Face Match
Face Match confirms that the person presenting themselves is the same person shown in the identity document, by comparing the live selfie against the portrait extracted from the ID.
This module typically runs right after the selfie and the ID document have been captured. It acts as the biometric link between the user and their document, returning a clear match / no-match result before the flow continues.
Where it fits in the flow
Face Match usually appears immediately after Face Capture and ID Capture, once both a selfie and an ID portrait are available.
The module compares the two images and surfaces the outcome to the user. On a successful match the flow continues to downstream verification or submission; on a no-match it presents a clear error state before handing control back to the application.
User Flow
The Face Match experience is short and reassuring. The user first sees their selfie and ID portrait side by side under a clear “Verifying identity” heading, confirming which two images are being compared. The system then runs the biometric comparison while a brief processing state lets the user know their photos are being checked.
Once the comparison completes, the user receives an explicit result: a success state confirming the faces matched, or an error state indicating the faces do not match. Both outcomes provide a clear next action so the user always knows how to continue.
Full Flow Map
This diagram presents the full sequence of screens involved in Face Match, from the side-by-side comparison and processing state through to the match and no-match outcomes.
Happy Path (Light & Dark)
The ideal journey, where the selfie and ID portrait belong to the same person and the comparison succeeds without interruption.
Light mode
Dark mode
Best Practices
Recommended guidelines for designing and implementing the Face Match experience.
✅ Do
- Make it obvious which two images are being compared (Selfie vs ID).
- Keep the processing state short and reassuring.
- Communicate both match and no-match outcomes explicitly.
- Provide a clear next action on every result screen.
- Maintain high contrast and readable text across light and dark themes.
❌ Don’t
- Don’t hide or merge the two source images in a way that obscures what is being compared.
- Don’t rely on color alone to communicate the match result.
- Don’t skip the no-match error state.
- Don’t add interaction during the processing step.
Updated 1 day ago