Polygonal Mesh Triangulation Consistency via Face-Vertex Properties
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional rendering systems face challenges in consistently indexing and triangulating polygonal mesh faces across multiple meshes representing the same object, leading to inconsistent rendering due to varying triangulations of the same face in different meshes.
Innovation Solution
A method is introduced to store and retrieve triangulations of polygonal mesh faces non-destructively by using face-vertex properties, where each vertex stores the integer count of triangles incident on it, allowing for consistent triangulation across multiple meshes by embedding a single canonical triangulation in each mesh prior to rendering.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If polygons are triangulated before processing by the rendering system, then the rendering system can process the meshes, but the triangulation may be inconsistent across multiple meshes representing the same object
Solution Approach 1:
The patent applies preliminary action by pre-computing and storing the triangulation of each polygonal face before rendering. The triangulation data is embedded in the mesh data structure during authoring, so when the mesh is loaded for rendering, the correct triangulation is already available and can be consistently applied across all meshes representing the same object, eliminating the need to re-triangulate during rendering.
Solution Approach 2:
The patent uses copying by creating a canonical triangulation from one reference mesh and then copying this triangulation data to all other meshes representing the same object. This ensures that all meshes use the same triangulation pattern, maintaining consistency across different poses and animations while avoiding the complexity of independently triangulating each mesh.
2Adaptability or versatility
If multiple polygonal meshes are authored for the same object in different poses, then the object can be animated, but the rendering system cannot consistently index vertices and triangles across the meshes
Solution Approach 1:
The patent applies universality by creating a universal triangulation scheme that works across all meshes representing the same object in different poses. The canonical triangulation serves as a universal template that can be applied to any mesh in the sequence, ensuring that vertex and triangle indexing remains consistent throughout the animation sequence, enabling both animation versatility and consistent rendering.
3Manufacturing precision
If Delaunay triangulation is performed to generate optimized triangulation, then thin or narrow triangles are avoided, but the triangulation may differ between multiple meshes
Solution Approach 1:
The patent performs Delaunay triangulation as a preliminary action during the authoring phase on a reference mesh to generate an optimized triangulation that avoids thin or narrow triangles. This high-quality triangulation is then copied to all other meshes, ensuring both the quality benefits of Delaunay triangulation and the consistency needed for rendering across multiple meshes.
Data Source
AI summary
A method, device, and computer-readable storage medium for retrieving a stored triangulation of a polygonal mesh. The method includes: receiving an input polygonal mesh that is a computer representation of a three-dimensional (3D) object; identifying one or more faces of the input polygonal mesh that have more than three edges; retrieving for each face of the one or more faces, integer counts of a number of triangles that are incident to each vertex of the face stored in face-vertex properties of the face, wherein a specific triangulation of the face is represented by the number of triangles that are incident to each vertex of the face; and generating a triangulated polygonal mesh corresponding to the input polygonal mesh by triangulating, for each face of the one or more faces, the face based on the integer counts of the number of triangles that are incident to each vertex of the face.