StableMaterials：通过半监督学习增强材质生成的多样性
StableMaterials: Enhancing Diversity in Material Generation via Semi-Supervised Learning
把 SDXL 的对抗式蒸馏用于潜扩散模型，可在无标注条件下扩展生成分布，这一思路可迁移到其他数据稀缺的生成任务（如特定风格纹理、领域受限图像）；其四步潜一致性蒸馏加去伪影的 tileable 采样技巧，对需要快速、无缝生成的视觉应用有直接参考价值。
原文依据 employs adversarial training to distill knowledge from existing large-scale image generation models
摘要译文
由模型翻译 · 术语统一 · 请对照原文核对关键结论我们提出 StableMaterials，一种将半监督学习与潜变量扩散模型（LDM）相结合的新方法，用于生成照片级真实感的基于物理的渲染（PBR）材质。该方法采用对抗训练从已有的大规模图像生成模型中蒸馏知识，从而最小化对标注数据的依赖并提升生成的多样性。这种蒸馏方式使生成材质的分布与来自 SDXL 模型的图像纹理分布对齐，从而能够生成初始训练数据集中不存在的新材质。此外，我们采用基于扩散的精炼模型来提升样本的视觉质量并实现高分辨率生成。最后，我们蒸馏出一个潜一致性模型，可在仅四步内完成快速生成，并提出一种新的可平铺技术，以消除通常与较少扩散步数相关的视觉伪影。我们详细介绍了 StableMaterials 的架构与训练过程，以及半监督训练在现有 LDM 框架中的集成方式。与最优方法的对比评估显示了 StableMaterials 的有效性，并凸显了其在计算机图形学及其他领域的潜在应用。StableMaterials 已公开于 https://gvecchio.com/stablematerials。
英文原文摘要 1354 字符
We introduce StableMaterials, a novel approach for generating photorealistic physically-based rendering (PBR) materials that integrate semi-supervised learning with Latent Diffusion Models (LDMs). Our method employs adversarial training to distill knowledge from existing large-scale image generation models, minimizing the reliance on annotated data and enhancing the diversity in generation. This distillation approach aligns the distribution of the generated materials with that of image textures from an SDXL model, enabling the generation of novel materials that are not present in the initial training dataset. Furthermore, we employ a diffusion-based refiner model to improve the visual quality of the samples and achieve high-resolution generation. Finally, we distill a latent consistency model for fast generation in just four steps and propose a new tileability technique that removes visual artifacts typically associated with fewer diffusion steps. We detail the architecture and training process of StableMaterials, the integration of semi-supervised training within existing LDM frameworks. Comparative evaluations with state-of-the-art methods show the effectiveness of StableMaterials, highlighting its potential applications in computer graphics and beyond. StableMaterials is publicly available at https://gvecchio.com/stablematerials.