[Paraview] Suggestion for STL import

[Eugene de Villiers]

Utkarsh,

Unfortunately, the STL that produced the issue is confidential and also quite large, so I am unsure how to go about producing something equivalent. Let me know what you find after looking at the reader and I will try to find something appropriate. To be honest, the reasoning is speculative, but it is clear that the STL reader has some drawbacks. At the very least we would be happy to test any alterations you might come up with on the original input.

Not merging the points is a reasonable work-around, but you will lose some utility (feature detection). I think it should be relatively easy to create or port an existing algorithm that is a bit more efficient (of course these kind of assumptions tend to be wrong, but anyway!).

Eugene

From: Utkarsh Ayachit [mailto:utkarsh.ayachit at kitware.com]
Sent: 05 July 2016 17:05
To: Eugene de Villiers <e.devilliers at engys.com>
Cc: paraview at paraview.org
Subject: Re: [Paraview] Suggestion for STL import

Eugene,

I am pretty sure the community will be interested in this! While I'll need to look into the reader to understand, to get things going, do you have a sample dataset to demonstrate the issue?

Utkarsh

On Tue, Jul 5, 2016 at 10:28 AM, Eugene de Villiers <e.devilliers at engys.com<mailto:e.devilliers at engys.com>> wrote:
Hi,

When importing STL format geometry or similar, where connectivity information is not implicit in the data structure, it appears that the connectivity is reconstructed via an  octree search. This is very inefficient when surfaces with large differences in edge length are imported – we have had a recent case where an STL took 45mins to load. If the same input geometry is converted to OBJ format via an external tool, the load time reduces to minutes.

A generally more efficient method is to calculate the distance of each point from a location outside the point cloud bounding box and then to bubble-sort the resulting list. Unless you are dealing with a pathological case your local search neighbourhood of identical distance points will be small and the algorithm very fast. More complex, multi-origin algorithms are also possible to counter pathological instances.

I can provide more details and sample code if you are interested.

Best regards,

Eugene de Villiers
Managing Director
e.devilliers at engys.com<mailto:e.devilliers at engys.com>
Mob: +44 (0) 77 89748490<tel:%2B44%20%280%29%2077%2089748490>
Tel: +44 (0)20 32393041<tel:%2B44%20%280%2920%2032393041> (ext. 102)
Fax: +44 (0)20 33573123<tel:%2B44%20%280%2920%2033573123>
[logo_red-black_fonts_signature]<http://www.engys.com/>

This message is intended only for the use of the addressee and may contain information that is privileged, confidential and exempt from disclosure under applicable law. If the reader of this message is not the intended recipient, or the employee or agent responsible for delivering the message to the intended recipient, you are hereby notified that any dissemination, distribution or copying of this communication is strictly prohibited. If you have received this e-mail in error, please notify us immediately by return e-mail and delete this e-mail and all attachments from your system.



_______________________________________________
Powered by www.kitware.com<http://www.kitware.com>

Visit other Kitware open-source projects at http://www.kitware.com/opensource/opensource.html

Please keep messages on-topic and check the ParaView Wiki at: http://paraview.org/Wiki/ParaView

Search the list archives at: http://markmail.org/search/?q=ParaView

Follow this link to subscribe/unsubscribe:
http://public.kitware.com/mailman/listinfo/paraview

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://public.kitware.com/pipermail/paraview/attachments/20160708/894b52c3/attachment.html>
-------------- next part --------------
A non-text attachment was scrubbed...
Name: image001.png
Type: image/png
Size: 2215 bytes
Desc: image001.png
URL: <http://public.kitware.com/pipermail/paraview/attachments/20160708/894b52c3/attachment.png>