Gaze Analysis Using Iris Sclera Geometry
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current gaze analysis technologies face issues such as safety concerns due to infrared ray proximity, high resolution lens requirements, and sensitivity to light illumination, as well as complexity in model matching and verification processes.
Innovation Solution
A gaze analysis method and apparatus that calculate a gazing angle based on iris and sclera information of the left and right eyes, determining whether the object is gazed by assessing the differential value between focusing and spacing distances, without needing high resolution lenses or infrared rays, thus reducing computational complexity and costs.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If cornea reflection manner with infrared ray is used to determine vision direction, then gaze analysis accuracy is improved, but safety issues arise due to infrared ray proximity to eyes
Solution Approach 1:
The patent converts the harmful infrared radiation into a beneficial approach by using visible light reflection from the cornea. Instead of using infrared rays that pose safety risks, the system utilizes the natural reflection properties of the cornea under visible light, thereby maintaining measurement precision while eliminating safety concerns.
Solution Approach 2:
The patent replaces expensive high-resolution infrared lenses with standard visible light cameras and simple reflective markers. By using readily available visible light sources and standard imaging sensors, the system achieves comparable or superior accuracy without the need for costly infrared optics.
2Measurement precision
If cornea reflection manner is used for gaze analysis, then gaze direction can be determined, but high resolution lens is required which increases cost
Solution Approach 1:
The patent creates a simplified optical model that copies the essential reflection behavior of the cornea without requiring complex high-resolution optics. By modeling the corneal reflection geometry and using standard camera sensors, the system replicates the functionality of expensive infrared setups with much cheaper components.
Solution Approach 2:
The system uses inexpensive visible light cameras and simple reflective markers instead of expensive infrared lenses and specialized sensors. The visible light source and standard imaging components are significantly cheaper than infrared equipment while achieving comparable measurement accuracy.
3Measurement precision
If iris center detection with eye model matching is used, then gaze analysis can be performed, but complex model matching and storage is required
Solution Approach 1:
The patent extracts only the essential geometric features needed for gaze analysis - specifically the corneal reflection position and pupil center - without requiring full eye model matching. By focusing on these key extracted features and their spatial relationships, the system achieves accurate gaze determination while eliminating the complexity of storing and processing complete eye models.
Solution Approach 2:
Instead of matching the captured eye image against pre-stored eye models to find the best fit, the patent inverts the approach by directly calculating gaze direction from geometric relationships of detected features. This eliminates the need for model storage and matching computations, replacing them with straightforward geometric calculations.
4Measurement precision
If pupil image recording with complex correction and verification is used, then gaze determination is achieved, but computation complexity increases
Solution Approach 1:
The patent extracts only the critical geometric parameters - corneal reflection position and pupil center coordinates - without requiring full pupil image recording and complex correction procedures. By working directly with these extracted geometric features, the system eliminates unnecessary computational steps while maintaining measurement accuracy.
Solution Approach 2:
The patent performs preliminary geometric calibration to establish the relationship between camera coordinates and eye coordinates once, then uses these pre-determined transformation parameters for rapid gaze calculation. This preliminary action eliminates the need for complex real-time correction and verification computations during actual gaze measurement.
Data Source
AI summary
A gaze analysis method has following steps. When detecting at least one face in an image, according to a face angle and an eye angle, a gazing angle is calculated, and whether the gazing angle falls within a first range is determined, wherein the eye angle is obtained according to information of irises and scleras of left and right eyes. A differential value between a focusing distance and a spacing distance from the face to the object falls within a second range is determined if the gazing angle falls within the first range, and a detecting result whether the differential value falls within the second range is used to determine whether the object is gazed, wherein the focusing distance is determined according to the information of the irises and scleras of the left and right eyes.