3D Character Generation via Base Mesh Shrink-Wrapping
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current methods for generating three-dimensional (3D) characters are labor-intensive and resource-intensive, requiring significant technical skill and human involvement, making it challenging to produce thousands of unique characters efficiently, especially in industries like film and gaming.
Innovation Solution
A computing device standardizes 3D models through a standardization procedure, applies a base mesh with material IDs, extracts polygonal mesh pieces, generates composite meshes and texture maps, and uses machine learning to ensure uniqueness, allowing for rapid generation of unique 3D characters from a limited number of source models.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Manufacturing precision
If sculpting/modeling and blend-shapes/morphing methods are used to create unique 3D characters, then character uniqueness and detail quality are improved, but labor intensity and resource consumption increase significantly
Solution Approach 1:
The patent segments the character generation process into distinct modules: base mesh selection, shrink-wrapping source models to the base mesh, material ID extraction, and composite mesh assembly. This segmentation allows automated processing of each stage, dramatically improving productivity while maintaining quality through systematic refinement at each step.
Solution Approach 2:
The patent creates copies of source 3D models through shrink-wrapping onto a standardized base mesh. Multiple source models are copied and their features are extracted and recombined to generate unique composite characters, eliminating the need for manual sculpting while preserving detailed characteristics from the source models.
2Manufacturing precision
If LIDAR, photogrammetry, and hand scanning are used to capture human models, then character realism and detail are improved, but the number of scans required and processing resources increase considerably
Solution Approach 1:
The patent merges features from multiple shrink-wrapped source models into a single composite character. By combining geometric data, texture maps, and material properties from different source models, the system generates realistic characters without requiring an equivalent number of individual scans for each character produced.
Solution Approach 2:
The base mesh serves as a universal template that can accept and integrate features from any number of source models. This multi-functional approach allows the same base structure to generate diverse characters by swapping and recombining extracted features, reducing the need for numerous unique scans.
3Manufacturing precision
If manual refinement processes are applied to scan data, then production quality is improved, but human involvement and time requirements increase significantly
Solution Approach 1:
The patent replaces manual mechanical refinement processes with automated computational algorithms. The shrink-wrapping algorithm automatically refines source models to match the base mesh topology, and machine learning models automatically refine texture maps and material properties, eliminating time-consuming manual intervention while maintaining high production quality.
Solution Approach 2:
The patent changes the refinement process from manual parameter adjustment to automated parameter optimization. By using algorithms to automatically adjust mesh density, texture resolution, and material properties based on predefined quality criteria, the system achieves production quality standards without human time investment.
4Adaptability or versatility
If thousands of unique characters are needed for film or gaming production, then narrative depth and consumer engagement are improved, but the feasibility of acquiring sufficient scan data deteriorates
Solution Approach 1:
The patent adds a combinatorial dimension to character generation. Instead of acquiring one scan per character, the system uses a limited set of source model scans and creates variety through combinatorial assembly of their features. This dimensional shift from quantity-based to combination-based generation makes producing thousands of unique characters feasible with minimal scan data.
Data Source
AI summary
Aspects described herein relate to three-dimensional (3D) characters and rapidly generating 3D characters from a plurality of 3D source models. In generating the 3D characters, one or more 3D source models may be standardized, applied to a base mesh with material ID assignments, and decomposed to isolate particular polygonal mesh pieces and/or texture map feature selections denoted by the material IDs of the base mesh. The disparate isolated polygonal mesh pieces and/or texture map feature selections may be assembled in a modular fashion to compose unique 3D characters unlike any of those of the one or more 3D models. The 3D characters may be further refined through the addition of features and/or accessories, and may also be processed through machine learning algorithms to further ascertain uniqueness.