Computer Graphics, Computer Vision and Image Processing
The Computer Graphics, Computer Vision and Image Processing group studies how artificial systems can interact, replicate and give insight into the visual sense of humans. From the computer graphics of Hollywood to the fundamental understanding of neuroscience related to vision, there is a wide range of advances and interesting discoveries to be made. Computer graphics is the foundation for video games and movie visual effects, and our researchers have links with Wellington's internationally recognised entertainment industry. Computer Vision seeks to improve the sensing of autonomous platforms from CCTV cameras to mobile robots and even anthropomorphic humanoids. Image Processing addresses the problem of extracting the information an image holds, from identifying unusual activity in airports to evolving detectors to autonomously outline objects of interest.
The research group is currently in the following main areas: Computer Graphics, Computer Vision, Image Processing, and Visualisation.
- Computer Graphics (CG) is a broad area that can incorporate many interests. Our group has particular strengths in data-driven techniques, character animation, mobile and real-time graphics, and algorithms for visual effects. We are also interested in research on perceptual aspects of graphics and other areas. Our computer graphics researchers teach courses in the new computer graphics programme developed in conjunction with visual effects studio Weta, game developer Sidhe (PikPok), and advance technology company Unlimited Realities.
- Computer Vision (CV) involves different levels of computer and machine vision, ranging from edge detection, through segmentation, feature extraction and selection, object classification and detection, to robot vision. The techniques in this area include the conventional methods and the artificial intelligence particularly computational intelligence methods such as neural networks, evolutionary computation, and fuzzy systems. The group has good international reputation in Evolutionary Computer Vision, and currently holds top-level positions in IEEE CIS Task Force in Evolutionary Computer Vision and Image Processing, European Joint Conference on Evolutionary Computing (EvoStar), IEEE Congress on Evolutionary Computation.
- Active Vision (AV) allows a system to "intelligently" use its sensing resources when the world is too complex to allow full processing of incoming image data. We are exploring a number of approaches that allow a machine to sensibly direct its cameras to capture the information that is most pertinent to the tasks that the machine is performing.
- Image Processing: (IP) is a broad field of research that seeks to enhance particular features of an image or extract information from images. Image processing techniques are used in much of the groups' research, but our research in this area is primarily focused on the development of a variety of feature detection algorithms for use in machine vision and astronomy. In particular we are interested in the development of efficient algorithms for deployment on GPU architectures.
- Information Visualisation: (IV) is a field of research that explores ways to present and interact with visual (and audio) representations of large-scale data that doesn't have natural geo-spatial properties. While the wider field of IV looks also at scaling and structuring the data so that it can be best visualised, our group mostly focuses on (a) graph layout and (b) navigation within multiple visual representations of the same underlying data model. We are currently interested in multiple domains, including such areas as software visualisation, sports visualisation, and network visualisation.
Find out more about our Research Team and how to work with them:
Members of the group have produced a number of publications and other outputs over the years:
- The new meeting time for T1 2013 will be posted shortly.