Automatic detection of camera translation in eye video recordings using multiple methods

Faisal Karmali, Mark Shelhamer

Research output: Contribution to journalArticle

Abstract

A concern with video eye movement tracking is that movement of the camera headset relative to the head creates an artifact of eye movement in pupil-detection software. We describe the development of, and compare the results of, three automatic image processing algorithms to measure camera movement. The best of the algorithms has an average accuracy of 1.3 pixels, equivalent to 0.49 deg with our eye tracking system.

Original languageEnglish (US)
Pages (from-to)470-476
Number of pages7
JournalAnnals of the New York Academy of Sciences
Volume1039
DOIs
StatePublished - 2005

Keywords

  • Edge detection
  • Eye
  • Eyelid
  • Image processing
  • Video

ASJC Scopus subject areas

  • Neuroscience(all)
  • Biochemistry, Genetics and Molecular Biology(all)
  • History and Philosophy of Science

Fingerprint Dive into the research topics of 'Automatic detection of camera translation in eye video recordings using multiple methods'. Together they form a unique fingerprint.

  • Cite this