高效且免训练的单图像扩散模型
Efficient and Training-Free Single-Image Diffusion Models
提出一种无需训练的扩散模型，利用单张图像的多尺度图像块数据集和闭式最优去噪器计算得分函数，可复用于图像生成、风格化、对称化等任务，并支持潜空间扩散和加速技术实现快速高分辨率生成。
原文依据 we model the image using a dataset of its patches at different scales
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考虑生成图像的问题，其内部结构——由多尺度图像块的分布定义——与单张参考图像匹配。近期方法通过在单张图像上训练扩散模型来解决此问题。但即使在这种设定下，训练计算代价高昂且需要数小时优化。相反，我们使用不同尺度下图像块的数据集来对图像建模。由于该数据集有限且图像块的维度较小，噪声图像块的得分函数可通过最优闭式去噪器精确计算，无需神经网络训练。我们将这种基于图像块的去噪器集成到高效、免训练的图像扩散模型中，并描述我们的方法如何与经典的基于图像块的图像复原技术相联系。与训练过的单图像扩散模型相比，我们的方法实现了最优的生成质量和多样性，并展示了应用，包括无条件图像生成、文本引导风格化、图像对称化和重定向。此外，我们证明我们的方法与潜空间扩散兼容，并展示了多种额外加速技术，可在一秒内实现百万像素单图像生成，在数分钟内实现十亿像素生成。
英文原文摘要 1355 字符
We consider the problem of generating images whose internal structure - defined by the distribution of patches across multiple scales---matches that of a single reference image. Recent approaches address this problem by training a diffusion model on a single image. But even in this setting, training is computationally expensive and requires hours of optimization. Instead, we model the image using a dataset of its patches at different scales. As this dataset is finite and the dimensionality of its patches is small, the score function for a noisy patch can be computed tractably using an optimal, closed-form denoiser, eliminating the need for neural network training. We integrate this patch-based denoiser into an efficient, training-free image diffusion model, and we describe how our method connects to classical patch-based image restoration techniques. Our approach achieves state-of-the-art generation quality and diversity compared to trained single-image diffusion models, and we demonstrate applications, including unconditional image generation, text-guided stylization, image symmetrization, and retargeting. Further, we show that our approach is compatible with latent space diffusion, and we show multiple additional acceleration techniques to achieve megapixel single-image generation in one second, and gigapixel generation in minutes.