Voxel Hashing for Sparse Volumetric Data Access
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current computing systems face challenges in processing and storing large volumetric data sets efficiently, particularly in augmented and mixed reality applications, where high-frame rates and low latency are required, often leading to memory and processing constraints.
Innovation Solution
A sparse volumetric data structure is introduced, utilizing a sparse sexaquaternary tree format that tags voxels as occupied or empty, allowing for reduced storage and faster processing by eliminating empty space, and is hardware-accelerated to minimize latency.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Quantity of substance
If dense volumetric data structures are used to store all voxels, then complete spatial coverage is achieved, but memory requirements increase significantly
Solution Approach 1:
The patent extracts and stores only the occupied voxels in the volumetric data structure, removing the unnecessary empty space representation. This selective extraction reduces memory requirements while maintaining complete spatial coverage through coordinate-based addressing of occupied regions only.
Solution Approach 2:
The patent applies different storage strategies to different regions of the volumetric space: occupied regions are stored with full detail while empty regions are implicitly represented by their absence from the data structure. This local differentiation optimizes memory usage based on actual data presence.
2Productivity
If high-frame rate processing is implemented in AR/VR systems, then real-time rendering is achieved, but processing latency increases causing motion sickness
Solution Approach 1:
The patent extracts and processes only the occupied voxels that contain actual data, eliminating processing of empty space. This selective processing reduces the computational workload per frame, enabling high frame rates while maintaining low latency by focusing resources on relevant data only.
Solution Approach 2:
The patent performs processing on a partial subset of the total volumetric data (only occupied voxels) rather than the complete volume. This partial action approach achieves sufficient processing throughput for real-time rendering without the excessive computation that would cause latency.
3Quantity of substance
If sparse volumetric data structures are used to reduce memory usage, then storage efficiency improves, but data access complexity increases
Solution Approach 1:
The patent replaces complex hierarchical navigation structures with a direct coordinate-based addressing mechanism. Instead of traversing tree structures or spatial indexes to locate voxels, the system directly computes memory addresses from voxel coordinates, substituting mechanical navigation with mathematical addressing to reduce access complexity.
Solution Approach 2:
The patent changes the addressing parameter from hierarchical level-based navigation to direct coordinate-based indexing. This parameter transformation enables O(1) access time complexity while maintaining sparse data storage efficiency, as the memory address can be directly computed from the voxel's x, y, z coordinates without traversal.
Data Source
AI summary
A particular voxel is identified within a volume and a hash table is used to obtain volumetric data describing the particular voxel within the volume. Values of x-, y- and z-coordinates in the volume associated with the particular voxel are determined an index value associated with the particular voxel is determined according to a hashing algorithm, where the index value is determined from summing weighted values of the x-, y- and z-coordinates, and the weighted values are based on a variable value corresponding to a dimension of the volume. A particular entry is identified in the hash table based on the index value, where the particular entry includes volumetric data, and the volumetric data identifies, for the particular voxel, whether the particular voxel is occupied.