Rejection of mismatched correspondences along the affine epipolar line

Identifying the correspondences of points between images is an important ta sk. It finds applications in depth recovery, motion recovery, photogrammetr y, and robotics. Commonly, one uses the epipolar geometry to constrain the problem from a two-dimensional to a one-dimensional search. However, the es timation of the epipolar parameters itself depends on finding a sufficientl y accurate set of initial point correspondences. Mismatched correspondences present themselves as outliers in this parameter estimation. Even if the e pipolar geometry is perfectly found, it is still not sufficient to disambig uate false matches whose points lie along the epipolar line. This paper att empts to identify (and thereby reject) such mismatched correspondences, wit hin the framework of an affine projection model and the use of three images . The robust Least Median of Squares (LMedS) approach is utilized in two st ages. First, it is used to estimate the epipolar parameters and reject thos e point-correspondences whose orthogonal distance to the epipolar lines are significant. Second, points are resolved along their epipolar lines and th e LMedS is used to solve for a consistent rotation for all points using thr ee images. (C) 2000 Elsevier Science B.V. All rights reserved.