Introduces a two-stream attention fusion model and joint loss for multispectral point cloud classification, releasing two new airborne datasets and reporting gains over prior methods.
Unsupervised occluded target detection based on spherical shell with multispectral point clouds,
2 Pith papers cite this work. Polarity classification is still indexing.
2
Pith papers citing it
fields
cs.CV 2years
2026 2verdicts
UNVERDICTED 2representative citing papers
Proposes a prior-free anomaly detection framework for sub-canopy UAV multispectral point clouds that estimates solar angle via inverse optimization and uses illumination-consistent background dictionaries to separate targets from shadows.
citing papers explorer
-
An Enhanced Geometric-Spectral Feature Learning Framework for Airborne Multispectral Point Cloud Classification
Introduces a two-stream attention fusion model and joint loss for multispectral point cloud classification, releasing two new airborne datasets and reporting gains over prior methods.
-
Illumination-Invariant Anomaly Detection for Sub-Canopy UAV Multispectral Point Clouds
Proposes a prior-free anomaly detection framework for sub-canopy UAV multispectral point clouds that estimates solar angle via inverse optimization and uses illumination-consistent background dictionaries to separate targets from shadows.