Irvin Aloise

I am part of the Mapping and Localization team, designing and developing state-of-the-art computer vision algorithms for head tracking, and shipping them into products.

• Architected, developed, productionized, and shipped to customers a state-of-the-art large-scale mapping and localization pipeline to enable massive multi-user experiences and content persistence with ML2 devices.
• Technical leadership of the project, including mentoring and coaching of several junior-to-senior engineers. Cross-team engineering coordination and multi-functional collaboration with different stakeholders.
• Productization of complex vision algorithms to fit compute budgets and performance requirements of AR glasses' embedded systems, writing high-performance C++ code.
• Since 2024, collaborating with Google to develop and ship computer vision algorithms to millions of users via Android XR. Hands on work with several AR products (e.g. Samsung Galaxy XR), impacting the next generation of AR glasses.

Conducted doctoral research within the Department of Computer, Control, and Management Engineering "Antonio Ruberti" (DIAG), under the supervision of Prof. Giorgio Grisetti.

• Investigated advanced methodologies in Simultaneous Localization and Mapping (SLAM) and Graph Optimization.
• Designed unified modular architectures for modular and plug-and-play SLAM solvers in C++.
• Developed robust chordal-based error formulations and geometric primitive matchables for 3D Pose-Graph Optimization.

Conducted collaborative research with the Photogrammetry & Robotics Lab at Rheinische Friedrich-Wilhelms-Universität Bonn, Germany.

• Researched 3D-LiDAR SLAM and high-precision sensor calibration techniques.
• Implemented calibration algorithms for 3D-LiDAR and camera extrinsic parameter estimation.
• Released open-source tooling for AprilTag-based multi-sensor calibration [code repository].

Department of Computer, Control, and Management Engineering "Antonio Ruberti". Supervisor: Prof. Giorgio Grisetti. Research focus on mobile robotics, SLAM, point-cloud registration, and graph optimization.

First Class with Honors (110/110 cum laude). Program taught in English. Master's thesis: Pose-Graph Optimization [PDF], supervised by Prof. Giorgio Grisetti.

Core coursework: Least-Squares SLAM, Vision and Perception, Artificial Intelligence, Mobile Robotics, Robotics, Interactive Graphics.

Undergraduate thesis addressed the stability of humanoid robots subject to external forces, supervised by Prof. Giuseppe Oriolo.