Glossary

Definitions of terms used across the IDV Platform documentation. See also Key Concepts for how the main records relate.

People & organizations

Company — an organization using the platform to verify its end users.
Integrator — a developer at the Company who connects the platform to the Company's systems.
User — a person going through a verification; not yet necessarily a Customer.
Customer — a verified person in the Company's database.
Tenant — a Company's isolated space on the platform.

Digital Identity — the record of a single verification attempt.
Customer record — the master record for a person, linking all their Digital Identities.
Primary identity — the Digital Identity that is the current source of truth for a Customer's data.
Assignment — a link between a Customer and something they were issued (SIM, account, policy).
Trust Factor — a configurable check that contributes to the accept / review / reject decision.
Trust Factor profile — a named set of Trust Factors and thresholds referenced by a workflow.
Workflow — the ordered sequence of steps a verification runs through.
Session — one run of a workflow for one User.

OCR — Optical Character Recognition; reading text from a document image.
VIZ — Visual Inspection Zone; the human-readable printed area of a document.
MRZ — Machine-Readable Zone; the coded lines at the bottom of many IDs and passports.
NFC / eMRTD — the cryptographically secured RFID chip in e-passports and eIDs (ICAO Doc 9303).
Face matching — see Face comparison 1:1
Face comparison 1:1 — comparing one face against one reference (e.g. selfie vs document portrait).
Identification 1:N — searching one face against many enrolled records.
Liveness / PAD — Presentation Attack Detection; confirming a live person, not a photo, screen, or mask.
Deepfake detection — detecting AI-generated or manipulated face images.
Injection-attack detection / IAD — detecting a fake video feed injected in place of a real camera.
Age estimation — estimating a person's age from their face.
Age verification — verifying a person's age from their date of birth on ID document.
Age validation — checking that the estimated age and the document age are consistent.
Palm verification — enrolling both palms of a Customer and authenticating them 1:1 with either palm; an optional modality next to the face.

Duplicity check — detecting that a person already exists in the records, by face or by document data.
Name vector — a non-reversible numeric representation of a person's name used for 1:N document duplicity search; tolerates spelling variants, cannot be turned back into the name.
Combined duplicity outcome — the single result (full_hit, similar_face, similar_name, face_missmatch, doc_missmatch, multiple_full_hits) that merges the face and the document duplicity signals.
Watchlist — a gallery searched by 1:N (Customer watchlist, Blocklist, Review watchlist, and in-flight tracking).
Blocklist — the watchlist of people to reject; a biometric match auto-rejects the attempt.
Anonymization — irreversible erasure of a person's personal data and biometrics (GDPR).

External ID — the identifier under which the Company's own systems know a Customer; unique per tenant, optionally composed by the workflow from document data.
SSO — single sign-on; dashboard users sign in through the identity provider of the Company.
TCS / usage reporting — the Innovatrics Transaction Counting System that receives operation counts and record counts for billing; no personal data.
Release version — the MAJOR.MINOR.PATCH number of the platform; the release history is in What's New.

APCER - Attack Presentation Classification Error Rate - Rate of attack presentation images that are classified as bona-fide presentations (false accepts). (formerly FAR).
BPCER: Bona-fide Presentation Classification Error Rate - rate of bona-fide presentation images that are classified as attacks (false rejects). (formerly FRR).
FMR — False Match Rate is the proportion of comparison trials that result in a false match on a given dataset with a given threshold.
FNMR — False Non-Match Rate is the proportion of comparison trials that result in a false non-match.
FAR — False Acceptance Rate
FRR — False Rejection Rate