Point Cloud Decoding by Compression Unit Type Selection
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Point cloud media encoding and decoding face challenges due to large data volume and redundancy, necessitating improved flexibility and efficiency in encoding and decoding processes.
Innovation Solution
The method involves encoding and decoding point cloud media by identifying and selectively decoding compression units based on type information, such as geometry headers and attribute slices, using a TLV format to enhance flexibility and reduce computational resources.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If all compression units in point cloud media data are decoded, then complete point cloud data is obtained, but decoding computational cost and time increase significantly
Solution Approach 1:
The patent extracts and identifies specific compression units that need to be decoded based on type information in media file data boxes. By selectively extracting only necessary compression units (geometry headers, attribute slices, etc.) rather than decoding all compression units, the system reduces decoding time and computational cost while obtaining the required point cloud data for specific viewing regions or attributes.
Solution Approach 2:
The patent implements partial decoding by determining whether each compression unit needs to be decoded based on type information and specific application requirements. Instead of performing excessive full decoding of all compression units, the system performs only the necessary partial decoding of relevant compression units, thereby reducing computational overhead while maintaining the quality and completeness of required point cloud data.
2Productivity
If point cloud media data is compressed and transmitted, then data transmission efficiency improves, but data redundancy and large transmitted data volume persist
Solution Approach 1:
The patent segments point cloud data into multiple compression units with different types (geometry headers, attribute slices, point cloud slices) and organizes them in a hierarchical structure within media file data boxes. This segmentation allows the system to transmit only the necessary segments based on type information, reducing overall data volume while maintaining transmission efficiency by enabling selective decoding of required segments.
3Productivity
If flexible decoding of specific compression units is implemented, then decoding efficiency improves, but system complexity increases due to type information processing
Solution Approach 1:
The patent uses type information (acting as identifiers or 'colors') to categorize different compression units into distinct types such as geometry headers, attribute slices, and point cloud slices. This classification system enables the decoding system to quickly identify and process only the relevant compression unit types based on type information, improving decoding efficiency while managing system complexity through structured categorization rather than unstructured processing.
Data Source
AI summary
In a method for decoding point cloud media, point cloud media data including a plurality of point cloud samples encapsulated in one or more tracks is received. A compression unit of a point cloud sample in the plurality of point cloud samples is obtained. A media file data box of the point cloud sample includes type information that indicates a type of the compression unit. The type of the compression unit is one of a geometry header indicating a parameter set of geometry information, an attribute header indicating a parameter set of attribute information, a geometry slice indicating point cloud slice data of the geometry information, and an attribute slice indicating point cloud slice data of the attribute information. Whether the compression unit is to be decoded is determined according to the type information. The compression unit is decoded to obtain point cloud data based on the determination.