Optimal Selection of Subset of Images with Highest Intra-Class Similarity For 3D Scene Reconstruction

Finding the accurate location of a mobile device based on images it acquires usually requires applying structure from motion (SFM) for 3D camera position reconstruction. Since the convergence of SFM depends on effectively selecting among the multiple retrieved images, we propose an optimization framework to do make the selection using the criterion of the highest intra-class similarity among images returned from retrieval pipeline. The selection process should consider only images with distinct GPS-tags. The selected images along with the query can be used to reconstruct a 3D scene and obtain relative camera positions. Experimental results demonstrate our method achieves a higher convergence rate in the SFM processing.

[1]  Changsheng Xu,et al.  Interaction Design for Mobile Visual Search , 2013, IEEE Transactions on Multimedia.

[2]  C. V. Jawahar,et al.  Accurate localization by fusing images and GPS signals , 2015, 2015 IEEE Conference on Computer Vision and Pattern Recognition Workshops (CVPRW).

[3]  Rashid Ansari,et al.  Improved Image Retrieval for Efficient Localization in Urban Areas Using Location Uncertainty Data , 2016, 2016 IEEE International Symposium on Multimedia (ISM).

[4]  Mubarak Shah,et al.  GPS-Tag Refinement Using Random Walks with an Adaptive Damping Factor , 2014, 2014 IEEE Conference on Computer Vision and Pattern Recognition.

[5]  Mahdi Salarian,et al.  Accurate localization in dense urban area using Google street view images , 2014, 2015 SAI Intelligent Systems Conference (IntelliSys).

[6]  Afshin Dehghan,et al.  GMCP-Tracker: Global Multi-object Tracking Using Generalized Minimum Clique Graphs , 2012, ECCV.

[7]  Toshihide Ibaraki,et al.  An Algorithm for Finding K Minimum Spanning Trees , 1981, SIAM J. Comput..

[8]  Xin Chen,et al.  City-scale landmark identification on mobile devices , 2011, CVPR 2011.

[9]  Avideh Zakhor,et al.  Location-based image retrieval for urban environments , 2011, 2011 18th IEEE International Conference on Image Processing.

[10]  C. Schmid,et al.  On the burstiness of visual elements , 2009, 2009 IEEE Conference on Computer Vision and Pattern Recognition.

[11]  Masatoshi Okutomi,et al.  Visual Place Recognition with Repetitive Structures , 2015, IEEE Transactions on Pattern Analysis and Machine Intelligence.

[12]  Torsten Sattler,et al.  Image Retrieval for Image-Based Localization Revisited , 2012, BMVC.

[13]  Nenghai Yu,et al.  AMIGO: accurate mobile image geotagging , 2012, ICIMCS '12.

[14]  Rashid Ansari,et al.  Accurate Image Based Localization by Applying SFM and Coordinate System Registration , 2016, 2016 IEEE International Symposium on Multimedia (ISM).

[15]  Luc Van Gool,et al.  Speeded-Up Robust Features (SURF) , 2008, Comput. Vis. Image Underst..

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

[17]  Matteo Fischetti,et al.  Weighted k-cardinality trees: Complexity and polyhedral structure , 1994, Networks.