Pioneering Federated Real-Time Video Analytics. Real-time video applications will fundamentally change our future work and lives. This project proposes deep-learning solutions combined with federated

Description

Pioneering Federated Real-Time Video Analytics. Real-time video applications will fundamentally change our future work and lives. This project proposes deep-learning solutions combined with federated analytics, to address fundamental limitations of existing solutions bottlenecked by large video size, limited bandwidth, and data privacy. The project will generate new knowledge of combined deep learning and distributed computing to achieve real-time video delivery and intelligent video analytics with low delay, high accuracy, and protected privacy. The expected outcome includes novel algorithms and principles, and a practical system to realise them in the real world. It will provide significant benefits for all organisations and users who utilise real-time videos in daily work and life.. Scheme: Discovery Projects. Field: 4604 - Cybersecurity and Privacy. Lead: A/Prof Bing Zhou