3D Point Cloud Encoding via Region Connectivity
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current three-dimensional data encoding methods lack the ability to efficiently select and decode specific encoded three-dimensional points from a large dataset, leading to inefficient data transmission and processing, particularly in applications like three-dimensional map rendering where only relevant data is needed.
Innovation Solution
A method that encodes three-dimensional points by region, generating connectivity information based on relationships between regions, including tile and relation information, to create a bitstream that allows for selective decoding of desired points, reducing unnecessary data processing and improving rendering efficiency.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If all three-dimensional points are encoded and transmitted, then complete data is available, but data transmission and processing time increase
Solution Approach 1:
The three-dimensional space is divided into multiple regions (tiles), and connectivity information is generated to represent relationships between these regions. This segmentation allows the system to transmit and process only the necessary regions rather than the entire point cloud, reducing transmission time while maintaining data completeness when needed.
Solution Approach 2:
The patent extracts and transmits only the essential connectivity information (region relationships) rather than all raw three-dimensional point data. This extraction approach allows selective decoding of specific regions, reducing overall data transmission time while preserving the ability to reconstruct complete data when required.
2Productivity
If region-based encoding with connectivity information is implemented, then selective decoding efficiency improves, but encoding complexity increases
Solution Approach 1:
Connectivity information representing region relationships is pre-calculated and embedded in the encoded data structure during the encoding phase. This preliminary action enables the decoder to quickly identify and selectively decode only the necessary regions without having to analyze the entire point cloud, thereby improving selective decoding efficiency while keeping the additional encoding complexity manageable.
3Adaptability or versatility
If connectivity information is generated between regions, then selective point selection capability improves, but data structure complexity increases
Solution Approach 1:
The patent introduces connectivity information as an intermediary data structure that mediates between the raw three-dimensional point data and the decoding process. This intermediary layer provides a simplified interface for selecting and decoding specific regions by representing spatial relationships in a structured format, thereby improving selective point selection capability while managing data structure complexity through organized region relationships.
Data Source
AI summary
A three-dimensional data encoding method includes: encoding three-dimensional points each located in any one of a plurality of regions, on a per region basis, to generate encoded three-dimensional points; generating connectivity information generated based on a relationship between a first region among the plurality of regions and second regions aside from the first region among the plurality of regions, the connectivity information including (i) tile information indicating values each uniquely assigned to a different one of the plurality of regions and (ii) relation information indicating that the first region and one of the second regions are related; and generating a bitstream including the connectivity information generated and the encoded three-dimensional points.