Point Cloud Index Determination via Perpendicular Plane Occupancy
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
The accuracy of index determination in point cloud compression is low, leading to reduced coding performance in existing technologies.
Innovation Solution
A method for index determination is provided, where the first index of a current node is determined based on occupied child nodes of at least one decoded or encoded neighbouring node on a plane perpendicular to a k-th axis, improving the accuracy of index determination and thus enhancing coding performance.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If the encoder uses spatial correlation between the current node and surrounding nodes for intra prediction of occupancy bits, then the coding performance is improved through context-based adaptive binary arithmetic coding, but the accuracy of the determination of the index of the current node is low
Solution Approach 1:
The patent extends the prediction from traditional 2D planar neighbors to 3D spatial neighbors by considering occupied child nodes on planes perpendicular to multiple axes (k-th axis, l-th axis, m-th axis). This dimensional extension allows the encoder to capture spatial correlation more effectively in three-dimensional space, improving index determination accuracy while maintaining coding efficiency.
Solution Approach 2:
The patent applies different prediction strategies based on the local spatial configuration of occupied child nodes. By examining the specific arrangement of occupied children on perpendicular planes and selecting appropriate prediction modes accordingly, the method adapts to local structural characteristics, thereby improving both accuracy and coding performance.
2Loss of information
If the encoder performs arithmetic encoding based on the index of the current node to implement context-based adaptive binary arithmetic coding, then the geometry bitstream is obtained, but the bit distortion rate and bit per index pixel ratio are suboptimal due to low index determination accuracy
Solution Approach 1:
The patent incorporates feedback from decoded neighboring nodes' occupied child nodes to improve the prediction of the current node's index. By utilizing the spatial correlation information from already decoded neighbors and feeding it back into the prediction process, the method achieves more accurate index determination, which directly reduces bit distortion rate and improves the bit per index pixel ratio in the arithmetic encoding process.
Data Source
AI summary
Embodiments of the disclosure provide a method for index determination, a decoder, an encoder, and a bitstream. The method includes the following. A first index of a current node is determined based on occupied child nodes of at least one decoded neighbouring node of the current node on a plane perpendicular to a k-th axis.