Myria3D

Segments airborne lidar into ground, vegetation, building, water, bridge and permanent structure.

Field
Remote sensing
Method
Deep learning
Data it takes
point-cloud
License
BSD-3-Clause

Myria3D is IGN’s RandLA-Net segmentation model for airborne lidar, distributed with the FRACTAL Lidar HD checkpoint. It assigns each point one of seven classes.

The checkpoint was trained on French national lidar. How well it carries to another survey depends on flying height, point density and land cover, and a survey unlike Lidar HD in those respects deserves a check against hand-labelled ground truth before the output is trusted.

The checkpoint is staged at run time.

Upstream project →

Catalog entry last checked 2026-09-08. All models →

Run Myria3D on your data

The lab runs this on its own compute, in a container, with the inputs and parameters recorded alongside the result. Initial scoping conversations are free.

Contact the Lab