Point Cloud Occupancy Coding With Neighbor-Based Context Reduction

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Current point cloud compression methods are inefficient in handling large datasets and fail to effectively manage contexts in context-adaptive binary entropy coding, leading to suboptimal compression and decoding performance.

Innovation Solution

The proposed solution involves encoding and decoding point clouds using a method that determines bit sequences for sub-volumes based on neighbor configurations, applying context reduction operations to reduce the number of contexts, and using binary entropy coding to improve compression efficiency while maintaining computational feasibility.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Loss of information

If context-adaptive binary entropy coding is applied to point cloud compression, then compression efficiency is improved, but the number of contexts to be managed increases excessively

Engineering Contradiction:
Improvecompression efficiencyVSAvoidnumber of contexts
Core Design Contradiction:
Loss of informationVSDevice complexity

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 system to manage a limited number of contexts per octant rather than requiring excessive contexts for the entire point cloud, thus resolving the contradiction between compression efficiency and context management complexity

Inventive Principle:
Principle #1Segmentation

Solution Approach 2:

The patent applies local quality by using neighbor configuration-dependent contexts where each context model is adapted to the specific local geometry of neighboring points. This allows efficient compression by exploiting local spatial correlations without requiring a globally excessive number of contexts, as each local region uses contexts tailored to its specific characteristics

Inventive Principle:
Principle #3Local quality

2Loss of information

If more contexts are used in binary entropy coding, then compression performance is improved, but computational complexity increases

Engineering Contradiction:
Improvecompression performanceVSAvoidcomputational complexity
Core Design Contradiction:
Loss of informationVSPower

Solution Approach 1:

The patent performs preliminary action by pre-defining a limited set of context models based on neighbor configurations before the actual entropy coding process. This allows the system to achieve good compression performance using a manageable number of pre-prepared contexts, avoiding the need to compute and manage excessive contexts during compression, thus reducing computational complexity while maintaining performance

Inventive Principle:
Principle #10Preliminary action

3Quantity of substance

If point cloud data is compressed using existing methods, then data storage is reduced, but compression and decoding speed is insufficient for large datasets

Engineering Contradiction:
Improvedata storageVSAvoidcompression and decoding speed
Core Design Contradiction:
Quantity of substanceVSProductivity

Solution Approach 1:

The patent segments the point cloud into multiple octants and processes each independently with parallelizable operations. This segmentation enables efficient compression and decoding of large datasets by allowing parallel processing of different octants, thus improving productivity while achieving effective data storage reduction

Inventive Principle:
Principle #1Segmentation

Solution Approach 2:

The patent applies partial action by focusing computational resources on encoding the most significant geometric features using neighbor configuration-dependent contexts. This approach achieves effective compression for large datasets by prioritizing the encoding of critical spatial relationships, improving processing speed without sacrificing essential point cloud information

Inventive Principle:
Principle #16Partial or excessive action

Data Source

PatentEP3514967B1Methods and devices for binary entropy coding of points clouds
Publication Date: 2021.09.08 BLACKBERRY LTD
  • EP3514967B1 patent drawingFigure 1~2
  • EP3514967B1 patent drawingFigure 3~4
  • EP3514967B1 patent drawingFigure 5~6

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.