Facial Recognition - FAQ about data protection and privacy

In this article we want to address the most frequent asked questions about the protection of the user’s data and we address the question how we ensure that the privacy of our users is protected:

1) How is the Visage Technologies App performing facial recognition?
Facial recognition is based on face descriptors. The system calculates the similarity between the input face descriptor of detected faces within the field of view of the camera and all face descriptors previously enrolled in the decentralized database. The goal is to find the face(s) from the database that are most similar to the input face.

  • no face images are stored or processed
  • database contains face descriptors and their corresponding names / IDs

2) What about the image sections that are displayed in Enrollment mode?
The image sections of the detected faces displayed in the live image are used exclusively to simplify the enrollment process and are discarded as soon as the associated face descriptors are enrolled or overwritten.

3) Is there a risk of misuse of the data by reverse engineering the face descriptors?
Since all biometric templates are exclusively mathemathical representations of users’ faces, biometric and personal information is strictly separated. This way of handling data ensures the highest level of privacy, even when dealing with extremely sensitive data.

  • Visage is developing their SDKs 100% in-house
  • Biometric templates cannot be exported or reverse-engineered to their original format

4) To what extent is the skin color, ancestry or origin of the recognized persons a factor in Face Recognition?
Since the face descriptors created are purely mathematical models of the taught-in faces, the software only takes into account the shape, texture and other special mathematical features of a face. At no time is an evaluation based on skin color, origin or other possibly typical facial characteristics of persons performed.