Human motion recognition with a convolution kernel

Dongwei Cao; Osama T. Masoud; Daniel L Boley

doi:10.1109/ROBOT.2006.1642359

Human motion recognition with a convolution kernel

Dongwei Cao, Osama T. Masoud, Daniel L Boley

Computer Science and Engineering

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

Abstract

We address the problem of human motion recognition in this paper. The goal of human motion recognition is to recognize the type of motion recorded in a video clip, which consists of a set of temporarily ordered frames. By defining a Mercer kernel between two video clips directly, we propose in this paper a recognition strategy that can incorporate both the information of each individual frame and the temporal ordering between frames. Combining the proposed kernel with the support vector machine, which is one of the most effective classification paradigms, the resulting recognition strategy exhibits excellent performance over real data sets.

Original language	English (US)
Title of host publication	Proceedings 2006 IEEE International Conference on Robotics and Automation, ICRA 2006
Pages	4270-4275
Number of pages	6
DOIs	https://doi.org/10.1109/ROBOT.2006.1642359
State	Published - 2006
Event	2006 IEEE International Conference on Robotics and Automation, ICRA 2006 - Orlando, FL, United States Duration: May 15 2006 → May 19 2006

Publication series

Name	Proceedings - IEEE International Conference on Robotics and Automation
Volume	2006
ISSN (Print)	1050-4729

Other

Other	2006 IEEE International Conference on Robotics and Automation, ICRA 2006
Country/Territory	United States
City	Orlando, FL
Period	5/15/06 → 5/19/06

Keywords

Convolution kernels
Human motion recognition
Support vector machines

Access

10.1109/ROBOT.2006.1642359

OpenUrl availability

Full text

Cite this

Human motion recognition with a convolution kernel. / Cao, Dongwei; Masoud, Osama T.; Boley, Daniel L.
Proceedings 2006 IEEE International Conference on Robotics and Automation, ICRA 2006. 2006. p. 4270-4275 1642359 (Proceedings - IEEE International Conference on Robotics and Automation; Vol. 2006).

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

Cao, D, Masoud, OT & Boley, DL 2006, Human motion recognition with a convolution kernel. in Proceedings 2006 IEEE International Conference on Robotics and Automation, ICRA 2006., 1642359, Proceedings - IEEE International Conference on Robotics and Automation, vol. 2006, pp. 4270-4275, 2006 IEEE International Conference on Robotics and Automation, ICRA 2006, Orlando, FL, United States, 5/15/06. https://doi.org/10.1109/ROBOT.2006.1642359

@inproceedings{76bfbef0dd0b4d34b51e4bf2422db573,

title = "Human motion recognition with a convolution kernel",

abstract = "We address the problem of human motion recognition in this paper. The goal of human motion recognition is to recognize the type of motion recorded in a video clip, which consists of a set of temporarily ordered frames. By defining a Mercer kernel between two video clips directly, we propose in this paper a recognition strategy that can incorporate both the information of each individual frame and the temporal ordering between frames. Combining the proposed kernel with the support vector machine, which is one of the most effective classification paradigms, the resulting recognition strategy exhibits excellent performance over real data sets.",

keywords = "Convolution kernels, Human motion recognition, Support vector machines",

author = "Dongwei Cao and Masoud, {Osama T.} and Boley, {Daniel L}",

year = "2006",

doi = "10.1109/ROBOT.2006.1642359",

language = "English (US)",

isbn = "0780395069",

series = "Proceedings - IEEE International Conference on Robotics and Automation",

pages = "4270--4275",

booktitle = "Proceedings 2006 IEEE International Conference on Robotics and Automation, ICRA 2006",

note = "2006 IEEE International Conference on Robotics and Automation, ICRA 2006 ; Conference date: 15-05-2006 Through 19-05-2006",

}

TY - GEN

T1 - Human motion recognition with a convolution kernel

AU - Cao, Dongwei

AU - Masoud, Osama T.

AU - Boley, Daniel L

PY - 2006

Y1 - 2006

N2 - We address the problem of human motion recognition in this paper. The goal of human motion recognition is to recognize the type of motion recorded in a video clip, which consists of a set of temporarily ordered frames. By defining a Mercer kernel between two video clips directly, we propose in this paper a recognition strategy that can incorporate both the information of each individual frame and the temporal ordering between frames. Combining the proposed kernel with the support vector machine, which is one of the most effective classification paradigms, the resulting recognition strategy exhibits excellent performance over real data sets.

AB - We address the problem of human motion recognition in this paper. The goal of human motion recognition is to recognize the type of motion recorded in a video clip, which consists of a set of temporarily ordered frames. By defining a Mercer kernel between two video clips directly, we propose in this paper a recognition strategy that can incorporate both the information of each individual frame and the temporal ordering between frames. Combining the proposed kernel with the support vector machine, which is one of the most effective classification paradigms, the resulting recognition strategy exhibits excellent performance over real data sets.

KW - Convolution kernels

KW - Human motion recognition

KW - Support vector machines

UR - http://www.scopus.com/inward/record.url?scp=33845601340&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=33845601340&partnerID=8YFLogxK

U2 - 10.1109/ROBOT.2006.1642359

DO - 10.1109/ROBOT.2006.1642359

M3 - Conference contribution

AN - SCOPUS:33845601340

SN - 0780395069

SN - 9780780395060

T3 - Proceedings - IEEE International Conference on Robotics and Automation

SP - 4270

EP - 4275

BT - Proceedings 2006 IEEE International Conference on Robotics and Automation, ICRA 2006

T2 - 2006 IEEE International Conference on Robotics and Automation, ICRA 2006

Y2 - 15 May 2006 through 19 May 2006

ER -

Human motion recognition with a convolution kernel

Abstract

Publication series

Other

Keywords

Access

OpenUrl availability

Other files and links

Fingerprint

Cite this