Boosting particle filter-based eye tracker performance through adapted likelihood function to reflexions and light changes

Riad Hammoud, Dan Witzner Hansen

Research output: Conference Article in Proceeding or Book/Report chapterBook chapterResearchpeer-review

Abstract

In this paper we propose a log likelihood-ratio function of foreground and background models used in a particle filter to track the eye region in dark-bright pupil image sequences. This model fuses information from both dark and bright pupil images and their difference image into one model. The tracker overcomes the issues of prior selection of static thresholds during the detection of feature observations in the bright-dark difference images. The auto-initialization process is performed using cascaded classifier trained using adaboost and adapted to IR eye images. Experiments show good performance in challenging sequences with test subjects showing large head movements and under significant light changes.
Original languageEnglish
Title of host publicationBoosting particle filter-based eye tracker performance through adapted likelihood function to reflexions and light changes
Number of pages6
Publication date2005
Pages111-116
ISBN (Print)0780393856
DOIs
Publication statusPublished - 2005

Keywords

  • log likelihood-ratio
  • particle filter
  • eye tracking
  • dark-bright pupil images
  • Adaboost classifier

Fingerprint

Dive into the research topics of 'Boosting particle filter-based eye tracker performance through adapted likelihood function to reflexions and light changes'. Together they form a unique fingerprint.

Cite this