Image re-ranking and rank aggregation based on similarity of ranked lists

In Content-based Image Retrieval (CBIR) systems, ranking accurately collection images is of great relevance. Users are interested in the returned images placed at the first positions, which usually are the most relevant ones. Collection images are ranked in increasing order of their distance to the query pattern (e.g., query image) defined by users. Therefore, the effectiveness of these systems is very dependent on the accuracy of the distance function adopted. In this paper, we present a novel context-based approach for redefining distances and later re-ranking images aiming to improve the effectiveness of CBIR systems. In our approach, distances among images are redefined based on the similarity of their ranked lists. Conducted experiments involving shape, color, and texture descriptors demonstrate the effectiveness of our method.

► Presentation of the contextual distance measure which is based on the similarity of ranked lists. ► Presentation of the RL-Sim re-ranking for improving effectiveness of CBIR systems. ► Description of different approaches for comparing ranked lists. ► Description of how RL-Sim algorithm can be used for the rank aggregation tasks. ► Experimental evaluation considering different datasets, descriptors, and baselines.