Multiple likelihoods and state noises based particle filter for long-lived full occlusion handling

Chengjiao Guo, Ying Lu, Xiangzhong Fang, Takeshi Ikenaga

Research output: Chapter in Book/Report/Conference proceedingConference contribution

1 Citation (Scopus)

Abstract

Reliable object tracking in complex visual environment is a challenging problem in the field of computer vision. One of the common problems in object tracking is partial and full object occlusions. And especially in the condition of long-lived full occlusion during which the full occlusion lasts for tens of frames, the tracking is more difficult. This paper proposes an occlusion handling scheme based on particle filter. Compared with the standard particle filter, multiple likelihood models - HSV color likelihood and gradient orientation likelihood, are employed in the observation model for occlusion handling. Also, multiple state noises are introduced under occlusion. Experiment results demonstrate the robust and accurate tracking performance in the condition of long-lived full occlusion.

Original languageEnglish
Title of host publication2010 6th International Conference on Wireless Communications, Networking and Mobile Computing, WiCOM 2010
DOIs
Publication statusPublished - 2010
Event2010 6th International Conference on Wireless Communications, Networking and Mobile Computing, WiCOM 2010 - Chengdu
Duration: 2010 Sep 232010 Sep 25

Other

Other2010 6th International Conference on Wireless Communications, Networking and Mobile Computing, WiCOM 2010
CityChengdu
Period10/9/2310/9/25

Fingerprint

Computer vision
Color
Experiments
experiment
performance

Keywords

  • Long-lived full occlusion
  • Object tracking
  • Particle filter
  • State transition model

ASJC Scopus subject areas

  • Computer Networks and Communications
  • Communication

Cite this

Guo, C., Lu, Y., Fang, X., & Ikenaga, T. (2010). Multiple likelihoods and state noises based particle filter for long-lived full occlusion handling. In 2010 6th International Conference on Wireless Communications, Networking and Mobile Computing, WiCOM 2010 [5601019] https://doi.org/10.1109/WICOM.2010.5601019

Multiple likelihoods and state noises based particle filter for long-lived full occlusion handling. / Guo, Chengjiao; Lu, Ying; Fang, Xiangzhong; Ikenaga, Takeshi.

2010 6th International Conference on Wireless Communications, Networking and Mobile Computing, WiCOM 2010. 2010. 5601019.

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Guo, C, Lu, Y, Fang, X & Ikenaga, T 2010, Multiple likelihoods and state noises based particle filter for long-lived full occlusion handling. in 2010 6th International Conference on Wireless Communications, Networking and Mobile Computing, WiCOM 2010., 5601019, 2010 6th International Conference on Wireless Communications, Networking and Mobile Computing, WiCOM 2010, Chengdu, 10/9/23. https://doi.org/10.1109/WICOM.2010.5601019
Guo C, Lu Y, Fang X, Ikenaga T. Multiple likelihoods and state noises based particle filter for long-lived full occlusion handling. In 2010 6th International Conference on Wireless Communications, Networking and Mobile Computing, WiCOM 2010. 2010. 5601019 https://doi.org/10.1109/WICOM.2010.5601019
Guo, Chengjiao ; Lu, Ying ; Fang, Xiangzhong ; Ikenaga, Takeshi. / Multiple likelihoods and state noises based particle filter for long-lived full occlusion handling. 2010 6th International Conference on Wireless Communications, Networking and Mobile Computing, WiCOM 2010. 2010.
@inproceedings{c57f6589f4b4469ba34e495a46199a06,
title = "Multiple likelihoods and state noises based particle filter for long-lived full occlusion handling",
abstract = "Reliable object tracking in complex visual environment is a challenging problem in the field of computer vision. One of the common problems in object tracking is partial and full object occlusions. And especially in the condition of long-lived full occlusion during which the full occlusion lasts for tens of frames, the tracking is more difficult. This paper proposes an occlusion handling scheme based on particle filter. Compared with the standard particle filter, multiple likelihood models - HSV color likelihood and gradient orientation likelihood, are employed in the observation model for occlusion handling. Also, multiple state noises are introduced under occlusion. Experiment results demonstrate the robust and accurate tracking performance in the condition of long-lived full occlusion.",
keywords = "Long-lived full occlusion, Object tracking, Particle filter, State transition model",
author = "Chengjiao Guo and Ying Lu and Xiangzhong Fang and Takeshi Ikenaga",
year = "2010",
doi = "10.1109/WICOM.2010.5601019",
language = "English",
isbn = "9781424437092",
booktitle = "2010 6th International Conference on Wireless Communications, Networking and Mobile Computing, WiCOM 2010",

}

TY - GEN

T1 - Multiple likelihoods and state noises based particle filter for long-lived full occlusion handling

AU - Guo, Chengjiao

AU - Lu, Ying

AU - Fang, Xiangzhong

AU - Ikenaga, Takeshi

PY - 2010

Y1 - 2010

N2 - Reliable object tracking in complex visual environment is a challenging problem in the field of computer vision. One of the common problems in object tracking is partial and full object occlusions. And especially in the condition of long-lived full occlusion during which the full occlusion lasts for tens of frames, the tracking is more difficult. This paper proposes an occlusion handling scheme based on particle filter. Compared with the standard particle filter, multiple likelihood models - HSV color likelihood and gradient orientation likelihood, are employed in the observation model for occlusion handling. Also, multiple state noises are introduced under occlusion. Experiment results demonstrate the robust and accurate tracking performance in the condition of long-lived full occlusion.

AB - Reliable object tracking in complex visual environment is a challenging problem in the field of computer vision. One of the common problems in object tracking is partial and full object occlusions. And especially in the condition of long-lived full occlusion during which the full occlusion lasts for tens of frames, the tracking is more difficult. This paper proposes an occlusion handling scheme based on particle filter. Compared with the standard particle filter, multiple likelihood models - HSV color likelihood and gradient orientation likelihood, are employed in the observation model for occlusion handling. Also, multiple state noises are introduced under occlusion. Experiment results demonstrate the robust and accurate tracking performance in the condition of long-lived full occlusion.

KW - Long-lived full occlusion

KW - Object tracking

KW - Particle filter

KW - State transition model

UR - http://www.scopus.com/inward/record.url?scp=78549238909&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=78549238909&partnerID=8YFLogxK

U2 - 10.1109/WICOM.2010.5601019

DO - 10.1109/WICOM.2010.5601019

M3 - Conference contribution

AN - SCOPUS:78549238909

SN - 9781424437092

BT - 2010 6th International Conference on Wireless Communications, Networking and Mobile Computing, WiCOM 2010

ER -