Point Cloud Attribute Coding With Planar Node Skip Decisions
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
The existing Geometry-based Point Cloud Compression (G-PCC) encoding and decoding framework faces high complexity due to sequential transforms and prediction processes, leading to inefficiencies in removing attribute redundancy and low coding efficiency.
Innovation Solution
A method is introduced to determine the number of voxel nodes and reconstructed nodes to decide whether to skip decoding or encoding, utilizing planar encoding for efficient compression by identifying planar nodes and reducing redundant encoding operations.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If RAHT transform is performed sequentially on nodes of each layer, then attribute information can be coded, but encoding and decoding complexity increases
Solution Approach 1:
The patent divides the point cloud attribute coding into two independent parts: geometry-based attribute coding and non-geometry-based attribute coding. This segmentation allows each part to be processed separately with optimized algorithms, reducing overall complexity while maintaining complete attribute representation.
Solution Approach 2:
The patent performs preliminary classification of attribute nodes into geometry-based and non-geometry-based categories before encoding. This preliminary action enables the selection of appropriate coding strategies for each category, avoiding unnecessary complex processing and reducing overall encoding/decoding complexity.
2Reliability
If RAHT transform is performed sequentially on nodes of each layer, then attribute information can be coded, but coding efficiency decreases
Solution Approach 1:
By segmenting attribute coding into geometry-based and non-geometry-based parts, the patent enables parallel processing and optimized algorithm selection for each segment, significantly improving coding efficiency while preserving all attribute information.
Solution Approach 2:
The patent changes the coding parameter strategy by selecting different coding methods (geometry-based vs. non-geometry-based) depending on the attribute characteristics. This parameter adaptation allows more efficient coding for each attribute type, improving overall coding efficiency.
3Loss of information
If traditional encoding is used for all voxel nodes, then complete attribute information is preserved, but encoding complexity and time increase
Solution Approach 1:
The patent applies different encoding qualities and methods to different types of nodes: geometry-based attributes use one encoding approach while non-geometry-based attributes use another. This local quality differentiation preserves necessary information while reducing unnecessary processing time.
Solution Approach 2:
The patent performs partial encoding by selectively applying encoding only to necessary attribute components based on their type. This partial action approach avoids redundant processing of all nodes with the same complex method, reducing encoding time while preserving essential information.
Data Source
AI summary
An encoding method, a decoding method and a storage medium are provided. The method includes: determining a first quantity of voxel nodes of a current unit and a second quantity of reconstruction nodes of the current unit, wherein the first quantity and the second quantity are used for determining whether to skip decoding the voxel nodes of the current unit; and on the basis of the first quantity and the second quantity, determining attribute reconstruction values of the voxel nodes of the current unit.