Bayesian Pose Graph Optimization via Bingham Distributions and Tempered Geodesic MCMC

We introduce Tempered Geodesic Markov Chain Monte Carlo (TG-MCMC) algorithm for initializing pose graph optimization problems, arising in various scenarios such as SFM (structure from motion) or SLAM (simultaneous localization and mapping). TG-MCMC is first of its kind as it unites asymptotically global non-convex optimization on the spherical manifold of quaternions with posterior sampling, in order to provide both reliable initial poses and uncertainty estimates that are informative about the quality of individual solutions. We devise rigorous theoretical convergence guarantees for our method and extensively evaluate it on synthetic and real benchmark datasets. Besides its elegance in formulation and theory, we show that our method is robust to missing data, noise and the estimated uncertainties capture intuitive properties of the data.

[1]  Javier González,et al.  Fast global optimality verification in 3D SLAM , 2016, 2016 IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS).

[2]  Suvrit Sra,et al.  First-order Methods for Geodesically Convex Optimization , 2016, COLT.

[3]  Hongdong Li,et al.  Rotation Averaging , 2013, International Journal of Computer Vision.

[4]  Javier González,et al.  Initialization of 3D pose graph optimization using Lagrangian duality , 2017, 2017 IEEE International Conference on Robotics and Automation (ICRA).

[5]  Arnak S. Dalalyan,et al.  Further and stronger analogy between sampling and optimization: Langevin Monte Carlo and gradient descent , 2017, COLT.

[6]  Andrea Torsello,et al.  Multiview registration via graph diffusion of dual quaternions , 2011, CVPR 2011.

[7]  Maxim Raginsky,et al.  Local Optimality and Generalization Guarantees for the Langevin Algorithm via Empirical Metastability , 2018, COLT.

[8]  Roland Badeau,et al.  Stochastic Quasi-Newton Langevin Monte Carlo , 2016, ICML.

[9]  Andrea Fusiello,et al.  Spectral Synchronization of Multiple Views in SE(3) , 2016, SIAM J. Imaging Sci..

[10]  Yudell L. Luke,et al.  Inequalities for generalized hypergeometric functions , 1972 .

[11]  Richard M. Murray,et al.  A Mathematical Introduction to Robotic Manipulation , 1994 .

[12]  Venu Madhav Govindu,et al.  Robust Relative Rotation Averaging , 2018, IEEE Transactions on Pattern Analysis and Machine Intelligence.

[13]  Jochen Trumpf,et al.  L1 rotation averaging using the Weiszfeld algorithm , 2011, CVPR 2011.

[14]  Slobodan Ilic,et al.  X-Tag: A Fiducial Tag for Flexible and Accurate Bundle Adjustment , 2016, 2016 Fourth International Conference on 3D Vision (3DV).

[15]  Joan Solà,et al.  Quaternion kinematics for the error-state Kalman filter , 2015, ArXiv.

[16]  B. Rossi,et al.  Robust Absolute Rotation Estimation via Low-Rank and Sparse Matrix Decomposition , 2014, 2014 2nd International Conference on 3D Vision.

[17]  Venu Madhav Govindu,et al.  Lie-algebraic averaging for globally consistent motion estimation , 2004, Proceedings of the 2004 IEEE Computer Society Conference on Computer Vision and Pattern Recognition, 2004. CVPR 2004..

[18]  Zhe Gan,et al.  Bridging the Gap between Stochastic Gradient MCMC and Stochastic Optimization , 2015, AISTATS.

[19]  Andrea Fusiello,et al.  Camera Motion from Group Synchronization , 2016, 2016 Fourth International Conference on 3D Vision (3DV).

[20]  Slobodan Ilic,et al.  Survey of Higher Order Rigid Body Motion Interpolation Methods for Keyframe Animation and Continuous-Time Trajectory Estimation , 2018, 2018 International Conference on 3D Vision (3DV).

[21]  Martial Hebert,et al.  Fully automatic registration of multiple 3D data sets , 2003, Image Vis. Comput..

[22]  ARNO KNAPITSCH,et al.  Tanks and temples , 2017, ACM Trans. Graph..

[23]  M. Girolami,et al.  Geodesic Monte Carlo on Embedded Manifolds , 2013, Scandinavian journal of statistics, theory and applications.

[24]  Ali Taylan Cemgil,et al.  Asynchronous Stochastic Quasi-Newton MCMC for Non-Convex Optimization , 2018, ICML.

[25]  Tianqi Chen,et al.  A Complete Recipe for Stochastic Gradient MCMC , 2015, NIPS.

[26]  Matus Telgarsky,et al.  Non-convex learning via Stochastic Gradient Langevin Dynamics: a nonasymptotic analysis , 2017, COLT.

[27]  John J. Leonard,et al.  SE-Sync: A certifiably correct algorithm for synchronization over the special Euclidean group , 2016, Int. J. Robotics Res..

[28]  Johan Fredriksson,et al.  Simultaneous Multiple Rotation Averaging Using Lagrangian Duality , 2012, ACCV.

[29]  Zhanxing Zhu,et al.  Stochastic Fractional Hamiltonian Monte Carlo , 2018, IJCAI.

[30]  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.

[31]  S. Mitter,et al.  Recursive stochastic algorithms for global optimization in R d , 1991 .

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

[33]  Xiaowei Zhou,et al.  A Survey on Rotation Optimization in Structure from Motion , 2016, 2016 IEEE Conference on Computer Vision and Pattern Recognition Workshops (CVPRW).

[34]  Vincent Lepetit,et al.  Monocular Model-Based 3D Tracking of Rigid Objects: A Survey , 2005, Found. Trends Comput. Graph. Vis..

[35]  Nassir Navab,et al.  Camera Pose Filtering with Local Regression Geodesics on the Riemannian Manifold of Dual Quaternions , 2017, 2017 IEEE International Conference on Computer Vision Workshops (ICCVW).

[36]  Ronen Basri,et al.  Stable Camera Motion Estimation Using Convex Programming , 2013, SIAM J. Imaging Sci..

[37]  Anders P. Eriksson,et al.  Rotation Averaging and Strong Duality , 2017, 2018 IEEE/CVF Conference on Computer Vision and Pattern Recognition.

[38]  Gary R. Bradski,et al.  Monte Carlo Pose Estimation with Quaternion Kernels and the Bingham Distribution , 2011, Robotics: Science and Systems.

[39]  Mert Gürbüzbalaban,et al.  Global Convergence of Stochastic Gradient Hamiltonian Monte Carlo for Non-Convex Stochastic Optimization: Non-Asymptotic Performance Bounds and Momentum-Based Acceleration , 2018, Oper. Res..

[40]  Noah Snavely,et al.  Robust Global Translations with 1DSfM , 2014, ECCV.

[41]  Christopher Bingham.  An Antipodally Symmetric Distribution on the Sphere , 1974 .

[42]  Jan-Michael Frahm,et al.  Pixelwise View Selection for Unstructured Multi-View Stereo , 2016, ECCV.

[43]  Slobodan Ilic,et al.  Online inspection of 3D parts via a locally overlapping camera network , 2016, 2016 IEEE Winter Conference on Applications of Computer Vision (WACV).

[44]  Jan-Michael Frahm,et al.  Structure-from-Motion Revisited , 2016, 2016 IEEE Conference on Computer Vision and Pattern Recognition (CVPR).

[45]  John J. Leonard,et al.  A Certifiably Correct Algorithm for Synchronization over the Special Euclidean Group , 2016, WAFR.

[46]  Yuchen Zhang,et al.  A Hitting Time Analysis of Stochastic Gradient Langevin Dynamics , 2017, COLT.

[47]  C. Hwang.  Laplace's Method Revisited: Weak Convergence of Probability Measures , 1980 .

[48]  Yijun Huang,et al.  Asynchronous Parallel Stochastic Gradient for Nonconvex Optimization , 2015, NIPS.

[49]  F. Sebastian Grassia,et al.  Practical Parameterization of Rotations Using the Exponential Map , 1998, J. Graphics, GPU, & Game Tools.

[50]  B. Leimkuhler,et al.  Molecular Dynamics: With Deterministic and Stochastic Numerical Methods , 2015 .

[51]  Eric Moulines,et al.  Stochastic Gradient Richardson-Romberg Markov Chain Monte Carlo , 2016, NIPS.

[52]  A. Morawiec,et al.  Rodrigues parameterization for orientation and misorientation distributions , 1996 .

[53]  Leslie Pack Kaelbling,et al.  Tracking the spin on a ping pong ball with the quaternion Bingham filter , 2014, 2014 IEEE International Conference on Robotics and Automation (ICRA).

[54]  Slobodan Ilic,et al.  CAD Priors for Accurate and Flexible Instance Reconstruction , 2017, 2017 IEEE International Conference on Computer Vision (ICCV).

[55]  Andrew T. A. Wood,et al.  On the derivatives of the normalising constant of the Bingham distribution , 2007 .

[56]  Venu Madhav Govindu,et al.  Combining two-view constraints for motion estimation , 2001, Proceedings of the 2001 IEEE Computer Society Conference on Computer Vision and Pattern Recognition. CVPR 2001.

[57]  Hong Cheng,et al.  Accelerated First-order Methods for Geodesically Convex Optimization on Riemannian Manifolds , 2017, NIPS.

[58]  Wolfram Burgard,et al.  G2o: A general framework for graph optimization , 2011, 2011 IEEE International Conference on Robotics and Automation.

[59]  Suvrit Sra,et al.  The multivariate Watson distribution: Maximum-likelihood estimation and other aspects , 2011, J. Multivar. Anal..

[60]  Gerhard Kurz,et al.  Recursive estimation of orientation based on the Bingham distribution , 2013, Proceedings of the 16th International Conference on Information Fusion.

[61]  Noah Snavely,et al.  When is Rotations Averaging Hard? , 2016, ECCV.

[62]  Yee Whye Teh,et al.  Bayesian Learning via Stochastic Gradient Langevin Dynamics , 2011, ICML.

[63]  Kostas Daniilidis,et al.  Statistical Pose Averaging with Non-isotropic and Incomplete Relative Measurements , 2014, ECCV.

[64]  Frank Dellaert,et al.  Initialization techniques for 3D SLAM: A survey on rotation estimation and its use in pose graph optimization , 2015, 2015 IEEE International Conference on Robotics and Automation (ICRA).

[65]  Yurii Nesterov,et al.  Introductory Lectures on Convex Optimization - A Basic Course , 2014, Applied Optimization.

[66]  Venu Madhav Govindu,et al.  On Averaging Multiview Relations for 3D Scan Registration , 2014, IEEE Transactions on Image Processing.

[67]  Radford M. Neal.  MCMC Using Hamiltonian Dynamics , 2011, 1206.1901.

[68]  Giuseppe C. Calafiore,et al.  Convex Relaxations for Pose Graph Optimization With Outliers , 2018, IEEE Robotics and Automation Letters.

[69]  Lawrence Carin,et al.  On the Convergence of Stochastic Gradient MCMC Algorithms with High-Order Integrators , 2015, NIPS.

[70]  Umut Simsekli,et al.  Fractional Langevin Monte Carlo: Exploring Levy Driven Stochastic Differential Equations for Markov Chain Monte Carlo , 2017, ICML.

[71]  Yuan Xu,et al.  Approximation Theory and Harmonic Analysis on Spheres and Balls , 2013 .

[72]  Sergei M. Sitnik,et al.  Inequalities and monotonicity of ratios for generalized hypergeometric function , 2009, J. Approx. Theory.

[73]  Yang Song,et al.  Stochastic Gradient Geodesic MCMC Methods , 2016, NIPS.

[74]  Venu Madhav Govindu,et al.  Efficient and Robust Large-Scale Rotation Averaging , 2013, 2013 IEEE International Conference on Computer Vision.