mapping returned from cv::aruco::estimatePose

asked 2019-10-07 15:38:46 -0500

genesys gravatar image

updated 2019-10-07 16:11:33 -0500

cv::aruco::estimatePose returns the mapping between marker space and camera space via the two variables rvec and tvec - but the documentation doesn't state how exactly. In particular I wonder whether it describes camera location/rotation in the marker space or whether it describes marker location/rotation in camera space?

Or stated differently, is it

point_in_marker_space = (Rodrigues(rvec) * point_in_camera_space) + tvec

or

point_in_camera_space = (Rodrigues(rvec) * point_in_marker_space ) + tvec

or something else?

edit retag flag offensive close merge delete

Comments

It is the second one.

You can test it easily by printing the tvec values. Here is the camera frame.

Eduardo gravatar imageEduardo ( 2019-10-09 13:37:47 -0500 )edit