Gaze Placement Determination Using Corneal Reflection Vectors
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current gaze estimation techniques face challenges in precision and comfort due to their complex hardware configurations and inability to accurately estimate gaze placement without intrusive methods, especially when the user's head moves naturally.
Innovation Solution
A method and device that determine gaze placement using ocular images by calculating characteristic vectors from reference speckles and pupil features, employing a gaze estimation model and support vector regression for head movement compensation, allowing for non-intrusive and accurate gaze estimation with reduced hardware requirements.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If complex hardware configurations are used for gaze estimation, then measurement precision may be improved, but device complexity increases
Solution Approach 1:
The patent extracts and utilizes naturally occurring optical phenomena (corneal reflections from environmental light sources) rather than requiring specialized hardware. By taking out the dependency on complex dedicated illumination systems and using instead the reflections from existing light sources, the system achieves gaze estimation with simpler hardware while maintaining measurement precision through mathematical modeling of the reflection geometry
Solution Approach 2:
The system employs a universal approach by using a single imaging device that serves multiple functions: capturing the eye image, detecting the pupil position, and identifying corneal reflection points. This multi-functional use of standard camera hardware eliminates the need for specialized gaze-tracking hardware components, reducing device complexity while maintaining gaze estimation capability
2Measurement precision
If intrusive methods are used to ensure accurate gaze estimation, then measurement precision is improved, but ease of operation deteriorates
Solution Approach 1:
The system employs self-service by utilizing the eye's own optical properties (corneal reflection) to provide the necessary measurement signals. The cornea naturally reflects environmental light sources, and the system simply captures and analyzes these self-generated optical signals, eliminating the need for intrusive interventions while maintaining measurement precision through the inherent optical characteristics of the eye
Solution Approach 2:
The patent changes the measurement parameters from requiring controlled eye movements or head positioning to analyzing the geometric relationships between pupil center and corneal reflection points. By changing to these passive geometric parameters that can be extracted from natural eye states, the system achieves accurate gaze estimation without requiring users to perform specific actions, thereby improving ease of operation
3Ease of operation
If the system does not compensate for head movements, then ease of operation is improved, but measurement precision deteriorates
Solution Approach 1:
The system transitions from two-dimensional image plane coordinates to three-dimensional spatial relationships by incorporating depth information through the known geometry of the corneal reflection. By using the distance between the camera and the eye, along with the positions of the pupil center and reflection points in the image plane, the system calculates the actual gaze direction in 3D space, thereby compensating for head movements without restricting natural operation
Solution Approach 2:
The system implements feedback by continuously monitoring the relative positions of the pupil center and corneal reflection points, and using this information to dynamically calculate and adjust the gaze placement. The mathematical model incorporates real-time geometric relationships to compensate for head position changes, providing accurate gaze estimation that adapts to natural head movements while maintaining measurement precision
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
The solution provides accurate and comfortable gaze estimation that tolerates natural head movements, reducing hardware complexity and improving user experience while maintaining high precision.
Implementation Method 1
a first reference speckle to a center of a pupil, wherein the first reference speckle is formed by a first reference source in the ocular image
Data Source
AI summary
Embodiments of the present disclosure provide a method and a device for determining a gaze placement and a computer readable storage medium. The method comprises acquiring an ocular image of a subject; determining at least one gaze characteristic vector based on an ocular image; determining a gaze placement of the subject, according to a gaze estimation model and the at least one characteristic vector. The characteristic vector comprises at least one of: a first characteristic vector from a first reference speckle center to a pupil center, wherein the first reference speckle is formed by a first reference source in the ocular image; a second characteristic vector from the pupil center to a second reference speckle center, wherein the second reference speckle is formed by a second reference source in the ocular image; a third characteristic vector from the second reference speckle center to the first reference speckle center.