continually optimizing state-of-the-art face recognition methods

Internal research and participation in international research projects involve the following subjects:

  • deep learning techniques
  • identification from photographs or video data using 3D facial images
  • face recognition with low-resolution images and video material
  • pose variation and partial occlusion in facial images
  • tracking people in crowds, even when the face is not visible

Facial image forgeries and manipulations

Recent years have seen significant advances in face forgery and manipulation methods that make it possible to create genuine-looking images and videos with modified facial identities.

Deepfakes in particular—the exchange of a face in an image or a video with the face of a different person—have become known to a wider public due to media coverage on high-profile cases.

Read about measures to prevent facial image forgeries entering biometric systems in our white paper.

Masked faces

In light of recent global developments, researchers at Cognitec have investigated the impact of face masks on face recognition accuracy, especially for unsupervised scenarios, where many faces are being matched against large databases. As expected, our current technology matches masked faces, but tests showed an overall decrease of face detection rates and an increase of false match rates, in comparison to unmasked faces.

Some face recognition companies are publishing doubtful claims about their technologies reaching the same matching accuracy for faces covered by masks. A report published by the National Institutes of Science and Technology (NIST) in July shows the performance of 89 face recognition algorithms when matching faces with digitally applied face masks with unmasked faces of the same person, with error rates between 5% and 50% . Cognitec did not participate in this test.

Since masks occlude a large area of the face, removing substantial information needed for comparison tasks, matching accuracy will evidently decrease. Therefore, unmasked faces will remain the gold standard for automated face recognition technologies. High-security applications that need to establish a person's identity, for example, border control or passport issuance, will most likely require the removal of masks. But as long as wearing masks is required or common in everyday life, recognizing masked faces is becoming an expected feature for systems that unlock phones, track and count people, measure demographics, or identify persons of interest.

Cognitec is responding to such market demands as we continuously improve our algorithms. Our research and development team constantly works on increasing the accuracy for partially occluded faces. In addition, we are using training data with masked faces, and current lab results already show significant advances for detecting and matching faces with masks. Stand by for product releases!

Conference sponsorships

Cognitec regularly sponsors international biometrics research conferences:

  • IEEE Sixth International Conference on Biometrics: Theory, Applications and Systems
  • International Conference on Biometrics

Working with universities

University of Applied Sciences (Hochschule für Technik und Wirtschaft) Dresden

Since 2010, Cognitec has been working with the department of artificial intelligence at HTW Dresden on the development of intelligent service assistance systems/humanoid robots that will react naturally and intuitively to the human user. Face recognition can lend robots a multitude of capabilities: detecting faces, estimating a person's gender and age, following faces in space and sensing moods.

University of Surrey

Cognitec is working with the University of Surrey on face models and improved face recognition for facial information extracted from multiple frames in video sequences. This involves fitting a 2D or 3D morphable face model to input face images, without restriction to the subject’s pose.

Technical University Dresden

Cognitec supports research projects at the Institute of Mathematics at the TU Dresden that study support vector machines and numerical optimization.

Collaborative projects


The project aims to gain new insights about how assistance robots can be employed in the care of the elderly. Possible tasks carried out by such robots include night watch, helping with executing cognitive and motoric exercises, and providing support for therapists and carers. The two-year project started in April 2017 and is funded by the Free State of Saxony and by the European Union through EFRE (Europäischer Fonds für regionale Entwicklung; English: European Fund for Regional Development) with more than 800.000 Euro. Cognitec intends to contribute to the project with improvements of the existing face analysis technologies, both for person identification and for the estimation of person characteristics such as age and gender, taking into account the specific operating conditions of the assistance robots.