Point Cloud Geometry Coding via Low-Correlation Feature Vectors

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Existing methods do not provide a solution for coding the geometry of a point cloud using a coordinate-based network.

Innovation Solution

An information processing apparatus and method that generates a feature vector expressing a spatial correlation lower than the spatial correlation of a parameter vector, using a coordinate-based network, and codes or decodes this feature vector to improve coding efficiency.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Loss of information

If a parameter vector of a coordinate-based network is used to represent geometry of 3D data, then the geometry can be encoded, but the spatial correlation is high which reduces coding efficiency

Engineering Contradiction:
Improvecoding efficiencyVSAvoidspatial correlation
Core Design Contradiction:
Loss of informationVSStability of the object's composition

Solution Approach 1:

The patent introduces a feature vector as an intermediary between the parameter vector and the coded data. This feature vector is generated by processing the parameter vector through a neural network to reduce spatial correlation, thereby improving coding efficiency while preserving geometric information

Inventive Principle:
Principle #24Intermediary (Mediator)

Solution Approach 2:

The patent transforms the parameter vector into a feature vector by changing its representation parameters. This transformation reduces the spatial correlation of the data while maintaining the essential geometric information, enabling more efficient compression

Inventive Principle:
Principle #35Parameter changes

2Productivity

If the spatial correlation of the parameter vector is reduced to improve coding efficiency, then compression performance improves, but additional processing steps are required

Engineering Contradiction:
Improvecoding efficiencyVSAvoidprocessing complexity
Core Design Contradiction:
ProductivityVSDevice complexity

Solution Approach 1:

The patent employs a neural network that automatically learns and performs the transformation from parameter vector to feature vector. The system self-adapts to reduce spatial correlation without requiring manual intervention or complex external processing, thereby improving coding efficiency while keeping the added complexity manageable

Inventive Principle:
Principle #25Self-service

Data Source

PatentUS20250371741A1Information processing apparatus and method
Publication Date: 2025.12.04 SONY GROUP CORP
  • US20250371741A1 patent drawing
  • US20250371741A1 patent drawing
  • US20250371741A1 patent drawing

AI summary

There is provided an information processing apparatus and method adapted to be capable of coding a geometry of a point cloud, using a coordinate-based network. The information processing apparatus and method generate, on the basis of a parameter vector of a coordinate-based network representing a geometry of 3D data, a feature vector expressing a spatial correlation lower than a spatial correlation of the parameter vector, and code the feature vector. Another information processing apparatus and method decode coded data to generate a feature vector expressing a spatial correlation lower than a spatial correlation of a parameter vector of a coordinate-based network representing a geometry of 3D data, and generate the geometry from the feature vector. The present disclosure is applicable to, for example, an information processing apparatus, an electronic device, an information processing method, a program, or the like.