Grid-Based Patch Generation for Point Cloud Coding
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing point cloud coding technologies face challenges in efficiently compressing and transmitting large volumes of 3D point cloud data, particularly in applications like autonomous driving and geographic information systems, where data reduction is necessary for faster transmission and reduced storage requirements.
Innovation Solution
The method involves grid-based patch generation for point cloud coding, where voxels in the point cloud are grouped into cubic grids. If the cubic dimension exceeds a threshold, a gridded point cloud is generated, and patches are created within this gridded structure. These patches are then converted back into the original point cloud, allowing for efficient data representation and transmission.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If traditional point cloud coding methods are used, then data compression is achieved, but the complexity of patch generation increases and compression efficiency decreases
Solution Approach 1:
The patent segments the point cloud into patches based on grid structures. Each patch is defined by specific grid indices and contains points within a localized region. This segmentation enables independent processing and compression of smaller data units, reducing overall computational complexity while maintaining compression efficiency.
Solution Approach 2:
The patent transforms the point cloud from a traditional 3D representation into a grid-based 2D index structure. By mapping 3D points to 2D grid coordinates and using texture maps, the system reduces dimensional complexity and enables more efficient compression algorithms to process the data.
2Speed
If point cloud data is reduced for faster transmission, then transmission speed improves, but data integrity and quality may be compromised
Solution Approach 1:
The patent applies different compression levels and processing methods to different patches based on their importance and spatial frequency characteristics. High-frequency details are compressed more aggressively in less critical regions, while preserving quality in important areas, thus balancing transmission speed with data integrity.
Solution Approach 2:
The patent creates a compressed representation of the point cloud using grid-based patches and texture maps as copies or summaries of the original data. This compressed copy can be transmitted efficiently while the original 3D structure is reconstructed from these simplified representations, maintaining data integrity through lossless or lossy compression techniques.
Data Source
AI summary
Methods and apparatuses of grid-based patch generation for point cloud coding are provided. The method includes obtaining an input that is a point cloud; grouping one or more voxels in the point cloud into one or more grids, the one or more grids having a cubic dimension; determining whether the cubic dimension associated with the one or more grids exceeds a threshold value; in response to determining the cubic dimension exceeds the threshold value, generating a gridded point cloud; generating one or more patches in the gridded point cloud; converting the one or more patches in the gridded point cloud to one or more patches in the input point cloud; outputting the one or more patches to one or more output devices.