Adaptive Patch Filtering in Video-Based Point Cloud Coding
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing point cloud compression technologies face challenges in efficiently reducing data volume while maintaining quality, particularly in video-based applications, due to the high dimensionality and complexity of point clouds.
Innovation Solution
Adaptive filtering methods are applied during patch generation, adjusting grid size parameters based on characteristics such as depth, geometry, and color statistics of connected components in point clouds to enhance compression efficiency and reduce artifacts.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Quantity of substance
If traditional point cloud compression technologies are used, then data size reduction is attempted, but compression efficiency remains insufficient for large-scale applications
Solution Approach 1:
The patent segments the point cloud into connected components (CCs) based on spatial proximity and attribute similarity. Each CC is processed independently to generate patches, enabling efficient compression while preserving the overall structure. This segmentation allows the system to handle large-scale point clouds by dividing them into manageable units that can be compressed individually.
Solution Approach 2:
The patent applies adaptive filtering with varying grid sizes to different connected components based on their local characteristics. Components with higher attribute variability receive finer grid sizes for better preservation, while uniform components use coarser grids for efficient compression. This local quality approach optimizes compression efficiency without sacrificing important geometric or attribute details.
2Measurement precision
If adaptive filtering with variable grid sizes is applied, then data representation accuracy is improved, but processing complexity increases
Solution Approach 1:
The patent dynamically adjusts the grid size parameter based on the characteristics of each connected component. Components with high attribute variability use smaller grid sizes for accurate representation, while uniform components use larger grid sizes for efficient processing. This parameter change strategy balances representation accuracy with processing complexity by adapting the filtering intensity to local data characteristics.
3Reliability
If outlier points are removed based on projection, then data quality is improved, but computational overhead increases
Solution Approach 1:
The patent performs preliminary projection of points onto a 2D plane before applying the adaptive filtering process. This preliminary action organizes the data structure to facilitate efficient outlier detection and removal. By pre-projecting points, the system can quickly identify and remove outliers based on their projected positions, reducing the computational overhead compared to methods that would require complex 3D analysis for outlier detection.
Data Source
AI summary
Provided are a method, apparatus, and non-transitory computer readable medium for adaptive filtering in patch generation for video-based point cloud coding. The method includes projecting points included in a connected component (CC) to a projection plane associated with the CC, and removing one or more outlier points to generate an updated CC; determining a first surface by selecting one or more points associated with the updated CC; adaptively filtering the one or more points selected for the first surface, by adjusting a grid size parameter based on one or more characteristics of the CC; determining a second surface by selecting one or more points associated with the updated CC; adding a patch corresponding to the filtered first surface and the second surface to a list of patches associated with an original point cloud; and coding the original point cloud based on the list of patches.