Point Cloud Data Transmission Device Shared LOD Generation
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current methods for processing point cloud data face challenges in efficiently transmitting and receiving large datasets, experiencing latency and encoding/decoding complexity, and require improvements in compression performance, especially in geometry-based point cloud compression (G-PCC), while also supporting parallel processing and efficient handling of fused point cloud data with multiple attributes.
Innovation Solution
The proposed solution involves a method and apparatus for point cloud data transmission and reception that encodes geometry information, including positions of points, generates Level Of Details (LODs), and searches neighbor points for attribute encoding, sharing these processes for color and reflectance information encoding and decoding, and transmits encoded data with signaling information, allowing for efficient encoding and decoding of geometry and attribute information.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Adaptability or versatility
If separate encoding processes are used for color information and reflectance information, then processing flexibility is improved, but encoding complexity and processing time increase
Solution Approach 1:
The patent merges the LOD generation and neighbor point search processes into a single shared operation that serves both color information encoding and reflectance information encoding. This reduces redundancy in computation while maintaining the ability to process different attribute types, thereby lowering encoding complexity without sacrificing processing flexibility.
2Manufacturing precision
If LOD generation and neighbor point search are performed separately for each attribute type, then attribute-specific optimization is improved, but processing time and computational load increase
Solution Approach 1:
The patent combines LOD generation and neighbor point search into unified processes that are shared across multiple attribute encodings. This eliminates redundant computations while preserving attribute-specific optimization capabilities through separate encoding stages that utilize the pre-computed spatial structures.
Solution Approach 2:
The patent performs LOD generation and neighbor point search as preliminary actions before the actual attribute encoding. By preparing these spatial structures in advance, the encoding process can proceed more efficiently without repeating computationally intensive operations for each attribute type.
3Loss of information
If full point cloud data is transmitted, then data completeness is improved, but transmission bandwidth and storage requirements increase
Solution Approach 1:
The patent extracts and transmits only the essential encoded attribute information (color and reflectance) along with necessary signaling data, rather than transmitting complete raw point cloud data. This selective extraction maintains data completeness for the required attributes while significantly reducing overall data volume for transmission and storage.
4Manufacturing precision
If independent encoding is performed for each attribute, then encoding precision is improved, but overall processing efficiency decreases
Solution Approach 1:
The patent merges computationally intensive preprocessing operations (LOD generation and neighbor point search) into shared processes, while maintaining independent encoding stages for color and reflectance attributes. This hybrid approach preserves encoding precision through attribute-specific processing while improving overall efficiency by eliminating redundant computations.
Data Source
Figure 1
Figure 2
Figure 3
AI summary
Disclosed herein is a method of transmitting point cloud data. The method may include acquiring point cloud data, encoding geometry information including positions of points in the point cloud data, generating Level Of Details (LODs) based on the geometry information and searching neighbor points of each point to be attribute encoded based on the generated LODs, encoding first attribute information of each point based on the searched neighbor points, encoding second attribute information of each point based on the searched neighbor points, and transmitting the encoded geometry information, the encoded first and second attribute information and signaling information.