Sightcorp logo

AI in Advertising Analysis

Everything about AI in Advertising Analysis


What is the role of AI in Advertising Analysis?

Until recently, one of the biggest challenges in the advertising industry has been figuring out how to track the performance of various types of advertising campaigns. How do you determine the ROI of a billboard campaign, for example? This is one of the challenges that AI is helping to solve.

In the context of digital signage:

When advertisers make use of digital signage, AI can provide insightful data that can be used to analyze ad performance and to adjust advertising campaigns in real time. Now you’re probably wondering, “How does this work?” Here is a quick explanation:

  • The area surrounding a digital advertising display is monitored by a camera.
  • The camera captures images/video of people while they are viewing the display.
  • The camera is connected to a computer running AI-powered face analysis software.
  • The face analysis software analyzes the video stream in real time and provides insights in form of anonymous data.
  • This audience data can be integrated with a CMS to enable real-time content triggering based on the type of audience standing in front of the digital display.


In the context of programmatic advertising:

When advertisers run programmatic advertising campaigns, they often benefit from AI algorithms that conduct real-time audience analysis and help to ensure that relevant ads are shown to relevant users at the right place and at the right time.

Click below to learn more about AI in programmatic advertising:

AI In Programmatic Advertising

What types of insights can AI provide in the context of advertising through digital signage?

As mentioned previously, data on the viewers of your digital signage ad is collected and processed by face analysis software. The face analysis software uses AI algorithms to detect faces, gather their demographic profiles, determine their attention time, and recognize emotions based on facial expressions. In practice, this means that you can use the face analysis software to:

  • Count how many people view your ad (viewers)
  • Count how many people walked by your ad (opportunity to see)
  • Determine the age and gender of the people viewing your ad
  • Determine the times at which your ad is being viewed
  • Determine whether your ad is being viewed more than once by the same person (anonymously)
  • See how long, on average, people spend looking at your ad
  • Discover the emotional responses that people have towards your ad
  • Make real-time decisions about what ad you want to show on your display


This opens up a whole new world of possibilities to advertisers who need to be able to provide quantitative data regarding ad performance (KPIs) in order to have their advertising budgets approved and meet their targets. It also means that advertising through digital signage is no longer a guessing game – you can see who you are advertising to, how those people are responding to your ads, and which type of content captures and holds their attention for the longest. These types of insights are powerful when it comes to maximizing the impact of your advertising campaigns.

Want to learn more about Sightcorp’s Face Analysis software solutions for advertising analytics?

Click on the product links below:

Crowdsight Toolkit DeepSight SDK

Here are other articles you might find interesting:



Technical Specifications

The table below shows how FaceMatch SDK performs on the Labelled Faces in the Wild (LFW) dataset:

FPRTPRThreshold (Inverse of distance)
0.10.99900 ±0.002130.55448
0.010.99667 ±0.005370.59791
0.0010.99367 ±0.006050.62989

FPR = False Positive Rate
TPR = True Positive Rate

These results are an indication only and are based on the specific dataset Labelled Faces in the Wild. Customers can expect similar performance, with possible variations due to hardware and the availability of annotated data.