Choose the evidence the object can reveal
ASCAND supports more than one route from a rotating object to reconstructed geometry. Vision, Laser and Combo workflows use different kinds of evidence. Multi-Scan adds complementary object orientations and merges overlapping reconstructions when one orientation cannot expose enough of the surface.
These are not four quality levels. The appropriate workflow depends on the object’s shape, the surfaces that must be represented, how those surfaces interact with light, and which optional hardware is available. A more complex workflow is useful only when its additional evidence addresses a real limitation.
Different reconstruction paths share one controlled capture foundation
ASCAND keeps the camera steady while the coded turntable rotates the object. The recorded video contains a structured sequence of viewpoints, and the coded reference around the turntable helps associate observations with rotation. The capture is uploaded to 3D-Scan.Online, where the selected reconstruction path interprets the evidence and prepares reviewable results.
What changes between scan types is not this foundation, but the evidence used to infer geometry:
- Vision works from image-derived boundaries and silhouettes.
- Laser observes a projected line and reconstructs surface points through calibrated triangulation.
- Combo uses complementary Vision and Laser evidence from the same object.
- Multi-Scan uses additional object orientations and aligns overlapping reconstructed geometry.
The optional Laser Extension Kit is required only for laser-assisted capture. Vision uses the standard camera-based capture path.
Learn why structured rotation matters
Vision, Laser, Combo and Multi-Scan at a glance
| Workflow | Primary evidence | Additional requirement | Useful when | Principal boundary |
|---|---|---|---|---|
| Vision Scan | Object boundaries and silhouettes across the rotation | No laser hardware | The outer form is the main requirement and the object separates clearly from its background | A visual hull cannot reproduce concavities that never affect a visible silhouette |
| Laser Scan | Visible projected laser line interpreted through calibrated geometry | Aligned Laser Extension Kit and suitable laser capture | Recesses, grooves or surface relief require evidence beyond the outer silhouette | Only illuminated and visible line observations can contribute; alignment and surface response matter |
| Combo Scan | Complementary Vision and Laser evidence | Both laser-off and laser-on evidence in a supported configuration | Outer-form context and laser-derived depth should support one another | Combining methods cannot restore geometry absent from both evidence sources |
| Multi-Scan / Multi Merge | Overlapping reconstructed geometry from different object orientations | Multiple suitable captures with enough common geometry | The support surface, underside or self-occlusion hides important regions in one orientation | Alignment needs overlap and cannot invent regions missing from every scan |
The table is an orientation aid, not an automatic object classifier. Surface response, framing, calibration, capture quality and intended use still affect the result.
Vision Scan: reconstruct the visual hull
A Vision Scan derives a three-dimensional outer form from image observations recorded as the object rotates. In simplified terms, processing begins with a volume that could contain the object. Each valid silhouette rules out parts of that volume that must be background. After observations from multiple rotation angles are evaluated, the remaining volume represents the object’s visual hull.
This makes Vision useful when the external form is the main requirement and the boundary between object and background can be interpreted consistently. It does not require the optional laser hardware and benefits from the controlled relationship between the fixed camera, rotating object and coded turntable.
Its defining limitation follows directly from the method. A deep opening, undercut or internal recess may remain inside every silhouette even though it is empty in the physical object. If no view reveals that empty region as background, silhouette-based carving has no evidence with which to remove it. The result can therefore preserve the outer envelope while closing or simplifying hidden concavities.
Vision is not the same as conventional multi-angle photogrammetry. ASCAND uses a structured rotating-object capture sequence and documented image-derived reconstruction routines. Surface transparency, reflectivity, low contrast or segmentation difficulty can still affect what the images reveal, so no material should be treated as automatically easy.
Understand Vision scanning and voxel carving
Laser Scan: triangulate the observed laser line
A Laser Scan adds an actively projected line to the controlled rotation. The camera detects where that line appears on the visible object surface. With the calibrated relationship between camera, laser plane and turntable orientation, processing can intersect camera rays with the laser plane and reconstruct three-dimensional surface points.
This provides geometric evidence that is different from an outer silhouette. When the laser line reaches a visible recess, groove or change in surface relief, triangulation can represent depth that a visual hull alone may not resolve.
Laser capture is not LiDAR, and it does not see through the object or around an occlusion. A point can contribute only when the projected line reaches the surface and the camera observes it reliably. Steep angles, shadows, self-occlusion, reflections, scattering or weak line contrast can create gaps or outliers.
The method is also sensitive to the physical relationship between the camera and laser. A shifted or tilted laser changes the assumed triangulation geometry and can distort the reconstruction. Correct assembly, alignment and current safety guidance are therefore part of the workflow. Laser output should not be presented as industrial metrology without verified measurement evidence.
Learn the laser-triangulation principle
Combo Scan: use complementary Vision and Laser evidence
A Combo Scan records the evidence needed by both Vision and Laser processing. This requires supported laser-on and laser-off capture of the same stable object. The clean, laser-off observations support Vision reconstruction; the laser-on observations support line detection and triangulation.
The value is complementary rather than additive in a simple numeric sense. The Vision result can provide an outer-form reference. Laser-derived points can contribute depth observations in regions where the line is visible. Documented combination stages can use those relationships to reject implausible laser outliers, reconcile geometry or provide Vision-derived support where the laser evidence contains gaps.
Combo does not mean that every point is measured twice, nor that the result automatically inherits only the strengths of both methods. The capture must remain geometrically compatible, the laser must be aligned, and both evidence paths must be usable. A surface hidden from the camera and laser in every contributing observation remains unsupported.
Choose Combo when both kinds of evidence address the object’s actual requirements—not merely because it is the most elaborate option. The current interface and hardware configuration determine whether the workflow is available and how its captures are recorded.
Understand complementary Combo evidence
Multi-Scan and Multi Merge: add complementary orientations
Every turntable scan has a support relationship. Part of the object rests on the platform, while overhangs, undersides or self-occluded regions may remain difficult to observe. Rotating the object provides views around one axis; it does not automatically reveal a surface that remains hidden throughout that orientation.
Multi-Scan addresses this coverage problem by capturing the object again in a different stable orientation. Each capture is reconstructed through an appropriate supported method. Multi Merge then uses common geometry to align the separate results and combine their complementary regions.
Overlap is essential. The reconstructions need enough reliable shared shape for alignment. If the second orientation exposes only new geometry without a trustworthy common region, the system has too little evidence to determine how the results belong together. Symmetry, repeated shapes, noise or strong reconstruction differences can also make alignment ambiguous.
Multi-Scan is therefore a coverage strategy, not a sensor type and not a repair command. It can add observed surfaces and reduce dependence on one resting orientation. It cannot recover a region missing from every capture, and it cannot make incompatible or poorly reconstructed scans agree automatically.
Learn about complementary orientations and Multi Merge
Start with the limiting feature, not the most complex method
Use the object and intended result to guide the first choice:
- Is the outer form the main requirement? Begin by evaluating Vision. Confirm that the object boundary can remain clear across the rotation and accept that hidden concavities may be simplified.
- Do important grooves or recesses need direct depth evidence? Evaluate Laser, provided the projected line can reach those surfaces, remain visible to the camera and the optional hardware can be aligned correctly.
- Would outer-form context and laser-derived depth complement one another? Evaluate Combo and confirm that compatible laser-off and laser-on evidence can be recorded without moving the object.
- Are important surfaces hidden by the support or first orientation? Plan a complementary orientation and Multi-Scan workflow with sufficient shared geometry for alignment.
- Does the object create several independent limitations? A sequence may involve Combo evidence within each orientation and Multi Merge across orientations, but each added stage also creates more capture and validation requirements.
No decision tree can replace a practical object assessment. Consider geometry, surface behavior, stability, background separation, laser visibility, framing, required coverage and downstream purpose together.
Every method remains limited by observation
All ASCAND workflows depend on usable capture evidence. The object must remain stable. The camera must remain fixed, focused and correctly framed. Illumination and background conditions must support the selected method. Coded references must be visible enough to interpret the structured rotation. Laser-assisted workflows additionally depend on a visible projected line and correct alignment.
Processing can interpret, filter, combine and surface the evidence that was recorded. It cannot reveal the underside of an object that was never exposed, restore a laser line the camera never observed, or prove accuracy from appearance alone. Meshing can close geometry, but closure is not the same as observation.
When a result is unsuitable, identify the earliest limitation: object preparation, capture, method selection, alignment, reconstruction, merging or output preparation. Changing the file format does not correct missing evidence.
Choose the workflow, then capture deliberately
Select the simplest workflow that records the evidence your intended use requires. Prepare the object and station for that method, follow the verified tutorial, and inspect the result against its purpose. Add laser-assisted evidence or another orientation when it solves a specific observed limitation.
Browse tutorials · Compare ASCAND configurationsWhen processing completes, your Scan Project may contain one or more representations of the reconstructed object. Review them before choosing a download. The aim is not simply to find a file that opens. It is to decide whether the captured and reconstructed geometry is suitable for what you intend to do next.