python calibrate.py [-W WIDTH] [-H HEIGHT]
                    [--grid_length] [--debug] data_path

                    -W: Number of rows - 1 
                    -H: Number of columns - 1 
                    --grid_length: Length of a single grid
                    --debug: Option to turn on debug messages and render the result
                    data_path: Path to the directory containing images and intrinsics.json

-W and -H are the number of rows - 1 and columns - 1 of the chessboard. This is because only the inner grids are used for detection.

Place the captured images in a folder. Each image must be named as {camera_id}.png, where camera_id is a unique id for each camera.

Create instrinsics.json which contains a list of camera IDs and their camera intrinsics as follows.

{
    camera_id_1: [fx, fy, ppx, ppy],
    camera_id_2: [fx, fy, ppx, ppy]
}

camera_id is a unique string for each camera. fx and fy are the focal legthm and ppx and ppy are the pixel coordinates of the principal point (center of projection).