Eye Tracking Recalibration via Glint-Pupil Vectors
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing head-mounted displays face challenges in accurately tracking eye movements due to movements of the device or user's head, leading to errors in gaze position determination, which can render gaze-dependent interfaces unusable.
Innovation Solution
A method and system for eye tracking that involves generating infrared light, scanning and detecting reflections, identifying glints, determining glint center positions, transforming these positions to gaze positions, and recalibrating the glint space using glint-pupil vectors to account for device and head movements, ensuring accurate gaze estimation.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If eye tracking is implemented using infrared light reflections, then gaze position can be determined, but device or head movements cause errors in gaze position determination
Solution Approach 1:
The system continuously monitors eye position using infrared reflections and uses this feedback to dynamically adjust the displayed content position, ensuring that content remains aligned with the user's actual gaze direction even during head movements
Solution Approach 2:
The patent introduces an intermediary computational layer that processes raw eye tracking data and compensates for movement artifacts before determining final gaze position, acting as a mediator between the physical eye movement and the interpreted gaze direction
2Ease of operation
If the display content is adjusted based on gaze position, then user interface usability improves, but errors in gaze position make the interface unusable
Solution Approach 1:
The system preemptively compensates for expected errors by applying correction factors to the gaze position calculation before using it to control the interface, preventing the harmful effect of measurement errors from manifesting in the user interface
Solution Approach 2:
The patent dynamically adjusts parameters such as the sensitivity of gaze detection and the threshold for triggering interface actions based on detected movement conditions, modifying system behavior to maintain usability despite varying measurement accuracy
Applied Scientific Principles
This section explains which scientific principles are used to turn an abstract innovation direction into a practical engineering solution.
Function Achieved in This Case
This approach provides reliable eye tracking by minimizing errors caused by device or head movements, enabling accurate gaze position determination and improving the usability of gaze-dependent interfaces in head-mounted displays.
Implementation Method 1
generating infrared light
Implementation Method 2
detecting reflections of the infrared light from the eye
Data Source
AI summary
A method of tracking an eye of a user includes generating an infrared light, scanning the infrared light over the eye, and detecting reflections of the infrared light from the eye over an eye tracking period. A plurality of glints is identified from the reflections of the infrared light detected. A glint center position of each glint in a glint space is determined and transformed to a gaze position in a display space. At least once during the eye tracking period, an image of the eye is reconstructed from a portion of the reflections of the infrared light detected. A pupil is detected from the image, and a pupil center position is determined. A glint-pupil vector is determined from the pupil center position and the glint center position of at least one glint corresponding in space to the pupil. The glint space is recalibrated based on the glint-pupil vector.