Robert McCraith
Papers
1
Total Citations
20
H-Index
1
About
Robert McCraith is a researcher whose work sits at the intersection of computer vision and robotics, with a primary focus on self-supervised learning for 3D scene understanding. His most-cited paper, "Monocular Depth Estimation with Self-supervised Instance Adaptation" (2020, 20 citations), tackles a critical challenge in robotics: learning accurate depth from a single camera without relying on expensive 3D ground truth data. McCraith’s key contribution lies in developing methods that adapt self-supervised depth estimation to real-world scenarios where multiple views of a scene may be unavailable—a common constraint in robotic deployment. This work bridges the gap between theoretical advances in self-supervised learning and practical, on-device applications, enabling robots to perceive depth reliably even under limited data conditions. By addressing the variability of visual input in dynamic environments, McCraith’s research has implications for autonomous navigation, manipulation, and mapping. His approach demonstrates how leveraging raw video data can reduce dependency on labeled datasets, making depth estimation more scalable and robust for field robotics.
Research Focus
Key Achievements
Top Papers
- 1Monocular Depth Estimation with Self-supervised Instance Adaptation20 citations · 2020