Importing and Exporting Point Clouds in Blender: A Step-by-Step Guide

What you’ll learn in this guide

This guide walks you through bringing point cloud data from a scanner or another 3D application into Blender, checking the key data, and exporting the file again. The example workflow uses the Point Cloud I/O add-on. It supports E57, PLY, LAS/LAZ, PCD, XYZ and PTS formats, and can bring values such as color, normals, intensity and classification from the file into Blender as point attributes.

The aim here isn’t to convert scan data into a polygon model, but to import it as a point cloud, assess its usability in the scene and export it in a suitable format. This distinction matters especially with large scans: the add-on creates the data as Blender’s native PointCloud geometry rather than as a mesh.

Requirements and compatibility

You’ll need Blender and the Point Cloud I/O add-on for this workflow. The source does not specify a required Blender version or account. Since the installation steps are provided through Blender’s Extensions section, start by opening Blender and making sure the main interface is available.

The add-on description says its dependencies are bundled as packages prepared for Python 3.13. Out of the box, supported systems are macOS arm64, Linux x86_64 and Windows x86_64. Intel-based Macs are not directly supported; dependencies may need to be built from source on those systems. Installation also requires neither an internet connection nor a manual pip step.

Check the file format you intend to use beforehand. E57, PLY, LAS/LAZ, PCD, XYZ and PTS are supported. Don’t confuse Blender’s built-in Stanford PLY tool with Point Cloud I/O: the built-in tool creates a mesh, while the add-on creates a PointCloud object.

Step-by-step import and inspection

1. Install the add-on. In Blender, go to Edit > Preferences > Get Extensions. Search for “Point Cloud I/O” and click the Install button in the results. Once installation is complete, have the scan file you want to import ready.

2. Choose the import command for your file format. Go to File > Import in the top menu and select the command for your file type. For E57, choose E57 Point Cloud; for PLY, choose PLY Point Cloud; and for LAS or LAZ, choose LAS/LAZ Point Cloud. Import options for PCD, XYZ and PTS follow the same naming pattern.

3. Inspect the object and its attributes. After import, a PointCloud object appears in the scene. Press N in the 3D Viewport to see the point count and available attributes in the Point Cloud tab. This panel also includes a point radius setting. The add-on sets an initial radius based on the cloud’s bounds and density; adjust it in the panel if needed for visibility or presentation.

4. Verify color and other data. The add-on can create a Principled BSDF material linked to the imported color or normal attribute. If points don’t appear correctly in Material Preview or Rendered view, check the attributes listed in the panel and the data present in the file. If an E57 file contains multiple scans, each scan may become a separate PointCloud object; you can also combine scans using the Merge Scans option.

5. Use the data in Geometry Nodes or shaders. Imported color, normals, intensity, classification and custom fields are stored as point attributes. According to the add-on description, these values can be used in Geometry Nodes or shader workflows. The key is to check whether the data is actually present in the file and whether the add-on lists the relevant attribute after import.

Exporting and avoiding data loss

To export, choose the command for your target format from the File > Export menu. For example, you can use the PLY Point Cloud or LAS/LAZ Point Cloud export options. Check the options in the export window before saving, as not every format carries the same data channels.

The Point Cloud I/O 0.6.0 release notes say that a Source Attributes section for export channels was added in the update dated 11 September 2026. You can select source attributes for channels such as color and intensity; LAS also offers options for classification, return number and number of returns. The default selection is Auto. To omit a channel, select None, or choose an appropriate attribute manually.

Read the warnings in the export window about attributes the target file won’t carry. The add-on notifies you before export when some data won’t be included in the file, and reports this again after the operation. This check is useful when transferring data between applications, so you don’t assume that all information in the source file will be preserved.

Common mistakes

• Confusing Mesh and PointCloud options: Blender’s built-in Stanford PLY tool and the add-on’s PLY Point Cloud tool don’t create the same type of object. If you’re working with point cloud data, use the add-on’s relevant import option.

• Overlooking coordinate and scale issues: Georeferenced LAS/LAZ data may have very large coordinate values. Center on Origin is enabled by default in the add-on; it helps keep the data near the origin and adds the offset back during export. Take this behavior into account in your workflow.

• Ignoring export warnings: A channel missing from the target format or an unselected source attribute can lead to data loss. Before delivering the file, review the warning list and the export report.

• Expecting the same point radius for every scan: Although the add-on sets the radius automatically, scan size and density can vary. If the points aren’t visible, check the radius in the Point Cloud panel.

• Assuming normals will be preserved when exporting E57: E57 export does not write normals. Imported normals may be present as attributes in Blender, but they aren’t written back to an E57 file.

Use in architecture, interiors and archviz workflows

A point cloud can serve as reference data for architecture and visualization teams bringing a scan of an existing building or site into a Blender scene. Working with a native PointCloud object lets you inspect the scan in the scene without converting it directly into a mesh. The ability to use fields such as color, intensity or classification as attributes also makes it possible to work with this data during organization and visualization.

On the other hand, the source provides no specific hardware requirements or performance measurements, so it’s not possible to say that large datasets will run at the same speed on every workstation. Before starting a project, test with the actual file on the computer you’ll use, check the visibility settings and confirm that the target application supports the export format. License terms should also be checked separately for the add-on and distribution method you plan to use; the supplied text from the Extensions page does not specify the add-on’s license.

Next steps

First, test the import and export cycle with a small sample file. Then determine the format you’ll use in the actual project and which attributes need to be transferred. If you have different sources such as E57, LAS/LAZ or PLY, check the data in the same scene and review the export warnings separately for each format. Deciding upfront whether the target for PLY is a point cloud or a mesh reduces the risk of choosing the wrong tool.

Source and license

The Point Cloud I/O add-on’s features and usage steps are based on its description on the Blender Extensions page; the supplied source text does not specify the add-on’s license. Blender’s Stanford PLY guide is licensed under CC BY-SA 4.0, and the information about PLY import/export options has been adapted into Turkish based on that document.