Online motion classification using support vector machines

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

Research output: Contribution to journalArticle

24 Citations (Scopus)

Abstract

We propose a motion recognition strategy that represents a videoclip as a set of filtered images, each of which encodes a short period of motion history. Given a set of videoclips whose motion types are known, a filtered image classifier is built using support vector machines. In offline classification, the label of a test videoclip is obtained by applying majority voting over its filtered images. In online classification, the most probable type of action at an instance is determined by applying the majority voting over the most recent filtered images, which are within a sliding window. The effectiveness of this strategy was demonstrated on real datasets where the videoclips were recorded using a fixed camera whose optical axis is perpendicular to the person's trajectory. In offline recognition, the proposed strategy outperforms a Principal Component Analysis based recognition algorithm. In online recognition, the proposed strategy can not only classify motions correctly and identify the transition between different types of motions, but also identify the existence of an unknown motion type. This latter capability and the efficiency of the proposed strategy make it possible to create a real-time motion recognition system that can not only make classifications in real-time, but also learn new types of actions and recognize them in the future.

Original languageEnglish (US)
Pages (from-to)2291-2296
Number of pages6
JournalProceedings - IEEE International Conference on Robotics and Automation
Volume2004
Issue number3
StatePublished - Jul 5 2004

Fingerprint

Support vector machines
Principal component analysis
Labels
Classifiers
Cameras
Trajectories

Keywords

  • Human motion recognition
  • Recursive filtering
  • Support vector machines

Cite this

Online motion classification using support vector machines. / Cao, Dongwei; Masoud, Osama T.; Boley, Daniel L; Papanikolopoulos, Nikolaos P.

In: Proceedings - IEEE International Conference on Robotics and Automation, Vol. 2004, No. 3, 05.07.2004, p. 2291-2296.

Research output: Contribution to journalArticle

@article{b1fa8ade84b845899eddcebd89956c5d,
title = "Online motion classification using support vector machines",
abstract = "We propose a motion recognition strategy that represents a videoclip as a set of filtered images, each of which encodes a short period of motion history. Given a set of videoclips whose motion types are known, a filtered image classifier is built using support vector machines. In offline classification, the label of a test videoclip is obtained by applying majority voting over its filtered images. In online classification, the most probable type of action at an instance is determined by applying the majority voting over the most recent filtered images, which are within a sliding window. The effectiveness of this strategy was demonstrated on real datasets where the videoclips were recorded using a fixed camera whose optical axis is perpendicular to the person's trajectory. In offline recognition, the proposed strategy outperforms a Principal Component Analysis based recognition algorithm. In online recognition, the proposed strategy can not only classify motions correctly and identify the transition between different types of motions, but also identify the existence of an unknown motion type. This latter capability and the efficiency of the proposed strategy make it possible to create a real-time motion recognition system that can not only make classifications in real-time, but also learn new types of actions and recognize them in the future.",
keywords = "Human motion recognition, Recursive filtering, Support vector machines",
author = "Dongwei Cao and Masoud, {Osama T.} and Boley, {Daniel L} and Papanikolopoulos, {Nikolaos P}",
year = "2004",
month = "7",
day = "5",
language = "English (US)",
volume = "2004",
pages = "2291--2296",
journal = "Proceedings - IEEE International Conference on Robotics and Automation",
issn = "1050-4729",
publisher = "Institute of Electrical and Electronics Engineers Inc.",
number = "3",

}

TY - JOUR

T1 - Online motion classification using support vector machines

AU - Cao, Dongwei

AU - Masoud, Osama T.

AU - Boley, Daniel L

AU - Papanikolopoulos, Nikolaos P

PY - 2004/7/5

Y1 - 2004/7/5

N2 - We propose a motion recognition strategy that represents a videoclip as a set of filtered images, each of which encodes a short period of motion history. Given a set of videoclips whose motion types are known, a filtered image classifier is built using support vector machines. In offline classification, the label of a test videoclip is obtained by applying majority voting over its filtered images. In online classification, the most probable type of action at an instance is determined by applying the majority voting over the most recent filtered images, which are within a sliding window. The effectiveness of this strategy was demonstrated on real datasets where the videoclips were recorded using a fixed camera whose optical axis is perpendicular to the person's trajectory. In offline recognition, the proposed strategy outperforms a Principal Component Analysis based recognition algorithm. In online recognition, the proposed strategy can not only classify motions correctly and identify the transition between different types of motions, but also identify the existence of an unknown motion type. This latter capability and the efficiency of the proposed strategy make it possible to create a real-time motion recognition system that can not only make classifications in real-time, but also learn new types of actions and recognize them in the future.

AB - We propose a motion recognition strategy that represents a videoclip as a set of filtered images, each of which encodes a short period of motion history. Given a set of videoclips whose motion types are known, a filtered image classifier is built using support vector machines. In offline classification, the label of a test videoclip is obtained by applying majority voting over its filtered images. In online classification, the most probable type of action at an instance is determined by applying the majority voting over the most recent filtered images, which are within a sliding window. The effectiveness of this strategy was demonstrated on real datasets where the videoclips were recorded using a fixed camera whose optical axis is perpendicular to the person's trajectory. In offline recognition, the proposed strategy outperforms a Principal Component Analysis based recognition algorithm. In online recognition, the proposed strategy can not only classify motions correctly and identify the transition between different types of motions, but also identify the existence of an unknown motion type. This latter capability and the efficiency of the proposed strategy make it possible to create a real-time motion recognition system that can not only make classifications in real-time, but also learn new types of actions and recognize them in the future.

KW - Human motion recognition

KW - Recursive filtering

KW - Support vector machines

UR - http://www.scopus.com/inward/record.url?scp=3042625140&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=3042625140&partnerID=8YFLogxK

M3 - Article

VL - 2004

SP - 2291

EP - 2296

JO - Proceedings - IEEE International Conference on Robotics and Automation

JF - Proceedings - IEEE International Conference on Robotics and Automation

SN - 1050-4729

IS - 3

ER -