Video based mobile location search with large set of SIFT points in cloud

In densely populated cities like Hong Kong, GPS alone is not sufficient in providing rich location based information services. In this work, we provide a mobile location search service by allowing users shooting a short video clip about the surrounding buildings and the Cloud will return a tagged image with a summary of the location and services available. The key technical challenges are the robustness of the SIFT based object matching in video sequences, and the computational complexity associated with the large scale of the repository. We solved this by spatio-temporal pruning of SIFT points in the video repository, and PCA projection and indexing of SIFT points in Cloud. Simulation results demonstrated the robustness of the PCAed and indexed SIFT points in building identification and overall effectiveness of the proposed solution on a small scale trial. Tagging and integration with large scale video repositories like YouTube, Tudou are underway with more interesting applications.

[1]  Shuicheng Yan,et al.  SIFT-Bag kernel for video event analysis , 2008, ACM Multimedia.

[2]  Xiao Zhang,et al.  Efficient indexing for large scale visual search , 2009, 2009 IEEE 12th International Conference on Computer Vision.

[3]  Cordelia Schmid,et al.  A performance evaluation of local descriptors , 2005, IEEE Transactions on Pattern Analysis and Machine Intelligence.

[4]  J. Borkowski,et al.  Enhanced performance of Cell ID+RTT by implementing forced soft handover algorithm , 2004, IEEE 60th Vehicular Technology Conference, 2004. VTC2004-Fall. 2004.

[5]  Mubarak Shah,et al.  A 3-dimensional sift descriptor and its application to action recognition , 2007, ACM Multimedia.

[6]  Ying Wu,et al.  Common Spatial Pattern Discovery by Efficient Candidate Pruning , 2007, 2007 IEEE International Conference on Image Processing.

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

[8]  Yan Ke,et al.  PCA-SIFT: a more distinctive representation for local image descriptors , 2004, Proceedings of the 2004 IEEE Computer Society Conference on Computer Vision and Pattern Recognition, 2004. CVPR 2004..

[9]  G LoweDavid,et al.  Distinctive Image Features from Scale-Invariant Keypoints , 2004 .

[10]  Qi Tian,et al.  Feature selection using principal feature analysis , 2007, ACM Multimedia.

[11]  George M. Giaglis,et al.  A taxonomy of indoor and outdoor positioning techniques for mobile location services , 2002, SECO.