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

Chengjiao Guo*, Ying Lu, Xiangzhong Fang, Takeshi Ikenaga

*Corresponding author for this work

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

2 Citations (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 Nov 25
Event2010 6th International Conference on Wireless Communications, Networking and Mobile Computing, WiCOM 2010 - Chengdu, China
Duration: 2010 Sept 232010 Sept 25

Publication series

Name2010 6th International Conference on Wireless Communications, Networking and Mobile Computing, WiCOM 2010

Conference

Conference2010 6th International Conference on Wireless Communications, Networking and Mobile Computing, WiCOM 2010
Country/TerritoryChina
CityChengdu
Period10/9/2310/9/25

Keywords

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

ASJC Scopus subject areas

  • Computer Networks and Communications
  • Communication

Fingerprint

Dive into the research topics of 'Multiple likelihoods and state noises based particle filter for long-lived full occlusion handling'. Together they form a unique fingerprint.

Cite this