Authors

Yunkai Huang

Type

Text

Type

Thesis

Advisor

Hong, Sangjin | Milder, Peter.

Date

2013-12-01

Keywords

Computer engineering | camera, gaze behavior detection, gaze tracking, opencv

Department

Department of Electrical Engineering.

Language

en_US

Source

This work is sponsored by the Stony Brook University Graduate School in compliance with the requirements for completion of degree.

Identifier

http://hdl.handle.net/11401/77468

Publisher

The Graduate School, Stony Brook University: Stony Brook, NY.

Format

application/pdf

Abstract

This thesis presents a design methodology of a low-cost noninvasive gaze tracking system to detect gaze behavior when user is browsing internet or reading material on computer. The user's face image is captured and processed in real-time. By means of C++ and OpenCV library, the system detects face, eye region with Haar feature-based cascade classifier. Eye center is detected by contouring dark area in eye region and finding the center of largest area among contoured dark areas. The detected eye center is mapped to gaze point on computer screen after four point calibration. The average angular error is 1.96 degree, which is comparable to other proposed techniques. During the experiment, the gaze point is displayed real-time with eye movement, and its coordinate as well as the gazed object are recorded in file. The system represents image information in unit area, object, scene, and frame hierarchy structure. With the gaze point data and image information, it is able to analyze gaze duration among objects and understand user's gaze behavior. | 44 pages

Share

COinS
 
 

To view the content in your browser, please download Adobe Reader or, alternately,
you may Download the file to your hard drive.

NOTE: The latest versions of Adobe Reader do not support viewing PDF files within Firefox on Mac OS and if you are using a modern (Intel) Mac, there is no official plugin for viewing PDF files within the browser window.