Kieran A. Murphy
Papers
1
Total Citations
7
H-Index
1
About
Kieran A. Murphy is a researcher whose work lies at the intersection of computer vision, robotics, and probabilistic machine learning, with a particular focus on pose estimation and geometric deep learning. His most notable contribution is the development of Implicit-PDF, a non-parametric representation for probability distributions on the rotation manifold, introduced in his 2021 paper. This work addresses two critical challenges in single image pose estimation: modeling uncertainty in predictions and handling symmetric objects with multiple valid poses. By representing distributions directly on the rotation manifold, Murphy’s approach enables more robust and principled pose estimation, moving beyond traditional point estimates. His work has garnered attention in the vision and robotics communities, with his most-cited paper accumulating 7 citations and influencing subsequent research in uncertainty-aware perception. Murphy’s contributions are particularly valuable for applications requiring reliable object pose estimation under ambiguity, such as robotic manipulation and augmented reality. His research exemplifies the growing importance of probabilistic reasoning in geometric computer vision, offering a foundation for more trustworthy and interpretable models in real-world systems.
Research Focus
Key Achievements
Top Papers
- 1