Point Cloud Projection Encoding for Spatial Correlation Compression
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing point cloud encoding technologies, such as octree-based and prediction tree-based methods, fail to fully reflect the spatial correlation of point clouds, leading to inefficient encoding and decoding due to high proportions of empty nodes and insufficient spatial correlation representation.
Innovation Solution
Projecting point clouds onto a two-dimensional regularization plane, performing regularization corrections, and encoding depth information maps to enhance spatial correlation representation, thereby improving encoding efficiency.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Device complexity
If octree-based geometric encoding is used, then point cloud data is divided into hierarchical structures, but the spatial correlation of point cloud cannot be fully reflected due to large proportion of empty nodes
Solution Approach 1:
The patent transforms the three-dimensional point cloud data into a two-dimensional regular grid structure through projection. This dimensional change converts the sparse 3D spatial distribution into a dense 2D matrix where spatial correlation is preserved and can be effectively captured through conventional image processing techniques, thereby resolving the issue of empty nodes in octree structures.
Solution Approach 2:
The patent creates a regular grid copy of the point cloud data by projecting 3D points onto a 2D plane. This copying process maintains the spatial relationships and correlation information while replacing the irregular octree structure with a systematic grid, enabling better prediction and encoding efficiency.
2Ease of operation
If prediction tree-based geometric encoding is used, then point cloud is sorted and classified by laser scanner, but the tree structure does not fully reflect the spatial correlation of point cloud
Solution Approach 1:
The patent replaces the hierarchical tree structure with a two-dimensional regular grid projection. This dimensional transformation maintains the organizational benefits of structured encoding while achieving superior spatial correlation representation through the grid's regular topology, which naturally captures neighborhood relationships in 3D space.
Solution Approach 2:
The patent changes the structural parameter from tree-based hierarchical organization to grid-based regular arrangement. This parameter change enables the use of conventional image processing algorithms for prediction and encoding, significantly improving spatial correlation capture while maintaining operational efficiency.
3Measurement precision
If three-dimensional point cloud data is directly encoded, then original spatial information is preserved, but data volume is large and transmission and storage are not conducive
Solution Approach 1:
The patent extracts the essential spatial correlation information from the three-dimensional point cloud by projecting it onto a two-dimensional regular grid. This extraction process retains the critical spatial relationships needed for encoding while discarding redundant information, thereby reducing data volume for transmission and storage while maintaining spatial information accuracy.
Solution Approach 2:
The patent reduces the dimensionality of the data representation from 3D to 2D through regular grid projection. This dimensional reduction compresses the data volume significantly while preserving the spatial correlation structure, making the data more suitable for transmission and storage without losing essential spatial information.
Data Source
AI summary
Disclosed are a point cloud encoding and decoding method and device based on a two-dimensional regularization plane projection. The encoding method includes: acquiring original point cloud data; performing two-dimensional regularization plane projection on the original point cloud data to obtain a two-dimensional projection plane structure; obtaining a plurality of pieces of two-dimensional image information according to the two-dimensional projection plane structure; and encoding the plurality of pieces of two-dimensional image information to obtain code stream information.