When an institution enrolls twins or very similar-looking siblings, seeing a name on screen is not enough. The goal is to apply a reliable verification process when matching is uncertain. Staff can check an unrecognized student; a record created under the wrong identity may affect attendance and parent messages.
1. Keep student records clearly separate
Use a distinct institution ID for each student. Double-check names, classes and photo associations during enrollment. Do not use a photo containing both siblings as an individual enrollment image. Similar names can cause confusion during bulk imports, so review profiles afterwards.
2. Understand the limits of recognition
Face recognition is a probabilistic matching process and should not be evaluated with a universal zero-error assumption. BE Soft applies additional comparison and checks across several frames for similar candidates; it may abstain when the distinction is unclear. Review conditions and records before treating this as a device fault.
3. Test each twin separately
During the pilot, test each student individually at the actual entrance device. The displayed name and recorded event must refer to that person. Include ordinary conditions such as morning lighting, standing distance and different clothing. One successful trial is not evidence for an entire term.
4. Do not approve an uncertain identity
If the wrong name appears, do not confirm it. Staff should verify the student using the agreed institutional method. If a record has already been created, review the event and any parent message. Contact support with device, time and event details rather than changing thresholds arbitrarily.
5. Set a simple daily rule
Students should approach the camera individually, avoiding two faces in the frame. Define where an unrecognized student waits and when staff intervene. This reduces pressure to approve an uncertain result during busy arrivals.
Review your institution’s scenario
Plan enrollment, pilot checks and staff intervention together. Contact BE Soft to discuss a verification workflow suited to your institution.