Eye Tracker Calibration Using Stored Profile Matching
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
The calibration process for eye-tracking devices is time-consuming and tedious, requiring users to look at multiple reference points to establish an accurate calibration profile, which can be improved for a better user experience.
Innovation Solution
A method and device that reduce calibration time by displaying fewer reference points, comparing acquired eye data to stored sets, and loading a matching calibration setting, with options to repeat the process if no match is found, using gaze direction offsets, interocular distance, and inter-pupillary distance for precise matching within a predetermined threshold.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If the conventional calibration process displays multiple reference points and collects eye data for each, then the calibration precision is improved, but the calibration time increases significantly
Solution Approach 1:
The system performs preliminary actions by storing calibration data from multiple users in advance. When a new user calibrates, the system pre-computes similarity comparisons between the new user's eye characteristics and stored calibration datasets, allowing rapid identification of a matching calibration profile without requiring extensive real-time calibration points
Solution Approach 2:
The system creates copies of calibration profiles from other users and applies them to the current user. By copying previously calibrated eye characteristics (pupil size, corneal radius, gaze patterns) from similar users, the system achieves acceptable calibration precision with significantly fewer reference points displayed to the user
2Reliability
If the calibration process collects comprehensive eye data for multiple reference points, then the calibration profile accuracy is improved, but the complexity of the calibration process increases
Solution Approach 1:
The system extracts only the most critical eye characteristics (pupil size, corneal radius, inter-pupillary distance) needed for calibration rather than collecting comprehensive eye data for multiple reference points. By taking out only the essential parameters, the system maintains calibration reliability while reducing process complexity
Solution Approach 2:
The system creates universal calibration profiles that can be applied across multiple users with similar eye characteristics. By developing calibration settings that work for groups of users rather than individualizing every parameter for each user, the system reduces calibration complexity while maintaining sufficient accuracy for reliable eye tracking
Data Source
Figure 1a~1b
Figure 2a~2b
Figure 3
AI summary
Disclosed is a method for calibrating an eye-tracking device to suit a user of the eye-tracking device (600), wherein a calibration setting of the eye-tracking device (600) associated with a user is calculated based on acquired eye data of the user when looking at a set of reference points (1). The method comprises displaying (202) a reference point of the set to the user; acquiring (204), by means of at least one camera (10) of the eye-tracking device (600), eye data for at least one of the eyes of the user when looking at the reference point; comparing (206) the acquired eye data to stored eye data sets related to the reference point, wherein each of the stored eye data sets is associated with a calibration setting of the eye-tracking device; and if the acquired eye data matches one of the stored eye data sets, abandoning (208) the calibration process and loading (210) the calibration setting associated with the matching stored eye data.