Development of a three dimensional audio-visual next generation speech recognition system. To overcome the disadvantages of current Audio-Visual Speech Recognition Systems, we propose a set of robust algorithms in three dimensional computer vision and speech processing. The proposed system will have far-reaching implications in various areas, for example, human-machine interaction for speech recognition in automated dialog systems and voice-to-text conversions.
Computational tools to analyse and exploit the social media revolution. We aim to create technologies to analyse social media communities, which are rapidly growing in reach, complexity, and content produced and shared. Powerful techniques to tap this resource will lead to commercial outcomes for marketing and search industries, alongside deeper insight into the cultural and social impact of this Internet revolution.
Reading facial expressions from real and virtual humans. This project aims to advance understanding of human emotional communication and improve human rapport with the virtual humans and avatars that are rapidly infiltrating our social world. Using two unique stimulus sets - naturalistic human expressions and highly realistic virtual faces - together with powerful genetic, experimental, and individual differences designs, the project expects to answer previously intractable questions in emotion ....Reading facial expressions from real and virtual humans. This project aims to advance understanding of human emotional communication and improve human rapport with the virtual humans and avatars that are rapidly infiltrating our social world. Using two unique stimulus sets - naturalistic human expressions and highly realistic virtual faces - together with powerful genetic, experimental, and individual differences designs, the project expects to answer previously intractable questions in emotion science, as well as deliver tangible outcomes, such as new psychological tests to better understand human social connection. This should provide significant benefits, by improving emotion communication and offering a new perspective on how artificial intelligence can best serve human social needs.
Read moreRead less
Rethinking auditory alarms. This project aims to investigate whether auditory displays that give listeners background information about the status of monitored processes will lead to better situation awareness than conventional auditory alarms. Recent efforts to combat alarm fatigue by reducing auditory alarms may cause listeners to lose awareness of early process deterioration and its context. This project will investigate benefits of intermittent displays summarising process status with sound ....Rethinking auditory alarms. This project aims to investigate whether auditory displays that give listeners background information about the status of monitored processes will lead to better situation awareness than conventional auditory alarms. Recent efforts to combat alarm fatigue by reducing auditory alarms may cause listeners to lose awareness of early process deterioration and its context. This project will investigate benefits of intermittent displays summarising process status with sound or time-compressed speech while participants perform specific ongoing tasks. Anticipated outcomes are safer, more effective and better tolerated ways of monitoring processes than conventional auditory alarms. Results apply to healthcare, transportation and process industries.Read moreRead less
A 'smarter image' for educational multimedia: Improving learning of dynamic content with animated and static graphics. Animated images were once used largely for entertainment but are now increasingly a feature of multimedia information and education systems. However, recent research casts doubt on animation's assumed superiority over static pictures for facilitating understanding. This project aims to find how animations should be designed, supported, and used in order to fulfill their vast pot ....A 'smarter image' for educational multimedia: Improving learning of dynamic content with animated and static graphics. Animated images were once used largely for entertainment but are now increasingly a feature of multimedia information and education systems. However, recent research casts doubt on animation's assumed superiority over static pictures for facilitating understanding. This project aims to find how animations should be designed, supported, and used in order to fulfill their vast potential for explaining and instructing. It will examine in fine detail how people with different backgrounds actually use animations when trying to comprehend familiar and unfamiliar topics. Research-based guidelines will be produced on how to design effective animations and animated-static graphic blends for use in multimedia. Read moreRead less
3D Diffusion Models for Generating and Understanding 3D Scenes. Diffusion models, such as DALL-E2 and Imagen, have achieved remarkable success in generating photorealistic images and hold promise to solve long-standing computer vision problems. However, 3D scene generation remains unexplored. This research project aims to bridge the gap by developing 3D diffusion models capable of generating complete 3D scenes. This will advance our theoretical understanding of diffusion in complex 3D environmen ....3D Diffusion Models for Generating and Understanding 3D Scenes. Diffusion models, such as DALL-E2 and Imagen, have achieved remarkable success in generating photorealistic images and hold promise to solve long-standing computer vision problems. However, 3D scene generation remains unexplored. This research project aims to bridge the gap by developing 3D diffusion models capable of generating complete 3D scenes. This will advance our theoretical understanding of diffusion in complex 3D environments and open up new possibilities for applications in fields such as virtual reality, architecture, and city planning. The proposed 3D diffusion models will also enhance the accuracy of computer vision tasks related to 3D scene understanding, such as object detection, tracking, and semantic segmentation.Read moreRead less
Creating evidence-based public open space standards for apartment residents. This project aims to examine how to create public open space for apartment residents. It will (1) identify how apartment dwellers use public open space, including which spaces are used and why; and (2) test whether public open space use compensates for reduced apartment/building space. With less private space, apartment residents are assumed to rely on public spaces for physical, social and recreational activities. Yet ....Creating evidence-based public open space standards for apartment residents. This project aims to examine how to create public open space for apartment residents. It will (1) identify how apartment dwellers use public open space, including which spaces are used and why; and (2) test whether public open space use compensates for reduced apartment/building space. With less private space, apartment residents are assumed to rely on public spaces for physical, social and recreational activities. Yet little is known about which spaces they use and why, and what they would prefer. Expected outcomes include tailored, equitable evidence-based recommendations for public open space planning and apartment design guidelines. Benefits include the delivery of sustainable high-density precincts that cater to community needs.Read moreRead less
Optimal Robust Fitting under the Framework of LP-Type Problems. The project aims to develop algorithms to support the development of robust and accurate computer vision systems. Real-world visual data (images, videos) is inherently noisy and outlier prone. To build computer vision systems that work reliably in the real world, it is necessary to ensure that the underlying algorithms are robust and efficient. The project aims to devise novel algorithms that can compute the best possible result giv ....Optimal Robust Fitting under the Framework of LP-Type Problems. The project aims to develop algorithms to support the development of robust and accurate computer vision systems. Real-world visual data (images, videos) is inherently noisy and outlier prone. To build computer vision systems that work reliably in the real world, it is necessary to ensure that the underlying algorithms are robust and efficient. The project aims to devise novel algorithms that can compute the best possible result given the input data in a short amount of time. The expected outcomes would support the construction of reliable and accurate computer vision-based systems, such as large-scale 3-D reconstruction from photo collections, self-driving cars and domestic robots.Read moreRead less
Active multispectral computer vision for defence and security. This project will develop new techniques to extract intelligent information from multispectral images in the visible and near infra-red spectrum. It will enable computers to automatically recognise objects, faces and human actions with unprecedented accuracy.
Person identification from multiple non-invasive iris and face biometrics in video. This project will undertake research to develop a prototype system for personal identification that can be used by law enforcement and security agencies to enrol people at points of entry at public places. The system will non-invasively acquire face and iris biometrics and match them against a database of known persons. The proposed system can be used in sensitive buildings for access control, eliminating the nee ....Person identification from multiple non-invasive iris and face biometrics in video. This project will undertake research to develop a prototype system for personal identification that can be used by law enforcement and security agencies to enrol people at points of entry at public places. The system will non-invasively acquire face and iris biometrics and match them against a database of known persons. The proposed system can be used in sensitive buildings for access control, eliminating the need to carry access cards or remember passwords. This research contributes to the national research priority of Safeguarding Australia. We will develop new techniques in computer vision and train new researchers in this area.Read moreRead less