CloudCompare forum

Good morning, I am a student from the University of Malaga(Spain) who is doing my final degree work and I would need someone to help me with an error in the mesh. My work seeks to go from a point cloud to a complete mesh of the bridge structure, for them I cleaned the point cloud, I obtained the normals, I defined them with the command (With minimun spanning tree) and then I misused with poisson recontructiom and the result is the following:
As you can see, the result does not fit the structure of the bridge and I do not know where the error is, since if I mesh each element separately (pillars, beams) it is done correctly,but the whole does that weird thing
Could someone help me and tell me where the fault is or how to solve it Thank you very much.

You need to have poison output the density as a scalar and then filter by density. This is a fairly common result/question.

Thank you very much and how is it done so that it emits the density as a scalar?

See https://www.cloudcompare.org/doc/wiki/i ... n_(plugin)

Thank you very much, all your answers have been helpful.