3D Point Cloud Encoding via 2D Projection for Data Compression
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing methods for processing three-dimensional (3D) data, particularly point clouds, face challenges in efficient compression and transmission due to their large memory and processor resource requirements, necessitating a technique to reduce data transmission and storage needs while maintaining effective reconstruction of 3D images.
Innovation Solution
The method involves estimating normal vectors for points in a point cloud, projecting grouped points onto predetermined planes, generating two-dimensional (2D) images by packing projected patches, and encoding these images for transmission, along with projection direction information, to reconstruct the original 3D data efficiently.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If point cloud data is stored and processed in its original 3D format, then the completeness and accuracy of 3D information is preserved, but the memory and processor resources required become excessively large
Solution Approach 1:
The patent transforms 3D point cloud data into 2D image data by projecting points onto 2D planes. This dimensionality reduction allows the data to be processed and stored as conventional 2D images, dramatically reducing memory and processor requirements while preserving essential 3D geometric information through the projection process
Solution Approach 2:
The patent creates 2D image copies of the 3D point cloud data through projection. These 2D representations serve as efficient substitutes that capture the essential spatial relationships and geometric features of the original 3D data, enabling processing with standard 2D image algorithms rather than resource-intensive 3D point cloud algorithms
2Quantity of substance
If point cloud data is compressed to reduce transmission size, then data transmission and storage requirements are reduced, but the complexity of the compression algorithm increases
Solution Approach 1:
The patent applies dimensionality change as a compression strategy by converting 3D point cloud data into 2D projected images. This transformation inherently reduces data size because 2D image representations require fewer bits to encode than dense 3D point cloud data, while maintaining the essential geometric information needed for reconstruction
Solution Approach 2:
The patent changes the fundamental parameters of data representation from 3D coordinates (x, y, z) to 2D image arrays with projection direction metadata. This parameter transformation enables the use of efficient 2D image compression algorithms instead of complex 3D point cloud compression methods
3Quantity of substance
If 3D data is projected onto 2D planes for compression, then data size is reduced, but information about depth and spatial relationships may be lost
Solution Approach 1:
The patent segments the 3D point cloud data into multiple groups and projects each group onto different 2D planes. By dividing the data and using multiple projection views, the patent preserves more complete spatial relationship information compared to a single projection, as different views capture different aspects of the 3D structure
Solution Approach 2:
The patent uses dimensionality change with multiple projection planes to preserve depth information. By projecting onto different 2D planes (e.g., front, back, left, right views), the patent encodes depth and spatial relationships in the arrangement and content of multiple 2D images rather than losing them in a single projection
Data Source
AI summary
A method of encoding a three-dimensional (3D) image including a point cloud includes grouping a plurality of points included in the point cloud into at least one segment; generating patches by projecting the points included in the segment onto a predetermined plane in a first direction or a second direction; generating two-dimensional (2D) images by packing the patches; and generating and outputting a bitstream including information about a direction in which each point is projected to generate the patches and information about the 2D images.