FabricGen: Microstructure-Aware Woven Fabric Generation

2026-03-07

AI Summary

Existing generative models struggle to synthesize realistic woven fabric materials from text that adhere to authentic weaving constraints and exhibit yarn-level detail. This work proposes FabricGen, an end-to-end framework that decouples macroscopic texture from microscopic weaving structure. It leverages a diffusion model to generate weaving-agnostic textures and introduces WeavingLLM—a novel, domain-finetuned large language model—to interpret weaving instructions and drive procedural geometric modeling, ensuring strict compliance with real-world weaving principles in yarn arrangement. By integrating linguistic understanding of textile specifications with physically grounded geometry generation, the method significantly enhances the richness and realism of synthesized fabrics, enabling direct use in high-quality rendering applications.

Technology Category

• Natural Language Processing: Generation
• Computer Vision: Diffusion Models for Vision
• Machine Learning: Large Multimodal Models (LMMs)

Application Category

• Graph Algorithms and Modeling for the Web: Foundation models and LLMs for Web-related graphs
• Web Mining and Content Analysis: Web data generation and simulation
• Economics, Online Markets and Human Computation: Economic ramifications for generative AI infrastructure and applications

Abstract

Woven fabric materials are widely used in rendering applications, yet designing realistic examples typically involves multiple stages, requiring expertise in weaving principles and texture authoring. Recent advances have explored diffusion models to streamline this process; however, pre-trained diffusion models often struggle to generate intricate yarn-level details that conform to weaving rules. To address this, we present FabricGen, an end-to-end framework for generating high-quality woven fabric materials from textual descriptions. A key insight of our method is the decomposition of macro-scale textures and micro-scale weaving patterns. To generate macro-scale textures free from microstructures, we fine-tune pre-trained diffusion models on a collected dataset of microstructure-free fabrics. As for micro-scale weaving patterns, we develop an enhanced procedural geometric model capable of synthesizing natural yarn-level geometry with yarn sliding and flyaway fibers. The procedural model is driven by a specialized large language model, WeavingLLM, which is fine-tuned on an annotated dataset of formatted weaving drafts, and prompt-tuned with domain-specific fabric expertise. Through fine-tuning and prompt tuning, WeavingLLM learns to design weaving drafts and fabric parameters from textual prompts, enabling the procedural model to produce diverse weaving patterns that stick to weaving principles. The generated macro-scale texture, along with the micro-scale geometry, can be used for fabric rendering. Consequently, our framework produces materials with significantly richer detail and realism compared to prior generative models.

Problem

• woven fabric generation
• microstructure-aware
• text-to-material
• yarn-level detail
• weaving rules

Innovation

• microstructure-aware generation
• diffusion model fine-tuning
• procedural geometric modeling
• WeavingLLM
• text-to-fabric synthesis