Skip to content
Focient

Cafe terrace camera, recorded footage; the person's head is blurred.

Your cameras, as intelligence.

Focient turns the cameras you already have into a record your team can ask questions of.

Your night, as it is today

Every camera records, and almost nothing is watched.

  • Around the clock. Your cameras record every hour of every night.

  • A few operators. At 03:00, a handful of people face a wall of feeds.

  • Finding it. After a call, finding the right recording means scrubbing through hours of video.

Car-park door at night, by Rec Everywhere on Pexels (Pexels License).

The alert

From a rule to an alert

Alert at 02:47:13

Your rules are checked as each event from a live camera is recorded, and a matching alert is sent at that point.

A rule can name a zone, a time in that zone and a schedule.

Operators mark each alert useful or not, and every rule keeps a quality score.

An example alert rule, as the rule editor shows it. It is an event rule for camera 03, the loading dock. It matches events tagged after hours and door, and carries a plain-language check: Confirm the side door is open. The data is fictional.

Screens on this page show fictional data.

An example event, as the event detail shows it: a snapshot of the side door at the loading dock at 02:47:13, marked Verified, the written description that the rule's check read, and a question asking whether the alert was useful, with Useful and Not useful buttons. The data is fictional.

Ask

Ask your cameras a question

Ask in plain language about what your cameras saw, and each answer cites the events and frames it rests on.

Under each answer you see the snapshot of the exact frame the vision model judged.

Example questions

What happened at the loading dock last night?

Focient

At 02:47 the side door at the loading dock opened on CAM 03. No activity was scheduled in that zone. At 02:51 a white van stopped at the dock gate on CAM 01. Times are in UTC.

Sources

  • CAM 03 · 02:47:13

    Side door opened after hours

    #1Verified
  • CAM 01 · 02:51:40

    White van stopped at the gate

    #2Verified

Ranked by how closely this matches your question. Not a probability.

Car park at night from above, by Tom Fisk on Pexels (Pexels License).

Describe

Find it by describing it

Describe what you are looking for in plain words, and Focient searches events and frames for it.

Matches are ranked by how closely they fit the question, and no match is shown as a percentage.

An example search, as the search palette shows it. The description white van stopped at the dock gate after midnight is typed into the search field. Two events and two frames are listed under Events and Frames in rank order, each with its camera and the frames with their time. No score or percentage is shown. The data is fictional.

The case

Turn what you found into a case

Put what you found in a case: a timeline across cameras, annotations and a relationship map of what was seen together.

A case briefing is drafted for an analyst to review, and exports as a Markdown file.

Each case has owner, editor and viewer roles, and only its members can see it.

A case's Timeline screen: the dock door opens and closes, and a white van stops at the gate, each with its time, camera and snapshot. The data is fictional.
The same case's Reports screen, with one briefing. The data is fictional.

Reports and patterns

A daily report, and a baseline for each camera

  • Reports

    A report is written for your team once a day, automatically, with the day's highlights, patterns and anomalies.

    An example daily report: the period, the day's event and alert counts, a summary that names cameras, highlights, the peak hour and a list of the most significant events. The data is fictional.
  • Memory and patterns

    Each camera has a baseline by day of week and hour, and once it has enough history deviations are flagged.

    The same report's Patterns and Anomalies sections. Each line names a camera, a day and an hour and says how it differs from the usual. The data is fictional.

Smart cameras compared

A smart camera answers yes or no. Ask in plain language.

A smart camera

Person, vehicle, animal. It reports only the classes it was built for.

Focient, on the same camera

What we saw

A pickup is stopped at the corner with its lights on. One person stands near the crossing.

The vision model writes its own description, tags and significance score for each frame that reaches it, and what it describes is kept as an event whether or not a rule is set. A rule can add an optional plain-language check against the event's description before anyone is notified.

Both panels show the same street corner at night: a pickup stopped at the crossing with its lights on, and one person standing near it. In the left panel a smart camera labels the scene Animal: no, Person: yes and Vehicle: yes, with the tag Smart camera · fixed classes. In the right panel Focient labels the pickup vehicle · #0746 and the person as person, on camera CAM 12 · CORNER, and says what it saw.

The labels are illustrative.

From camera to answer

From your cameras to a cited answer.

Works with the RTSP cameras you already run, with no new cameras to buy.

  1. 01

    Your cameras

    The ones you already run

  2. 02

    Focient describes

    What each camera sees, in words

  3. 03

    One record

    Events, with their frames

  4. 04

    Your team

    Alerts, answers, reports, cases

Your sector

End of shift

See it in a demo.

Everything this site describes is built into the product today. In a demo we show you the console on recorded footage, under NDA.

Tell us about your cameras and what you need to ask of them. We will reply by email to arrange a time.