Relevant Kinematic Feature Selection to Support Human Action Recognition in MoCap data

This paper presents a feature selection comparison oriented to human action recognition only with the kinematic features of skeleton representation. For this purpose, three relevance methods are used to rank the contribution of kinematic features for classifying an action is employed. Particularly,...

Full description

Autores:
Pulgarín Giraldo, Juan Diego
Alvarez Meza, Andres Marino
Ruales Tores, A. A.
Castellanos Dominguez, German
Tipo de recurso:
Part of book
Fecha de publicación:
2017
Institución:
Universidad Autónoma de Occidente
Repositorio:
RED: Repositorio Educativo Digital UAO
Idioma:
eng
OAI Identifier:
oai:red.uao.edu.co:10614/11615
Acceso en línea:
http://hdl.handle.net/10614/11615
https://link.springer.com/chapter/10.1007/978-3-319-59740-9_23
https://link.springer.com/content/pdf/10.1007%2F978-3-319-59740-9.pdf
https://doi.org/10.1007/978-3-319-59773-7_51
Palabra clave:
Center kernel alignment
Feature selection
Human motion
Kinematics
Relevance
ReliefF
Motion capture data
Principal Component Analysis
Kinematics
Human beings--Attitude and movement
Cinemática
Actitud y movimiento del hombre
Motion
Mechanical movements
Movimiento
Movimientos mecánicos
Rights
openAccess
License
Derechos Reservados - Universidad Autónoma de Occidente
id REPOUAO2_fd2646e948c3814bdcd4cb6d78806b72
oai_identifier_str oai:red.uao.edu.co:10614/11615
network_acronym_str REPOUAO2
network_name_str RED: Repositorio Educativo Digital UAO
repository_id_str
dc.title.eng.fl_str_mv Relevant Kinematic Feature Selection to Support Human Action Recognition in MoCap data
title Relevant Kinematic Feature Selection to Support Human Action Recognition in MoCap data
spellingShingle Relevant Kinematic Feature Selection to Support Human Action Recognition in MoCap data
Center kernel alignment
Feature selection
Human motion
Kinematics
Relevance
ReliefF
Motion capture data
Principal Component Analysis
Kinematics
Human beings--Attitude and movement
Cinemática
Actitud y movimiento del hombre
Motion
Mechanical movements
Movimiento
Movimientos mecánicos
title_short Relevant Kinematic Feature Selection to Support Human Action Recognition in MoCap data
title_full Relevant Kinematic Feature Selection to Support Human Action Recognition in MoCap data
title_fullStr Relevant Kinematic Feature Selection to Support Human Action Recognition in MoCap data
title_full_unstemmed Relevant Kinematic Feature Selection to Support Human Action Recognition in MoCap data
title_sort Relevant Kinematic Feature Selection to Support Human Action Recognition in MoCap data
dc.creator.fl_str_mv Pulgarín Giraldo, Juan Diego
Alvarez Meza, Andres Marino
Ruales Tores, A. A.
Castellanos Dominguez, German
dc.contributor.author.none.fl_str_mv Pulgarín Giraldo, Juan Diego
dc.contributor.author.spa.fl_str_mv Alvarez Meza, Andres Marino
Ruales Tores, A. A.
Castellanos Dominguez, German
dc.subject.eng.fl_str_mv Center kernel alignment
Feature selection
Human motion
Kinematics
Relevance
ReliefF
Motion capture data
Principal Component Analysis
topic Center kernel alignment
Feature selection
Human motion
Kinematics
Relevance
ReliefF
Motion capture data
Principal Component Analysis
Kinematics
Human beings--Attitude and movement
Cinemática
Actitud y movimiento del hombre
Motion
Mechanical movements
Movimiento
Movimientos mecánicos
dc.subject.lemb.eng.fl_str_mv Kinematics
Human beings--Attitude and movement
dc.subject.lemb.spa.fl_str_mv Cinemática
Actitud y movimiento del hombre
dc.subject.armarc.eng.fl_str_mv Motion
Mechanical movements
dc.subject.armarc.spa.fl_str_mv Movimiento
Movimientos mecánicos
description This paper presents a feature selection comparison oriented to human action recognition only with the kinematic features of skeleton representation. For this purpose, three relevance methods are used to rank the contribution of kinematic features for classifying an action is employed. Particularly, the method with the best results includes the supervised information regarding the action to find out a relevant set of features, encoding the most discriminative information. Experimental results are obtained on a well-known public data (MSR Action3D). Results are encouraging to use kernel theory methods to get better kinematic feature selection for each action with a good generalization indistinct to the subject
publishDate 2017
dc.date.issued.spa.fl_str_mv 2017-05-27
dc.date.accessioned.spa.fl_str_mv 2019-11-28T20:35:57Z
dc.date.available.spa.fl_str_mv 2019-11-28T20:35:57Z
dc.type.spa.fl_str_mv Capítulo - Parte de Libro
dc.type.coarversion.fl_str_mv http://purl.org/coar/version/c_970fb48d4fbd8a85
dc.type.coar.spa.fl_str_mv http://purl.org/coar/resource_type/c_3248
dc.type.content.spa.fl_str_mv Text
dc.type.driver.spa.fl_str_mv info:eu-repo/semantics/bookPart
dc.type.redcol.spa.fl_str_mv https://purl.org/redcol/resource_type/CAP_LIB
dc.type.version.spa.fl_str_mv info:eu-repo/semantics/publishedVersion
format http://purl.org/coar/resource_type/c_3248
status_str publishedVersion
dc.identifier.citation.eng.fl_str_mv Pulgarin-Giraldo J.D., Ruales-Torres A.A., Alvarez-Meza A.M., Castellanos-Dominguez G. (2017) Relevant Kinematic Feature Selection to Support Human Action Recognition in MoCap Data. In: Ferrández Vicente J., Álvarez-Sánchez J., de la Paz López F., Toledo Moreo J., Adeli H. (eds) Biomedical Applications Based on Natural and Artificial Computing. IWINAC 2017. Lecture Notes in Computer Science, vol 10338. Springer, Cham
dc.identifier.issn.spa.fl_str_mv 1611-3349 (en línea)
0302-9743 (impresa)
978-3-319-59773-7 (en línea)
9783319597720 (impreso)
dc.identifier.uri.spa.fl_str_mv http://hdl.handle.net/10614/11615
https://link.springer.com/chapter/10.1007/978-3-319-59740-9_23
https://link.springer.com/content/pdf/10.1007%2F978-3-319-59740-9.pdf
dc.identifier.doi.spa.fl_str_mv https://doi.org/10.1007/978-3-319-59773-7_51
identifier_str_mv Pulgarin-Giraldo J.D., Ruales-Torres A.A., Alvarez-Meza A.M., Castellanos-Dominguez G. (2017) Relevant Kinematic Feature Selection to Support Human Action Recognition in MoCap Data. In: Ferrández Vicente J., Álvarez-Sánchez J., de la Paz López F., Toledo Moreo J., Adeli H. (eds) Biomedical Applications Based on Natural and Artificial Computing. IWINAC 2017. Lecture Notes in Computer Science, vol 10338. Springer, Cham
1611-3349 (en línea)
0302-9743 (impresa)
978-3-319-59773-7 (en línea)
9783319597720 (impreso)
url http://hdl.handle.net/10614/11615
https://link.springer.com/chapter/10.1007/978-3-319-59740-9_23
https://link.springer.com/content/pdf/10.1007%2F978-3-319-59740-9.pdf
https://doi.org/10.1007/978-3-319-59773-7_51
dc.language.iso.eng.fl_str_mv eng
language eng
dc.relation.eng.fl_str_mv Biomedical Applications Based on Natural and Artificial Computing : International Work-Conference on the Interplay Between Natural and Artificial Computation, IWINAC 2017, Corunna, Spain, June 19-23, 2017, Proceedings, Part II. Páginas 501-509, (2017)
dc.relation.haspart.eng.fl_str_mv Lecture Notes in Computer Science. 10337. Theoretical Computer Science and General Issues. 10337
dc.rights.spa.fl_str_mv Derechos Reservados - Universidad Autónoma de Occidente
dc.rights.coar.fl_str_mv http://purl.org/coar/access_right/c_abf2
dc.rights.uri.spa.fl_str_mv https://creativecommons.org/licenses/by-nc-nd/4.0/
dc.rights.accessrights.spa.fl_str_mv info:eu-repo/semantics/openAccess
dc.rights.creativecommons.spa.fl_str_mv Atribución-NoComercial-SinDerivadas 4.0 Internacional (CC BY-NC-ND 4.0)
rights_invalid_str_mv Derechos Reservados - Universidad Autónoma de Occidente
https://creativecommons.org/licenses/by-nc-nd/4.0/
Atribución-NoComercial-SinDerivadas 4.0 Internacional (CC BY-NC-ND 4.0)
http://purl.org/coar/access_right/c_abf2
eu_rights_str_mv openAccess
dc.format.spa.fl_str_mv application/pdf
dc.format.extent.spa.fl_str_mv Páginas 501-509
dc.coverage.spatial.spa.fl_str_mv Universidad Autónoma de Occidente. Calle 25 115-85. Km 2 vía Cali-Jamundí
dc.publisher.eng.fl_str_mv Springer, Cham
dc.source.spa.fl_str_mv instname:Universidad Autónoma de Occidente
reponame:Repositorio Institucional UAO
instname_str Universidad Autónoma de Occidente
institution Universidad Autónoma de Occidente
reponame_str Repositorio Institucional UAO
collection Repositorio Institucional UAO
dc.source.bibliographiccitation.spa.fl_str_mv Kale, G.V., Patil, V.H.: A study of vision based human motion recognition and analysis. IJACI 7(2), 75–92 (2016)
Mandery, C., Plappert, M., Sol, J.B., Asfour, T.: Dimensionality reduction for whole-body human motion recognition. In: 19th International Conference on Information Fusion, FUSION, Heidelberg, Germany, 5–8 July 2016, pp. 355–362 (2016)
Althloothi, S., Mahoor, M.H., Zhang, X., Voyles, R.M.: Human activity recognition using multi-features and multiple kernel learning. Pattern Recogn. 47(5), 1800–1812 (2014)
Li, M., Leung, H., Liu, Z., Zhou, L.: 3D human motion retrieval using graph kernels based on adaptive graph construction. Comput. Graph. Pergamon 54, 104–112 (2016)
Wang, L., Zhao, G., Cheng, L., Pietikainen, M.: Machine Learning for Vision-Based Motion Analysis, 1st edn. Springer, Heidelberg (2011)
Wei, X.K., Chai, J.: Modeling 3D human poses from uncalibrated monocular images. In: IEEE 12th International Conference on Computer Vision, ICCV, Kyoto, Japan, 27 September–4 October 2009, pp. 1873–1880 (2009)
Li, W., Zhang, Z., Liu, Z.: Action recognition based on a bag of 3D points. In: IEEE Conference on Computer Vision and Pattern Recognition, CVPR Workshopps 2010, San Francisco, CA, USA, 13–18 June 2010, pp. 9–14 (2010)
Ravet, T., Tilmanne, J., d’Alessandro, N.: Hidden Markov model based real-time motion recognition and following. In: Proceedings of the 2014 International Workshop on Movement, Computing, MOCO 2014, pp. 82–87. ACM, New York (2014)
Jiang, Y., Saxena, A.: Modeling high-dimensional humans for activity anticipation using Gaussian process latent CRFs. In: Robotics: Science and Systems X, University of California, Berkeley, USA, 12–16 July 2014 (2014)
García-Vega, S., Álvarez-Meza, A.M., Castellanos-Domínguez, C.G.: MoCap data segmentation and classification using kernel based multi-channel analysis. In: Ruiz-Shulcloper, J., Sanniti di Baja, G. (eds.) CIARP 2013. LNCS, vol. 8259, pp. 495–502. Springer, Heidelberg (2013). doi: 10.1007/978-3-642-41827-3_62
Pulgarin-Giraldo, J.D., Alvarez-Meza, A.M., Melo-Betancourt, L.G., Ramos-Bermudez, S., Castellanos-Dominguez, G.: A similarity indicator for differentiating kinematic performance between qualified tennis players. In: Beltrán-Castañón, C., Nyström, I., Famili, F. (eds.) CIARP 2016. LNCS, vol. 10125, pp. 309–317. Springer, Cham (2017). doi: 10.1007/978-3-319-52277-7_38
Diaz-Martinez, N.F., Pulgarin-Giraldo, J.D., Vinasco-Isaza, L.E., Agredo, W.: Analysis of the alignment angles and flexion angle in women with patellofemoral pain syndrome. In: Torres, I., Bustamante, J., Sierra, D. (eds.) CLAIB 2016. IFMBE Proceedings. Springer, Singapore (2016)
Yang, X., Tian, Y.: Effective 3D action recognition using eigenjoints. J. Vis. Commun. Image Represent. 25(1), 2–11 (2014)
Yang, X., Tian, Y.: Eigenjoints-based action recognition using Naïve-Bayes-Nearest-Neighbor. In: CVPR Workshops, pp. 14–19. IEEE Computer Society (2012)
Ofli, F., Chaudhry, R., Kurillo, G., Vidal, R., Bajcsy, R.: Sequence of the most informative joints (SMIJ): a new representation for human skeletal action recognition. J. Vis. Commun. Image Represent. 25(1), 24–38 (2014)
Alvarez-Meza, A.M., Velasquez-Martinez, L.F., Castellanos-Dominguez, G.: Time-series discrimination using feature relevance analysis in motor imagery classification. Neurocomputing 151(Part 1), 122–129 (2015)
Brockmeier, A.J., Choi, J.S., Kriminger, E.G., Francis, J.T., Principe, J.C.: Neural decoding with kernel-based metric learning. Neural Comput. 26(6), 1080–1107 (2014)
Zeng, X., Wang, Q., Zhang, C., Cai, H.: Feature selection based on reliefF and PCA for underwater sound classification. In: Proceedings of 3rd International Conference on Computer Science and Network Technology, pp. 442–445 (2013)
bitstream.url.fl_str_mv https://red.uao.edu.co/bitstreams/72fbd1a1-5342-4d6a-b9ae-2198e6fbffce/download
https://red.uao.edu.co/bitstreams/24db5418-7436-4323-ad07-e0dccc87ad41/download
bitstream.checksum.fl_str_mv 4460e5956bc1d1639be9ae6146a50347
20b5ba22b1117f71589c7318baa2c560
bitstream.checksumAlgorithm.fl_str_mv MD5
MD5
repository.name.fl_str_mv Repositorio Digital Universidad Autonoma de Occidente
repository.mail.fl_str_mv repositorio@uao.edu.co
_version_ 1814259853151961088
spelling Pulgarín Giraldo, Juan Diegovirtual::4145-1jdpulgaring@unal.edu.coAlvarez Meza, Andres Marino5154abb266e67961a71f8dc28e883cbf-1Ruales Tores, A. A.3e3a8475001cc9f7d3845a5b97b052ad-1Castellanos Dominguez, German3de8cb9245317beddda7d5f7dade0b1b-1Universidad Autónoma de Occidente. Calle 25 115-85. Km 2 vía Cali-Jamundí2019-11-28T20:35:57Z2019-11-28T20:35:57Z2017-05-27Pulgarin-Giraldo J.D., Ruales-Torres A.A., Alvarez-Meza A.M., Castellanos-Dominguez G. (2017) Relevant Kinematic Feature Selection to Support Human Action Recognition in MoCap Data. In: Ferrández Vicente J., Álvarez-Sánchez J., de la Paz López F., Toledo Moreo J., Adeli H. (eds) Biomedical Applications Based on Natural and Artificial Computing. IWINAC 2017. Lecture Notes in Computer Science, vol 10338. Springer, Cham1611-3349 (en línea)0302-9743 (impresa)978-3-319-59773-7 (en línea)9783319597720 (impreso)http://hdl.handle.net/10614/11615https://link.springer.com/chapter/10.1007/978-3-319-59740-9_23https://link.springer.com/content/pdf/10.1007%2F978-3-319-59740-9.pdfhttps://doi.org/10.1007/978-3-319-59773-7_51This paper presents a feature selection comparison oriented to human action recognition only with the kinematic features of skeleton representation. For this purpose, three relevance methods are used to rank the contribution of kinematic features for classifying an action is employed. Particularly, the method with the best results includes the supervised information regarding the action to find out a relevant set of features, encoding the most discriminative information. Experimental results are obtained on a well-known public data (MSR Action3D). Results are encouraging to use kernel theory methods to get better kinematic feature selection for each action with a good generalization indistinct to the subjectapplication/pdfPáginas 501-509engSpringer, ChamBiomedical Applications Based on Natural and Artificial Computing : International Work-Conference on the Interplay Between Natural and Artificial Computation, IWINAC 2017, Corunna, Spain, June 19-23, 2017, Proceedings, Part II. Páginas 501-509, (2017)Lecture Notes in Computer Science. 10337. Theoretical Computer Science and General Issues. 10337Derechos Reservados - Universidad Autónoma de Occidentehttps://creativecommons.org/licenses/by-nc-nd/4.0/info:eu-repo/semantics/openAccessAtribución-NoComercial-SinDerivadas 4.0 Internacional (CC BY-NC-ND 4.0)http://purl.org/coar/access_right/c_abf2instname:Universidad Autónoma de Occidentereponame:Repositorio Institucional UAOKale, G.V., Patil, V.H.: A study of vision based human motion recognition and analysis. IJACI 7(2), 75–92 (2016)Mandery, C., Plappert, M., Sol, J.B., Asfour, T.: Dimensionality reduction for whole-body human motion recognition. In: 19th International Conference on Information Fusion, FUSION, Heidelberg, Germany, 5–8 July 2016, pp. 355–362 (2016)Althloothi, S., Mahoor, M.H., Zhang, X., Voyles, R.M.: Human activity recognition using multi-features and multiple kernel learning. Pattern Recogn. 47(5), 1800–1812 (2014)Li, M., Leung, H., Liu, Z., Zhou, L.: 3D human motion retrieval using graph kernels based on adaptive graph construction. Comput. Graph. Pergamon 54, 104–112 (2016)Wang, L., Zhao, G., Cheng, L., Pietikainen, M.: Machine Learning for Vision-Based Motion Analysis, 1st edn. Springer, Heidelberg (2011)Wei, X.K., Chai, J.: Modeling 3D human poses from uncalibrated monocular images. In: IEEE 12th International Conference on Computer Vision, ICCV, Kyoto, Japan, 27 September–4 October 2009, pp. 1873–1880 (2009)Li, W., Zhang, Z., Liu, Z.: Action recognition based on a bag of 3D points. In: IEEE Conference on Computer Vision and Pattern Recognition, CVPR Workshopps 2010, San Francisco, CA, USA, 13–18 June 2010, pp. 9–14 (2010)Ravet, T., Tilmanne, J., d’Alessandro, N.: Hidden Markov model based real-time motion recognition and following. In: Proceedings of the 2014 International Workshop on Movement, Computing, MOCO 2014, pp. 82–87. ACM, New York (2014)Jiang, Y., Saxena, A.: Modeling high-dimensional humans for activity anticipation using Gaussian process latent CRFs. In: Robotics: Science and Systems X, University of California, Berkeley, USA, 12–16 July 2014 (2014)García-Vega, S., Álvarez-Meza, A.M., Castellanos-Domínguez, C.G.: MoCap data segmentation and classification using kernel based multi-channel analysis. In: Ruiz-Shulcloper, J., Sanniti di Baja, G. (eds.) CIARP 2013. LNCS, vol. 8259, pp. 495–502. Springer, Heidelberg (2013). doi: 10.1007/978-3-642-41827-3_62Pulgarin-Giraldo, J.D., Alvarez-Meza, A.M., Melo-Betancourt, L.G., Ramos-Bermudez, S., Castellanos-Dominguez, G.: A similarity indicator for differentiating kinematic performance between qualified tennis players. In: Beltrán-Castañón, C., Nyström, I., Famili, F. (eds.) CIARP 2016. LNCS, vol. 10125, pp. 309–317. Springer, Cham (2017). doi: 10.1007/978-3-319-52277-7_38Diaz-Martinez, N.F., Pulgarin-Giraldo, J.D., Vinasco-Isaza, L.E., Agredo, W.: Analysis of the alignment angles and flexion angle in women with patellofemoral pain syndrome. In: Torres, I., Bustamante, J., Sierra, D. (eds.) CLAIB 2016. IFMBE Proceedings. Springer, Singapore (2016)Yang, X., Tian, Y.: Effective 3D action recognition using eigenjoints. J. Vis. Commun. Image Represent. 25(1), 2–11 (2014)Yang, X., Tian, Y.: Eigenjoints-based action recognition using Naïve-Bayes-Nearest-Neighbor. In: CVPR Workshops, pp. 14–19. IEEE Computer Society (2012)Ofli, F., Chaudhry, R., Kurillo, G., Vidal, R., Bajcsy, R.: Sequence of the most informative joints (SMIJ): a new representation for human skeletal action recognition. J. Vis. Commun. Image Represent. 25(1), 24–38 (2014)Alvarez-Meza, A.M., Velasquez-Martinez, L.F., Castellanos-Dominguez, G.: Time-series discrimination using feature relevance analysis in motor imagery classification. Neurocomputing 151(Part 1), 122–129 (2015)Brockmeier, A.J., Choi, J.S., Kriminger, E.G., Francis, J.T., Principe, J.C.: Neural decoding with kernel-based metric learning. Neural Comput. 26(6), 1080–1107 (2014)Zeng, X., Wang, Q., Zhang, C., Cai, H.: Feature selection based on reliefF and PCA for underwater sound classification. In: Proceedings of 3rd International Conference on Computer Science and Network Technology, pp. 442–445 (2013)Center kernel alignmentFeature selectionHuman motionKinematicsRelevanceReliefFMotion capture dataPrincipal Component AnalysisKinematicsHuman beings--Attitude and movementCinemáticaActitud y movimiento del hombreMotionMechanical movementsMovimientoMovimientos mecánicosRelevant Kinematic Feature Selection to Support Human Action Recognition in MoCap dataCapítulo - Parte de Librohttp://purl.org/coar/resource_type/c_3248Textinfo:eu-repo/semantics/bookParthttps://purl.org/redcol/resource_type/CAP_LIBinfo:eu-repo/semantics/publishedVersionhttp://purl.org/coar/version/c_970fb48d4fbd8a85Publication33e9b6b4-bd6d-4b86-b500-ae237e1e9a98virtual::4145-133e9b6b4-bd6d-4b86-b500-ae237e1e9a98virtual::4145-1https://scholar.google.com.co/citations?user=Bwuc2BkAAAAJ&hl=envirtual::4145-10000-0002-6409-5104virtual::4145-1https://scienti.minciencias.gov.co/cvlac/visualizador/generarCurriculoCv.do?cod_rh=0000207497virtual::4145-1CC-LICENSElicense_rdflicense_rdfapplication/rdf+xml; charset=utf-8805https://red.uao.edu.co/bitstreams/72fbd1a1-5342-4d6a-b9ae-2198e6fbffce/download4460e5956bc1d1639be9ae6146a50347MD52LICENSElicense.txtlicense.txttext/plain; charset=utf-81665https://red.uao.edu.co/bitstreams/24db5418-7436-4323-ad07-e0dccc87ad41/download20b5ba22b1117f71589c7318baa2c560MD5310614/11615oai:red.uao.edu.co:10614/116152024-03-13 14:21:16.824https://creativecommons.org/licenses/by-nc-nd/4.0/Derechos Reservados - Universidad Autónoma de Occidentemetadata.onlyhttps://red.uao.edu.coRepositorio Digital Universidad Autonoma de Occidenterepositorio@uao.edu.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