So, I have been using my Miraco a lot recently at work to scan some of the sculptures our sculptors are doing, and as of late some of our projects have been rather large sculptures that quite simply scanning in near mode will cause a lot of scan drift, but far mode can accurately capture. So currently to resolve this issue, I first scan the sculpture in far mode, or use my company’s Artec Eva, then I will do spot scans using near mode single shot to get multiple scans that I then align to the lower res scan that I am confident is dimensionally accurate. Then it kinda occured to me, With single shot mode as long as you have a point cloud, the miraco can continue the scan at any time. Since we now have the ability to do global marker scans to scan large objects, what about a global point cloud? Say you first scan an object with far mode to get a base point cloud that can be then used in a near mode single shot high res scan to help with the alignment process as well as fill in any gaps that may have been missed during the near mode capture process. This would be especially helpful in areas of a scan that have limited features to register to. I don’t see a reason that this process couldn’t work on the current hardware, as you would just be importing a point into a scan project and aligning to that as you scan.
near mode is for scanning very small objects and sections , scanning bigger section will result in drifting, that where Global market can help .
Miraco Plus fixed the issues with drifting by using photogrammetry to pre capture the whole object so literally having big and small surface elements and details regardless the distance or modes . That is exactly what you needed .
For that reason there will be nothing else added to the base version of MIRACO as its hardware is already on its limitation to what it can do .
I did tested the photogrammetry on the basic MIRACO but the results was unacceptable for that reason a new better version was build ( MIRACO PLUS )
Regarding you other idea about single shots , single shots are not really capturing 3D , they are single shots of 2D surfaces and require at least 10 frames at different angles to capture the 3D dimension correctly , compare single shot to a bump or displacement maps where the details are only captured in Z direction per one shot .
You idea of course is great , but wishful thinking as it will not happen to already released product that reached its hardware limitation .
This is the basic description of my photogrammetry workflow, where you start far away and work in for detail only where it is needed! unless you have a few terabytes of RAM ( working on it ) or the armature method where you crate the frame work to hang the detail scans off
What about placing a few markers only in the detailed areas?
That way, the Miraco could easily regain tracking locally without having to follow the entire object. I’d first capture a global scan in Far mode for the overall shape, then do targeted Near scans on those marked zones, aligning them later with the main mesh.
I wonder if this approach could work.
Yes you can cut off parts and replace them later with higher resolutions parts, I do it all time , scanning full body then merge details like hands and face … It works but you will get small "scar " around the areas that need to be edited , since merging high and low accuracy models will never create smooth transition due to different resolution and details so it need to be manually smooth out and edited in 3D editing software .