Capture the evidence the model will be built from
ASCAND records a suitable object as it rotates on the coded turntable while the smartphone remains steady. The resulting video is not merely a presentation of the object. It is the ordered visual evidence from which later processing can derive geometry.
Consistency matters more than dramatic photography. The object must remain stable, relevant surfaces must stay visible, and the recording must preserve clear observations throughout the rotation. Processing can interpret and refine what was captured, but it cannot fully recover information that never entered the camera.
The camera stays steady while the object rotates
ASCAND uses controlled rotational acquisition. The smartphone is held in one stable position while the turntable presents successive sides of the object to it. Video records these views in a predictable order without requiring the operator to move a camera by hand around the object.
This stable camera-to-turntable relationship creates a consistent observational framework. Where the applicable workflow uses it, visible information on the coded outer region can also help associate captured observations with turntable orientation.
Keeping the camera steady does not make every object easy to reconstruct. Focus, illumination, object stability, surface behavior, framing and visibility still matter. The benefit is that these factors are managed within a controlled physical relationship rather than combined with arbitrary handheld movement.
A controlled capture in seven steps
1. Assess and prepare the object
Start with the object rather than the recording controls. Check its size, support, stability, surface, visible features and geometry. Transparent, reflective, very dark or visually ambiguous surfaces can be difficult for optical reconstruction. Deep recesses and overlapping parts may remain hidden from a single camera level or object orientation.
Preparation should make the object consistently observable without changing the geometry you need to preserve. Use the dedicated suitability guidance before treating a difficult material or surface.
2. Arrange the capture station
Place the turntable on a stable surface and establish a clear working area around it. The smartphone holder, turntable and object must remain mechanically stable throughout the recording. Vibration or movement changes the acquisition geometry and can reduce the consistency of the observation set.
Use a controlled background and soft, even illumination. Avoid a setup in which people, shadows or surrounding objects move through the capture area during the rotation.
3. Place the object securely
Position the object so it remains stable as the platform rotates. Consider both its support footprint and the surfaces the camera needs to see. A larger platform may provide more physical support, but it does not automatically improve framing or reveal hidden geometry. A smaller-object configuration may help an appropriate object occupy more of the usable image area, but it does not turn the system into a microscope.
Rotate the platform carefully before recording to check for wobble, clearance problems and changing occlusions.
4. Keep the smartphone fixed
Secure the smartphone so its position and viewing direction do not change during capture. ASCAND is designed around a fixed camera and a rotating object, not an arbitrary handheld orbit.
The chosen position must keep the object and the required turntable region observable throughout the rotation. Do not touch, adjust or refocus the phone while recording. If the setup must be changed, stop and begin a new controlled capture rather than introducing movement into the existing sequence.
5. Set framing, focus and illumination
Frame the complete object and the complete coded turntable with clear margins around them. Check the view over a full manual rotation, not only from the starting angle. Protrusions or an off-centre object may leave the frame later.
Focus must remain crisp on the relevant object region. Illumination and exposure should remain consistent as the surface turns. The objective is stable computational input, not a cinematic image. Hard reflections, deep shadows, rapid brightness changes and poor separation from the background can make later interpretation more difficult.
6. Record a complete controlled rotation
Start the video and allow the object to pass through the complete rotation required by the current capture procedure. Keep the camera, support and lighting unchanged. The ordered video sequence gives later processing multiple observations while preserving constant camera geometry.
Do not infer a recording duration, rotation speed or camera setting from older material. Follow the current instructions for the hardware and capture path in use. Incomplete rotational coverage leaves viewpoints absent from the original evidence.
7. Inspect the source video
Before uploading, review enough of the video to identify avoidable faults. Confirm that the object stayed still relative to the platform, remained in frame and in focus, and completed the intended rotation. Look for vibration, motion blur, interrupted illumination, obstruction and unexpected background movement.
For a capture path that depends on the coded region or a projected laser line, confirm that the relevant evidence remains observable. If a clear acquisition fault is present, a new capture is usually more useful than asking later processing to recover absent information.
What the camera needs to see
A suitable capture preserves useful evidence across the complete rotation. The starting frame alone is not enough. The object may become wider from another side, a protrusion may cross the image boundary, or a reflective surface may change dramatically under the same light.
For a clear instructional setup:
- The complete object remains inside the frame.
- The complete turntable and outer coded region remain visible where required.
- Clear margin surrounds the complete rotating setup.
- Focus remains crisp on the relevant object region.
- Illumination and exposure remain stable.
- The object stays visually separable from its surroundings.
- Nothing passes between the camera and the object.
Problematic input includes cropped geometry, incorrect focus, vibration, moving shadows, incomplete rotation and an object that shifts on the platform. A clean-looking frame is not by itself proof of a successful scan, but these checks protect the quality and consistency of the observations supplied to reconstruction.
Read capabilities and limitations
Capture requirements depend on the selected scan path
Vision capture
Vision reconstruction uses image-derived evidence, including object silhouettes and visual information present in the recorded views. Clear boundaries, stable framing and consistent observations are therefore important. Vision capture does not require the optional laser hardware, but it remains limited by what the images reveal.
Laser capture
Laser-assisted capture uses an optional, correctly aligned laser setup. The camera must observe the projected line as the object rotates so geometric evidence can be derived through calibrated triangulation. Laser visibility depends on the surface, illumination, focus, alignment and exposure. Follow the current setup and safety instructions rather than improvising the configuration.
Combo capture
A Combo workflow can use complementary vision and laser-derived evidence. Its purpose is to bring different observations into the wider reconstruction process, not to guarantee that one method repairs every limitation of the other. Capture each required source according to the current workflow and preserve the controlled geometry of the setup.
One rotation cannot reveal every surface
The camera can reconstruct only from surfaces it can observe. The underside resting on the platform is hidden. Deep recesses, overhangs and overlapping parts may also conceal geometry as the object rotates. A complete horizontal rotation does not automatically produce complete coverage of the top, bottom and every cavity.
Repositioning the object can expose surfaces that were hidden in the first orientation. A complementary scan may then provide additional geometry, and Multi Merge can align overlapping reconstructed regions into a more complete representation.
This works only when the scans contain sufficient usable evidence and overlap. Merging cannot invent a surface that was absent from every capture. Plan complementary orientations around the missing views and preserve enough shared geometry for later alignment.
Learn about Multi-Scan and Multi Merge
Check the recording before you upload it
- The object remained stable on the platform.
- The smartphone and holder did not move.
- The intended rotation was completed.
- The complete object remained inside the frame.
- The required turntable region remained visible.
- Focus stayed crisp on the relevant region.
- Illumination and exposure remained reasonably consistent.
- No hand, person or moving object obstructed the view.
- The selected capture path recorded its required evidence.
- Important hidden surfaces have been noted for a possible complementary scan.
This check can prevent avoidable processing attempts. It cannot guarantee a successful model, because suitability and reconstruction also depend on the object and the evidence available to the selected processing path.
When the recording is ready, continue to processing
The source video can now become input to a Scan Project in 3D-Scan.Online. The next stage explains how captured material enters the browser-based workflow and how the selected reconstruction path processes the available observations.