AttenTech
Identity APIs

Face Verification

Match a selfie to an ID photo and get a liveness-checked decision in one call.

Get API AccessPrivate beta only
// OVERVIEW

How Face Verification works

Face Verification compares a live selfie against a reference ID photo and returns a calibrated match score, a liveness check, and a pass/fail decision — the exact check onboarding and step-up authentication flows need before trusting a claimed identity. Built to hold up against phone cameras, poor lighting, and spoofing attempts, not just studio headshots.

Direct face matching

A selfie is compared directly against the ID photo on file, returning a calibrated similarity score rather than a bare yes/no.

Liveness detection

Passive liveness checks catch printed photos, replayed video, and masks before a spoofed face ever reaches the match step.

Threshold-tuned decisions

Match score and liveness result combine into a single decision against a threshold you control, so risk tolerance stays yours to set.

face
attentechth.com/api/identity/face-verify
01// verify a selfie against an ID photo
02import { Face } from '@attentech/sdk'
03
04const face = new Face({
05 region: 'th-bangkok',
06 threshold: 0.85,
07})
08
09const result = await face.verify({
10 selfie: selfieImage,
11 reference: idPhoto,
12})
13
14if (result.decision === 'pass') {
15 const { matchScore, liveness } = result
16 approve(result)
17}
// PROCESS

How Face Verification matches a selfie to an ID photo

Both faces are normalized and encoded, then compared for similarity while a passive liveness check runs in parallel — so a match score alone never has to carry the whole trust decision.

MATCH SCORE
LIVENESS
DECISION
Capture
Analyze
Score
Structured Output

Ready to integrate Face Verification?

Get API access in days, or bring us in to run the full workflow for you — sovereignty by design at every step.

Get API AccessPrivate beta only