3D Point Cloud Encoding Hierarchical Weighting
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current three-dimensional data encoding methods lack accuracy in representing and transmitting point cloud data, which is essential for applications like autonomous vehicles and infrastructure inspection, due to inefficiencies in compressing and decompressing large datasets.
Innovation Solution
A method that transforms attribute information of three-dimensional points into coefficient values using hierarchical weighting calculations, separating them into high-frequency and low-frequency components, and encodes these values into a bitstream, with the option to fix weights in layers to reduce loss during transformation.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If point cloud data is compressed for transmission and storage, then data transmission efficiency is improved, but decoding accuracy deteriorates
Solution Approach 1:
The patent segments the point cloud data into multiple layers based on depth distance, with each layer containing points at similar distances from the camera. This segmentation allows differential encoding strategies to be applied to different depth regions, improving overall compression efficiency while maintaining accuracy where needed.
Solution Approach 2:
The patent applies different encoding precision to different spatial regions. Points in near-field regions (closer to camera) use higher precision encoding with more bits per point, while far-field points use lower precision encoding. This local quality approach maintains decoding accuracy for important near objects while reducing overall data volume.
Solution Approach 3:
The patent dynamically adjusts encoding parameters including bits per point, quantization precision, and attribute information retention based on the layer depth and point density. By changing these parameters adaptively across different layers, the system optimizes the balance between compression ratio and decoding accuracy.
2Quantity of substance
If compression ratio is increased to reduce data volume, then storage efficiency is improved, but reconstruction accuracy deteriorates
Solution Approach 1:
The patent divides point cloud data into multiple layers based on depth, allowing different compression ratios to be applied to different layers. This segmentation enables the system to achieve high overall compression ratios while maintaining high reconstruction accuracy in critical near-field regions.
Solution Approach 2:
The patent implements local quality by preserving higher precision for near-field points and accepting lower precision for far-field points. This approach reduces total data volume significantly while maintaining reconstruction accuracy where it matters most for applications like autonomous navigation.
Solution Approach 3:
The patent employs adaptive parameter changes including variable bits per point, dynamic quantization levels, and selective attribute information encoding. These parameter adjustments across different layers enable high compression ratios while preserving reconstruction accuracy in important regions.
Data Source
AI summary
A three-dimensional data encoding method includes: transforming pieces of attribute information of three-dimensional points included in point cloud data into coefficient values; and encoding the coefficient values to generate a bitstream. In the transforming, weighting calculation is performed hierarchically to generate the coefficient values belonging to one of layers, the weighting calculation separating each of the pieces of attribute information into a high-frequency component and a low-frequency component. In the weighting calculation, the weighting calculation is performed using weights fixed or not fixed in the layers. The bitstream includes first information indicating whether to fix the weights in the layers.