Motion history histograms for human action recognition

Meng, Hongying and Pears, Nick and Freeman, Michael and Bailey, Chris (2008) Motion history histograms for human action recognition. In: Embedded Computer Vision (Advances in Pattern Recognition). Springer, UK. ISBN 9781848003033

Documents
Motion history histograms for human action recognition
[img]
[Download]
Request a copy
[img] PDF
ecvChapterYork-v04.pdf
Restricted to Repository staff only

1MB

Abstract

In this chapter, a compact human action recognition system is presented with a view to applications in security systems, human-computer interaction and intelligent environments. There are three main contributions: Firstly, the framework of an embedded human action recognition system based on a support vector machine (SVM) classifier and some compact motion features has been proposed. Secondly, the limitations of the well known Motion History Image (MHI) are addressed and a new Motion History Histograms (MHH) feature is proposed to represent the motion information in the video. MHH not only provides rich motion information, but also remains computationally inexpensive. We combine MHI and MHH into a low dimensional feature vector for the system and achieve improved performance in human action recognition over comparable methods, which use tracking-free temporal template motion representations. Finally, an simple system based on SVM and MHI has been implemented on a reconfigurable embedded computer vision architecture for real-time gesture recognition.

Item Type:Book Section
Additional Information:In this chapter, a compact human action recognition system is presented with a view to applications in security systems, human-computer interaction and intelligent environments. There are three main contributions: Firstly, the framework of an embedded human action recognition system based on a support vector machine (SVM) classifier and some compact motion features has been proposed. Secondly, the limitations of the well known Motion History Image (MHI) are addressed and a new Motion History Histograms (MHH) feature is proposed to represent the motion information in the video. MHH not only provides rich motion information, but also remains computationally inexpensive. We combine MHI and MHH into a low dimensional feature vector for the system and achieve improved performance in human action recognition over comparable methods, which use tracking-free temporal template motion representations. Finally, an simple system based on SVM and MHI has been implemented on a reconfigurable embedded computer vision architecture for real-time gesture recognition.
Keywords:Embedded system, FPGA
Subjects:G Mathematical and Computer Sciences > G411 Computer Architectures
G Mathematical and Computer Sciences > G760 Machine Learning
G Mathematical and Computer Sciences > G400 Computer Science
G Mathematical and Computer Sciences > G740 Computer Vision
Divisions:College of Science > School of Computer Science
ID Code:1973
Deposited By:INVALID USER
Deposited On:12 Aug 2009 13:26
Last Modified:13 Mar 2013 08:33

Repository Staff Only: item control page