When a face is detected on the image, inSight marks it with a rectangle, where you can see gender, age and mood estimates. This works for one or many faces on the same image.

On the right panel, you can see how many faces are being detected on the image, and the actual average for age, gender and each mood state (angry, neutral, happy, sad, scared and surprised).


That's why we don't record any images or video feeds. If you want to know more, read about our privacy concerns below.


We consider privacy a very important issue, and believe in being transparent about it. So, let's be clear: when you allow us access to your camera, your video feed will be transmitted to and processed on CyberLabs servers.

However, no images will be saved or recorded. Only facial features (just numbers), age, gender and mood estimates will be saved in order to demonstrate InSight's functionalities and help CyberLabs improve its solutions.


How does the face recognition technology work?

  1. Video stream: your camera (webcam or CCTV) feed is transferred to our servers

  2. Face detection: The program searches for any face in each video frame

  3. Facial feature extraction: Each recognized face is converted to a vector using deep learning techniques. Moreover, estimations of gender, age and emotion are associated to each processed face

  4. Database: Vector data are recorded in our database

  5. Final processing: Vectors are clustered together and compared for identification

  6. Visualization: Our API outputs can show recurrence rate, public flow and segmentation in the form of graphs and indicators

How can I get better insights from the data??

We at CyberLabs understand that you don't have the time to read through all your data. We want to deliver only relevant insightful information. We study in depth sectoral needs and cross our data with other indicators to deliver real useful insights.

Is it suitable for my company?

We are working hard to better serve out of home media and retail industries. We already have working applications that can give insightful, relevant information for players in these industries. Our goal is to bring physical world data that is already used in digital media and deliver new insights.We have been asked to develop solutions for other markets and industries. Maybe yours will be the next!

Do you save any images?

CyberLabs cares about your privacy and takes this point very seriously. Our software runs in a trusted environment. All images are processed and transformed into numbers immediately and without any images ever being saved. NB: It is impossible to reverse the numbers back into an images!

Is there trial version?

We are working hard in bringing you a free product in the following weeks. Soon there will be a panel where you can test our applications.