计算散斑图案干涉术
Computational Speckle Pattern Interferometry
将散斑干涉的成像模型分解为空间与时间因子，并借鉴 Horn-Schunck 光流框架从单帧图像恢复亚微米级位移与振动，为无精密仪器的单帧相位测量提供了可复用思路。
原文依据 we recover the scene dynamics using a reconstruction algorithm modeled after the classic Horn-Schunck method for estimating optical flow
摘要译文
由模型翻译 · 术语统一 · 请对照原文核对关键结论视觉上不可见的表面形变编码了关于场景的丰富信息，从物体的力学性质到周围环境中存在的声学振动。光学干涉技术可以揭示这些细微变化，通常通过捕获一系列测量来执行时序相移。本文提出计算散斑图案干涉术（CSPI），一种用于估计逐像素位移与运动的新型单帧方法。我们的关键洞察是，散斑图案干涉的成像模型可以分解为空间和时间因子，每个因子表示为一个向量。在对空间项进行校准后，我们使用一种仿照经典 Horn-Schunck 光流估计方法的重建算法来恢复场景动态。与传统干涉方法不同，CSPI 无需精密仪器即可执行相位步进。我们通过测量亚微米尺度的逐像素位移与运动、可视化音叉和克拉尼板的高频振动，以及从这些振动中间接恢复声音，证明了其有效性。
英文原文摘要 1205 字符
Visually imperceptible surface deformations encode rich information about a scene, from the mechanical properties of an object to the acoustic vibrations present in the surrounding environment. Optical interferometric techniques can reveal these subtle changes, typically by capturing a sequence of measurements to perform temporal phase shifting. In this paper, we introduce Computational Speckle Pattern Interferometry (CSPI), a novel single-shot approach to estimating per-pixel displacement and motion. Our key insight is that the image formation model for speckle pattern interferometry can be decomposed into spatial and temporal factors, each represented as a vector. After calibrating for the spatial term, we recover the scene dynamics using a reconstruction algorithm modeled after the classic Horn-Schunck method for estimating optical flow. Unlike traditional interferometric methods, CSPI requires no precision instrumentation to perform phase stepping. We demonstrate its effectiveness by measuring per-pixel displacements and motions at sub-micrometer scales, visualizing high-frequency vibrations of a tuning fork and a Chladni plate, and recovering sound indirectly from these vibrations.