3-D shape recovery across two views based on approximate geometric constraints

In this paper, we propose a new approach to object matching and shape reconstruction across two views. The approach relies on the topological relationship among the feature points and a crude rigidity constraint (epipolar condition) in carrying the object from the mnodel view to the scene view to a first good approximation. The correspondences are then refined by fitting an active contour model to the transferred feature points on the scene view, which automatically assumes the shape of the object on correct matching. Under the assumption of weak perspective projection, it could be shown that the establishment of such correct correspondences would result in the full determination of the relative depths of all the matched object points in 3-D space and thus recovery of the complete 3-D structure of the object.