Good day dear community,
I have been dealing with the issue of camera calibration for a few days. In a project we have four cameras on a robot in a fixed position. Furthermore the intrinsic camera parameters are known. We need the external parameters of the cameras. In addition, we would like to use camera 1 as a reference coordinate system for all other cameras.
For this I have the following idea: We use a large ChArUco board, which serves on the ground to get known 3D points in world coordinates. As far as I understood correctly, we can then use cv.solvepnp() to get the rotation and translation for each camera. Where I am unsure is how can I subsequently specify camera 1 as the reference coordinate system? To better illustrate the problem, I have uploaded a picture. Here the triangles represent the cameras. Can anyone tell me if I am on the right track with my approach. I am also open to advice.
