avatar

Alessio is a researcher with 10 years experience and expertise in computer vision and deep learning. Alessio contributed computer vision and deep learning methods and audio-visual datasets for human-robot collaborations, decentralised vision methods for collaborative 3D reconstruction and navigation, and benchmarks of graph neural networks for visual privacy and explainability. As a Postdoctoral Research Assistant with the School of Electronic Engineering and Computer Science at Queen Mary University of London (QMUL), UK, he is currently investigating vision-language models for zero-shot open-vocabulary object detection, learning-based models for visual privacy (image privacy classification), and perception-based pipelines for human-to-robot handovers.

Alessio obtained the Ph.D. in Electronic Engineering with a research focus on Computer Vision from Queen Mary University of London, UK, in September 2020. Alessio obtained the Master Degree in Telecommunications Engineering and the Bachelor Degree in Electronics and Telecommunication Engineering from the University of Trento in March 2015 and February 2012, respectively.

Research interests

Computer Vision | Machine Learning | Deep learning | Image and video processing | Perception for robotics | Visual privacy


News

2025/05/09 Pre-print of our article "Visual Affordances: Enabling robots to understand object functionality", currently under review, available on ArXiv
2025/05/09 Workshop "Benchmarking via Competitions in Robotic Grasping and Manipulation (2nd Edition)" accepted at IROS 2025 (see webpage)
2025/05/07 Artifact for PoPETs paper available with code, demo, curated datasets, model card, and trained models (see github repo)
2025/04/01 Article, titled "Stereo Hand-Object Reconstruction for Human-to-Robot Handover", accepted at IEEE Robotics and Automation Letters (RA-L) [arxiv]
2025/03/01 Paper accepted at Privacy Enhancing Technologies Symposium (PETS) 2025 and in the journal Proceedings on Privacy Enhancing Technologies (PoPETs), Volume 2025, Issue 3, 2025

All news

Recent blog posts


Autonomous Robotic Grasping and
Manipulation in Real-World Applications


Guest editor
IEEE RA-P Special Collection - Call for Papers

Awards
Outstanding Reviewer at ECCV 2024

GraphNEx
CHIST-ERA Projects Seminar 2024

Awards

Award ECCV2024 Award CVPR2024 Award ICASSP23 Award ICASSP22 Award ICIP2020
Outstanding Reviewer
ECCV 2024
(top 3% of 7,293 reviewers) 		Outstanding Reviewer
IEEE/CVF CVPR 2024
(top 2% of 9,872 reviewers) 		Outstanding Reviewer
IEEE ICASSP 2023
 Outstanding Reviewer
IEEE ICASSP 2022
(31 reviewers selected) 		Top Reviewer
IEEE ICIP 2020
(top 3% - 22 reviewers selected)


Selected publications

TIP2020
A spatio-temporal multi-scale binary descriptor
IEEE Transactions on Image Processing 2020

Cross-Camera View-Overlap Recognition
International Workshop on Smart Distributed Cameras at European Conference on Computer Vision, 2022 		ICIP2018
MORB: a multi-scale binary descriptor
IEEE International Conference on Image Processing 2018


The CORSMAL benchmark for the prediction of the properties of containers
IEEE Access, vol. 10, 2022


Audio-Visual Object Classification for Human-Robot Collaboration
IEEE International Conference on Acoustic, Speech and Signal Processing 2022 		ICASSP2020
Multi-view shape estimation of transparent containers
IEEE International Conference on Acoustic, Speech and Signal Processing 2020