Gaze Tracking Parameter Determination via Vector Decomposition
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing gaze tracking devices face complexity in gaze estimation and glint prediction due to complex calculation algorithms and issues with light source matching, leading to inaccurate gaze direction determination.
Innovation Solution
A method and device that acquire and process images using a preset model to obtain first and second vectors, determining parameters such as glint position and gaze direction by decomposing image data into reference and viewing angle spaces, simplifying the estimation algorithm and increasing processing speed.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If a 3D approximately spherical model with multiple cameras and light sources is used for gaze estimation, then measurement precision is improved, but device complexity increases and calculation becomes complex
Solution Approach 1:
The patent segments the gaze estimation problem into two independent sub-problems: (1) glint position detection using a simplified 2D model, and (2) gaze direction determination using the detected glint positions. This segmentation allows each sub-problem to be solved with appropriate simplifications, reducing overall system complexity while maintaining accuracy.
Solution Approach 2:
The patent extracts the glint position detection as a separate preprocessing step before gaze estimation. By extracting and solving the glint detection problem independently using a 2D model, the complex 3D gaze estimation is simplified to a more manageable task based on already-detected glint positions.
2Measurement precision
If complex gaze estimation algorithms are used, then measurement precision is improved, but processing speed decreases
Solution Approach 1:
The patent performs preliminary detection of glint positions using a simplified 2D model before conducting the full gaze estimation. This preliminary action prepares the data in advance, allowing the subsequent gaze estimation to proceed more efficiently with pre-processed information, thereby improving processing speed without sacrificing accuracy.
Solution Approach 2:
The algorithm is segmented into distinct processing stages: glint detection phase and gaze estimation phase. Each phase uses appropriately optimized methods, with the first phase being computationally lighter, thus improving overall processing throughput while maintaining measurement precision in the second phase.
3Measurement precision
If multiple light sources are used for comprehensive gaze tracking, then measurement precision is improved, but glint matching reliability decreases due to camera position limitations
Solution Approach 1:
The patent applies different quality requirements to different parts of the system: the 2D glint detection model uses simplified assumptions appropriate for its local task, while the 3D gaze estimation uses more rigorous models where needed. This local quality approach allows the system to tolerate certain limitations in individual components while maintaining overall reliability.
Solution Approach 2:
The patent introduces an intermediary step of glint position detection that bridges the gap between light source emission and gaze direction determination. This intermediary process handles the matching complexity by detecting glints in image space first, then using those detected positions as reliable inputs for the subsequent 3D gaze estimation, thereby improving overall matching reliability.
Data Source
AI summary
The disclosure discloses a method and device for determining parameter for a gaze tracking device. The method includes that: an image to be detected and a characteristic matrix of the image to be detected are acquired; the image to be detected is processed according to a preset model to obtain a first vector and second vector of the image to be detected; and a parameter of the image to be detected is determined according to the first vector and the second vector, the parameter of the image to be detected including at least one of: a position of a glint in the image to be detected and a gaze direction of eyes in the image to be detected.