Hannes Schulz
Papers
10
Total Citations
446
H-Index
6
About
Hannes Schulz is a robotics and computer vision researcher whose work spans robot perception, scene understanding, and autonomous robot behavior. He is perhaps best known for his highly influential 2015 paper on RGB-D object recognition and pose estimation using pre-trained convolutional neural networks, which has garnered over 320 citations and demonstrated how transfer learning can dramatically reduce the cost of training manipulation robots on new datasets. This contribution helped establish deep CNN-based approaches as a practical tool for real-world robotic perception. Beyond object recognition, Schulz has made meaningful contributions to semantic scene segmentation, combining geometric and semantic features to enable robots to interpret complex indoor environments — work that reflects a consistent focus on making autonomous agents more contextually aware. His earlier research addressed foundational robotics challenges, including field-line-based localization for RoboCup soccer robots and hierarchical reactive control for humanoid robot teams, with his team earning recognition as RoboCup 2011 Humanoid League Winners. He has also explored human-robot interaction through real-time hand gesture recognition systems for mobile robots. Taken together, Schulz's body of work reflects a career dedicated to bridging perception, learning, and autonomous action in robotics systems.
Research Focus
Key Achievements
Top Papers
- 1
- 2
- 3Real time interaction with mobile robots using hand gestures28 citations · 2012
- 4
- 5
- 6RoboCup 2011 Humanoid League Winners7 citations · 2012
- 7Semantic segmentation priors for object discovery5 citations · 2016
- 8Hierarchical reactive control for a team of humanoid soccer robots5 citations · 2007
- 9
- 10Line structure-based localization for soccer robots2 citations · 2009