Level Set Methods For Image Segmentation And 3D Reconstruction

In this book level-set methods are used to deal with two problems in the computer vision field, image segmentation and surface reconstruction. For the first problem we use the level-set methods to segment image objects, which have a given parametric shape based on energy functional. We demonstrate the proposed approach on the extraction of objects with explicit shape parameterization, such as linear image segments. We also demonstrate the successful application of the proposed method to the problem of calibrating and removing camera lens distortion. For the second problem, we develop and implement a variational framework for surface reconstruction starting from multiple 2D images taken by a calibrated camera. The approach works directly in 3D Euclidean space based on a level set formulation. The proposed method is evaluated using real datasets which are obtained using experimental setup which we have built for the experiments. One important advantage of the proposed algorithm is that any available a priori information about the surface shape can be modeled easily. So the thesis investigates the incorporation of the shape a priori information in the 3D reconstruction framework