Everyfr A me JT Digital

Scale video. One engine, Every job.

Everyframe 2026
Rec 00:00:06:04

One engine.
As many videos as you need.

Everyframe is video software we built and host in Europe. It takes footage you already own and returns finished video at the volume the work needs. Every example below is a real run on real client footage.

01  /  Redaction  /  Tour de France
Detected
Pixelate
Black

Ship the footage without clearing every face in it.

369 faces found across 186 frames, all pixelated except the one rider you want on screen. He is held by his track identity, so no face recognition and no boxes drawn by hand. The same run also exports a blacked-out version.

Black / pixelate / blur Runs on-device

369

Faces detected
186 frames

02  /  Composition  /  National innovation awards

1,410 award videos. Nobody opened an editor.

One design and one list of entries. Every video came back finished in its own region and voice, with zero manual re-edits.

235 entries  /  6 voice variants  /  13 regions

1,410

Finished videos

One cell, one rendered video

03  /  Motion counting  /  Warehouse floor

Every lift counted. Nobody wearing a sensor.

The engine reads 17 joints per person from the camera already on the floor. It finds the repeating motion on its own, so there is no rule to write per exercise and no hardware to buy, fit or charge. Give it footage that does not repeat and it refuses to count. It will not guess.

17

Tracked joints
No sensor worn

Left: the camera  /  right: what the engine reads 17 tracked joints

Six more clips the same run counted

01 / 01
04  /  Edge tracking  /  Drone ISR and target lock
Uplink Trk TGT-01 Conf 0.38 Power 17 W Record ~68 B Video sent: none

Vehicles read from the air. The video never leaves the drone.

A Jetson board on the aircraft does the seeing at 17 W, fully offline. Home gets about 68 bytes per record and never a frame, so the link stays usable on a bad day and the footage stays where the rules say it has to stay.

68

Bytes per record
17 W on the board

34

Frames carried by the filter
out of 250

The lock, frame by frame

Left is the drone feed with the tracker's box drawn on it. Right is the same target magnified. The coloured box is the raw reading for that frame, and the colour says which confidence gate it cleared. The dashed box and the ring around it are the engine's guess and how far off that guess could be. Watch the ring swell when the detector loses the car, then snap tight the moment a reading comes back.

Drone feed · the tracker's view Box = tracked estimate
Lock view
Tracker state ACQUIRE
Confidence
--
Gate
--
Speed px/frame
--
Uncertainty
--
Track age
--
Frame
--
Waiting for the first frame.
Reading, high confidence Reading, low confidence, kept anyway No reading, the filter carries the box

20 frames clear the high gate. 196 are low confidence and kept anyway. 34 have nothing to read, so the filter carries the box. That is all 250 frames, at a mean confidence of 0.38 .

What happens when the target fades out.

Holding one vehicle is the hard part. A car drops below what the detector can see and the box keeps moving anyway. Confidence gates decide what still counts as the same vehicle, and a filter carries the box forward while there is nothing to read. A solid box is a reading and a dashed box is the engine's guess.

“Perception, not force.” The engine reads a scene and hands it to a person. A human stays in every decision.

Seven passes the engine has run on real footage

01 / 01
01 Depth
Depth, on the same board
02 Segmentation
On the road, or off it
03 Geo-projection
From a track to a coordinate
04 Pose
What the target is doing
05 Convoy
A column, not a pixel
06 MTI
Movers, no label needed
07 Target lock
The solid box is a reading  /  the dashed box is a prediction Predicted 34/250
05  /  The through line  /  One engine behind all four

Why one engine covers all of them.

Other tools read a JSON template on their own servers every time a video runs. Everyframe compiles a typed description into one portable program. That is compiler work, and it is what every run above rests on.

Models are typed functions. A detector goes from video to detections. If one model's output matches the next one's input they chain. If it does not, the pipeline does not compile.

Video Detections Tracks Redact Count Render Report

Typed before it runs

Mistakes surface when the pipeline is compiled. A batch of 1,400 videos does not fail on the 900th video.

One portable artifact

The same compiled program runs in our cloud, inside your own SDK, or on a machine with no network at all.

06  /  In production  /  Who is running it

Three organisations. Over a thousand finished videos.

  • GZS Inovacije Video presentations built from what each participant submitted, for the national innovation awards.
  • Varnoška Workplace safety drill videos, one variant per site.
  • Faculty of Logistics Redaction on training footage before it is shown to students.
Everyfr A me JT Digital

Different examples, the same engine behind all of them.

Nothing on this page is a mockup. Every showcase is a run on footage a client already had, and the numbers beside each one come from that run.