3D Point Cloud Encoding Layered Search Range Optimization
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
There is a demand to reduce the processing amount of three-dimensional data encoding and decoding, as existing methods are inefficient in handling the large amounts of data involved in three-dimensional data representation.
Innovation Solution
A method is proposed where three-dimensional points are assigned to layers based on geometry information, and a predicted value of attribute information is calculated using selected surrounding points, with different search ranges for same and higher layers.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If three-dimensional data is represented using point cloud scheme with positions and colors stored, then the shape information is captured accurately, but the data amount becomes massive requiring compression
Solution Approach 1:
The patent segments the point cloud data into multiple layers based on depth information, with each layer containing points at similar distances from the camera. This segmentation allows selective processing and compression of different spatial regions, reducing the overall data amount while preserving essential shape information through hierarchical representation.
Solution Approach 2:
The patent extracts and processes only the necessary attribute information (such as color, material properties) for each point while separating it from the geometric data. By extracting only relevant attributes and using prediction models to reconstruct missing information, the system reduces data quantity while maintaining measurement precision for critical shape characteristics.
2Device complexity
If prediction structure is determined after layer generation, then the encoding process is simple, but the processing amount increases significantly
Solution Approach 1:
The patent performs preliminary actions by determining the prediction structure during the layer generation process itself, rather than after. The prediction relationships between points are established as layers are created, allowing the encoding process to proceed with pre-computed prediction structures. This preliminary determination significantly reduces the processing amount required in subsequent encoding steps.
Solution Approach 2:
The patent merges the layer generation process with the prediction structure determination process into a single integrated operation. By combining these steps, the system avoids separate processing phases and reduces overall computational complexity, achieving both simplicity in encoding and efficiency in processing.
3Device complexity
If uniform search range is used for all layers when selecting surrounding points, then the search process is simple, but the processing efficiency decreases for higher layers
Solution Approach 1:
The patent applies local quality by using different search ranges for different layers. For higher layers (points farther from the camera), a smaller search range is used, while lower layers use larger search ranges. This adaptive approach optimizes processing efficiency for each layer according to its specific characteristics, reducing unnecessary computations for distant points while maintaining accuracy for nearby points.
Data Source
AI summary
A three-dimensional data encoding method includes: assigning three-dimensional points to one of layers, based on items of geometry information of the three-dimensional points; searching three-dimensional points surrounding a current three-dimensional point to be encoded, to select, from the three-dimensional points, a three-dimensional point to be referred to when a predicted value of attribute information of the current three-dimensional point is calculated, the current three-dimensional point belonging to a first layer among the layers; and calculating the predicted value of the attribute information of the current three-dimensional point using the three-dimensional point selected. In the searching of the three-dimensional points, a search range for a same layer as the current three-dimensional point is different from a search range for a layer higher than the first layer.