#18 · Primary category: Data Annotation & Labeling Tools

labelCloud

3d-object-detection 6d-pose-estimation annotation bounding-boxes computer-vision labeling machine-learning point-clouds tool

A lightweight tool for labeling 3D bounding boxes in point clouds.

Project last updated:05/02/26

GitHub Stars

803

Forks

134

Contributors

9

License

GPL-3.0

Why we included this project

Training a 3D object detector or a 6-DoF pose model means you eventually have to draw ground-truth boxes on raw point clouds, and labelCloud handles that step without much ceremony. It is a small Python app you install with pip, point at a folder of point clouds, and start labeling. Two workflows cover the actual annotation: pick a front-top edge and scroll to set rotation, or span length, width, and height across four vertices. Afterward you correct translation, dimension, and rotation with mouse and keyboard shortcuts. It ships with an example point cloud and a config file for tuning, so a small team can run it locally and skip standing up a heavier annotation platform.

Articles for this project

No articles for this project yet.

To suggest a topic or contribute an article, contact us.

Related projects in this category