Multiple motion analysis using 3d orientation steerable filters

In this paper, we address multiple motion analysis from the standpoint of orientation analysis. Using the fact that multiple motions are equivalent to multiple planes in the derivative space or in the spectral domain, we apply a new kind of 3D steerable filter in motion estimation. This new method is based on the decomposition of the sphere with a set of overlapping basis filters in the feature space. It is superior to principal axis analysis based approaches and current 3D steerability approaches in achieving high orientation resolution. Our approach is similar to the 3D Hough transform, but more efficient and robust. It also improves the performance of the expectation-maximization algorithm. Based on the spatial coherence in image sequences we use the “shift-and-subtract ” technique to localize occlusion boundaries and to track their movement in occlusion sequences. This technique can also be used to distinguish occlusion from transparency and to decompose transparency scenes into multi-layers. 1