Point Cloud Decoding Using Local Spatial Coordinates
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing point cloud decoding methods in the Geometry-based Point Cloud Compression (G-PCC) framework suffer from large geometric errors during spatial scalability due to the uniform representation of points within octree nodes, disregarding the varying distributions of points inside different nodes.
Innovation Solution
A point cloud decoding method that determines geometric center coordinates of the current node and its neighboring nodes based on spatial correlation, utilizing local spatial coordinate means to reconstruct geometric information, thereby reducing geometric errors and improving decoding efficiency.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Device complexity
If all points within a cube range of octree nodes are represented by one geometric point, then the decoding process is simplified, but large geometric errors occur in geometry reconstruction processes after spatial scalability
Solution Approach 1:
The patent applies local quality by differentiating the treatment of points based on their spatial distribution characteristics. Instead of uniformly representing all points in a node with a single geometric point, the method identifies and processes regions with different point distribution densities separately. This allows high-density regions to be reconstructed with higher precision while maintaining simplified processing in low-density regions, thereby resolving the contradiction between decoding simplicity and geometric reconstruction accuracy.
2Productivity
If uniform representation of points within octree nodes is used, then the encoding and decoding process is more efficient, but the varying distributions of points inside different nodes are not considered leading to large geometric errors
Solution Approach 1:
The patent employs parameter changes by dynamically adjusting the representation parameters based on the actual point distribution within each octree node. The method calculates point distribution metrics (such as density or variance) and uses these to determine the appropriate reconstruction strategy for each node. This adaptive parameter adjustment allows the system to maintain high decoding efficiency while achieving accurate geometric reconstruction by matching the representation complexity to the actual data characteristics.
Data Source
AI summary
A point cloud decoding method includes that: a decoder decodes a bitstream, and determines a mode parameter; in response to the mode parameter indicating to use a spatial scalability mode, the decoder determines geometric center coordinates of a current node, and determines, in the point cloud, geometric center coordinates of neighbouring nodes of the current node based on the geometric center coordinates of the current mode; the decoder determines local spatial coordinate means corresponding to the current node based on the geometric center coordinates of the current node and the geometric center coordinates of the neighbouring nodes; and the decoder determines reconstructed geometric information of the current node based on the local spatial coordinate means. Further, a decoder and a non-transitory computer-readable storage medium are also provided.