Intelligent video analytics software: questions we get asked

Your recorder mostly does pixel difference detection with a person filter bolted on top, which is why rain, headlights and a branch set it off. This classifies what the moving object is before anything alerts. To be fair to your recorder: for one camera on a door in good light, its built-in detection is genuinely fine. It falls apart at four cameras outdoors at night, which is where this starts paying for itself.

Accurate enough that the alert channel becomes readable, and not accurate enough to be left alone with a fence line. It struggles with heavy rain directly on the dome, near total darkness with no working infrared, and a subject that is mostly out of frame. We would rather name those than quote a percentage we cannot support. Every event carries a confidence value, so you can see when the model was unsure.

The event stays in the timeline marked unclassified, with its real motion data intact, and the reason is shown. We never substitute a guessed label, because a wrong label is worse than an honest gap.

Yes, and this is usually the single biggest reduction on any site. Draw an exclude polygon over the road and events whose centre of motion falls inside it are dropped before anything alerts. You can watch that happen live in the demo.

Only if you ask it to. Class filters and schedules mean people during opening hours can go to the timeline and not to a phone, while the same camera pages someone at 03:00.

No, and it never will. It classifies categories: person, vehicle, animal, package, other motion. There are no face templates, no face matching, no gait analysis and no re-identification across cameras or across days. This is the shape of the product, not a setting, which is what makes it defensible under the Illinois biometric law and its equivalents in Texas and Washington.

In the demo on this site, nowhere: your clip is decoded by your own browser and is never uploaded. Only small still crops around each detected event are sent to be classified. On a live deployment the detection runs next to your cameras and you choose the retention window, with configurable deletion.

Yes, from the Operate tier. A security review is included from Multi-Site, and a custom MSA is an Enterprise term. The sub-processor list is published.

Roles start on Operate with Viewer and Operator, and Multi-Site adds Admin, custom roles and per site scoping so a contractor sees one yard. Every view, zone change and export is in the audit log, for 30 days on Operate and a year on Multi-Site.

Almost certainly, and you probably already do. Camera law is local and we are not your lawyers. The fact that most often changes the answer is that this product collects no biometric identifiers, which is the trigger for the strictest statutes.

Operate, $149 a month, covers 15 cameras, so twelve fits with room to grow. That is $9.93 per camera per month, or $119 a month billed annually, which is $1,428 a year.

No. Every tier is paid. The demo is the free part and it is complete and ungated. We do not offer a free production tier, because a security tool nobody pays for is a security tool nobody checks.

You confirm your email with a 6 digit code and sign in with that address. Pick the tier that matches your camera count, pay by card in Stripe's own checkout, and the console opens: cameras, exclude zones, the classified event timeline and alerts. Nothing is charged until you pick a tier, and the demo on this site stays open and stays free.

Yes. Watch, Operate and Multi-Site are bought by card on the pricing page, monthly or annually, and cancelled from your own console. Only Enterprise is quoted in writing.

Camera health monitoring and pan, tilt and zoom control. An enterprise platform does both and this does not: it reads what the camera sees and classifies it. If that is the job, your existing NVR keeps doing it alongside.