This is a computer vision system that allows players to play chess with a real-life chessboard. The system uses computer vision and a convolutional neural network to detect and track the movement of chess pieces on the board.
To install the system, follow these steps:
- Clone this repository onto your local machine.
- Install the required libraries and dependencies by running pip install -r requirements.txt in the command line.
- Connect a camera to your computer and position it over the chessboard. The camera should be positioned directly above the board, with the chessboard fully visible in the camera's field of view.
- Run the main.py script to start the system.
The computer vision system uses OpenCV and a convolutional neural network (CNN) to detect and track the movement of chess pieces on the board. The CNN was trained on a large dataset of images of chess pieces in different positions and orientations. The dataset was provided by [insert your name or organization here].
If you have your own chessboard and would like to train a model specifically for it, you can do so from the setting.
To play chess with the system, simply move the chess pieces on the board as you would in a regular game of chess. The system will detect the movement of the pieces and update the game board on the screen. You can play against another human player or against the computer.
This system was developed by Pietro Mele