Università della Svizzera italiana

Geometric deep learning for shape analysis : extending deep learning techniques to non-Euclidean manifolds

Boscaini, Davide ; Bronstein, Michael (Dir.) ; Masci, Jonathan (Codir.)

Thèse de doctorat : Università della Svizzera italiana, 2017 ; 2017INFO009.

The past decade in computer vision research has witnessed the re-emergence of artificial neural networks (ANN), and in particular convolutional neural network (CNN) techniques, allowing to learn powerful feature representations from large collections of data. Nowadays these techniques are better known under the umbrella term deep learning and have achieved a breakthrough in performance in a...

Università della Svizzera italiana

Spectral methods for multimodal data analysis

Kovnatsky, Artiom ; Bronstein, Michael (Dir.)

Thèse de doctorat : Università della Svizzera italiana, 2016 ; 2016INFO006.

Spectral methods have proven themselves as an important and versatile tool in a wide range of problems in the fields of computer graphics, machine learning, pattern recognition, and computer vision, where many important problems boil down to constructing a Laplacian operator and finding a few of its eigenvalues and eigenfunctions. Classical examples include the computation of diffusion ...