.png)
.png)
.png)
A Demo Is More Than an On-Site Preview
An effective demo should use a handheld simultaneous localization and mapping (SLAM) scanner such as the SHARE SLAM S20, SHARE SLAM S20 SE, SHARE SLAM S100, or the SHARE3DCAM C10 Series to capture data at a real site and process it in SHARE PointClouds Studio through the agreed deliverables and checks. The following checklist applies to presales demos, procurement evaluations, and trial runs by delivery teams.
Figure 1: A real-world project pilot test proceeds through preparation, capture, office processing, independent checks, DXF verification, and results documentation.
Pre-Test Preparation Checklist
Project and Model
- Document the project type, area, number of floors, indoor/outdoor scope, and key risk areas.
- Specify the candidate models and the reasons for selecting them: SHARE SLAM S20, SHARE SLAM S20 SE, SHARE SLAM S100, and the SHARE3DCAM C10 Series may be included in testing based on the scenario.
- Assign a data capture operator, office-processing operator, computer-aided design (CAD) reviewer, and customer acceptance representative.
On-Site Route and Data
- Mark the start and end points, loops, stairs, elevators, long corridors, glass, black surfaces, and indoor-outdoor transition areas on the floor plan.
- For multi-segment or repeated scans, reserve fixed corresponding feature points that are identifiable, appropriately distributed, and non-collinear.
- If testing engineering coordinates, prepare a control-point CSV and keep the control points separate from the independent check points.
- If testing post-processed kinematic (PPK), prepare the rover, base station/observation station, and navigation files. Generate an RTKLIB POS trajectory file in an external PPK tool, then select it only as the
Trajectory Sourcein global navigation satellite system (GNSS) fusion throughGNSS Fusion.
Deliverables and Opening Environment
- Confirm the point cloud format: LAS, PLY, PCD, E57, or RCS.
- Use the CAD drawing export, which simultaneously generates drawing exchange format (DXF) and DWG files and includes them in the exported ZIP package, and record the target CAD software and version.
- RVT, IFC, and complete BIM models are downstream professional conversion or modeling outputs. Clearly assign their generation and manual review to professional CAD, building information modeling (BIM), or SketchUp personnel.
- If 3D Gaussian splatting (3DGS) is required, list “photorealistic viewing” as a separate validation item rather than using it as a basis for dimensional, accuracy, or CAD acceptance.
What Steps Should Be Performed During the Demo and Office Processing?
- Complete copy and project creation: Preserve the original directory structure and filenames, and confirm that the data passes validation and can be added to the project.
- Point cloud reconstruction and trajectory inspection: Review task progress, point cloud coverage, and trajectory continuity. When photo verification is required, synchronize the floor plan, trajectory, photo points, photos, and 3D viewport.
- Manual corresponding-point registration: Select the reference point cloud and the point cloud to be registered, pick corresponding points distributed across the entire area, inspect the error, and output
original_name_registered.las. - Coordinate transformation: Match at least three valid control points, or enter confirmed XYZ translation, XYZ rotation, and scale parameters. After verification, output
original_name_converted.las. - Independent quality checks: Perform separate 3D and elevation checks and save the reports. Control points must not also serve as check points.
- Photorealistic viewing: If testing 3DGS, verify the photo prerequisites, graphics processing unit (GPU) memory, and stability in large scenes, and clearly define its viewing limitations.
What Should the Results Record Include at a Minimum?
| Record Item | Required Information |
|---|---|
| Test conditions | Site, weather, route, model, configuration, and workstation |
| Coverage and trajectory | Missing areas, abnormal locations, trajectory continuity, and photo correspondence |
| Registration and coordinates | Corresponding-point distribution, errors, number of control points, parameters, and output filenames |
| Accuracy checks | 3D report, elevation report, independent check points, and project tolerances |
| CAD deliverables | Section width, exported DXF and DWG files, target software version, and issues found during manual review |
| Stability and support | Task success/failure/cancellation, retry results, logs, and issue descriptions |
| Manual effort | Time spent on capture, processing, correction, and review |
Failed or canceled tasks can be retried. When support is required, retain the raw data, project, deliverables, and local diagnostic information. Without an available network, login may not complete, some advanced functions such as 3DGS may not work normally, and logs cannot be uploaded.
How Should Pass Criteria Be Defined?
Pass criteria should be established in writing before the pilot test.
For example: complete coverage of key areas; traceable _registered.las and _converted.las files; independent 3D and elevation checks that meet project tolerances; and an exported ZIP package containing both DXF and DWG files, with both opening in the target software version and passing manual review. Do not substitute “looks good” for quantitative evaluation, and do not generalize the results from one scenario to other projects.
Frequently Asked Questions
Q: Can we schedule only half a day of on-site scanning without office processing? This can provide a hands-on experience, but it is not enough to support a purchasing decision. At a minimum, complete reconstruction, quality checks, and verification that the target format opens correctly.
Q: Why should a demo include manual corresponding-point registration? The ease of aligning segmented, repeated, or cross-area captures directly affects delivery efficiency. Registration should generate a _registered.las file; merging cannot replace registration.
Q: Why does coordinate transformation require at least three valid control points? The software needs enough valid correspondences to calculate the transformation parameters. You should also verify the control-point distribution, XYZ order, and units, and validate the _converted.las file using independent check points.
Q: How should 3DGS be accepted during a pilot test? Evaluate it based on photorealistic viewing, photo quality, and operational stability. Dimensions, accuracy, and CAD acceptance should still be based on the point cloud and inspection reports.
Q: What should be submitted after a test task fails? Retain the complete raw data, project, deliverables, and logs. Record the workstation environment, task stage, and observed error. Failed or canceled tasks can be retried in Task Management.
