device
classification
dance analysis
author_facet Eftychios Protopapadakis
Athanasios Voulodimos
Anastasios Doulamis
Stephanos Camarinopoulos
Nikolaos Doulamis
Georgios Miaoulis
author_sort Eftychios Protopapadakis
title Dance Pose Identification from Motion Capture Data: A Comparison of Classifiers
title_short Dance Pose Identification from Motion Capture Data: A Comparison of Classifiers
title_full Dance Pose Identification from Motion Capture Data: A Comparison of Classifiers
title_fullStr Dance Pose Identification from Motion Capture Data: A Comparison of Classifiers
title_full_unstemmed Dance Pose Identification from Motion Capture Data: A Comparison of Classifiers
title_sort dance pose identification from motion capture data: a comparison of classifiers
publisher MDPI AG
series Technologies
issn 2227-7080
publishDate 2018-03-01
description In this paper, we scrutinize the effectiveness of classification techniques in recognizing dance types based on motion-captured human skeleton data. In particular, the goal is to identify poses which are characteristic for each dance performed, based on information on body joints, acquired by a Kinect sensor. The datasets used include sequences from six folk dances and their variations. Multiple pose identification schemes are applied using temporal constraints, spatial information, and feature space distributions for the creation of an adequate training dataset. The obtained results are evaluated and discussed.
topic pose identification
Kinect sensor
motion capture device
classification
dance analysis
url http://www.mdpi.com/2227-7080/6/1/31
work_keys_str_mv AT eftychiosprotopapadakis danceposeidentificationfrommotioncapturedataacomparisonofclassifiers
AT athanasiosvoulodimos danceposeidentificationfrommotioncapturedataacomparisonofclassifiers
AT anastasiosdoulamis danceposeidentificationfrommotioncapturedataacomparisonofclassifiers
AT stephanoscamarinopoulos danceposeidentificationfrommotioncapturedataacomparisonofclassifiers
AT nikolaosdoulamis danceposeidentificationfrommotioncapturedataacomparisonofclassifiers
AT georgiosmiaoulis danceposeidentificationfrommotioncapturedataacomparisonofclassifiers
_version_ 1725355630782316544

Similar Items