Michael Voit

Karlsruhe Institute of Technology, Fraunhofer Institute of Optronics, System Technologies and Image Exploitation, Fraunhofer Society

Papers

Total Citations

146

H-Index

About

Michael Voit is a leading researcher in human-robot interaction and computer vision, with a particular focus on enabling seamless, natural communication between humans and machines. His foundational work on the Karlsruhe Humanoid Robot, published in 2007 and garnering 135 citations, established key frameworks for multimodal interaction by integrating spontaneous speech recognition, visual perception, and dialogue processing. This early contribution laid the groundwork for more intuitive robotic systems. More recently, Voit has advanced the field of action understanding, proposing novel deep learning approaches for unsupervised 3D skeleton-based action recognition using cross-attention mechanisms, and conducting comprehensive surveys on action anticipation—a critical capability for autonomous driving and proactive robotics. His work also extends to practical cooperative systems, such as an HCI assembly station that uses dynamic projections and individualized assistance based on worker identity and pose. By bridging foundational interaction technologies with cutting-edge deep learning for action prediction, Voit’s research continues to shape how robots perceive, anticipate, and respond to human behavior in real-world settings.