Skip to content

Liveness and Face2Face

Liveness

For the user, Liveness has one structure: instruction, camera with an oval, processing, result. The set of on-camera tasks depends on Workflow settings.

Instruction

Shown before the camera when the instruction page is enabled in the Workflow.

  • Heading "Instruction", subheading "Keep the camera at eye level".
  • Text "Make sure your face is evenly lit and the camera image is not overexposed. Remove headwear and accessories".
  • "Continue" button. A short spinner may follow: the interface loads models and checks the camera.

The instruction can be a separate page or a pop-up, depending on customization.

Camera

The user sees their video and an oval. Tasks and hints appear above the oval.

Tasks (set and order depend on settings):

Task On-screen text
Centering "Place your head in the center of the oval"
Approach "Move closer"
Turn "Turn your head left / right"
Gaze "Look up / down / left / right"
Hold gaze "Keep your head straight and look at this green bar"
Condition check "Please wait, we are checking the lighting and whether your face is visible"

Hints when something is wrong:

Hint What to do
"Face not found" Get your face into the oval
"Too far. Come closer" / "Too close" Change the distance to the camera
"Please turn on the light, we cannot see your face" Improve lighting
"Remove the mask" / "Remove the glasses" Remove the item covering the face
"Make sure you are alone in the frame" Stay alone in the frame
"Open your eyes" Look at the camera

Each task has a time limit. When it runs out the attempt counts as failed.

Processing

A waiting screen with "Preparing the camera…" before start and "Processing data…" after. No action is required.

Attempt result

  • Success: the next technology starts, no separate screen.
  • Failure with attempts left: the retry screen with the reason and tips. The number of attempts is set in Liveness settings in the cabinet.
  • Failure with no attempts left: the final screen with a failure.

Retry-screen messages and their causes are listed under Failure reasons.

Face2Face

In production this is a background step: the user only sees a waiting screen, no input is needed. The comparison result is not shown on screen; a negative result ends the check with a failure. The client's system retrieves comparison details through the result API.

In demo mode, if no photos exist yet, a "Face similarity" screen appears with two fields, "First photo" and "Second photo" ("Tap here to add an image"), and a "Compare" button.

Face2Face NPCK

Comparison with the reference photo from the state database (COID). The user sees a waiting screen "Face similarity via the COID service" and "Processing data". At the end: "Verification finished". No input is needed.