3D Attribute Signal Compression Using Hierarchical Wavelet Encoding
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current methods for compressing 3D meshes and point clouds focus primarily on geometry data rather than vertex attributes, which are critical for high-quality rendering, and struggle with the irregular and sparse structure of unstructured point clouds, leading to inefficient compression and decompression.
Innovation Solution
A computer-implemented method using discrete wavelet transforms with wavelet convolution filters to compress physical attributes in a predetermined space, exploiting space correlation and handling irregular structures, while maintaining a high compression ratio and fast processing.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Quantity of substance
If discrete wavelet transform with wavelet convolution filters is used to compress vertex attributes, then compression ratio is improved and real-time processing is enabled, but computational complexity increases
Solution Approach 1:
The patent applies segmentation by dividing the space into hierarchical subdivisions and processing vertex attributes in discrete wavelet transform subbands. This allows the compression algorithm to work on segmented data structures (hierarchical subdivisions) rather than raw unstructured data, reducing overall computational complexity while maintaining compression efficiency.
Solution Approach 2:
The patent changes parameters by transforming vertex attributes from spatial domain to frequency domain using discrete wavelet transform. This parameter transformation enables energy compaction where most information is concentrated in fewer coefficients, achieving high compression ratios. The use of different wavelet bases and transform levels provides additional parameter control for optimizing the compression-decompression tradeoff.
2Productivity
If space correlation is exploited through hierarchical subdivision structure, then compression efficiency is improved, but handling irregular and sparse structures becomes more difficult
Solution Approach 1:
The patent segments the irregular and sparse point cloud data into hierarchical subdivisions, creating a structured representation that preserves spatial relationships. This segmentation approach allows the algorithm to exploit space correlation within each subdivision while maintaining the ability to handle the overall irregular structure through the hierarchical organization.
Solution Approach 2:
The patent implements nesting through hierarchical subdivisions where smaller subdivisions are nested within larger ones. This nested structure allows the compression algorithm to operate at multiple levels of detail, exploiting space correlation at each level while efficiently representing the irregular and sparse nature of the original data through the nested hierarchy.
3Quantity of substance
If vertex attributes are compressed separately from geometry data, then compression of attributes is improved, but overall data integration becomes more complex
Solution Approach 1:
The patent segments the data into geometry components and attribute components, processing each separately through appropriate compression methods. Vertex attributes are compressed using discrete wavelet transform while geometry data uses separate compression techniques. This segmentation allows optimization for each data type while the hierarchical subdivision structure provides a framework for integrating the compressed results.
Solution Approach 2:
The patent creates a universal compression framework based on hierarchical subdivisions that can handle both geometry data and vertex attributes. The discrete wavelet transform approach serves multiple functions: it compresses attribute data efficiently, preserves spatial correlation, and integrates with the overall compression system. This multi-functional approach simplifies data integration compared to completely separate processing pipelines.
Data Source
AI summary
The invention notably relates to a computer-implemented method for compressing data representing values of a physical attribute in a predetermined space. The method comprises providing a signal that includes a mapping from leaf cells of a hierarchical subdivision of the predetermined space each onto a respective coefficient representative of a value of the physical attribute at the respective leaf cell. The method also comprises determining a discrete wavelet transform of the signal and encoding the determined discrete wavelet transform. The method provides an improved way to compress a modeled object that represents a real object.