Gaze Tracking for VR Rendering Power Reduction
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current head-mounted display (HMD) systems for virtual reality (VR) require high frame rates to prevent motion sickness, leading to increased power consumption and computational load due to constant rendering of visuals, even during vision-interrupting events like blinks and saccades, where visual information is not perceived by the user.
Innovation Solution
Implementing eye gaze tracking to detect the onset and duration of blinks and saccades, allowing the graphics rendering system to delay rendering frames until the event ends, and using foveated rendering to concentrate computational resources on the foveal region, reducing unnecessary rendering and power consumption.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If high frame rate rendering is performed continuously to prevent motion sickness, then user comfort and visual quality are improved, but power consumption and computational load increase
Solution Approach 1:
The system implements periodic rendering by detecting vision-interrupting events (blinks and saccades) and suspending frame rendering during these periods. Rendering operates at full frame rate during normal periods and reduces or stops during detected events, creating a periodic on-off pattern that matches human visual perception cycles
Solution Approach 2:
The system discards frames that would be rendered during vision-interrupting events since they would not be perceived by the user. Computational resources are recovered during these events and can be reallocated or used to pre-render frames for display after the vision-interrupting event concludes
2Manufacturing precision
If frames are rendered at full resolution across the entire display, then visual quality is improved, but computational load and power consumption increase
Solution Approach 1:
The system applies foveated rendering that renders the foveal region (central vision area) at high resolution while rendering peripheral regions at lower resolution. This local differentiation of quality matches the human visual system's varying acuity across the visual field, maintaining perceived visual quality while reducing overall computational load
Solution Approach 2:
The system performs partial rendering by concentrating full computational effort only on the foveal region rather than the entire display area. Peripheral regions receive reduced rendering effort, implementing partial action that is sufficient for perceived quality while avoiding excessive computation in low-acuity areas
3Stability of the object's composition
If rendering is performed during all time periods, then frame continuity is maintained, but unnecessary computational resources are wasted during vision-interrupting events
Solution Approach 1:
The rendering system uses eye tracking data from the vision system to automatically detect vision-interrupting events and adjust its own rendering output accordingly. The system serves itself by using its own sensory input (eye position data) to control its computational output (frame rendering), eliminating wasted resources during periods when the user cannot perceive the display
Data Source
AI summary
Gaze tracking data may be analyzed to determine the onset and duration of a vision interrupting event, such as a blink or saccade. Presentation of images to a viewer may then be suspended during the vision interrupting event and resumed in sufficient time to ensure that the viewer sees the image at the time the vision interrupting event has concluded.