Estimation value for three dimension reconstruction

Abstract

This paper deals with a fundamental problem for 3D model acquisition after camera calibration [1]. We present an approach to estimate a robust fundamental matrix for camera calibration [2, 3]. Single axis motion can be described in terms of its fixed entities, those geometric objects in space or in the image that remain invariant throughout the sequence. In particular, corresponding epipolar lines between two images intersect at the image of the rotation axis. This constraint is then used to remove the outliers and provides new algorithms for the computing the fundamental matrix. In the simulation results, our method can be used to compute the fundamental matrix for camera calibration more efficiently. © 2013 Springer Science+Business Media Dordrecht(Outside the USA).