CAM-80
The shop doorway
The doorway of a clothing shop seen from inside, looking out at the street: both glass doors wide open, a dark mat on the threshold and afternoon sun falling across the tiled floor. A woman with a bag in her hand is coming in through the middle, another is heading out towards the street and a customer is walking to the door from inside; along the sides, mannequins, rails of coloured shirts and a display table with handbags. Outside on the pavement, people walk past without coming in. IRIS counts those who cross the doorway, those who pass without entering and how long each one stays, and estimates the age-band and sex split of the group; it does not identify anyone.
Turn on IRIS Vision
AI-generated image
What IRIS understands
What it recognises in this scene
- open doorway
- coming in
- going out
- shopper inside
- walking past
What this camera gives back
Nobody gets an alert here. Here you get numbers.
- 412+38Entries today
- 3 140Walk past
- 13 / 100Come in per 100 who pass
- 14Inside right now
- 11 min+2 minAverage time inside
- 18:00 – 19:00Peak hour
The figures on this screen are an example from the illustration, not any client's measurement.
IRIS incorporates NeuralPax, a separate licence certified by the Spanish Metrology Centre for people counting, occupancy and crowd-flow control. That the figure is right is guaranteed by a third party, not by us, and that is why it holds up against the regulations in force. The certification covers that measurement — counting people — not the whole product.
The series, not one figure
One number says nothing. A curve does.
Who comes in and who walks past
- Come in
- Pass without coming in
| Slot | Come in | Pass without coming in |
|---|---|---|
| 06 | 12 | 140 |
| 08 | 38 | 320 |
| 10 | 61 | 470 |
| 12 | 74 | 560 |
| 14 | 52 | 430 |
| 16 | 44 | 380 |
| 18 | 88 | 690 |
| 20 | 63 | 520 |
| 22 | 19 | 180 |
Estimated split by age and sex
- Women
- Men
| Slot | Women | Men |
|---|---|---|
| Under 25 | 14% | 9% |
| 25 to 40 | 22% | 15% |
| 40 to 60 | 17% | 11% |
| Over 60 | 8% | 4% |
How long the people who come in stay
- Minutes inside
| Slot | Minutes inside |
|---|---|
| 06 | 6 min |
| 08 | 8 min |
| 10 | 11 min |
| 12 | 13 min |
| 14 | 9 min |
| 16 | 10 min |
| 18 | 15 min |
| 20 | 12 min |
| 22 | 5 min |
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 are two lines and one area here, and all three are drawn onto the image. The first line crosses the threshold: whoever passes it inwards is an entry, whoever passes it outwards is an exit, and the difference gives you how many people are in the shop at any moment. The second line runs along the pavement, in front of the window, and counts those who walk past without coming in. The area covers the mat and the first few metres of the floor, and measures how long the people who come in stay. None of this is asked for in words: asking in plain language is for finding things and for describing a situation, but counting needs to know exactly where the line runs and in which direction, and only a stroke drawn onto the image settles that. That is why Tuesday's figure and Saturday's are measured the same way, and why the number is repeatable and auditable.
- doorway line · Counting line
- pavement line · Counting line
- entrance area · Dwell zone
What changes
What changes for whoever runs the store
The store manager decides Saturday's shift by how the supervisor feels about it, and the window display gets argued over with opinions. The door line gives him how many people really come in and at what hour, and the pavement line tells him how many walked past and did not come in: if three thousand go past and four hundred come in, the conversation stops being about what sells inside and becomes about the window. The entrance area adds how long the people who come in stay, which is what separates a shop people walk through from a shop people look around in. The split by age band and sex is a group estimate with its own margin: it is there to decide the range, the window and the shift roster, and it is not for anything to do with one particular person. And in case anyone asks, which they will: no face is recognised here and nobody's identity is stored. People counting is type-approved by the Spanish Metrology Centre, so the figure has been verified by an independent third party and not by us; the numbers on the illustration are an example, not any shop's measurement.
What people usually ask
Does IRIS know the age or sex of the people who come in?
No. IRIS estimates which age band and which category a person resembles from their appearance and their silhouette, and that is a group estimate with its own margin, not a fact about a person: it does not know anyone's age. It recognises no faces, compares them against no database, stores nobody's identity, does not follow a person from one visit to the next or from one shop to another, and is not cross-referenced with the loyalty scheme or the receipt. What comes out is the split of a group — "of those who came in this afternoon, this many in every hundred looked like this age band" — and no record of any individual comes out at all. It is there to decide the range, the window and the shift roster, and it must not be used for anything to do with one particular person. In a real deployment this also needs its own legal basis and a visible sign at the door, in all eight markets where we work; that is signed off by the organisation installing the camera, not by us.
What is the point of counting people who walk past and do not come in?
To know whether the problem is inside or in the window. The door line says how many came in; the pavement line says how many had the chance to. If three thousand walk past and four hundred come in, the conversation stops being about what sells inside: the window, the sign and the opening hour become decisions with a number behind them, and the week of the new display can be compared with the week before. Without that second line you only know what you already knew, that it was a quiet day, but not whether few people came in or whether nobody was walking down the street that day.
Who guarantees that the people count is correct?
An independent third party. IRIS people counting is type-approved by the Spanish Metrology Centre, meaning the measurement has been verified by an outside body and not by us, and at a shop doorway — which is where occupancy gets counted — that is exactly the guarantee you need. The scope is worth stating precisely, because stretching it would be a lie: that approval covers people counting, not the whole product, and it does not cover the age-band or sex estimate, which is a different thing and is not approved. On top of that, the line is drawn onto the image: anyone can see where it is and why the figure came out as it did.
A different case every time
Watch what happens. Then watch what IRIS does with it.
Not a recorded video and not a drawing: it is a summary of the real tool, with a different case each time round. The whole tool does not fit in a box.
- Checks the whole system
- Flags what is failing
- Puts it on one screen
Next step
What if the outage does not start with an attack?
13 cases, and in every one of them IRIS raises an alert when something happens. You see each image as it is, and then with what it understands on top.
And who watches the cameras?
0 cases, and in every one of them IRIS raises an alert when something happens. You see each image as it is, and then with what it understands on top.
And the best part
You have cameras to spare.What you run out of is eyes to watch them.
One person can follow two or three screens. Nobody can follow forty. IRIS watches them all and speaks up when someone needs to decide.
Tell us your problem and we will say whether IRIS understands it, or not yet.
We reply the same working day.