Comparison between asymptotic Bayesian approach and Kalman filter-based technique for 3D reconstruction using an image sequence

Reconstructing 3D informations of a scene from a sequence of 2D images is an important problem in computer vision. This paper compares two statistical approaches for 3D reconstruction from an image sequence: the asymptotic Bayesian surface reconstruction and the Kalman filter-based depth estimation. Both techniques are recursive algorithms where relevant information contained in previously taken images are summarized in a prior term (prior to the taking of the next image), which means that the reconstruction results are based upon informations from all images but the storage and computation required do not grow dramatically. The experiments with both real images and computer generated images demonstrate that the asymptotic Bayesian approach achieve better results than the Kalman filter-based approach does, mainly due to the better problem formulation.