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

Comparions - David vs Photofly/ARC3D/My3DScanner

Comparions - David vs Photofly/ARC3D/My3DScanner

Postby shoesam » Mon Oct 17, 2011 2:37 pm

Hello all,

Has anybody out there done an actual comparison of DAVID to these online Photo-to-Mesh web portals?

thanks,

Sam
shoesam
 
Posts: 39
Joined: Tue Jun 07, 2011 7:36 pm

Re: Comparions - David vs Photofly/ARC3D/My3DScanner

Postby MagWeb » Mon Oct 17, 2011 6:41 pm

Hi,

maybe I´ve not the biggest experience in doing a photogrammetry scan. But all I tried was poorer by far than a scan by DAVID.
Nowadays I cann´t get almost equal results as DAVID scanning results. If there´s anybody out there who can, please post and we could do a comparison.

Photogrammetry (at its current state) is a possibility to capture big scenes where DAVID sucks (e.g to capture a whole building) but on things that fit into a corner.... An advantage of photo based systems is the better quality of texture (for you can process high res images). Its mesh-results might become better and better in the future . Nowadays, IMO, it is only an option if you are interested in a rough mesh that carries a good texture (e.g. for 3D web presentation of a product).

Sure, there are some stunning photogrammetry results of body scans out there. But, as far as I can see, those are made using a bunch of calibrated and simultaneously triggered cams, not only a single digicam you move after each shot.

To add an affordable photo-based local application, see: www.agisoft.ru

just to add my 2 cents

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

Re: Comparions - David vs Photofly/ARC3D/My3DScanner

Postby shoesam » Tue Oct 18, 2011 11:23 am

Greetings Magweb,

Here are some results I got from Photofly......I did quick comparisons to the same object in David, and for Macro comparison, it was pretty good.

I took this mesh, and overlayed one from David, and although I cannot get actual measurements on the differences, it was actually pretty good.
photfly test textured.jpg

photfly test .jpg
shoesam
 
Posts: 39
Joined: Tue Jun 07, 2011 7:36 pm

Re: Comparions - David vs Photofly/ARC3D/My3DScanner

Postby hunkatibor2 » Tue Oct 18, 2011 1:07 pm

Hi Soesam
With which software you prepared the coloured sock last????
Tibor :roll: :roll: :roll:
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: Comparions - David vs Photofly/ARC3D/My3DScanner

Postby shoesam » Tue Oct 18, 2011 2:00 pm

Gunter,

I tried the AGISOFT program........very good results......just CPU intensive.....results were very comparable to the others. TY for the link

Tibor,
I scanned foot in DAVID, then used RESURF in Rhino to develop smooth surfaces. Cut with 3-axis CNC in two parts and assembled.


Sam
shoesam
 
Posts: 39
Joined: Tue Jun 07, 2011 7:36 pm

Re: Comparions - David vs Photofly/ARC3D/My3DScanner

Postby MagWeb » Tue Oct 18, 2011 6:25 pm

Hi Sam,

Your result looks quite good. Seems the color information of the sock helps a lot to get the images aligned.
To compare your results you could:

- load both results, the photogrammetry result and the fused DAVID mesh in DAVID Shapefusion.
- Align both meshes using the free mode.
- Save both meshes without fusion as single scans A and B
- Load both scans to Rapidform Explorer by multiselection.
- Go to MenueBar/Measurement/MeshDeviation
- In the upper left of the 3D viewer there´s the comparison interface now.
- choose one mesh as target (should be in the upper textbox now)
- choose the other mesh to compare (in the lower textbox)
- click on the little arrow to continue
- Rapidform now shows a colored mesh deviation. If it is of a same color all over, modify the scale of the vertical color bar to show smaller differences.
- Hovering over the meshes with the mouse now, gives measurements of the distance between the surfaces.

Just one way to compare results.

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

Re: Comparions - David vs Photofly/ARC3D/My3DScanner

Postby chris_lwgr » Tue Oct 18, 2011 6:31 pm

Actually if you use Active photography (use a projector to project some noise patterns on objects)you can have very good results with Photoscan(Agisoft).
Time taken for reconstruction about 12-15 min .Raw scan, no processing (smooth ,bump maps etc) only cleaning some artifacts from borders.
You can achieve quicker times if you reduce the number of photos.
With 2-3 photos you can have acceptable results in less than 3-4 min.
No calibration needed .The photos must have EXIF data embedded.
This test was made with 5MP photos.
Attachments
active_projection.jpg
scan_shoe_detail.jpg
chris_lwgr
 
Posts: 46
Joined: Fri Aug 28, 2009 5:36 pm

Re: Comparions - David vs Photofly/ARC3D/My3DScanner

Postby icstra » Tue Oct 18, 2011 7:08 pm

Hi Chris,
Could you give us more information about Active photography noise projection?
Regards.
Thierry
MacBookPro 17" I5 Dual Core 2xSSD 128Go, RAM 8Go, OSX Lion, Bootcamp, Win7 X64
EOS 550d and Logitech HD Pro C910
LG HX300G 1024x768

www.icstra,ch
icstra
 
Posts: 44
Joined: Sat Sep 10, 2011 9:12 am
Location: Châtel-St-Denis Switzerland

Re: Comparions - David vs Photofly/ARC3D/My3DScanner

Postby MagWeb » Tue Oct 18, 2011 8:17 pm

But if you use some noise projection:
Is there any advantage compared to a SL scan? You also need certain ambient light conditions...
All these images have to use a single position of the projector. The projected patterns have to be equal in all grabbed images you process. Means, once again you have to do partial scans, align them and finally fuse them....maybe you can get data in projection shadows too. Also grabbing the shots will take some time .... DAVID grabs and processes a sequence in seconds not minutes
No doubt, Chris, this is an awesome result for photogrammetry and I never thought about that trick , but photogrammetry claims to use images only.
MagWeb
Moderator
 
Posts: 2499
Joined: Wed Jul 18, 2007 8:48 pm

Re: Comparions - David vs Photofly/ARC3D/My3DScanner

Postby chris_lwgr » Tue Oct 18, 2011 8:18 pm

You will need a projector and a noise pattern.
Project the noise to the object and make sure to focus (projector).
Take the pictures from different angles and use them to make the reconstruction.
With this method assist the program to find more matching points between the photographs.
Make sure to take good pictures with correct exposures and iso so to avoid the loose of the pattern on the object.
Attachments
noise_pattern.jpg
chris_lwgr
 
Posts: 46
Joined: Fri Aug 28, 2009 5:36 pm

Re: Comparions - David vs Photofly/ARC3D/My3DScanner

Postby chris_lwgr » Tue Oct 18, 2011 8:24 pm

You are correct Gunter.No advantage over SL scanning and the pattern must be in the same position for all the photos.
It was just an experiment to see if i can take better results than using plain photos.
Of course this will not work in external environment but if the objects have textures with great detail and contrast maybe will.
The only advantage that i see is that you dont need calibration and very expensive camera (like Ueye , BlueFox etc).
Cheers,

Chris.
chris_lwgr
 
Posts: 46
Joined: Fri Aug 28, 2009 5:36 pm

Re: Comparions - David vs Photofly/ARC3D/My3DScanner

Postby dickda1 » Tue Oct 18, 2011 8:30 pm

Here is a flyby of a 3D image I created with PhotoFly using about 40 still camera images. The stitching of the pictures went very well - flat lighting and sharp photos are a must for success. I swung the virtual camera around rather carelessly, but it does demonstrate quite a complete model. I can fly through the head and look out from the inside. The mesh was generated automatically.
http://www.youtube.com/watch?v=T-Ye8uk0SUM

I follow developments in photogrammetry, in particular development work done at Stanford here in the States and by colleagues at EPFL in Lausanne.

This is in fact what brought me to DAVID.

Photofly is not quite there yet. If I wanted to scan something about the size of a human, I still would choose DAVID over PhotoFly.

-Dick
User avatar
dickda1
 
Posts: 37
Joined: Tue Dec 22, 2009 1:23 am
Location: San Francisco

Re: Comparions - David vs Photofly/ARC3D/My3DScanner

Postby MagWeb » Tue Oct 18, 2011 8:47 pm

Thinking over it...

Just to add an advantage of Chris's experiment:
You could use a simple still image projector instead of an expensive data projector.

@ texture quality:
Think, the texture will show the projected pattern, which helped to align the images to get better depth information. And aligning different scans and fusing their textures should give some interferences of the different scan textures. Is this significant?

Chris, did you already try to align different scans. Do they fit together? I ask this really interested and not skeptical at all.

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

Re: Comparions - David vs Photofly/ARC3D/My3DScanner

Postby chris_lwgr » Tue Oct 18, 2011 9:17 pm

I did some experiments
I t came with different scale but mesh lab han an option to register and scale the objects
Worked fine.I need more tests to be done.
As for the textures you capture some without the noise (simple quality photos) and use them with Zbrush(or any 3d painting software like mudbox or Mari) to project the clean texture on object
Last edited by chris_lwgr on Tue Oct 18, 2011 9:23 pm, edited 1 time in total.
chris_lwgr
 
Posts: 46
Joined: Fri Aug 28, 2009 5:36 pm

Re: Comparions - David vs Photofly/ARC3D/My3DScanner

Postby MagWeb » Tue Oct 18, 2011 9:22 pm

Did you consider reflecting your projection all around using a mirror system?
For it has not to be coded but arbitrary, you could get an all around scan this way.
MagWeb
Moderator
 
Posts: 2499
Joined: Wed Jul 18, 2007 8:48 pm

Next

Return to Software

Who is online

Users browsing this forum: No registered users and 1 guest