GBL Face - facial recognition technology for face detection, verification and identification



  • 1.95% rank-10 probability among 10K photos
  • 2.88% rank-10 probability among 1M photos
  • 3.99% verification accuracy
  • 4.Half-second search time for 1 billion photos
  • 4.The index needs just 20 GB for 500 million photos



Safer Cities. Facial recognition and person identification from thousands of cameras. City streets, airports and railway stations, subway, transportation hubs, malls and government establishments, event venues and stadiums, sensitive sites


On many occasions, the only available clue to solve a criminal case is a photo or video camera recording. With our unique facial recognition technology, any face captured on a photo or a video is now actionable evidence, with the person easily identified among millions (or even billions) of loaded photos.


Unique neural network demonstrates utmost accuracy among all commercially available solutions. The false accept rate (FAR) is as good as 1 in 1,000,000, making for superior security, whether used alone or combined with other authentication methods. Facial recognition is the easiest security authentication method yet. Just think about how many times a day your employees type in their passwords, PIN codes, or swipe their access cards? With face-based authentication, there’s no need anymore — now every face is an ID.


With GBLFace's state-of art facial recognition technology, you can automatically confirm that the user in front of a computer or a smartphone matches the photo they have on file. Moreover, with our unique ability to search among billions of photos in less than a second, you can further improve scoring accuracy by checking a user's social network profiles or against a database of known fraudulent people.


With facial recognition technology accurate enough to identify a visitor from a camera's video stream and quickly search hundreds of faces per second in a dataset of millions (even billions) of faces, impressive customer analytics techniques can come to offline word.


When you’re running an event registration booth, how cumbersome is it to search names or scan barcodes? Who really wants to scan badges to check eligibility or exchange contact details? How long does it take to find a relevant photo in an event’s enormous photo album?