Realwave’s Intelligent Video Surveillance Cloud is powered by its autonomous monitoring engine, DIVE. Realwave’s cloud-native design offers the user an intuitive platform to stream, search, and share video in realtime, while DIVE works autonomously in the background. DIVE is a proprietary system framework that applies deep learning with both integrated device data and image processing to monitor continuously for realtime exceptions…
The underlying architecture of DIVE is specifically designed to be fast, robust, and scalable. Using an event-driven, non-blocking I/O model, DIVE is lightweight and efficient, easily handling data-intensive real-time processes across distributed devices. The DIVE foundational core includes:
Artificial Intelligence Deep Learning models are embedded in local Realwave hardware (e.g. Realwave Edge Exchange-REX) and within the Realwave Cloud. This combination provides efficient processing at the edge and additional analysis in the cloud. 
DIVE is designed for the Internet of Things (IoT) with a RESTful API and an extensive library of native device integrations and data parsers.
DIVE’s object classification, identification, and recognition services leverage Convolutional Neural Networks (CNN) or Deep Neural Networks (DNN) for solving advanced application problems. Sample problems include Shopper Demographics, Queue Management, Audience Sentiment, Elevated Temperature Detection, Access Control, Occupancy Counts, Time-to-Service Tracking, etc.
DIVE’s Exception Monitoring analyzes all video and operational data inputs as well as user behavior to continuously improve prediction accuracy. Exceptions trigger workflow outputs based on priority for searchable log events, messaging, reporting and long-term cloud archiving.