Historically, the AI in video analytics is usually single-threaded and purpose-built to perform a specific function (e.g.,facial recognition, object detection, license plate recognition). This limits an organisation’s ability to have a more complete understanding of a scene from video surveillance. This also means that the metadata is limited to the single use case, disconnected from the other algorithms, and unable to holistically make sense of a scene.

“Video analytics is still in its infancy. Today’s video analytics provide limited insights which can translate to poor and missed alarms,” according to Dieter Joecker, Oosto chief technology officer. “Wandering through ISC East and talking to enterprise customers and the partners who serve them, it's clear that the world of physical security needs to move out of the dark ages and help organisations prevent incidents of violence and vandalism, or to minimize the damage immediately after.”

Joecker added, “Tomorrow, commercial enterprises will harness the power of contextual scene analysis – multithreaded AI – to deliver real-time contextual analytics and precision alerts based on a more holistic understanding of live video surveillance. For example, when someone falls down in a store, this could mean any number of things: Were they actually bending over to tie their shoes? Were they pushed by another person? Was it a staged ‘slip & fall’ to generate a fraudulent claim? Is the person having a legitimate medical emergency? How an organisation responds in real-time is based on the timeliness and quality of the alert.”

“Leading Vision AI companies are starting to exploit the power of semantic segmentation and edge computing to deliver real-time contextual analytics and alerts to achieve a more holistic understanding of video footage,” added Professor Marios Savvides, Director of Carnegie Mellon University’s CyLab Biometrics Center and Oosto Chief AI Scientist. “We can also start to capture more metadata to gain greater context about what’s occurring in real-time and ensure that the right types of alerts are sent to the right personnel. Instead of relying on surveillance professionals to monitor video footage 24x7, security teams can use their skills to respond to very precise alerts. It’s not just about adding more algorithms. Leading companies are going to use the power of semantic scene understanding and edge computing to deliver real-time contextual analytics and alerts based on a more granular and complete understanding of a scene.”