llocus

Eye tracking, without the eye tracker

We make eye tracking work on any webcam.

Eye tracking has always meant buying a dedicated sensor. We are removing it. Our gaze engine turns the cameras already sitting in laptops, kiosks, and machines into an input you can build on.

Talk about an integration See how it works

The problem we work on

Detecting eyes is solved. Keeping gaze useful for hours is not.

Almost anything can track a person's gaze for ten seconds. Then they shift in the chair, the afternoon light crosses the desk, the laptop is a different one, and hour two looks nothing like the calibration. That gap, not the demo, is what has kept eye tracking out of shipping products. We bridge it, so eye tracking becomes possible on your product.

So we measure the part that is hard.

Gaze accuracy is quoted as an angle: the gap between where someone looked and where the engine said they did. Smaller is better. Here is where we sit.

1.43° sitting still

The angle between where a person was actually looking and where we said they were. At normal laptop distance that is roughly a centimetre and a half on screen, close enough to know which line someone is reading.

1.94° while moving

Measured with the user actually moving, because movement is what breaks most webcam gaze systems. Staying under two degrees through it is the number we care about most.

Across 15 hardware setups

Real laptops, uncontrolled rooms, changing lighting, more than one face. Not a lab bench, not one favourable webcam, and not a single tester.

1.64° for the reference webcam system

That is Tobii Nexus, the best known webcam based eye tracking system, working from the same class of camera we do. We measure 1.43°. Dedicated eye tracking sensors are a separate category and can still be more accurate, but they need a sensor bolted to the product. We do not.

What you integrate is one engine.

Camera frames go in. A gaze signal comes out. Calibration, per-user adaptation, and movement handling live inside it instead of on your roadmap.

It fits the person in front of it

No two faces, cameras, or seating positions produce the same picture, so we do not ship one frozen model and hope. A short calibration fine-tunes the engine to the individual user, on the machine they are actually using.

It keeps learning after calibration

Real use tells the engine when it was probably wrong. We feed that back in to keep improving during normal use, instead of sending people back to a calibration grid every time something drifts.

It runs on about one CPU core

Capture, inference, and smoothing together. No GPU budget, no dedicated silicon, and enough headroom left over for the product the gaze signal is feeding.

Gaze data that never goes anywhere.

A camera pointed at a person is a sensitive thing to add to a product. We built the engine so that adding it does not create a new place for that data to go.

The frames never leave the device

Gaze is computed on the machine holding the camera. Webcam video does not travel anywhere to be turned into a gaze signal, because the work happens where the camera already is.

There is no llocus server in the loop

The engine does not need us to be reachable to do its job. Nothing about a user, their face, or where they looked has to reach us for the product around it to work.

Nothing new to defend in a security review

For an embedded or industrial deployment, camera data staying on the device is usually the difference between a short conversation with a security team and a long one.

Where this goes

Anywhere a camera already points at a person.

  • Industrial and machine interfaces
  • Manufacturing environments
  • Accessibility
  • Research instrumentation
  • Human-computer interaction
  • Productivity software
  • Education
  • Device and OEM integration

Why we also ship an app

llocus desktop is a real product and the engine's test surface at the same time.

No lab produces a thousand different webcams, faces, rooms and postures. People do. llocus desktop lets anyone read and navigate with their eyes, and every hour of that use shows us where the engine is still wrong, before the same engine goes into somebody else's product.

See llocus desktop

Commercial inquiries

Gaze should be a normal input.

Microphone, camera, keyboard, touch. Each one is just a signal software can ask for. Eye tracking is the last one still locked behind special hardware, and that is the whole reason we exist.

Tell us what you want to point a camera at: the device, the conditions it runs in, and what a gaze signal would let you do. You will get a straight answer about fit from the people building the engine.