Video Analytics 101: Understanding the Video Intelligence Ecosystem

Choosing the Best Solution for your Business

Choosing a video analytics solution is a significant decision. With so many options available, you want to make sure you are investing in the right choice for your needs. While we already walked you through the ins and outs of video analytics deployment, demystifying “cloud” and “on-prem” decisions, today, we want to take a step back and outline the bigger picture. Let’s dig into the video intelligence ecosystem so you can understand the options available and make informed choices for your organization.

The Big Picture: Video Technology Ecosystem

The video analytics ecosystem lives within the greater sphere of video surveillance technology, which consists of all the video surveillance components, such as cameras, Video Management System (VMS), underlying hardware, and even the systems integrators that drive the sale and optimal deployment of the video surveillance infrastructure. Because video surveillance is such a robust industry, it is easy to get lost in all these parts which are complete ecosystems in and of themselves.

The Video Analytics Landscape

Today, even within the video analytics market, there are different types of intelligent solutions available which fall into 3 main categories:

• Embedded, On-Camera Analytics
• Video Management System (VMS) Analytics
• Server-Based Analytics Platforms

We’re going to lay out the advantages and disadvantages of each technology so you can move forward with selecting the video analytics solution best suited to your environment and specific business requirements.

1. Embedded Analytics

Embedded analytics are analytics that are conducted on the camera or another hardware device, where there is no need to transmit data to a centralized server for processing and analysis. Edge and point solutions fall under this category. Point solutions are specialized analytics solutions that focus on specific tasks like License Plate Recognition (LPR), motion detection, people counting, face mask detection and more.

Advantages:

• • Easy to get up and running.
  • Effective for focused use cases.
  • Low latency because video data is processed right on the camera and not transferred back and forth between the camera and the cloud.
  • Efficient bandwidth usage because data is not constantly sent back and forth, saving on bandwidth costs as well.

Disadvantages:

• • Difficulty tracking and identifying objects in a complex scene – like traffic intersections, college campuses, airport arrival and departures, and more.
  • Restricted face recognition (FR) and license plate recognition (LPR) features, which cannot be performed completely on-camera, because the faces or license plates must be matched against faces in a database on a server or in the cloud.
  • Limited processing power due to the spatial constraints of cameras, i.e., the number of processors or System on Chip (SoCs) that can be accommodated within the casing are limited.

2. Video Management System (VMS) Analytics

The VMS powers video collection, recording, interface, and management. VMS analytics perform analysis on the video collected from cameras and other sources connected to the system.

Advantages:

• • Connected to multiple cameras on a network – enabling data retrieval from all cameras in an environment.
  • Typically enables event management, allowing users to set up rule-based alerts that trigger alarms or notifications for specific events. This is a critical capability for real-time situational awareness.
  • Eliminates the need for edge or platform analytics since the capabilities are available right on the VMS.

Disadvantages:

• • Analytics are often basic and detection-based with limited accuracy, and they are not vendor agnostic, i.e., they operate exclusively with the VMS providers’ platform, This is challenging for hybrid environments comprised of multi-vendor cameras and VMS deployments.
  • Lack of compatibility makes this solution inflexible and difficult to scale. For instance, integrating VMS analytics with other systems can be challenging due to patents or intellectual property.
  • Resource intensive – VMS analytics typically require substantial processing power and storage capacity.

3. Platform Analytics

Platform analytics are software-based and hosted on a server. Also referred to as server-based analytics, platform solutions are designed to address sophisticated and ever-evolving end user needs, with extensible, flexible, and scalable technology.

Advantages:

• • Driven by powerful GPUs that can execute advanced analytics compared to on-camera or VMS analytics.
  • Enables features like face recognition (FR), license plate recognition (LPR), and vehicle make and model recognition (VMMR).
  • Enables business intelligence by aggregating all metadata from each camera and VMS into user-friendly, visual dashboards.
  • Integrates with existing camera and VMS investments to create a truly vendor agnostic solution.

Disadvantages:

• • Systems can be expensive with high upfront costs for hardware and maintenance.
  • More GPUs are needed for more concurrent video streams as each GPU can only process so much video at a given time.
  • Limited storage capacity requires users to either reduce the amount of video footage that is stored or add additional server storage.

Video Analytics for Your Business

Video analytics is a powerful extension of video intelligence ecosystem, offering several solution categories, each with unique benefits and challenges. Platform analytics is the most powerful, flexible, and scalable, but can be financially out of reach or unnecessary for many users. Edge analytics are affordable and easy to implement but solve limited problems. VMS analytics take the middle ground with “good enough” analytics conveniently built into an existing investment, but limit flexibility because of specific VMS dependence.

Ultimately, when it comes to video analytics, there is no right or wrong answer, but rather what is right for you. You need to select the solution that best meets your needs, so you are empowered to translate video into impact for your business, organization, or community.

Discover how the strong commercial cooperation and deep technological integration between BriefCam and LenelS2 enables LenelS2 OnGuard users to unlock additional video analytics value. Talk to us about getting started and enhancing your LenelS2 OnGuard implementation with streamlined alarm monitoring and rule-based automations.