ScienceDirect® Home Skip Main Navigation Links
You have guest access to ScienceDirect. Find out more.
 
Home
Browse
My Settings
Alerts
Help
 Quick Search
 Search tips (Opens new window)
    Clear all fields    
Pattern Recognition
Volume 35, Issue 6, June 2002, Pages 1389-1401
 
Font Size: Decrease Font Size  Increase Font Size
 Abstract - selected
Article
Purchase PDF (417 K)

 
 
 
Related Articles in ScienceDirect
View More Related Articles
 
View Record in Scopus
 
doi:10.1016/S0031-3203(01)00116-9    How to Cite or Link Using DOI (Opens New Window)
Copyright © 2002 Pattern Recognition Society. Published by Elsevier Science B.V.

A method of detecting and tracking irises and eyelids in video

S. Sirohey1, A. RosenfeldCorresponding Author Contact Information, E-mail The Corresponding Author and Z. Duric2

Center for Automation Research, Institute for Advanced Computer Studies, Computer Vision Laboratory, University of Maryland, College Park, MD 20742-3275, USA

Received 20 February 2001;
revised 12 April 2001;
accepted 30 April 2001
Available online 28 February 2002.

Purchase the full-text article



References and further reading may be available for this article. To view references and further reading you must purchase this article.

Abstract

We locate the eye corners, eyelids, and irises in every frame of an image sequence, and analyze the movements of the irises and eyelids to determine changes in gaze direction and blinking, respectively. Using simple models for the motions of the head and eyes, we determine the head-independent motions of the irises and eyelids by stabilizing for the head motion. The head-independent motions of the irises can be used to determine behaviors like saccades and smooth pursuit. Tracking the upper eyelid and using the distance between its apex and the center of the iris, we detect instances of eye closure during blinking. In experiments on two short image sequences, in one of which the subject was wearing glasses, we successfully located the irises in every frame in which the eyes were fully or partially open, and successfully located the eyelids 80% of the time. When motion information in the form of normal flow was used, the irises were successfully tracked in every frame in which the eyes were fully or partially open, and the eyelids were successfully located and tracked 90% of the time.

Author Keywords: Eye detection; Eyelid detection; Iris detection; Gaze tracking; Blink detection

Article Outline

1. Introduction
2. Literature review
3. Eye part detection
3.1. Iris detection
3.2. Eyelid detection
4. Frame-to-frame tracking
4.1. Iris motion
4.2. Eyelid motion
4.3. Experimental results
5. Flow-based tracking
5.1. Normal flow measurement
5.2. Regions of interest and motion models
5.3. Head motion
5.4. De-coupling motion components by stabilization
5.5. Iris motion
5.6. Eyelid motion
6. Conclusions
References
Vitae














Pattern Recognition
Volume 35, Issue 6, June 2002, Pages 1389-1401
 
Home
Browse
My Settings
Alerts
Help
Elsevier.com (Opens new window)
About ScienceDirect  |  Contact Us  |  Information for Advertisers  |  Terms & Conditions  |  Privacy Policy
Copyright © 2008 Elsevier B.V. All rights reserved. ScienceDirect® is a registered trademark of Elsevier B.V.