GloveChess is an implementation of the chess game in augmented reality which exploit a hand tracker system as input for chess move.
The whole application use Python, OpenCV and OpenGL

Set up

Python v. 3.7.6
Main python dependencies:

• OpenCV 3.4.2
• PyOpenGl
• Numpy
• Pillow
• Pygame

The projection is marker-based using ARuco module. In the project the following marker has been used, so be sure to get one

Clone the repository

git clone https://github.com/xAlessandroC/GloveChess

Install the dependencies

pip3 install -r requirements.txt

Unfortunately there is some kind of problem with installing PyOpenGl with pip since the freeglut binaries won't be installed. To solve this problem you have to search here https://www.lfd.uci.edu/~gohlke/pythonlibs/#pyopengl the PyOpenGl and PyOpenGl-accelerate binaries for your python version and install them. In the resources/wheels you can find mine.

Launch the application

python GloveChess_application.py

This is what you should get

Working application

Step 1: At the beginning the application need to acquire the color for hand detection, so put the hand in front of the green square and press 'd'

Step 2: Now the game starts

How to make chess move

Since the projection of the chessboard happen over the hand we add a pointer that helps you to keep trace of the selected piece/cell. Use your whole hand to move the selector

Use two finger to selected the starting or ending cell

Hint: you can use the terminal to know if the selection has been computed properly, sometimes some mistakes can happen

Credits

Thanks to Pietro Morisco for the logo!
Thanks to printable_models for the chess 3D models