In this paper, we describe our approach for the MediaEval 2015 Retrieving Diverse Social Images Task. The proposed approach exploits available user-generated textual descriptions and the visual content of the images in a combination with common, unsupervised clustering techniques in order to increase the diversification of retrieval results. Preliminary experiments indicate that the approach generalizes well for different datasets and achieves comparable results for single- and multi-topic queries.
M. Zaharieva, L. Diem: "MIS @ Retrieving Diverse Social Images Task 2015"; Talk: MediaEval Benchmarking Initiative for Multimedia Evaluation, Wurzen, Deutschland; 09-14-2015 - 09-15-2015; in: "CEUR Workshop Proceedings of the MediaEval 2015 Multimedia Benchmark Workshop", CEUR-WS, Vol-1436 (2015), ISSN: 1613-0073; Paper ID 33, 3 pages.
Click into the text area and press Ctrl+A/Ctrl+C or ⌘+A/⌘+C to copy the BibTeX into your clipboard… or download the BibTeX.