光流匹配：将光流重构为连续输运动力学
Optical Flow Matching: Reframing Optical Flow as Continuous Transport Dynamics
把光流从逐帧离散匹配重构为学习时变速度场的连续输运问题，其三角速度协同（TVS）几何约束与 Euler ODE 求解可直接替换现有光流网络的位移回归头，为时序平滑与跨数据集泛化提供新思路。
原文依据 we introduce Optical Flow Matching (OFM), a continuous formulation that learns a time-dependent velocity field to transport pixel coordinates
摘要译文
由模型翻译 · 术语统一 · 请对照原文核对关键结论现代光流估计虽受益于近期深度神经网络架构，但仍根植于经典视觉沿袭而来的离散对应范式。多数网络推断帧间位移，只捕捉像素移动到哪里，而未刻画运动如何随时间连续演化。然而，现实世界中的物理运动遵循由底层速度场支配的平滑动力学，这在流体力学与输运理论中早已确立。为弥合这一差距，我们提出光流匹配（OFM），一种连续形式化方法，学习时变速度场，使像素坐标沿运动分布相干的轨迹输运。OFM 的关键组件是三角速度协同（TVS），一种简洁而实用的几何机制，提供稳定且具物理意义的速度构造，确保连续输运保持良定义。结合基于 Euler 的 ODE 求解器，OFM 产生时序平滑、几何一致且过程可解释的光流场。在 Sintel、KITTI 与 Spring 上的实验表明，OFM 达到最优精度、增强的时序稳定性以及显著更强的跨数据集泛化能力，推动光流估计从对应关系推断走向连续动力学推理。
英文原文摘要 1336 字符
Modern optical flow estimation, though empowered by recent deep neural architectures, remains rooted in the discrete correspondence paradigm inherited from classical vision. Most networks infer frame-to-frame displacements, capturing where pixels move but not how motion evolves continuously through time. Yet physical motion in the real world follows smooth dynamics governed by underlying velocity fields, as long established in fluid mechanics and transport theory. To bridge this gap, we introduce Optical Flow Matching (OFM), a continuous formulation that learns a time-dependent velocity field to transport pixel coordinates along motion distribution coherent trajectories. A key component of our OFM is Triangle Velocities Synergy (TVS), a neat yet pragmatic geometric mechanism that provides a stable and physically meaningful velocity construction, ensuring that continuous transport remains well-defined. Combined with an Euler-based ODE solver, OFM yields flow fields that are temporally smooth, geometrically consistent, and process-interpretable. Experiments on Sintel, KITTI, and Spring demonstrate that OFM achieves state-of-the-art accuracy, enhanced temporal stability, and notably stronger cross-dataset generalization, advancing optical flow estimation from correspondence inference to continuous dynamical reasoning.