SegmentAnyTree

Separates a forest lidar point cloud into individual trees

Field: Forestry & vegetation
Method: Deep learning
Data it takes: point-cloud
License: MIT

SegmentAnyTree labels every point in a forest lidar cloud with the tree it belongs to. Separating trees enables per-tree measurement of height, crown dimensions and stem position across a full survey rather than sample plots.

Input is LAS, LAZ or COPC. Output is a segmented cloud with per-tree attributes.

TreeLearn covers ground-based scans. 3DFin is the deterministic alternative where a non-learned baseline is required.

Upstream project →

Catalog entry last checked 2026-08-17. All models →

Run SegmentAnyTree 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