Automatic Machine Learning with Imperfect Data for Video Analysis . This project aims to propose new algorithms and technologies for constructing an efficient video analysis system, which will be aligned with Australia’s science and research priorities. Specifically, during this project, a novel network structure search method based on auto machine learning will be proposed, an unsupervised domain adaptation algorithm will be developed, and a generative data augmentation method will be construct ....Automatic Machine Learning with Imperfect Data for Video Analysis . This project aims to propose new algorithms and technologies for constructing an efficient video analysis system, which will be aligned with Australia’s science and research priorities. Specifically, during this project, a novel network structure search method based on auto machine learning will be proposed, an unsupervised domain adaptation algorithm will be developed, and a generative data augmentation method will be constructed. All of these will construct a stable and efficient deep neural network, which is able to process large size videos captured from real scenarios in high efficiencies. Various fields, such as health care service and cybersecurity, will benefit hugely from this project.Read moreRead less
Making Meta-learning Generalised . This project aims to develop novel machine learning techniques, termed generalised meta-learning, to make machines better utilise past experience to solve new tasks with few data. It expects to reduce the undesirable dependence of current machine learning on labelled data and significantly expand its application scope. Expected outcomes of the project consist of new theoretical results on meta-learning and a set of innovative algorithms that can support the bui ....Making Meta-learning Generalised . This project aims to develop novel machine learning techniques, termed generalised meta-learning, to make machines better utilise past experience to solve new tasks with few data. It expects to reduce the undesirable dependence of current machine learning on labelled data and significantly expand its application scope. Expected outcomes of the project consist of new theoretical results on meta-learning and a set of innovative algorithms that can support the building of next generation of computer vision systems to work in open and dynamic environments. This should be able to produce solid benefits to the science, society, and economy of Australian via the application of these advanced intelligent systems.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
Automatic Training Data Search and Model Evaluation by Measuring Domain Gap. We aim to investigate computer vision training data and test data, using automatically generated data sets for facial expression recognition and object re-identification. This project expects to quantify and understand the domain gap, the distribution difference between training and test data sets. Expected outcomes of this project are insights on measuring the domain gap, the ability to estimate model performance witho ....Automatic Training Data Search and Model Evaluation by Measuring Domain Gap. We aim to investigate computer vision training data and test data, using automatically generated data sets for facial expression recognition and object re-identification. This project expects to quantify and understand the domain gap, the distribution difference between training and test data sets. Expected outcomes of this project are insights on measuring the domain gap, the ability to estimate model performance without accessing expensive test labels and improvements to system generalisation. This should provide significant benefits for computer vision applications that currently require expensive labelling, and commercial and economic benefits across sectors such as transportation, security and manufacturing.Read moreRead less
Australian Laureate Fellowships - Grant ID: FL170100117
Funder
Australian Research Council
Funding Amount
$3,208,192.00
Summary
On snapping up semantics of dynamic pixels from moving cameras. The project aims to develop a suite of original models and algorithms for processing and understanding videos captured by moving cameras, and to establish the mathematical foundations for deep learning-based computer vision to provide theoretical underpinnings. The project expects to generate new knowledge that will transform moving-camera computer vision with step-changes in visual quality enhancement, compression and acceleration ....On snapping up semantics of dynamic pixels from moving cameras. The project aims to develop a suite of original models and algorithms for processing and understanding videos captured by moving cameras, and to establish the mathematical foundations for deep learning-based computer vision to provide theoretical underpinnings. The project expects to generate new knowledge that will transform moving-camera computer vision with step-changes in visual quality enhancement, compression and acceleration technologies, and solutions for fundamental computer vision tasks. A new concept of feature complexity for measuring the discriminant and learnable abilities of features from deep models will also be defined. The outcomes of the project will be critical for enabling autonomous machines to perceive and interact with the environment.Read moreRead less
Learning kernel-based high-order visual representation for image retrieval. Image retrieval plays a key role in many practical applications. The recent increase of real-world applications calls for higher retrieval accuracy. This project aims to address this issue by exploring advanced visual representation that models the high-order information of image content. This project expects to generate new knowledge in the area of computer vision by developing a novel image retrieval framework. Expecte ....Learning kernel-based high-order visual representation for image retrieval. Image retrieval plays a key role in many practical applications. The recent increase of real-world applications calls for higher retrieval accuracy. This project aims to address this issue by exploring advanced visual representation that models the high-order information of image content. This project expects to generate new knowledge in the area of computer vision by developing a novel image retrieval framework. Expected outcomes include theory development on visual representation and more effective retrieval techniques. This should provide significant benefits, such as improving public information access services, facilitating environmental monitoring, and enhancing smart traffic management.Read moreRead less
Assistive micro-navigation for vision impaired people. This project aims to develop novel algorithms to transform a simple camera into a smart sensor, that can enable a vision-impaired person to navigate freely and without additional aids in a crowded area. Such a smart sensor will be endowed with the capability to detect and locate obstacles, identify the walking path, recognise objects and traffic signs and convey step-by-step instructions to the user. The project outcomes are expected to impr ....Assistive micro-navigation for vision impaired people. This project aims to develop novel algorithms to transform a simple camera into a smart sensor, that can enable a vision-impaired person to navigate freely and without additional aids in a crowded area. Such a smart sensor will be endowed with the capability to detect and locate obstacles, identify the walking path, recognise objects and traffic signs and convey step-by-step instructions to the user. The project outcomes are expected to improve the well-being and accessibility to public areas for vision-impaired people and reduce physical access disparities for this disadvantaged and vulnerable group. Furthermore, technologies developed in this project can potentially be adapted for use in related special navigation applications such as road safety, self-driving vehicles, and autonomous robots.Read moreRead less
Transfer Learning Handling Causally Bilateral Shift . Transfer learning is a core step for machines to transfer knowledge. This Project aims to equip machines with the ability to harness complex causal structures for transfer learning. The Project expects to produce the next great step for artificial intelligence – the potential to explore and exploit complex causal information to better understand, reason, and trust transfer learning. Expected outcomes of this Project include theoretical founda ....Transfer Learning Handling Causally Bilateral Shift . Transfer learning is a core step for machines to transfer knowledge. This Project aims to equip machines with the ability to harness complex causal structures for transfer learning. The Project expects to produce the next great step for artificial intelligence – the potential to explore and exploit complex causal information to better understand, reason, and trust transfer learning. Expected outcomes of this Project include theoretical foundations for transfer learning utilising causality and the next generation of intelligent systems to accommodate data with complex causal structures. This should benefit science, society, and the economy nationally and internationally through the applications to analysing their corresponding complex data.Read moreRead less