Nirovision uses the face(s) included in someone's profile to recognise them when they appear in front of Doorkeeper(s) and IP camera(s). That's why unrecognised activity is generally caused by low-quality profile images.
Examples of low-quality profile images include: blurry faces, zoomed-in faces, faces smaller than 100x100 pixels, distorted faces, partial and occluded faces, sideways faces and faces with very little pixel variance.
The best way to address this is to add at least one good quality face to the person's profile. However, the process is different based on the Nirovision product you use:
Check-in activity from Doorkeeper.
IP camera events from Monitor.
What happens when someone that is already an Identity in Nirovision doesn't get recognised by Doorkeeper?
They will be treated like any other unknown person and guided through the onboarding flow you've designed to enrol new profiles in the system. As they complete their onboarding requirements, Doorkeeper will look for possible matches against the provided email and phone number.
If there is a match, Doorkeeper will add the face collected during this event to the existing identity. Adding this high-quality face improves this individual's identity, and they will most likely be recognised every time coming forward.
If there is no match, a new identity will be created with the details provided during this enrolment. If this ends up being a mistake (a duplicate), these two identities will need to be merged. Learn more 👈
What happens when someone that is already an Identity in Nirovision doesn't get recognised by an IP camera using our Monitor software?
An event is created in the Activity view, with a No Match result associated.
Even if there is No Match, there may be a similar (might be) identity in Events that closely resemble an identity in the system but did not meet our recognition threshold at that time. It is worth enabling the Person > Similar filter in the Activity (Events) view to expose this activity, as that will help you collect additional faces to improve future results for this person.
However, it's also possible that Nirovision has nothing to compare that face against, if the person does not exist in the system as an Identity. Nirovision matches faces against your Identity database at the time; creating an Identity using a high-quality face for this person will ensure Nirovision will match that person coming forward.
On the Nirovision Web App 🖥️ go to Activity and navigate to Events.
Filter events by applying the Person > Similar filter on the left side of the window. This will only expose events that have "No Match" and have a "might be" message below the "No Match".
This indicates that the face detected in this event is up to the standards of facial recognition to improve the detection of this person in Nirovision.
To assign the face to an Identity, simply click on the face at the bottom of the event.
A module will appear on the right side of the screen with a section indicating that this face is similar to an identity in the Nirovision database. Click on the Edit button next to the "No Match".
Simply click on the similar identity and click Mark as same.
A prompt will ask if you would like to edit the check-in times for this identity, you can click cancel on this step if the person has checked in earlier.
This will take you back to the activity view and your event will now be named as that person.
On the Nirovision Web App 🖥️ go to Activity and navigate to the Events activity page.
Every time a face of a person is detected it will appear in the activity view as a "No match"
To create an identity for a new person detected in the Events activity, simply click on the face at the bottom of the event
A window will pop up. Choose the Create Identity option, on the bottom left of the window.
In the middle complete the data fields and add labels.
Click the Save button on the bottom right.
This will take you back to the activity view and your event will now be named as that person and a new identity will have been created.