Bounded Volume Hierarchy for Visual Data Compression
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current data compression technologies face challenges in efficiently compressing and decompressing visual elements, such as pixels in two-dimensional images or voxels in three-dimensional holograms, especially in maintaining quality and complexity in graphics-intensive applications like virtual and augmented reality.
Innovation Solution
The implementation of a bounded volume hierarchy data structure, where parent nodes represent subsets of visual elements with child nodes indicating offsets from a representative value, allowing for efficient data compression and partial decompression, and enabling improved entropy encoding and rendering.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If full data decompression is performed to reconstruct the complete uncompressed data structure, then visual quality is maintained, but processing time and computational resources increase significantly
Solution Approach 1:
The patent implements partial decompression by allowing the system to decompress only the specific portions of compressed visual data that are currently needed for display, rather than decompressing the entire data structure. This is achieved through a viewport management system that identifies which tiles or regions are visible and decompresses only those portions, significantly reducing processing time while maintaining visual quality for the visible areas.
Solution Approach 2:
The visual data structure is divided into multiple independent tiles or regions that can be compressed, stored, and decompressed separately. This segmentation allows the system to selectively decompress only the necessary tiles based on viewport position and size, rather than processing the entire large-scale visual data structure, thus reducing computational overhead while preserving visual quality where needed.
2Quantity of substance
If compressed data structures are used to reduce storage and transmission requirements, then data size is reduced, but decompression complexity increases
Solution Approach 1:
The data is pre-compressed into a hierarchical tile-based structure with metadata that organizes compression levels, tile boundaries, and dependency relationships. This preliminary organization of compressed data allows the decompression system to efficiently navigate and process only the necessary portions without having to analyze the entire compressed data structure, reducing decompression complexity while maintaining high compression ratios.
3Loss of energy
If high-level compression is applied to reduce data transmission, then bandwidth usage decreases, but rendering speed may be affected
Solution Approach 1:
The system dynamically adjusts the decompression and rendering process based on real-time conditions such as viewport position, device capabilities, and available processing resources. When bandwidth is constrained, higher compression levels are maintained with selective decompression of visible tiles. When processing resources are abundant, the system can decompress and render more aggressively, optimizing the balance between bandwidth usage and rendering speed adaptively.
Data Source
AI summary
A machine is configured to perform data compression or data decompression with respect to visual elements, such as two-dimensional pixels of an image or three-dimensional voxels of a hologram. The machine accesses a set of visual elements and performs compression on the set by generating a bounded volume hierarchy of nodes. After the bounded volume hierarchy has been generated, with or without modification, the machine provides a compressed version of the set by providing at least a portion of the bounded volume hierarchy. The same machine or another machine accesses a compressed first representation of the set by accessing at least part of the bounded volume hierarchy. The machine may perform partial decompression of the first representation by generating a second representation of the set, based on only a portion of the bounded volume hierarchy. The machine then causes rendering and display of the second representation.