[x3d-public] Here's something worth tracking down Roy: Normal accompanying PointSet

vmarchetti at kshell.com vmarchetti at kshell.com
Thu Oct 26 12:18:57 PDT 2017


"Normal" might be too exacting a term to apply to this important idea.
What we really want to do is to have some way of attaching a direction to at least some of the points in a point cloud.

One use would be flow-visualizations, displaying the velocity field at certain points.

In scanning, where a point cloud is used to estimate a surface of a solid -- a directional vector that is known to point out of the solid is important to create
a mesh for the surface. If the scan is done optically, a basic arrow that points from the measured point surface to the optical sensor serves this purpose

Some contact probes intended for surfaces in machining applications can generate both an accurate point location and an accurate normal to the hard metal surface at that point.

> On Oct 26, 2017, at 11:47 AM, Leonard Daly <Leonard.Daly at realism.com> wrote:
> 
> [Responding to the most recent message on this topic]
> 
> It will be necessary to very carefully define what this means. Individual points do not have a normal -- it is mathematically impossible. In fact it takes at least three points to get the number of normals to a finite number. If normals are going to be defined based on a collection of points, then it will be necessary to carefully define how the collection is obtained and ordered.
> 
> If a normal is assigned to each point, then it defines a orientated (front/back) plane that is perpendicular to the normal.
> 
> If an algorithm is going to be used to construct normals, than I believe the X3D is not the right place for that calculation. It is a display tool and not an analysis or constructor tool.
> 
> Leonard Daly
> 
> 
> On 10/26/2017 8:21 AM, Don Brutzman wrote:
>> I recommend that inclusion of Normal as allowed with PointSet and ILS/LS be only found in X3D v4 JSON schema. 
>> 
>> I will accept the action to update this design-pattern example to X3D v4 with corresponding content-model validation by v4 XML Schema and DTD. 
>> 
>> all the best, Don 
> 
> -- 
> Leonard Daly
> 3D Systems & Cloud Consultant
> LA ACM SIGGRAPH Past Chair
> President, Daly Realism - Creating the Future
> _______________________________________________
> x3d-public mailing list
> x3d-public at web3d.org
> http://web3d.org/mailman/listinfo/x3d-public_web3d.org




More information about the x3d-public mailing list