Giulia Vezzani
Papers
12
Total Citations
155
H-Index
7
About
Giulia Vezzani is a leading roboticist whose research sits at the intersection of perception, manipulation, and learning for autonomous systems. Her core contributions center on enabling robots—particularly humanoid platforms like the iCub—to interact with unknown objects in unstructured environments. She pioneered a superquadric-based grasping approach that models both object shape and graspable volume, a method that has garnered over 50 citations and remains foundational in the field. Vezzani’s work extends to robust 6D object pose tracking (MaskUKF, 13 citations) and markerless visual servoing, both critical for precise real-world manipulation. She also contributed to the GRASPA benchmark (31 citations), providing standardized protocols for evaluating grasping pipelines. Her recent involvement in RoboCat (2023) reflects a shift toward multi-embodiment, self-improving agents that learn from heterogeneous robotic experience. Beyond vision, she has advanced tactile object recognition and soft fingertip sensor integration, bridging the gap between sensing and control. With a publication record spanning over a decade and citations exceeding 150, Vezzani’s work consistently pushes the boundaries of dexterous, perceptive, and generalizable robotic manipulation.
Research Focus
Key Achievements
Top Papers
- 1A grasping approach based on superquadric models53 citations · 2017
- 2GRASPA 1.0: GRASPA is a Robot Arm graSping Performance BenchmArk31 citations · 2020
- 3Improving Superquadric Modeling and Grasping with Prior on Object Shapes13 citations · 2018
- 4
- 5RoboCat: A Self-Improving Generalist Agent for Robotic Manipulation9 citations · 2023
- 6
- 7A novel Bayesian filtering approach to tactile object recognition7 citations · 2016
- 8A novel pipeline for bi-manual handover task6 citations · 2017
- 9
- 10