In this paper, we describe our approach and its results for the MediaEval 2015 Retrieving Diverse Social Images task. The main strength of the proposed approach is its flexibility that permits to filter out irrelevant images, and to obtain a reli- able set of diverse and relevant images. This is done by first clustering similar images according to their textual descrip- tions and their visual content, and then extracting images from different clusters according to a measure of user’s cred- ibility. Experimental results shown that it is stable and has little fluctuation in both single-concept and multi-concept queries.
|Titolo:||Multimodal-based Diversified Summarization in Social Image Retrieval|
|Data di pubblicazione:||2015|
|Tipologia:||4.1 Contributo in Atti di convegno|