Media & Enterprise
Is this real? The question turns up as a live feed, a file somebody uploads, a meeting invitation and a page open in a browser. FaceOff covers all of those routes and reports the same way for every one of them, so a team learns to read one result and can act on any of them
Overview
Check video from any source, whether it arrives as a live feed, a file someone uploads or a page open in a browser.
Most organisations never decided to handle video from many sources. It simply happened. A camera network was already there, then files started arriving by email and portal, then meetings moved onto video, and eventually anything anybody needed to judge turned up as a link in a browser.
The usual response is to solve each route separately, and that is where the gap opens. Different tools answer in different formats, teams learn one and not the others, and material that would have been checked on its usual route goes unchecked because it came by another. Anybody trying to get something past you only has to choose the route with the least attention on it.
The checks below cover those routes and report in the same language. A result from an upload reads like a result from a live call, so a team that has learned one already knows how to read the next, and coverage stops depending on which door the video came through.
What you deploy
These are the products organisations deploy to cover every route video takes, from a file sitting in an inbox to a page on screen to a call already in progress.
Key capabilities
What this sector asks for most often. Each one runs at the moment a decision is being made, and each returns a result a reviewer can explain to somebody else.
Video and images arriving as uploads, attachments or portal submissions are checked for face swaps, generated footage and editing, and the finding points to the parts worth reviewing rather than leaving one verdict stretched across the whole file.
The person on a live video call is confirmed as a real participant during the call, across the meeting platforms your teams already use, so a warning lands before a decision is made rather than after everybody has hung up.
A finding from an upload, from a browser page and from a live call is written the same way, so a team learns to read one and can act on all three. Consistency is what stops a check being skipped simply because it looked unfamiliar.
Protection that follows only the expected path leaves the unexpected one open, and that is the one that gets used. Working across live, uploaded and on screen video means a check does not depend on the sender choosing a convenient way to send something.
Media met online can be checked where it is being viewed, with a result arriving while the content is still on screen. That covers the most common situation of all, which is somebody looking at something now and needing to know whether to believe it.
No finding arrives on its own. Each carries what was noticed and where, so it can be acted on now, reviewed by somebody else later and explained to an auditor, a client or a court without anybody having to go back to the file and start again.
Where it is used
Every entry names a product on this site and describes it the way its own page does.
DeepVideoGuard checks recorded video for face swaps and generated footage, and DeepImageGuard does the same for still images.
DeepBrowserGuard checks the images and video people meet online, while they are on screen.
DeepMeetGuard confirms the person on a video call is real, while the call is happening.
Common questions
No. They report in the same language, which is what lets one team read a result from an upload, a live call and a browser page without having to learn three separate things.
Yes. DeepBrowserGuard checks images and video people meet online while they are being viewed and returns a result while the content is still open, rather than afterwards.
Yes. DeepMeetGuard confirms the person on a video call is real while the call is running, across Zoom, Teams and Meet, so a warning arrives in time to affect what gets decided on the call.
DeepVideoGuard checks recorded video for face swaps and generated footage and points to the moments that need review, and DeepImageGuard does the same for still images and highlights the areas that look manipulated.
Analysis can run inside your own environment where material is sensitive or must stay in the country. Checking video that people trusted you with should not require handing it to somebody else first.
Yes. DeepBrowserGuard rolls out to a workforce as a browser extension, which is how a check reaches the people who actually meet questionable media in the course of their ordinary work.
The finding points to the moments worth watching rather than returning a single verdict across the whole file, which is what makes a long recording reviewable in the time a person genuinely has.
No. It adds a check to the routes video already takes into your organisation and leaves your recording, meeting and case systems exactly as they are. Findings come back in a form those processes can carry.
Yes. Any method of detection can be mistaken in either direction, human judgement included, so a result is an evidence based assessment rather than a guarantee. The reasoning comes with it so that it can be examined rather than simply believed.
Bring your own footage. We will run it through the Adaptive Cognito Engine and walk you through the findings.
