The Apostle research program delivered the commercial software product: Fivecast Insight.
The Apostle program was born from a need for national security and law enforcement agencies to be able to sift through vast data sources to efficiently produce useful and relevant insights.
The program helped agencies understand the data available to them, and how machine-learning and artificial intelligence can extract useful insights from those vast data sources.
The outputs from the Apostle research program are being commercialised by D2D CRC spinout company, Fivecast.
The Apostle research program delivered the commercial software product: Fivecast Insight.
The Apostle research program filed two patent applications for their innovations.
The Apostle research program produced 132 publications and technical reports over the five years of D2D CRC.
Ross led the D2D CRC development and operations engineering project team. This team supported research streams by collecting and curating various data sets. It also integrates outputs of the research streams and supports agencies in trialling new capabilities.
Professor Xie was the research lead for the Picturing Knowledge research stream. This stream aimed to develop core techniques to learn image-centric knowledge graphs by connecting large collections of image/video and their descriptions to existing knowledge bases.
Associate Professor Wang was the research lead for the Knowledge Graph Construction and Knowledge Graph Query research streams. This stream focused on transforming the 'noisy' data found on the internet into a structured form with the goal of enabling analysts to find the information they need faster.
Professor Yang was the research lead for the Semantic Indexing of Large Scale Video Archives research stream. This stream sought novel methods for semantic concept detection in videos. The project provided analysts with hassle free analytical tools for big video data management and utilisation.
Associate Professor Wobcke was the research lead for the Knowledge Mining research stream. The stream developed techniques for extracting knowledge (events and their associated entities) from a broad range of data sources.
Professor van den Hengel was the research lead for the Exploiting Contextual Cues in Large Scale Machine Learning and Visual Question Answering streams. This research developed technologies to accurately depict specific objects in images. It also developed image understanding technologies to access information in images as easily as text information.
