About

Yohan Dupuis is a leading researcher at the intersection of robotics, computer vision, and human-robot collaboration. His work centers on developing intelligent perception systems that enable robots to understand and navigate complex environments. Dupuis’s most significant contribution is his pioneering study on the positioning performance of Vicon motion capture systems, which has garnered over 416 citations and serves as a foundational reference for researchers in biomechanics, clinical gait analysis, and robotics. He has also made substantial advances in creating synthetic data for machine learning, most notably through his work on digital twins for industrial workstations—a method that auto-labels data for human action recognition in collaborative robotics. His development of the OmniScape dataset (46 citations) addresses a critical gap in omnidirectional vision, providing ground-truth semantic and depth data for autonomous systems. Dupuis has further innovated in sensor fusion, combining omnidirectional and PTZ cameras for robust face detection and tracking, and has surveyed collaborative mobile robotics for semantic mapping. His research consistently pushes the boundaries of how robots perceive and interact with dynamic, unstructured environments.

Research Focus

Key Achievements

6
H-Index
10
Papers
605
Total Citations
61
Avg Citations/Paper
🏆 Most Cited Paper
A Study of Vicon System Positioning Performance
416 citations · 2017
📈 Most Prolific Year: 2017 (3 Papers)
🤝 Key Collaborators: 23
🏛 Institutions: Centre d'Études et d'Expertise sur les Risques, l'Environnement, la Mobilité et l'Aménagement, Centre d'Etudes Superieures Industrielles, École Supérieure d'Ingénieurs en Génie Électrique, Institut de Recherche sur les Systèmes Atomiques et Moléculaires Complexes, Normandie Université, Embedded Systems (United States)

Top Papers

  1. 1
  2. 2
  3. 3
    The OmniScape Dataset
    46 citations · 2020
  4. 4
  5. 5
  6. 6
  7. 7
  8. 8
  9. 9
  10. 10

Key Collaborators

Contact & Links

Available for collaboration
Content generated · 13 days ago