An Approach to Projective Reconstruction from Multiple Views

We present an original multiple views method to perform a robust and detailed 3D reconstruction of a static scene from several images taken by one or more uncalibrated cameras. Making use only of fundamental matrices we are able to combine even heterogeneous video and/or photo sequences. In particular we give a characterization of camera matrices space consistent with a given fundamental matrix and provide a straightforward bottom-up method, linear in most practical uses, to fulfil the 3D reconstruction. We also describe shortly how to integrate this procedure in a standard vision system following an incremental approach.