Bias compensation in visual odometry

Empirical evidence shows that error growth in visual odometry is biased. A projective bias model is developed and its parameters are estimated offline from trajectories encompassing loops. The model is used online to compensate for bias and thereby significantly reduces error growth. We validate our approach with more than 25 km of stereo data collected in two very different urban environments from a moving vehicle. Our results demonstrate significant reduction in error, typically on the order of 50%, suggesting that our technique has significant applicability to deployed robot systems in GPS denied environments.

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

[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]  Supun Samarasekera,et al.  Ten-fold Improvement in Visual Odometry Using Landmark Matching , 2007, 2007 IEEE 11th International Conference on Computer Vision.

[4]  Larry Matthies,et al.  Two years of Visual Odometry on the Mars Exploration Rovers: Field Reports , 2007 .

[5]  Frans C. A. Groen,et al.  Bias reduction for stereo based motion estimation with applications to large scale visual odometry , 2009, 2009 IEEE Conference on Computer Vision and Pattern Recognition.

[6]  Matthijs C. Dorst.  Distinctive Image Features from Scale-Invariant Keypoints , 2011 .

[7]  T. Clarke,et al.  The Development of Camera Calibration Methods and Models , 1998 .

[8]  Paul R. Cohen,et al.  Camera Calibration with Distortion Models and Accuracy Evaluation , 1992, IEEE Trans. Pattern Anal. Mach. Intell..

[9]  Janusz,et al.  Geometrical Methods in Robotics , 1996, Monographs in Computer Science.

[10]  Richard Szeliski,et al.  Visual odometry and map correlation , 2004, Proceedings of the 2004 IEEE Computer Society Conference on Computer Vision and Pattern Recognition, 2004. CVPR 2004..

[11]  D. Gennery,et al.  Calibration and Orientation of Cameras in Computer Vision , 2001 .

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

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

[14]  Clark F. Olson,et al.  Rover navigation using stereo ego-motion , 2003, Robotics Auton. Syst..

[15]  Peter F. Sturm,et al.  On Camera Calibration with Linear Programming and Loop Constraint Linearization , 2012, International Journal of Computer Vision.

[16]  金谷 健一.  Statistical optimization for geometric computation : theory and practice , 2005 .

[17]  Javier Civera,et al.  Camera self-calibration for sequential Bayesian structure from motion , 2009, 2009 IEEE International Conference on Robotics and Automation.

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

[19]  Paul L. Rosin.  Robust pose estimation , 1999, IEEE Trans. Syst. Man Cybern. Part B.

[20]  Peter Meer,et al.  Optimal rigid motion estimation and performance evaluation with bootstrap , 1999, Proceedings. 1999 IEEE Computer Society Conference on Computer Vision and Pattern Recognition (Cat. No PR00149).

[21]  Duane C. Brown,et al.  Close-Range Camera Calibration , 1971 .

[22]  Kurt Konolige,et al.  FrameSLAM: From Bundle Adjustment to Real-Time Visual Mapping , 2008, IEEE Transactions on Robotics.

[23]  Jan-Michael Frahm,et al.  A Comparative Analysis of RANSAC Techniques Leading to Adaptive Real-Time Random Sample Consensus , 2008, ECCV.

[24]  F. Park.  Distance Metrics on the Rigid-Body Motions with Applications to Mechanism Design , 1995 .

[25]  Frank Dellaert,et al.  Flow separation for fast and robust stereo odometry , 2009, 2009 IEEE International Conference on Robotics and Automation.

[26]  Patrick Rives,et al.  Accurate Quadrifocal Tracking for Robust 3D Visual Odometry , 2007, Proceedings 2007 IEEE International Conference on Robotics and Automation.

[27]  Pascal Fua,et al.  On benchmarking camera calibration and multi-view stereo for high resolution imagery , 2008, 2008 IEEE Conference on Computer Vision and Pattern Recognition.

[28]  Gijs Dubbelman,et al.  Intrinsic statistical techniques for robust pose estimation , 2011 .

[29]  Zhengyou Zhang,et al.  A Flexible New Technique for Camera Calibration , 2000, IEEE Trans. Pattern Anal. Mach. Intell..

[30]  James R. Bergen,et al.  Visual odometry for ground vehicle applications , 2006, J. Field Robotics.

[31]  Janne Heikkilä,et al.  A four-step camera calibration procedure with implicit image correction , 1997, Proceedings of IEEE Computer Society Conference on Computer Vision and Pattern Recognition.

[32]  Klamer Schutte,et al.  Efficient trajectory bending with applications to loop closure , 2010, 2010 IEEE/RSJ International Conference on Intelligent Robots and Systems.

[33]  Frans C. A. Groen,et al.  Accurate and robust ego-motion estimation using expectation maximization , 2008, 2008 IEEE/RSJ International Conference on Intelligent Robots and Systems.

[34]  Vincent Lepetit,et al.  View-based Maps , 2010, Int. J. Robotics Res..

[35]  Gerd Hirzinger,et al.  More accurate pinhole camera calibration with imperfect planar target , 2011, 2011 IEEE International Conference on Computer Vision Workshops (ICCV Workshops).

[36]  Richard Szeliski,et al.  Modeling the World from Internet Photo Collections , 2008, International Journal of Computer Vision.

[37]  Tobias Hanning.  High Precision Camera Calibration , 2011 .

[38]  David Nister,et al.  Bundle Adjustment Rules , 2006 .

[39]  Jiri Matas,et al.  Locally Optimized RANSAC , 2003, DAGM-Symposium.

[40]  Yakup Genc,et al.  Nonlinear Mean Shift for Robust Pose Estimation , 2007, 2007 IEEE Workshop on Applications of Computer Vision (WACV '07).

[41]  Juho Kannala,et al.  Geometric Camera Calibration , 2008, Wiley Encyclopedia of Computer Science and Engineering.