- Do you have a PhD in computer science, engineering, mathematics or a similar discipline?
- Like the sound of developing and deploying AI algorithms and workflows for underwater ocean sensing?
- Join our world-leading Distributed Sensing Systems group in this 2-year postdoctoral role
CSIRO Early Research Career (CERC) Postdoctoral Fellowships provide opportunities to scientists and engineers who have completed their doctorate and have less than three years of relevant postdoctoral work experience. These fellowships aim to develop the next generation of future leaders of the innovation system.
We are seeking a talented and motivated CERC Fellow to join our Distributed Sensing Systems (DSS) group, which is a world-leading research group specialised on large-scale, decentralized, and intelligent sensing technologies.
The research focus of this role is on the development and deployment of AI algorithms and workflows for underwater ocean sensing. The CERC Fellow with work with CSIRO mobile surveillance platforms for oceans that collect image/video data streams, location and motion information, and environmental parameters from under water. The goal of the project will be to enhance the efficiency and quality of data/knowledge acquisition on large scales through employing artificial intelligence (AI) algorithms that run on sensing devices, at edge, or in the cloud. We work directly with external users of the technology and translate scientific research to improvements in productivity and sustainability of various industries. Technologies developed by the DSS group have been deployed at continental scales in Australia and around the world in a broad range of environmental, agricultural, and industrial applications.
The main research area of the CERC Fellow will be machine learning (ML)/AI for on-device processing/edge computing. They will develop new concepts, algorithms, and software tools to speed up the development of on-device analytics for ocean survey sensor platforms. The main objectives include analytics for multi-modal data streams, such as video, motion, and location:
- Automation of the data annotation using ML models, weakly/semi-supervised training of ML algorithms, and active learning based on human input
- Incorporation of position and location time-series data to improve vision-based ML models, for example, through reconstruction of metric dimensions of objects.
- The CERC Fellow will be exposed to a range of datasets from real-life ocean surveys collected for different scientific and industrial purposes.