Publications (24)
Generating Graphs via Spectral Diffusion
Giorgia Minello, Alessandro Bicciato, Luca Rossi +2
In this paper, we present GGSD, a novel graph generative model based on 1) the spectral decomposition of the graph Laplacian matrix and 2) a diffusion process. Specifically, we pro…
Isospectralization, or how to hear shape, style, and correspondence
Luca Cosmo, Mikhail Panine, Arianna Rampini +3
The question whether one can recover the shape of a geometric object from its Laplacian spectrum ('hear the shape of the drum') is a classical problem in spectral geometry with a b…
Generalized Multi-Source Inference for Text Conditioned Music Diffusion Models
Emilian Postolache, Giorgio Mariani, Luca Cosmo +2
Multi-Source Diffusion Models (MSDM) allow for compositional musical generation tasks: generating a set of coherent sources, creating accompaniments, and performing source separati…
Partial Functional Correspondence
Emanuele RodolÃ, Luca Cosmo, Michael M. Bronstein +2
In this paper, we propose a method for computing partial functional correspondence between non-rigid shapes. We use perturbation analysis to show how removal of shape parts changes…
Latent Autoregressive Source Separation
Emilian Postolache, Giorgio Mariani, Michele Mancusi +3
Autoregressive models have achieved impressive results over a wide range of domains in terms of generation quality and downstream task performance. In the continuous domain, a key…
Unsupervised Source Separation via Bayesian Inference in the Latent Domain
Michele Mancusi, Emilian Postolache, Giorgio Mariani +4
State of the art audio source separation models rely on supervised data-driven approaches, which can be expensive in terms of labeling resources. On the other hand, approaches for…
Latent-Graph Learning for Disease Prediction
Luca Cosmo, Anees Kazi, Seyed-Ahmad Ahmadi +2
Recently, Graph Convolutional Networks (GCNs) have proven to be a powerful machine learning tool for Computer-Aided Diagnosis (CADx) and disease prediction. A key component in thes…
SelfGeo: Self-supervised and Geodesic-consistent Estimation of Keypoints on Deformable Shapes
Mohammad Zohaib, Luca Cosmo, Alessio Del Bue
Unsupervised 3D keypoints estimation from Point Cloud Data (PCD) is a complex task, even more challenging when an object shape is deforming. As keypoints should be semantically and…
Spectral Maps for Learning on Subgraphs
Marco Pegoraro, Riccardo Marin, Arianna Rampini +3
In graph learning, maps between graphs and their subgraphs frequently arise. For instance, when coarsening or rewiring operations are present along the pipeline, one needs to keep…
Naturalistic Music Decoding from EEG Data via Latent Diffusion Models
Emilian Postolache, Natalia Polouliakh, Hiroaki Kitano +4
In this article, we explore the potential of using latent diffusion models, a family of powerful generative models, for the task of reconstructing naturalistic music from electroen…
FolAI: Synchronized Foley Sound Generation with Semantic and Temporal Alignment
Riccardo Fosco Gramaccioni, Christian Marinoni, Emilian Postolache +4
Traditional sound design workflows rely on manual alignment of audio events to visual cues, as in Foley sound design, where everyday actions like footsteps or object interactions a…
Graph-in-Graph (GiG): Learning interpretable latent graphs in non-Euclidean domain for biological and healthcare applications
Kamilia Mullakaeva, Luca Cosmo, Anees Kazi +3
Graphs are a powerful tool for representing and analyzing unstructured, non-Euclidean data ubiquitous in the healthcare domain. Two prominent examples are molecule property predict…
Learning disentangled representations via product manifold projection
Marco Fumero, Luca Cosmo, Simone Melzi +1
We propose a novel approach to disentangle the generative factors of variation underlying a given set of observations. Our method builds upon the idea that the (unknown) low-dimens…
GNN-LoFI: a Novel Graph Neural Network through Localized Feature-based Histogram Intersection
Alessandro Bicciato, Luca Cosmo, Giorgia Minello +2
Graph neural networks are increasingly becoming the framework of choice for graph-based machine learning. In this paper, we propose a new graph neural network architecture that sub…
COCOLA: Coherence-Oriented Contrastive Learning of Musical Audio Representations
Ruben Ciranni, Giorgio Mariani, Michele Mancusi +4
We present COCOLA (Coherence-Oriented Contrastive Learning for Audio), a contrastive learning method for musical audio representations that captures the harmonic and rhythmic coher…
STAGE: Stemmed Accompaniment Generation through Prefix-Based Conditioning
Giorgio Strano, Chiara Ballanti, Donato Crisostomi +3
Recent advances in generative models have made it possible to create high-quality, coherent music, with some systems delivering production-level output. Yet, most existing models f…
Differentiable Graph Module (DGM) for Graph Convolutional Networks
Anees Kazi, Luca Cosmo, Seyed-Ahmad Ahmadi +2
Graph deep learning has recently emerged as a powerful ML concept allowing to generalize successful deep neural architectures to non-Euclidean structured data. Such methods have sh…
Bending Graphs: Hierarchical Shape Matching using Gated Optimal Transport
Mahdi Saleh, Shun-Cheng Wu, Luca Cosmo +3
Shape matching has been a long-studied problem for the computer graphics and vision community. The objective is to predict a dense correspondence between meshes that have a certain…
Shape registration in the time of transformers
Giovanni Trappolini, Luca Cosmo, Luca Moschella +3
In this paper, we propose a transformer-based procedure for the efficient registration of non-rigid 3D point clouds. The proposed approach is data-driven and adopts for the first t…
Graph Kernel Neural Networks
Luca Cosmo, Giorgia Minello, Alessandro Bicciato +4
The convolution operator at the core of many modern neural architectures can effectively be seen as performing a dot product between an input matrix and a filter. While this is rea…
Universal Spectral Adversarial Attacks for Deformable Shapes
Arianna Rampini, Franco Pestarini, Luca Cosmo +2
Machine learning models are known to be vulnerable to adversarial attacks, namely perturbations of the data that lead to wrong predictions despite being imperceptible. However, the…
Multi-Source Diffusion Models for Simultaneous Music Generation and Separation
Giorgio Mariani, Irene Tallini, Emilian Postolache +3
In this work, we define a diffusion-based generative model capable of both music synthesis and source separation by learning the score of the joint probability density of sources s…
LIMP: Learning Latent Shape Representations with Metric Preservation Priors
Luca Cosmo, Antonio Norelli, Oshri Halimi +2
In this paper, we advocate the adoption of metric preservation as a powerful prior for learning latent representations of deformable 3D shapes. Key to our construction is the intro…
Learning Spectral Unions of Partial Deformable 3D Shapes
Luca Moschella, Simone Melzi, Luca Cosmo +5
Spectral geometric methods have brought revolutionary changes to the field of geometry processing. Of particular interest is the study of the Laplacian spectrum as a compact, isome…