Human Molar
This is a test of a hard-surface object with little features and it has been a test of the Alignment Transfer method of the MICROscanR.
Alignment Transfer is the method where a calibration object is being shot from the precise same viewpoints as the object to be scanned (human molar in this case - I know, interesting only for dental students perhaps). The calibration object is shot without Focus-Stacking and a precise camera alignment is being found using Reality Scan. That camera alignment is then transferred to the molar and from these camera positions a dataset of the molar is being created in RS for 3DGS. In order for this method to succeed, the repeatability of the viewpoints between the calibration- and the molar scan has to be precise to a few micrometers (translation) and a few arcsecond of rotation (1 arcsec = 1/3600 degree). This what the MICROscanR has been built for.
Capture
_____________________________________
- Camera: SonyA7R5 camera (61 Mp) - 1/40 sec. - f16 - ISO 100.
- Lens: Laowa 90 mm 1-2 x Macro at ~ 0.7 x
Scan
_____________________________________
- Scanned with: MICROscanR
- Total viewpoints: 417
- Stack-depth: 4-15 frames (dynamically assigned per viewpoint).
- Total pictures: 3948
- Scan time: 59m.
- Scan date: 15-09-2026.
Lighting
______________________________________
- Lights: Object in diffuse light field between 2x side-beam round (28 cm) LED panels, 2 x 24 Watt
+ 1x Microscope LED beamer, with diffuse focus on subject fixed on front of lens.
- Background: black velvet.
Post processing
______________________________________
- Focus-stacked with Helicon.
- Camera alignment + masking + 3DGS dataset created with Reality Scan
- 3DGS made with Lichtfelt (masked, 3840 px image width,1M splats).