Smooth Pursuit Gaze Tracking Calibration
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional eye tracking calibration methods can be distracting and annoying for users, as they require explicit attention to a known object, which may not accurately reflect natural gaze behavior.
Innovation Solution
A method that calibrates gaze by displaying a moving object on a screen and tracking eye movements, identifying patterns indicative of smooth pursuit, allowing for unobtrusive calibration and determining if content is viewed by associating eye position with the object's position on the screen.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If conventional calibration methods require explicit user attention to known objects, then calibration accuracy can be achieved, but user experience deteriorates due to distraction and annoyance
Solution Approach 1:
The system performs calibration automatically by detecting smooth pursuit eye movements during normal content viewing, without requiring user initiation or explicit cooperation. The calibration process serves itself by utilizing naturally occurring eye movements that occur when users watch moving objects in videos or games.
Solution Approach 2:
Moving objects in displayed content serve as intermediaries that naturally attract and hold user attention, enabling the system to capture calibration data during authentic viewing behavior. These objects mediate between the user's natural viewing habits and the calibration objective.
2Measurement precision
If explicit calibration procedures are implemented, then gaze tracking accuracy improves, but calibration time increases due to user interaction requirements
Solution Approach 1:
The system performs calibration actions preliminarily and continuously in the background during normal viewing, so that when gaze tracking is needed, calibration is already complete or continuously updated, eliminating dedicated calibration time requirements.
Solution Approach 2:
Calibration is performed continuously during normal content consumption rather than as a discrete separate step. The useful action of watching content and the calibration action occur simultaneously, eliminating time loss.
3Ease of operation
If traditional calibration requires user cooperation and attention, then calibration can be performed, but natural gaze behavior is compromised
Solution Approach 1:
The system captures natural gaze behavior by detecting smooth pursuit movements that occur spontaneously during content viewing, without requiring user awareness or cooperation. The calibration process exploits the user's own natural eye movement patterns.
Solution Approach 2:
Instead of requiring users to deliberately follow calibration targets, the system inverts the approach by detecting natural eye movements during normal viewing and using those for calibration, thereby capturing authentic gaze behavior.
Data Source
AI summary
A system and method to calibrate a gaze point on a screen of an eye as the eye watches an object move on the screen. A movement of an eye may be tracked in a series of images, and a pattern of the movement of the eye may be detected. Detection of a pattern of eye movements that is associated with watching the moving object on a screen may be an indication that the eye was watching the object at the time that the pattern of eye movements was detected. The identified gaze point may be part of a calibration for purpose of gaze tracking. The detected pattern may be included in an indication that the viewer looked at the moving object.