Biometric Face Comparison
Face comparison determines whether the images of faces belong to the same person in identity verification.
Another use case is second factor authentication for login, when the user presents his face to be verified with a previously stored face image of him.
Comparison threshold and comparison decision
The final comparison decision if the two face images belong to the same person is determined by the comparison score and the threshold. If the score is above the threshold this can be interpreted as a match, if the score is below the threshold it is a non-match.
Comparison accuracy
Comparison thresholds affects the decision mistakes. False match is when comparison decides that face images of two different persons are matching. False Match Rate FMR (formerly FAR) is the proportion of comparison trials that result in a false match on a given dataset with a given threshold. False non-match is when comparison decides that two face images of the same person do not match. False Non-match Rate FNMR (formerly FRR) is the proportion of comparison trials that result in a false non-match on the same dataset and with the same threshold. The combination of these two FMR @ x% FNMR defines the accuracy of comparison.
Thresholds for comparison of ID document portrait vs selfie
The following characteristics have been measured on our face testing dataset using Digital Identity Service (accurate extraction mode) for the comparing selfie to document.
| FMR - Use case type | FMR [%] | FNMR [%] | Score threshold |
|---|---|---|---|
| 1:5,000 - Convenience | 0.02 | 0.405 | 0.287 |
| 1:10,000 | 0.01 | 0.681 | 0.31 |
| 1:15,000 - High Security | 0.006 | 1 | 0.329 |
Thresholds for comparison of selfie vs selfie
This use case is for comparing two selfies in face authentication process.
| FMR - Use case type | FMR [%] | FNMR [%] | Score threshold |
|---|---|---|---|
| 1:10,000 - Convenience | 0.01 | 0.472 | 0.553 |
| 1:25,000 | 0.004 | 1.00 | 0.589 |
| 1:50,000 - High Security | 0.002 | 1.27 | 0.603 |
Example
If we require in the document to selfie comparison an FMR rate below 0.01, we have to set the threshold of the comparison score to 0.33. Then, if we have a representative set of 100,000 mated face image pairs (selfie and of the same person), statistically 1000 will in this case be incorrectly decided as a non-match. If we have 100,000 not mated face image pairs, statistically 6 will be wrongly decided as a match.
Trust factors for face comparison
The matching is in IDV Platform configured by the following Trust factors:
- Document face match
- Smile liveness matching threshold
- Multi-range liveness matching threshold
- Document NFC face match
- Face identification threshold
During the initial configuration of the system, two thresholds should be set. If the score is below the bottom threshold, the result is automatically set to reject. If the score is above the top threshold it is automatically accepted. If the score is between the two thresholds, images go for review to a back office operator for a final decision. Over time, the range between the two thresholds can be minimized or even eliminated by setting a single decision threshold if not needed anymore.
When setting the face comparison threshold, please consider the following:
- Image quality - if image quality is low, accuracy of the comparison decreases
- Age difference between the images - time difference between the capture of the two images is several years, the person's appearance might have changed significantly.