Zippering SEI Message for 3D Mesh Reconstruction
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current methods for compressing 3D meshes in the V-PCC standard struggle to efficiently encode connectivity and attributes of triangle faces, leading to visual artifacts like holes and cracks between sub-meshes due to independent encoding of sub-meshes.
Innovation Solution
The introduction of a zippering SEI message that determines border points in sub-meshes and merges them using various mesh zippering implementations, such as defined search distance, maximum distance, and matching index methods, to reduce gaps and improve mesh reconstruction accuracy.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Loss of energy
If sub-meshes are encoded independently using V-PCC, then encoding efficiency and bitrate are improved, but visual artifacts such as holes and cracks appear on the mesh surface
Solution Approach 1:
The patent introduces border point identification and matching as an intermediary mechanism between independently encoded sub-meshes. By detecting border points on each sub-mesh and matching them with corresponding points on adjacent sub-meshes, the system creates a bridge that ensures geometric continuity while preserving the benefits of independent encoding. This mediator process eliminates holes and cracks without requiring full mesh dependency.
Solution Approach 2:
The patent modifies geometric parameters of border vertices by adjusting their positions to achieve precise alignment between adjacent sub-meshes. By changing the position parameters of border points based on matching criteria, the system resolves discontinuities and visual artifacts while maintaining overall mesh integrity and encoding efficiency.
2Loss of information
If mesh connectivity is encoded using traditional mesh compression methods, then connectivity information is preserved, but the method requires dense meshes and cannot efficiently handle sparse point clouds
Solution Approach 1:
The patent applies different encoding strategies to different regions of the mesh. Border points, which are critical for connectivity but constitute only a portion of the overall mesh, are identified and processed with specialized matching algorithms. This local focus allows the system to preserve connectivity information where needed while maintaining efficiency for the majority of mesh points, enabling applicability to both dense and sparse structures.
Solution Approach 2:
The patent segments the mesh into multiple independently encodable sub-meshes while identifying and preserving connectivity at the borders. By dividing the mesh into manageable segments that can be encoded separately yet connected through border point matching, the system achieves both connectivity preservation and adaptability to various mesh densities including sparse point clouds.
3Productivity
If vertices are reorganized to improve connectivity encoding, then coding efficiency is improved, but the original mesh structure and vertex order are altered
Solution Approach 1:
The patent performs preliminary identification and recording of border point positions and their corresponding relationships before final mesh reconstruction. By establishing the matching relationships in advance during the encoding process, the system maintains coding efficiency while ensuring that the original mesh structure and vertex relationships are preserved in the decoded output through the border point matching mechanism.
Data Source
AI summary
A new SEI message for the V-DMC standard is described herein, the zippering SEI. The zippering SEI message can be used by the decoder for the mesh reconstruction, where in the case of multiple sub-meshes, the zippering SEI provides ways to reduce common artifacts caused by independent sub-mesh encoding, such as holes and cracks on the mesh surface.