CAM-538
Eighteen hundred walk past and two hundred stop
Dusk on the shopping street. On the right, a lit window under a dark awning, with mannequins and bags. Two people are standing still in front of the glass and a third has stopped a little earlier. People keep moving along the street: a couple with shopping bags walks past without looking, a passer-by is out in the middle, and further back the rest of the flow comes down. IRIS counts how many walk past and how many stop, and how long each stop lasts.
Turn on IRIS Vision
AI-generated image
What IRIS understands
What it recognises in this scene
- lit shop window
- shop awning
- two people stopped at the window
- person looking at the window
- second window on the street
- couple with bags walking straight past
- flow of people further down the street
- passer-by crossing the walkway
What this camera gives back
Nobody gets an alert here. Here you get numbers.
- 1 840 / hPeople per hour walking past
- 210 / hPeople per hour who stop
- 38 sAverage time stopped
- 3 min 10 sLongest stop this hour
- 19:00 – 20:00Busiest hour for footfall
- 19:40Last updated
The figures on this screen are an example from the illustration, not any client's measurement.
How the counting works
Counting is not something you ask for. You draw a line.
There is a line drawn across the walking axis: crossing it means you walked past. And there is a zone painted over the pavement next to the glass: entering it and standing still for a few seconds means you stopped to look. Asking is for finding; counting is for measuring, and they are not done the same way. A line cannot be described in a sentence: it has to be traced onto the image, and that is why the number is repeatable and auditable, and why this Tuesday compares with last Tuesday.
- line across the walking axis · Counting line
- pavement zone in front of the window · Dwell zone
And also, one rule
You decide once. IRIS applies it always.
If
in a busy hour almost nobody stops in front of the window
Then
- send the figure to the shop manager with the hour and the footfall there was
- save a still of the window at that hour so it can be seen exactly as it was
- leave the comparison with the same hour a week earlier
The instruction that set it up«Tell me every day how many people walked past the window and how many stopped to look at it.»
What changes
What changes for operations, for asset management and for security
A shop manager knows what they sell, but almost never knows how many people walked past without coming in. That is the number that changes the conversation: it says whether the window works, whether the awning hides it, whether the eight o'clock lighting does anything, and whether the street has footfall at seven rather than at five. For the centre, it is the figure that organises the street and defends a rent. And here is the limit, plainly: IRIS sees who walks past and who stops in front of the glass. Whether that person went in and whether they bought anything is told by the retailer's own till data, not by the camera.
What people usually ask
Does IRIS know whether the person who stopped ended up buying?
No, and we are not going to pretend otherwise. This camera sees who walks past and who stops in front of the glass, and measures how long that stop lasts. Whether that person went in and whether they bought anything comes from the retailer's own till or door data. Put side by side, the two numbers explain a great deal; on their own, each tells half the story. What we will not do is credit the camera with a figure it does not have.
How does it tell someone stopped from someone walking slowly?
By the time spent inside the drawn zone. A threshold is set — a handful of seconds — and anyone inside for longer counts as a stop. The centre sets that threshold, not us, and it is kept the same all year so weeks can be compared. Someone walking through the zone is in and out before the threshold and does not count. Changing the threshold halfway through the year breaks the series, so it is fixed once and written down.
Does the camera need to see faces?
No, and it is better if it does not. To count walk-pasts and stops, the silhouette and the position on the ground are enough. IRIS does not recognise faces, does not compare them with any database and does not store who anyone is. The figure comes out just as well with the camera where it belongs: four metres up, angled down, which is how it is actually mounted on a street.
Other cameras
- CAM-535 · How many people come in through each door, and when
- CAM-536 · An empty corridor twenty metres from a packed wing
- CAM-537 · Before, during and after: what the campaign actually brought in
- CAM-539 · The ground floor packed and the upper walkway deserted
- CAM-540 · How many get out of the car and which door they end up using
- CAM-541 · The queue is not moving and there are barely any free tables left
The real interface, step by step
We won't just tell you. We'll show you.
It is a summary of the real screen: it plays by itself, step by step, and shows only the path of one case. The actual tool has far more than fits in here.
- The alert fires
- You open it
- You confirm or dismiss
Next step
What does an airport see before the queue forms?
14 cases: 7 raise an alert when something happens and 7 only measure. You see each image as it is, and then with what IRIS understands on top.
What can a camera see without pointing at anyone?
13 cases: 9 raise an alert when something happens and 4 only measure. You see each image as it is, and then with what IRIS understands on top.
And the best part
There are nights of thick fogwhen the image gives you nothing. We do not dress that up.
Backlight, heavy rain, a street lamp that has blown. IRIS works with what the camera puts in front of it; when there is no image, it says so instead of inventing an answer.
Tell us your problem and we will say whether IRIS understands it, or not yet.
We reply the same working day.