In search of relevant things: A novel approach for image analysis. This project aims to investigate how experts’ cognitive processes may be transferred to computers for the automatic recognition of visual features. By merging computer and brain sciences, the project will characterise the way the brains of experts understand what is seen, in order to translate such a process in a new computer vision tool. This should provide significant benefits, such as automatic detection of threats or diseases ....In search of relevant things: A novel approach for image analysis. This project aims to investigate how experts’ cognitive processes may be transferred to computers for the automatic recognition of visual features. By merging computer and brain sciences, the project will characterise the way the brains of experts understand what is seen, in order to translate such a process in a new computer vision tool. This should provide significant benefits, such as automatic detection of threats or diseases in satellite and diagnostic imaging, respectively, among other applications. For the first time, the combination of how a computer analyses an image and how an expert interprets it will be used as a common language to enable machines to process visual information in a manner that mimics the way human brains do.Read moreRead less
Superhuman place recognition with a unified model of human visual processing and rodent spatial memory. Current robotic and personal navigation systems leave much to be desired; GPS only works in open outdoor areas, lasers are expensive and cameras are highly sensitive to changing environmental conditions. In contrast, nature has evolved superb navigation systems. This project aims to solve the challenging problem of place recognition, a key component of navigation, by modelling the visual recog ....Superhuman place recognition with a unified model of human visual processing and rodent spatial memory. Current robotic and personal navigation systems leave much to be desired; GPS only works in open outdoor areas, lasers are expensive and cameras are highly sensitive to changing environmental conditions. In contrast, nature has evolved superb navigation systems. This project aims to solve the challenging problem of place recognition, a key component of navigation, by modelling the visual recognition skills of humans and the rodent spatial memory system. This project looks to combine the best understood and most capable components of place recognition in nature to create a whole more capable than its parts, produce advances in robotic and personal navigation technology and lead to breakthroughs in understandings of the brain.Read moreRead less
Automatic video annotation by learning from web data. This project aims to study next-generation video annotation technologies to automatically tag raw videos using a huge set of semantic concepts. The project will study new domain adaptation schemes and frameworks in order to substantially improve video annotation performance. The resulting prototype system can be directly used by ordinary users worldwide to search their personal videos using textual queries. The system is also applicable to vi ....Automatic video annotation by learning from web data. This project aims to study next-generation video annotation technologies to automatically tag raw videos using a huge set of semantic concepts. The project will study new domain adaptation schemes and frameworks in order to substantially improve video annotation performance. The resulting prototype system can be directly used by ordinary users worldwide to search their personal videos using textual queries. The system is also applicable to video surveillance applications, which can enhance Australia’s homeland security.Read moreRead less
Multiview Complete Space Learning for Sparse Camera Network Research. Data analytics in video surveillance and social computing is a problem because data are represented by multiple heterogeneous features. This project will develop a multiview complete space learning framework to exploit heterogeneous properties to represent images obtained from sparse camera networks. It will integrate multiple features to identify people and understand behaviour, to build a database of activities occurring in ....Multiview Complete Space Learning for Sparse Camera Network Research. Data analytics in video surveillance and social computing is a problem because data are represented by multiple heterogeneous features. This project will develop a multiview complete space learning framework to exploit heterogeneous properties to represent images obtained from sparse camera networks. It will integrate multiple features to identify people and understand behaviour, to build a database of activities occurring in a wide area of surveillance. It will expand frontier technologies and safeguard Australia by providing warnings for hazardous (for example, overcrowding, trespassing), criminal, and terrorist situations. Results will be applicable internationally and enhance Australia’s role in machine learning and computer vision communities.Read moreRead less
AUSLearn: AUtomated Sample Learning for Object Recognition. This project aims to enable computers to learn how to effectively use training samples for object recognition. Training sample is the only source used by computers to learn recognising objects. This project creates a new research direction that will enable the first full exploration of the power of samples. The aims will be enabled by leveraging the recent advances in reinforcement learning, fast training algorithms, and by developing n ....AUSLearn: AUtomated Sample Learning for Object Recognition. This project aims to enable computers to learn how to effectively use training samples for object recognition. Training sample is the only source used by computers to learn recognising objects. This project creates a new research direction that will enable the first full exploration of the power of samples. The aims will be enabled by leveraging the recent advances in reinforcement learning, fast training algorithms, and by developing novel deep learning algorithms. The new algorithms will benefit a wide range of applications, e.g. to effectively use car crash training samples for accurately identifying potential road crashes in transport and to effectively use rare medical imaging training data for robustly diagnosing diseases in health.Read moreRead less
Delivering information suitable for studying spatial and temporal variability in benthic habitats using autonomous underwater vehicles. This project will develop the tools required to transform observations, made from autonomous underwater vehicles (AUV) of benthic habitats, into information that supports a better understanding of variability in benthic environments. This will allow for a coordinated and collaborative approach for data analysis and mapping to be undertaken.
Robust and Explainable 3D Computer Vision. Computer vision is increasingly relying on deep learning which is fragile, opaque and fails catastrophically without warning. This project aims to address these problems by developing new theory in graph representation of 3D geometric and image data, hierarchical graph simplification and novel modules designed specifically for deep learning over geometric graphs. Using these modules, it aims to design graph convolutional network architectures for self-s ....Robust and Explainable 3D Computer Vision. Computer vision is increasingly relying on deep learning which is fragile, opaque and fails catastrophically without warning. This project aims to address these problems by developing new theory in graph representation of 3D geometric and image data, hierarchical graph simplification and novel modules designed specifically for deep learning over geometric graphs. Using these modules, it aims to design graph convolutional network architectures for self-supervised learning that are robust to failures and provide explainable decisions for object detection and scene segmentation. The outcomes are expected to advance theory in robust deep learning and benefit 3D mapping, surveying, infrastructure monitoring, transport and robotics industries.Read moreRead less
Adapting Deep Learning for Real-world Medical Image Datasets. The project aims to investigate new deep learning modelling approaches to leverage real-world large-scale image data sets that contain noisy and incomplete labels and imbalanced class prevalence – to enable the use of these data sets for modelling deep learning classifiers. Expected outcomes include an innovative method for modelling deep learning classifiers. The research will involve new inter-disciplinary and international collabor ....Adapting Deep Learning for Real-world Medical Image Datasets. The project aims to investigate new deep learning modelling approaches to leverage real-world large-scale image data sets that contain noisy and incomplete labels and imbalanced class prevalence – to enable the use of these data sets for modelling deep learning classifiers. Expected outcomes include an innovative method for modelling deep learning classifiers. The research will involve new inter-disciplinary and international collaborations with machine learning and medical image analysis research institutions. This should provide significant benefits, such as better understanding of deep learning theory, new deep learning applications that can use previously unexplored data sets, and training for the future Australian workforce.Read moreRead less
The role of strong duality in computer vision. This project aims to undertake research in the fields of computer vision and optimization that will have a significant impact on the design of numerical algorithms for solving a wide range of problems in Computer Vision, Virtual Reality and Robotic Navigation. This project will advance understanding of a broad class of problems related to how computers interpret images. An expected outcome is the generation of novel mathematical theory and numerical ....The role of strong duality in computer vision. This project aims to undertake research in the fields of computer vision and optimization that will have a significant impact on the design of numerical algorithms for solving a wide range of problems in Computer Vision, Virtual Reality and Robotic Navigation. This project will advance understanding of a broad class of problems related to how computers interpret images. An expected outcome is the generation of novel mathematical theory and numerical algorithms capable of fundamentally changing the way problems relevant to a wide range of vision-related applications are solved. This should offer Australia a strong competitive advantage as a leader in scientific innovation in the areas of Computer Vision, Virtual Reality and Robotics and Autonomous Systems.Read moreRead less
Declarative Networks: Towards Robust and Explainable Deep Learning. The aim of this project is to develop declarative machine learning techniques that exploit inherent structure and models of the world. Deep learning has become the dominant approach for machine learning with many products and promises built on this technology. But deep learning is expensive, opaque, brittle and relies solely on human labelled data. This project intends to make deep learning more reliable by establishing theory a ....Declarative Networks: Towards Robust and Explainable Deep Learning. The aim of this project is to develop declarative machine learning techniques that exploit inherent structure and models of the world. Deep learning has become the dominant approach for machine learning with many products and promises built on this technology. But deep learning is expensive, opaque, brittle and relies solely on human labelled data. This project intends to make deep learning more reliable by establishing theory and algorithms that allow physical and mathematical models to be embedded within a deep learning framework, providing performance guarantees and interpretability. This would likely benefit machine learning based products that can understand the world and interact with humans naturally through vision and language.Read moreRead less