Sensitivity Analysis of Projective Geometry 3D Reconstruction
暂无分享,去创建一个
One of the most powerful methods for 3D scene reconstruction without camera calibration is based on Projective Geometry (Coxeter, 1994; Faugeras, 1992; Mohr and Arbogast, 1991) This method is very elegant and relies on the solution of a series of non-linear equations that allow the determination of the 3D coordinates of a point given the identity and 3D position of at least 6 reference points relying on two intersecting planes. The method is an ideal testbed for application of the variance propagation methodology for performance evaluation proposed by Haralick (1994).
[1] Roger Mohr,et al. It can be done without camera calibration , 1991, Pattern Recognit. Lett..
[2] Robert M. Haralick,et al. Propagating covariance in computer vision , 1994, Proceedings of 12th International Conference on Pattern Recognition.
[3] Olivier D. Faugeras,et al. What can be seen in three dimensions with an uncalibrated stereo rig , 1992, ECCV.
[4] Josef Kittler,et al. Error Guided Design of a 3D Vision System , 1998, IEEE Trans. Pattern Anal. Mach. Intell..