CAM-127
Hallway closed off by visitors
An admissions hallway with forty-odd visitors closed in a ring, wall to wall. In the middle of the ring there’s a cart with someone lying on it and two professionals in scrubs leaning over; a third, in a white coat, is caught at the edge and can’t get out. On the left, an open bay with a free bed and the monitor on: that’s where you have to get to. On the right, the desk with two people behind it. There’s no way through: right now a second cart can’t get in. IRIS counts people and sees where they are; it does not know who any of them are.

Turn on IRIS Vision
AI-generated image
What IRIS understands
What it recognizes in this scene
- hallway with no way through
- cart being attended
- professional boxed in
- free bay
- nurses’ station
What this camera measures
An alert says something’s happening. The reading says how much.
- 46People in the hallway
- 31+14Around the cart
- 5 minClosed off for
- 18Limit for this stretch
- 4Professionals inside the ring
- 18:26Time of the alert
The figures on this screen are an example from the illustration, not any client’s measurement.
The rule
You decide once. IRIS applies it always.
If
that stretch of hallway holds more people than the hospital has set and the way through stops being clear
Then
- alert the station with the camera and the stretch that has clogged up
- say how many people there are and since when, so you know who to send
- log the time, so the hospital learns what hour that hallway fills up every day
The instruction that set it up«Tell me when the admissions hallway gets so full that a cart can’t get through, and tell me while it is filling up, not once it is already closed off.»
What changes
What changes for whoever runs the unit
Seven in the evening, visiting hours, and the admissions hallway has closed wall to wall. Nobody here’s to blame: people gather where the person they love is, and that isn’t fixed with signs. What’s a problem is that a second cart can’t get through right now and that the family who have just arrived can’t find anyone to ask. With IRIS the control desk knows while the hallway is closing, with the camera, the stretch and how many people are there, and there’s time to open a route. And the figure remains afterward: what time it fills up each day, which is what makes it possible to change how visits are organized instead of pushing people around.
What people usually ask
Does IRIS know who the visitors are?
No. IRIS does not identify anyone: it does not recognize faces, it does not compare them against any database, it does not know who any patient or visitor is, it keeps no file on anyone and it has no access to medical records. Nor does it follow a person from one hallway to another. To know that a cart can’t get down a hallway you only need to count bodies and see where they are, and that’s exactly all it does here.
Is this the same as the room’s capacity?
No. Capacity is a limit on the building, measured at the door: how many people are inside. This measures something else — how tightly packed one specific stretch of hallway is — and they’re two different figures. A lobby can be well under its capacity and still have a hallway no cart can get down. The second figure is the one that stops people working, which is why it’s the one that matters here.
What if a lot of people is normal at that hour?
The limit is set by the hospital, and it can be different per stretch and per time band. During visiting hours you raise the limit for the lobby, where the crowd is expected, and leave the limit for the cart stretch as it is, because that has to stay passable at all times. That’s what keeps the alert meaningful: if it fires every afternoon at the same hour, within a week nobody looks at it.
A different case every time
A new case each time. Always the same screen.
It plays by itself and shows one case from start to finish. It’s a summary: the real screen has far more functions than fit in this box.
- The rule gets built
- It goes to every camera
- Published, it watches alone
- And then it fires
Next step
Banking
Who is watching the cameras of a whole branch network?
13 cases: 12 raise an alert when something happens and one only measures. You see each image as it is, and then with what IRIS understands on top.
Casinos
How long does it take to find a hand played two hours ago?
12 cases: 9 raise an alert when something happens and 3 only measure. You see each image as it is, and then with what IRIS understands on top.
And the best part
We’d rather promise less.Let the first week be the surprise, not the brochure.
What it can’t do gets said on day one, not on the last day. A short list that holds beats a long one that falls apart the moment it is turned on.
Tell us your problem and we’ll say whether IRIS understands it, or not yet.
A person answers, the same working day.