Abstract

A content-based image retrieval system based on multinomial relevance feedback is proposed. The system relies on an interactive search paradigm where at each round a user is presented with k images and selects the one closest to their ideal target. Two approaches, one based on the Dirichlet distribution and one based the Beta distribution, are used to model the problem motivating an algorithm that trades exploration and exploitation in presenting the images in each round. Experimental results show that the new approach compares favourably with previous work.

Link to Paper »

Authors

Dorota Glowacka, Yee Whye Teh, John Shawe-Taylor

Year

2016

Journal

arXiv preprint arXiv:1603.09522.

Keywords

Complex issues, Complex networks, and Complex Systems