EyeTrackVR #
DIY eye tracking for any headset. Here are the steps to get the software working on linux.
Prerequisites #
- tkinter
pacman -S tkorapt install python3-tk - Python 3.11 for it to work out-of-the-box, available through the AUR, build yourself! Otherwise tk will not work!
- Build after installing tkinter, otherwise it will not be included
- poetry
apt install python3-poetry
Clone the repository #
git clone https://github.com/EyeTrackVR/EyeTrackVR.git
cd EyeTrackVR
Patching for Python 3.13 #
Eyetrackapp runs fine off of Python 3.13, but you need to change a couple package versions. Follow this if you don’t want to build or install Python 3.11
Change these three lines in pyproject.toml and delete poetry.lock
@@ -7,11 +7,11 @@ license = "LICENSE: Babble Software Distribution License 1.0"
repository = "https://github.com/EyeTrackVR/EyeTrackVR"
[tool.poetry.dependencies]
-python = "~3.11.0"
+python = "~3.13.0"
python-osc = "^1.8.0"
requests = "^2.28.1"
opencv-python = "^4.6.0.66"
-numpy = "~1.24.3"
+numpy = "~1.26.4"
pye3d = "^0.3.2"
pysimplegui-4-foss = "^4.6.4.1"
pydantic = "^2.4.2"
@@ -29,7 +29,7 @@ matplotlib = "^3.10.7"
[tool.poetry.group.dev.dependencies]
black = "^22.10.0"
-pyinstaller = "^5.6.2"
+pyinstaller = "^6.10.0"
flake8 = "^5.0.4"
[tool.taskipy.tasks]
Building EyeTrackApp #
poetry install --no-root
cd EyeTrackApp
poetry run pyinstaller eyetrackapp.spec
and just like that you now should now have a fully functional eyetrack app in directory dist
Your serial cams should be up under /dev/ttyACM*
Alternatively, instead of building an executable you can run the app directly from the code by running poetry run python3 eyetrackapp.py within the EyeTrackApp directory.
VRCFT #
Be sure to check out VRCFaceTracking.Avalonia for cross-platform VRCFT.