Point clouds of the same object captured by multiple fixed cameras in different coordinate systems

I have two fixed-position cameras and want to merge their recorded SVO2 files into a complete point cloud. While I can obtain individual point clouds from each camera, I don’t know how to use Python to fuse these two point clouds into a unified whole.

“pose1”: “1.000000 0.000000 0.000000 0.000000 0.000000 1.000000 0.000000 0.000000 0.000000 0.000000 1.000000 0.000000 0.000000 0.000000 0.000000 1.000000”,

“pose2”: “-0.997949 0.003248 0.063927 -0.114481 -0.000056 0.998667 -0.051616 1.679325 -0.064009 -0.051514 -0.996619 3.249537 0.000000 0.000000 0.000000 1.000000”,

SDK:4.2.5
Python:3.8
ZED2i

Could you please provide me with some example code, demonstrating how to convert and merge files? I would be extremely grateful.

Hi @BUZHA,

Point clouds are retrieved in a structured manner, through a sl.Mat which contains the XYZRGBA information for each pixel in the image.

For your use case, as you are attempting to fuse multiple point clouds, you will have to convert the sl.Mat in a destructured format, apply the transformation to your second point cloud, and add the point clouds to each other.

I would recommend using a point cloud processing library such as Open3D in python to perform these operations on your point clouds.

1 Like