Case Study: Insentry achieves real-time multi-camera video analytics with SimbirSoft

Insentry scales real-time video analytics for 100+ cameras with SimbirSoft

The customer, Insentry, a leading provider of integrated smart video surveillance solutions, faced the challenge of improving the performance and synchronization of video analytics in their deployed platform. Their specific needs included achieving higher processing throughput for many concurrent streams, grouping video feeds for synchronized rendering of AI outputs, and minimizing latency for operators using wearable cameras and drones. They engaged the vendor, SimbirSoft, for software development services to enhance these AI features.

SimbirSoft delivered an integrated solution that included a synchronization mechanism for graphical annotations, techniques for low-latency rendering using motion vectors, and a distributed inference system with autoscaling. The results were a system that reliably overlays detections on many streams with minimal latency, which reduces operator load and improves reaction accuracy. The scalable design also reduces infrastructure costs for large installations of over 100 cameras, as confirmed by the customer's Head of Video Analytics.

Insentry