Integrations & SDKs
Add face-verified pickup, attendance and access control to your platform
SooperVision's Authorised Pickup and Educator Attendance modules run on any phone or tablet, verify faces on the device in under a second, and keep working offline. When online, they sync records securely with our servers in Sydney. Use them standalone or build them into your own software with our SDKs.
Accurate, private and built for the real world
Independently benchmarked by NIST
The face recognition algorithms behind SooperVision are evaluated by the US National Institute of Standards and Technology (NIST), part of the US Government, in its Face Recognition Technology Evaluation (FRTE). Results are public, not marketing claims.
Runs on the device
Face matching happens on the phone or tablet itself, so verification is fast and never waits on a cloud service. Pickup and attendance records sync securely to our servers in Sydney; face images are never uploaded.
Works offline
Verification keeps working when the internet drops. When the device is online it syncs records with our secure servers in Sydney, Australia, and catches up automatically after an outage.
Any phone or tablet
Native Android and iOS, plus Flutter and React Native. Use a low-cost tablet at the door, a staff phone, or devices already in your customers' centres. No special camera or reader is needed.
Liveness built in
Passive liveness detection checks that a real person is in front of the camera, not a photo or a screen, with no blinking or head-turning required.
Privacy by design
At enrolment, the photo is turned into an encrypted face template and the original image is deleted. Adults give consent before enrolment and can withdraw it at any time. Children are never enrolled or identified.
Accuracy you can check
of travellers correctly matched in NIST's Paperless Travel evaluation, the US Government test that simulates boarding an aircraft by face instead of a boarding pass (420-person gallery, false positive rate held at 3 in 10,000).
Boarding a plane by face is the same problem as verifying an authorised adult at the pickup door: a known group of people, one camera, a decision in under a second.
1:1 verification (is this person who they claim to be?). True match rate at a false match rate of 1 in 100,000. NIST FRTE 1:1, mugshot dataset.
1:N identification (who is this, from a list?). Rank-1 identification rate with 12 million people enrolled. NIST FRTE 1:N.
Results are for FaceTagr's algorithms as submitted to NIST. They have been refined since and validated through millions of real-world verifications. NIST Paperless Travel results → · NIST 1:1 results →
A false match rate of 1 in 100,000 means that, on average, only one in 100,000 comparisons of two different people is wrongly accepted.
Used by enterprises in 9+ countries
of real-world face verifications in production
with enterprise deployments in workforce attendance, access control and transport
certified information security; data hosted in Sydney on SOC 2-compliant servers
Three ways to use it
Standalone
Centres use the SooperVision portal and dedicated device. No other software required.
Ready-made portal and tablet app. No development needed.
Integrated via SDK
Software providers embed pickup verification and educator attendance in their own app, branded as part of their product.
Your app, your brand. Face matching runs inside your app on the user's device.
Access control
Verified adults open or log entry at doors and gates through compatible entry hardware. Approvals and revocations sync in real time.
Proven with Dormakaba and Magnetic Access gate systems, and configurable to other door and gate hardware.
Two SDKs, one simple integration
| SDK | What it does | Typical use |
|---|---|---|
| 1:1 Verification SDK | Confirms a person matches their own enrolled template, with passive liveness | Authorised pickup at handover, educator sign-in on a shared tablet |
| 1:N Identification SDK | Finds a person among everyone enrolled at a centre or group, with passive liveness | Hands-free door entry, walk-up attendance, blocked-person alerts at the gate |
Authorised Pickup features
Enrol consenting adults · time-limited pickup permissions (date, range or number of pickups) · director-level court-order and AVO blocks that parents cannot override · verification result at handover · timestamped pickup log.
Educator Attendance features
Consent-based sign-in and sign-out · anonymous class and room child counts · automated attendance reports for ratio evidence · shift and on-premises visibility for directors.
Platforms: Native Android · Native iOS · Flutter · React Native.
Who it's for
- Childcare management and parent-communication platforms
- Rostering, attendance and compliance software
- CCTV installers and security integrators
- Access control providers
- Workforce, HR and payroll platforms that need verified attendance for childcare staff
Designed for Australian services
- Children are never enrolled or identified
- Consent required before any adult is enrolled; withdrawal and deletion on request
- Encrypted templates; original enrolment photos deleted
- Verification on the device; face images are never uploaded
- Records sync to secure servers in Sydney, Australia; works offline in between
- Separate from CCTV Supervision, which never recognises anyone
How partnering works
- 1Tell us about your platform or customers
- 2Get SDK documentation and a sandbox (shared with approved partners)
- 3Build and test with our team
- 4Launch to your centres, with Australian support from GreeneDesk
Request SDK access
We reply within one business day with SDK documentation, a sandbox and partner pricing.
SDK documentation is shared with approved partners.
Frequently asked questions
Can Authorised Pickup and Educator Attendance be used without CCTV Supervision?
Yes. Each module works standalone or alongside the others.
Do you offer SDKs?
Yes. SDKs are available for Authorised Pickup and Educator Attendance so software providers can integrate them into their own products.
Can the modules control door or gate access?
Yes, with compatible entry hardware. Verified adults can open or log entry, and access is revoked instantly when permissions change.
Are children's faces enrolled?
No. Only consenting adults are enrolled. Children are never enrolled or identified.
Where is data stored?
Face verification runs on the device; supporting services are hosted in Sydney, Australia on an ISO 27001-certified platform.
Is CCTV used for face recognition?
No. Face recognition is switched off on CCTV Supervision.
How accurate is the face verification?
The algorithms are independently benchmarked by NIST, part of the US Government. In NIST's Paperless Travel evaluation, which simulates boarding an aircraft by face, 99.91% of travellers were correctly matched. In NIST's 1:1 evaluation the true match rate is 99.55% at a false match rate of 1 in 100,000, and in 1:N identification with 12 million people enrolled the rank-1 identification rate is 99.71%.
Does it need an internet connection?
No. Verification runs on the phone or tablet and works offline. When connected, the device syncs records with our secure servers in Sydney; after an outage it catches up automatically.
Which devices does it work on?
Any modern Android or iOS phone or tablet. SDKs are available for native Android, native iOS, Flutter and React Native.
Can someone fool it with a photo?
Passive liveness detection checks that a live person is present, not a printed photo or a screen.
Do you see or store our users' faces?
No face images are stored. Enrolment turns the photo into an encrypted template and deletes the image. Matching runs on the device. Pickup and attendance records sync to secure servers in Sydney.
What is the difference between 1:1 and 1:N?
1:1 checks that a person is who they say they are, for example a parent at handover. 1:N finds a person among everyone enrolled, for example walk-up entry at the door.
Which access control hardware does it work with?
It has proven integrations with Dormakaba and Magnetic Access gate systems and can be configured for other door and gate hardware.