无需训练即可实现精准图像修复，速度更快且结果更自然。
LanPaint: Training-Free Diffusion Inpainting with Asymptotically Exact and Fast Conditional Sampling
- 用精心设计的朗之万动力学实现无梯度采样
- 在多种任务中生成更连贯、条件更精确的修复结果
- 兼容快速ODE采样器，适合追求高效与质量的开发者
扩散模型在图像生成中擅长联合像素采样，但缺乏高效的无训练部分条件采样方法（如已知像素的图像修复）。以往方法通常将此问题建模为难以求解的逆问题，依赖粗略的变分近似、需昂贵反向传播的启发式损失或缓慢的随机采样。这些限制导致：(1) 修复结果分布匹配不准确，(2) 无法实现无梯度的高效推理，(3) 与快速的ODE采样器不兼容。为此，我们提出LanPaint：一种针对基于ODE和修正流扩散模型的无训练、渐近精确的部分条件采样方法。通过精心设计的朗之万动力学，LanPaint实现了无需反向传播的快速蒙特卡洛采样。实验表明，该方法在多种任务中均实现更优性能，具备精确的部分条件约束与视觉上连贯的修复效果。
原文摘要 · Abstract (English)
Diffusion models excel at joint pixel sampling for image generation but lack efficient training-free methods for partial conditional sampling (e.g., inpainting with known pixels). Prior work typically formulates this as an intractable inverse problem, relying on coarse variational approximations, heuristic losses requiring expensive backpropagation, or slow stochastic sampling. These limitations preclude: (1) accurate distributional matching in inpainting results, (2) efficient inference modes without gradient, (3) compatibility with fast ODE-based samplers. To address these limitations, we propose LanPaint: a training-free, asymptotically exact partial conditional sampling methods for ODE-based and rectified flow diffusion models. By leveraging carefully designed Langevin dynamics, LanPaint enables fast, backpropagation-free Monte Carlo sampling. Experiments demonstrate that our approach achieves superior performance with precise partial conditioning and visually coherent inpainting across diverse tasks.
Thank you to arXiv for use of its open access interoperability. PaperDance 不是 arXiv 官方产品；中文卡片由大模型生成，请以原文为准。