Why is point-based rendering used?

[jogging]

Hi,all
At one fair, I saw one 3D scanner. On the computer the application software will
process the point cloud of each captured frame. I find that the software processes the point cloud directly, not first convert the point cloud to triangle mesh presentation.I think it is convenient to process the mesh. On the screen it seems that the software uses point-based rendering to render the point cloud.I am not familiar with point-based rendering. It is easy to render triangle meshes.Why is point-based rendering used?

Thanks in advance.

Regards
Jogging

[Daniel Pitts]

Let's ask the same question, but in the context of a 2d image, shall we?

Why does 2d scanner software use raster graphics rather than vector
graphics? The answer is that it more directly maps to the physical
process used to acquire the data. I imagine that the 3D scanner
hardware isn't finding meshes, but is in fact finding a point cloud.

I can imagine that someone using the 3d scanned data may want to process
it in several different ways, and converting that data into a mesh may
interfere.

Also there is ambiguous solutions to triangle meshes, which would mean
that an automatic process might come up with a mesh that doesn't match
what the user needs.

[Nobody]

On Thu, 15 Nov 2012 22:15:15 -0800, jogging wrote:

> It is easy to render triangle meshes. Why is point-based rendering used?

Converting a point cloud to a mesh requires knowing which vertices are
part of the same triangle, and that information presumably isn't available.