iSearch: mining retrieval history for content-based image retrieval

Relevance feedback is a powerful technique to bridge the gap between high-level concepts and low-level features, and has been successfully applied to the field of Content-Based Image Retrieval (CBIR) to improve the query accuracy in recent years. In this paper, we propose a novel model (iSearch) which predicts user´s information need based on past retrieval history. Based on the prediction, we then transform the feature space based on the user´s feedback and employ an Expectation Maximization (EM) approach to simulate the new space by a mixture of Gaussian distributions. The experimental results show that the proposed method is effective and captures the user´s information need more precisely.