Robust place recognition by avoiding confusing features and fast geometric re-ranking

There are millions of mobile phone applications based on location. Using a photo to precisely locate users location is useful and necessary. However, real-time location recognition or retrieval system is a challenging problem due to the really big differences between the query and the dataset in scale, viewpoint and lighting, or the noise existed in the foreground or background etc. To address this problem, we design a place recognition system and a new famous buildings dataset with ground truth labels. By adding a fast geometric image matching procedure before using RANSAC and applying a relative camera orientation calculation algorithm to filter the dataset collected from the Internet, we can substantially improve the efficiency of spatial verification and recognition accuracy.

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

[2]  David G. Lowe,et al.  Distinctive Image Features from Scale-Invariant Keypoints , 2004, International Journal of Computer Vision.

[3]  Andrew J. Davison,et al.  Active Matching , 2008, ECCV.

[4]  Jiri Matas,et al.  Total recall II: Query expansion revisited , 2011, CVPR 2011.

[5]  Andrew Zisserman,et al.  Three things everyone should know to improve object retrieval , 2012, 2012 IEEE Conference on Computer Vision and Pattern Recognition.

[6]  Michael Isard,et al.  Total Recall: Automatic Query Expansion with a Generative Feature Model for Object Retrieval , 2007, 2007 IEEE 11th International Conference on Computer Vision.

[7]  Michael Isard,et al.  Object retrieval with large vocabularies and fast spatial matching , 2007, 2007 IEEE Conference on Computer Vision and Pattern Recognition.

[8]  Michael Isard,et al.  Lost in quantization: Improving particular object retrieval in large scale image databases , 2008, 2008 IEEE Conference on Computer Vision and Pattern Recognition.

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

[10]  Bernd Girod,et al.  Fast geometric re-ranking for image-based retrieval , 2010, 2010 IEEE International Conference on Image Processing.

[11]  Richard I. Hartley,et al.  In Defense of the Eight-Point Algorithm , 1997, IEEE Trans. Pattern Anal. Mach. Intell..

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

[13]  Luc Van Gool,et al.  World-scale mining of objects and events from community photo collections , 2008, CIVR '08.

[14]  Cordelia Schmid,et al.  Hamming Embedding and Weak Geometric Consistency for Large Scale Image Search , 2008, ECCV.

[15]  Robert Pless,et al.  Toward Fully Automatic Geo-Location and Geo-Orientation of Static Outdoor Cameras , 2008, 2008 IEEE Workshop on Applications of Computer Vision.

[16]  Cordelia Schmid,et al.  Scale & Affine Invariant Interest Point Detectors , 2004, International Journal of Computer Vision.

[17]  Thomas Deselaers,et al.  ClassCut for Unsupervised Class Segmentation , 2010, ECCV.

[18]  Lifeng Sun,et al.  Efficient relative camera orientation detection for mobile applications , 2011, MLBS '11.

[19]  JUSTIN ZOBEL,et al.  Inverted files for text search engines , 2006, CSUR.

[20]  Tomás Pajdla,et al.  Avoiding Confusing Features in Place Recognition , 2010, ECCV.

[21]  Cordelia Schmid,et al.  Improving Bag-of-Features for Large Scale Image Search , 2010, International Journal of Computer Vision.