Eye Gesture Calibration for Head-Mountable Devices
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current near-eye display technologies for head-mountable devices (HMDs) face challenges in accurately recognizing and executing eye gesture commands due to variations in user inputs and potential noise or involuntary gestures, which affects the reliability and efficiency of task execution.
Innovation Solution
The implementation of a method and system for calibrating eye gesture recognition in HMDs, where signals indicative of detected eye gestures are compared to reference signals to determine if they represent valid commands, with adjustments made to the reference signals based on confirmed commands, allowing for improved recognition and execution of tasks, including implicit and explicit calibration techniques.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Ease of operation
If eye gesture recognition is implemented in HMDs, then hands-free operation capability is improved, but recognition accuracy deteriorates due to variations in user inputs and noise
Solution Approach 1:
The system performs preliminary calibration by collecting multiple eye gesture signals from the user and establishing reference signals before normal operation. This preliminary action creates a personalized baseline that improves subsequent recognition accuracy despite variations in user inputs.
Solution Approach 2:
The system continuously compares detected eye gesture signals against reference signals and adjusts the reference signals based on the comparison results. This feedback mechanism dynamically adapts to user behavior patterns, maintaining high recognition accuracy while enabling hands-free operation.
2Measurement precision
If reference signals are adjusted dynamically, then recognition accuracy is improved, but system complexity increases
Solution Approach 1:
The system performs self-calibration by automatically collecting user eye gestures, analyzing patterns, and updating reference signals without requiring external intervention. This self-service approach improves accuracy while minimizing the complexity burden on users and operators.
Solution Approach 2:
The system uses a simplified calibration process that collects a sufficient number of samples rather than attempting to capture all possible variations. This partial action approach achieves practical accuracy thresholds without the excessive complexity of comprehensive calibration.
Data Source
AI summary
Examples of methods and systems for providing calibration for eye gesture recognition are described. In some examples, calibration can be executed via a head-mountable device. A method for calibration of a system may account for changes in orientation of the head-mountable device, update recognition of the eye gestures, or increase efficiency of the system, for example. The head-mountable device may be configured to receive signals indicative of eye gestures from an eye gesture-detection system and in response to receiving a second command confirming that the signal is indicative of an eye gesture command, to make adjustments to the eye gesture recognition system and/or the reference signals. The head-mountable device may calibrate an eye gesture recognition system via implicit or explicit calibration, for example.