Privacy First People Counting Software

People Counting

All in one logo solution icon blue
DeepSight Toolkit
  • Get started quickly and easily
  • No need for hard coding
  • Ready Application
  • Real-time analysis
  • People Counting Software
  • Privacy by default: face blurring
anroid linux windows logo
SDK icon blue
DeepSight SDK
  • Software Development Kit
  • Build your own solution
  • Written in C++ with additional wrappers (e.g. C#, Python) 
  • Works offline (one-time internet connection)
  • No GPU required
linux android windows dcoker icon
Technology icon blue

Use Cases

digital signage totem face analysis crowd mall
DOOH Audience Reach

Using people counting technology, media owners can accurately determine the audience reach of their digital displays regardless of their location and charge for ad space based on impression data. The anonymized data supplied by our product DeepSight, provides data transparency and works in real-time to achieve the most accurate results.

people detection station people walking technology detection
DOOH Ad Performance

Media buyers can use impression data insights, provided by our people counting technology, to identify prime locations for the placement of their ads before purchasing ad space in order to maximize audience reach and engagement. POP data is then also accessible post ad play to measure campaign performance and success. 


mall shopping woman shop
In-Store Analytics

Our traffic counting software provides retailers with helpful insights for establishing store peak hours, calculating conversion rates when compared to sales data and discovering patterns in the way visitors navigate through the store. This information can be used to optimize store layout and product placement as well as staff allocation.

Our Partners

Intuiface logo
broadsign logo
grassfish logo blue
simple nuc logo
telelogos lofo
IOT Alliance Logo

Accurate and Anonymous


crowd logo black
What is People Counting Software?​

People counting technology is used to measure the number of people visiting a location, area, or space. Our software uses body detection to anonymously detect and count passersby. This method is completely privacy compliant thanks to our automatic face blur feature incorporated in DeepSight.

How it works: 

1. A camera placed at head height, pointed towards the flow of traffic is used to detect and count people passing by. 

2. Human bodies detected by the body detector are used to count people passing by the camera with high accuracy. The body detector can count people even if their face is not fully visible which leads to a larger proportion of the audience being captured using body detection.

3. The face detector runs in parallel with people counting to quantify the number of people actually viewing the screen.

4.The resulting data (nr. of OTS, nr. of impressions, nr. of viewers and their respective dwell and attention times) is then aggregated and displayed in a Cloud-based dashboard to be used by the media owner and the advertiser.

people detection mall people walking technology detection
family shopping mall shop
Why People Counting Software?

Through the use of Sightcorp’s People Counting Technology, DOOH network owners can automatically measure the number of people passing by their screens and use this traffic data to increase their CPM. Retailers are able to tally people entering their stores and compare this data to sales figures to understand visitor conversion. Forget about outdated manual counting, automate everything, and optimize your business in a simple, fast, and convenient way.

What you can expect:

1. Fully anonymous and privacy compliant traffic counting system

2. Face blur applied automatically

3. Highly accurate real-time counts

4. Robust technology that operates effectively in crowded areas

5. Easy integration with other applications


Digital Signage Icon black
People Counting Software for Digital Signage, DOOH & OOH Media

Sightcorp’s audience analytics software combines two counting technologies into one product to guarantee the most precise measure of the audience at any given location. This enables you to detect people at variable distances from your digital screen to generate the most accurate and inclusive counts possible.

Body Detection: Detects people within a larger distance range (up to 10 meters). It is robust and works perfectly for capturing OTS data in highly crowded areas where there is high (face) occlusion. Measure all potential viewers within the screen viewability area!

Face Detection: Detects people within a shorter distance range (up to 6 meters) and is ideal for counting viewers and impressions and attributing these to specific screens, as well as measuring how long they spent viewing the ad content.

people detection station people walking technology detection

People Counting with Body Detection FAQ's

Our DeepSight SDK performs people counting on the basis of our body detection model. The people counting feature comes as an addition to the existing audience analytics features available for any user of the SDK. The software functions through the use of cameras that are placed in areas with high traffic of people. DeepSight is able to detect individuals that pass through the viewability area of the digital display. Screens of different sizes have a different viewability index therefore the people counting technology needs to be adaptable to this changing scenario. Our counting technology can cater to many different screen placements, both indoor and outdoor. Thanks to the synergy of face detection with body detection, we are able to count more impressions than ever before. Body detection works with a larger object (bodies) compared to face detection so it’s suitable for counting people at distance, while face detection is most commonly used for counting people that pass closer to the screen and converting them into a number of impressions and viewers.

Camera based audience measurement is the only technology that provides a representable picture of all people that have the opportunity to see a display. Unlike mobile or wifi tracking, video analytics does not rely on tracking devices carried by individuals, instead, it analyzes individuals in an anonymous way using cameras attached to screens that detect if passersby have viewed the ad content and for how long.

Our people counting capabilities are offered as a part of DeepSight. This product line is distributed in two different formats, the SDK and a Toolkit. 

The SDK is a development kit primarily designed for developers or tech companies looking to integrate DeepSight into their own systems. Users of the SDK can create their own white label applications according to the customer’s requirements and needs. 

The Toolkit on the other hand, is a ready-made plug & play application that can be run on Windows, Linux and Android. It doesn’t require any coding expertise, all you need to do is install it on your machine and connect it to a USB or IP camera. Afterwards, the system is immediately ready to start collecting audience data.

Both the SDK and Toolkit offer all-in-one face detection and body detection capabilities that can be used for counting people with the opportunity to see content, actual impressions for the content, and true viewers of the content.

We provide a two-week free trial of our People Counting software. With this free trial, you will be able to access not only this, but a range of other features including face mask detection, demographic analysis, mood detection etc. Following this trial, we will reach out to discuss the next steps. We offer 1-year licences during which time you can carry out all the analysis you want.

Traffic data, collected by the anonymous people counting software can be displayed on Sightcorp’s data visualization tool, DeepSight Data Studio. This Data Studio is a fully managed, cloud-based dashboard that offers real-time view only reports. Alternatively, there are also ready plugins available for integration with 3rd party dashboards.

Yes, the people counting system is very robust and has been designed to detect people with high precision. Using body detection technology, it is easier to count people in large crowds, even when they are walking in different directions or facing away from the camera itself. The large detection area (bodies) ensures that even if partial occlusions occur, people will still be detected and included in the traffic data. The people counter can also detect individuals up to large range distances (10 meters) ensuring that the traffic data is the most inclusive and accurate.  

When compared to other counting technologies, using 3D cameras for instance, body counting is the one that takes the edge on effectiveness and scalability. This is mainly because it does not require any additional hardware next to the standard audience measurement kit. End-customers can use the same cameras for counting people, analyzing their demographic profiles and tracking viewing behaviour. Most other technologies require specific sensors and cameras which cannot be used for multiple purposes. That is why we believe that audience measurement using 2D video cameras is the most versatile solution out there that can satisfy the requirements of many types of users.

However, the accuracy of such a solution can be influenced by external factors which need to be controlled before running an analysis. The accuracy of any computer vision application generally depends on 3 factors; positioning and viewing angle of the camera, processing power of the device running the software and lightning conditions within the environment. With proper camera placement and a suitable hardware device, the accuracy of the body detector can reach its maximum values. If any of these components is not within the recommended threshold, the quality of the data output can suffer. Therefore before starting any POC or a commercial deployment, it is essential to evaluate the scenario in which the camera and software will be placed in order to obtain optimal results and carry out a successful project. Our team offers extensive documentation and guidance throughout this whole process to make sure you succeed on the first try. We are only successful if you are successful!

Our solutions are completely anonymized and privacy proof as the software counts people through video analytics in real time, meaning it doesn't record and store any actual video footage. Moreover, the live video footage is also combined with a Face Blur technology that hides faces of individuals detected by the camera. This feature can be activated by default to make sure every individual’s identity is protected and that there is no possibility for a breach. Using DeepSight as a people counting tool guarantees compliance with all privacy and anonymity standards within the industry.

People counting software can be used in any industry where visitor data, in particular traffic counting data is useful to analyse. It can be applied in the Retail, DOOH, Corporate and Education sectors & more. The technology is perfect for helping businesses understand the number of visitors that they have on a daily, weekly or monthly basis and enables them to carry out effective market research, monitor occupancy levels and much more. Sightcorp’s counting software is easy to integrate, scalable and affordable, making it the perfect solution for any business. The SDK also makes it possible for developers to build the counting software into existing technologies.  

There is no maximum number of visitors that the technology can count. The counting software will continue counting each visitor until a limit has been decided or set by the operator.  

Start Optimizing Your Business Now

Analyze your traffic, measure performance of marketing strategies, increase conversion rates & get the most accurate and actionable data now