Convert PTS to STL

A 3D printer, CAD system, or mesh editor may require a triangular STL surface instead of the scanned point cloud stored in a PTS file. The conversion must reconstruct faces from the points; renaming .pts to .stl does not produce usable geometry.

What the PTS format is

In 3D modeling, PTS usually means an ASCII point-cloud format exported by terrestrial laser scanners, 3D scanners, surveying applications, and scan-processing software. A file commonly starts with a point count, followed by records containing X Y Z coordinates and, depending on the producer, intensity and RGB values. PTS has no single universal record layout, so import settings must match the source software's documentation.

PTS stores sampled positions, not connected faces. Noise, duplicate points, gaps, overlapping scans, and unorganized points may be present. An STL exporter therefore has to estimate a surface before it can write triangles.

What the STL format is

STL represents a surface as triangular facets and is accepted by most 3D-printing slicers, printers, rapid-prototyping systems, CAD programs, and mesh editors. Binary STL is normally preferable because it is substantially smaller than ASCII STL, although both contain the same basic triangle data.

STL does not retain PTS intensity, RGB color, textures, materials, scan stations, or point-level structure. It also has no reliable unit declaration; millimeters are common in 3D printing, but the import and export scale must be checked explicitly.

Convert the point cloud with CloudCompare

CloudCompare is a suitable free desktop option because it can import point clouds, remove outliers, compute normals, reconstruct a mesh, and export STL.

1. Install CloudCompare from the official project and open the file through File → Open. Select the correct coordinate, intensity, and color columns when the importer asks for them. Confirm the preview, coordinate order, scale, and orientation.
2. Keep only the scan region required for the model. Use segmentation to remove other scans, registration targets, floor areas, and obvious outliers; apply a statistical outlier filter when isolated noise is widespread.
3. Select the cleaned cloud and calculate normals through Edit → Normals → Compute. Choose a neighborhood large enough to include several nearby samples but not unrelated surfaces. Normal quality strongly affects reconstruction.
4. Run Poisson surface reconstruction through the installed Poisson reconstruction plugin. Plugin names and menu locations vary between CloudCompare builds. Increase reconstruction depth only when the point density supports finer detail; higher values increase memory use and can reproduce noise.
5. Inspect the generated mesh. Remove floating components and unwanted background surfaces, and check areas outside the scanned region because Poisson reconstruction can close holes or create geometry across sparse data.
6. Select the mesh and use File → Save or File → Save As to choose STL. Select binary STL unless an ASCII file is specifically required.

Use MeshLab as an alternative

MeshLab is free and can perform the same workflow when the installed build supports PTS point-cloud import. Open the file through File → Import Mesh, calculate point-set normals through Filters → Point Set → Compute normals for point sets, then use Poisson reconstruction or Ball Pivoting under the point-set surface-reconstruction filters. Export the generated mesh through File → Export Mesh As and select STL.

Poisson reconstruction generally suits closed organic objects with fairly uniform coverage and tends to fill holes. Ball Pivoting can follow sampled surface detail more closely, but it requires an appropriate ball radius and usually leaves gaps where the cloud is sparse. Menu names and available filters differ between MeshLab releases.

Online conversion and automation

Use an online service only when its current input list explicitly accepts 3D point-cloud PTS and its workflow includes surface reconstruction. Many browser converters translate one existing mesh format to another; they cannot create faces from unconnected points. A service such as Aspose.3D can be tested for a non-sensitive file when its upload page lists PTS support, but verify the resulting mesh, scale, and reconstruction settings rather than assuming that an extension conversion preserves the scan. Do not upload confidential survey, architectural, or industrial scans without checking the service's retention policy.

Quality and compatibility limits

• This is a reconstruction, not an exact format translation. Point spacing, noise filtering, normal orientation, reconstruction method, hole filling, and mesh settings determine the result.
• Features smaller than the local point spacing cannot be recovered reliably. Sparse regions can create holes, bridges, or false surfaces; noisy dense regions can produce rough or oversized geometry.
• For printing, inspect the STL in a mesh checker or slicer for watertightness, consistently oriented normals, non-manifold edges, self-intersections, disconnected fragments, and the intended physical dimensions. Decimate or remesh an unnecessarily dense result after preserving a higher-resolution master.
• Retain the original PTS and, when color or intensity matters, export a color-capable format such as PLY in addition to STL. STL cannot carry that scan information.