
Intelligent & Visual Computing (IVC) Research explores the various aspects of Artificial Intelligence and Visual Information Processing. In the next 10 to 20 years, we expect to see the rapid emergence of true intelligent computational systems. Such systems will not be confined to desktop machines but will pervade everyday activities. Some of the techniques that our members are currently working on include computational intelligence approaches such as neural networks and evolutionary algorithms. IVC research group also performs research in both fundamental and applied problems in signal, image and visual processing, object recognition, object tracking, image compression, video coding and computer vision. We also work on real-time visual information systems using DSP, FPGA and WSN computing. Our research focuses on developing the next generation of intelligent computing environments with smart systems and human computing interfaces for everyone. Our research activities include artificial intelligence, machine learning, visual information processing, biometrics and applications for society such as for affective computing and business intelligence.
Machine Learning focuses on the design and development of algorithms that learn from existing incomplete data, generalize it to create rules of behaviour, and then make intelligent predictions of future behaviour of missing or future data. Bayesian belief networks, artificial neural networks and support vector machines have been successfully utilized to effect machine learning algorithms. Applications include measuring trust, object relationships extracted into ontologies, and adaptive neural networks for dynamic learning.
We address the topic of intelligent systems in a number of critical areas. We work on a range of intelligent and hybrid technologies, including neural networks, fuzzy systems, genetic / evolutionary algorithms, hybrid intelligent systems, intelligent systems in cognitive radio networks, wireless sensor networks, computer vision, emulation of sensory aspects of the biological brain in hardware and software. There is a strong emphasis on embedded systems, and in particular the incorporation of greater intelligence into embedded systems.
Our research has multidisciplinary foundations; including signal, image and video processing, object recognition, object tracking, image compression and video coding. For visual information engineering, we work on real-time visual information systems using DSP, FPGA and WSN computing. We also develop advanced vision systems for driver assistance and safety applications. The research on automotive vision systems focuses on real-time visual processing technologies. Others are visual surveillance and applications of visual and information systems for society.
The research aims to develop intelligent biometric, multi-biometric and audio-visual systems. One of our research focuses is audio-visual authentication or recognition system over IP or wireless sensor network. Other research activities are to develop real-time multi-biometric systems e.g. face-iris, face-speech, face-fingerprint. Other projects are audio-visual speech recognition, automatic 3D model-based face recognition, etc. The research works will have benefits for the technology and contribute to the community in terms of increased security.
The research of handwritten character recognition is one of the topics of pattern recognition which is classified into two categories: online handwritten character recognition and offline handwritten character recognition. Handwritten character recognition is one of the challenging aspects of the Chinese character recognition system due to varieties in the number of strokes, the order of strokes and shapes. We study on the various methods of Chinese character recognition on unconstrained online Chinese handwriting.
HCI is the study, planning, and design of the interaction between people and computers. One of our research activities focuses on the study, planning and design of the interaction between people and public display in order to create an engaging and informative environment. The research falls between the computer science and behavioural science fields of study. We focus the interaction between users and public display at the user interface level, which includes both software (to display the contents) and hardware (to use camera as sensor for tracking the gesture). Another research focus is multimodal interface, for example, audio visual speech recognition for HCI. We also develop HCI systems for people with disability eg. hearing impaired, visual impaired, etc.
Affective Computing research combines engineering and computer science with psychology, cognitive science, etc. The research brings together individuals with a diversity of technical, artistic, and human abilities in a collaborative spirit to push the boundaries of what can be achieved to improve human affective experience with technology. Our research activities include emotion recognition, intelligent techniques in affective computing, etc.