Waveguide Correction Map Compression for AR Displays
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current augmented and virtual reality systems face challenges in efficiently managing waveguide non-uniformity, which leads to visual artifacts and increased memory requirements due to the large number of correction maps needed for accurate image rendering across different eye positions.
Innovation Solution
The system employs compressed correction maps and interpolation techniques to compensate for waveguide non-uniformity, using eye tracking to dynamically determine eye positions and apply pre-computed correction maps with low spatial resolution, which are then up-sampled to match the image resolution, reducing memory usage and improving image smoothness.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If correction maps are stored for all pre-determined eye positions with high resolution, then waveguide non-uniformity compensation accuracy is improved, but memory requirements increase significantly
Solution Approach 1:
The correction maps are segmented into multiple resolution levels. A low-resolution base correction map is stored in memory, while high-resolution correction data is stored externally or computed on-demand. This segmentation allows the system to maintain accurate compensation while reducing in-memory storage requirements.
Solution Approach 2:
The correction map structure uses a nested approach where a low-resolution base map is stored in device memory, and additional high-resolution correction layers are stored externally or generated as needed. This nested structure enables progressive refinement of correction quality without proportionally increasing base memory requirements.
2Manufacturing precision
If high-resolution correction maps are applied directly, then image quality is improved, but system processing load and memory bandwidth increase
Solution Approach 1:
The system dynamically adjusts the correction map resolution based on current operational requirements. For static or slowly changing images, lower-resolution correction maps are used to reduce processing load. For dynamic content requiring precise eye tracking, the system transitions to higher-resolution correction maps, optimizing the balance between image quality and processing demands.
Solution Approach 2:
Instead of applying full high-resolution correction maps to all images, the system applies partial correction using low-resolution base maps for content where maximum precision is not critical. This selective approach reduces overall processing load while maintaining adequate image quality for most use cases.
3Quantity of substance
If correction maps are up-sampled from low to high resolution, then memory usage is reduced, but processing time increases
Solution Approach 1:
The low-resolution base correction maps are pre-computed and stored in device memory during manufacturing or initial setup. This preliminary action eliminates the need to compute high-resolution correction maps in real-time, as the system only needs to perform efficient up-sampling operations during operation, significantly reducing processing time compared to generating correction maps from scratch.
4Quantity of substance
If interpolation is performed on compressed correction maps, then memory requirements are reduced, but interpolation accuracy may decrease
Solution Approach 1:
The system changes the parameter of correction map resolution by storing and performing interpolation on low-resolution base maps. This parameter change reduces memory requirements while the interpolation algorithm compensates for the lower input resolution, maintaining adequate correction accuracy for the application's needs.
Data Source
AI summary
A computing system may determine, for a current frame, a viewer's current eye position with respect to a waveguide of a display, identify eye positions that collectively form a grid of eye positions surrounding the current eye position, obtain, from a memory on the display, compressed arrays of scaling factors for correcting non-uniformities of the waveguide at the identified eye positions, perform interpolation based on the compressed arrays to generate an array of scaling factors for the current eye position, adjust pixel values of the current frame based on the customized array, and output the current frame with the adjusted pixel values to the display. The compression operation may include dithering or converting pixel values to a different color space. The interpolation may be performed on the compressed arrays or on results of a decompression operation. The customized array may be up-sampled prior to adjusting the pixel values.