Indoor localization with occlusion removal

A novel 3D image-based indoor localization system integrated with an obstacle removal component is proposed. In contrast with existing state-of-the-art localization techniques focusing on static outdoor or indoor environments, the adverse effects generated by moving obstacles, which are very common in busy indoor spaces, is considered in our work. In particular, this problem is converted into a separation of moving foreground and static background. We use a low-rank and sparse matrix decomposition approach to solve this problem efficiently. Our system has been tested on data sets established to emphasize the dynamic situations caused by deforming obstructions appearing in front of a static background scene that may contain useful features for localization. We demonstrate that the localization effectiveness is increased significantly after removing the dynamic occluding objects. The performance of our system is evaluated based on quantitative experimental results.

[1]  Dacheng Tao,et al.  GoDec: Randomized Lowrank & Sparse Matrix Decomposition in Noisy Case , 2011, ICML.

[2]  David G. Lowe,et al.  Fast Approximate Nearest Neighbors with Automatic Algorithm Configuration , 2009, VISAPP.

[3]  Torsten Sattler,et al.  Scalable 6-DOF Localization on Mobile Devices , 2014, ECCV.

[4]  Xueming Qian,et al.  Image Location Estimation by Salient Region Matching , 2015, IEEE Transactions on Image Processing.

[5]  Torsten Sattler,et al.  Fast image-based localization using direct 2D-to-3D matching , 2011, 2011 International Conference on Computer Vision.

[6]  Roberto Cipolla,et al.  An Image-Based System for Urban Navigation , 2004, BMVC.

[7]  Jana Kosecka,et al.  Qualitative image based localization in indoors environments , 2003, 2003 IEEE Computer Society Conference on Computer Vision and Pattern Recognition, 2003. Proceedings..

[8]  Torsten Sattler,et al.  Improving Image-Based Localization by Active Correspondence Search , 2012, ECCV.

[9]  Jan-Michael Frahm,et al.  From structure-from-motion point clouds to fast location recognition , 2009, 2009 IEEE Conference on Computer Vision and Pattern Recognition.

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

[11]  Horst Bischof,et al.  From structure-from-motion point clouds to fast location recognition , 2009, CVPR.

[12]  Yi Ma,et al.  Robust principal component analysis? , 2009, JACM.

[13]  David Nistér,et al.  Scalable Recognition with a Vocabulary Tree , 2006, 2006 IEEE Computer Society Conference on Computer Vision and Pattern Recognition (CVPR'06).

[14]  Patrick Pérez,et al.  Dense estimation and object-based segmentation of the optical flow with robust techniques , 1998, IEEE Trans. Image Process..

[15]  Noah Snavely.  Photo Tourism : Exploring image collections in 3D , 2006 .

[16]  Wei Zhang,et al.  Image Based Localization in Urban Environments , 2006, Third International Symposium on 3D Data Processing, Visualization, and Transmission (3DPVT'06).

[17]  Richard Szeliski,et al.  City-Scale Location Recognition , 2007, 2007 IEEE Conference on Computer Vision and Pattern Recognition.

[18]  Dacheng Tao,et al.  Bilateral random projections , 2012, 2012 IEEE International Symposium on Information Theory Proceedings.

[19]  Daniel P. Huttenlocher,et al.  Location Recognition Using Prioritized Feature Matching , 2010, ECCV.

[20]  Ian T. Jolliffe,et al.  Principal Component Analysis , 2002, International Encyclopedia of Statistical Science.

[21]  Torsten Sattler,et al.  Exploiting Spatial and Co-visibility Relations for Image-Based Localization , 2016, Large-Scale Visual Geo-Localization.