draw-calibration returns a different pattern

[RiSpa5]

Hello everyone, I'm having an issue with the draw-calibration function since it returns a reconstruction image that is not identical to the real ChAruCo board I used for calibration. Here are the settings for the calibration in my config.toml file.

# Calibration marker configuration
[calibration]
# The type of board used for calibration, choose between: "checkerboard" / "charuco" / "aruco"
board_type = "charuco" 
# The size of the board
board_size = [4,6] # width (X) and height (Y) of grid
# ONLY IF CHARUCO/CHECKERBOARD - length of the side of the square
board_square_side_length = 65 # mm
# ONLY IF CHARUCO - length of the side of the marker within the square
board_marker_length = 48 # mm
# ONLY IF ARUCO/CHARUCO - specify the number of bits of the markers within the square (e.g. 4X4)
board_marker_bits = 4
# ONLY IF ARUCO/CHARUCO - specify how many types of markers are in the markers dictionary
board_marker_dict_number = 50
# ONLY IF ARUCO - length of marker separation
# board_marker_separation_length = 1 # mm

# Animal is used for calibration?
animal_calibration = false
# Fisheye lens used for recordings?
fisheye = false

Here are the files of the real ChAruCo I used for calibration and its reconstruction by anipose. The real one: The anipose reconstruction:

Does anyone have a possible explanation for this? Is it possible that this issue might affect negatively the board detection and the calibration process? And if so, is there any possibility to recover the calibration I did using this ChAruCo or should I do a new one? Any help and opinion is deeply appreciated! Thanks in advance!

[RiSpa5]

Any help on this issue? I still haven't found a solution for it. Any suggestion is deeply appreciated.