Point Cloud LOD Sampling for Consistent Encoding and Decoding
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing point cloud encoding and decoding methods using a fixed sampling period for Level of Detail (LOD) division lack system consistency, affecting encoding and decoding efficiencies due to uncertainty in the sampling period range.
Innovation Solution
Determine a maximum allowable value of the sampling period for LOD division and set a preset value based on this maximum, ensuring the sampling period does not exceed it, thereby enhancing system consistency and improving encoding and decoding efficiencies.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Device complexity
If a fixed sampling period method is used for LOD division, then the complexity is relatively low and the distribution of initial points can be captured well, but system consistency is not considered and encoding and decoding efficiencies are affected
Solution Approach 1:
The patent changes the sampling period parameter from a fixed value to a variable that is adaptively determined based on the point cloud data characteristics. The encoder determines a sampling period for LOD division, and this parameter is transmitted to the decoder, allowing both encoder and decoder to use the same sampling period value, thus ensuring system consistency while improving encoding and decoding efficiencies.
2Ease of manufacture
If the sampling period is not constrained with a maximum allowable value, then the method is simpler to implement, but system consistency deteriorates and encoding/decoding efficiency is reduced
Solution Approach 1:
The patent applies preliminary action by determining and constraining the sampling period range before the actual LOD division process. The encoder determines a maximum allowable value for the sampling period and ensures the selected sampling period does not exceed this maximum. This preliminary constraint is communicated to the decoder, ensuring both sides operate within the same parameter bounds, thus maintaining system consistency without significantly complicating the implementation.
Data Source
AI summary
Disclosed are a point cloud encoding and decoding method, and a decoder. The method includes: acquiring geometric information and attribute information of an input point cloud; determining a maximum allowable value of a sampling period when the input point cloud is subjected to level-of-detail (LOD) division; determining a preset value of the sampling period on the basis of the maximum allowable value of the sampling period; processing the input point cloud according to the preset value of the sampling period and the geometric information, so as to obtain at least one refinement layer and at least one detail layer; and encoding the attribute information by using the at least one refinement layer and the at least one detail layer, so as to generate a code stream, and writing the preset value of the sampling period into the code stream.