SOM-based human action recognition using local feature descriptor CHOG3D

Yanli Ji, Atsushi Shimada, Hajime Nagahara, Rin Ichiro Taniguchi

Research output: Contribution to journalArticle

Abstract

Human action recognition is applied in a wide field, such as video surveillance, intelligent interface, and intelligent robots. However, since various action classes, complex surrounding, interaction with objects, et al., it is still a complex problem to be solved. In this paper, we propose a method combining the Self-Organizing Map(SOM) and the classifier k-Nearest Neighbor algorithm (k-NN) to recognize human actions. We represent human actions in the form of local features using a compact descriptor, a histogram of oriented gradient in spatio-temporal 3D space(CHOG3D), which was proposed by us in the paper 1). Then we adopt SOM for feature training to extract key features of action information. With these key features, we adopt k-NN for action recognition. In our experiments, we test the optimal map size of SOM and the proper value k of k-NN for correct recognition. Our method is tested on KTH, Weizmann and UCF datasets, and results certify its efficiency.

Original languageEnglish
Pages (from-to)1-8
Number of pages8
JournalResearch Reports on Information Science and Electrical Engineering of Kyushu University
Volume17
Issue number1
Publication statusPublished - May 2012

Fingerprint

Self organizing maps
Intelligent robots
Classifiers
Experiments

All Science Journal Classification (ASJC) codes

  • Computer Science(all)
  • Electrical and Electronic Engineering

Cite this

@article{95a0fa8d58394bc89fea2811e4a1cc2e,
title = "SOM-based human action recognition using local feature descriptor CHOG3D",
abstract = "Human action recognition is applied in a wide field, such as video surveillance, intelligent interface, and intelligent robots. However, since various action classes, complex surrounding, interaction with objects, et al., it is still a complex problem to be solved. In this paper, we propose a method combining the Self-Organizing Map(SOM) and the classifier k-Nearest Neighbor algorithm (k-NN) to recognize human actions. We represent human actions in the form of local features using a compact descriptor, a histogram of oriented gradient in spatio-temporal 3D space(CHOG3D), which was proposed by us in the paper 1). Then we adopt SOM for feature training to extract key features of action information. With these key features, we adopt k-NN for action recognition. In our experiments, we test the optimal map size of SOM and the proper value k of k-NN for correct recognition. Our method is tested on KTH, Weizmann and UCF datasets, and results certify its efficiency.",
author = "Yanli Ji and Atsushi Shimada and Hajime Nagahara and Taniguchi, {Rin Ichiro}",
year = "2012",
month = "5",
language = "English",
volume = "17",
pages = "1--8",
journal = "Research Reports on Information Science and Electrical Engineering of Kyushu University",
issn = "1342-3819",
publisher = "Kyushu University, Faculty of Science",
number = "1",

}

TY - JOUR

T1 - SOM-based human action recognition using local feature descriptor CHOG3D

AU - Ji, Yanli

AU - Shimada, Atsushi

AU - Nagahara, Hajime

AU - Taniguchi, Rin Ichiro

PY - 2012/5

Y1 - 2012/5

N2 - Human action recognition is applied in a wide field, such as video surveillance, intelligent interface, and intelligent robots. However, since various action classes, complex surrounding, interaction with objects, et al., it is still a complex problem to be solved. In this paper, we propose a method combining the Self-Organizing Map(SOM) and the classifier k-Nearest Neighbor algorithm (k-NN) to recognize human actions. We represent human actions in the form of local features using a compact descriptor, a histogram of oriented gradient in spatio-temporal 3D space(CHOG3D), which was proposed by us in the paper 1). Then we adopt SOM for feature training to extract key features of action information. With these key features, we adopt k-NN for action recognition. In our experiments, we test the optimal map size of SOM and the proper value k of k-NN for correct recognition. Our method is tested on KTH, Weizmann and UCF datasets, and results certify its efficiency.

AB - Human action recognition is applied in a wide field, such as video surveillance, intelligent interface, and intelligent robots. However, since various action classes, complex surrounding, interaction with objects, et al., it is still a complex problem to be solved. In this paper, we propose a method combining the Self-Organizing Map(SOM) and the classifier k-Nearest Neighbor algorithm (k-NN) to recognize human actions. We represent human actions in the form of local features using a compact descriptor, a histogram of oriented gradient in spatio-temporal 3D space(CHOG3D), which was proposed by us in the paper 1). Then we adopt SOM for feature training to extract key features of action information. With these key features, we adopt k-NN for action recognition. In our experiments, we test the optimal map size of SOM and the proper value k of k-NN for correct recognition. Our method is tested on KTH, Weizmann and UCF datasets, and results certify its efficiency.

UR - http://www.scopus.com/inward/record.url?scp=84904507697&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84904507697&partnerID=8YFLogxK

M3 - Article

AN - SCOPUS:84904507697

VL - 17

SP - 1

EP - 8

JO - Research Reports on Information Science and Electrical Engineering of Kyushu University

JF - Research Reports on Information Science and Electrical Engineering of Kyushu University

SN - 1342-3819

IS - 1

ER -