Skip to Content

Face Forward

OmniPerception’s facial-recognition technology protects privacy as well as property.

In today’s security-conscious world, better access control-whether it’s a company restricting entry into its building or a government monitoring entry into a country-has become a priority. One solution gaining popularity is biometrics, systems that use specific biological traits such as fingerprints or facial features to identify individuals. Face recognition is an especially appealing technique, because capturing an image of the face is simple and nonintrusive. But using face recognition for applications such as border control can require querying a database of thousands to millions of photos, which is time consuming and raises privacy concerns.

To get around these problems, OmniPerception, a spinoff from the University of Surrey in England, has combined its facial-recognition technology with a smart-card system. This could make face recognition more robust and better suited to applications such as passport authentication and building access control, which, if they use biometrics at all, rely mainly on fingerprint verification, says David McIntosh, the company’s CEO. With OmniPerception’s technology, an image of a person’s face is verified against a “facial PIN” carried on the card, eliminating the need to search a central database and making the system less intimidating to privacy-conscious users.

OmniPerception’s technology creates a PIN about 2,500 digits long from its analysis of the most distinctive features of a person’s face. The number is embedded in a smart card-such as those, say, that grant access to a building-and used to verify that the card belongs to the person presenting it. A user would place his or her card in or near a reader and face a camera, which would take a photo and feed it to the card. The card would then compare the PIN it carried to information it derived from the new photo and either accept or reject the person as the rightful owner of the card. The technology could also be used to ensure passport or driver’s license authenticity and to secure ATM or Internet banking transactions, says McIntosh.

The key differences between the various face recognition systems hitting the market reside mainly in the algorithms that create digital code from their analysis of faces. Kittler says his algorithms require less computation, allowing all the processing to happen on the card, instead of in an external computer, which results in nearly instantaneous identity verification. Fast processing would be particularly useful at, for example, a busy airport or ATM.

Biometrics companies typically validate their technologies by participating in competitions that test their accuracy at identification. OmniPerception won a European competition last year but has yet to go head-to-head with more prominent U.S. players in the field’s premier competition, which is sponsored by the U.S. government. And it must also contend with more established biometric techniques-namely fingerprinting.

Even so, with plans to demonstrate its technology this year to U.S. passport authorities, British and U.S. driver’s-license agencies, and security companies worldwide, OmniPerception believes it has a head start toward the future of face recognition.

Others in Facial Recognition
A4Vision (Cupertino, CA) 3-D facial recognition for access control and surveillance
Acsys Biometrics (Burlington, Ontario) Neural-network facial recognition for building and border access control
Identix (Minnetonka, MN) Facial-feature analysis for access control and surveillance
Viisage (Littleton, MA) “Eigenface” technology for access control and surveillance

Keep Reading

Most Popular

Large language models can do jaw-dropping things. But nobody knows exactly why.

And that's a problem. Figuring it out is one of the biggest scientific puzzles of our time and a crucial step towards controlling more powerful future models.

The problem with plug-in hybrids? Their drivers.

Plug-in hybrids are often sold as a transition to EVs, but new data from Europe shows we’re still underestimating the emissions they produce.

Google DeepMind’s new generative model makes Super Mario–like games from scratch

Genie learns how to control games by watching hours and hours of video. It could help train next-gen robots too.

How scientists traced a mysterious covid case back to six toilets

When wastewater surveillance turns into a hunt for a single infected individual, the ethics get tricky.

Stay connected

Illustration by Rose Wong

Get the latest updates from
MIT Technology Review

Discover special offers, top stories, upcoming events, and more.

Thank you for submitting your email!

Explore more newsletters

It looks like something went wrong.

We’re having trouble saving your preferences. Try refreshing this page and updating them one more time. If you continue to get this message, reach out to us at with a list of newsletters you’d like to receive.