Visual learning and understanding

The Visual Learning and Understanding research program investigates the fundamental problem of how a robot can learn to reliably interpret its environment, and build an internal representation of its surroundings in order to decide on its actions.

The program addresses research questions such as how machine learning for visual perception can be made safe, trustworthy, and reliable; how robots can understand and represent the geometry, semantics, and functionality of their surroundings and the task-relevant objects therein; and how robots can use this internal representation and learn to decide or plan their next actions in order to accomplish a useful task in a safe way.


PhD Topics


Led by Associate Professor Niko Sünderhauf