What does IRIS do?
IRIS sees, understands, decides and acts.
Cameras already see the world; IRIS is what makes them understand it.
It is three in the morning. Forty screens are on and one person is watching them. On screen 34, someone climbs the perimeter fence for six seconds. The camera recorded it. The only question is whether anyone was looking at that screen right then. Cameras multiplied. Human attention did not.
IRIS is a system that understands what your cameras see and does something with it. It connects to the ones you already have, watches them all at once and recognises what is in front of them: people, vehicles, objects, zones and how long they have been there. So it stops being ‘there is a person’ and becomes ‘a person has been inside the loading area for eleven seconds while a lorry manoeuvres’.
Your organisation decides once what deserves attention and never has to remember it again. IRIS checks that rule on every camera, at any hour, without getting tired at four in the morning. Detecting is not deciding: an alert says something happened; a decision knows what to do about it.
When a situation matches the rule, IRIS opens the incident, alerts the right person, brings the camera up on screen, records the figure and keeps the evidence. What changes for your team is the whole shift: instead of watching forty screens, they handle the situations that reach them, with the context and the next step attached. Everything comes in. Only what matters gets through.
Sees what is already recording
IRIS connects to the cameras you already have, as long as their video stream is available. Seeing was never the problem: the cameras were already doing that on their own.
Understands what is going on
It recognises people, vehicles, objects, zones and durations, and how they relate to each other. It moves from ‘there is something in the image’ to ‘this is what is going on’.
Decides by your rule
It compares what happens with the rules your organisation has defined. Decide once; IRIS applies it every time.
Acts and writes it down
It opens the incident, alerts, brings up the camera, records the figure and stores the evidence. The team receives a situation ready to be decided on, not an hour of footage.
Does IRIS make decisions for us?
No. Your organisation defines the rules and the limits, and IRIS takes care of applying them over and over across every camera. Situations that call for human judgement reach a person together with the footage, the context and the history, so the decision is made with everything in view.
Do I have to replace my cameras to use IRIS?
Not necessarily. IRIS is designed to make use of existing video infrastructure where the integration and the available streams allow it. A survey beforehand goes camera by camera: which ones work as they are, which are worth re-aiming, and which do not give the image quality the detection you want requires.
How is IRIS different from video analytics?
IRIS differs from video analytics in what it does after detecting. Analytics spots an isolated fact: there is a person, a line was crossed, twelve vehicles went past. IRIS links that fact to the zone, the elapsed time and the context, checks the rule your organisation set, and carries out what should happen next. Detection is the beginning, not the end.
Are detecting and understanding the same thing?
No, and this is one of the most confused points. Looking is three different things. The first FINDS: it knows there is a person there and at exactly which point of the image; it is fast and always gives the same result on the same image, which is why it is the only one you can count with and have a number come out the same twice. The second NAMES: it tells a van from a car, a hard hat from a cap; it works with a list, and what is not on the list it does not see. The third UNDERSTANDS: it reads the whole scene and how things relate — not «a person» and «a car», but «a person on the ground beside a car stopped in the middle of the junction» —, it needs no list and answers a sentence written in your own words, but it is not the one you count with. IRIS carries all three as standard, not as separate add-ons: on the same camera you usually need them at once. That is why we say asking is for searching and counting is for measuring, and that they are not done the same way.
Can IRIS learn to see something that only happens at my site?
Yes. A part made only here, a movement done only in this trade, a defect that only shows up with this material: none of that is on any list, and it is not going to be. A model of your own is trained on images from your cameras and on your cases, and it becomes one more detection inside IRIS, written into a rule like all the others. And that model is yours: it is trained on your images and on the judgement of your people, so it remains your property. It is not reused with another customer, it is not folded into a product sold to your competition, and you do not have to remain a customer to keep using it. What it takes to train it: images of that happening — real ones, not staged — and someone on your team who can say which is right and which is not. If the thing you want to detect cannot be seen in the image, no model will pull it out.
What it does not do
Before you find this out in a demo, we would rather tell you here.
IRIS is not for everyone, and saying so now costs less than finding out halfway through a project. If you have four cameras and one person who takes them all in at a glance, you do not need any of this. IRIS starts to pay off when there are more cameras than eyes.
Nor does IRIS replace your team. It takes on the repetitive part — watching, counting, timing, searching — and leaves the part that calls for judgement to the people who have it. IRIS does not replace human judgement: it stops you spending it where it is not needed.
And it does not guess what matters to you. Someone in your organisation has to say what deserves an alert, in which zones and within what limits, and those rules are tested against real footage before they are left running. That is work at the start of a project, not a switch you flip.
Not for four cameras
With few cameras and one alert person, the problem does not exist. IRIS solves a problem of scale, and that problem only shows up once you can no longer watch everything.
It does not empty the control room
It does the tiring part. Decisions that call for judgement still belong to a person, and they arrive with the footage, the context and the history in front of them.
It does not get it right every time
Some scenes are hard: backlight, fog, heavy rain, an object far away. That is why every rule is tested against real footage from that camera before it is switched on.
It does not install itself
The cameras have to be reviewed, the zones drawn and the first rules written with your team. Nobody plugs this in on a Friday afternoon and goes home.
The real interface, step by step
You set the rule once. IRIS applies it every time.
You will see a summary of the interface, played step by step and hands-free. Each round starts with another case. The complete tool does not fit in a demo.
- The rule gets built
- It goes to every camera
- Published, it watches alone
- And then it fires
See it somewhere else
Who is watching a whole quay at three in the morning?
14 cases: 13 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.
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.
And the best part
Ten cameras or four thousand:not one of them goes unwatched.
The day IRIS starts, they all start: the ten in a school or the four thousand in a whole network. And it only raises its hand when the thing you asked for happens.
Tell us your problem and we will say whether IRIS understands it, or not yet.
We reply the same working day.