Semi Supervised Image Search Re-Ranking

Abstract

Image search methods usually fail to capture the user‟s intention when the query term is ambiguous. It gives unsatisfactory result. Therefore, reranking with user interactions is highly demanded to effectively improve the search performance. The essential problem is how to identify the user‟s intention effectively. To complete this goal, this paper presents a structural information based active sample selection strategy to reduce the user‟s labeling efforts. Furthermore, to localize the user‟s intention in the visual feature space, a novel local-global discriminative dimension reduction algorithm is proposed. In this algorithm, a submanifold is learned by transferring the local geometry and the discriminative information from the labeled images to the whole (global) image database.