Thomas Deselaers

Papers

Total Citations

H-Index

About

Thomas Deselaers is a leading researcher in computer vision and machine learning, with a particular focus on real-time face detection, recognition, and image retrieval. His most influential work, "Randomized trees for real-time one-step face detection and recognition" (2008, 14 citations), introduced a pioneering system that simultaneously detects and recognizes faces in a single, efficient step, enabling instant learning of new identities—a critical capability for mobile service robotics. This contribution advanced the field by demonstrating how randomized decision trees could achieve real-time performance without sacrificing accuracy, bridging the gap between detection and recognition tasks. Deselaers’ broader research spans large-scale image classification, medical image analysis, and visual object recognition, where he has developed methods that improve both speed and robustness. His work has been widely cited and applied in interactive robotics and surveillance systems, reflecting its practical impact. Beyond his technical contributions, Deselaers has contributed to the community through collaborative projects and open-source tools, making him a respected figure in applied computer vision.