Skip to main content
Our age estimation process includes automated analysis to verify the age of users through biometric processing. This page outlines the potential warnings and errors you might encounter during the age estimation process.

​
Verification Process

When performing age estimation, the system attempts to:
  • Detect and analyze facial features in the captured images/video
  • Apply machine learning algorithms to estimate the user’s age
  • Compare the estimated age against specified age requirements
  • Validate the quality and confidence of the age estimation data

​
Configurable Verification Settings

Applications can configure how the system handles various verification issues:
  1. Age Below Minimum
    • Risks: AGE_BELOW_MINIMUM
    • Configurable options:
      • Age threshold: Set the specific minimumage requirement (e.g., 18, 21) users must meet
      • ID verification fallback: Automatically initiate ID verification for borderline cases when enabled
  2. Low Liveness Score
    • Risks: LOW_LIVENESS_SCORE
    • Configurable thresholds:
      • Review threshold: Sessions with scores below this threshold are set to “In Review”
      • Decline threshold: Sessions with scores below this threshold are automatically declined
  3. Possible Duplicated Face
    • Risks: POSSIBLE_DUPLICATED_FACE
    • Configurable options:
      • Similarity threshold: Define the similarity threshold for detecting duplicate faces
      • Action on detection: Configure whether to reject, review, or flag sessions with potential duplicates

​
Age Estimation Warnings

Tag
Description
AGE_BELOW_MINIMUM
The estimated age falls below the minimum age requirement for the service or product being accessed.
AGE_NOT_DETECTED
The system was unable to estimate the user’s age, which may be due to image quality, lighting issues, or other technical factors.
LOW_LIVENESS_SCORE
The liveness check resulted in a low score, indicating potential use of non-live facial representations or poor-quality biometric data.
NO_FACE_DETECTED
The system couldn’t identify a face during the liveness check, which may be due to poor image quality, improper positioning, or technical issues.
LIVENESS_FACE_ATTACK
The system detected a potential attempt to bypass the liveness check.
FACE_IN_BLOCKLIST
The detected face matches an entry in the system’s blocklist, preventing verification.
POSSIBLE_DUPLICATED_FACE
The system detected significant similarity between the user’s face and previously verified faces, indicating potential duplicate registration.

​
Warning Types

Each risk is assigned a warning type based on your application’s configuration settings. Warnings fall into three severity categories: