LittleBrother: Vision Systems Supporting Detection of Offenders in Public Places. Current visual surveillance systems can track people in an area only if complete camera coverage is provided. This project will develop a visual surveillance system able to track and record people's movements in a public building requiring only limited visual coverage. We will propose novel ways of matching images of a single individual from distant cameras by using features such as color histograms decomposed for ....LittleBrother: Vision Systems Supporting Detection of Offenders in Public Places. Current visual surveillance systems can track people in an area only if complete camera coverage is provided. This project will develop a visual surveillance system able to track and record people's movements in a public building requiring only limited visual coverage. We will propose novel ways of matching images of a single individual from distant cameras by using features such as color histograms decomposed for the different body parts, estimated height, and build type. Creating a record with this tracking information will effectively support security officers in the identification of responsible parties in the event of an offence.Read moreRead less
Automatic Recognition of Human Activities in Surveillance Videos: Overcoming the Curse of Dimensionality. This project will deliver a technology capable of automatically recognising human activities of interest in surveillance videos. The project will tackle the challenging, huge complexity inherent in the recognition of human activities by novel statistical pattern recognition techniques. The outcome of this project will be an effective activity recognition technology that will help monitor the ....Automatic Recognition of Human Activities in Surveillance Videos: Overcoming the Curse of Dimensionality. This project will deliver a technology capable of automatically recognising human activities of interest in surveillance videos. The project will tackle the challenging, huge complexity inherent in the recognition of human activities by novel statistical pattern recognition techniques. The outcome of this project will be an effective activity recognition technology that will help monitor the security and safety of environments and support the further development of the Australian video surveillance industry.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
Generative Visual Pre-training on Unlabelled Big Data. This project aims to develop a generative visual pre-training of large-scale deep neural networks on unlabelled big data. Developing pre-trained visual models that are accurate, robust, and efficient for downstream tasks is a keystone of modern computer vision, but it poses challenges and knowledge gaps to existing unsupervised representation learning. Expected outcomes include new theories and algorithms for unsupervised visual pre-training ....Generative Visual Pre-training on Unlabelled Big Data. This project aims to develop a generative visual pre-training of large-scale deep neural networks on unlabelled big data. Developing pre-trained visual models that are accurate, robust, and efficient for downstream tasks is a keystone of modern computer vision, but it poses challenges and knowledge gaps to existing unsupervised representation learning. Expected outcomes include new theories and algorithms for unsupervised visual pre-training, which are anticipated to deepen our understanding of visual representation and make it easier to build and deploy computer vision applications and services. Examples of benefits include modernising machines in manufacturing and farming with visual intelligence. Read moreRead less
A Novel System for Surveillance of Moving Objects. Surveillance of moving objects is critical in numerous applications such as detection and recognition of motor vehicles. It is important for detection to be fast and accurate with low cost. In this project, we aim to implement a surveillance system consisting of an efficient algorithm on a PC network with a camera. Our detection algorithm will be achieved with an advanced and computationally powerful image representation for fast computation. It ....A Novel System for Surveillance of Moving Objects. Surveillance of moving objects is critical in numerous applications such as detection and recognition of motor vehicles. It is important for detection to be fast and accurate with low cost. In this project, we aim to implement a surveillance system consisting of an efficient algorithm on a PC network with a camera. Our detection algorithm will be achieved with an advanced and computationally powerful image representation for fast computation. Its accuracy will be enhanced by adapting a well recognized theory for fast removal of image noise. Our implementation on the PC network will provide a flexible and extensible platform for parallel computing to further reduce detection time while keeping costs low.Read moreRead less
Online Learning for Large Scale Structured Data in Complex Situations. Online Learning (OL) is the process of predicting answers for a sequence of questions. OL has enjoyed much attention in recent years due to its natural ability of processing large scale non-structured data and adapting to a changing environment. However, OL has three weaknesses: it does not scale for structured data; it often assumes that all of the data are equally important; it often considers that all of the data are compl ....Online Learning for Large Scale Structured Data in Complex Situations. Online Learning (OL) is the process of predicting answers for a sequence of questions. OL has enjoyed much attention in recent years due to its natural ability of processing large scale non-structured data and adapting to a changing environment. However, OL has three weaknesses: it does not scale for structured data; it often assumes that all of the data are equally important; it often considers that all of the data are complete and noise-free. These weaknesses limit its utility, because real data such as those that must be analysed in processing social networks, fraud detection do not satisfy the restrictions. The aim of this project is to develop theoretical and practical advances in OL that overcome the existing weaknesses.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
Two-way Auslan: Automatic Machine Translation of Australian Sign Language. This project aims to develop an automatic two-way machine-translation system between Auslan (Australian Sign Language) and English by researching and leveraging advanced computer vision and machine learning technology. The project expects to advance research in AI technology on topics including visual recognition, language processing and deep learning. This will boost Australia's national research capacity and global com ....Two-way Auslan: Automatic Machine Translation of Australian Sign Language. This project aims to develop an automatic two-way machine-translation system between Auslan (Australian Sign Language) and English by researching and leveraging advanced computer vision and machine learning technology. The project expects to advance research in AI technology on topics including visual recognition, language processing and deep learning. This will boost Australia's national research capacity and global competitiveness. Expected outcomes of this project will help to break the communication barriers between the Deaf and hearing population. This should provide significant benefits to Deaf communities through enhanced communication and improved quality-of-life, leading to a fair, more inclusive and resilient Australian society.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
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