Research on Human Action Recognition Based on Global and Local Mixed Features

DOI

10.2991/mce-14.2014.175How to use a DOI?

Keywords

Action Recognition; Mixed Features; Global; Local; Cognitive Map

Abstract

In recent years, human action recognition has become a hot topic in the computer visual and pattern recognition fields, which has a wide application prospect, such as intelligent monitoring, human-computer interaction, virtual reality, content-based video retrieval and home robotics services. Although progresses have been made in the existing action recognition technology, it is still in the early stages of recognizing a minority of standard gesture and simple action experimental samples. This paper attempts to realize human action recognition through global features and local features structure variables and mixed feature descriptors reflecting the feature causal relationship, and thus improves the recognition rate of human action. This paper has certain practical significance and application values.

Copyright

© 2014, the Authors. Published by Atlantis Press.

Open Access

This is an open access article distributed under the CC BY-NC license (http://creativecommons.org/licenses/by-nc/4.0/).

Download article (PDF)

TY  - CONF
AU  - Xueping Liu
AU  - Yibo Li
PY  - 2014/03
DA  - 2014/03
TI  - Research on Human Action Recognition Based on Global and Local Mixed Features
BT  - Proceedings of the 2014 International Conference on Mechatronics, Control and Electronic Engineering
PB  - Atlantis Press
SP  - 778
EP  - 782
SN  - 1951-6851
UR  - https://doi.org/10.2991/mce-14.2014.175
DO  - 10.2991/mce-14.2014.175
ID  - Liu2014/03
ER  -