Karl Amende

Technische Universität Ilmenau, Valeo (France)

Papers

Total Citations

474

H-Index

About

Karl Amende is a prominent researcher in the field of autonomous driving perception, with a particular focus on real-time 3D object detection using LiDAR point cloud data. His most influential contribution, the Complex-YOLO framework, introduced an innovative Euler-Region-Proposal approach that enabled efficient, real-time 3D object detection on sparse point clouds — a notoriously challenging problem in computer vision. First presented in 2018 and refined in a 2019 publication that has since accumulated over 376 citations, Complex-YOLO addressed a critical bottleneck in autonomous vehicle pipelines by linking environmental understanding directly to motion planning and prediction. Building on this foundation, Amende extended his work with Complexer-YOLO, a novel fusion architecture that integrates neural network-based 3D detection with visual semantic segmentation to enable simultaneous object detection and tracking on semantic point clouds. This body of work has had a meaningful impact on robotics, augmented reality, and autonomous systems research. With a growing citation record and a clear trajectory from detection to full scene understanding, Amende has established himself as a key contributor to the rapidly evolving field of intelligent vehicle perception.