Numerical Methods for Model-Based Pose Recovery

In this paper we review and compare several techniques for model-based pose recovery (extrinsic camera calibration) from monocular images. We classify the solutions reported in the literature as analytical perspective, affine and numerical perspective. We also present reformulations for two of the most important numerical perspective solutions: Lowe''s algorithm and Phong-Horaud''s algorithm. Our improvement to Lowe''s algorithm consists of eliminating some simplifying assumptions on its projective equations. A careful experimental evaluation reveals that the resulting fully projective algorithm has superexponential convergence properties for a wide range of initial solutions and, under realistic usage conditions, it is up to an order of magnitude more accurate than the original formulation, with arguably better computation-time properties. Our extension to Phong-Horaud''s algorithm is, to the best of our knowledge, the first method for independent orientation recovery that actually exploits the theoretical advantages of point correspondences over line correspondences. We show that in the context of a specific real-life application (visual navigation), it is either more accurate than other similar techniques with the same computational cost, or more efficient with the same accuracy.

[1]  Michael Brady,et al.  On the Appropriateness of Camera Models , 1996, ECCV.

[2]  Olivier D. Faugeras,et al.  Monocular pose determination from lines: critical sets and maximum number of solutions , 1993, Proceedings of IEEE Conference on Computer Vision and Pattern Recognition.

[3]  David W. Jacobs.  Optimal matching of planar models in 3D scenes , 1991, Proceedings. 1991 IEEE Computer Society Conference on Computer Vision and Pattern Recognition.

[4]  Vishvjit S. Nalwa,et al.  A guided tour of computer vision , 1993 .

[5]  E. Mjolsness,et al.  Online computation of exterior orientation with application to hand-eye calibration , 1996 .

[6]  Stéphane Christy,et al.  Euclidean Reconstruction: From Paraperspective to Perspective , 1996, ECCV.

[7]  Michael Brady,et al.  Closing the loop on multiple motions , 1995, Proceedings of IEEE International Conference on Computer Vision.

[8]  D. W. Thompson,et al.  Three-dimensional model matching from an unconstrained viewpoint , 1987, Proceedings. 1987 IEEE International Conference on Robotics and Automation.

[9]  Geoffrey D. Sullivan,et al.  A Filter for Visual Tracking Based on a Stochastic Model for Driver Behaviour , 1996, ECCV.

[10]  Masayoshi Kakikura,et al.  A 3-D Sensor System for Teaching Robot Paths and Environments , 1987 .

[11]  Olivier D. Faugeras,et al.  Determination of Camera Location from 2-D to 3-D Line and Point Correspondences , 1990, IEEE Trans. Pattern Anal. Mach. Intell..

[12]  Kiriakos N. Kutulakos,et al.  Affine object representations for calibration-free augmented reality , 1996, Proceedings of the IEEE 1996 Virtual Reality Annual International Symposium.

[13]  Francis L. Merat,et al.  Introduction to robotics: Mechanics and control , 1987, IEEE J. Robotics Autom..

[14]  Pietro Perona,et al.  Dynamic rigid motion estimation from weak perspective , 1995, Proceedings of IEEE International Conference on Computer Vision.

[15]  Roger Y. Tsai,et al.  A versatile camera calibration technique for high-accuracy 3D machine vision metrology using off-the-shelf TV cameras and lenses , 1987, IEEE J. Robotics Autom..

[16]  Tao Daniel Alter.  3-D Pose from 3 Points Using Weak-Perspective , 1994, IEEE Trans. Pattern Anal. Mach. Intell..

[17]  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.

[18]  S. Altmann.  Rotations, Quaternions, and Double Groups , 1986 .

[19]  Charles V. Stewart,et al.  MINPRAN: A New Robust Estimator for Computer Vision , 1995, IEEE Trans. Pattern Anal. Mach. Intell..

[20]  金谷 健一.  Group-theoretical methods in image understanding , 1990 .

[21]  Richard A. Volz,et al.  Estimating 3-D location parameters using dual number quaternions , 1991, CVGIP Image Underst..

[22]  Mongi A. Abidi,et al.  A New Efficient and Direct Solution for Pose Estimation Using Quadrangular Targets: Algorithm and Evaluation , 1995, IEEE Trans. Pattern Anal. Mach. Intell..

[23]  Ian D. Reid,et al.  Recursive Affine Structure and Motion from Image Sequences , 1994, ECCV.

[24]  Helder Araújo,et al.  A Fully Projective Formulation to Improve the Accuracy of Lowe's Pose-Estimation Algorithm , 1998, Comput. Vis. Image Underst..

[25]  Joseph S.-C. Yuan.  A general photogrammetric method for determining object position and orientation , 1989, IEEE Trans. Robotics Autom..

[26]  Edward M. Riseman,et al.  Hybrid weak-perspective and full-perspective matching , 1992, Proceedings 1992 IEEE Computer Society Conference on Computer Vision and Pattern Recognition.

[27]  Eric Krotkov,et al.  Position estimation from outdoor visual landmarks for teleoperation of lunar rovers , 1996, Proceedings Third IEEE Workshop on Applications of Computer Vision. WACV'96.

[28]  Editors , 1986, Brain Research Bulletin.

[29]  Larry S. Davis,et al.  Pose Determination of a Three-Dimensional Object Using Triangle Pairs , 1988, IEEE Trans. Pattern Anal. Mach. Intell..

[30]  Claus B. Madsen.  Viewpoint variation in the noise sensitivity of pose estimation , 1996, Proceedings CVPR IEEE Computer Society Conference on Computer Vision and Pattern Recognition.

[31]  Alan M. McIvor,et al.  An Analysis of Lowe's Model-based Vision System , 1988, Alvey Vision Conference.

[32]  David G. Lowe,et al.  Three-Dimensional Object Recognition from Single Two-Dimensional Images , 1987, Artif. Intell..

[33]  David G. Lowe,et al.  Fitting Parameterized Three-Dimensional Models to Images , 1991, IEEE Trans. Pattern Anal. Mach. Intell..

[34]  S. Sitharama Iyengar,et al.  A New Generalized Computational Framework for Finding Object Orientation Using Perspective Trihedral Angle Constraint , 1994, IEEE Trans. Pattern Anal. Mach. Intell..

[35]  Demetri Terzopoulos,et al.  Real-time computer vision , 1995 .

[36]  Radu Horaud,et al.  An analytic solution for the perspective 4-point problem , 1989, Proceedings CVPR '89: IEEE Computer Society Conference on Computer Vision and Pattern Recognition.

[37]  W. Eric L. Grimson,et al.  Recognizing 3D objects from 2D images: an error analysis , 1992, Proceedings 1992 IEEE Computer Society Conference on Computer Vision and Pattern Recognition.

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

[39]  Larry S. Davis,et al.  Iterative Pose Estimation Using Coplanar Feature Points , 1996, Comput. Vis. Image Underst..

[40]  Michael Brady,et al.  Ground Plane Motion Camera Models , 1996, ECCV.

[41]  Alistair J. Bray.  Tracking objects using image disparities , 1990, Image Vis. Comput..

[42]  Fadi Dornaika,et al.  Object pose: links between paraperspective and perspective , 1995, Proceedings of IEEE International Conference on Computer Vision.

[43]  Ehud Rivlin,et al.  Optimal servoing for active foveated vision , 1996, Proceedings CVPR IEEE Computer Society Conference on Computer Vision and Pattern Recognition.

[44]  Helder Araujo,et al.  A Fully Projective Formulation for Lowe's Tracking Algorithm , 1996 .

[45]  Thierry Pun,et al.  Combining Multiple Motion Estimates for Vehicle Tracking , 1996, ECCV.

[46]  David W. Murray,et al.  A unifying framework for structure and motion recovery from image sequences , 1995, Proceedings of IEEE International Conference on Computer Vision.

[47]  F McLauchlanPhilip,et al.  Active Camera Calibration for a Head-Eye Platform Using the Variable State-Dimension Filter , 1996 .

[48]  Geoffrey D. Sullivan,et al.  Model based perspective inversion , 1989, Image Vis. Comput..