Eagles Eye — see more, understand more, act faster

EDUCATIONAL RESEARCH PLATFORM

EVERY CAMERA YOU ALREADY HAVE, ACTUALLY WATCHED

Fifteen agents watch every camera continuously and raise explained incidents — crowd density, unattended baggage, restricted-zone entry and cross-camera person search. Processing runs on your own GPU, so footage never has to leave your site.

FIFTEEN AGENTS

Crowd, behaviour, objects, identity and video events - all reading one shared picture of every frame.

YOUR OWN GPU

Processing runs on your machine. Footage never has to leave the site; only conclusions travel.

EXPLAINED ALERTS

Every incident carries the arithmetic behind its score. The system recommends, a human decides.

PRIVACY FIRST

Bystander faces blurred before frames leave the machine; identity escalation needs approval.

WHAT IT WATCHES

Crowd intelligence

Density, flow, compression, counter-flow and surge, sampled continuously. A density model counts crowds too dense to separate into individual people.

Unattended objects

Bags are linked to the person who carried them. When one is left behind, the escalation carries the owner's last trajectory.

Behaviour

Loitering, counter-flow, running, erratic movement and restricted-zone entry - thresholded by your policy, not ours.

 

Person of interest

Authorised, audited watchlist matching across every camera in scope, answered in plain language.

Anomaly and video events

Weakly-supervised models trained on public CCTV datasets, each reported with its measured confidence.

Digital twin and field view

A live site map with camera coverage, and a field screen that sorts incidents by distance from the officer.

15

Specialist agents

92

% cross-camera match

35

fps per camera

185

Automated tests

Measured on held-out data during training and benchmarking, not estimated.

HOW IT WORKS

One frame, five stages, about twenty-nine milliseconds.
  1. 01

    Take the frame

    RTSP, USB or a recorded file, decoded on whichever GPU sits closest to the camera — the server here, or an operator's own machine acting as an edge node.

    12 fps target per camera
  2. 02

    Find and follow

    YOLO11m locates people and objects; ByteTrack, written from scratch rather than pulled in, keeps an identity on each one as it moves.

    7.5 ms detect on GPU, 54 ms on CPU
  3. 03

    Recognise them again

    An OSNet appearance embedding per person, computed for the whole crowd in one batch, so somebody who leaves one camera and enters another stays the same person.

    rank-1 92.8% on Market-1501
  4. 04

    Work out what it means

    Fifteen agents read the tracks: loitering, following, counter-flow, sudden running, crowd surge, an unattended bag, a face on the watchlist.

    thresholds are yours, editable at runtime
  5. 05

    Raise it with the evidence

    An incident carries the clip, the frames and the trail that produced it, and every read of it is written to the audit log.

    nothing is asserted without its source
  6. 06

    Search across cameras

    Ask for one person and the gallery answers with every camera that saw them, in order, with the time they appeared at each.

    a 15-minute window, your cameras only
  7. 07

    Ask in plain words

    Questions about what happened are answered from the incidents themselves, and every answer cites the one it came from. With no model configured it falls back to fixed wording rather than guessing.

    answers carry their source
  8. 08

    Forget on schedule

    Bystanders are blurred before a frame leaves the machine, identity escalation needs an operator's approval, and footage past its retention window is deleted without being asked.

    30-day default, yours to set

Measured, not claimed. 1080p with every agent running holds about 35 fps on a GPU and 11 fps on CPU. Those numbers come from benchmarking this pipeline, not from a datasheet.

It says when it does not know. Models are promoted only if they beat the baseline on held-out data — two were trained this month and refused. When a component is unavailable the platform reports the backend it actually used, so nothing implies a trained model where there is none.

WHERE IT RUNS

Footage never has to leave the place it was filmed.

On your own server

Point it at the recorder you already have. Cameras, zones and thresholds are configured once and the agents run continuously from there.

8 cameras per node by default

On an operator's GPU

A machine with a graphics card can register as an edge node and take the cameras nearest to it. It polls outward over HTTPS, so nothing needs to be opened to the internet.

3x the throughput of CPU-only

Anywhere you watch from

The dashboard is a browser tab — no install, no plug-in. It draws the overlays locally, so a laptop can follow a site it is nowhere near.

the browser renders, the node infers

Built for airports, railway stations, campuses and public venues — places with cameras already installed and nobody able to watch all of them at once.

Register your site and attach your first camera

Accounts are approved per organisation — an airport sees only its own cameras and incidents, never another site's.

CREATE AN ACCOUNT