Abstract: 3D reconstruction problem from images can be classified into three strata each of which is equivalent to the estimation of a specific geometry group. The simplest being projective, then affine, next metric and finally Euclidean structure. The advantage of stratification is that the images do not need to be from calibrated cameras in order to obtain reconstruction. In this paper results for both camera calibration and reconstruction are presented to verify that it is possible to obtain a 3D model directly from features in the images for man-made world.