International audienceIn our data driven world, clustering is of major importance to help end-users and decision makers understanding information structures. Supervised learning techniques rely on ground truth to perform the classification and are usually subject to overtraining issues. On the other hand, unsupervised clustering techniques study the structure of the data without disposing of any training data. Given the difficulty of the task, unsupervised learning tends to provide inferior results to supervised learning. To boost their performance, a compromise is to use learning only for some of the ambiguous classes. In this context, this paper studies the impact of pairwise constraints to unsupervised Spectral Clustering. We introduce a new generalization of constraint propagation which maximizes partitioning quality while reducing annotation costs. Experiments show the efficiency of the proposed scheme

In our data driven world, categorization is of major importance to help end-users and decision makers understanding information structures. Supervised learning techniques rely on annotated samples that are often difficult to obtain and training often overfits. On the other hand, unsupervised clustering techniques study the structure of the data without disposing of any training data. Given the difficulty of the task, supervised learning often outperforms unsupervised learning. A compromise is to use a partial knowledge, selected in a smart way, in order to boost performance while minimizing learning costs, what is called semi-supervised learning. In such use case, Spectral Clustering proved to be an efficient method. Also, Deep Learning outperformed several state of the art classification approaches and it is interesting to test it in our context. In this paper, we firstly introduce the concept of Deep Learning into an active semi-supervised clustering process and compare it with Spectral Clustering. Secondly, we introduce constraint propagation and demonstrate how it maximizes partitioning quality while reducing annotation costs. Experimental validation is conducted on two different real datasets. Results show the potential of the clustering methods.

When considering multimedia database growth, one current challenging issue is to design accurate navigation tools. End user basic needs, such as exploration, similarity search and favorite suggestions, lead to investigate how to find semantically resembling media. One way is to build numerous continuous dissimilarity measures from low-level image features. In parallel, an other way is to build discrete dissimilarities from textual information which may be available with video sequences. However, how such different measures should be selected as relevant and be fused ? To this aim, the purpose of this paper is to compare all those various dissimilarities and to propose a suitable ranking fusion method for several dissimilarities. Subjective tests with human observers on the CITIA animation movie database have been carried out to validate the model.