Hypergraphes : clustering, réduction et marches aléatoires orientées pour la segmentation d'images et de vidéo

Graphs have been successfully used in Pattern Recognition and Computer Vision domains for a number of years, due to their ability to represent relational patterns. However, their main limitation comes from the fact that they are only able to model pairwise affinities between elements of interest. In this work, we focus on exploiting the advantages of hypergraphs, which generalize the graph concept…