Papers
4
Total Citations
74
H-Index
3
About
Elia Bonetto is a robotics and computer vision researcher whose work sits at the intersection of autonomous aerial systems, human motion capture, and synthetic data generation. He is best known for pioneering methods that enable teams of unmanned aerial vehicles (UAVs) to perform markerless 3D human motion capture in unstructured, outdoor environments — a problem that had long been constrained by the need for calibrated cameras and offline processing. His 2022 paper *AirPose*, which has accumulated 35 citations, introduced a multi-view fusion network allowing UAVs to estimate human pose and body shape in real time using only onboard RGB cameras and computation. Complementing this, his deep reinforcement learning-based formation controller *AirCapRL* (2020, 34 citations) demonstrated how multi-robot systems could autonomously coordinate to track and capture human motion. More recently, Bonetto has expanded his focus to simulation and synthetic data with the *GRADE* framework, which leverages NVIDIA Isaac Sim to generate physically realistic, dynamic environments tailored specifically for robotics research — addressing a critical gap between computer vision datasets and robotics applications. His body of work reflects a consistent drive to bring robust, autonomous perception capabilities into real-world, uncontrolled settings.
Research Focus
Key Achievements
Top Papers
- 1
- 2
- 3
- 4