Securing Biometrics for Finance

Computer vision and security R&D to help banks leverage facial authentication in mobile apps.


A leading banking security provider in the EMEA region wanted to expand its product offering with face recognition biometrics, for use cases such as mobile banking app login, KYC (know your customer), and transaction authorization. The challenge was not only to measure face similarity against photos in our database but also to make sure that the user is a legitimate, living person (the task known as “liveness detection”).


We created a pilot application consisting of an AI-enabled authentication backend and iOS and Android user-facing apps. We also conducted a thorough security research and wrote a white paper detailing all major attack vectors and the respective countermeasures, including their technical feasibility. Both the pilot application and the white paper will help the company develop a solid business case and test customer demand in real life, before investing in a full-scale product.

Looking for something similar?



  • Deep neural networks
  • Attack vectors examined: 2D presentation attacks (paper, photo, video), 3D presentation attacks (various types of masks and 3D-printed heads), face alterations, attacks physically exploiting the victim (sleeping, unconscious, hostage, or even dead), twins, attacking neural networks using image noise, data injection, brute force attacks
  • Countermeasures examined: depth detection, skin texture detection, heartbeat detection, iris detection & eye movement, face micro-motions, deepfake detection, behavioral tests

Read More Case Studies