Infrared Eye Tracking for AR Head-Up Displays
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
High-performance head-up displays, such as augmented reality HUDs, face challenges in projecting complex images without distortion, particularly in tracking the user's eye location to maintain undistorted views and fused virtual and real image registration as the user moves.
Innovation Solution
A system incorporating an infrared light source and camera, with a controller that filters infrared noise using pulse illumination and optical spatial filters, tracks the user's eye location by reflecting infrared light off the windshield and performing frequency domain analysis to demodulate noise, enabling dynamic distortion compensation and improved image alignment.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If infrared illumination is used to track eye location, then tracking capability is improved, but infrared noise interferes with image quality
Solution Approach 1:
The system uses periodic pulsed infrared illumination at a known frequency to illuminate the scene. By pulsing the light source periodically rather than continuously, the system creates a time-varying signal that can be distinguished from constant infrared noise through frequency domain analysis, thereby improving eye location tracking precision while managing infrared noise interference
Solution Approach 2:
The system performs frequency domain analysis on the captured infrared images to identify and isolate the periodic illumination signal from the noise. By analyzing the frequency characteristics of the received signal and comparing it to the known illumination frequency, the system can extract accurate eye location information while filtering out infrared noise through feedback-based signal processing
2Adaptability or versatility
If the user moves during viewing, then viewing flexibility is improved, but image distortion increases
Solution Approach 1:
The system dynamically adjusts the projected image parameters in real-time based on continuously tracked eye location data. As the user moves, the system updates the projection transformation to compensate for changes in viewing angle and position, maintaining accurate image registration and undistorted views throughout the eyebox region
Solution Approach 2:
The system uses real-time eye location feedback from the infrared camera to continuously adjust the image projection parameters. This closed-loop control ensures that as the user moves within the viewing zone, the projected image automatically compensates for positional changes, maintaining proper alignment and reducing distortion throughout the viewing session
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 solution effectively reduces motion parallax and maintains augmented reality image registration by accurately tracking the user's eye position, allowing for real-time adjustments to image rendering and maintaining an undistorted view as the user moves.
Implementation Method 1
An infrared camera within the housing captures a sequence of images including infrared noise combined with a captured reflection of the pulse illumination
Implementation Method 2
The infrared light source can be configured to reflect off a windshield of a vehicle including the head-up display to an eyebox defining an expected viewing area of a user
Implementation Method 3
An optical spatial filter can be operably coupled to the infrared camera, where the optical spatial filter is configured to block light external to the eyebox
Implementation Method 4
A bandpass filter can be operably coupled to the infrared camera, where the bandpass filter is configured to filter non-infrared wavelengths
Data Source
AI summary
A system includes an infrared light source, an infrared camera, and a controller. The controller includes processing circuitry configured to control a pulse illumination of the infrared light source, capture a sequence of images from the infrared camera including infrared noise combined with a captured reflection of the pulse illumination, filter the infrared noise from the sequence of images based on a comparison of the captured reflection of the pulse illumination to an expected result of the pulse illumination, and track a feature of interest in the sequence of images after filtering of the infrared noise.