robosuite
robosuite copied to clipboard
How to get camera quat based on (camera_pos, lookat_dir, up) for mujoco renderer
How would you go from (camera_pos, lookat, up)
to quat for the mujoco renderer (not e.g. nvsii) to set the camera pose?
My use case is that I want to initialize the cameras in a circle and point them all towards the center of the table; and it'd save a lot of time to not have to manually copy paste quat values from tune_camera.py
. If anyone has pointers, that'd be much appreciated!