Point Cloud Entropy Coding with Neighbor-Based Occupancy Contexts
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing methods for compressing point clouds are inefficient and do not effectively utilize context-adaptive binary entropy coding, leading to excessive context management and suboptimal compression performance.
Innovation Solution
The proposed solution involves encoding and decoding point clouds using a method that determines the bit sequence for sub-volumes based on a sub-volume neighbour configuration, which depends on the occupancy pattern of neighbouring volumes. This approach selects probabilities for entropy coding based on the sub-volume neighbour configuration, reducing the number of contexts required and improving compression efficiency.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If existing compression methods are used for point clouds, then compression is achieved, but compression efficiency is poor and context management becomes excessive
Solution Approach 1:
The patent segments the point cloud data into multiple octants (8 sub-volumes) and processes each octant separately with its own context model. This segmentation allows the encoder to manage contexts locally within each octant rather than globally, reducing the overall context management complexity while improving compression efficiency through localized adaptation to each octant's specific characteristics
Solution Approach 2:
The patent implements local quality by maintaining separate context models for each octant of the point cloud data. Each octant has its own probability context that is independently updated based on local occupancy patterns, allowing the compression algorithm to adapt to local variations in point density and distribution rather than using a single global context model
2Productivity
If context-adaptive binary entropy coding is applied to point clouds, then compression performance improves, but the number of contexts required becomes excessive
Solution Approach 1:
The patent divides the point cloud into 8 octants and allocates separate context models to each octant. This segmentation reduces the total number of contexts needed compared to a global context model while maintaining the benefits of context-adaptive coding, as each octant's context model only needs to track probabilities for that specific region
Solution Approach 2:
Instead of using a single global context model that would require tracking all possible context states across the entire point cloud, the patent inverts the approach by using multiple local context models, one per octant. This inversion reduces the complexity of context management while preserving compression performance through localized adaptation
Data Source
AI summary
Methods and devices for encoding a point cloud. A bit sequence signalling an occupancy pattern for sub-volumes of a volume is coded using binary entropy coding. For a given bit in the bit sequence, a context may be based on a sub-volume neighbour configuration for the sub-volume corresponding to that bit. The sub-volume neighbour configuration depends on an occupancy pattern of a group of sub-volumes of neighbouring volumes to the volume, the group of sub-volumes neighbouring the sub-volume corresponding to the given bit. The context may be further based on a partial sequence of previously-coded bits of the bit sequence.