{"links":{"self":"http://dataportal.arc.gov.au/NCGP/API/grants/DE260100509"},"data":{"type":"grant-details","id":"DE260100509","attributes":{"code":"DE260100509","administering-organisation":"University of Technology Sydney","announcement-administering-organisation":"University of Technology Sydney","scheme-name":"Discovery Early Career Researcher Award","grant-status":"Active","funding-commencement-year":2026,"years-funded":3,"project-start-date":"2026-11-09","anticipated-end-date":"2029-11-08","grant-summary":"Agile Vision: Adapting Vision Foundation Models in Real-World Contexts. This project aims to develop an adaptive framework for vision foundation models to operate effectively in data-scarce, resource-constrained, and evolving environments. It expects to generate new knowledge in adaptive artificial intelligence by integrating data-efficient learning, parameter-efficient fine-tuning, and latency-efficient inference techniques. The expected outcomes include enhanced deployment of vision models in precision agriculture, environmental monitoring, and industrial automation. This will provide substantial benefits, including increased AI efficiency in critical Australian industries, improved decision-making in dynamic environments, and strengthened sovereign AI capabilities aligned with the National AI Strategy. ","funding-current":496523.00,"funding-at-announcement":492678,"investigators-current":[{"title":"Dr","firstName":"Yanbin","familyName":"Liu","roleName":"Discovery Early Career Researcher Award","roleCode":"DECRA","isFellowship":true,"orcidIdentifier":"0000-0003-4724-8065 "}],"investigators-at-announcement":[{"title":"Dr","firstName":"Yanbin","familyName":"Liu","roleName":"Discovery Early Career Researcher Award","roleCode":"DECRA","isFellowship":true,"orcidIdentifier":"0000-0003-4724-8065 "}],"organisations-current":[{"organisationName":"University of Technology Sydney","roleName":"Administering Organisation","state":"NSW"}],"organisations-at-announcement":[{"organisationName":"University of Technology Sydney","roleName":"Administering Organisation","state":"NSW"}],"field-of-research":[{"isPrimary":false,"code":"460304","name":"Computer Vision","type":"FOR20"},{"isPrimary":true,"code":"4605","name":"Data Management and Data Science","type":"FOR20"},{"isPrimary":false,"code":"460501","name":"Data Engineering and Data Science","type":"FOR20"},{"isPrimary":false,"code":"460502","name":"Data Mining and Knowledge Discovery","type":"FOR20"}],"socio-economic-objective":[{"code":"220302","name":"Electronic Information Storage and Retrieval Services","type":"SEO20"},{"code":"220403","name":"Artificial Intelligence","type":"SEO20"}],"international-collaboration":["Germany","Japan","Korea, Republic of (South)","New Zealand"],"lief-register":[],"achievement-summary":null,"national-interest-test-statement":"Vision Foundation Models (VFMs) are cutting-edge AI models that enable machines to understand and interpret visual data similar to human perception. Trained on large image datasets, VFMs can improve productivity in agriculture, environmental monitoring, and manufacturing. However, deploying these models in real-world settings is challenging due to limited data, high computing demands, and changing conditions. This project will adapt VFMs to work reliably in places with limited resources, such as farms with poor data access or remote areas with little infrastructure for real-time analysis. In agriculture, VFMs will help farmers monitor crop health, detect pests early, and predict yields, reducing costs and improving food security. In environmental monitoring, VFMs will analyse satellite images to track land use changes, identify environmental risks and monitor biodiversity by detecting shifts in habitats and ecosystems. The project will also enhance manufacturing by identifying defects in production lines, improving efficiency and reducing waste. By developing AI models suited to Australia’s conditions, this research will make AI solutions more affordable and reduce reliance on foreign technologies. The project will collaborate with government, industry, and small businesses, sharing findings through open-access publications, software tools, and outreach activities such as workshops and demonstrations to ensure long-term benefits for Australia."}}}