About

Dimitrios Kosmopoulos is a leading researcher at the intersection of computer vision, robotics, and assistive technologies. His work spans three core areas: robust visual perception for robotic manipulation, statistical machine learning for sensor data, and socially assistive systems. Kosmopoulos made foundational contributions to superquadric-based segmentation, developing methods that fuse region and boundary information to recover 3D object models from range images—critical for bin-picking and depalletizing tasks (cited over 60 times collectively). His 2010 variational Bayesian methodology for hidden Markov models using Student’s-t mixtures (62 citations) advanced robust statistical modeling of sensor data. In visual servoing, he proposed a robust Jacobian estimation technique (59 citations) that improved robot control under uncertainty. Beyond industrial robotics, Kosmopoulos pioneered rehabilitation gaming systems for cerebral palsy patients and developed visual competencies for socially assistive robots in the HOBBIT project. His work on real-time depalletizing systems demonstrated practical deployment of active vision in manufacturing. With over 200 total citations across these domains, Kosmopoulos continues to bridge theoretical advances in perception and learning with tangible robotic applications that enhance both industrial automation and human quality of life.

Research Focus

Key Achievements

6
H-Index
9
Papers
207
Total Citations
23
Avg Citations/Paper
🏆 Most Cited Paper
A variational Bayesian methodology for hidden Markov models utilizing Student's-t mixtures
62 citations · 2010
📈 Most Prolific Year: 2002 (3 Papers)
🤝 Key Collaborators: 18
🏛 Institutions: Institute of Informatics & Telecommunications, National Centre of Scientific Research "Demokritos", National Technical University of Athens, The University of Texas at Arlington, Hella (Germany), Institute for Medical Informatics and Biostatistics

Top Papers

  1. 1
  2. 2
  3. 3
  4. 4
  5. 5
  6. 6
  7. 7
  8. 8
  9. 9

Key Collaborators

Contact & Links

Available for collaboration
Content generated · 14 days ago