Conventional or traditional AI can be defined by a dependence on supervised learning, where the AI model is trained on labelled information or data. Today’s Generative AI models are capable of leveraging both supervised and unsupervised learning; however, they excel at unsupervised learning – meaning the model excels in identifying patterns and structures in data without the need for labelled information.
This ability to make sense of unlabelled information represents a critical inflection point – we are moving from the need to train Video Analytic Engines to identify specific instances, actions, or objects to a significantly more flexible query-based solution.
In an increasingly connected world, our challenge lies not in deploying new sensors across different sectors and environments; rather it stems from the ability to make sense of and act on huge amounts of information that is readily available to us.
Contemporary video analytic engines do a good job of alerting us to pre-defined scenarios and we anticipate that this will continue to be an important foundational aspect of surveillance. AGIL® Vision represents a powerful complementary solution that today, addresses the need to be able to deploy searches during unanticipated situations.
By way of example, our traditional systems can quickly identify a known Person of Interest from a watchlist; but when faced with the challenge of finding a lost child in a crowded mall, human intervention becomes necessary.
AGIL Vision enables security personnel to use simple natural language queries to search for the child, not by facial recognition but by generic fields such as clothing colour, brand apparel, hair length, or any combination of these permutations.
The power of these flexible search queries can extend to objects (in the case of searching for unattended bags, lost and found items, or even weapons) or even actions (such as the act of falling, fighting, running, or climbing), creating a robust tool that enables more rapid investigation.
While this capability translates well towards larger Command Centre type of environments, we have productised AGIL Vision into a small but powerful form factor that will facilitate the deployment of this capability to a range of sectors and scenarios with requirements both large and small.
We envision our solution to be a revolutionary tool that can be deployed across use cases including security & surveillance management, workplace safety enforcement, healthcare and patient management, warehousing and logistics, and many more.