Stephen Moseson
Papers
1
Total Citations
662
H-Index
1
About
Stephen Moseson is a leading researcher in robotic manipulation and computer vision, best known for pioneering data-driven approaches to robotic grasping. His most influential contribution is the development of a novel rectangle representation for grasp detection, introduced in his landmark 2011 paper "Efficient grasping from RGBD images," which has accumulated over 660 citations. This work revolutionized how robots interact with novel objects by enabling them to estimate full 7-dimensional gripper configurations—including 3D location, 3D orientation, and gripper opening width—directly from RGBD images. Moseson's approach was among the first to successfully apply machine learning to grasp objects never seen before by the robot, bridging the gap between perception and action. His research has had profound implications for industrial automation, service robotics, and assistive technologies, establishing foundational methods that continue to influence modern grasping algorithms. By combining efficient visual processing with practical robotic control, Moseson has helped make autonomous manipulation more reliable and adaptable, earning recognition as a key figure in the advancement of intelligent robotic systems.
Research Focus
Key Achievements
Top Papers
- 1