Gaze Point Positioning via Zone-Based Mapping in VR
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current eye tracking systems in virtual reality have low positioning precision due to the assumption of global pupil movement, leading to poor user experience.
Innovation Solution
A method that captures real-time eye images, determines pupil center points, and divides the target screen into zones to establish specific mapping ratios for each zone, accounting for the pupil's movement on a spherical surface, improving gaze point positioning accuracy.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Device complexity
If a global mapping model is used for eye tracking, then the system complexity is reduced, but the gaze point positioning precision deteriorates
Solution Approach 1:
The patent divides the display screen into multiple zones and creates separate mapping models for each zone. This segmentation allows the system to account for local variations in pupil movement characteristics across different regions, improving positioning precision without requiring a completely complex global model. Each zone has its own mapping ratio calculated from reference eye images, enabling localized optimization.
Solution Approach 2:
The patent applies local quality by treating different zones of the screen with different mapping characteristics. Instead of using a uniform global mapping model, the system calculates specific mapping ratios for each zone based on local pupil movement patterns. This allows the mapping model to adapt to local variations in eye movement, improving overall positioning accuracy.
2Measurement precision
If a zone-based mapping approach is used, then the gaze point positioning precision is improved, but the device complexity increases
Solution Approach 1:
The screen is divided into multiple zones, and the patent establishes separate mapping models for each zone. This segmentation enables the system to capture local variations in pupil movement while maintaining a manageable structure. The complexity is distributed across zones rather than requiring a single complex global model.
Solution Approach 2:
The patent performs preliminary calibration by capturing reference eye images at multiple preset points across different zones before actual use. This preliminary action establishes the mapping ratios for each zone in advance, so that during actual gaze tracking, the system can simply apply the pre-calculated mappings without performing complex real-time calculations, thus reducing operational complexity.
3Measurement precision
If reference eye images are captured at multiple preset points, then the mapping accuracy is improved, but the time required for calibration increases
Solution Approach 1:
The patent captures reference eye images at multiple preset points during an initial calibration phase. This preliminary action establishes the mapping relationships for different zones in advance. Although this requires additional time during calibration, it enables fast and accurate gaze tracking during actual use by pre-computing the mapping ratios.
Solution Approach 2:
The patent combines multiple reference eye images captured at different preset points to establish comprehensive mapping models for all zones. By merging the information from these multiple images, the system creates accurate zone-specific mappings that improve overall mapping accuracy while consolidating the calibration process into a structured sequence.
Data Source
AI summary
Disclosed herein includes a method, an apparatus, a display device and storage medium storing computer executable instructions for positioning a gaze point. The method for obtaining a gaze point in a display device may comprise capturing a real time eye image, obtaining a real time pupil center point from the real time eye image, determining a gaze target zone based on the real time pupil center point and obtaining a gaze point on a target screen generated by the display device based on the real time pupil center point and a mapping ratio for the gaze target zone.