Gaze Estimation Method Using Analysis of Electrooculogram Signals and Kinect Sensor

Area of Science:

• Biomedical Engineering
• Human-Computer Interaction
• Assistive Technology

Background:

• Existing gaze estimation systems are crucial for communication among individuals with severe disabilities.
• Previous eye tracking methods using electrooculogram (EOG) signals were limited by suboptimal accuracy.
• There is a need for enhanced accuracy in EOG-based eye tracking for practical assistive communication.

Purpose of the Study:

• To identify specific electrooculogram (EOG) signal components strongly correlated with eye movement variations.
• To evaluate the feasibility of improving gaze estimation accuracy by integrating EOG signals with Kinect sensor data.
• To advance communication methods for individuals unable to use conventional speech or gesture.

Main Methods:

• Conducted experiments involving object observation solely through eye movements.
• Performed experiments where participants used combined eye and face movements to view objects.
• Utilized electrooculogram (EOG) signals and data from a Kinect sensor during experimental tasks.

Main Results:

• Identified specific EOG signal characteristics that correlate with changes in eye movement.
• Demonstrated that the integration of EOG signals and Kinect sensor data enhances gaze estimation potential.
• Experimental results indicate a viable pathway for improved eye tracking accuracy.

Conclusions:

• The study confirms the potential of using electrooculogram (EOG) signals for gaze estimation.
• Combining EOG signals with Kinect sensor data offers a promising approach to increase eye tracking accuracy.
• This research contributes to developing more effective communication tools for severely disabled individuals.