Point Cloud Cleanup

Requires the Max or Studio license

Point Cloud Cleanup removes noisy or unwanted points from imported Quick, Heavy, Dense, and mesh-related point cloud objects so the reconstruction becomes easier to inspect and use.

How to use it

  1. 1 Import a point cloud result.
  2. 2 Select the target point cloud in the cleanup panel.
  3. 3 Adjust neighbor count and aggressiveness.
  4. 4 Run cleanup.
  5. 5 Use Ctrl+Z if the result is too aggressive and rerun with lighter settings.

Point Cloud Cleanup Panel

Point Cloud Cleanup removes noisy, floating points from a tracking point cloud with Statistical Outlier Removal (SOR). For every vertex it looks at the nearest neighbors and removes points that sit far away from their local cluster, while keeping the overall structure of the scene.

Target Object

Pick the point cloud in the Target Object field. The panel shows its vertex count so you can see how much a cleanup removed.

Pick the points, not the parent. Choose the point cloud mesh itself, not the empty or parent object that holds it.

Cleanup parameters

  • •
    Neighbor Count (default 30): how many neighbors each point is compared with. Higher values give a smoother result but can remove fine detail; low values may leave some noise.
  • •
    Aggressiveness (default 0.1): lower values clean more aggressively and remove more points; higher values are gentler and may leave noise behind.

Running the cleanup

Click Cleanup Point Cloud. If the result is too aggressive, press Ctrl+Z and try again with gentler settings. Cleanup does not fix a bad solve; if the cloud is fundamentally wrong, improve the footage, masks or Heavy settings and track again.

Technical notes

Cleanup is useful before alignment, object placement, mesh inspection, and layout. It reduces clutter from floating points or unwanted reconstruction regions.

It does not fix a bad solve by itself. If the point cloud is fundamentally wrong, rerun tracking with better footage, masks, or Heavy settings.

Docs Dense Point Clouds

Dense Point Cloud generation creates a high-detail colorized representation of the tracked scene from Heavy Tracking results. It is useful when sparse points are not enough to understand the environment.

Docs Floor Alignment

The Aligner panel turns arbitrary reconstruction orientation into a usable Blender scene. It can snap selected floor points to the XY plane and flip inverted tracks.

Docs Import Results

Import Results separates processing from scene import. Scan one folder to choose among indexed Quick, Heavy, Tripod, Planar, Dense, or Mesh results and to scope the solved-lens source used by calibration and STMap export.

 Guide Point cloud generation for Blender

Generate useful sparse and dense point clouds from tracked footage

 Guide Blender camera tracking tutorial

A practical camera tracking workflow from footage to Blender scene