Deep Learning-Powered Video Analytics Technology
Agent Vi’s Deep Learning technology boasts a unique object classification methodology that actively and continuously learns how to accurately classify objects, so that it can distinguish between a variety of object types, i.e. persons, cars, motorcycles, bicycles, buses, trucks and static object types.
Agent Vi’s powerful technology trains continuously on millions of images extracted from 15 years-worth of real-life surveillance video footage, resulting in unparalleled detection accuracy and an extremely low false alarm rate.
How Does innoVi Work to Make Any Camera “Smart”?
Agent Vi’s patented architecture distributes Video Analytics processing between the innoVi Edge appliance and innoVi Core (a server located either in the cloud or on-premise).
Cameras are connected to the compact appliance pre-loaded with Agent Vi’s powerful video analytics – innoVi Edge. The appliance processes the video locally and communicates with the innoVi Core for alarm response, health monitoring, model updates, commissioning and central management. Automatic SaaS updates / immediate on-premise update notifications ensure that innoVi Edge and innoVi Core are always up-to-date.