Bertram Drost

Papers

Total Citations

246

H-Index

About

Bertram Drost is a leading researcher in 3D computer vision, with a primary focus on advancing object detection and pose estimation for industrial and robotic applications. His most impactful work, the introduction of the **MVTec ITODD dataset** (2017, 159 citations), established a critical benchmark for 3D object recognition in realistic industrial settings, providing a public standard that has driven progress in the field. Drost’s earlier seminal contribution, **"3D Object Detection and Localization Using Multimodal Point Pair Features"** (2012, 85 citations), pioneered a novel multimodal feature that fuses intensity and depth data, enabling scale- and rotation-invariant detection—a foundational technique for robust robotic manipulation and inspection tasks. His research also tackles fundamental computational challenges, such as fast 3D neighbor search and the detection of both rigid and deformable objects in point clouds. Through these contributions, Drost has significantly enhanced the speed, accuracy, and robustness of 3D perception systems, directly impacting automation and quality control in industry. His work remains highly cited and influential, shaping modern approaches to 3D object recognition.