Unsupervised similarity learning through cartesian product of ranking references for image retrieval tasks

Abstract

Despite the consistent advances in visual features and other Content-Based Image Retrieval techniques, measuring the similarity among images is still a challenging task for effective image retrieval. In this scenario, similarity learning approaches capable of improving the effectiveness of retrieval in an unsupervised way are indispensable. A novel method, called Cartesian Product of Ranking References (CPRR), is proposed with this objective in this paper. The proposed method uses Cartesian product operations based on rank information for exploiting the underlying structure of datasets. Only subsets of ranked lists are required, demanding low computational efforts. An extensive experimental evaluation was conducted considering various aspects, four public datasets and several image features. Besides effectiveness, experiments were also conducted to assess the efficiency of the method, considering parallel and heterogeneous computing on CPU and GPU devices. The proposed method achieved significant effectiveness gains, including competitive state-of-the-art results on popular benchmarks.