In this repository you will find the final project of the Computer Vision subject at the University of Málaga fully implemented by the student of Computer Science Marcos Hidalgo. As you will soon discover, our goal is to control the mouse with the eyes by estimating how much the center of the iris has moved between frames of the input video. The project is composed of two main phases which are:
- A Jupyter Notebook containing the code and corresponding explanations. Check the Jupyter Notebook here!
- The final qualification is obtained after the realization of an exhibition. Check the presentation slides here!
The project was develop with academic porposes and does not represent a proffesional work so it's not a viable tool to use. Due to imprecisions made during the track of the iris, the program does not respond as expected. Nevertheless, the code used to implemented is compleatly functional and needs to be reinforced with advanced CV techniques that I am unable to develop with my current knowledge.
You must have previuslly installed Jupyter Notebook to succesfully run the Python code within it self. Check this link to properly install them! Additionaly, there are function that have to be imported from external libraries manually:
Most of the functions used are from OpenCV. If not installed already, just copy paste on the console the following.
pip install opencv-pythonOur program needs to be able to handle the clicks without using the mouse. With that porpose, Pynput provides us a Controller to detect them allongside some valuable functions and constanst to represent mouse interactions.
pip install pynput