ForestMamba replaces Transformer attention with linear Mamba modeling and seeds queries from a multi-scale canopy height model to achieve higher accuracy and lower compute on forest point cloud segmentation across seven regions.
Title resolution pending
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.CV 1years
2026 1verdicts
UNVERDICTED 1representative citing papers
citing papers explorer
-
ForestMamba: Sparse Mamba with Geometry-guided Queries for 3D Forest Point Cloud Segmentation
ForestMamba replaces Transformer attention with linear Mamba modeling and seeds queries from a multi-scale canopy height model to achieve higher accuracy and lower compute on forest point cloud segmentation across seven regions.