2.5D Textile Appearance Model for Ply-Level Fiber Detail
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current computer-generated imagery (CGI) techniques fail to accurately model the appearance of woven and knitted textiles, particularly in representing micro details like threads and fibers, as they require excessive computational resources and storage, and struggle to simulate light reflection and transmission through these fabrics.
Innovation Solution
A 2.5D appearance model that uses a smooth geometric surface to simulate the appearance of woven and knitted textiles by computing properties such as two-dimensional distributions of heights, tangents, and normals, and leveraging the repetitive nature of weave patterns to account for self-shadowing and light scattering by individual fibers, without explicitly modeling each fiber.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If detailed volumetric representation with CT scanning is used to model textiles, then fiber detail accuracy is improved, but memory requirements and computational resources become excessive
Solution Approach 1:
The patent extracts only the essential geometric information (surface heights, tangents, normals) needed to represent textile appearance, discarding the excessive volumetric detail that consumes memory. Instead of modeling every fiber volume, the invention extracts 2D distribution maps of surface properties that capture the visual appearance without requiring full 3D volumetric data.
Solution Approach 2:
Instead of starting with detailed volumetric fiber data and trying to render it efficiently, the patent inverts the approach by directly computing surface appearance properties (heights, tangents, normals) from 2D weave patterns and fabric images, bypassing the need for volumetric representation entirely.
2Reliability
If volumetric description with radiative transfer modeling is used, then light interaction is simulated, but surface reflection and transmission cannot be properly simulated
Solution Approach 1:
The patent applies different modeling approaches to different aspects of light interaction: using 2D distribution maps of surface heights, tangents, and normals for surface reflection, and separate transmission models for light passing through fabric. This local quality approach allows accurate simulation of both reflection and transmission without the limitations of pure volumetric modeling.
Solution Approach 2:
The patent segments the light interaction simulation into distinct components: surface reflection (handled by 2D distribution maps of geometric properties), transmission (handled by separate optical models), and absorption. This segmentation allows each component to be modeled with the most appropriate method, achieving comprehensive accuracy.
3Measurement precision
If every fiber within the textile is explicitly modeled, then fiber-level detail is achieved, but storage requirements and computational complexity become prohibitive
Solution Approach 1:
Instead of modeling each individual fiber, the patent creates simplified 2D distribution maps that copy the statistical properties of fiber arrangements (heights, tangents, normals) at the macro scale. These maps reproduce the visual appearance and light interaction of detailed fiber structures without requiring individual fiber geometry, dramatically reducing computational complexity.
Solution Approach 2:
The patent merges the information from numerous individual fibers into aggregated 2D distribution maps that represent the collective behavior of fiber bundles. By combining multiple fiber properties into statistical distributions, the model achieves fiber-level appearance detail without the computational burden of tracking each fiber separately.
4Manufacturing precision
If classic BRDF and SVBRDF models are used, then smooth surface appearance is accurately represented, but micro details of fabric threads are not captured
Solution Approach 1:
The patent transitions from classic 2D BRDF/SVBRDF models to a 2.5D approach by introducing height maps and surface normal distributions that add vertical dimension information. This dimensional enhancement allows the model to represent both the smooth macro surface (captured by traditional BRDF) and the micro geometric details of fabric threads (captured by 2D distributions of heights, tangents, and normals).
Applied Scientific Principles
This section explains which scientific principles are used to turn an abstract innovation direction into a practical engineering solution.
Function Achieved in This Case
This approach enables a more accurate and computationally efficient simulation of textile appearances, allowing for realistic representation of light reflection and transmission, and can handle larger pieces of clothing, while reducing the need for detailed volumetric representations and fiber-by-fiber modeling.
Implementation Method 1
simulate both reflection and transmission through clothing
Implementation Method 2
simulate both reflection and transmission through clothing
Implementation Method 3
account for self-shadowing and light scattering by individual fibers
Implementation Method 4
account for self-shadowing and light scattering by individual fibers
Data Source
AI summary
Systems and methods configured to determine appearance of woven and knitted textiles at the ply-level are presented herein. Exemplary embodiments may: obtain an input pattern of a textile, the input pattern comprising a two-dimensional weave pattern; obtain appearance information, the appearance information including one or more of color, transparency, or roughness; determine ply curve geometry based on ply-level fiber details making up individual plys; generate an image simulating an appearance of the textile based on the two-dimensional weave pattern, the appearance information, and the ply curve geometry so that the image simulating the appearance of the textile takes into account the ply-level fiber details; and/or perform other operations.