EyeGuardian: a framework of eye tracking and blink detection for mobile device users.

ABSTRACTComputer Vision Syndrome (CVS) is a common problem in the "Information Age", and it is becoming more serious as mobile devices (e.g. smartphones and tablet PCs) with small, low-resolution screens are outnumbering the home computers. The simplest way to avoid CVS is to blink frequently. However, most people do not realize that they blink less and some do not blink at all in front of the screen. In this paper, we present a mobile application that keeps track of the reader's blink rate and prods the user to blink if an exceptionally low blink rate is detected. The proposed eye detection and tracking algorithm is designed for mobile devices and can keep track of the eyes in spite of camera motion. The main idea is to predict the eye position in the camera frame using the feedback from the built-in accelerometer. The eye tracking system was built on a commercial Tablet PC. The experimental results consistently show that the scheme can withstand very aggressive mobility scenarios.