sksurgeryarucotracker.algorithms.registration_2d3d module

Classes and functions for 2D to 3D registration

sksurgeryarucotracker.algorithms.registration_2d3d.estimate_poses_no_calibration(marker_corners, aruco_board)

Returns tracking data for a camera with no calibration data. x and y are the screen pixel coordinates. z is based on the size of the tag in pixels, there is no rotation. No account is taken of the size of the model marker pattern, so it will be bit flakey.

sksurgeryarucotracker.algorithms.registration_2d3d.estimate_poses_with_calibration(marker_corners2d, marker_ids, aruco_board, camera_projection_matrix, camera_distortion)

Estimate the pose of a single tag or a multi-tag rigid body when the camera calibration is known. :param marker_corners2d: a list of 2d marker corners, 1 row per tag,

8 columns per tag

Parameters:

• model_points – Matched list of of corresponding model points, 1 row per tag, 15 columns per tag: corner points and centre point
• camera_projection_matrix – a 3x3 camera projection matrix
• camera_distortion – camera distortion vector

:return : a tracking rotation, translation and a quality