Edge Video Analytics

What is Edge Video Analytics?

Edge video analytics is a method of collecting and analyzing data from live camera streams in a continuous process at the camera or at a sensor nearby. Information being processed from these cameras is used for real-time decisions, such as in traffic control, retail store monitoring, surveillance and security, as well as consumer applications including digital assistants and AR.

What does edge video analytics do?

Edge-based analytics run on raw video data, where a device uses hardware near or inside the camera to decode video, detect objects, and receive and process events before sending its results downstream. The key difference is that the video-processing pipelines are within the device or a nearby sensor, allowing the analytics to gather more sensitive and accurate data than if the information was being encoded and sent to a central processor. In addition, it allows the analytics to control the sensor and enable optimized video input for the analytic engine.

Why is edge video analytics needed?

Because edge-based analytic devices do not rely on servers, their network bandwidth requirements are minimal. This makes them ideal for deployments where infrastructure is not available and transmitting video of high quality to a centralized server is not an option, such as remote locations with limited bandwidth. They are often used in facilities that need to monitor large perimeters, complex campus environments, or geographically dispersed open spaces.

Related HPE Solutions, Products, or Services

What are the benefits of edge video analytics?

There are several benefits to using video analytics at the edge, where the camera itself processes streaming data in real time. Because the intelligence is created entirely independently, edge video analytics is superior to using a separate dedicated server for the following reasons:


With the ability to process what is happening as it happens and discern if a relevant alert is triggered automatically, corrective action can be taken much faster than if the video needs to be sent to a server for processing.


Using compute at the camera allows for correspondingly lower provisioning (or usage) in the cloud. In addition, irrelevant or immaterial parts of the video can be filtered out (using motion-detection techniques, for example), thus dramatically reducing the bandwidth that needs to be provisioned.


With video analytics on the edge, whether the network is down has no effect on running the analytics. They would continue regardless. Such autonomy means a remote camera can detect threats and trigger actions at an isolated site regardless of connectivity.


Because of the high data volumes involved in video processing, the cost to send raw video to the cloud or a central server can be expensive. It is less expensive to process in or near the camera, so that only the metadata is sent to the cloud.

HPE and edge video analytics

With the proliferation of connected devices and a combination of hybrid infrastructure and rising consumer demands, organizations face rising complexity as they struggle to stay competitive. The need for faster computing in the data center and intelligent edge technologies with the power to execute deep analytics and AI right at the device—these are some of the real hurdles enterprise IT must overcome.

HPE has a broad and deep set of experts in Edge/IoT services who can position your developers and data scientists for a successful edge video analytics solution deployment. Our advisory services from HPE Pointnext Services and HPE GreenLake can help you get started with an in-depth assessment and carry you though to deployment and production phases with a simple, repeatable process gained from helping hundreds of customers with their data modernization initiatives. With comprehensive support, your IT operation teams can bridge their knowledge gap and sustain your analytics solutions at enterprise scale.

In addition, the HPE GreenLake edge-to-cloud platform offers an enterprise-grade AI/ML and analytics cloud-like experience so you can simplify the complexities of edge video analytics at scale, leading to better, faster decisions no matter where your data resides.