Metric mapping with single camera + 2d lidar + imu + encoder?

[zonkedsquare]

Hi,
I understand that we lose scale if we try to map the world with a single camera. Can we use RTABMAP to build a map using a single monocular camera, but supplement with scale information from a combination of a 2d lidar, an imu and encoder odometry?
If this is possible, is it then possible to localize up to scale using a single monocular camera once we have ametric map of the world?
Thanks,
zonkedsquare

[matlabbe]

It is possible when activating "Mem/StereoFromMotion". However, it assumes that odometry is relatively accurate between two rtabmap updates (e.g., during 1 sec) while the robot is moving.