Human motion recognition using support vector machines

Dongwei Cao; Osama T. Masoud; Daniel Boley; Nikolaos Papanikolopoulos

doi:10.1016/j.cviu.2009.06.002

Human motion recognition using support vector machines

Dongwei Cao, Osama T. Masoud, Daniel Boley, Nikolaos Papanikolopoulos

Computer Science and Engineering

Research output: Contribution to journal › Article › peer-review

11 Scopus citations

Abstract

We propose a motion recognition strategy that represents each videoclip by a set of filtered images, each of which corresponds to a frame. Using a filtered-image classifier based on support vector machines, we classify a videoclip by applying majority voting over the predicted labels of its filtered images and, for online classification, we identify the most likely type of action at any moment by applying majority voting over the predicted labels of the filtered images within a sliding window. We also define a classification confidence and the associated threshold in both cases, which enable us to identify the existence of an unknown type of motion and, together with the proposed recognition strategy, make it possible to build a real-time motion recognition system that cannot only make classifications in real-time, but also learn new types of motions and recognize them in the future. The proposed strategy is demonstrated on real datasets.

Original language	English (US)
Pages (from-to)	1064-1075
Number of pages	12
Journal	Computer Vision and Image Understanding
Volume	113
Issue number	10
DOIs	https://doi.org/10.1016/j.cviu.2009.06.002
State	Published - Oct 2009

Bibliographical note

Funding Information:
The authors would like to express their gratitude to the anonymous reviewers for their thoughtful comments. This work has been supported in part by the National Science Foundation through Grants #IIS-0219863, #IIS-0208621, #IIS-0534286, #CNS-0224363, #CNS-0324864, #CNS-0420836, #IIP-0443945, #IIP-0726109, #CNS-0708344 and #CNS-0821474, the US Army Research Laboratory, the US Army Research Office under Contract No. 911NF-08-1-0463 (Proposal 55111-CI), the Minnesota Department of Transportation, and the ITS Institute at the University of Minnesota.

Keywords

Human motion recognition
Recursive filtering
Support vector machine

Access

10.1016/j.cviu.2009.06.002

Cite this

@article{6b7e79b2d8514e0bb0aa9bcef25f2b43,

title = "Human motion recognition using support vector machines",

abstract = "We propose a motion recognition strategy that represents each videoclip by a set of filtered images, each of which corresponds to a frame. Using a filtered-image classifier based on support vector machines, we classify a videoclip by applying majority voting over the predicted labels of its filtered images and, for online classification, we identify the most likely type of action at any moment by applying majority voting over the predicted labels of the filtered images within a sliding window. We also define a classification confidence and the associated threshold in both cases, which enable us to identify the existence of an unknown type of motion and, together with the proposed recognition strategy, make it possible to build a real-time motion recognition system that cannot only make classifications in real-time, but also learn new types of motions and recognize them in the future. The proposed strategy is demonstrated on real datasets.",

keywords = "Human motion recognition, Recursive filtering, Support vector machine",

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

note = "Funding Information: The authors would like to express their gratitude to the anonymous reviewers for their thoughtful comments. This work has been supported in part by the National Science Foundation through Grants #IIS-0219863, #IIS-0208621, #IIS-0534286, #CNS-0224363, #CNS-0324864, #CNS-0420836, #IIP-0443945, #IIP-0726109, #CNS-0708344 and #CNS-0821474, the US Army Research Laboratory, the US Army Research Office under Contract No. 911NF-08-1-0463 (Proposal 55111-CI), the Minnesota Department of Transportation, and the ITS Institute at the University of Minnesota. ",

year = "2009",

month = oct,

doi = "10.1016/j.cviu.2009.06.002",

language = "English (US)",

volume = "113",

pages = "1064--1075",

journal = "Computer Vision and Image Understanding",

issn = "1077-3142",

publisher = "Academic Press Inc.",

number = "10",

}

TY - JOUR

T1 - Human motion recognition using support vector machines

AU - Cao, Dongwei

AU - Masoud, Osama T.

AU - Boley, Daniel

AU - Papanikolopoulos, Nikolaos

N1 - Funding Information: The authors would like to express their gratitude to the anonymous reviewers for their thoughtful comments. This work has been supported in part by the National Science Foundation through Grants #IIS-0219863, #IIS-0208621, #IIS-0534286, #CNS-0224363, #CNS-0324864, #CNS-0420836, #IIP-0443945, #IIP-0726109, #CNS-0708344 and #CNS-0821474, the US Army Research Laboratory, the US Army Research Office under Contract No. 911NF-08-1-0463 (Proposal 55111-CI), the Minnesota Department of Transportation, and the ITS Institute at the University of Minnesota.

PY - 2009/10

Y1 - 2009/10

N2 - We propose a motion recognition strategy that represents each videoclip by a set of filtered images, each of which corresponds to a frame. Using a filtered-image classifier based on support vector machines, we classify a videoclip by applying majority voting over the predicted labels of its filtered images and, for online classification, we identify the most likely type of action at any moment by applying majority voting over the predicted labels of the filtered images within a sliding window. We also define a classification confidence and the associated threshold in both cases, which enable us to identify the existence of an unknown type of motion and, together with the proposed recognition strategy, make it possible to build a real-time motion recognition system that cannot only make classifications in real-time, but also learn new types of motions and recognize them in the future. The proposed strategy is demonstrated on real datasets.

AB - We propose a motion recognition strategy that represents each videoclip by a set of filtered images, each of which corresponds to a frame. Using a filtered-image classifier based on support vector machines, we classify a videoclip by applying majority voting over the predicted labels of its filtered images and, for online classification, we identify the most likely type of action at any moment by applying majority voting over the predicted labels of the filtered images within a sliding window. We also define a classification confidence and the associated threshold in both cases, which enable us to identify the existence of an unknown type of motion and, together with the proposed recognition strategy, make it possible to build a real-time motion recognition system that cannot only make classifications in real-time, but also learn new types of motions and recognize them in the future. The proposed strategy is demonstrated on real datasets.

KW - Human motion recognition

KW - Recursive filtering

KW - Support vector machine

UR - http://www.scopus.com/inward/record.url?scp=69549088139&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=69549088139&partnerID=8YFLogxK

U2 - 10.1016/j.cviu.2009.06.002

DO - 10.1016/j.cviu.2009.06.002

M3 - Article

AN - SCOPUS:69549088139

VL - 113

SP - 1064

EP - 1075

JO - Computer Vision and Image Understanding

JF - Computer Vision and Image Understanding

SN - 1077-3142

IS - 10

ER -

Human motion recognition using support vector machines

Abstract

Bibliographical note

Keywords

Access

Other files and links

Fingerprint

Cite this