3D Reconstruction by Gluing Pair-Wise Euclidean Reconstructions, or "How to Achieve a Good Reconstruction from Bad Images"

This paper presents a new technique for estimating a multi-view reconstruction given pair-wise Euclidean reconstructions up to rotations, translations and scales. The partial reconstructions are glued by the following three step procedure: (i) Camera rotations consistent with all reconstructions are estimated linearly. (ii) All the pair-wise reconstructions are modified according to the new rotations and refined by bundle adjustment while keeping the corresponding rotations same. (iii) The refined rotations are used to estimate both camera translations and 3D points using Second Order Cone Programming by minimizing the L1- norm. We introduce a new criterion for evaluating importance of an epipolar geometry in influence on the overall 3D geometry. The estimated importance is used to re-weight data in the above algorithm to better handle unequiponderantly captured images. The performance of the proposed method is demonstrated on difficult wide base-line image sets.