An interactive framework for content-based image retrieval

In recent years, numerous content-based image retrieval (CBIR) systems have been proposed for locating relevant information within visual databases. These systems rely on visual features such as colour and texture to represent and compare images. The possible mismatch between this low-level machine representation and the high-level user concepts as well as the subjectivity of human judgement lead to several technical challenges in CBIR.; This work proposes a novel framework for content-based image retrieval systems. The similarity calculation problem is formulated as a decision making process based on the information provided by the various visual features. This framework provides an effective and flexible tool for user dependent similarity calculation for modelling the human subjectivity.; The subjectivity inherent in interpretation of visual content necessitates the development of online interactive learning. This work presents a novel method for interactive image retrieval, Query Feedback. Query Feedback learns the user query as well as the correspondence between high level user concepts and their low-level machine representation by performing retrievals according to multiple queries.