Object-Based Accumulated Motion Feature for the Compressed Domain Human Action Analysis

DOI

10.2991/jcis.2006.262How to use a DOI?

Keywords

Compressed video,Video segmentation,Object-based accumulative motion vector (OAMV),Hidden Markov Models.

Abstract

This paper proposed an effective and robust method to detect the rare behavior events within the compressed video directly. New motion feature called object-based accumulative motion vector (OAMV) is generated to extract a prominent motion feature and then polar histograms are used to describe the distribution patterns for each human action. The various kinds of human actions are identified by the HMM method. Experimental results show that the human actions may be identified accurately.

Copyright

© 2006, 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  - Cheng-Chang Lien
AU  - Chen-Yu Hong
AU  - Yu-Ting Fu
PY  - 2006/10
DA  - 2006/10
TI  - Object-Based Accumulated Motion Feature for the Compressed Domain Human Action Analysis
BT  - Proceedings of the 9th Joint International Conference on Information Sciences (JCIS-06)
PB  - Atlantis Press
SN  - 1951-6851
UR  - https://doi.org/10.2991/jcis.2006.262
DO  - 10.2991/jcis.2006.262
ID  - Lien2006/10
ER  -