Speaker
Caterina Fenu
(Università di Cagliari)
Description
In this talk, we propose an approach to perform multi-label segmentation. Starting from an image, we construct the associated weighted graph and assign a label to a small number of pixels (seeds). The aim is to assign each unlabeled pixel to a label to identify different regions. Our algorithm uses the notion of communicability from complex networks theory to compute the ease for an unlabeled pixel to reach a labeled one. We can perform a good image segmentation by assigning each pixel to the label for which the largest communicability is calculated. Moreover, an extension to coloured images is proposed.
Primary author
Caterina Fenu
(Università di Cagliari)
