Image-based localization using Gaussian processes

Visual localization is the process of finding the location of a camera from the appearance of the images it captures. In this work, we propose an observation model that allows the use of images for particle filter localization. To achieve this, we exploit the capabilities of Gaussian Processes to calculate the likelihood of the observation for any given pose, in contrast to methods which restrict the camera to a graph or a set of discrete poses. We evaluate this framework using different visual features as input and test its performance against laser-based localization in an indoor dataset, showing that our method requires smaller particle filter sizes while having better initialization performance.

[1]  Paul Newman,et al.  FAB-MAP: Appearance-Based Place Recognition and Mapping using a Learned Visual Vocabulary Model , 2010, ICML.

[2]  J. M. M. Montiel,et al.  ORB-SLAM: A Versatile and Accurate Monocular SLAM System , 2015, IEEE Transactions on Robotics.

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

[4]  Dieter Fox,et al.  KLD-Sampling: Adaptive Particle Filters , 2001, NIPS.

[5]  Sebastian Thrun,et al.  Probabilistic robotics , 2002, CACM.

[6]  Geoffrey E. Hinton,et al.  ImageNet classification with deep convolutional neural networks , 2012, Commun. ACM.

[7]  Stefan Carlsson,et al.  CNN Features Off-the-Shelf: An Astounding Baseline for Recognition , 2014, 2014 IEEE Conference on Computer Vision and Pattern Recognition Workshops.

[8]  Luc Van Gool,et al.  SURF: Speeded Up Robust Features , 2006, ECCV.

[9]  Gordon Wyeth,et al.  Towards persistent indoor appearance-based localization, mapping and navigation using CAT-Graph , 2012, 2012 IEEE/RSJ International Conference on Intelligent Robots and Systems.

[10]  Francisco Angel Moreno,et al.  Stereo vision specific models for particle filter-based SLAM , 2009, Robotics Auton. Syst..

[11]  Javier González,et al.  Training a Convolutional Neural Network for Appearance-Invariant Place Recognition , 2015, ArXiv.

[12]  Carl E. Rasmussen,et al.  Gaussian processes for machine learning , 2005, Adaptive computation and machine learning.

[13]  Karen Spärck Jones.  A statistical interpretation of term specificity and its application in retrieval , 2021, J. Documentation.

[14]  Michael Greenspan,et al.  Visual indoor positioning with a single camera using PnP , 2015, 2015 International Conference on Indoor Positioning and Indoor Navigation (IPIN).

[15]  Gary R. Bradski,et al.  ORB: An efficient alternative to SIFT or SURF , 2011, 2011 International Conference on Computer Vision.

[16]  Dieter Fox,et al.  Gaussian Processes for Signal Strength-Based Location Estimation , 2006, Robotics: Science and Systems.

[17]  Michael Milford,et al.  Convolutional Neural Network-based Place Recognition , 2014, ICRA 2014.

[18]  Gordon Wyeth,et al.  SeqSLAM: Visual route-based navigation for sunny summer days and stormy winter nights , 2012, 2012 IEEE International Conference on Robotics and Automation.

[19]  Florian Pregizer,et al.  Coverage gaps in fingerprinting based indoor positioning: The use of hybrid Gaussian Processes , 2015, 2015 International Conference on Indoor Positioning and Indoor Navigation (IPIN).

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

[21]  Gordon Wyeth,et al.  Continuous appearance-based trajectory SLAM , 2011, 2011 IEEE International Conference on Robotics and Automation.

[22]  Florian Schweiger,et al.  TUMindoor: An extensive image and point cloud dataset for visual indoor localization and mapping , 2012, 2012 19th IEEE International Conference on Image Processing.