3D Point Cloud to 2D Image Conversion Preserving Axial Depth

Resolve Bottlenecks,
Find Innovative SolutionsGenerate Solutions

Solution Overview

Problem

Three-dimensional point cloud data conversion to two-dimensional data for identification often loses depth information, leading to inaccurate identification results due to the inability to directly input three-dimensional data into convolutional neural networks.

Innovation Solution

A method involving data creation, layering, gridding, and data processing to identify axial depth in three axes, allowing for the conversion of three-dimensional point cloud data into two-dimensional images that preserve depth information for use in convolutional neural networks.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Adaptability or versatility

If three-dimensional point cloud data is compressed and converted into two-dimensional data, then the data can be processed by convolutional neural networks, but depth information is lost

Engineering Contradiction:

Improvecompatibility with convolutional neural networkVSAvoiddepth information

Core Design Contradiction:

Adaptability or versatilityVSLoss of information

Solution Approach 1:

The patent applies dimensionality change by converting 3D point cloud data into 2D data while preserving depth information through encoding. Specifically, the method divides the 3D space into multiple depth layers and encodes depth values into the 2D representation, allowing CNNs to process the data while retaining depth characteristics through the encoded information.

Inventive Principle:

Principle #17Another dimension (Dimensionality change)

2Loss of information

If three-dimensional point cloud data is directly input into convolutional neural network, then depth information is preserved, but the network architecture cannot process three-dimensional data

Engineering Contradiction:

Improvedepth informationVSAvoidcompatibility with convolutional neural network

Core Design Contradiction:

Loss of informationVSAdaptability or versatility

Solution Approach 1:

The patent introduces an intermediary processing step that transforms 3D point cloud data into a 2D format suitable for CNN input while encoding depth information. This intermediary representation serves as a bridge between the 3D data structure and the 2D CNN architecture, allowing depth information to be preserved through encoding schemes while maintaining compatibility with standard CNN processing.

Inventive Principle:

Principle #24Intermediary (Mediator)

3Device complexity

If data processing is performed only for a certain dimension, then processing complexity is reduced, but three-dimensional information of point cloud data is reduced

Engineering Contradiction:

Improveprocessing complexityVSAvoidthree-dimensional information

Core Design Contradiction:

Device complexityVSLoss of information

Solution Approach 1:

The patent resolves this contradiction by transforming 3D data into 2D while encoding depth information, thereby reducing processing complexity to match standard 2D CNN architectures while preserving three-dimensional information through the encoded depth values embedded in the 2D representation.

Inventive Principle:

Principle #17Another dimension (Dimensionality change)

Data Source

PatentUS11348261B2Method for processing three-dimensional point cloud data

Publication Date: 2022.05.31 NATIONAL YUNLIN UNIVERSITY OF SCIENCE AND TECHNOLOGY

AI summary

A method for processing three-dimensional point cloud data includes a data creation step, a layering step, a gridding step, a data processing step and a two-dimensional image generation step, so that the three-dimensional point cloud data can be converted into a two-dimensional image, and the two-dimensional image can correspond to, identify and store the axial depth and information point features of the point cloud data in three axes.