3D Forest Semantic Segmentation Using Multispectral LiDAR and 3D Deep Learning

Narges Takhtkeshha, Lauris Bocaux, Lassi Ruoppa, Fabio Remondino, Gottfried Mandlburger, Antero Kukko, Juha Hyyppä · PFG – Journal of Photogrammetry, Remote Sensing and Geoinformation Science · 2025

DOI

Status

Reading

Topics

forest-segmentation, datasets

Feeds

First paper · height sampling

Used in

Notebook 01: PointNet and PointNet++ baselines on ForestSemantic-MS, Notebook 02: sampling retention gate

Dataset

ForestSemantic-MS

What they did

Releases ForestSemantic-MS, multispectral helicopter LiDAR of six boreal plots with six semantic classes, and benchmarks 3D deep networks; Point Transformer V3 with multispectral features scores best at 69.1% mIoU.

My note

No note yet.

BibTeX

@article{takhtkeshha2025-forestsemantic-ms,
  title = {{3D Forest Semantic Segmentation Using Multispectral LiDAR and 3D Deep Learning}},
  author = {Narges Takhtkeshha and Lauris Bocaux and Lassi Ruoppa and Fabio Remondino and Gottfried Mandlburger and Antero Kukko and Juha Hyyppä},
  journal = {PFG – Journal of Photogrammetry, Remote Sensing and Geoinformation Science},
  year = {2025},
  doi = {10.1007/s41064-025-00369-4}
}