By hearing the name Point Cloud, it is understood that the cloud of points. That is, creating a cloud of where the points are.
Depth of any object cannot be understood in normal camera or two dimensional camera. But in three-dimensional image or camera, the x,y,z axis information of the object is available. Let’s say we are going to create a point cloud of a person, then we need to know the depth information of that person first. If a person stands in front of the camera, depth information will be available where there is a figure and where there is no depth information will be infinite. Now we get more points where we got depth information and less where we don’t. In this way, the point cloud is calculated by calculating the depth of each point.
There is a separate library for point clouds. Library Link: http://pointclouds.org/