3D Mesh Decoding with Boundary-Aware Edge Division

Resolve Bottlenecks,
Find Innovative SolutionsGenerate Solutions

Solution Overview

Problem

Existing methods for encoding and decoding three-dimensional data are in need of further improvement to enhance the processing efficiency and accuracy.

Innovation Solution

A decoding method that involves determining the number of iterations of division for boundary and non-boundary edges in submeshes and applying different division processes to ensure proper combination of submeshes during reconstruction of three-dimensional meshes.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Ease of operation

If the same division process is applied to all edges in submeshes, then the decoding process is simple, but the submeshes cannot be properly combined during reconstruction

Engineering Contradiction:

Improvedecoding process simplicityVSAvoidsubmesh combination accuracy

Core Design Contradiction:

Ease of operationVSReliability

Solution Approach 1:

The patent applies different division processes to different types of edges based on their local characteristics. Boundary edges (shared between submeshes) undergo a first division process that preserves compatibility across submesh boundaries, while non-boundary edges undergo a second division process optimized for internal mesh regions. This local differentiation enables both simple decoding operations and accurate submesh reconstruction.

Inventive Principle:

Principle #3Local quality

2Manufacturing precision

If different division processes are applied to boundary and non-boundary edges, then submesh reconstruction accuracy is improved, but the decoding process complexity increases

Engineering Contradiction:

Improvemesh reconstruction accuracyVSAvoiddecoding process complexity

Core Design Contradiction:

Manufacturing precisionVSDevice complexity

Solution Approach 1:

The patent segments the edge set into two distinct categories: boundary edges and non-boundary edges. This segmentation allows the application of specialized division processes tailored to each category's requirements. Boundary edges receive a division process that maintains continuity across submesh boundaries, while non-boundary edges receive a different process optimized for their internal position, thereby achieving accurate reconstruction without unnecessary complexity.

Inventive Principle:

Principle #1Segmentation

Solution Approach 2:

The patent performs preliminary identification and classification of edges into boundary and non-boundary categories before applying the division process. This preliminary action enables the subsequent division operations to proceed in a structured manner, where each edge type receives the appropriate treatment in advance, simplifying the overall decoding process while ensuring accurate mesh reconstruction.

Inventive Principle:

Principle #10Preliminary action

Data Source

PatentUS20260057559A1Decoding method, encoding method, decoding device, and encoding device

Publication Date: 2026.02.26 PANASONIC INTELLECTUAL PROPERTY CORP OF AMERICA

AI summary

A decoding method according to one aspect of the present disclosure includes: decoding, from a bitstream, (1) position information of vertices forming a first submesh obtained by dividing a three-dimensional mesh and (ii) connection information regarding a connection relationship between the vertices; determining whether a first count that is a number of iterations of division performed on edges forming the first submesh and a second count that is a number of iterations of division performed on a boundary edge shared between the first submesh and a second submesh obtained by dividing the three-dimensional mesh are the same; and when the first count and the second count are different, dividing the boundary edge by performing a first division process and dividing a non-boundary edge by performing a second division process different from the first division process.