Combining Features by Query-time Weights Determination for Image Retrieval

Abstract

In this paper a Content Based Image Retrieval approach is presented that combines weighted features. The determination of feature’s weight involves examination by K additional queries formed by the first K retrieved images. For each additional query, the user query is added to the database and the standard deviation of its rank is utilized to calculate the feature’s weight. Finally it is shown the performance of the
system.

Authors

  • Nikolay Neshov

Venue

International Scientific Conference on Information, Communication and Energy Systems and Technologies (ICEST), 2014.

Links

http://rcvt.tu-sofia.bg/ICEST2014_1_42.pdf

Categories

, , ,