Gaze Target Determination via Display Event Synchronization
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional gaze tracking systems struggle to accurately determine the object of focus among multiple objects on a screen, especially when user posture or position changes, due to limitations in detection accuracy and error-prone calibration methods.
Innovation Solution
A gaze target determination device that generates display events for each object, detecting gaze direction and calculating gaze trajectories to synchronize with object movements, allowing for accurate determination based on time differences between event detection and occurrence times, even with low detection accuracy.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Device complexity
If the corneal reflex method is used to detect eye gaze direction, then the detection mechanism is simple and direct, but the detection range is limited and user posture flexibility is reduced
Solution Approach 1:
The patent replaces the optical-mechanical corneal reflex detection system with a video-based visual field analysis system. Instead of using infrared LEDs and corneal reflection physics, the system captures video images of the user's face and eyes, then processes these images computationally to determine gaze direction and identify gazed objects, thereby enabling flexible user postures without additional optical hardware constraints
Solution Approach 2:
The patent introduces a video camera and image processing system as an intermediary between the user's eye and the gaze detection process. Rather than directly measuring corneal reflection, the system uses video images as an intermediate representation to extract eye position and gaze direction information, allowing adaptation to various user postures through computational analysis
2Adaptability or versatility
If the three-dimensional eye model method is used to detect gaze direction, then user posture flexibility is improved, but measurement accuracy deteriorates due to modeling errors
Solution Approach 1:
The patent creates a visual field map that copies and represents the actual screen display content from the user's perspective. Instead of relying on theoretical 3D eye models, the system generates a digital copy of what the user sees and compares detected gaze positions against this actual visual content, thereby improving accuracy by referencing real display information rather than modeled eye geometry
Solution Approach 2:
The patent changes the approach from fixed 3D geometric parameters (eyeball radius, center position) to dynamic visual field parameters. The system adjusts the visual field map based on detected eye position and orientation, creating a flexible reference frame that adapts to different user postures while maintaining accurate gaze determination through comparison with actual screen content
3Device complexity
If gaze position is directly compared with object display position to determine gaze target, then the determination process is simple, but accuracy deteriorates when multiple objects are present or user posture changes
Solution Approach 1:
The patent adds a temporal dimension to gaze target determination by analyzing gaze trajectories over time. Instead of making instantaneous point-to-point comparisons between gaze position and object positions, the system tracks gaze movement paths and determines targets based on trajectory patterns, thereby improving accuracy in distinguishing between multiple objects and maintaining reliability under varying user postures
Applied Scientific Principles
This section explains which scientific principles are used to turn an abstract innovation direction into a practical engineering solution.
Function Achieved in This Case
Enables precise identification of the object being gazed at by analyzing synchronization structures between object movements and eye movements, reducing false determinations and improving accuracy across various user postures and positions.
Implementation Method 1
a gaze direction detection unit which detects the gaze direction of the user
Data Source
Figure 1
Figure 2
Figure 3A~3B
AI summary
The object at which a user is gazing is accurately determined from among objects displayed on a screen. A gaze target determination device (10) includes: an event setting unit (13) which generates, for each object at different times, a display event which triggers a movement of the gaze direction of the user who is gazing at the object and which indicates a movement or a change of the object; a gaze trajectory calculation unit (12) which calculates a gaze trajectory based on the detected user's gaze direction; an eye movement event detection unit (14) which detects event detection time which is time at which the gaze direction moves according to the display event, based on the calculated gaze trajectory; a synchronization structure analysis unit (15) which calculates, for each object, time difference between the event detection time and event occurrence time; and a gaze target determination unit (16) which determines the object at which the user is gazing from among the objects, based on the calculated time difference.