快速且忠实:面向反问题的原则性条件流匹配
FIRM: Flow-based Imaging via Regularized Minimization
- WashU(圣路易斯华盛顿大学)
- UW-Madison(威斯康星大学麦迪逊分校)
机构由 AI 辅助整理,请以论文原文为准。
AI总结:
针对成像反问题,提出一种原则性条件流匹配参数化,显式嵌入数据一致性,实现端到端训练,以50倍更少函数评估达到最先进PSNR,并支持测试时失真-感知权衡控制。
AI中文摘要:
流匹配方法在成像反问题中通常以两种方式纳入测量信息。基于条件的方法将测量衍生的信息作为网络输入,通常通过拼接方式实现,而基于推理引导的方法则将无条件速度场与单独的数据一致性更新相结合。在这些常见公式中,前向模型并未在学习的条件速度场中被显式强制执行。我们提出了一种测量条件速度场的原则性参数化方法来解决反问题。在线性插值下,我们将条件速度$v(x_t,t,y)$表示为后验均值$E[x_1 | x_t,y]$的函数,并将该均值刻画为一个变分目标的唯一最小化器,其中数据一致性项是显式的。我们进一步证明了该速度场定义了一个从源分布到测量条件后验的概率流。将变分目标拆分,得到一个具有算子相关数据一致性更新的条件速度参数化,我们在流匹配目标下进行端到端训练,推理时无需额外引导。我们的方法在PSNR上达到了最先进的水平,并且与最强的流基线相比,函数评估次数减少了$50\ imes$。改变采样步数可以在不重新训练的情况下,在测试时控制失真-感知权衡。
英文摘要:
Flow matching methods for imaging inverse problems typically incorporate measurements through network conditioning or guidance during sampling. Neither approach explicitly applies the forward operator within the learned conditional velocity field. We develop a principled measurement-conditional velocity parameterization that does. For a linear interpolation path, we express the optimal velocity through the posterior mean $E[x_1|x_t, y]$ and show that this mean is the unique minimizer of a variational objective with an explicit data-consistency term. The velocity defined by this minimizer provably transports the source distribution to the measurement-conditioned posterior. This result leads to a forward operator-aware velocity field that is trained end-to-end and requires no separate guidance during sampling. Across five imaging tasks, our method achieves leading reconstruction quality with up to $50\times$ fewer network evaluations than competitive flow-based methods. Varying the number of sampling steps also controls the distortion-perception trade-off without retraining.