HMD Eye Gesture Calibration via Proximity Sensor Clustering
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current calibration processes for head-mountable displays (HMDs) require users to perform specific eye gestures, which can be burdensome and disruptive, especially in real-time applications like augmented or virtual reality.
Innovation Solution
The HMD observes proximity sensor data over time to generate a distribution of magnitude changes, identifying clusters for normal eye movements and blinks, allowing for automatic calibration without user intervention by mapping these clusters to detect blinks and winks based on typical magnitude changes.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If typical calibration processes require the wearer to stop and perform certain eye gestures, then calibration accuracy can be improved, but user convenience and continuity of use deteriorate
Solution Approach 1:
The system performs calibration automatically without requiring user action. The processor analyzes sensor data from proximity sensors to detect eye gestures and generate updated reference data autonomously, allowing the calibration process to serve itself rather than requiring the wearer to perform specific gestures manually
Solution Approach 2:
The system continuously collects and analyzes sensor data in the background to prepare calibration information in advance. By monitoring proximity sensor data over time and identifying clusters of magnitude changes corresponding to blinks and winks, the system has calibration data ready before it is actually needed, eliminating the need for interruptive calibration sessions
2Measurement precision
If calibration is performed frequently to account for changes in HMD position and lighting, then measurement accuracy is improved, but loss of time and disruption to usage increase
Solution Approach 1:
The calibration process operates continuously in the background without interruption to normal device usage. The processor continuously receives sensor data, identifies eye gestures, and updates reference data while the wearer uses the HMD for its intended purposes, ensuring calibration is maintained without requiring dedicated calibration time
Solution Approach 2:
The system performs calibration preparations in advance by continuously monitoring and analyzing sensor data patterns. By identifying clusters of magnitude changes and mapping them to eye gestures proactively, the system has calibration updates ready before environmental changes or position shifts affect detection accuracy
Data Source
AI summary
Examples methods and systems for distinguishing winks from other eye motions are described. In some examples, a method for distinguishing winks from other eye motions may account for changes in orientation of a computing device relative to a user, calibrate eye gesture recognition, or increase efficiency of the computing device. A computing device may be configured to receive sensor data corresponding to an eye and determine a distribution of magnitude changes in the sensor data over an interval of time. The computing device may identify clusters that correspond to ranges of magnitude changes within the distribution and use the clusters as reference data to identify ranges of sensor data for different types of eye motions, including differentiating winks from clusters indicative of ranges of normal eye activity and blinks.