In this paper, based on Bayesian relevance feedback methods, we propose a novel interactive face retrieving model based on two objective functions, one is the Maximum a Posterior (MAP) and the other is maximization of mutual information. The proposed bi-objective optimization model aims at minimizing both the number of interactive iterations and the average length of iterations. Moreover, we deduce a top-bottom search algorithm to solve the proposed. Experiments with real testers prove that the proposed algorithm could largely improve the interactive searching efficiency in face databases. © 2011 Springer-Verlag Berlin Heidelberg.
CITATION STYLE
Fang, Y., Cai, Q., Luo, J., Dai, W., & Lou, C. (2011). A Bi-objective optimization model for interactive face retrieval. In Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics) (Vol. 6524 LNCS, pp. 393–400). https://doi.org/10.1007/978-3-642-17829-0_37
Mendeley helps you to discover research relevant for your work.