Skip to content
This repository has been archived by the owner on Apr 15, 2024. It is now read-only.


Repository files navigation

EyeLoop License: GPL v3 contributions welcome Build Status version lab beta


EyeLoop is a Python 3-based eye-tracker tailored specifically to dynamic, closed-loop experiments on consumer-grade hardware. Users are encouraged to contribute to EyeLoop's development.


  • High-speed > 1000 Hz on non-specialized hardware (no dedicated processing units necessary).
  • Modular, readable, customizable.
  • Open-source, and entirely Python 3.
  • Works on any platform, easy installation.


How it works

EyeLoop consists of two functional domains: the engine and the optional modules. The engine performs the eye-tracking, whereas the modules perform optional tasks, such as:

  • Experiments
  • Data acquisition
  • Importing video sequences to the engine

The modules import or extract data from the engine, and are therefore called Importers and Extractors, respectively.

One of EyeLoop's most appealing features is its modularity: Experiments are built simply by combining modules with the core Engine. Thus, the Engine has one task only: to compute eye-tracking data based on an imported sequence, and offer the generated data for extraction.

How does the Engine work?
How does the Importer work?
How does the Extractor work?

Getting started


Install EyeLoop by cloning the repository:

git clone

Dependencies: python -m pip install -r requirements.txt

Using pip: pip install .

You may want to use a Conda or Python virtual environment when installing eyeloop, to avoid mixing up with your system dependencies.

Using pip and a virtual environment:

python -m venv venv

source venv/bin/activate

(venv) pip install .


  • numpy: python pip install numpy
  • opencv: python pip install opencv-python

To download full examples with footage, check out EyeLoop's playground repository:

git clone


EyeLoop is initiated through the command-line utility eyeloop.


To access the video sequence, EyeLoop must be connected to an appropriate importer class module. Usually, the default opencv importer class (cv) is sufficient. For some machine vision cameras, however, a vimba-based importer (vimba) is neccessary.

eyeloop --importer cv/vimba

Click here for more information on importers.

To perform offline eye-tracking, we pass the video argument --video with the path of the video sequence:

eyeloop --video [file]/[folder]

EyeLoop can be used on a multitude of eye types, including rodents, human and non-human primates. Specifically, users can suit their eye-tracking session to any species using the --model argument.

eyeloop --model ellipsoid/circular

In general, the ellipsoid pupil model is best suited for rodents, whereas the circular model is best suited for primates.

To learn how to optimize EyeLoop for your video material, see EyeLoop Playground.

To see all command-line arguments, pass:

eyeloop --help

Designing your first experiment

In EyeLoop, experiments are built by stacking modules. By default, EyeLoop imports two base extractors, namely a FPS-counter and a data acquisition tool. To add custom extractors, e.g., for experimental purposes, use the argument tag --extractors:

eyeloop --extractors [file_path]/p (where p = file prompt)

Inside the extractor file, or a composite python file containing several extractors, define the list of extractors to be added:

extractors_add = [extractor1, extractor2, etc]

Extractors are instantiated by EyeLoop at start-up. Then, at every subsequent time-step, the extractor's fetch() function is called by the engine.

class Extractor:
    def __init__(self) -> None:
    def fetch(self, core) -> None:

fetch() gains access to all eye-tracking data in real-time via the core pointer.

Click here for more information on extractors.

Open-loop example

As an example, we'll here design a simple open-loop experiment where the brightness of a PC monitor is linked to the phase of the sine wave function. We create anew python-file, say "", and in it define the sine wave frequency and phase using the instantiator:

class Experiment:
    def __init__(self) -> None:
        self.frequency = ...
        self.phase = 0

Then, by using fetch(), we shift the phase of the sine wave function at every time-step, and use this to control the brightness of a cv-render.

    def fetch(self, engine) -> None:
        self.phase += self.frequency
        sine = numpy.sin(self.phase) * .5 + .5
        brightness = numpy.ones((height, width), dtype=float) * sine
        cv2.imshow("Experiment", brightness)

To add our test extractor to EyeLoop, we'll need to define an extractors_add array:

extractors_add = [Experiment()]

Finally, we test the experiment by running command:

eyeloop --extractors path/to/

See Examples for demo recordings and experimental designs.

For extensive test data, see EyeLoop Playground


EyeLoop produces a json-datalog for each eye-tracking session. The datalog's first column is the timestamp. The next columns define the pupil (if tracked):

((center_x, center_y), radius1, radius2, angle)

The next columns define the corneal reflection (if tracked):

((center_x, center_y), radius1, radius2, angle)

The next columns contain any data produced by custom Extractor modules

Graphical user interface

The default graphical user interface in EyeLoop is minimum-gui.

EyeLoop is compatible with custom graphical user interfaces through its modular logic. Click here for instructions on how to build your own.

Running unit tests

Install testing requirements by running in a terminal:

pip install -r requirements_testing.txt

Then run tox: tox

Reports and results will be outputted to /tests/reports

Known issues

  • Respawning/freezing windows when running minimum-gui in Ubuntu.


If you use any of this code or data, please cite [Arvin et al. 2021] (article).

  title    = "{EyeLoop}: An open-source system for high-speed, closed-loop
  author   = "Arvin, Simon and Rasmussen, Rune and Yonehara, Keisuke",
  journal  = "Front. Cell. Neurosci.",
  volume   =  15,
  pages    = "494",
  year     =  2021


This project is licensed under the GNU General Public License v3.0. Note that the software is provided "as is", without warranty of any kind, express or implied.


Lead Developer: Simon Arvin,


Corresponding Author: Keisuke Yonehera,




EyeLoop is a Python 3-based eye-tracker tailored specifically to dynamic, closed-loop experiments on consumer-grade hardware.




Code of conduct





No packages published

Contributors 4