University of Illinois Urbana-Champaign

Web Cam Eye Gaze Estimation with Lighting Interaction System

Park, Doheum

This item is only available for download by members of the University of Illinois community. Students, faculty, and staff at the U of I may log in with your NetID and password to view the item. If you are trying to access an Illinois-restricted dissertation or thesis, you can request a copy through your library's Inter-Library Loan office or purchase a copy directly from ProQuest.

Permalink

https://hdl.handle.net/2142/46535

Description

Title
Web Cam Eye Gaze Estimation with Lighting Interaction System
Author(s)
Park, Doheum
Contributor(s)
Hoiem, Derek
Issue Date
2012-12
Keyword(s)
  • human-computer interaction
  • computer interfaces
  • eye gaze estimation
  • eye tracking
Abstract
Our research goal is to enable accurate systems at lower cost and to incorporate user interaction to improve robustness to difficult lighting environments, which would exist in many commercial applications. Our research aims to replace the complex equipment currently used for eye tracking with a simple web cam. Currently, our system sacrifices a small amount of accuracy but greatly reduces cost, a good trade-off for many commercial uses. Our system detects eye center locations using an isocenter estimation technique. Eye corner locations are detected using normalized cross-correlation template matching. Taking advantage of sequential video frames from the web cam, we incorporated Kalman filtering to more accurately track the eye locations. Our system uses the eye location data to estimate visual gaze direction, mapping between location data and screen pixel coordinates. Additionally, our work focused on improving robustness to lighting conditions. Inappropriate lighting conditions make it difficut to detect eye centers and corners required for visual gaze estimation. We alleviate the problem of lighting conditions using human-computer interaction, which has also been proposed in previous work. Our system uses exact image histogram specification and gamma correction to correct problematic lighting conditions. If the lighting condition of an image is not adequately improved with feedback, the interaction system notifies the user to manually adjust the lighting environment.
Type of Resource
text
Language
en
Permalink
http://hdl.handle.net/2142/46535

Owning Collections

Senior Theses - Electrical and Computer Engineering PRIMARY
The best of ECE undergraduate research