Fabian Manhardt

Google (United States), Technical University of Munich

Papers

5

Total Citations

144

H-Index

5

About

Fabian Manhardt is a computer vision and robotics researcher whose work sits at the intersection of 6D object pose estimation, robotic grasping, and embodied AI. His research addresses some of the most persistent challenges in enabling robots to perceive and interact with the physical world, particularly under real-world constraints such as limited training data, photometrically challenging objects, and the need to generalize across large numbers of object categories. Manhardt's most influential contributions include MonoGraspNet (2023, 42 citations), which advances 6-DoF robotic grasping using only a single RGB image — a significant step beyond depth-sensor-dependent methods — and his CPS and CPS++ frameworks (2020, 35 and 14 citations), which introduced class-level 6D pose and shape estimation from monocular images, enabling pose estimation at scale across hundreds of object instances rather than just a handful. His work on DemoGrasp (2021, 29 citations) further demonstrated innovation in few-shot learning for grasping guided by human demonstration, while SG-Bot (2024, 24 citations) extended his reach into scene-level object rearrangement using scene graphs. Collectively, his research has meaningfully advanced the practical deployment of intelligent robotic systems in unstructured environments.

Research Focus

Key Achievements

5
H-Index
5
Papers
144
Total Citations
29
Avg Citations/Paper
🏆 Most Cited Paper
MonoGraspNet: 6-DoF Grasping with a Single RGB Image
42 citations · 2023
📈 Most Prolific Year: 2020 (2 Papers)
🤝 Key Collaborators: 17
🏛 Institutions: Google (United States), Technical University of Munich

Top Papers

  1. 1
  2. 2
  3. 3
  4. 4
  5. 5

Key Collaborators

Contact & Links

Available for collaboration
Content generated · 14 days ago