Face API · 1:N face search

Face recognition API.

Identify enrolled people in any photo. SIMSIM searches each detected face against your gallery and returns the best match, a confidence score and its location.

Start free Talk to sales

POST /v1/recognize

Endpoint

One call. Structured JSON back.

Same token, same error model and the same response conventions as every other SIMSIM endpoint.

POST/v1/recognizeIdentify enrolled people in a photo (1:N) with a confidence score per match.

API reference

Capabilities

What you can build.

Gallery search (1:N)

Match every face in an image against the people you've enrolled.

Confidence per face

A calibrated score so your rules — not ours — decide what counts as a match.

Multi-face images

Group photos and camera frames: each face is matched independently.

Enrol & delete by API

Add people as templates; remove them with one call when consent ends.

Use cases

Where teams use it.

iGaming

Age assurance, one-player-one-account and self-exclusion enforcement.

See use case →

Retail & logistics

Staff access control, depot security and opt-in loyalty experiences.

See use case →

Trading

Confirm the account holder before margin, limits or withdrawals change.

See use case →

FAQ

Face Recognition questions.

What's the difference between recognition and verification?

Recognition (1:N) asks "who is this?" against a gallery; verification (1:1) asks "is this the same person?" for two images. Most KYC flows need verification; access control and watchlists use recognition.

How large can my gallery be?

Up to your plan's face-storage allowance — from 100 templates on Free to 54,000 on Enterprise. Contact us for more.

Is 1:N identification regulated differently?

Often, yes. In the EU, remote biometric identification systems are classed as high-risk under the AI Act, and real-time use in public spaces is heavily restricted. Check Western Europe.

Learn more