Single Camera Eye Tracking Using Corneal Reflection Parallax
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing methods for determining the distance to a subject in a real space using two cameras are costly and require complex calibration, prompting the need for a more efficient approach that can calculate distance using a single imaging device.
Innovation Solution
A position determination system that uses one imaging device to capture images of both eyes of a user, processing cornea images to calculate the position of an object point in real space by analyzing the parallax between the images formed on the cornea of each eye, allowing for accurate distance determination without the need for multiple imaging devices.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If two cameras are used to calculate distance to a subject, then measurement precision is improved, but device complexity and calibration requirements increase
Solution Approach 1:
The patent segments the measurement task into two parts: (1) capturing cornea images reflecting the subject, and (2) calculating distance from corneal reflection geometry. This allows using a single camera instead of two, reducing device complexity while maintaining measurement precision through the unique geometric properties of corneal reflections
Solution Approach 2:
The cornea acts as an intermediary optical element that reflects the subject image. By capturing the corneal reflection rather than directly imaging the subject, the system can determine distance from the geometric relationship between the camera, cornea, and reflected subject, eliminating the need for multiple cameras while preserving measurement accuracy
2Measurement precision
If two cameras are used for stereo imaging, then distance calculation accuracy is improved, but calibration complexity increases
Solution Approach 1:
The cornea's fixed anatomical position and known optical properties serve as a self-calibrating reference. The system uses the cornea's inherent geometry (radius of curvature, position relative to the camera) to automatically determine distance without requiring external calibration procedures or multiple camera calibrations, thereby maintaining accuracy while eliminating calibration complexity
3Device complexity
If a single imaging device is used, then device cost and complexity are reduced, but measurement precision deteriorates
Solution Approach 1:
The patent changes the measurement parameter from direct subject imaging to corneal reflection imaging. By measuring the position and size of the subject's reflection in the cornea rather than imaging the subject directly, the system can calculate distance with high precision using a single camera, as the corneal reflection provides geometric constraints that enable accurate distance determination
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 reduces costs and calibration efforts while accurately determining object positions in real space, enabling efficient and precise distance measurement with a single imaging device.
Implementation Method 1
a cornea image of a first eye and a cornea image of a second eye of a user who is viewing the object point, in a captured image generated by imaging the first and second eyes of the user by an imaging device
Implementation Method 2
processing cornea images to calculate the position of an object point in real space by analyzing the parallax between the images formed on the cornea of each eye
Data Source
AI summary
A controller is configured to determine a position of an object point in a real space, based on a cornea image of a first eye and a cornea image of a second eye of a user who is viewing the object point, in a captured image generated by imaging the first and second eyes of the user by an imaging device.