Skip to content
Search

CloudCompare

lastmod 2026-08-09created 2026-04-028 edits

CloudCompare is a 3D point cloud (and triangular mesh) processing software. It was originally designed to perform comparison between two 3D points clouds (such as the ones obtained with a laser scanner) or between a point cloud and a triangular mesh. It relies on an octree structure that is highly optimized for this particular use-case. It was also meant to deal with huge point clouds (typically more than 10 million points, and up to 120 million with 2 GB of memory).

Install 

sudo apt install cloudcompare

Usage 

CloudCompare pcd_images/aubo_i10/aubo_cloud_out.pcd

Labeling methods

Manual segmentation (Segmentation)

The most basic labeling method: select a specific object directly, separate it, and attach a label.

  • Use the Segment tool: click the 'Segment' icon (scissors) in the toolbar.
  • Draw a polygon: left-click to draw a polygon around the area to label. Right-click to finish.
  • Separate: separate the points inside or outside the polygon by pressing the 'Segment in' or 'Segment out' button.
  • Save: save the separated part as a separate cloud and label it (e.g. 'car', 'tree').

Connected Components segmentation

A method that automatically groups and labels spatially connected points.

  • Run the tool: run 'Label Connected Components' (blue + white square icon) from the toolbar.
  • Settings: set 'Min distance' (minimum distance) to separate objects that are apart.
  • Result: each set of connected points automatically gets a unique ID label.

LiDAR data classification (Classification)

Mainly used to classify the ground (Ground) and other objects in LiDAR datasets.

  • CSF (Cloth Simulation Filter): use Plugins -> CSF Filter to automatically classify (Ground/Non-ground) and label the ground.
  • Use attribute fields: classified points are stored in the 'Classification' attribute field and can be used as a deep learning training dataset.

Semi-automatic labeling and AI system integration

CloudCompare provides workflows for efficiently labeling 3D point clouds using unsupervised learning algorithms or plugins.

  • Automatic classification plugins: some recent versions or plugins use clustering algorithms to automatically classify points with similar characteristics.

Alignment and labeling using point pairs (Align Two Point Clouds)

Used to match the positions of two point clouds, or to label matching parts.

  • 'Align' tool: select 4 or more equivalent points to align one cloud to the other.

Tips

  • Visualization: use Colors -> Scalar Fields to color by elevation, intensity, or classified Classification values, which makes labeling easier.
  • Save format: it is best to save labeled data in .las or .ply format so that it can include the label information.

Plugins

3DMASC

See also

Favorite site

CloudCompare wiki

author Zer0 sitemap