Multi-Resolution Frame Descriptors for XR Localization Accuracy
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing XR systems face challenges in efficiently localizing XR devices in large and very large scale environments with reduced time and improved accuracy, leading to computational inefficiencies and latency in rendering virtual content relative to real objects.
Innovation Solution
The system employs multi-resolution frame descriptors, utilizing neural networks to compute frame descriptors of varying resolutions, with higher resolution used in cloud-based processing and lower resolution on local devices, to streamline image frame comparisons and reduce computational burden, enabling efficient localization and map merging across multiple XR devices.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If high-resolution frame descriptors are used for localization, then localization accuracy is improved, but computational load and processing time increase
Solution Approach 1:
The patent segments the frame descriptor processing by creating multiple versions at different resolutions (e.g., 256-bit, 512-bit, 1024-bit descriptors). Low-resolution descriptors are used for initial rapid comparison and coarse localization, while high-resolution descriptors are applied only when needed for final precise localization. This segmentation allows the system to achieve high localization accuracy without consistently incurring the full computational cost of high-resolution processing.
Solution Approach 2:
The system applies partial action by using low-resolution frame descriptors for the majority of localization operations and reserving high-resolution descriptors for cases requiring enhanced precision. This selective application of computational resources based on operational needs optimizes the balance between localization accuracy and processing efficiency.
2Measurement precision
If high-resolution frame descriptors are used, then localization accuracy is improved, but device complexity and computational resources required increase
Solution Approach 1:
The patent implements local quality by allowing different computational resolutions at different stages and locations in the localization pipeline. Edge devices use lower-resolution descriptors for local processing, while cloud-based systems can leverage higher-resolution descriptors when performing more demanding localization tasks. This distributed quality approach optimizes resource utilization across the system architecture.
Solution Approach 2:
The system creates multiple copies of frame descriptors at different resolution levels (256-bit, 512-bit, 1024-bit versions). These copied descriptors are stored and can be selectively applied based on the specific localization scenario, allowing the system to avoid computing high-resolution descriptors from scratch and reducing overall computational burden on individual devices.
3Productivity
If multi-resolution frame descriptors are implemented across distributed devices, then localization efficiency is improved, but system complexity and coordination requirements increase
Solution Approach 1:
The patent establishes a universal multi-resolution descriptor framework that can be implemented across diverse XR devices and distributed systems. The same set of resolution levels (256-bit, 512-bit, 1024-bit) and comparison algorithms are used throughout the system, enabling consistent localization performance regardless of which device performs the computation. This universality simplifies system integration despite the distributed architecture.
Solution Approach 2:
The system uses frame descriptors as an intermediary representation that mediates between raw sensor data and localization decisions. By translating visual data into standardized multi-resolution descriptors, the system creates a common language for comparison across different devices and processing contexts, simplifying the coordination required in distributed environments.
Data Source
AI summary
A distributed, cross reality system efficiently and accurately compares location information that includes image frames. Each of the frames may be represented as a numeric descriptor that enables identification of frames with similar content. The resolution of the descriptors may vary for different computing devices in the distributed system based on degree of ambiguity in image comparisons and/or computing resources for the device. A descriptor computed for a cloud-based component operating on maps of large areas that can result in ambiguous identification of multiple image frames may use high resolution descriptors. High resolution descriptors reduce computationally intensive disambiguation processing. A portable device, which is more likely to operate on smaller maps and less likely to have the computational resources to compute a high resolution descriptor, may use a lower resolution descriptor.