Our research experts
Dr. Jürgen Lüttin
Cognitive systems – Machine learning and semantic technologies meet artificial intelligence
“We build real-world intelligent cognitive systems that assist people in their daily life in a context-dependent and personalized way. To do this, we combine methods from different disciplines such as machine learning and semantic technologies.”
I am a research engineer in the Dependable Connected Systems Department (CR/AEX). I work in the area of cognitive systems and semantic technologies. My research interests include machine learning, knowledge management, semantic inference, and reasoning. Application areas of my research include driver assistance, speech recognition, computer vision, face recognition, smart agriculture, handwriting recognition, and video surveillance.
Robert Bosch GmbH
2002 – 2008
Systems Engineer in the business unit Driver Assistance
Development of video-based driver assistance systems including night vision, traffic sign recognition, lane keeping assistant, headlight assistant and emergency brake assistant.
Ascom AG, Switzerland
2000 - 2002
Head of the Pattern Recognition Group
Research and product development of intelligent video surveillance systems.
Johns Hopkins University, USA
1997 and 2000
Invited Scientist, Center for Language and Speech Processing
Research in audiovisual speech recognition, collaboration with U.S. Dep. of Defense and IBM.
IDIAP Research Institute, Switzerland
1997 - 2000
Head of the Computer Vision Group
Research in speech recognition, computer vision, biometrics and script recognition.
G. Potamianos et al. (2012)Audiovisual automatic speech recognition
- G. Potamianos, C. Neti, J. Luettin, I. Matthews
- Book chapter in Audiovisual Speech Processing, Cambridge University Press, p. 193-247
B. Fasel & J. Luettin (2003)Automatic Facial Expression Analysis: A Survey
- Pattern recognition, vol. 36, issue 1, p. 259-275
C. Kaas et al. (2002)Evaluation of a Self-learning Event Detector
- C. Kaas, J. Luettin, R. Mattone, K. Zahn
- Book chapter in: Video-Based Surveillance Systems, Springer
J. Luettin et al. (2001)Asynchronous stream modeling for large vocabulary audio-visual speech recognition
- J. Luettin, G. Potamianos, C. Neti
- IEEE ICASSP (1) 169-172
A. Vinciarelli & J. Luettin (2000)Off-Line Cursive Script Recognition based on Continuous Density HMM
- Frontiers in Handwriting Recognition, 493-498
S. Dupont & J. Luettin (2000)Audio-Visual Speech Modeling for Continuous Speech Recognition
- IEEE Transactions on Multimedia, vol. 2, issue 3, 141-151
S. Ben-Yacoub et al. (1999)Fast Face Detection using MLP and FFT
- S. Ben-Yacoub, B. Fasel, J. Luettin
- Audio Visual Biometric Person Authentication, 31-36
P. Jourlin et al. (1997)Acoustic-Labial Speaker Verification
- P. Jourlin, J. Luettin, D. Genoud, H. Wassner
- Pattern Recognition Letters 18, 853-858
J. Luettin & N. A. Thacker (1997)Speech Reading using Probabilistic Models
- Computer Vision and Image Understanding, vol. 65, issue 2, p. 163-178
J. Luettin et al. (1996)Active shape models for visual speech feature extraction
- J. Luettin, N. A. Thacker, S. Beet
- Book chapter in NATO ASI Speechreading by Man and Machine, Springer Verlag, vol. 95, issue 44
Interview with Dr. Jürgen Lüttin
Research Scientist Semantics and Machine Learning based Cognitive Systems
Please tell us what fascinates you most about research.
It fascinates me to work at the cutting edge of research, on problems that nobody has solved before, to deep dive on a specific problem, to look at all aspects and all perspectives and to bring technology one step forward.
What makes research done at Bosch so special?
Bosch has a great research team. We work in multicultural, international teams with diverse minds and backgrounds, which I enjoy very much. Bosch offers one of the best environments in the world to do research. Its research centers are located around the world, and Bosch’s success is built on a history of over 130 years. Our research topics are as diverse as our worldwide business units and customers, and if you have an idea in a new product area, you can even initiate a start-up inside Bosch to realize it. The Bosch values of openness, trust, and work-life balance provide the basis for our “Passion for innovation".
What research topics are you currently working on at Bosch?
I work on cognitive assistance systems in the automotive domain. These systems know all kinds of things about a user. They learn their habits, their preferences, and what they like and dislike. This personal information is combined with world knowledge in order to provide personal assistance. Really useful assistance is very much dependent on the context the user is currently in and on finding his particular needs in that situation. We therefore sense different kind of context information to provide context-dependent personal assistance. Our system provides assistance before the user has even asked for it.
What are the biggest scientific challenges in your field of research?
Human language is the basis for human communication and the way our knowledge of the world is presented. To understand this knowledge, the semantics and pragmatics by machine is however non-trivial. The knowledge needs to be presented in a form that a machine can use for knowledge extraction, understanding and reasoning in order to solve complex tasks in assistance system. How to do this is one of the main challenges.
How do the results of your research become part of solutions "Invented for life"?
In our research projects we work in close collaboration with the business units that will take up our results and exploit them in the next innovative products. For example, in one project, I contributed to video-based driver assistance systems. These Bosch products and their successors are now used in millions of cars to make driving safer – in the full sense of "Invented for life".