'How to improve the point cloud accuracy?

I am retrieving the point cloud from ZED camera and i have a question about point cloud and 3D pixel coordinate estimation. If i placed my camera at highest point in my room do i get better point cloud results?

I thought the problem was the light or shadows spots but always i get a wrong point cloud coordinate, i know it because a plot the X and Z coordinate points (from [X,Y,Z] pixel coordinates) on graph so i can see where is the point location based on camera reference which is [0,0,0].



Sources

This article follows the attribution requirements of Stack Overflow and is licensed under CC BY-SA 3.0.

Source: Stack Overflow

Solution Source