|
This group conducts research into algorithms and computer architectures for
a broad range of topics in signal and pattern processing.
It has well-established programmes of fundamental and applied research in:
- Computer vision
- Image processing
- Underwater acoustics (the University of Birmingham has been associated with this area for over 80 years)
- Computer architectures
- Speech and language processing, and
- Multimodal interaction.
Current research topics include:
- Model-based image interpretation.
- Unsupervised image segmentation.
- Motion analysis.
- 3D and multi-modal imaging.
- Heterogeneous computer architectures for integrated numerical and symbolic
computing.
- Fine-grain parallel architectures.
- New paradigms for high-accuracy speeech recognition and synthesis.
- Speaker, language and accent identification.
- Speech and Language Technologies in Education (SLaTE).
- Multimodal interaction (including speech, gaze and gesture).
- Speech technologies for children.
- Technology aspects of accents of the British Isles.
In addition to this basic research, the group is involved in the development
of large-scale engineering prototypes.
There is also a broad and expanding programme of multidisciplinary research
involving other groups in the University and external collaborators.
Examples include medical image processing, industrial inspection,
petrochemical surveying, vehicle hazard warning (RadarNet), automated
environmental analysis, powder flow modelling in chemical engineering, and
speech technology in education.
The group is engaged in collaborative research with a number of companies,
including
British Telecom,
BMW,
Daimler Chrysler,
QinetiQ
and DSTL,
Ensigma Technologies,
Fiat,
Jaguar,
Meritor,
Shell,
Siemens AT,
Sundance Multiprocessor Technology,
Unilever,
Volvo, and
Xilinx.
There are six academic staff, four research staff, and twenty odd
full-time research students and a number of external,
industrially-based PhD students.
It has traditionally attracted large quantities of external funding from
Government research laboratories, the European Union, industry, and
EPSRC.
Research students and staff come from diverse backgrounds, ranging from
electronic and electrical engineering, mechanical engineering, physics,
chemistry and computer science
through to speech processing and linguistics.
The group is internationally known for its post-experience courses on
Underwater Acoustics and Sonar Signal Processing, which are attended by
sonar system designers from around the world.
-
Vision, touch and balance (3D motion tracking system, CI Russell, PI Prof Alan Wing, School of Psychology) (1 year, 2007-2008)
-
A unified model for speech recognition and synthesis (3 years, EPSRC, 2005-2008)
-
Articulatory modelling for speech recognition and speaker identification (3 years, 2004-2007)
-
PF-STAR - Preparing Future multisensorial
interaction research (2 years, FP7, 2002-2004). The University of Birmingham led WP5: Speech Technologies for Children
-
Speech in multi-modal environments
(HP,
computing facilities).
-
Speaker recognition using an ‘articulatory’ segmental HMM (1year, EOARD (USA))
-
PUMA-Personalised user Interfaces for Information Management and
Authorisation (3 years, EPSRC-DTI, 2000-2003)
-
An integrated multiple-level statistical model for speech pattern processing
(Balthasar) (3 years, EPSRC, 1998-2001)
-
Biometrics for keyless automobile access (Meritor Light Vehicle Systems (UK)
Ltd, £45k, 1998-2001)
-
Development of data compression techniques for coronary angiography video
sequences (EPSRC, £160k, 1998-2001)
All multi-modal interaction laboratory group members can be contacted at:
Multi-modal Interaction Laboratory,
School of Electronic, Electrical and Computer Engineering,
The University Of Birmingham,
Edgbaston,
Birmingham B15 2TT,
United Kingdom
General Enquiries:
Fax:
Telex:
|
+44 (0)121 414 4285
+44 (0)121 414 4291
333762 UOBHAM G
|
Individual group members' contact information may be found on the people page.
|