Image by Alexandre Debiève

CAWAMO TECHNOLOGY EXPLAINED

Cawamo Hybrid AI Network (CHAIN)

The AI Video Analytics Market

AI-enabled video analysis requires massive computational resources. For this reason, many companies chose cloud-based implementation of their video analytics solutions. Yet, while cloud computing offers some computing and storage advantages, they also consume huge amounts of bandwidth for upload (and download) transmission of video streaming, which is also increasing costs. A cloud-based solution also relies on 100% Internet availability and may not be able to address sensitive security events in real-time in cases of internet connectivity failures.

 

Cloud-based solutions are also not engaged with all on-site systems, such as lacking the ability to trigger actions in response to alerts. Other AI video analytic solutions are based on Edge devices.  While there are many advantages to having a local Edge unit on-premises, this approach by itself faces several challenges and limitations, including limited CPU power that impact performance and functionality, high-cost replacement of new AI-based devices such as AI cameras with sufficient processing power, and being prone to theft or vandalism, losing all video information.

Cawamo’s Unique Architecture

Our CHAIN Architecture (Cawamo Hybrid AI Network) exploits the benefits of both cloud and local edge solutions, working with an accurate synchronization mechanism to offer the best cost/performance solution in the market!

This approach brings high-quality AI-analytics that is suitable for all market segments: residential, commercial and government, small and large enterprises, and mass markets, with numerous remote locations. The Cawamo Edge unit is installed on-site and performs the first AI-based detection which works in synchronization with the Cawamo cloud neural AI network and generates high-quality, accurate alerts.

Cawamo’s Cloud is the core of our solution.  This is a powerful neural network specifically trained to address numerous security scenarios. Huge amounts of data that our research scientists are continuously collecting, augmented by our unique “crowed-source” inputs (inputs provided by our users and partners), is being stored on the cloud and is becoming part of our cloud "knowledge."  While the cloud is offering unlimited computing power, flexibility, and scalability, implementing the edge device allows us to control the site security with great bandwidth efficiency and a full flow end-to-end solution including connectivity to any on-site accessories.