I’m currently using a monocular RGB camera + LiDAR setup where I detect a target in the image (pixel u, v) and convert it to real-world coordinates using a GSD-based approach with height from LiDAR.
I’m planning to switch to the ZED X camera and want to simplify this pipeline.
From my understanding, ZED provides depth and point cloud data, so instead of using GSD, I should be able to directly get:
(u, v) → (X, Y, Z)
My questions:
-
Does ZED X directly provide per-pixel 3D coordinates (X, Y, Z) in metric units?
-
Are these coordinates in meters by default, and can they be configured to millimeters?
-
Is using
retrieve_measure(..., MEASURE.XYZ)the correct way to access these 3D points?