Dirk Gehrig
Papers
4
Total Citations
60
H-Index
4
About
Dirk Gehrig is a researcher whose work sits at the intersection of computer vision, human-robot interaction, and machine learning, with a particular focus on enabling robots to understand and anticipate human behavior. His most significant contributions center on developing multi-level recognition systems that allow humanoid robots to interpret human intention, activity, and motion in real-time using only monocular camera input. This work, detailed in his highly cited 2011 paper (with 27 and 22 citations across two versions), integrates domain knowledge with visual data to create robust, online recognition pipelines—a critical step toward deploying robots in unstructured household environments. Gehrig also advanced the field of human motion analysis through his 2008 study on feature selection for motion recognition from joint-angle trajectories, and his 2010 work on semantic segmentation of motion sequences, which aimed to parse continuous human movement into meaningful, discrete actions. Collectively, his research has laid foundational groundwork for context-aware robotic systems, earning over 60 citations and demonstrating a sustained focus on making robots not just observers, but active, intelligent partners in daily life.
Research Focus
Key Achievements
Top Papers
- 1
- 2
- 3Selecting relevant features for human motion recognition6 citations · 2008
- 4Towards Semantic Segmentation of Human Motion Sequences5 citations · 2010