Dynamic Over-Rendering for AR Late-Warping Power Reduction
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Augmented reality (AR) devices experience motion-to-photon latency, causing virtual content to lag behind real-world movements, which diminishes the user experience.
Innovation Solution
A method for dynamically adjusting the size of the over-rendered area based on angular and linear velocity, recent pose, and previous warp poses to reduce the number of shaded pixels and minimize power consumption.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If a large over-rendered area is used to compensate for head movement during late-warping, then rendering quality and smoothness are improved, but computational cost, power consumption, and thermal impact increase
Solution Approach 1:
The patent applies dynamics by making the over-rendering area size variable rather than fixed. The system dynamically adjusts the over-rendering area based on real-time head movement characteristics (angular velocity, linear velocity, prediction error metrics). When head movement is detected to be minimal, the over-rendering area is reduced; when movement is significant, the area is expanded. This dynamic adaptation resolves the contradiction by optimizing the balance between rendering smoothness and power consumption based on actual usage conditions.
Solution Approach 2:
The patent changes the parameter of over-rendering area size based on measured head movement parameters (angular velocity, linear velocity, prediction error). By monitoring these parameters and adjusting the over-rendering area accordingly, the system achieves rendering quality only when necessary, thereby reducing unnecessary computational cost and power consumption while maintaining reliability when head movement requires it.
2Reliability
If a large over-rendered area is used to compensate for head movement during late-warping, then rendering quality and smoothness are improved, but computational cost increases
Solution Approach 1:
The system dynamically adjusts the over-rendering area size based on real-time head movement characteristics. By making the rendering workload adaptive rather than static, the system maintains rendering smoothness only when head movement requires it, thereby improving computational efficiency by avoiding unnecessary rendering calculations during periods of minimal movement.
Solution Approach 2:
The patent adjusts the over-rendering area parameter based on measured head movement parameters (angular velocity, linear velocity, prediction error). This parameter change strategy ensures that computational resources are allocated efficiently - increasing rendering quality only when head movement metrics indicate it is necessary, thus resolving the contradiction between rendering smoothness and computational efficiency.
3Reliability
If a large over-rendered area is used to compensate for head movement during late-warping, then rendering quality and smoothness are improved, but thermal impact increases
Solution Approach 1:
The patent applies dynamics by making the over-rendering area size variable based on real-time head movement characteristics. When head movement is minimal, the over-rendering area is reduced, thereby decreasing computational workload and thermal generation. When head movement is significant, the area is expanded to maintain rendering smoothness. This dynamic approach resolves the contradiction by managing thermal impact adaptively.
Solution Approach 2:
The system changes the over-rendering area parameter based on head movement metrics (angular velocity, linear velocity, prediction error). By adjusting this parameter dynamically, the system maintains rendering quality only when necessary, thereby reducing unnecessary thermal generation while preserving rendering smoothness when head movement requires it.
4Use of energy by moving object
If the over-rendered area is reduced to minimize power consumption, then energy efficiency is improved, but rendering quality may deteriorate during head movement
Solution Approach 1:
The patent implements feedback by continuously monitoring head movement characteristics (angular velocity, linear velocity, prediction error metrics) and using this information to adjust the over-rendering area size. This closed-loop control ensures that rendering quality is maintained only when head movement metrics indicate it is necessary, thereby optimizing power consumption while preventing rendering quality deterioration during actual head movement.
Solution Approach 2:
The system performs preliminary assessment of head movement characteristics before adjusting the over-rendering area. By evaluating angular velocity, linear velocity, and prediction error in advance, the system can proactively adjust the rendering area to prevent quality deterioration while minimizing unnecessary power consumption during periods of stable viewing.
Data Source
Figure 1
Figure 2
Figure 3
AI summary
A method for adjusting an over-rendered area of a display in an AR device is described. The method includes identifying an angular velocity of a display device, a most recent pose of the display device, previous warp poses, and previous over-rendered areas, and adjusting a size of a dynamic over-rendered area based on a combination of the angular velocity, the most recent pose, the previous warp poses, and the previous over-rendered areas.