Algorithm 1 Training
1:	$\mathrm{Input}: \mathrm{target} \mathrm{image} X_1$$; \mathrm{velocity} \mathrm{model} v_{\theta }$$: R^d\to R^d$$\mathrm{with}\mathrm{parameter}\theta ;$ $\mathrm{Physics-Based} \mathrm{Noise} \mathrm{Model} C(X)$
2:	$\mathrm{Sample}: X_0\sim N\left(0,1\right)$
3:	$\mathrm{Sample}: T=C\left(X_1\right)$
4:	$\mathrm{Sample}: t\sim Uniform(\left[0,1\right])$
5:	$X_{\mathrm{t}}=t{X}_1+\left(1-t\right)X_0$
6:	$\stackrel{\wedge }{\theta }=\min {‖X_1-X_0-v_{\theta }\left(X_t,{\rm T},t\right)‖}_1$
7:	$d{Z}_t=v_{\stackrel{\wedge }{\theta }}\left(Z_t,T,t, \right)dt$
8:	$\mathrm{Return}: Z=\left\{Z_t:t\in \left[0, 1\right]\right\}$