Center for Processing Speech and Images
The center for Processing Speech and Images (PSI) is with 70 people one of the big research units within the Electrical Engineering (ESAT) Department. Research covers a large area in the field of audio-visual information processing.
The combination of speech and image processing within the same group is an important advantage for the future, which is quite unique within the European context. Our researchers come from different backgrounds: engineers, physicists, mathematicians, linguists,... This multi disciplinarity gives PSI the possibility to couple fundamental innovation with practical solutions in a broad spectrum of applications.
The large number of European projects and 2 ERC grant holders in our ranks, underlines our international reputation. A high position in impact ranking (e.g. h83, h53, h41 and h31), numerous awards (e.g. David Marr prize, Henry Ford Conservation Award, European IST prize, etc.) and positions in the organization of top conferences and editorial functions is a further proof of the groups strength. PSI has also been a starting point of successful spin-offs and a reliable partner for industry.
The center consists of three subgroups:
- MIC is the Medical Image Computing group. Work is done on Medical Imaging (reduction metal artefacts (CT), echocardiography simulation, reconstruction for fast MR-imaging), Quantitative Image Analysis (image segmentation, rigid & non-rigid registration, shape analysis, brain morphometry), visualisation (3D surgery planning with soft tissue deformation, augmented reality & immersive visualisation), Multimedia & Telecommunication (PACS/HIS, hypermedia, teleworking).
- VISICS (VISion for Industry Communications and Services) concentrates on image processing for industrial applications, communication and the service sector. The most important areas of research are virtual reality, multimedia and image compression, 3D reconstruction, surveillance, space applications, robot vision/navigation, remote sensing and visual inspection.
The SPEECH group works on automatic speech recognition by means of Hidden Markov Models and Neural networks, processing and quality improvement of audio signals in noisy environments, speech corpora for Dutch, speech driven user interfaces and speech coding and speech transformations.
You can reach us at
- Tel.: +32-(0)16-32.17.13
- Fax: +32-(0)16-32.17.23