Point Cloud RAHT Axis Ordering for Higher Coding Efficiency
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing point cloud encoding methods, particularly in geometry-based point cloud compression (G-PCC), suffer from low transform efficiency due to the use of a fixed order for regional adaptive hierarchical transform (RAHT), which reduces coding efficiency.
Innovation Solution
Adaptive adjustment of the coordinate processing order during encoding and decoding by signaling the processing order in the bitstream, allowing RAHT to be performed according to a dynamically determined order rather than a fixed one, thereby enhancing transform efficiency.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Ease of manufacture
If point cloud data is encoded using conventional methods (e.g., OCV, ATC), then the encoding process is simple, but the decoded point cloud quality deteriorates due to loss of geometric information and increased quantization noise
Solution Approach 1:
The patent segments the point cloud data into multiple groups based on spatial distribution and geometric characteristics. Each group is processed independently with customized encoding parameters, allowing preservation of geometric information while maintaining encoding simplicity. This segmentation approach resolves the contradiction by enabling detailed processing without overwhelming complexity.
Solution Approach 2:
The patent applies different encoding strategies to different regions of the point cloud data based on their geometric importance and spatial distribution. Critical regions with rich geometric information receive enhanced encoding treatment, while less important regions use simpler encoding. This local quality approach maintains overall encoding simplicity while improving specific areas of point cloud quality.
2Quantity of substance
If point cloud data is compressed to reduce data volume, then storage and transmission efficiency improve, but geometric information is lost and quality deteriorates
Solution Approach 1:
The patent dynamically adjusts compression parameters based on the characteristics of each point cloud group and region. The compression ratio and quantization levels are adaptively modified to balance data volume reduction with geometric information preservation. This dynamic approach allows aggressive compression where geometric detail is less critical and more conservative compression where geometric information is essential.
Solution Approach 2:
The patent changes encoding parameters such as quantization step sizes, bit allocation, and compression ratios based on the geometric properties and spatial location of different point cloud regions. By adjusting these parameters locally, the system achieves effective compression while preserving critical geometric information, resolving the contradiction between data volume reduction and quality maintenance.
3Productivity
If conventional encoding methods are used, then processing speed is fast, but decoded point cloud quality is poor due to high quantization noise and loss of geometric details
Solution Approach 1:
The patent divides the point cloud processing into segmented operations that can be executed in parallel. By processing different spatial groups independently, the system maintains high processing speed while applying enhanced encoding to preserve geometric details. The segmentation enables optimized processing of each region without significantly increasing overall computational complexity.
Solution Approach 2:
The patent performs preliminary organization and classification of point cloud data into meaningful groups before applying encoding. This preliminary action prepares the data structure to facilitate faster and more efficient encoding while preserving geometric information. The pre-organization reduces the computational burden during encoding, maintaining processing speed while improving quality.
Data Source
Figure 1
Figure 2
Figure 3
AI summary
A point cloud encoding and decoding method, an encoder, a decoder, and a computer storage medium are provided. The encoder determines a processing order of point cloud data during point cloud encoding, where the processing order indicates a coordinate axis processing order of three-dimensional coordinates of the point cloud data and the point cloud data is all or part of data in point cloud. The encoder determines a coordinate-axis-order index corresponding to the processing order. The encoder encodes the coordinate-axis-order index and signals encoded bits into a bitstream. The encoder processes the point cloud data according to the processing order, to obtain point cloud data to-be-encoded. The encoder encodes the point cloud data to-be-encoded and signals encoded bits into the bitstream. The decoder parses a bitstream to obtain a coordinate-axis-order index. The decoder determines a processing order of point cloud data during point cloud decoding according to the coordinate-axis-order index. The decoder parses the bitstream to obtain recovered data of the point cloud data. The decoder determines a position of coordinate data of the point cloud data in a storage unit of the recovered data according to the processing order.