The Taub Faculty of Computer Science Events and Talks

Computing the epipolar geometry between cameras with very different viewpoints is often problematic as matching points are hard to find. In these cases, it has been proposed to use information from dynamic objects in the scene for suggesting point and line correspondences. We introduce an approach that improves by two orders of magnitude the performance over state-of-the-art methods, by significantly reducing the number of outliers in the putative matches. Our approach is based on (a) a novel method to *directly* recover corresponding epipolar lines across views without access to the fundamental matrix, and (b) formulation of the correspondences problem as constrained flow optimization, providing efficient optimal iterative solution. Our method was validated on four standard datasets providing accurate calibrations across very different viewpoints.