Skip to content

WestPoint · OpenEye

The cloud manages. The drive at the site carries on recording.

OpenEye is the middle road: the video is recorded on a unit inside your installation, as always, and what goes up to the cloud is the administration — users, updates, the status of every site and getting in from outside. For a lot of people it is the split they wanted and did not know existed.

The problem was not where it records. It was who looks after it.

When somebody starts thinking about moving to the cloud, it is almost never about the cloud. It is about not having to look after twenty machines.

A recorder at each site works well for the first year. What does not work is the rest: different versions at every site, drives filling up without warning, passwords going round by email, a camera that stopped recording in March and is found out in October.

What this platform proposes is separating two things that always go together and do not have to: the video stays recording on the unit at the site and the management goes up. From a web page you see the status of every site, you update the units, you give and take away users and you get in to watch without opening a port on any router.

The problem it solves, in one sentence: having twenty or a hundred sites with local recording and finding out what is going on in them without going there. And without the upload bill of a platform that sends all the video out, because here the video does not leave.

How it is built.

It is a genuinely hybrid architecture, not a cloud with a consolation drive.

  • The unit at the site records and stores, and it has to be sized just as always: channels, resolution, real days of retention and the drive space that asks for. The cloud does not remove that calculation, and anyone who gets it wrong will have the same problem they had before.
  • The cloud is the management plane: centralised users and permissions, health status of every unit and every camera, updates pushed without going to the site, remote configuration and access from outside. That is the product, and that is what you pay for.
  • Cameras: it works with third-party equipment over the standard protocols and there are also cameras of its own. The difference from a closed-hardware platform shows up in the bill for a refresh — if what is installed is good enough, it stays.
  • Redundancy: what protects the video is the unit at the site, with mirrored drives and the option of sending a copy of what matters off site. And that unit is still something that can fail and that can be carried away, so where it goes is a project decision and not a stores decision.
  • Access control and intruder detection are integrated, not unified: events that mark the video and alerts that travel wherever they have to go. There is an API for pulling status, events and video out to other systems — it is not the most open platform on this list, and for what it is normally used for it is enough.

The five questions about the cloud.

With this split, three have short answers. And the fifth has to be asked more carefully than anywhere else.

Where the recordings live: in your installation, on your drive, and that settles half the conversation with the legal department. What is outside is the users, the permissions, the status of the units and the route you come in through to watch — and the video is seen over that route, so by design the supplier has an administration door. Minimum permissions, two-factor and a visible access log: the same as with the others.

If the line goes down everything carries on recording at the site, with nothing lost, because the recorder does not need the cloud to work. What goes down is the management and the access from outside. For places with bad connections, that is the main argument.

The cost of uploading video: small and predictable, because the video is not uploaded. Status data and thumbnails go up, and video when somebody watches or when a copy of something specific has been configured. A site with a modest line copes with this, which with a lot of small sites is no small detail.

RGPD and international transfers: there is less surface than with a pure cloud because the images are not stored outside, and there is still processing —accounts, access logs, metadata and the ability to view through the service—. Ask for the usual: a processor contract, guarantees for the transfer, the region in writing and information given to the staff.

And the dependency, which here has to be asked about more carefully: what happens to the unit at the site if you leave the service? Does it carry on recording? Can you get in to watch it and to export locally, without the cloud, and with which functions? You need that answer from the manufacturer, in writing, before you buy. On the good side: the cameras are standard and they stay.

The subscription, and how it grows.

The recording unit is bought and the management service is paid for by camera and by term: the console, the updates, the support, the health status and the access from outside. If there is an archive off site, it goes separately and is chosen camera by camera.

As you grow the cost is linear and boring, which is a compliment. What gets neglected is the hardware: the unit copes with the cameras it was bought for, and the day that site goes from sixteen to twenty-four you have to work out drive space and capacity again. And the honest comparison with a pure cloud: here you pay for a unit per site and less line; there you pay for no unit and you pay for upload every month. Which one works out cheaper is worked out with the numbers from your case.

Who it fits, and who it does not.

Here the border is sharp and it is nearly always set by the connection at the sites.

It fits if…

You have a lot of sites with cameras already installed, nobody technical in them and ordinary or bad lines. If you need long, complete retention —weeks or months— that will not fit in a pure cloud because of the upload. If the legal department or the specification demand that the images stay on your premises. And if what you are after is to stop looking after recorders without changing the recording model.

It does not fit if…

You want to take all the hardware out of the site: here there is still a machine, with its warranty and its end of life. Nor if you need a large control room with unified platform functions, nor if the project asks for a lot of bespoke integration over the API. And it is worth checking the manufacturer's presence and support in your country before committing a hundred sites.

Analytics: what comes with it, what you can add and where the ceiling is.

As standard, what you need to operate: search by movement and by time, alerts by zone and by time of day, and a watch on system health that in a chain is worth its weight in gold — a camera not recording, a drive running out, a unit not answering, retention that has dropped below what was agreed.

What you add: the analytics that come inside the cameras, which is the natural route here. The camera detects and the system picks up the event, marks it and raises an alert. Number plates, line crossing, counting: you choose the camera by the analytics it has to do, and that gets decided during the project and not the following year.

The ceiling of what is included, plainly: it detects that something has moved or crossed a line, and it does not understand the scene. In a warehouse with forklifts, movement analytics raise a hundred alerts a day about normal things and within a week nobody looks at the alerts. It is not a fault of this platform: it is where its work stops.

And the disclosure that is due: in the same house there is IRIS Neural, the NVMS from Infinity Neural —the other company in the group—, so we recommend it with an interest of our own and we say so first. In many installations they live together: the VMS governs cameras, recording and retention, and IRIS brings the understanding of the scene. Here living together is straightforward, because the cameras are standard and both systems can take the image inside the site, at full quality and without loading the line.

What we do.

A hybrid platform means work on both sides: the classic recording calculation and the service part.

  • The sizing of each site the old-fashioned way, because it is still needed here: channels, resolution, frames, real days of retention and the drive space that asks for, with room for the cameras that will come later.
  • The decision about what goes up and what does not: status always, video when somebody watches, and a copy off site only of what has to be kept even if the unit disappears.
  • The implementation: units, a video network separate from the company's, cameras checked by day and by night, permissions by role, two-factor, and the health status configured so that it alerts somebody with a name and not a generic mailbox.
  • What we ask the manufacturer for in writing before the customer signs: what can be done locally without the cloud, what happens to the unit if the service is dropped, the region the data sits in and the processing contract, and support and replacements in your country.
  • The migration and the maintenance: here the cameras often stay and what changes is the recorder. They run side by side for a few weeks, whatever has to be kept is exported, and after that fields of view, drives, versions, retention measured —not estimated— and an annual review of the camera and subscription inventory.

Who operates it after we leave.

This split is the one best understood inside a company with IT, because it looks like the way they already manage the rest of their equipment.

  • IT: accounts, two-factor, the network at each site, the update window and keeping an eye on status. Unlike a pure cloud, here there are physical machines with a warranty, an end of life and drives that get replaced, and that goes into the IT inventory like any other server.
  • Security, or whoever is in charge of each site: watching, searching, bookmarking and exporting. It is done from the browser, it is learned in a morning and it lets you give each manager access to their own cameras only.
  • The border that has to be made clear: IT watches the health status, but the person who decides that a camera which has not recorded for two weeks is urgent is not IT. You have to write down what gets watched, how often and who it gets escalated to.
  • What it demands of the integrator is knowing about recording and about networks, and also being able to explain the model. A lot of customers arrive believing that «cloud» means the video is outside, and here it is the other way round; if the integrator does not explain that difference, the customer signs expecting something else.
  • Training: a morning for day-to-day use, with a sheet per site. And a session with IT on the local unit —drives, replacements, updates, what to do when a site does not answer— and on what can be done with no connection. That last part gets forgotten and it is the one you need on the day the line goes.
  • Documentation: an inventory of units per site with their warranty and their drive, a sheet per camera with what it sees and the real measured retention, a map of profiles, what goes up to the cloud and what does not, a subscription table with dates, and the administrator accounts in the customer's name.

Get started

Do you want to stop looking after recorders without sending the video out?

That is the split this platform makes, and it does not suit everybody. If you tell us how many sites you have, what cameras are there, how many days of recording you need and what line the worst of them has, we put the calculation next to the one for a pure cloud and you decide with both numbers in front of you.