Robust analysis of P3P pose estimation

la this paper, the robustness of pose estimation, which is relative to the input error, is investigated from the view of actual projects, with the P3P problem having unique solution. Firstly the relation between pose error and input error is deduced theoretically. Then we principally analyze the relation between the robustness of pose estimation which relates to measuring error of image coordinates, and intrinsic parameters of pose estimation system. Through theoretical deduction under simple condition and simulation experiments under common condition, we obtain several useful conclusions, i.e. the pose error caused by the measuring error of image coordinates is inversely proportional to camera's focus ratio; the pose error is independent of image center coordinate; however, it decreases when the interval between the control points of the target model increases, a contrary instance happens on the measuring distance. The conclusion above may provide guidance for designing a pose estimation system.

[1]  Hao Ying 3D Visual Methods for Object Pose Measurement , 2002 .

[2]  GaoXiao-Shan,et al.  Complete Solution Classification for the Perspective-Three-Point Problem , 2003 .

[3]  Robert C. Bolles,et al.  Random sample consensus: a paradigm for model fitting with applications to image analysis and automated cartography , 1981, CACM.

[4]  Zhu Feng A Note on Unique Solution Conditions of the P3P Problem , 2003 .

[5]  Xinhua Zhuang,et al.  Robust 3-D-3-D Pose Estimation , 1994, IEEE Trans. Pattern Anal. Mach. Intell..

[6]  Xinhua Zhuang,et al.  Robust 3D-3D pose estimation , 1993, 1993 (4th) International Conference on Computer Vision.

[7]  Robert M. Haralick,et al.  Analysis and solutions of the three point perspective pose estimation problem , 1991, Proceedings. 1991 IEEE Computer Society Conference on Computer Vision and Pattern Recognition.

[8]  Xinhua Zhuang,et al.  Pose estimation from corresponding point data , 1989, IEEE Trans. Syst. Man Cybern..

[9]  Jianliang Tang,et al.  Complete Solution Classification for the Perspective-Three-Point Problem , 2003, IEEE Trans. Pattern Anal. Mach. Intell..

[10]  B. Achiriloaie,et al.  VI REFERENCES , 1961 .