Pay by face at the checkout, an age hint at the self-checkout, traffic and queues in the store — without your own database of shoppers’ faces. Face biometrics in a store works through the Unified Biometric System (UBS) and your acquiring bank; we run the liveness check before matching. We count the shop floor, we do not recognise people.
Each card shows which store task the solution covers and which products to connect for it. Detailed pages appear in the section as they are released; the 18+ checkout and the shop floor are briefly described below.
Biometric payment
The shopper pays by face through the UBS, by palm — in a closed environment.
A terminal at the checkout and an ATM without a card: your acquiring bank matches the shopper’s face against the UBS, and before matching we check that a live person is in front of the camera, not a photo or a screen.
Payment by the palm vein pattern — in closed environments: a company canteen, a campus, a resort, with written consent and another way to pay nearby.
Checkout and shop floor
Age 18+ at the self-checkout and shop-floor analytics without recognising people.
Age estimation by face at the self-checkout — a hint that a document is needed; gender and age group — anonymised, for the audience report. The person is not identified.
Traffic, conversion by receipts, queues at the checkouts and a heat map of the floor on your cameras. We count people, we do not recognise them: no face templates.
Staff entrance and loyalty
Warehouse staff and bonuses that do not end up with fraudsters.
Page planned. Staff entrance, warehouse and the cash-office area: entry by face confirmed by a guard, or by card for those who declined biometrics. Tell us about your task — we will pick a scenario.
Page planned. Fake accounts for bonuses and points withdrawal: a document and Liveness at registration, a large redemption or a transfer of points — without face search.
Three groups of rules decide what a store may do with the camera at the checkout and on the floor. Below is a summary as of October 2026: a guide for a conversation with a lawyer, not a legal opinion.
572‑FZ, Arts. 3, 15, 16
Civil Code of the Russian Federation, Art. 152.1; 152‑FZ, Art. 11; 572‑FZ, Art. 1
171‑FZ, Art. 16; 15‑FZ, Art. 20
The same camera can work lawfully or not — it all depends on the scheme. We have collected the scenarios retail chains ask about and the route by which each can be launched after 572‑FZ.
We check that a live person is in front of the camera; estimate age without identifying the person; count visitors; recognise a document and match the face against the photo in it at registration.
Matching the face against UBS vectors and debiting the money is done by your acquiring bank through an accredited organisation. The shopper gives consent to biometric payment in their own bank.
Whether to sell alcohol, let a person into the warehouse or open a third checkout — the cashier, the guard and the manager decide. We give them a signal, not a verdict.
Neither we nor you store a database of shoppers’ faces: for payment it is in the UBS, for analytics it is not needed at all.
Saving shoppers’ faces to recognise them at the checkout or at the entrance — a direct route to the ban in cl. 1, Part 1, Art. 15 of 572‑FZ. For payment there is the UBS, for analytics faces are not needed.
A checkout where you can pay only by face violates Part 13, Art. 3 of 572‑FZ: refusing biometrics cannot be a reason to refuse a sale. A card and cash stay available.
Age estimation by face is off by years in both directions. Alcohol and tobacco are sold after a document or data from MAX; the camera only hints.
Anonymised traffic stitched to the cards and receipts of specific shoppers stops being anonymised. Keep shop-floor analytics and loyalty separate.
A terminal without Liveness can be fooled by a photograph or a phone screen. A check before matching rules out spoofing right at the checkout.
572-FZ
Arts. 3, 15, 16
At a self-checkout there is no seller nearby, and the law requires not selling alcohol and tobacco to minors. Age estimation by face helps call a member of staff only when needed.
The checkout sees an item with an age restriction and turns the camera on. Without such an item, age estimation does not start.
A neural network estimates age by face and does not compare it with any database. The frame is not saved, the name is not established.
If the estimate is well above the threshold, the employee gets a quiet mark and confirms the sale faster by document or data from MAX.
If the age is close to the threshold or the camera is unsure, the checkout asks you to wait for an employee. The decision on the sale is made by them, as the law requires.
Cameras the store installed for security can also answer the manager’s questions. For that you do not need to know who is in front of the camera — counting is enough. How video analytics works.
How many people came in by hour and how many of them bought — by the number of receipts from your point-of-sale system.
An event to the administrator when there are more people at the checkouts than set, and a report on the hours when checkouts are not enough.
Where shoppers stop and which shelves they pass by — for merchandising and promotions.
The share of men and women and age groups — in the report, without faces and without a link to loyalty cards.
Together with your lawyer and the acquiring bank we choose what to launch: payment through the UBS, an 18+ hint, shop-floor analytics. For each scenario there is a route from the block “which scenario is lawful and by which route”.
We connect the checkout and cameras of one store and set the age threshold and the floor zones. For a month we review the results with the manager and the security service.
We transfer the settings to the other stores, combine the reports into one table and send events to your system through the API.
A liveness check before face matching: against a photo, a screen, a mask and a deepfake. Accuracy — 99.9 %.
1:1 comparison of a selfie with the photo in a document and age estimation by face without identifying the person.
Visitor counting, queues, heat maps and events in staff areas on your cameras.
Liveness and a ready-made capture interface for the terminal; matching against the UBS is done by the acquiring bank.
A document and Liveness online — for 18+ loyalty programmes, pre-orders and services with an age threshold.
Remote identification through the Unified Biometric System — for partner banks and the chain’s financial services.
Online checks are paid per session according to the price matrix. Checkouts, terminals and video analytics are priced for the chain: the price depends on the number of stores, checkouts and cameras.
from 29 ₽
Per check: document, Liveness, face matching against the photo and age by face against the date of birth — for loyalty and 18+ services.
The checkout or cameras of one store: an 18+ hint, traffic and queues, a review of the results with the manager.
on request
Checkouts and terminals across the whole chain, video analytics, installation on your servers with no data leaving the perimeter.
Prices for online checks are in the price matrix. Equipment, installation and the acquiring bank’s fee for biometric payment are separate, from your suppliers and the bank.
A store needs a supplier that covers the checkout, the floor and the staff entrance and does not ask you to store shoppers’ faces.
For each scenario we show the route in advance: through the UBS, without identifying the person or with a decision by staff. The lawyer and the acquiring bank only need to check the scheme, not invent it.
The scheme comes before the equipment, not after an inspection.
Checkout, self-checkout, floor and staff entrance — one platform and one contract.
The project «Development of a multimodal system using artificial intelligence "Neuromed" for managing the diagnostic processes of a clinic on the basis of the "Neurovision" computer vision platform» is supported by the Innovation Promotion Foundation within the federal project «Domestic Solutions»
The Neurovision computer vision platform is registered with Rospatent (certificate No. 2023661508) and included in the Russian Software Registry (entry No. 19972 of 13.11.2023)
The right to use the Platform is granted under a licence agreement concluded by Neurovision with its customers
Words you will meet in a contract with the bank, in a conversation with a lawyer and in reports for managers.
Yes, but not in just any way. Payment by face goes through the UBS and the acquiring bank. Age estimation and visitor counting work without identifying the person. Organisations are prohibited from keeping their own database of shoppers’ faces to recognise them (cl. 1, Part 1, Art. 15 of 572‑FZ).
Biometrics are human characteristics used to identify a person and used for that purpose (Part 1, Art. 11 of 152‑FZ). Ordinary video recording of the floor and anonymised people counting are not biometrics; recognising a specific shopper is.
No. That would require storing face templates outside the UBS, which is prohibited. Regular customers are recognised by loyalty card or phone, and video analytics counts visits without linking them to people.
The shopper hands over biometrics to the UBS once and connects payment in their own bank. At the checkout the terminal captures the face, we check liveness, the acquiring bank matches the face against the UBS and debits the money. The store does not store faces. Details are on the page about biometric payments.
The UBS currently stores only a face image and a voice recording, so payment by palm through the UBS is not yet possible. The palm is used in closed environments — a company canteen, a campus — with written consent and another way to pay nearby.
No. Refusing biometrics cannot be a reason to refuse a sale or service (Part 13, Art. 3 of 572‑FZ). A biometric checkout always accepts a card and cash.
No. In case of doubt about age the seller may require a document or data through MAX (subcl. 11, cl. 2, Art. 16 of 171‑FZ). The camera estimates age and hints when to call a member of staff, and they make the decision.
No. In payment the frame goes to the bank for matching against the UBS; age estimation and visitor counting do not save faces. A face template is kept only during an online document check — to match a selfie with the passport photo within one session.
Yes, if the guard makes the decision: they see the camera frame and the photo from the list and open the door themselves — 572‑FZ does not apply to such a check (cl. 2, Part 2, Art. 1). For employees who declined biometrics, a card remains.
By checking a document and Liveness at registration, on a large redemption and on a transfer of points. Searching for duplicate members by face is not allowed — that is storing biometrics for identification outside the UBS.
An online document check with Liveness and age by face — from 29 ₽ per session. Checkouts, terminals and video analytics are priced for the chain; the first month of a pilot in one store is free.
Yes. Face recognition, Liveness and video analytics are installed in your environment with no data leaving the perimeter. Matching against the UBS for payment is still done by the acquiring bank.
Tell us how many stores and self-checkouts you have, which acquiring bank you use and whether there are cameras on the floor — we will suggest a scenario and a pilot in one store.
Can't put your request into words? Book a call and our team will contact you at a convenient time
Book a callThe company is included in the register of IT companies and the register of personal data operators of the Russian Ministry of Digital Development.
Address: Moscow, Presnenskaya Embankment 12 — Moscow City business centre, Federation Tower. Nearest metro stations: Delovoy Tsentr and Mezhdunarodnaya.
