These days, digital signs are a frequent presence in stores, shopping malls, airports, train stations, and many other public venues. As this medium is becoming increasingly popular for advertising, network operators are incorporating anonymous video analytics to more effectively measure the audiences viewing them. But what is anonymous video analytics (AVA), and how does it measure the audience that is viewing a particular message on a digital sign?
What is Anonymous Video Analytics (AVA)?
Anonymous video analytics is a computer vision technology able to gather metrics about digital signage audience engagement. AVA typically involves camera mounted on a digital signage unit linked to a computer running face detection and analysis software to capture data on the individuals passing by and viewing the signs.
It is important to note that the data captured by AVA is completely anonymous. The algorithms in the software are designed for detection and analysis only, not for identification or verification of individuals. No images are stored and no personally identifiable information is collected. The only data that is stored is of an anonymous, aggregate, statistical nature – in the form of code, not images.
Many people still find this confusing and often refer to AVA as facial recognition. However, that’s not correct. The right term is face analysis.
Let’s take a look at the difference between the two.
The difference between face analysis and face recognition
Both face analysis and face recognition software use face detection to locate faces within images. High-performance face detection is an important starting point, as it plays a vital role in the accuracy of the technology that it is built on top of it. After face detection takes place, however, there are big differences between how face analysis and face recognition technology work.
Face analysis algorithms anonymously analyze facial images to determine characteristics such as age, gender, emotion, head pose, and attention time. Knowledge of these characteristics is useful to retailers, advertisers, and digital network owners, as it enables them to make informed, data-driven decisions that promote audience engagement.
In contrast to face analysis, face recognition aims to determine or confirm the identity of an individual. In other words, face recognition is used to answer the question: “Who is this person?” or “Is this person who they say they are?” Face recognition is useful in situations such as security, access control and payment verification.
Why use anonymous video analytics?
If you’re running an advertising campaign using digital signage, you’ll most likely want to know how many people are seeing your ad. You’ll also want to know how effective your campaign is and whether there is anything you can do to optimize your ad spend.
By using anonymous video analytics, you can collect important viewership metrics for your digital signage content, including the following:
- Views (how many people looked at the ads)
- Demographics (gender and age bracket of the people who viewed the ads)
- Dwell time (how long each person looked at the ads)
- Time of day (when people viewed the ads)
With AVA-enabled digital signage, you can better understand who your audience is and at what time of the day your promotional content is mostly viewed. This is quite a major leap from traditional audience measurement techniques, where teams of people physically counted the numbers of passersby, and other viewership metrics were largely unavailable or unreliable.
The data that you obtain by using AVA can help you to determine what type of content to display in which locations and at which times of day to improve the customer experience and to ensure maximum ROI.
Interested in implementing anonymous video analytics software in your own business?
At Sightcorp, we offer various face analysis products, including ready-to-use Toolkits as well as SDKs that you can use to integrate anonymous video analytics into your own systems.
Other interesting articles: