George Vosselman
Papers
8
Total Citations
470
H-Index
6
About
George Vosselman is a prominent researcher specializing in photogrammetry, point cloud processing, and semantic scene understanding, with significant contributions bridging remote sensing, computer vision, and robotics. His most influential work centers on developing benchmark datasets and deep learning methods for 3D and aerial data interpretation. Most notably, Vosselman contributed to the UAVid dataset for semantic segmentation of UAV imagery, which has garnered over 367 citations and become a foundational resource for researchers working on autonomous systems and aerial scene understanding. His research extends into airborne laser scanning (ALS), where he has applied deep learning architectures such as PointNet++ for tree species classification, advancing environmental monitoring capabilities. Vosselman has also tackled practical challenges in indoor mobile laser scanning, proposing line segmentation techniques for SLAM-based systems, and explored point cloud visibility analysis in close-range photogrammetry. More recently, his work has expanded into 3D scene graph construction from monocular cameras and LLM-enhanced indoor scene synthesis, reflecting a forward-looking engagement with emerging AI technologies. Across his career, Vosselman has consistently shaped how researchers acquire, process, and semantically interpret complex 3D spatial data.
Research Focus
Key Achievements
Top Papers
- 1UAVid: A semantic segmentation dataset for UAV imagery367 citations · 2020
- 2Visibility analysis of point cloud in close range photogrammetry39 citations · 2014
- 3
- 4
- 5The UAVid Dataset for Video Semantic Segmentation10 citations · 2018
- 6UAVid: A Semantic Segmentation Dataset for UAV Imagery9 citations · 2018
- 7
- 8