Heterogeneous sources of information, such as images, videos, text and metadata are often used to describe di erent or complementary views of the same multimedia object, especially in the online news domain and in large annotated image collections. The retrieval of multimedia objects, given a mul- timodal query, requires the combination of several sources of information in an e cient and scalable way. Towards this direction, we provide a novel unsuper- vised framework for multimodal fusion of visual and textual similarities, which are based on visual features, visual concepts and textual metadata, integrating non-linear graph-based fusion and Partial Least Squares Regression. The fu- sion strategy is based on the construction of a multimodal contextual similarity matrix and the non-linear combination of relevance scores from query-based similarity vectors. Our framework can employ more than two modalities and high-level information, without increase in memory complexity, when com- pared to state-of-the-art baseline methods. The experimental comparison is done in three public multimedia collections in the multimedia retrieval task. The results have shown that the proposed method outperforms the baseline methods, in terms of Mean Average Precision and Precision@20.

This work was partially supported by the European Commission by the projects MULTISENSOR (FP7-610411) and KRISTINA (H2020-645012).