V-PCC Mesh Coding Without Occupancy Maps
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing video coding methods for 3D meshes, such as MPEG 3DG/PCC's Test Model for Category 2 (TMC2), inefficiently project multiple geometry and texture/attribute information onto a 2D image, leading to suboptimal compression efficiency and artifacts in dynamic textured meshes.
Innovation Solution
A novel approach that encodes 3D meshes using V-PCC video-based schemes without occupancy maps, by projecting meshes into patches and encoding edge contours, incorporating fast implicit re-meshing and filtering methods to enhance reconstructed meshes.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Loss of information
If multiple geometry and texture/attribute information are projected onto the same position of a 2D image, then compression efficiency is improved, but artifacts and loss of information occur in dynamic textured meshes
Solution Approach 1:
The patent divides the mesh into multiple patches and processes each patch separately. Instead of projecting all mesh information onto a single 2D image, the method segments the mesh surface into discrete patches that can be individually encoded and reconstructed, reducing information loss while maintaining compression efficiency
Solution Approach 2:
The patent transitions from traditional 2D projection methods to a hybrid approach that incorporates 3D mesh structure information. By utilizing depth maps and attribute maps in addition to 2D projections, the method adds dimensional information back into the reconstruction process, reducing artifacts and information loss
2Manufacturing precision
If occupancy maps are used in V-PCC encoding, then reconstruction accuracy is improved, but encoder and decoder complexity increases
Solution Approach 1:
The patent extracts and transmits only the essential border point information of patches instead of using complete occupancy maps. By taking out only the critical boundary data needed for reconstruction and transmitting this separately, the method maintains reconstruction accuracy while significantly reducing the complexity burden on encoders and decoders
Solution Approach 2:
The patent creates simplified representations of patch borders by copying only the essential geometric information needed for reconstruction. Instead of transmitting full occupancy maps, the method copies and transmits border point coordinates that can be used to reconstruct patch boundaries accurately with minimal processing complexity
3Productivity
If mesh patches are reconstructed in parallel, then processing speed is improved, but coordination and synchronization become more difficult
Solution Approach 1:
The patent performs preliminary organization of patch data during the encoding phase, arranging patch information and border points in a structured format that enables straightforward parallel reconstruction. By preparing the data structure in advance with proper indexing and ordering, the method facilitates parallel processing while minimizing coordination complexity during decoding
Data Source
AI summary
An approach extends a Video-Point Cloud Compression codec approach to store mesh information and provide a good efficiency and a good objective and subjective quality. To reduce the quantity of data that needs to be transmitted, the approach removes occupancy maps that are used to know which parts of the patches are occupied. These occupancy maps create a lot of artifacts on the borders of the patches due to the occupancy map sub-resolution used to transmit the occupancy information. Instead, the approach transmits the 3-Dimensional borders of the patches (list of the 3D points constituting the borders of the patches) and uses this information to identify which parts of the 2D patches are occupied and connect the 3D reconstructed patches according to the real 3D borders of the patches.