Iuri Frosio
Papers
2
Total Citations
35
H-Index
2
About
Iuri Frosio is a leading researcher in robotics and computer vision, with a primary focus on bridging the sim-to-real gap for robotic manipulation. His work addresses one of the most fundamental challenges in deep learning for robotics: enabling policies trained in simulation to transfer effectively to real-world systems. In his highly cited 2017 paper, Frosio demonstrated a novel approach for sim-to-real transfer of accurate grasping, achieving the precise manipulation of a tiny 1.37 cm sphere using an eye-in-hand camera and continuous control. By decomposing the system into a vision module and a closed-loop controller, he established a practical framework that has influenced subsequent work in the field. Building on this, his 2020 research introduced segmentation as a domain-invariant state representation, explicitly identifying and addressing both dynamics and visual sources of the sim-real gap. This work has been instrumental in advancing the reliability of simulation-trained policies. With over 35 citations across his most prominent papers, Frosio’s contributions continue to shape how researchers approach robust, transferable robotic learning in complex, real-world environments.
Research Focus
Key Achievements
Top Papers
- 1
- 2How to Close Sim-Real Gap? Transfer with Segmentation!6 citations · 2020