Camera intelligence · early warning · events, not footage

Rules you write in words.

Point it at the cameras you already have and write the rule in a sentence — “tell me when a pallet blocks the fire exit.” When it happens you get an event: the time, a line explaining why, and a still with the faces hidden. Not a recording. It looks for what comes before an accident (a person close to a moving machine, a gate left open, a rack overloaded, a fire starting) so someone can act while there is still time.

Blocked exitSpillNo helmetAfter-hoursWrong bayLeft running
Why it is different

Three things the usual video analytics does not do

Anyone can bolt a model onto a camera. The difference is in the three things below — and the third is the one a serious buyer asks about first.

A rule is a sentence

A new rule does not need labelled footage and a trained detector. It needs a sentence. You add or change a rule the same afternoon your process changes — not in the next project.

We keep the event, not the footage

The box stores the moment, the reason and a still with faces and plates blurred. Not a recording of your site. That is a selling point and a legal stance at once: you hold evidence of an event, not an archive of people.

We tell you how often it is wrong

The false-alarm rate per rule, measured on your site before you roll out. The same evaluation discipline we apply to text models, turned on pictures.

How it works

A cheap gate first, the model only on what matters

Running a vision-language model on every frame is not affordable, so it never sees every frame. A cheap gate watches for motion and scene change; only the candidates reach the model, which judges them against your rules; a reasoning step suppresses repeats and decides what to escalate.

CameraEvery frame, all day.
GateMotion, scene change. Most stop here.
ModelOnly the candidates, judged against your rules.
EventWhat actually matters.

Cameras

Your existing camerasRTSP · no new wiring
Cheap gatemotion · scene change

The box, on your site

Vision-language modelreads the candidate against your rules
is the rule met?whyconfidencededupeescalate
Stream watched, never recorded

Event

Time and the rule that fired
A line of reasoning
A still with faces and plates blurred
Alert: e-mail, Telegram or a webhook
The gate is the economicsThe model runs on a handful of candidates per camera, not on a video stream, so the cost per camera stays sane.
The reasoning step is the hard partA raw “is this notable?” fires too often. Suppressing repeats and escalating only what matters keeps you from being woken for the same thing twice.
On your termsThe box runs on your site or on hardware you rent in the EU. The events are yours.
What you can ask

Rules written the way you would say them

Each of these is one sentence to the system. You keep the ones that matter to you and change them whenever the site does.

safetyTell me when a pallet or anything else blocks a fire exit.
PPETell me when someone in the yellow zone has no helmet or vest.
spillTell me when there is liquid on the floor in the aisle.
accessTell me when a person is in the plant room after working hours.
logisticsTell me when a truck is in the wrong loading bay.
equipmentTell me when a machine is left running with nobody present.
Event, not footage

We hold evidence of an event, not an archive of people

Monitoring a workplace is specifically regulated in Slovenia (ZVOP-2) and under the GDPR, and an archive of recordings is the heavy part. Perception Watch avoids the archive: it keeps the event and a still in which faces are pixelated (licence plates too, when you switch that on). The original is not kept, so it cannot be recovered.

09:42:17face ×2 · pixelated
09:42:17plate ×1 · pixelated
09:42:17recording · not kept
09:42:18event + still · kept
No recording keptThe stream is watched, not stored. What persists is the event and a redacted still.
Faces hidden firstFaces are pixelated on the way in, before anything else sees the frame; plates too when switched on. Pixelation, not blur, so it cannot be undone.
Nothing to restoreBecause the original frame is never stored, there is no archive to leak, subpoena or misuse. The event and its redacted still are what you keep.
A stance you can defendTo an inspector or a works council, “we keep the event, not the footage” is the difference between “interesting” and “allowed”.
Where it fits

Many cameras, a few rules that keep changing

The easiest first customer is not a factory with a built-in VMS and a two-year procurement cycle. It is somewhere with a lot of cameras and a handful of rules that change often.

A good fit

  • Utilities and municipal yards
  • Construction sites and depots
  • Logistics terminals and warehouses
  • Building and facility management
  • Agriculture and outdoor sites

Not what this is for

  • Exact counting of items through a point
  • Precise localisation or measurement
  • Reading a licence plate character by character
  • Anything where a classic detector, trained once, is simply the right tool — we will say so
What we deliver

Two ways to get it

Either way you write the rules in sentences and the same events reach you. The difference is who looks after the equipment.

The service on your cameras

  • We connect to the cameras you already have
  • We set up the rules with you and tune them on your own site
  • You keep the cameras and the network as they are

A complete installation

  • We also provide and set up the equipment, cameras included where you need more of them
  • One partner for the whole chain, from the picture to the alert
  • Same rules, same events, same privacy stance
Honest limits

An extra pair of eyes, not a safety system

It assists peoplePerception Watch is not a certified safety system. It does not replace guards, interlocks, fences or procedures.
It warns early, it does not promiseIt notices conditions that come before an accident, such as a person close to a moving machine or a gate left open. Rules about something still developing need the most careful, slowest mode, and we have so far tried them on generated test clips, not yet on a customer site.
False alarms existWe tune each rule with you and measure how often it is wrong before you rely on it. Some rules will need a better camera angle, or will not suit this tool at all, and we say so.
The proof · early access

Four weeks, one site, three rules, a number

Perception Watch is in early access and we are looking for the first sites. We do not ask you to believe a brochure — we measure on your cameras.

1 · Scoping callYou describe the site and the three rules that would matter most. We say honestly whether they suit this tool or a classic detector.
2 · The box goes inWe connect to the cameras you already have. No new wiring, no cloud account, no recording.
3 · We measureFour weeks later you have the false-alarm rate per rule on your own footage — the number that decides whether it is worth rolling out.
4 · You decideIf the number works, we roll out more cameras and more rules. If it does not, you learned that cheaply and early.
FAQ

Questions buyers ask

How is this different from the video analytics my camera vendor sells?

Two ways. A new rule is a sentence, not a trained detector with labelled data, so you can add 'tell me when a pallet blocks the fire exit' the same afternoon. And we keep the event, not the footage: the box stores the moment, a line of reasoning and a still with faces and plates blurred, not a recording of your site.

Does the video leave the building?

The stream is read on the box and is never recorded. Faces are hidden before anything else sees a frame. What is stored is the event and a still with the faces hidden; only if you switch alerts on does that event, with its still, go to where you point them (e-mail, Telegram or a webhook).

Can it count things or read a licence plate exactly?

For exact counting and precise localisation a classic detector is better, and we will tell you so. Perception Watch is for 'tell me when something happens that I did not list in advance' — a blocked exit, a spill, a person in a zone after hours.

How do I know it is not crying wolf?

That is the part we measure. A four-week paid proof on one of your sites with three of your rules gives you the false-alarm rate per rule before you roll anything out. If it is above a workable threshold, you find out early and cheaply.

What do I need on site?

The cameras you already have and one box we supply or you rent. No new wiring. Rules are written in a sentence and edited whenever your process changes.