Multitiple scan locations for accuracy

Hi, I am new here and looking at getting a Revopoint to scan sculptures, which are between 0.5m and 1.5m in height for a art project.

I currently use photogrammetry however since the artwork I am scanning here is very smooth I am unable to get enough point density on the surface to create good scans that way.

I am trying to find out if i can scan a single object from different heights by mounting the scanner on a tripod and moving it up and down ?

this way i can do bottom-up to get detail under that artwork, pointing up, face front for the full piece, facing dead ahead, and top down for detail on the top of the artwork.

If i do 3 or 4 different levels, and angles, using a tripod and a large turntable (tripod wont physically move it will be static) - will they all align properly to give me a high quality scan ?

Also if i am shooting this way, would a Pop 2 be better or a Range ?

Thanks in advance for your help !

Thanks !

I don’t have any experience with the range - however - I do have a POP2. I would probably recommend in your use-case a Range as scanning things greater in size than 300mm x 300mm x 300mm would be a bit difficult regularly (it can do it but takes a lot of care/skill).

POP2 & Mini are good for less than 300mm in my opinion and a Range would be better for bigger objects.

Note: A range will capture less detail - but if your features are bigger than ~10-20mm you should be fine I would think and it will be far better than Photogrammetry.

Range would be best choice since it produce better details in Feature mode at distance of 30cm than POP2 in Body mode at 25cm

To scan 0.5m to 1.5m with POP2 would be rather more difficult… you will need the object to have lots of features .
Range needs less features only 8cm apart and it was designed for scanning this size of objects .

I have made some test with range on a statue I’m still learnign but
the result was interesting…



another





2 Likes

Great community and thanks for all the answers ! So Range it is then.

Can you let me know if this will work well on a tripod with a turntable ? My statues are large / heave so I have a industrial table with speed control.

I would be doing around 3 passess… startig at the bottom, poingint up, then moving up to the top pass pointing down. I would do a full 360 for each pass then adjust the tripod for the next pass.

Will this work ok ? not sure with the 3d scanner but with photogrammetry usually we need overlaping frames both horizontally and vertically

thanks again !

1 Like

It’s hard to definitely say it will work without seeing an example.

However your described work flow seems very ideal. Rotating the piece instead of the camera is easier to track and you can merge multiple scans into a single point cloud.

Yes @soulbank that is the best workflow here … my favorite as well
You going to have best results this way .

Got my Range scanner ! Will share my results but so far testing has been great. I do have a question regarding rotation… the scanner software works at 12 fps, what is the best speed to rotate the turntable at and do i need to pause it as it rotates for best results ?

my turntable can pause after a certain amount of degrees have passed, like it can rotate slowly but pause every 10 degrees for a few seconds then continue.

also do you have support for other turntables ? my turntable works the other way like most photography ones, it has a trigger it sends back to the camera to take the photo whenever it stops on its rotation. i need to use my tabel since my items are large and heavy…

Hi @soulbank you don’t pause the turntable , Revopoint scanners are portable scanners that shot multiple frames per second, for that reason please avoid any turntable pausing while scanning , also limit the amount of rescanning of the same surfaces .
Use the turntable at the highest speed and if you need to pause, pause just the scanning process and nothing else .

There is no support for other turntables but you can use any turntable you want .

Remember your last post with your workflow? Just follow it for best results .

Awesome - thanks for the help this will save me a lot of time. Are there any tutorials on how to stitch multiple scans together for best accuracy ?

Also are there any recommendations for overlap degree in the passess when scanning object using multiple passes ?

Thanks again !

The best accuracy will come from your point cloud , for that make sure you scanning all angles at the same distance , changing distanced while scanning will lower the final accuracy and that is completely normal when using structured light scanner.

Use the manual marker point stitching mode for .ore precise result .

You need more but not less than 10% overlapping for each section .

Please remove big parts from point clouds that going to be merged if the part overlapping too much the rest of the object .

You will need also clean the merged scan for Overlapping in case there are some lose points .