nsavinov/semantic3dnet

Point cloud semantic segmentation via Deep 3D Convolutional Neural Network

49
/ 100
Emerging

This project helps classify individual points within large 3D scans (point clouds) into meaningful categories like buildings, cars, or vegetation. It takes raw 3D point cloud data as input and outputs a classified point cloud, where each point is labeled with its semantic class. This tool is for researchers and practitioners working with detailed 3D spatial data who need to automatically understand and segment objects within their scans.

188 stars. No commits in the last 6 months.

Use this if you need to automatically categorize millions of points in a 3D scan into semantic classes, such as for urban planning, autonomous vehicle perception, or environmental mapping.

Not ideal if you don't have access to a powerful Nvidia GPU with at least 6GB of memory and 8GB of system RAM, or if you need to classify objects in 2D images instead of 3D point clouds.

3D-scanning urban-mapping geospatial-analysis LIDAR-processing object-segmentation
Stale 6m No Package No Dependents
Maintenance 0 / 25
Adoption 10 / 25
Maturity 16 / 25
Community 23 / 25

How are scores calculated?

Stars

188

Forks

77

Language

C++

License

BSD-3-Clause

Last pushed

Aug 30, 2022

Commits (30d)

0

Get this data via API

curl "https://pt-edge.onrender.com/api/v1/quality/computer-vision/nsavinov/semantic3dnet"

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