RECOVERING 3D MOTION OF MULTIPLE OBJECTS USING ADAPTIVE HOUGH TRANSFORM

We present a method to determine 3D motion and structure of multiple o bjects from two perspective views, using adaptive Hough transform. In our method, segmentation is determined based on a 3D rigidity constrai nt. Instead of searching candidate solutions over the entire five-dime nsional translation and rotation parameter space, we only examine the two-dimensional translation space. We divide the input image into over lapping patches, and, for each sample of the translation space, we com pute the rotation parameters of patches using least-squares fit. Every patch votes for a sample in the five-dimensional parameter space. For a patch containing multiple motions, we use a redescending M-estimato r to compute rotation parameters of a dominant motion within the patch . To reduce computational and storage burdens of standard multidimensi onal Hough transform, we use adaptive Hough transform to iteratively r efine the relevant parameter space in a ''coarse-to-fine'' fashion. Ou r method can robustly recover 3D motion parameters, reject outliers of the flow estimates, and deal with multiple moving objects present in the scene. Applications of the proposed method to both synthetic and r eal image sequences are demonstrated with promising results.