Eye Odometer Tracking for Low-Power HMD Eye Sensing
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing eye tracking systems in virtual and mixed reality applications consume significant power due to frequent 3D reconstruction and high frame rates, which is inefficient and drains the battery of mobile devices like head-mounted displays (HMDs).
Innovation Solution
Implementing head motion sensors (head odometers) and eye motion sensors (eye odometers) to detect and track relative movement, reducing the need for frequent 3D reconstruction and lowering the frame rate of eye tracking cameras, combined with capturing low-resolution frames to conserve power.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If 3D reconstruction is performed frequently and eye tracking cameras operate at high frame rates, then eye tracking accuracy is improved, but power consumption increases significantly
Solution Approach 1:
The system performs 3D reconstruction periodically rather than continuously, using inertial sensor data to interpolate eye positions between reconstruction frames. This reduces computational load and power consumption while maintaining tracking accuracy through sensor fusion of periodic 3D models with continuous inertial measurements.
Solution Approach 2:
Inertial sensors (accelerometers and gyroscopes) serve as intermediary components that continuously track head and eye motion between 3D reconstruction events. These sensors provide intermediate measurement data that bridges the gap between periodic high-accuracy 3D reconstruction frames, enabling accurate tracking without continuous heavy computation.
2Measurement precision
If 3D reconstruction is performed continuously, then eye position accuracy is maintained, but computational resources and processing time are excessively consumed
Solution Approach 1:
3D reconstruction is executed periodically at reduced frame rates rather than continuously at full frame rate. The system uses inertial sensor data to maintain accurate eye position estimates between reconstruction events, significantly reducing computational resource consumption while preserving measurement accuracy through sensor fusion.
Solution Approach 2:
The system performs preliminary 3D reconstruction at lower computational cost using inertial pre-data or coarse measurements, then refines eye position estimates using full 3D reconstruction only when necessary. This preliminary action reduces the frequency and computational burden of intensive 3D reconstruction operations.
3Measurement precision
If high-resolution frames are captured continuously, then eye tracking precision is improved, but bandwidth usage and power consumption increase
Solution Approach 1:
The eye tracking camera captures high-resolution frames periodically rather than continuously, using inertial sensor data to track eye motion between captured frames. This periodic capture approach maintains eye tracking precision through sensor fusion while significantly reducing power consumption and bandwidth usage associated with continuous high-resolution imaging.
Solution Approach 2:
The system uses partial action by capturing only essential eye region information at high resolution periodically, while using lower-resolution inertial sensor data for continuous tracking. This partial high-resolution capture approach maintains necessary precision while reducing overall energy consumption compared to continuous full-frame high-resolution imaging.
Data Source
AI summary
Low-power eye tracking for detecting position and movements of a user's eyes in a head-mounted device (HMD). An eye tracking system for an HMD may include eye tracking cameras and eye odometers that may reduce power consumption of the system. The eye odometers may be used as a low-power component to track relative movement of the user's eyes with respect to the HMD between frames captured by the eye tracking cameras, which may allow the frame rate of the eye tracking cameras to be reduced.