Gaze Point Mapping Function Determination via Segmented Parameter Solving
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
The existing eye-tracking technology requires users to stare at multiple calibration points to determine the gaze point mapping function, resulting in a high workload and unfavorable user experience.
Innovation Solution
A method and device that divide the parameter solving process into two stages, where parameters are first solved based on the user's eye image and gaze point information, and then secondary parameters are determined using solutions from multiple users, reducing the need for multiple calibration points.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If multiple calibration points are used to determine the gaze point mapping function, then the accuracy of gaze point determination is improved, but the user workload increases
Solution Approach 1:
The parameter solving process is divided into two distinct stages: first solving for primary parameters (pupil center coordinates, light reflection point coordinates) from the user's own eye images, then solving for secondary parameters (gaze point mapping function parameters) using aggregated data from multiple users. This segmentation allows the system to maintain high accuracy while reducing the calibration burden on individual users.
Solution Approach 2:
The system performs preliminary actions by collecting and processing eye image data from multiple users in advance to establish reference models for the gaze point mapping function. This preliminary data collection and model establishment reduces the calibration requirements for subsequent individual users, as the system can leverage pre-computed parameter relationships.
2Measurement precision
If all parameters are solved from individual user's eye image, then the accuracy for that user is improved, but the calibration time increases
Solution Approach 1:
The calibration process is segmented into essential parameters that must be solved for each user (pupil center, light reflection point) and secondary parameters that can be derived from group data (gaze point mapping function parameters). This segmentation significantly reduces the time required for individual calibration while preserving user-specific accuracy for the essential parameters.
Solution Approach 2:
The system uses data from multiple users to create a universal model for the gaze point mapping function that can be applied across different users. This multi-functional approach allows the system to leverage collective data to reduce individual calibration time while maintaining user-specific accuracy where needed.
3Reliability
If multiple calibration points are required, then the reliability of gaze tracking is improved, but the user experience deteriorates
Solution Approach 1:
The system performs preliminary data collection from multiple users to establish reliable reference models for the gaze point mapping function before individual calibration. This preliminary action ensures that the system has sufficient data to maintain reliable gaze tracking while minimizing the calibration burden on individual users, thereby improving overall user experience.
Solution Approach 2:
The system enables individual users to complete calibration with fewer points by leveraging pre-established models from group data. The system essentially serves itself by using aggregated user data to create templates that facilitate faster, more convenient calibration for subsequent users while maintaining tracking reliability.
Data Source
AI summary
A method for determining a gaze point mapping function includes that: all parameters to be solved in a gaze point mapping function are combined to obtain a parameter vector of the gaze point mapping function, and the parameters to be firstly solved and parameters to be secondly solved corresponding to the parameter vector are determined; the parameters to be firstly solved for a first user are solved according to an eye image of the first user and corresponding gaze point information; solutions of the parameters to be secondly solved for the first user are determined according to the parameter vector which is solved by each of multiple second users respectively; and the gaze point mapping function of the first user is determined according to solutions of the parameters to be firstly solved for the first user and the solutions of the parameters to be secondly solved for the first user.