3D Point Cloud Decoding With Hierarchical Resolution Output
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing technologies struggle to efficiently output three-dimensional data of varying resolutions, particularly in the context of point cloud compression and decoding.
Innovation Solution
A decoding device and encoding device that generate and output low-resolution and high-resolution three-dimensional data by restoring vertices and faces from encoded data, enabling flexible resolution output.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Quantity of substance
If point cloud compression is applied to reduce data amount, then data transmission and storage efficiency is improved, but the ability to output three-dimensional data of different resolutions becomes limited
Solution Approach 1:
The patent segments the three-dimensional data into two distinct data structures: a first data structure containing low-resolution data (first vertices and first face information) and a second data structure containing high-resolution enhancement information (second vertices and second face information). This segmentation allows the compressed data to be flexibly decoded at different resolutions by selectively processing the appropriate data structure, thereby resolving the contradiction between data compression and resolution output flexibility.
2Loss of information
If massive amount of point cloud data is stored, then three-dimensional data completeness is improved, but data transmission and processing load increases
Solution Approach 1:
The patent changes the parameter of data representation by using a hierarchical data structure where the first data structure provides a coarse representation and the second data structure provides fine-grained enhancement. By adjusting the decoding parameters to process only the necessary level of detail, the system maintains complete three-dimensional data information while reducing transmission and processing loads according to actual needs.
3Measurement precision
If high-resolution three-dimensional data is output, then data quality is improved, but data transmission and processing load increases
Solution Approach 1:
The patent implements partial action by providing a two-tier data structure where only the essential first data structure is required for basic three-dimensional representation, while the second data structure with enhancement information is optional. This allows systems to achieve adequate quality with partial data processing, or to achieve higher quality by adding the enhancement layer, thereby resolving the contradiction between data quality and processing efficiency.
Data Source
AI summary
A decoding device includes circuitry and memory coupled to the circuitry. In operation, the circuitry: restores first vertices based on encoded data; generates first face information including a first face by connecting the first vertices restored; restores second vertices based on the encoded data and the first face; generates second face information including a second face by connecting a point cloud including the first vertices restored and the second vertices restored; and outputs the first vertices, the second vertices, the first face information, and the second face information.