Related Experiment Video
Updated: Jan 1, 2026
12:39
A Methodology for Capturing Joint Visual Attention Using Mobile Eye-Trackers
Published on: January 18, 2020
8.1K
A probabilistic approach to online eye gaze tracking without explicit personal calibration
Summary
This study introduces a novel probabilistic eye gaze tracking system that eliminates the need for personal calibration. The system achieves high accuracy by gradually improving eye and gaze estimation through natural interaction, enhancing human-computer interfaces.
Area of Science:
- Computer Science
- Human-Computer Interaction
- Biomedical Engineering
Background:
- Traditional eye gaze tracking necessitates explicit personal calibration, which is inconvenient for natural human-computer interaction.
- Current methods deterministically estimate eye parameters using known gaze points, limiting adaptability.
Purpose of the Study:
- To propose a novel probabilistic eye gaze tracking system that operates without explicit personal calibration.
- To enable natural human-computer interaction by removing the calibration barrier.
Main Methods:
- Developed a probabilistic eye gaze tracking system estimating probability distributions of eye parameters and gaze.
- Utilized an incremental learning framework for gradual improvement of estimations during natural user interaction.
- Eliminated the need for pre-use personal calibration.
Main Results:
- The proposed system achieved an accuracy of less than 3 degrees.
- Demonstrated effectiveness across different individuals without requiring explicit personal calibration.
- Showcased the feasibility of gaze tracking in natural interaction settings.
Conclusions:
- The developed system offers a viable alternative to traditional eye gaze tracking methods.
- Probabilistic estimation and incremental learning enable accurate, calibration-free gaze tracking.
- This approach significantly enhances the naturalness and convenience of human-computer interaction systems.
