An Approach for 3D Reconstruction from Uncalibrated Images

Abstract

Three-dimensional (3D) reconstruction considering of image sequences is an unresolved and active research topic on the cross-section of computer vision and digital photogrammetry. In this paper, we focus on 3D object shape reconstruction from uncalibrated images and put forward a rapid algorithm to recover the object height. The recovering 3D shape comprises two steps, for the one hand, calculate homography transformation to obtain the outlines, for the other hand calculate the reconstructed object height by reference height proposed by the paper, and then recover the 3D object shape. This method requires no camera calibration or the estimation of the fundamental matrix; hence, it reduces the computational complexity by eliminating the requirement for abundant conjugate points. The experiment shows that the method is much validated and something useful is obtained.