Blink detection and eye tracking for eye localization

Author

Bhaskar, T.N. ; Keat, Foo Tun ; Ranganath, Surendra ; Venkatesh, Y.V.

Author_Institution

Dept. of Electr. & Comput. Eng., Nat. Univ. of Singapore, Singapore

Volume

2

fYear

2003

fDate

15-17 Oct. 2003

Firstpage

821

Abstract

A method of using frame differencing coupled with optical flow computation for eye blink detection is proposed. Frame differencing allows quick determination of possible motion regions. If they are detected, optical flow is computed within these regions. The direction and magnitude of the flow field are then used to determine whether a blink has occurred. The eyes are then tracked using the Kanade Lucas Tomasi (KLT) tracker. We obtained a success rate of 97.0% in blink detection using the proposed method, and localised the eyes automatically at an average rate of 22 frames per second.

Keywords

eye; image motion analysis; image sequences; tracking; KLT tracker; Kanade Lucas Tomasi tracker; blink detection; eye localization; eye tracking; flow field direction; flow field magnitude; frame differencing; motion regions; optical flow computation; Cameras; Eyes; Face detection; Image motion analysis; Karhunen-Loeve transforms; Motion detection; Optical computing; Optical coupling; Optical detectors; Security;

fLanguage

English

Publisher

ieee

Conference_Titel

TENCON 2003. Conference on Convergent Technologies for the Asia-Pacific Region

Print_ISBN

0-7803-8162-9

Type

conf

DOI

10.1109/TENCON.2003.1273293

Filename

1273293