The evolution of video surveillance: 2000s-2020s

The video surveillance market is crowded, complicated, and confusing. There are countless players and products that all appear to be similar. Now, AI has added another layer of complexity to the video surveillance market, causing organizations to spend more time understanding the role that AI will play in their next camera system.

How did we get here with more options than ever and what’s at stake for IT and security teams when choosing their next video surveillance system? Let’s start by unpacking how we got to this point.

2000s: The IP camera era

Prior to the 2000s, analog cameras and VHS tapes were the primary methods of recording. In the early 2000s, the shift to digital cameras began. Digital Video Recorders (DVRs) replaced VHS tapes, allowing for improved recording, resolution, storage, and flexibility.

The introduction of Internet Protocol (IP) cameras allowed video data to be transmitted over a network. This led to the development of the Network Video Recorder (NVR), enabling remote monitoring and more efficient storage solutions.

Despite technological advancements, the network video recorder still presented a number of challenges. While other software applications had transitioned to the cloud, the IP Cameras lagged behind and users could only access their footage on site or remotely via a Virtual Private Network (VPN). Solutions of this kind also put the onus of system management on IT teams.

2010s: The cloud era

In the 2010s, cloud-based surveillance emerged, allowing businesses to store footage remotely, providing accessibility from anywhere with an internet connection. Building on this technology, two new types of solution hit the market, commonly referred to as the Cloud Camera and Cloud NVR.

We are at an inflection point where artificial intelligence is becoming increasingly powerful and useful, transforming the Video Surveillance system. These rapid improvements have led to the emergence of a new architecture of Video Surveillance products — known as Hybrid AI — to keep up and leverage this wave of technology transformation. Hybrid AI leverages an appliance connected to the network called an Intelligent Video Recorder (IVR). These are similar to NVRs, but have GPUs or Tensor Processing Units (TPUs) built in that allow for AI processing at the edge, as well as in the cloud.

As AI Camera System models are entering the market, purpose built for AI, existing Cloud Camera and Cloud NVR vendors are attempting to retrofit their offerings with AI. On the surface, the products may appear similar, but it’s important to understand how to cut through the marketing and select a system that meets your business needs.

Cloud cameras and cloud camera systems

Vendors like Verkada and Rhombus brought cloud cameras and cloud camera systems with a cloud-managed architecture that combined recording, storage, and processing on the cameras themselves, eliminating the need for DVRs and NVRs. This type of cloud-managed architecture provided benefits over the traditional DVR vs. NVR solutions, offering remote access, automatic software updates, and a modern user interface. However, these cloud camera solutions required customers to rip and replace their existing cameras with proprietary cameras, as well as large-scale upgrades to their IT infrastructure.

Cloud NVR

Vendors like Eagle Eye led the Cloud NVR category, with an architecture that allowed buyers to keep their existing cameras while still gaining the benefits of a single pane of glass. These solutions operate on the edge by connecting an NVR to the internet to store and process video data locally and in the cloud.

The approach provided customers with the security, bandwidth, and cost benefits of an on-prem NVR, and the flexibility and convenience of the cloud. An NVR is still required, but these solutions seamlessly enable a single pane of glass for all cameras across multiple locations. This mitigates upfront costs associated with the full replacement of cameras.

Today, cloud capabilities are the expectation.

2020s: The AI era

Every piece of software will get reinvented with AI, and camera systems are no exception. The AI Camera System is the future of camera systems.

GPU compute power has drastically increased in the years since - doubling every year. The ability to train large and complex AI models has followed in lockstep. This, along with the emergence of Foundation Models, has led AI capabilities to improve at rates faster than ever before.

Hybrid AI

These rapid improvements have led to the emergence of a new architecture of video surveillance products — known as Hybrid AI. This new architecture leverages an appliance connected to the network called an Intelligent Video Recorder (IVR). These are similar to NVRs, but have GPUs or Tensor Processing Units (TPUs) built in that allow for AI processing at the edge, as well as in the cloud.

A new set of vendors are leveraging Hybrid AI architecture and bringing AI Camera Systems to market. These products offer all the benefits of the cloud as a baseline, but also allows for AI applications that are performant, usable, interoperable, and scalable to accommodate the rapid developments in AI technology.

As AI Camera System models are entering the market, purpose built for AI, existing cloud camera and cloud NVR vendors are attempting to retrofit their offerings with AI. On the surface, the products may appear similar, but it’s important to understand how to cut through the marketing and select a system that meets your business needs.

{{WEBINAR_BANNER}}

Learn more about AI Camera Systems and get AirPods for free.
Join the webinar