Related Experiment Video
Updated: Sep 22, 2025
07:48
Eye Tracking During A Complex Aviation Task For Insights Into Information Processing
Published on: April 4, 2025
607
Stable Gaze Tracking with Filtering Based on Internet of Things
1School of Computer and Communication Engineering, University of Science and Technology Beijing, Beijing 100083, China.
Sensors (Basel, Switzerland)
|May 20, 2022
Summary
This study enhances gaze tracking accuracy by using a Kalman filter and a correlation filter. The new method significantly reduces estimation errors and speeds up pupil detection for more stable gaze tracking.
Area of Science:
- Human-Computer Interaction
- Computer Vision
- Signal Processing
Background:
- Gaze tracking is crucial for Internet of Things applications.
- Noise interference causes fluctuations in estimated points of regard (PORs).
- Existing methods struggle with stable gaze parameter estimation.
Purpose of the Study:
- To improve the performance and stability of gaze tracking systems.
- To reduce the fluctuation range of estimated PORs.
- To adapt filtering based on gaze motion for enhanced accuracy.
Main Methods:
- Introduced a Kalman filter (KF) to filter gaze parameters.
- Designed measurement noise that varies with gaze speed.
- Employed a correlation filter-based method for rapid pupil localization.
Main Results:
- Variance of estimation error decreased by 73.83%.
- Size of extracted pupil image reduced by 93.75%.
- Pupil extraction speed increased by 1.84 times.
Conclusions:
- The proposed method offers more stable and accurate gaze tracking.
- The adaptive filtering and efficient tracking enhance system performance.
- The algorithm is applicable to various eye camera-based gaze trackers.
