Developing and Commercialising Intelligent Video Analytics Solutions for Public Safety
Lead Participant:
VISION SEMANTICS LIMITED
Abstract
Effective automatic intelligent video analysis of large scale data from public spaces is an important tool in the fight against crime and for safeguarding public safety. Yet, it is challenging to extract critical information from very large scale unstructured surveillance videos with a very high ratio of mundane data subject to severe visual ambiguity and clutters. For example, the Scotland Yard viewed 6,000hrs of CCTV footage in order to search the London 7/7 bombings terrorists. Automatic video analytics capable of rapid processing (faster than real-time) of very large surveillance data is critically required in a global market including China and the UK. Whilst there are video analytics systems in the market, the challenge for domain-transferable video abnormal event detection with people and vehicle re-identification (search) in multi-source open data is still unsolved. This project will develop innovative and scalable systems for abnormal event detection with joint person and vehicle search in unconstrained public spaces. The system will achieve 80% detection rate of 5 abnormal events and 3x more accurate than human experts and 100x faster than real-time for searching people/vehicles in large scale data from distributed urban spaces. Our goal is to increase the analyst productivity by a factor of 5.
Lead Participant | Project Cost | Grant Offer |
---|---|---|
VISION SEMANTICS LIMITED | £499,750 | £ 348,045 |
  | ||
Participant |
||
QUEEN MARY UNIVERSITY OF LONDON | ||
QUEEN MARY UNIVERSITY OF LONDON | £149,969 | £ 149,969 |
People |
ORCID iD |
Sean Gong (Project Manager) |