Dynamic Visual Scene Gist Recognition using a Probabilistic Inference Framework. How can we see the forest without intentionally looking for the trees? How can we tell traffic is flowing smoothly on a busy highway without identifying vehicles or measuring their speed? These are the questions that inspire this research project. Humans are endowed with the ability to grasp the ‘gist’ or overall meaning of a complex visual scene from a single glance and without attention to details. The aim of this ....Dynamic Visual Scene Gist Recognition using a Probabilistic Inference Framework. How can we see the forest without intentionally looking for the trees? How can we tell traffic is flowing smoothly on a busy highway without identifying vehicles or measuring their speed? These are the questions that inspire this research project. Humans are endowed with the ability to grasp the ‘gist’ or overall meaning of a complex visual scene from a single glance and without attention to details. The aim of this project is to develop new computational vision models that combine biological visual processing with probabilistic inference for gist recognition. The developed models will be able to mimic human vision by analysing a complex dynamic scene rapidly and classifying its semantic categories, without identifying individual objects.Read moreRead less
Hybrid optimisation for automatic large-scale video annotation. Optimization is the basis for solving many problems in Computer Vision, such as three-dimensional geometry recovery, image segmentation, scene labeling and object recognition. This project will develop new optimisation techniques and demonstrate their suitability for large-scale video annotation, which is key to visual data mining and scene understanding.
Discovery Early Career Researcher Award - Grant ID: DE220101379
Funder
Australian Research Council
Funding Amount
$417,000.00
Summary
Towards Transferable Visual Understanding in the Real World. This project aims to investigate how to improve the transferability of visual understanding algorithm and system in the real-world applications. This project expects to innovate and advance knowledge in the fields of visual transfer learning and generalizable visual representation learning. Expected outcomes of this project include techniques and algorithms to make the visual understanding system robust to diverse real-world scenarios. ....Towards Transferable Visual Understanding in the Real World. This project aims to investigate how to improve the transferability of visual understanding algorithm and system in the real-world applications. This project expects to innovate and advance knowledge in the fields of visual transfer learning and generalizable visual representation learning. Expected outcomes of this project include techniques and algorithms to make the visual understanding system robust to diverse real-world scenarios. This project should provide significant benefits, such as improving the robustness and safety of autonomous vehicles in transportation area, and reducing the cost of destructive data collection for intelligent fault detection in advanced manufacturing area.Read moreRead less
Robust and scalable change detection in geo-spatial data. A flood of data in the form of text, images and video emanate from a proliferation of sensors. These data are collected but rarely analysed, rendering it meaningless. This project aims to develop new software and techniques to detect changes over time in large scale geographically referenced data (for example photomaps) for use across numerous domains.
Discovery Early Career Researcher Award - Grant ID: DE130101311
Funder
Australian Research Council
Funding Amount
$375,000.00
Summary
Predicting health status of geriatric patients from user trusted multimedia observations. The information technology developed in this project will provide health care specialists with a better window into the lives of elderly patients. Their behaviour can then be accurately interpreted, potentially leading to earlier recognition of problems and better treatment.
Discovery Early Career Researcher Award - Grant ID: DE130101775
Funder
Australian Research Council
Funding Amount
$375,000.00
Summary
Distributed large-scale optimisation methods in computer vision. With the number of images and video available over the internet reaching billions and growing, the need for new tools for handling and interpreting such huge amounts of data is quickly becoming apparent. This project will focus on developing new optimisation methods for efficiently computing solutions for a broad class of large-scale problems.
Added depth: automated high level image interpretation. Humans are very good at understanding the world through imagery, but computers lack this fundamental capacity because they lack experience of what they might see. This project will provide this experience by combining the large volumes of imagery on the Internet with three dimensional information generated by humans for other purposes.
Discovery Early Career Researcher Award - Grant ID: DE190101473
Funder
Australian Research Council
Funding Amount
$387,000.00
Summary
Feature-dependent label noise learning for big data analytics. This project aims to equip machines with the ability to robustly harness feature-dependent label noise from big data. The project expects to produce the potential to explore and exploit the weakly supervised information to better understand, interpret, and infer big data. Expected outcomes of this project include theoretical foundations for learning with label noise in the real-world scenarios and the next generation of intelligent s ....Feature-dependent label noise learning for big data analytics. This project aims to equip machines with the ability to robustly harness feature-dependent label noise from big data. The project expects to produce the potential to explore and exploit the weakly supervised information to better understand, interpret, and infer big data. Expected outcomes of this project include theoretical foundations for learning with label noise in the real-world scenarios and the next generation of intelligent systems to accommodate noisily annotated big data. This project should benefit science, society, and the economy nationally and internationally through the applications in the areas of artificial intelligence, cybersecurity, and big data analytics.Read moreRead less
Discovery Early Career Researcher Award - Grant ID: DE170101415
Funder
Australian Research Council
Funding Amount
$365,000.00
Summary
Life-long learning in the understanding of big data. This project aims to design and develop computational systems and algorithms that learn as humans do (life-long learning). This will enable systems to automatically interpret Big Data from social media, social network and public surveillance. This project will apply the knowledge learned in auxiliary Big Data sources to effectively interpret target tasks and analyse the communication network (one variant of social network). This project is exp ....Life-long learning in the understanding of big data. This project aims to design and develop computational systems and algorithms that learn as humans do (life-long learning). This will enable systems to automatically interpret Big Data from social media, social network and public surveillance. This project will apply the knowledge learned in auxiliary Big Data sources to effectively interpret target tasks and analyse the communication network (one variant of social network). This project is expected to benefit science, society and the economy, and help governments to better serve the public by improving transport logistics, modelling and regulation, and preventing crime and terrorism.Read moreRead less
Discovery Early Career Researcher Award - Grant ID: DE120101161
Funder
Australian Research Council
Funding Amount
$375,000.00
Summary
Compressive sensing based probabilistic graphical models (PGM). The aim of the project is to develop fast, large scale probabilistic graphical models (PGM) learning and inference methods. The resulting system will be able to process large scale PGMs on a standard PC, and will be easily extendable to computer clustering for larger scale PGMs requiring higher precision.