PVSS: portable visual search service for researchers

Image retrieval researchers aim to build good algorithms and devise methods to produce accurate ranked lists. However, the search results themselves appear rather repetitious, and the experience of viewing them is akin to listening to the same song 100 times a day. Imagine a better search in which the researcher has a ready-to-use demonstration interface for showing his results to audiences who can interactively send image queries and see the results effortlessly; this is our aim. We present a Portable Visual Search Service for a researcher who desires to make his or her algorithm appear to be as "cool" as the best interactive demos at image retrieval conferences. Our service is a handy package, powered by VMWare virtualization technology, that is booted and served locally within a local network on a medium-spec laptop PC. The package contains all the necessary stuff to launch an on-site demo, e.g., a database, algorithm, and libraries. Moreover, our service is accessible using standard web browsers and client architectures, e.g. PC, iPhone, iPad, and Android. This interactive system has been tested together with a large database at conferences and other events. The audience of a demo/poster session was impressed by the features of our service. In this context, our service handles the user interaction with the audience and gives researchers more freedom to explain their algorithm.

[1]  Shin'ichi Satoh,et al.  Connect commercial films with realities , 2013, ICMR '13.

[2]  Michael S. Brown,et al.  Offline Mobile Instance Retrieval with a Small Memory Footprint , 2013, 2013 IEEE International Conference on Computer Vision.

[3]  Bernd Girod,et al.  Residual enhanced visual vector as a compact signature for mobile visual search , 2013, Signal Process..

[4]  Bernd Girod,et al.  Comparison of local feature descriptors for mobile visual search , 2010, 2010 IEEE International Conference on Image Processing.

[5]  Bernd Girod,et al.  Mobile Visual Search , 2011, IEEE Signal Processing Magazine.

[6]  Yan-Ying Chen,et al.  Enabling low bitrate mobile visual recognition: a performance versus bandwidth evaluation , 2013, MM '13.

[7]  Shin'ichi Satoh,et al.  Tell Me about TV Commercials of This Product , 2014, MMM.