Reference Gaze Calibration From Pupil Size Changes
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current eye tracking systems require external inputs or cumbersome methods to accurately determine where a user is looking, leading to inefficiencies and reduced accuracy.
Innovation Solution
An eye tracking system that determines reference gaze data by analyzing pupil size changes in response to local luminance level differences, without the need for external inputs, using processing circuitry to capture eye images and scene information to update gaze calibration settings.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If external input devices (mouse, keyboard, microphone, click button) are used to determine user gaze, then interaction accuracy is improved, but operation speed and user convenience deteriorate
Solution Approach 1:
The system uses the user's own physiological response (pupil size change) to determine gaze, eliminating the need for external input devices. The pupil naturally reacts to luminance changes in the viewed scene, providing automatic gaze confirmation without requiring the user to perform additional actions.
Solution Approach 2:
The system monitors pupil size changes as feedback to determine whether the user is actually looking at the calibration target or interactable object. This physiological feedback loop allows the system to automatically verify gaze accuracy without external inputs.
2Device complexity
If minimum gazing time or blink detection is used to determine user attention, then system complexity is reduced, but gaze determination accuracy deteriorates
Solution Approach 1:
The system replaces mechanical timing mechanisms and blink detection algorithms with a physiological measurement approach. By measuring pupil size changes in response to luminance variations, the system achieves more accurate gaze determination without relying on arbitrary time thresholds or complex behavioral analysis.
3Measurement precision
If traditional calibration methods requiring external inputs are used, then calibration accuracy is improved, but calibration time and productivity deteriorate
Solution Approach 1:
During calibration, the user's pupil naturally responds to the luminance differences between the calibration target and surrounding areas. The system captures this automatic physiological response to determine calibration points, eliminating the need for users to manually indicate each calibration point with external devices.
Solution Approach 2:
The system prepares the calibration scene with specific luminance contrasts before the user looks at it. The calibration target is designed to produce detectable pupil size changes, so when the user naturally looks at the target, the calibration data is automatically captured without requiring additional user actions.
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
Enables accurate and efficient gaze determination without external inputs, allowing for improved interaction and calibration, and supports applications like foveated rendering.
Implementation Method 1
determine, based on the first eye image, a first pupil size... determine, based on the second eye image, a second pupil size... determine a pupil size change between the first pupil size and the second pupil size, the pupil size change indicating that the user is looking at the second region
Data Source
Figure 1
Figure 2A~2B
Figure 3
AI summary
The invention is related to an eye tracking system for determining reference gaze data of a user in a scene exposing a pupil of the user. The eye tracking system comprising processing circuitry configured to obtain a first eye image comprising the pupil of the user, the first eye image being captured during a first time period; determine, based on the first eye image, a first pupil size; obtain a second eye image comprising the pupil of the user, the second eye image being captured during a second time period; determine, based on the second eye image, a second pupil size; obtain scene information of the scene exposing the pupil of the user, the scene information comprising at least the first luminance level, the second luminance level and spatial information of the second region during the second time period; determine a pupil size change between the first pupil size and the second pupil size, the pupil size change indicating that the user is looking at the second region; determine reference gaze data of the user during the second time period, if the pupil size change is larger than a pupil size change threshold. The invention further relates to a head-mounted device, a method, a computer program and a carrier.