Earn money by answering our surveys

Register now!
Marketing, customer satisfaction and loyalty
Satisfied customers will follow you everywhere

CaixaBank: ATMs integrating facial recognition

Facial recognition is becoming a part of our daily lives increasingly, especially with phones that are unlocked by detecting the characteristics of our faces through artificial intelligence programs. In this field, CaixaBank is innovating by integrating facial recognition into its ATMs.


  1. Withdrawing money without a PIN code
  2. Facial recognition, biometric technology

Withdrawing money without a PIN code

The purpose of using facial recognition at CaixaBank is to replace the PIN codes (Personal Identification Numbers) that everyone must enter into the ATM to withdraw money. It is always possible to choose the means of withdrawal and customers do not have to use facial recognition. They can also enter their PIN code.

Before the customer can make a cash withdrawal without using their PIN code, they must register and have themselves photographed so that their facial features can be stored and reused by the ATM.

There are already initiatives integrating facial recognition around the world, whether for fraud detection (Ping An, China) or in addition to the PIN code (Macau, China). However, the Spanish bank would be the first to use facial recognition as the only means of control between the bank and its customer when withdrawing money.

The first facial recognition ATMs are already active in 20 withdrawal points and are expected to spread throughout Spain during 2019 gradually, CaixaBank being the most extensive banking network in Spain.

Facial recognition, biometric technology

CaixaBank was already the first bank to launch facial recognition to allow mobile payments on specific mobile devices. This initiative follows the same pattern as it is still being developed using biometric technology.

Facial biometrics

A sensor, whether 2D or 3D, records facial features and transforms them into digital data using an algorithm and many comparisons with an extensive database.

The aim is to improve the user and consumer experience while continuing to guarantee the security of the customer and his data. Indeed, the face analysed via 16,000 control points such as eye-spacing, lip corners, nose edges, and so on.

However, we have already explained in a previous article how biased facial recognition algorithms can be. This can very quickly be compounded by poor lighting or low contrast between the foreground – the user’s face – and the background, which can promptly make facial recognition systems difficult.

Image: Shutterstock


Share This Post On

Submit a Comment

Your email address will not be published. Required fields are marked *