Effective User Relevance Feedback for Image Retrieval with Image Signatures

Abstract

Content-based image retrieval (CBIR) has attracted much attention due to the exponential growth of digital image collections that have become available in recent years. Relevance feedback (RF) in the context of search engines is a query expansion technique, which is based on relevance judgments about the top results that are initially returned for a given query. RF can be obtained directly from end users, inferred indirectly from user interactions with a result list, or even assumed (aka pseudo relevance feedback). RF information is used to generate a new query, aiming to re-focus the query towards more relevant results. This paper presents a methodology for use of signature based image retrieval with a user in the loop to improve retrieval performance. The significance of this study is twofold. First, it shows how to effectively use explicit RF with signature based image retrieval to improve retrieval quality and efficiency. Second, this approach provides a mechanism for end users to refine their image queries. This is an important contribution because, to date, there is no effective way to reformulate an image query; our approach provides a solution to this problem. Empirical experiments have been carried out to study the behaviour and optimal parameter settings of this approach. Empirical evaluations based on standard benchmarks demonstrate the effectiveness of the proposed approach in improving the performance of CBIR in terms of recall, precision, speed and scalability.