Data Augmented Dynamic Time Warping for Skeletal Action Classification

Ju Yong CHANG; Yong Seok HEO

doi:10.1587/transinf.2017EDP7275

Data Augmented Dynamic Time Warping for Skeletal Action Classification

Ju Yong CHANG, Yong Seok HEO

Full Text Views

0

Cite this

Summary :

We present a new action classification method for skeletal sequence data. The proposed method is based on simple nonparametric feature matching without a learning process. We first augment the training dataset to implicitly construct an exponentially increasing number of training sequences, which can be used to improve the generalization power of the proposed action classifier. These augmented training sequences are matched to the test sequence with the relaxed dynamic time warping (DTW) technique. Our relaxed formulation allows the proposed method to work faster and with higher efficiency than the conventional DTW-based method using a non-augmented dataset. Experimental results show that the proposed approach produces effective action classification results for various scales of real datasets.

Publication: IEICE TRANSACTIONS on Information Vol.E101-D No.6 pp.1562-1571

Publication Date: 2018/06/01

Publicized: 2018/03/01

Online ISSN: 1745-1361

DOI: 10.1587/transinf.2017EDP7275

Type of Manuscript: PAPER

Category: Pattern Recognition

Authors

Ju Yong CHANG
Kwangwoon University
Yong Seok HEO
Ajou University

Keyword

action classification, dynamic time warping, data augmentation

Cite this

Copy

Ju Yong CHANG, Yong Seok HEO, "Data Augmented Dynamic Time Warping for Skeletal Action Classification" in IEICE TRANSACTIONS on Information, vol. E101-D, no. 6, pp. 1562-1571, June 2018, doi: 10.1587/transinf.2017EDP7275.
Abstract: We present a new action classification method for skeletal sequence data. The proposed method is based on simple nonparametric feature matching without a learning process. We first augment the training dataset to implicitly construct an exponentially increasing number of training sequences, which can be used to improve the generalization power of the proposed action classifier. These augmented training sequences are matched to the test sequence with the relaxed dynamic time warping (DTW) technique. Our relaxed formulation allows the proposed method to work faster and with higher efficiency than the conventional DTW-based method using a non-augmented dataset. Experimental results show that the proposed approach produces effective action classification results for various scales of real datasets.
URL: https://global.ieice.org/en_transactions/information/10.1587/transinf.2017EDP7275/_p

Copy

@ARTICLE{e101-d_6_1562,
author={Ju Yong CHANG, Yong Seok HEO, },
journal={IEICE TRANSACTIONS on Information},
title={Data Augmented Dynamic Time Warping for Skeletal Action Classification},
year={2018},
volume={E101-D},
number={6},
pages={1562-1571},
abstract={We present a new action classification method for skeletal sequence data. The proposed method is based on simple nonparametric feature matching without a learning process. We first augment the training dataset to implicitly construct an exponentially increasing number of training sequences, which can be used to improve the generalization power of the proposed action classifier. These augmented training sequences are matched to the test sequence with the relaxed dynamic time warping (DTW) technique. Our relaxed formulation allows the proposed method to work faster and with higher efficiency than the conventional DTW-based method using a non-augmented dataset. Experimental results show that the proposed approach produces effective action classification results for various scales of real datasets.},
keywords={},
doi={10.1587/transinf.2017EDP7275},
ISSN={1745-1361},
month={June},}

Copy

TY - JOUR
TI - Data Augmented Dynamic Time Warping for Skeletal Action Classification
T2 - IEICE TRANSACTIONS on Information
SP - 1562
EP - 1571
AU - Ju Yong CHANG
AU - Yong Seok HEO
PY - 2018
DO - 10.1587/transinf.2017EDP7275
JO - IEICE TRANSACTIONS on Information
SN - 1745-1361
VL - E101-D
IS - 6
JA - IEICE TRANSACTIONS on Information
Y1 - June 2018
AB - We present a new action classification method for skeletal sequence data. The proposed method is based on simple nonparametric feature matching without a learning process. We first augment the training dataset to implicitly construct an exponentially increasing number of training sequences, which can be used to improve the generalization power of the proposed action classifier. These augmented training sequences are matched to the test sequence with the relaxed dynamic time warping (DTW) technique. Our relaxed formulation allows the proposed method to work faster and with higher efficiency than the conventional DTW-based method using a non-augmented dataset. Experimental results show that the proposed approach produces effective action classification results for various scales of real datasets.
ER -