Gaze tracking based on active appearance model and multiple support vector regression on mobile devices

Eui Chul Lee; You Jin Ko; Kang Ryoung Park

doi:10.1117/1.3158992

1 July 2009 Gaze tracking based on active appearance model and multiple support vector regression on mobile devices

Eui Chul Lee, You Jin Ko, Kang Ryoung Park

Author Affiliations +

Optical Engineering, Vol. 48, Issue 7, 077002 (July 2009). https://doi.org/10.1117/1.3158992

Abstract

Gaze tracking technology is a convenient interfacing method for mobile devices. Most previous studies used a large-sized desktop or head-mounted display. In this study, we propose a novel gaze tracking method using an active appearance model (AAM) and multiple support vector regression (SVR) on a mobile device. Our research has four main contributions. First, in calculating the gaze position, the amount of facial rotation and translation based on four feature values is computed using facial feature points detected by AAM. Second, the amount of eye rotation based on two feature values is computed for measuring eye gaze position. Third, to compensate for the fitting error of an AAM in facial rotation, we use the adaptive discrete Kalman filter (DKF), which applies a different velocity of state transition matrix to the facial feature points. Fourth, we obtain gaze position on a mobile device based on multiple SVR by separating the rotation and translation of face and eye rotation. Experimental results show that the root mean square (rms) gaze error is 36.94 pixels on the 4.5-in. screen of a mobile device with a screen resolution of 800×600 pixels.

©(2009) Society of Photo-Optical Instrumentation Engineers (SPIE)

Citation Download Citation

Eui Chul Lee, You Jin Ko, and Kang Ryoung Park "Gaze tracking based on active appearance model and multiple support vector regression on mobile devices," Optical Engineering 48(7), 077002 (1 July 2009). https://doi.org/10.1117/1.3158992

Published: 1 July 2009

ACCESS THE FULL ARTICLE

INSTITUTIONAL
Select your institution to access the SPIE Digital Library.

SELECT YOUR INSTITUTION

PERSONAL
Sign in with your SPIE account to access your personal subscriptions or to use specific features such as save to my library, sign up for alerts, save searches, etc.

PERSONAL SIGN IN

No SPIE Account? Create one

PURCHASE THIS CONTENT

SUBSCRIBE TO DIGITAL LIBRARY

50 downloads per 1-year subscription

Members: $195

Non-members: $335 ADD TO CART

25 downloads per 1 - year subscription

Members: $145

Non-members: $250 ADD TO CART

PURCHASE SINGLE ARTICLE

Includes PDF, HTML & Video, when available