3D Mesh Retopology via Template Alignment for Animation
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing technologies face challenges in generating three-dimensional (3D) meshes for virtual characters with good topology, leading to poor animation quality and high computational loads.
Innovation Solution
A system and method that involve obtaining a source 3D mesh, generating a second 3D mesh with fewer polygons, determining a 2D parameterization, aligning landmarks with a template 3D mesh, and reconstructing the parameterization into a target 3D mesh with improved topology.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Extent of automation
If a 3D mesh is generated automatically using generative machine learning techniques, then the mesh generation process is automated and can represent surface features accurately, but the topology is poor and animation quality is compromised
Solution Approach 1:
The patent introduces a template 3D mesh as an intermediary that guides the retopology process. The template serves as a mediator between the source mesh (with good surface representation) and the target mesh (with good topology). By projecting the source mesh onto the template and using the template's vertex set as constraints, the system automatically generates a retopologized mesh that combines the surface accuracy of the source with the topological quality of the template, thus resolving the contradiction between automation and topology precision.
2Manufacturing precision
If a 3D mesh with many polygons is used to represent surface features accurately, then the mesh can capture detailed geometry, but the computational load increases and animation performance decreases
Solution Approach 1:
The patent extracts only the necessary geometric information from the source mesh by projecting it onto the template mesh. Instead of using all polygons from the high-poly source mesh directly for animation, the system extracts key geometric data and redistributes it across the template's vertex set. This extraction process creates a retopologized mesh that maintains surface feature accuracy while using fewer polygons, thereby improving animation performance and reducing computational load.
3Productivity
If the number of polygons in a 3D mesh is reduced to decrease computational load, then animation performance improves, but the ability to represent detailed surface features deteriorates
Solution Approach 1:
The patent applies local quality by concentrating geometric detail where needed through the projection process. When the source mesh is projected onto the template, areas with complex surface features automatically receive more attention and retain higher detail, while flat or simple areas use fewer polygons. This localized distribution of geometric quality allows the retopologized mesh to maintain surface representation accuracy in critical areas while reducing overall polygon count for better computational efficiency.
Data Source
AI summary
Implementations relate to methods, systems, and computer-readable media to automatically perform rigging of physics based three dimensional objects. In some implementations, the method may include obtaining a source three-dimensional (3D) mesh of a face of an avatar, wherein the source 3D mesh includes a first plurality of polygons, generating a second 3D mesh based on the source 3D mesh, wherein a count of polygons in the second 3D mesh is fewer than a count of polygons in the source 3D mesh, determining a two-dimensional (2D) parameterization of the second 3D mesh, determining a target 2D parameterization based on a template 3D mesh and the 2D parameterization of the second 3D mesh such that one or more landmarks identified in the source 3D mesh are aligned with corresponding landmarks on the template 3D mesh, and reconstructing the target 2D parameterization into a target 3D mesh.