Eye Gaze Correction via Pursuit Vector Analysis
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Eye gaze tracking systems are prone to noise and imprecision, leading to errors in user interface item selection, as they often misinterpret the user's intent due to angular errors in gaze detection, limiting their acceptance in user interfaces.
Innovation Solution
A calibration method that does not require specific on-screen content, allowing for calibration at any time and triggered by user input or error monitoring, involves taking a snapshot of the user's gaze area, animating it to determine a pursuit vector, and calculating a correction factor to adjust the reported gaze location accurately.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If traditional eye gaze calibration procedures are used, then measurement precision is improved, but device complexity and calibration time increase
Solution Approach 1:
The system performs calibration automatically without requiring specific on-screen content or user interaction. The calibration process is triggered by error monitoring and uses pursuit vectors derived from normal user gaze movements to self-correct accuracy issues, eliminating the need for complex manual calibration procedures
Solution Approach 2:
The system dynamically adjusts calibration parameters by calculating pursuit vectors from gaze movement data. Instead of fixed calibration values, the system continuously updates correction factors based on observed pursuit movements, adapting to changing gaze patterns and improving accuracy without additional complexity
2Measurement precision
If traditional eye gaze calibration procedures are used, then measurement precision is improved, but loss of time increases
Solution Approach 1:
The system performs calibration in the background during normal operation without requiring a separate calibration phase. By continuously monitoring gaze errors and performing incremental calibration, the system prepares correction data preliminarily, eliminating time loss associated with explicit calibration procedures
Solution Approach 2:
The calibration process operates continuously alongside normal gaze tracking, rather than requiring interruption for separate calibration sessions. The system maintains continuous correction factor updates based on ongoing pursuit vector calculations, ensuring accuracy without time loss
3Ease of operation
If eye gaze tracking is used for item selection, then ease of operation is improved, but reliability worsens due to noise and imprecision
Solution Approach 1:
The system implements feedback by monitoring gaze errors and using pursuit vector analysis to detect when calibration is needed. This feedback loop continuously adjusts correction factors to maintain reliable item selection, reducing errors caused by noise and imprecision while preserving ease of operation
Solution Approach 2:
The system replaces traditional mechanical calibration adjustments with computational pursuit vector analysis. By substituting mathematical correction based on gaze movement patterns for physical calibration mechanisms, the system achieves higher reliability without compromising operational ease
Data Source
AI summary
Representative embodiments disclose mechanisms for calibrating an eye gaze selection system. When the calibration is triggered, a snapshot of an area around the current user's gaze point is taken. The snapshot area is then animated to cause motion of the snapshot area. As the snapshot is animated, the user's gaze will naturally track the thing the user was focusing on. This creates an eye tracking vector with a magnitude and direction. The magnitude and direction of the eye tracking vector can then be used to calculate a correction factor for the current user's gaze point. Calibration can be triggered manually by the user or based on some criteria such as error rates in item selection by the user.