Rembert Daems
Papers
1
Total Citations
4
H-Index
1
About
Rembert Daems is a computer vision researcher whose work focuses on advancing 6D object pose estimation, a critical capability for robotics, augmented reality, and autonomous systems. His most notable contribution, CenDerNet (2023), introduces a novel render-and-compare framework that leverages center and curvature representations to achieve precise pose estimation from monocular images. By decoupling object center detection from curvature-based refinement, Daems' approach addresses long-standing challenges in handling occlusions and textureless objects, achieving state-of-the-art accuracy on standard benchmarks. Though early in his career, his work has already garnered attention (4 citations), signaling growing influence in the field. Daems' research bridges geometric deep learning and 3D vision, offering practical solutions for real-world applications like robotic grasping and scene understanding. His innovative use of curvature priors represents a promising direction for robust pose estimation in cluttered environments.
Research Focus
Key Achievements
Top Papers
- 1