Estimate the position of eyes in the frame using computer Vision Techniques.
Control.RGB.LED.s.Color.with.Eyes.mp4
Iris Tracking Medipipe Opencv python Video Tutorial 📹
out_medium.demo.mp4
-
Mediapipe is way faster, has more landmarks, it can reach up to 30 fps as well. while dilb reaches about 10fps.
-
Mediapipe is easy to install, just with the pip command
-
mediapipe does not require extra things, Visual Studio, CMake, etc on Windows machine,
Here you can find Project Repository
Control Things with Eyes Tutorial
- Eyes landmark Detection 👀
- Blink Detector and Counter
- Extracting Eyes using Masking Techniques
- Threshold Eyes
- Dividing Eye into Three Parts Right Center Left
- Counting Black in Each part and Estimating The Position of Eye
- Controlling RGB LED's Color With Eyes
Steps are involved to run the code.
-
install Dlib
-
In order to install Dilb on windows machines you need the following: 😏
-
Visual Studio
-
Visual Studio Build Tools
-
Cmake
for More Detail check out this Blog post. here
-
pip install cmakepip install dlib-
Just you need Cmake that all here on Linux and Mac OS
-
install Dilb using Pip command
pip3 install cmakepip3 install dlibfor more details see
-
-
Install opencv-python
pip install opencv-pythonFor Linux or Mac OS replacepipwithpip3and you are good to go... 😉 -
Download landmarks Predictor
- Extract that file and put it into the Predictor folder or directory.
image shows the face landmarks available in, the dilb face landmark detector pre-trained network.
If you found this Helpful, please star ⭐ it.
You can Watch my Video Tutorial on Computer Vision Topics, just check out my YouTube Channel
If you have any questions or need help with the project, feel free to DM me on Instagram!