ch-sa/labelCloud

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

65
/ 100
Established

This tool helps engineers and researchers accurately label 3D objects within point cloud data. You input raw point cloud scans (like from LiDAR or depth sensors), then interactively draw and adjust 3D bounding boxes around objects of interest. The output is a structured file containing the precise location, dimensions, and orientation of each labeled object, suitable for training machine learning models or spatial analysis.

770 stars. Available on PyPI.

Use this if you need to manually outline objects in 3D point clouds with high precision for tasks like autonomous driving datasets, robotics, or augmented reality.

Not ideal if you're dealing with standard 2D images or videos, or if your labeling task primarily involves semantic segmentation of individual points rather than entire objects.

3D-data-labeling autonomous-vehicles robotics spatial-analysis LiDAR-data
Maintenance 6 / 25
Adoption 10 / 25
Maturity 25 / 25
Community 24 / 25

How are scores calculated?

Stars

770

Forks

134

Language

Python

License

GPL-3.0

Last pushed

Nov 09, 2025

Commits (30d)

0

Dependencies

5

Get this data via API

curl "https://pt-edge.onrender.com/api/v1/quality/computer-vision/ch-sa/labelCloud"

Open to everyone — 100 requests/day, no key needed. Get a free key for 1,000/day.