CV/Resume

Skills

• Front end development: HTML, CSS, Javascript;
• Back end development: PHP, SQL, Node.js;
• Other languages: C, C++, Python;
• Other: Machine Learning Algorithms, Neural Networks, Deep Learning, Reinforcement Learning, Computer Vision, PyTorch, TensorFlow, Keras, OpenCV.

Education

• Ph.D student in Computer Vision, La Sapienza University of Rome - Oct. 2024 - Current
• Master’s Degree in Artificial Intelligence and Robotics, La Sapienza University of Rome - Oct. 2020 Oct. 2023
• Bachelor’s Degree in Computer Engineering, Tor Vergata University of Rome - Oct. 2016 Feb. 2020

Work experience

• Junior research fellowship, La Sapienza University of Rome - Jan. 2024 - Oct. 2024
  • Research about Ground-to-Aerial image matching task using Neural Networks, Python, TensorFlow, and OpenCV;
  • Implementation of a road object detection and tracking system for a traffic analysis pipeline, using PyTorch, TensorFlow and OpenCV.
• Software and Web Developer, Tor Vergata students racing team, formula SAE - Oct. 2019 - Oct. 2020
  • Developed a unique online accessible calendar module, shared among all the team members, based on their permissions, using HTML, CSS, Vue.js, Node.js;
  • Projected a telemetry system to transfer data from the car to the base station.
• Web Developer, Oct. 2020
  • Developed in team a food delivery website using HTML, CSS, Vue.js, Node.js.

Projects

• Ground-to-Aerial image matching for geospatial applications
  • Thesis project in Computer Vision, the work has been done in collaboration with the ESA-φ-lab of Frascati, Rome;
  • Developed a Neural Networks-based architecture to match a query ground view image with a satellite one representing the same location, using Python and TensorFlow.
• Distance estimation of traffic signs in driving environment
  • Developed an offline Neural Networks-based architecture to estimate the distances to the traffic signs from a camera mounted on a car, using Python and TensorFlow.
• Covid-19 detection from chest x-ray
  • Developed a Neural Networks-based architecture to detect patients affected by COVID-19 through chest x-ray images, first training and testing three well-known networks and then developing two networks from scratch. The project has been implemented using Python and TensorFlow;
  • Collected online accessible chest x-ray images to create an ad hoc dataset.
• Computer vision - paper “fighting fake news: image splice detection via learned self-consistency”
  • Implemented and tested the mentioned paper Neural Network-based architecture with the aim of learning real images metadata, to discriminate them from fake ones.
• Assembly function recognizer
  • Implemented a method to recognize assembly functions among 4 classes, using Python to exploit two Machine Learning Algorithms, the Support Vector Machine and the Generative Model.
• Rasa NLP - Mall receptionist
  • Implemented a mall receptionist to assist customers in finding shops, restaurants, and related information, by filters. The project has been implemented with the SpeechRecognition Python library combined with the Rasa Framework for the Language Understanding part.
• Reinforcement learning - PPO Algorithm implementation
  • Developed the Reinforcement Learning PPO Algorithm in a simulated environment using Neural Networks. It is implemented with Python and PyTorch.
• Reliable UDP Protocol
  • Implemented a reliable version of the UDP Protocol to reduce the loss of information in a communication between two endpoints. This project was implemented using C.
• Restaurant database system
  • Developed a Database System for a pizza restaurant, to track clients’ orders, waiters’ work shifts, menu, and the pantry with the quantity of each product. The project has been implemented with C and SQL.

Publications

Learning from Unlabelled data with Transformers: Domain Adaptation for Semantic Segmentation of High Resolution Aerial Images

N. Dionelis, F. Pro, L. Maiano, I. Amerini and B. L. Saux, "Learning from Unlabelled data with Transformers: Domain Adaptation for Semantic Segmentation of High Resolution Aerial Images," IGARSS 2024 - 2024 IEEE International Geoscience and Remote Sensing Symposium, Athens, Greece, 2024, pp. 8167-8172, doi: 10.1109/IGARSS53475.2024.10641541.

A semantic segmentation-guided approach for ground-to-aerial image matching

F. Pro, N. Dionelis, L. Maiano, B. L. Saux and I. Amerini, "A Semantic Segmentation-Guided Approach for Ground-to-Aerial Image Matching," IGARSS 2024 - 2024 IEEE International Geoscience and Remote Sensing Symposium, Athens, Greece, 2024, pp. 2630-2635

Distance Estimation of Fixed Objects in Driving Environments

Giorgio Leporoni, Valerio Ponzi, Francesco Pro, & Christian Napoli (2023). Distance Estimation of Fixed Objects in Driving Environments. In Proceedings of the 9th Scholar's Yearly Symposium of Technology, Engineering and Mathematics, Rome, Italy, December 3-6, 2023 (pp. 17–24). CEUR-WS.org