Biometric Identification

Comparing a live face to stored identity images or watchlists to verify who a person is or flag a match in security screening.

Business impact

• faces scanned — Cambridgeshire Police reported scanning 29,361 faces in one six-hour operation.
• arrests made — Cambridgeshire Police reported three arrests after live facial recognition matches.
• identification attempts — In the Chinese government case, the system attempted more than 500,000 identifications in Sanmenxia over about a month.
• time per identity check — U.S. Customs and Border Protection said the airport facial-recognition check takes less than two seconds.
• crowd size processed — Jiangxi Province Police identified a person of interest in a concert crowd of nearly 60,000.

Data requirements

• face images (image) — Captured from cameras, live video, or a kiosk camera.
• passport chip photo (image) — Used as the identity reference in airport screening.
• watchlist entries (tabular) — Used to compare scanned faces against known people of interest.
• database of known stalkers or persons of interest (tabular) — Used as the comparison set in security and police deployments.

AI methods and techniques

• Face matching — Compares a captured face with a stored reference image or a watchlist.
• Real-time video analysis — Processes live camera feeds and checks faces as people pass by.
• Biometric comparison — Matches a traveler’s face to the photo embedded in a passport chip.

Industries

Real-world evidence

8 documented case studies on record.

Companies using this: Rose Bowl Stadium, Alibaba Group / Alibaba Cloud, Chinese government, U.S. Department of Homeland Security (Customs and Border Protection), Orlando Police Department, Cadillac Fairview, Jiangxi Province Police, Cambridgeshire Police.