Francesca Pistilli

Assistant Professor (non tenure-track) · DAUIN, Politecnico di Torino

I am a researcher working at the intersection of computer vision and robot learning. My work focuses on structured, graph-based representations of scenes and human activity — from point-cloud processing and egocentric video understanding to scene graphs that capture how objects, people and their relations evolve over time. I am interested in how such explicit, editable scene representations can help robots understand their surroundings, anticipate how a scene will change as actions unfold, and act and cooperate with people in dynamic environments. I also work on efficient models that make these capabilities practical on real systems.

I am an Assistant Professor (non tenure-track) at the Department of Control and Computer Engineering (DAUIN) of Politecnico di Torino. Before that, I was a post-doctoral researcher in the VANDAL Lab, and previously I received my Ph.D. in 2023 from the Image Processing and Learning (IPL) group at Politecnico di Torino, under the supervision of Enrico Magli. I completed an M.Sc. in Electronic Engineering at Politecnico di Torino and an M.Sc. in Electrical and Computer Engineering at the University of Illinois at Chicago, in 2019 and 2020 respectively.

Research interests: graph neural networks, egocentric vision, human behaviour understanding, scene graphs, 3D point clouds, learning for robotics.