Dynamic Mesh Subdivision for Lower-Triangle 3D Volumetric Decoding
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing mesh compression standards do not effectively handle dynamic meshes with time varying connectivity information and attribute maps, leading to inefficiencies in data storage and transmission, particularly in real-time applications like AR and VR.
Innovation Solution
Implementing a square root of three subdivision method that iteratively subdivides mesh faces, inserting vertices into existing faces and forming new faces while controlling the number of subdivided faces to be less than four times the original, and differentiating boundary edges from interior edges to optimize mesh refinement.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Manufacturing precision
If mid-point subdivision method is used to refine decimated mesh sequences, then the mesh geometry can be recovered, but the granularity is large and the number of triangles increases rapidly (quadruples each iteration)
Solution Approach 1:
The patent segments the mesh refinement process into multiple levels of detail (LOD) by applying square root of three subdivision iteratively. Instead of applying one coarse mid-point subdivision, the method applies finer subdivisions at different LOD levels, allowing selective refinement of only certain mesh regions or levels, thus controlling the total number of triangles while maintaining geometry recovery precision.
Solution Approach 2:
The patent implements local quality by allowing different regions of the mesh to have different levels of refinement. By using square root of three subdivision with LOD, certain important regions can be refined to higher precision while other regions remain at coarser levels, optimizing the balance between geometry recovery and triangle count.
2Measurement precision
If multiple iterations of mid-point subdivision are applied to increase mesh detail, then the mesh resolution improves, but the computational complexity and data volume increase significantly
Solution Approach 1:
The patent introduces dynamics by making the mesh resolution adaptive through LOD levels. Instead of uniformly applying multiple iterations of mid-point subdivision across the entire mesh, the system dynamically selects which LOD levels to apply and which regions to refine, reducing computational complexity while maintaining necessary mesh resolution for visual quality.
Solution Approach 2:
The patent applies partial action by selectively refining only certain portions of the mesh to the required resolution rather than applying full subdivision iterations to the entire mesh. This is achieved through LOD-based selective refinement, where only necessary regions receive detailed processing, reducing overall computational complexity.
3Quantity of substance
If existing mesh compression standards are used, then compression can be achieved, but they do not effectively handle dynamic meshes with time varying connectivity information and attribute maps
Solution Approach 1:
The patent implements universality by creating a compression framework that handles multiple types of mesh data simultaneously - geometry, time-varying connectivity information, and attribute maps - within a single unified approach. The square root of three subdivision with LOD provides a common foundation that works across all these different data types, making the compression standard versatile for various dynamic mesh applications.
Solution Approach 2:
The patent applies preliminary action by pre-defining the square root of three subdivision structure and LOD levels before compression. This pre-established framework allows efficient encoding of dynamic mesh properties by referencing the predetermined subdivision pattern, reducing the data volume required to represent time-varying connectivity and attributes while maintaining adaptability.
Data Source
AI summary
A method and apparatus comprising computer code configured to cause a processor or processors to obtain, from a bitstream, a mesh representing an encoded volumetric data of at least one three-dimensional (3D) visual content, and decode the encoded volumetric data based on displacement vectors of vertices of the mesh, and the displacement vectors are based on a square root of three subdivision in which faces of the mesh are iteratively subdivided in a manner in which a number of subdivided faces resulting from subdividing the mesh in an iteration is less than four times a number of faces from before subdividing the mesh in the iteration.