发表机构
University of Electronic Science and Technology of China; Tsinghua University(电子科技大学; 清华大学)
机构由 AI 辅助整理,请以论文原文为准。AI 中文总结
该研究针对扩散逆问题,提出尺度一致后验动力学方法,通过设计SDE模型、离散化算法并完成理论证明,在FFHQ和ImageNet的超分辨率等任务中取得竞争力结果。
AI 中文摘要
利用预训练扩散先验进行后验采样由条件得分控制,其中间似然分量通常难以处理。我们从理想的单参数后验SDE族出发,其中随机参数控制概率流传输与随机探索,且不改变后验边际。为获得易处理的模型,我们在重缩放的干净图像坐标中表示似然,并使用对数信噪比(log-SNR)组织得到的后验代理。通过正向算子投影扩散不确定性,得到噪声条件协方差路径,其目标趋近于干净后验。由于这些目标的端点一致性无法保证替代传输遵循它们,我们将传输与冻结目标的朗之万校正器交替进行,生成连续替代SDE。我们用外部李-特罗特分裂(Lie--Trotter splitting)和方差匹配的分步IMEX预测器离散化该模型,该预测器显式处理学习到的先验、隐式处理线性似然,并在隐式求解后处理随机创新。我们证明了理想族的边际不变性、混合与传输缺陷条件下连续替代的后验收敛性,以及离散算法的一阶弱误差界。在FFHQ和ImageNet上进行100次得分评估的实验表明,该方法在超分辨率和去模糊任务中具有竞争力的重建保真度。一项控制100张图像的 ablation研究将尺度一致性与随机增量位置、延续及校正器分配的有限步效应分离开来。另一项无噪声框内修复研究显示,仅当匹配的创新在刚性似然求解后注入时,大规模探索才能达到性能平台。
英文摘要
Pretrained diffusion models represent image distributions through a continuum of progressively smoothed distributions. This multiscale structure organizes generation from global structure to fine detail and supports high-quality, diverse samples. We exploit the same multiscale diffusion prior for linear imaging inverse problems. Rather than using the pretrained model only as a denoiser in an outer iteration, we define a surrogate likelihood whose center is aligned with the clean-image coordinate and whose covariance accounts for residual diffusion uncertainty. This construction defines an explicit surrogate posterior path, from which we derive continuous posterior dynamics. A tunable Langevin component supports target tracking and allows the amount of posterior exploration to be adapted to the application. We prove endpoint consistency and a finite-horizon tracking bound and, in the exact-score setting, first-order weak accuracy. For computation, we derive the Posterior-Dynamics Implicit--Explicit sampler (PD-IMEX), a stable method using one score evaluation per diffusion scale and an implicit data-consistency update. Experiments on deblurring, super-resolution, and inpainting show strong reconstruction quality at 100 score evaluations, coarse-grid stability, and controllable fidelity--diversity behavior.
Comments26 pages, 5 figures, 3 tables