Sensor-Aware 3D Data Encoding for Selective Point Cloud Extraction

Resolve Bottlenecks,
Find Innovative SolutionsGenerate Solutions

Solution Overview

Problem

Existing methods for representing and compressing three-dimensional data, such as point cloud data, face challenges in efficiently encoding and decoding data for applications like autonomous vehicles and infrastructure inspection, where appropriate extraction of point cloud data is desired but not adequately supported.

Innovation Solution

A method and device for encoding three-dimensional data that includes generating a bitstream with sensor information, allowing for the appropriate extraction of point cloud data by indicating the corresponding sensor, and a decoding method that utilizes this information to reconstruct the point cloud data accurately.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Productivity

If point cloud data is compressed for accumulation and transmission, then data storage and transmission efficiency is improved, but the ability to appropriately extract point cloud data for specific applications deteriorates

Engineering Contradiction:

Improvedata compression efficiencyVSAvoidextractability for applications

Core Design Contradiction:

ProductivityVSAdaptability or versatility

Solution Approach 1:

The point cloud data is divided into multiple layers with different levels of detail and accuracy. Each layer can be independently extracted based on application requirements, allowing selective decomposition of the compressed data structure to maintain both compression efficiency and extractability.

Inventive Principle:

Principle #1Segmentation

Solution Approach 2:

The patent introduces a new dimension of organization by embedding sensor identification information and hierarchical structure markers within the compressed bitstream. This additional dimensional organization enables applications to filter and extract data from specific sensors or layers without decompressing the entire dataset, resolving the contradiction between compression and extractability.

Inventive Principle:

Principle #17Another dimension (Dimensionality change)

2Quantity of substance

If massive amount of point cloud data is compressed, then storage space is reduced, but data extraction capability for specific sensors deteriorates

Engineering Contradiction:

Improvedata volumeVSAvoidsensor identification information

Core Design Contradiction:

Quantity of substanceVSLoss of information

Solution Approach 1:

Sensor identification information and metadata are extracted and embedded as separate header elements in the compressed bitstream structure. This allows receiving devices to identify and extract data from specific sensors without needing to decompress the entire point cloud dataset, preventing information loss while maintaining compression benefits.

Inventive Principle:

Principle #2Taking out (Extraction)

Solution Approach 2:

Sensor identification and categorization information is prepared and embedded in advance during the compression process. This preliminary organization of data with metadata enables receiving devices to perform selective extraction based on sensor type or application needs without losing critical identification information.

Inventive Principle:

Principle #10Preliminary action

Data Source

PatentUS12457362B2Three-dimensional data encoding method, three-dimensional data decoding method, three-dimensional data encoding device, and three-dimensional data decoding device

Publication Date: 2025.10.28 PANASONIC INTELLECTUAL PROPERTY CORP OF AMERICA

AI summary

A three-dimensional data encoding method includes: encoding point cloud data to generate encoded data; and generating a bitstream including the encoded data. The bitstream includes first information indicating a sensor corresponding to the point cloud data.