Bilkent University
Department of Computer Engineering


Mobile Visual Search


Fatih Çalışır
MSc Student
Computer Engineering Department
Bilkent University

The rapid development of technologies in both software and hardware have turned the mobile phones into powerful image and video processing devices equipped with high-resolution cameras and internet connection. This property of the mobile phones directs the users into mobile visual search rather than text-based visual search. The properties which are accessibility, ease of use and expression power are the major properties that make mobile visual search popular. Such applications can be used for identifying and comparing products or getting detailed information about them. There are some mobile visual search applications such as goggles, point and find, snaptell which are the first deployments of this kind of systems. Mobile image-retrieval area has a unique set of challenges like in terms of computation power or bandwidth and client-server architecture used for mobile visual search should address them. Classical mobile image retrieval methods use bag of visual words approach and single query image.

However, multiple query images can be used to increase the performance of the system because with multiple images more information is gathered about the object or the scene. Moreover, the performance of the system can be increased with using the interaction property of the mobile phone.


DATE: 24 March, 2014, Monday @ 15:40