Stefano Fiorini: Unlocking Reassembly Tasks - Reconstructing 2D and 3D Worlds with Deep Learning
28 November 2023 - 28 November 2023
East Campus USI-SUPSI, Room B1.09
Reassembly tasks are fundamental human skills acquired during early developmental stages; therefore, we believe it is necessary to develop these tasks to approach general artificial intelligence (AI). The resolution of reassembly tasks in both 2D and 3D holds significant relevance across various fields such as biology, computer vision, and cultural heritage. However, existing approaches tend to focus solely on specific tasks and modalities, lacking a unified framework. In this presentation, we will introduce a novel unified framework based on a Graph Neural Network architecture. We will also explore the benefits and issues related to the adoption of a diffusion process, where we introduce noise into the elements' positions and orientations, followed by iterative denoising to reconstruct their coherent poses. Through our study, we will reveal the shared fundamentals between 2D and 3D tasks, emphasizing the significance of rotation-equivariant representation as a common inductive bias that enhances performance in both modalities.

The Speaker

Stefano Fiorini is currently a postdoctoral researcher at the Pattern Analysis and Computer Vision (PAVIS) department of the Italian Institute of Technology (IIT) in Genova, supervised by Dr Alessio Del Bue. He is working on the interaction between graph neural networks and diffusion models for the resolution of reassembly tasks. He also has a strong interest in graph neural network theory and its applications in other fields. Stefano completed his Ph.D. in Computer Science at the University of Milano-Bicocca in 2023, supervised by Professor Michele Ciavotta.