Paul Lesur

German Research Centre for Artificial Intelligence

Papers

Total Citations

144

H-Index

About

Paul Lesur is a leading researcher in computer vision and robotics, with a primary focus on augmented reality, object pose estimation, and autonomous navigation in unstructured environments. His most impactful work, "Deep Multi-state Object Pose Estimation for Augmented Reality Assembly" (100 citations), pioneered the application of neural networks to estimate the pose of multi-part objects, enabling precise AR-guided assembly—a critical advancement for manufacturing and maintenance. Lesur also made significant contributions to field robotics with "SLAM in the Field: An Evaluation of Monocular Mapping and Localization on Challenging Dynamic Agricultural Environment" (44 citations), where he demonstrated a hybrid system combining sparse visual SLAM with Multi-View Stereo to achieve robust localization in complex, dynamic agricultural settings. This work directly addresses the challenges of deploying autonomous vehicles in real-world, non-static environments. With over 140 total citations, Lesur’s research bridges the gap between laboratory-grade computer vision algorithms and practical, real-world deployment, particularly in agriculture and industrial AR. His achievements highlight a career dedicated to making autonomous systems more reliable and context-aware in demanding operational conditions.