Affine-modeled video extraction from a single motion blurred image

Re-assign community