Federico Dassiè

Researcher and developer active in research on automated artifact scanning and intelligent data analysis. Passionate about innovative technological solutions at the intersection of humanities and engineering.

Interested in |

terminal

I primarily work with Python, but I also enjoy complementing my projects with frontend development. My experience includes image processing, 3D reconstruction (SfM, NeRF, Gaussian Splatting), and AI frameworks. Currently, I am exploring the potential of VLMs for cultural heritage applications, with a particular focus on automated textual analysis.

Projects

AI ColonySim

Social experiment testing SOTA AI agents behavior in a simple multiplayer survival environment simulation.

Redis RAG

Retrieval Augmented Generation (RAG) system using Redis and Agentic AI for efficient data retrieval, strict query checking, processing and cross-analysis of historical documents.

Palladia

Introducing a benchmark tool for Vision LLMs on historical image documents, based on the GT4HistOCR dataset.

CTE-Genova

Integration of 3D scanning and robotic systems for autonomous operations for the digitization of archeological artifacts in the context of the CTE-Genova project.

PyCrop

Image post-processing app with object detection, cropping, compression, and white balancing.

Mimic Me

Minigame webapp about mimicking given tasks using emojis, with different game modes available.

I Conti con la Storia

WebDoc (online documentary) on Italy's racial laws in media, combining scholarly study and web documentary formats.

Publications

DARS: A Dual-Arm Robotic System for Autonomous 3D Artifacts Scanning

J. Ahmad, F. Dassiè, S. Frascella, F. Cannella et al. – (submitted to) IROS 2025

A fully autonomous 3D scanning system using dual-arm robotics to preserve cultural heritage artifacts.

2025

AAPOE: Automated Artifacts Position and Orientation Estimation

J. Ahmad, S. Frascella, F. Dassiè et al. – MESA 2024

Novel method for estimating artifact orientation in cultural heritage using automated systems.

2024

Machine Learning and Computer Vision in the Humanities

F. Dassiè – Master's Thesis, Ca' Foscari

Exploring the intersection of machine learning and computer vision to advance humanities research.

2022

Work Experience

Research Fellow at the Center for Cultural Heritage Technology (CCHT-IIT), Venice

2024 - 2025

Application of computer vision techniques and robotics scripting to automate the scanning and analysis of cultural heritage artifacts, focusing on 3D reconstruction and data extraction.

Technologies used: Python, Robotics (UR, Robotiq), COLMAP, NerF/Gaussian Splatting, 3D Scanners and cameras (Artec, Polyga, Intel Realsense, Zivid), Genesis

Python Developer at Fondazione Giorgio Cini (ARCHiVe), Venice

2022 - 2024

Development of a pipeline for the automated post-production of images of ancient books and documents, involving object detection and metadata recognition + extraction.

Technologies used: Python, Pytorch, Detectron2, OpenCV, Leaflet, Adobe Lightroom, RawTherapee, Canon cameras