Statistical Methods of Model Fitting and Segmentation in Computer Vision. Electronic sensors such as cameras and lasers can provide a rich source of information about the position, shape, and motion of objects around us. However, to extract this information in a reliable, automatic, and accurate way requires a sophisticated statistical theory of the process. Example applications include: video surveillance (better automatic detection of moving people and vehicles and of characterising what those ....Statistical Methods of Model Fitting and Segmentation in Computer Vision. Electronic sensors such as cameras and lasers can provide a rich source of information about the position, shape, and motion of objects around us. However, to extract this information in a reliable, automatic, and accurate way requires a sophisticated statistical theory of the process. Example applications include: video surveillance (better automatic detection of moving people and vehicles and of characterising what those people and vehicles are doing), industrial prototyping and inspection (measuring the size and shape of objects), urban planning (laser scanning streetscapes to create computer models of cities), entertainment industry (movie special effects and games), etc. Read moreRead less
Visualisation of large, complex networks through small, beautiful diagrams. Data is increasingly organised as networks. Visualisation is a key way to understand networks. This project plans to develop a new paradigm for this task. Using modern generic constrained optimisation techniques it will produce layouts for small graphs whose quality is similar to that produced by hand, something that is not possible with current approaches. These algorithms will then be used to visualise large graphs. In ....Visualisation of large, complex networks through small, beautiful diagrams. Data is increasingly organised as networks. Visualisation is a key way to understand networks. This project plans to develop a new paradigm for this task. Using modern generic constrained optimisation techniques it will produce layouts for small graphs whose quality is similar to that produced by hand, something that is not possible with current approaches. These algorithms will then be used to visualise large graphs. Instead of simply trying to visualise every node and link in the graph. The project will develop techniques to extract useful subsets or abstractions that are as small possible, yet sufficient to answer targeted queries. The techniques for producing small high-quality diagrams will then be applicable to presenting these focused visualisations.Read moreRead less
Multiview video coding using cuboid data compression. This project investigates novel approaches to multiview video coding that use new data compression techniques and explicit occlusion handling. These new approaches complement the state-of-the-art, improving interactivity with instantaneous view change and VCR functionality, reducing encoding complexity, and increasing compression efficiency.
Effective profiling of large scale combinatorial optimisation problems. Finding optimum solutions to problems is one of the most common challenges in planning. It pervades all aspects of our social, environmental and economic life. However, designing programs that can solve optimisation problems effectively requires an iterative process that is often extremely challenging, time consuming and costly. For large-scale problems, this process can become impractical. This project will investigate meth ....Effective profiling of large scale combinatorial optimisation problems. Finding optimum solutions to problems is one of the most common challenges in planning. It pervades all aspects of our social, environmental and economic life. However, designing programs that can solve optimisation problems effectively requires an iterative process that is often extremely challenging, time consuming and costly. For large-scale problems, this process can become impractical. This project will investigate methods to profile and understand program performance. The results will help users to design scalable, efficient optimisation programs. This will in turn allow organisations large and small to reap the benefits of optimisation technology and, thus, make more efficient use of their resources.Read moreRead less
Methodologies for face recognition under varying imaging conditions. Face recognition systems are heavily dependent on the nature of the input to the system. Variability in appearance due to changes in illumination, expression, pose, etc. can reduce the recognition results of the existing systems. The aim of this project is to develop new techniques to improve the recognition accuracy in natural environment where unwanted image variations exist. The development of such techniques will be of grea ....Methodologies for face recognition under varying imaging conditions. Face recognition systems are heavily dependent on the nature of the input to the system. Variability in appearance due to changes in illumination, expression, pose, etc. can reduce the recognition results of the existing systems. The aim of this project is to develop new techniques to improve the recognition accuracy in natural environment where unwanted image variations exist. The development of such techniques will be of great importance to Australia's security and safety. The outcome of this research will provide the first steps towards formulating the next generation recognition systems that will improve the suitability of the face recognition for use in security, surveillance, intelligent robotics, banking, and smart environments.Read moreRead less
Accurate analysis of combinatorial problems: from the particular to the general. Combinatorial problems pervade all aspects of our social, environmental and economic life, but finding good solutions to these problems can take too much computer time. This project will develop new analysis tools that are effective at reducing this time, thus allowing for better solutions to be found.
Learning from learning solvers. Finding optimum solutions to everyday problems is one of the most common challenges in decision making. This project aims to design and implement effective analysis and transformation methods to improve models of combinatorial optimisation problems. Better models will enable more scalable and robust deployment of resources in all these areas, and do so immediately and at low risk and cost. The results will help users design better models while spending less time a ....Learning from learning solvers. Finding optimum solutions to everyday problems is one of the most common challenges in decision making. This project aims to design and implement effective analysis and transformation methods to improve models of combinatorial optimisation problems. Better models will enable more scalable and robust deployment of resources in all these areas, and do so immediately and at low risk and cost. The results will help users design better models while spending less time and money. This will in turn allow organisations large and small to reap the benefits of optimisation technology and, thus, make more efficient use of their resources.Read moreRead less
Semantic Vectorisation: From Bitmaps to Intelligent Representations. The objective of this innovative project is to provide a solution to the open question of representing natural images by semantically rich vector graphics. The challenges are to identify key visual and temporal elements for images and videos, and efficiently decompose the visual data into semantic vector representations that are faithful to original data, compact and editable. The project aims to investigate new bitmap-to-vecto ....Semantic Vectorisation: From Bitmaps to Intelligent Representations. The objective of this innovative project is to provide a solution to the open question of representing natural images by semantically rich vector graphics. The challenges are to identify key visual and temporal elements for images and videos, and efficiently decompose the visual data into semantic vector representations that are faithful to original data, compact and editable. The project aims to investigate new bitmap-to-vector conversion methods. It is expected to develop a framework where semantic labels and hyperlinks can be embedded in visual data automatically. It hopes to pioneer the creation of a web of images where the links are on image/video regions. New image simplification, stylisation, and non-photorealistic rendering methods are expected to be provided.Read moreRead less
Surviving the data deluge: Scalable feature extraction, discrimination and analysis for computer vision tasks using compressed sensed data. Strategically, our pioneering solutions besides being technically and socially significant, open fresh options for sensor-agnostic data analysis. The technical significance lies through the creation of new technologies for the critical national and global security markets, currently overwhelmed by data. The social significance arises from our solutions being ....Surviving the data deluge: Scalable feature extraction, discrimination and analysis for computer vision tasks using compressed sensed data. Strategically, our pioneering solutions besides being technically and socially significant, open fresh options for sensor-agnostic data analysis. The technical significance lies through the creation of new technologies for the critical national and global security markets, currently overwhelmed by data. The social significance arises from our solutions being privacy preserving, providing new avenues for the production of novel, socially acceptable products for aged care monitoring. Our methods spearhead future advancement in diverse disciplines due to the wide applicability of the methods to other sensor networks (Square Kilometre Array) and data types, providing new frameworks for addressing crucial problems of data management. Read moreRead less
Towards realistic verbal interactions between people and computers-a probabilistic approach. This project aims to facilitate natural spoken interactions between people and computer systems, addressing obstacles to the acceptance of these systems. We will investigate computational models for relevant aspects of spoken dialogue, which will be implemented in computer systems for diverse tasks (for example, home devices and phone-enabled services).