Predictive Point Cloud Coding via Occupancy Patterns

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Existing point cloud compression methods face inefficiencies due to the sparsely populated nature of point clouds, making it challenging to effectively encode and decode geometric data, particularly with predictive coding, where residual 3D vectors are costly to process.

Innovation Solution

The method involves predictive coding by determining a predicted set of points within a volumetric space, using a bit sequence to indicate occupancy patterns, and context-adaptively entropy coding based on selected context sets, which enables efficient encoding and decoding of point clouds by determining whether sub-volumes contain predicted points and adjusting context sets accordingly.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Productivity

If predictive coding is used to encode point clouds, then compression efficiency is improved, but the cost of encoding and decoding residual 3D vectors increases

Engineering Contradiction:
Improvecompression efficiencyVSAvoidcost of encoding and decoding residual 3D vectors
Core Design Contradiction:
ProductivityVSDevice complexity

Solution Approach 1:

The patent segments the volumetric space into multiple sub-volumes and uses occupancy patterns to indicate which sub-volumes contain points. This segmentation allows the system to focus computational resources only on occupied sub-volumes rather than processing the entire volume, thereby reducing the cost of encoding and decoding residual vectors while maintaining compression efficiency through predictive coding.

Inventive Principle:
Principle #1Segmentation

Solution Approach 2:

The patent applies predictive coding selectively only to occupied sub-volumes identified through occupancy patterns, rather than applying it uniformly to the entire point cloud. This partial action approach reduces unnecessary computational overhead for empty regions while maintaining compression benefits where needed, resolving the contradiction between compression efficiency and processing cost.

Inventive Principle:
Principle #16Partial or excessive action

2Measurement precision

If occupancy patterns are used to indicate sub-volume containment, then encoding accuracy is improved, but the number of contexts to track increases

Engineering Contradiction:
Improveencoding accuracyVSAvoidnumber of contexts to track
Core Design Contradiction:
Measurement precisionVSDevice complexity

Solution Approach 1:

The patent performs preliminary encoding of occupancy patterns to identify which sub-volumes contain points before applying predictive coding. This preliminary action establishes a foundation for accurate encoding by pre-identifying relevant regions, allowing the system to track fewer contexts specifically for occupied sub-volumes rather than maintaining contexts for the entire volumetric space, thus improving encoding accuracy while reducing contextual complexity.

Inventive Principle:
Principle #10Preliminary action

Data Source

PatentEP3553747B1Methods and devices for predictive coding of point clouds
Publication Date: 2021.09.01 BLACKBERRY LTD
  • EP3553747B1 patent drawingFigure 1~2
  • EP3553747B1 patent drawingFigure 3~4
  • EP3553747B1 patent drawingFigure 5~6

AI summary

Methods and devices for encoding a point cloud. A bit sequence signaling an occupancy pattern for sub-volumes of a volume is coded. Predictive coding is used to find a set of predicted points, from which a corresponding predicted occupancy pattern may be determined. The predicted occupancy pattern may be used to determine the contexts for entropy coding the occupancy pattern. The determination may include determining for each sub-volume, whether it contains a predicted point or not and, in some cases, the count of predicted points within the sub-volume. Various threshold numbers of predicted points may cause the selection of different context sets for coding the occupancy pattern. The predictive coding may be enabled by determining that the prediction of occupancy of a parent volume to the volume met a threshold level of accuracy.