In driver assistance system, human eye gaze direction is an important feature described some driver's situation such as distraction and fatigue. This paper proposes a method to track driver's gaze direction by using deformable template matching. The method is divided into three steps: first, identifying the face area. Second, localizing the eye area. Finally, combining the eye region model and sight algorithm to determine the driver's gaze direction. Experimental results show that this method can effectively identify human's eye and track gaze motion in a real-time running basis.
Driver gaze tracker using deformable template matching
2011-07-01
230821 byte
Conference paper
Electronic Resource
English
Eye-Gaze Detection from Monocular Camera Image Using Parametric Template Matching
Springer Verlag | 2007
|CALIBRATION APPARATUS OF GAZE TRACKER AND METHOD THEREOF
European Patent Office | 2017
|