CloudCompare is a free, open-source program for processing and comparing 3D point clouds and triangle meshes, and in dental work it is used to measure how far one scan deviates from another. Researchers, dental technicians and students use it for trueness and precision checks: a printed model against its source STL, two intraoral scans taken at different times, or a milled part against its design. If you need a CloudCompare download, the current stable release is 2.13.2 for Windows, macOS and Linux.
The project began in 2003 in a collaboration between Telecom ParisTech and EDF R&D, as Daniel Girardeau-Montaut’s PhD work on change detection in laser scans. It is now an independent GPL project. It knows nothing about teeth, but its registration and distance tools are exactly what a dental accuracy study needs.
CloudCompare at a glance
| Item | Details |
|---|---|
| Latest version | 2.13.2 “Kharkiv” (July 2024); a 2.14 beta was published in September 2026 |
| Licence | Free and open source, GPL; free for any use, including commercial |
| Platforms | Windows (64-bit installer and archive), macOS (Apple silicon and Intel), Linux (Flathub) |
| Developer | Daniel Girardeau-Montaut and the CloudCompare open-source community |
| Official website | cloudcompare.org |
| Best for | Deviation maps and accuracy checks between dental scans, models and printed parts |
What it does
- Reads and writes mesh formats STL, PLY, OBJ, OFF, FBX and VTK, and point formats such as LAS, E57, PCD and ASCII XYZ.
- Registration: manual alignment by picking point pairs, then fine registration with ICP.
- Distances: cloud-to-cloud and cloud-to-mesh, stored as a per-point scalar field and shown as a colour map.
- Statistics on scalar fields (mean, standard deviation, histogram) for reporting deviations.
- Interactive segmentation with a polyline to cut away bases, soft tissue or scan artefacts.
- Point picking to measure distances, angles and add labels.
- Plugins such as M3C2 distance, Poisson surface reconstruction and RANSAC shape detection.
- ccViewer, a light viewer-only companion for quick inspection.
How dental teams use it
Check the accuracy of a printed model
- Scan the printed model and open both the scan and the original STL.
- Crop both to the area of interest (for example, the dentition without the base) with the Segment tool.
- Tools > Registration > Align (point pairs picking): pick three or more matching points such as cusp tips.
- Tools > Registration > Fine registration (ICP) to refine the fit.
- Tools > Distances > Cloud/Mesh Dist with the original as reference; adjust the colour scale and read the histogram.
Compare two intraoral scans over time
- Register the scans on stable areas (for example the palate or unchanged teeth) rather than the whole arch.
- Compute the distance and set a symmetric colour scale so wear or movement stands out.
- Export the coloured mesh or a screenshot for the patient record or a presentation.
Clean up a scan quickly
- Use Segment to cut away unwanted areas, then delete the selected part.
- Remove isolated points with the noise or statistical outlier filters, then save as STL or PLY.
Install and first run
- Windows: use the 64-bit installer or the portable archive from the official release page.
- macOS: builds for Apple silicon and Intel are linked from the official release page.
- Linux: install from Flathub; distribution packages may be older.
- Navigation: left drag rotates, right drag pans, the wheel zooms.
- Gotcha: set units consistently. Dental STL files are normally in millimetres, but some scanners export metres; scale before comparing.
- Gotcha: a distance is only as good as the registration. Always refine with ICP and check the RMS result.
- The 2.14 version is still a beta; use 2.13.2 for work you need to reproduce.
Limitations
- No dental tools, reports or templates; you build the workflow yourself.
- No DICOM/CBCT reading; segment CBCT in another tool and import the STL.
- Limited mesh repair compared with MeshLab.
- Interface is technical and oriented to surveying and engineering users.
- Not a medical device; results are measurements, not diagnoses.
CloudCompare vs alternatives
MeshLab is the better choice for repairing and decimating meshes; CloudCompare is better for deviation statistics and colour maps. Blender and Blender for Dental are for modelling, not measurement. Commercial metrology software such as Geomagic Control X offers automated inspection reports, but CloudCompare covers the core distance workflow at no cost.
FAQ
Is CloudCompare free?
Yes. CloudCompare is free and open source under the GPL, with no paid edition.
Is CloudCompare free for commercial use?
Yes. The project states you can use it as is for any purpose; the GPL conditions apply only if you redistribute or reuse its code.
Is there CloudCompare for Mac?
Yes. macOS builds for Apple silicon and Intel are linked from the official release page.
What is CloudCompare used for?
Comparing 3D datasets: aligning scans and measuring distances between clouds and meshes. In dentistry that means trueness and precision checks of scans, models and printed parts.
CloudCompare vs MeshLab?
Use CloudCompare to measure deviations and MeshLab to repair and simplify meshes. Many labs use both.
How do I measure distance in CloudCompare?
For a single distance, use the point picking tool and select two points. For a full comparison, use Tools > Distances > Cloud/Mesh Dist.
How do I export a mesh from CloudCompare?
Select the mesh in the DB Tree, choose File > Save and pick STL, PLY or OBJ. PLY keeps the colour map of a comparison.
Is CloudCompare safe?
Yes, when downloaded from cloudcompare.org or the project’s GitHub releases. Avoid repackaged copies on download portals.
Last checked against official sources: 26 September 2026 (developer website: cloudcompare.org). Versions and licence terms change — confirm on the developer's site before deploying in a clinic.





