FabricGen：微观结构感知的机织织物生成
FabricGen: Microstructure-Aware Woven Fabric Generation
把材质生成拆成宏观纹理（微调扩散模型）与微观编织几何（LLM 驱动的程序化模型）两路，为视觉研究者提供了一种「扩散模型负责外观、程序化模型负责结构」的混合生成范式，可直接迁移到其他具有强结构规则的材质（针织、编织、木纹）生成与渲染。
原文依据 A key insight of our method is the decomposition of macro-scale textures and micro-scale weaving patterns.
摘要译文
由模型翻译 · 术语统一 · 请对照原文核对关键结论机织织物材料在渲染应用中被广泛使用，但设计出逼真的样例通常涉及多个阶段，需要编织原理与纹理创作方面的专业知识。近期的进展探索了用扩散模型来简化这一流程；然而，预训练扩散模型往往难以生成符合编织规则的复杂纱线级细节。为解决这一问题，我们提出 FabricGen，一个从文本描述生成高质量机织织物材料的端到端框架。本方法的关键洞见在于将宏观尺度纹理与微观尺度编织图案进行分解。为生成不含微观结构的宏观尺度纹理，我们在收集的无微观结构织物数据集上微调预训练扩散模型。对于微观尺度编织图案，我们开发了一个增强的程序化几何模型，能够合成带有纱线滑移与飞散纤维的自然纱线级几何。该程序化模型由一个专门的大语言模型 WeavingLLM 驱动，它在带标注的格式化编织图数据集上经过微调，并使用领域特定的织物专业知识进行提示词调优。通过微调与提示词调优，WeavingLLM 学会从文本提示词设计编织图与织物参数，使程序化模型能够生成遵循编织原理的多样编织图案。生成的宏观尺度纹理与微观尺度几何可用于织物渲染。因此，与先前的生成模型相比，我们的框架所生成的材质在细节与真实感上显著更丰富。
英文原文摘要 1653 字符
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.