A British tech start-up, Calipsa, has launched its video surveillance technology built on Deep Learning models. The company says that it’s developed algorithms that can process and analyse hours of video feeds to provide alerts and detailed reports for applications in real-time video monitoring, including traffic enforcement, public spaces, transport hubs, road accidents and public disorder.
The Calipsa engine uses a feedback loop to evolve and improve. Human operators can ‘teach’ the artificial intelligence using a point and click interface, which automates repetitive parts of their jobs. Designed to work with any camera or video source, the technology can be deployed via the cloud or on-premise, with no retrofit required. According to the company founders, Mohammad Rashid Khan and Boris Ploix, the product is adaptable to all weather and lighting conditions, with 95 per cent accuracy.
Mohammad Rashid Khan, co-founder and CEO at Calipsa, says: “There are around 250 million video surveillance cameras in operation worldwide today, capturing 1.6 trillion hours of video annually. Yet despite the fact that we live in the 21st century, the majority of video surveillance carried out by the police, military, transport operators and security companies, is still done by humans. This doesn’t make sense, as it’s very time consuming, expensive and inefficient. You can imagine that manually viewing huge quantities of video data for hours and hours can lead to fatigue, loss of attention, and most important, errors – at a time when video surveillance has never been more critical in our society.”
Calipsa was on show in June at IFSEC 2017. Typical scenarios, the firm says, include monitoring traffic flows and congestion, traffic surveys and incident detection, and mass transit (airports, bus and train stations). The company is looking to recruit partners covering hardware, security and integration.
Calipsa, which has expertise from universities including Cambridge, Imperial College London and UCL, has raised a seven figure seed funding round from venture capitalists, including Localglobe, Horizons Ventures and Entrepreneur First. The company adds that it’s already working with customers in the UK, France and India and is now seeking strategic partners to take its technology to market.