2D Plane Projection Encoding for Sparse Point Cloud 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 due to high proportions of empty nodes and insufficient entropy encoding.
Innovation Solution
Project the point cloud onto a two-dimensional regularization plane, perform regularization correction, and encode the resulting two-dimensional image information, including depth maps, to enhance spatial correlation representation and reduce redundancy.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Device complexity
If octree-based geometric encoding is used, then the point cloud can be divided into hierarchical structures, but the spatial correlation of the point cloud cannot be fully reflected due to strong spatial sparsity leading to large proportions of empty nodes
Solution Approach 1:
The patent transforms the three-dimensional point cloud data into a two-dimensional projection plane structure. By projecting points from 3D space onto a 2D plane and organizing them in raster scan order, the method creates a dimensional transformation that eliminates the sparsity problem inherent in 3D octree structures. This 2D organization allows adjacent points in the projection to be spatially correlated, enabling effective prediction and encoding while fully preserving spatial relationships.
2Adaptability or versatility
If prediction tree-based geometric encoding is used, then the tree structure can be established according to laser scanners, but the spatial correlation of the point cloud is still not fully reflected
Solution Approach 1:
The patent applies dimensional transformation by projecting 3D point cloud data onto a 2D plane and organizing it in raster scan order. This transformation creates a 2D array structure where spatially adjacent points are positioned next to each other, enabling effective prediction based on neighboring points. The method maintains adaptability to different laser scanner configurations while fully capturing spatial correlations through the 2D projection organization.
3Productivity
If traditional point cloud encoding methods are used, then the encoding process can be performed, but the encoding efficiency is insufficient due to high proportions of empty nodes and inadequate entropy encoding
Solution Approach 1:
The patent transforms 3D point cloud data into a 2D projection plane organized in raster scan order, eliminating empty nodes by projecting only actual points onto the 2D plane. This dimensional transformation creates continuous spatial relationships between adjacent points, enabling effective prediction encoding and entropy encoding. The method significantly improves encoding efficiency by removing spatial redundancy while preserving all essential point cloud 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. According to the present disclosure, the strong correlation representation of a point cloud on the two-dimensional projection plane structure is obtained, so that the spatial correlation of the point cloud is better reflected, and the encoding efficiency of the point cloud is improved. Moreover, a placeholder information map is used for assisting in encoding the depth information map, so that the encoding efficiency is further improved.