Motion History Histograms for Human Action Recognition

Hongying Meng; Nick Pears; Michael Freeman; Chris Bailey; Branislav Kisacanin; Shuvra S. Bhattacharyya; Sek Chai

Motion History Histograms for Human Action Recognition

Hongying Meng, Nick Pears, Michael Freeman, Chris Bailey, Branislav Kisacanin (Editor), Shuvra S. Bhattacharyya (Editor), Sek Chai (Editor)

Computer Science

Research output: Chapter in Book/Report/Conference proceeding › Chapter

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 presented. Secondly, the limitations of the well-known motion history image (MHI) are addressed and a new motion history histograms (MHH) feature is introduced 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 that use tracking-free temporal template motion representations. Finally, a simple system based on SVM and MHI has been implemented on a reconfigurable embedded computer vision architecture for real-time gesture recognition.

Original language	English
Title of host publication	Embedded Computer Vision
Publisher	Springer
Pages	139-162
Number of pages	24
Publication status	Published - 2009

Publication series

Name	Advances in Pattern Recognition
Publisher	Springer London

Bibliographical note

10.1007/978-1-84800-304-0_7

Access to Document

http://dx.doi.org/10.1007/978-1-84800-304-0_7

Cite this

@inbook{f89deb6a08ec4007b3c24baf77bc3d4c,

title = "Motion History Histograms for Human Action Recognition",

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 presented. Secondly, the limitations of the well-known motion history image (MHI) are addressed and a new motion history histograms (MHH) feature is introduced 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 that use tracking-free temporal template motion representations. Finally, a simple system based on SVM and MHI has been implemented on a reconfigurable embedded computer vision architecture for real-time gesture recognition.",

author = "Hongying Meng and Nick Pears and Michael Freeman and Chris Bailey and Branislav Kisacanin and Bhattacharyya, {Shuvra S.} and Sek Chai",

note = "10.1007/978-1-84800-304-0_7",

year = "2009",

language = "English",

series = "Advances in Pattern Recognition",

publisher = "Springer",

pages = "139--162",

booktitle = "Embedded Computer Vision",

address = "Germany",

}

TY - CHAP

T1 - Motion History Histograms for Human Action Recognition

AU - Meng, Hongying

AU - Pears, Nick

AU - Freeman, Michael

AU - Bailey, Chris

A2 - Kisacanin, Branislav

A2 - Bhattacharyya, Shuvra S.

A2 - Chai, Sek

N1 - 10.1007/978-1-84800-304-0_7

PY - 2009

Y1 - 2009

N2 - 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 presented. Secondly, the limitations of the well-known motion history image (MHI) are addressed and a new motion history histograms (MHH) feature is introduced 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 that use tracking-free temporal template motion representations. Finally, a simple system based on SVM and MHI has been implemented on a reconfigurable embedded computer vision architecture for real-time gesture recognition.

AB - 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 presented. Secondly, the limitations of the well-known motion history image (MHI) are addressed and a new motion history histograms (MHH) feature is introduced 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 that use tracking-free temporal template motion representations. Finally, a simple system based on SVM and MHI has been implemented on a reconfigurable embedded computer vision architecture for real-time gesture recognition.

M3 - Chapter

T3 - Advances in Pattern Recognition

SP - 139

EP - 162

BT - Embedded Computer Vision

PB - Springer

ER -