Evaluating Pose Estimation Methods for Stereo Visual Odometry on Robots

Structure-From-Motion (SFM) methods, using stereo data, are among the best performing algorithms for motion estimation from video imagery, or visual odometry. Critical to the success of SFM methods is the quality of the initial pose estimation algorithm from feature correspondences. In this work, we evaluate the performance of pose estimation algorithms commonly used in SFM visual odometry. We consider two classes of techniques to develop the initial pose estimate: Absolute Orientation (AO) methods, and Perspective-n-Point (PnP) methods. To date, there has not been a comparative study of their performance on robot visual odometry tasks. We undertake such a study to measure the accuracy, repeatability, and robustness of these techniques for vehicles moving in indoor environments and in outdoor suburban roadways. Our results show that PnP methods outperform AO methods, with P3P being the best performing algorithm. This is particularly true when stereo triangulation uncertainty is high due to a wide Field of View lens and small stereo-rig baseline.

[1]  Berthold K. P. Horn,et al.  Closed-form solution of absolute orientation using unit quaternions , 1987 .

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

[3]  Christopher G. Harris,et al.  A Combined Corner and Edge Detector , 1988, Alvey Vision Conference.

[4]  S. Umeyama,et al.  Least-Squares Estimation of Transformation Parameters Between Two Point Patterns , 1991, IEEE Trans. Pattern Anal. Mach. Intell..

[5]  Robert B. Fisher,et al.  A Comparison of Four Algorithms for Estimating 3-D Rigid Transformations , 1995, BMVC.

[6]  Andrew W. Fitzgibbon,et al.  Bundle Adjustment - A Modern Synthesis , 1999, Workshop on Vision Algorithms.

[7]  Long Quan,et al.  Linear N-Point Camera Pose Determination , 1999, IEEE Trans. Pattern Anal. Mach. Intell..

[8]  Gregory D. Hager,et al.  Fast and Globally Convergent Pose Estimation from Video Images , 2000, IEEE Trans. Pattern Anal. Mach. Intell..

[9]  B. Triggs,et al.  Camera Pose Revisited -- New Linear Algorithms , 2000 .

[10]  Bernhard P. Wrobel,et al.  Multiple View Geometry in Computer Vision , 2001 .

[11]  Zhanyi Hu,et al.  A Note on the Number of Solutions of the Noncoplanar P4P Problem , 2002, IEEE Trans. Pattern Anal. Mach. Intell..

[12]  Andrew J. Davison,et al.  Real-time simultaneous localisation and mapping with a single camera , 2003, Proceedings Ninth IEEE International Conference on Computer Vision.

[13]  Lihong Zhi,et al.  A complete symbolic-numeric linear method for camera pose determination , 2003, ISSAC '03.

[14]  Manolis I. A. Lourakis,et al.  The design and implementation of a generic sparse bundle adjustment software package based on the Le , 2004 .

[15]  Jianliang Tang,et al.  Some Necessary Conditions on the Number of Solutions for the P4P Problem , 2004, IWMM/GIAE.

[16]  James R. Bergen,et al.  Visual odometry , 2004, Proceedings of the 2004 IEEE Computer Society Conference on Computer Vision and Pattern Recognition, 2004. CVPR 2004..

[17]  Kurt Konolige,et al.  Visual Odometry Using Sparse Bundle Adjustment on an Autonomous Outdoor Vehicle , 2006, AMS.

[18]  Robert M. Haralick,et al.  Review and analysis of solutions of the three point perspective pose estimation problem , 1994, International Journal of Computer Vision.

[19]  Tom Drummond,et al.  Scalable Monocular SLAM , 2006, CVPR.

[20]  Vincent Lepetit,et al.  Accurate Non-Iterative O(n) Solution to the PnP Problem , 2007, 2007 IEEE 11th International Conference on Computer Vision.

[21]  Kurt Konolige,et al.  Large-Scale Visual Odometry for Rough Terrain , 2007, ISRR.

[22]  Larry H. Matthies,et al.  Two years of Visual Odometry on the Mars Exploration Rovers , 2007, J. Field Robotics.

[23]  Andrew Howard,et al.  Real-time stereo visual odometry for autonomous ground vehicles , 2008, 2008 IEEE/RSJ International Conference on Intelligent Robots and Systems.