CV 💁🏻‍♂️

Research Scholar

Polytechnic University of Turin

Research Scholarship: 'Understanding Visual Place Recognition - Explainable and Trustworthy Models / Comprendere i modelli di Visual Place Recognition e studiare la loro affidabilità' under the supervision of Professors Tatiana Tommasi and Carlo Masone @ Vandal Lab

• Research
• Computer Vision
• Artificial Intelligence
• Machine Learning
• Deep Learning

Teaching assistant for laboratories

Polytechnic University of Turin

Part-time collaboration for teaching assistance in the laboratories for Professor Tatiana Tommasi's courses: 'Advanced Machine Learning' and 'Data analysis and Artificial Intelligence'

• Artificial Intelligence
• Machine Learning
• Deep Learning
• Data Analysis

MSc

Polytechnic University of Turin

Computer Engineering - Artificial Intelligence & Data Analytics

• Big Data Processing and Analytics
• Data Science and Database Technologies
• Web Applications I
• Machine Learning and Pattern Recognition
• Architectures of Processing Systems
• System Programming
• Network Technologies and Services
• Explainable and Trustworthy AI
• Bioinformatics
• Image Processing and Computer Vision
• Information Systems Security
• Computational Intelligence
• Advanced Machine Learning
• Formal Languages and Compilers
• Advanced Numerical Methods

BSc

University of Palermo

Computer Engineering

• Algebra
• Mathematical Analysis C.I.
• Electronic Computers C.I.
• Physics I
• Geometry
• Algorithms and Data Structures
• Databases and Information Systems
• Computer-aided Design
• Physics II
• Mathematical and Numerical Methods
• Programming
• Signal Theory
• Automatic Controls
• Fundamentals of Electronics
• Software Engineering
• Web and Mobile Programming
• Computer Networks and Internet
• Operating Systems

High School

Istituto Tecnico Economico Damiani Almeyda - Crispi

Istituto Tecnico Economico

Albo Nazionale delle Eccellenze

MIUR - Ministero dell'Istruzione e del Merito

National Roll of Graduates with 100 cum laude

To Match or Not to Match: Revisiting Image Matching for Reliable Visual Place Recognition

Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR) Workshops

Visual Place Recognition (VPR) is a critical task in computer vision, traditionally enhanced by re-ranking retrieval results with image matching. However, recent advancements in VPR methods have significantly improved performance, challenging the necessity of re-ranking. In this work, we show that modern retrieval systems often reach a point where re-ranking can degrade results, as current VPR datasets are largely saturated. We propose using image matching as a verification step to assess retrieval confidence, demonstrating that inlier counts can reliably predict when re-ranking is beneficial. Our findings shift the paradigm of retrieval pipelines, offering insights for more robust and adaptive VPR systems.