Dear DAVID Community,

As you are probably aware, the entire DAVID company, team and product range has been integrated into HP Inc.
This forum is now read-only. Relevant content has been migrated to HP and merged into HP's Support Forums on November 1, 2016.

To start new discussions for 3D Scanning, please register and post your new topic at the HP Support Forums

Cropped Duplicate Align Strategy

The place for all topics related to the 'Shapefusion' tool of DAVID.

Cropped Duplicate Align Strategy

Postby MagWeb » Wed Dec 18, 2013 9:35 am

Hi out there,

This example shows a common align problem and a possible strategy to solve it.
This is the set of meshes I have to align.

Head10.png


Note: For this description the partial results aren't native scans but parts of a split up already modified fusion result. I made this nonsense just to demonstrate that you might run into this problem even with perfectly fitting meshes.

To demonstrate the problem I align mesh 2 to mesh1, mesh3 to mesh2, mesh4 to mesh3 aso.

Head9.png


This procedure ends in this result:

Head8.png
MagWeb
Moderator
 
Posts: 2499
Joined: Wed Jul 18, 2007 8:48 pm

Re: Cropped Duplicate Align Strategy

Postby MagWeb » Wed Dec 18, 2013 10:01 am

Aligning a set of meshes just along the order you grabbed the scans (e.g. clockwise) isn't ideal in this case for a still unknown :wink: reason.

A better strategy than this is a basic rule:
Align the partial results along their dominance of topology.
The target for the first alignment should be the one with the most details.
In this case this is mesh1.
So I align mesh2 to mesh1 and mesh7 to mesh1.
Now also mesh3 to mesh2 and mesh6 to mesh7.
That works fine.
Also aligning mesh5 to mesh6 isn't any problem.

Head7.png


but the remaining mesh4 shows the either trouble. Using FreeAlign it is aligned faulty:

Head6.png


Also grouping all meshes but mesh4 does not give a right result:

Head5.png
MagWeb
Moderator
 
Posts: 2499
Joined: Wed Jul 18, 2007 8:48 pm

Re: Cropped Duplicate Align Strategy

Postby MagWeb » Wed Dec 18, 2013 10:22 am

The reason for this behaviour is pretty simple: DAVID finds a bigger overlap placing mesh4 somewhere else.

You might consider to use ContactPair but you might face a similar problem: Due to the poor details it is very hard to find corresponding regions.

Here comes the trick:
Duplicate mesh4. Here I name this duplicate "mesh4a"
Hide mesh4 and the other source meshes just to make sure not to delete some parts of these.
Select the inner region of mesh4a.

Head4.png


Delete this selection and show the grouped meshes.

Head3.png


Align mesh4a to the group and show mesh4

Head2.png
MagWeb
Moderator
 
Posts: 2499
Joined: Wed Jul 18, 2007 8:48 pm

Re: Cropped Duplicate Align Strategy

Postby MagWeb » Wed Dec 18, 2013 10:36 am

Now it's easy to align mesh4 to mesh4a:

head1.png


You may delete mesh4a now.

Problem fixed!

Hope that helps a bit

Gunter
MagWeb
Moderator
 
Posts: 2499
Joined: Wed Jul 18, 2007 8:48 pm

Re: Cropped Duplicate Align Strategy

Postby Sven » Thu Dec 19, 2013 7:45 am

Thanks Gunter, great tricks!

Starting the alignment at the center of the face has another advantage: IF the scans don't align perfectly (distorted scans), the important face region will look nice and bad alignment at the back can be retouched more easily.

Making a copy of a scan, then deleting all that is irrelevant for alignment, is similar to another idea we still have not had time to implement :oops: : marking relevant contact areas with a brush (mouse).

Sven
User avatar
Sven
DAVID Developer
 
Posts: 1541
Joined: Mon Jan 22, 2007 8:59 am
Location: Braunschweig, Germany

Re: Cropped Duplicate Align Strategy

Postby MagWeb » Thu Dec 19, 2013 12:24 pm

Sven wrote:Thanks Gunter, great tricks!

Starting the alignment at the center of the face has another advantage: IF the scans don't align perfectly (distorted scans), the important face region will look nice and bad alignment at the back can be retouched more easily.

Making a copy of a scan, then deleting all that is irrelevant for alignment, is similar to another idea we still have not had time to implement :oops: : marking relevant contact areas with a brush (mouse).

Sven


Hi Sven,

sure, some way to select some AreaOfInterest would be great. Basically the shown way is nothing else... But as long as you do not find the time to implement such stuff: There's a way to solve such problems with current tools.
Maybe another idea:
DAVID is able to ignore the scan's borders at a set range for alignment. Means there is already some algorythm to detect borderfaces and their neighbors. What if you'd implement a second limit (means: IGNORE faces nearer than parameter1 AND farer than parameter2 away from the scanborders). This should give the same result as I did manually above. This restriction should exclude a large portion of the surface to be aligned....
MagWeb
Moderator
 
Posts: 2499
Joined: Wed Jul 18, 2007 8:48 pm

Re: Cropped Duplicate Align Strategy

Postby Sven » Tue Dec 24, 2013 12:10 pm

Right, Gunter, sounds promising. We'll have to test that. Maybe computing the border distance for entire meshes takes a lot of computation time...
The limits should probably be relative, not absolute values.
I'll try it!
Sven
User avatar
Sven
DAVID Developer
 
Posts: 1541
Joined: Mon Jan 22, 2007 8:59 am
Location: Braunschweig, Germany

Postby hal » Sat Dec 28, 2013 4:44 pm

Hello Gunter,
Thanks to share the trick. It is good one.
I've tried a similar way time ago, and I can confirm that it works nicely. Just make longer the jobs, especially if we have many of them to align.
So... we can just wait the "SelectRelevantContacts" function. :wink:

Best regards,
Mattia
User avatar
hal
Moderator
 
Posts: 1527
Joined: Thu Mar 01, 2007 4:08 pm
Location: Cavaria con Premezzo (VA), Italy.

Re: Cropped Duplicate Align Strategy

Postby hunkatibor2 » Thu Aug 14, 2014 10:06 am

Hi Sven
Select the mesh are interested in software how bad is it? When will I start to do what we can.
Tibor
http://forum.david-3d.com/viewtopic.php?f=7&t=6215
This is a problem with the porcelain and ceramics are examined.
Intel Xeon E5410 2.33GHZ 16GB RAM Windows 7 64bit David 4.5.0
2 Basler A622F camera 2 Tamron Asfirical AF 28-80mm lense
LG DLP Projector PF1500G Canon D40 Tokina 17-70mm 1:2.8-4,5
Humanti Calibration panel system and Alignmed&Two camera
hunkatibor2
 
Posts: 1408
Joined: Tue Mar 16, 2010 4:26 pm
Location: Hungary

Re: Cropped Duplicate Align Strategy

Postby volume » Sun Sep 14, 2014 3:38 pm

Good tuto, I do the same now and get a better alignment, but not perfect
SelectRelevantContacts waiting
volume
 
Posts: 51
Joined: Wed Nov 18, 2009 9:35 pm


Return to Shape Fusion

Who is online

Users browsing this forum: No registered users and 1 guest

cron