Abstract
Detection of gaze requires estimation of the position and the relation between user’s pupil and glint. This position is mapped into the region of interest using different edge detectors by detecting the glint coordinates and further gaze direction. In this research paper, a Gaze Direction Estimation (GDE) model has been proposed for the comparative analysis of two standard edge detectors Canny and Sobel for estimating automatic detection of the glint, its coordinates and subsequently the gaze direction. The results indicate fairly good percentage of the cases where the correct glint coordinates and subsequently correct gaze direction quadrants have been estimated. These results can further be used for improving the accuracy and performance of different eye gaze based systems.
Highlights
Eye gaze, by replacing the computer mouse and keyboard as the input method, provides a more immediate way of communication for the selection of any task, icon or menu by the user
Total one hundred input images have been taken for creating an image database DB[i] which consists of images of both male and female, all with normal eye vision in the ratio of 14:6
In this research paper Gaze Direction Estimation (GDE) model has been proposed for the detection of glint coordinates and gaze quadrant of the input eye image using edge detection operators
Summary
By replacing the computer mouse and keyboard as the input method, provides a more immediate way of communication for the selection of any task, icon or menu by the user. Communication through the direction of eyes is faster than any other mode of human communication (Bednarik, Gowases, & Tukiainen, 2009; Sibert & Jacob, 2000; Sharma & Abrol, 2013). Gaze based system requires the estimation and detection of gaze with high accuracy so as to identify the region of interest (RoI) in the input image. Based on the detected position of RoI, the gaze direction can be used to activate any icon or object on the screen. Gaze estimation is generally done in relation to gaze direction of the user’s eye position and movements which. Direction estimation model for gaze controlled systems.
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have