||Multi-modal 3D shape representations of objects. Develop novel Object-Centric Robot Perception strategies (e.g. shapes, object properties) by leveraging multi-modal information (RGB-D, Lidar, Force/Torque readings) and Task-driven Object representations. The algorithms should integrate multi-modal data capable of providing uncertainty estimation for vision-based robotic manipulation tasks.
Research Scientist - SpaceR group
SnT – Interdisciplinary Centre for Security, Reliability and Trust
UNIVERSITÉ DU LUXEMBOURG