space coordinates. about multiview_calib

[lcit]

if you want a 3D bounding box around the object in the scene, you can obtain it easily from the triangulated landmarks i.e. the points in 'global_triang_points.json' after the global registration

[wagaabderrahim]

Thank you for your response. Indeed, to obtain the complete 3D representation of an object in the scene, I understand that it is necessary to recover its 3D bounding box. However, I would like to clarify how to retrieve the total depth of the environment along the three axes (X, Y, Z), as indicated in the camera's calibration file __
Knowing that I have multiple cameras. . Thank you in advance for any clarification you can provide.

[lcit]

I do not understand what you mean with total depth. And the JSON file you posted is not generated by this repository. I do not know what the values mean.

[wagaabderrahim]

As you can see in the file, we have this section:

"Space": {
"MaxU": 2860,
"MinU": -4200,
"MaxV": 3180,
"MinV": -2960,
"MaxW": 2100,
"MinW": 100,
"VoxelSizeInMM": 20
}

But at the same time, I have multiple cameras, and each camera has intrinsic and extrinsic parameters. Therefore, these parameters represent the minimum and maximum values of each axis in the real scene.
I want to retrieve these values because we performed triangulation using 4 cameras, and I still have these values that I need to recover.