Closed rjeye closed 2 years ago
Hi, the point clouds have been extracted using a structure-from-motion (SfM) framework and, hence, they are not in any metric units. Without a dedicated depth-sensing device, such information cannot be inferred by simply observing 2D images of objects (which is what the SfM reconstruction pipeline does).
In order to roughly normalize coordinate systems across scenes, we center each point cloud around zero and scale-normalize so that the point cloud points have unit standard deviation across all 3 axes on average.
Are the units of the pointclouds in inches or mm? And how does that relate to the camera translation vector (extrinsics) units ?