Who is watching the cameras of a whole branch network?
Many branches, few eyes. You set the rule once and it holds for all of them.
We show thirteen cameras here, in a branch and on its street: eleven that raise an alert when something happens, one that alerts nobody — it just counts — and one that shows how you find the exact moment afterwards. And one thing worth saying early: IRIS identifies nobody, it does not recognise faces and it accuses no one. It sees facts — two bodies and a single door opening, someone who stays, a bag that does not move — and hands them over with the camera and the minute. The one who concludes is a person.

Without IRIS this is a lot of branches recording and nobody watching them.
AI-generated image
What is at stake in a branch network
A branch network is a lot of cameras and very few people to watch them, and those few already have their own work in front of them. The ATM lobby stays open all night with nobody inside. The machine is opened for servicing and the street keeps walking past. Cash crosses eight metres of pavement. An operation that takes two people sometimes ends up being done by one, and that only shows at the cash count. And when a complaint arrives about something that happened at eight o'clock on a Tuesday, somebody spends the afternoon going through hours of video from four cameras.
IRIS watches for you. It analyses every camera the branch already has, all at once, and only interrupts when a rule your own team set is met: two people through a door on a single authorisation, a restricted door left standing open, an operation that takes two people with one person in the room, someone who has spent half an hour in the ATM lobby, a machine standing open while people come closer, a cash box leaving the secure area, a person on the floor, a bag left on its own. The alert arrives with the camera, the time and the video, while it is happening. And it identifies nobody: it sees two bodies and one door opening, it does not say who anyone is or what they came for. The one who concludes is a person.
And where there is nothing to alert about, IRIS measures. How long the queue is at eleven and how long at six is what tells you how many desks you need at each hour, and until now that got counted by hand or not at all. Measuring is not set up by talking: it is set up by drawing a line or a zone on the image, because a line cannot be described in a sentence, it has to be drawn. That is why a Tuesday's number and a Friday's are measured the same way and can be audited. And here is something that does not happen in other sectors: branches look like one another. The rule is set once and it applies to the whole network.
The real scale
Thirteen cameras here. A whole network of branches that look alike.
This demo shows thirteen cameras because thirteen fit on a screen. A branch network does not. Here the axis is not the kilometres of a road or the passengers of a station: it is the branches. Lobby, counter, restricted door, counting room, secure area, the machines facing the street and the pavement in front. And every branch looks like the next one: the same lobby, the same door, the same machine. That is why a rule is worth more here than in any other sector. It is set once, tried in one branch and applied to the rest without starting over. The day a new branch opens, it already knows how to watch.
Examples
These thirteen are just examples.
Eleven raise an alert when something happens, one alerts nobody — it only counts — and one shows how you find the exact moment afterwards. There is the door and the counting room, the cash and the secure area, the ATM lobby at night and the people coming in from the street. None of them says that anyone is going to do anything: they say what is happening, where and at what time. The one who concludes is a person. These are not the thirteen things IRIS can do in a branch; they are the ones that fit on a screen. Open any of them and you will see the image as it is, and then with what IRIS understands on top.
Simulation · how IRIS works
No extra alarms here. Just what needs looking at.
Not a recorded video, not an explanatory drawing: a summary of the real screen, with a different case every time it starts. There is a lot more behind it.
- Four cameras at once
- Each understands its own view
- The numbers climb by themselves
Now think about your own branches
Imagine knowing all of it while the branch is still open.
The two who went through the restricted door on a single authorisation at ten past nine on Tuesday. The door left standing open for twenty minutes with somebody waiting outside. The counting room with one person in an operation that takes two. The one who had spent half an hour in the ATM lobby at three in the morning. The machine that stood open for fourteen minutes while two people came closer. The cash box that left the secure area at ten at night, and is now on record with its time. The customer who fell by the entrance and took four minutes to get up. And the eleven o'clock queue, twice as long as the six o'clock one.
All of that is happening in front of cameras the branches already have up. And right now nobody knows: there are three people working in the branch and not one of them can stand watching a screen. IRIS watches them all at once and says so while it is happening. And when the complaint arrives two weeks from now, the moment is no longer hunted for over hours: you describe it in one sentence and it comes up.
Next step
The building site
Who is watching when the load swings over somebody?
11 cases: 9 raise an alert when something happens and 2 only measure. You see each image as it is, and then with what IRIS understands on top.
The prison
Who is watching the yard screen when the fight starts?
17 cases: 16 raise an alert when something happens and 1 only measure. You see each image as it is, and then with what IRIS understands on top.
And the best part
By the eighth hour of a shiftnobody watches the way they did in the first. That is not a flaw.
It is being human. A camera has no eighth hour: at four in the morning it understands what it understood at noon, and it speaks up just the same.
Tell us your problem and we will say whether IRIS understands it, or not yet.
We reply the same working day.












