A category-level geometry embedding and consistency loss improve domain-generalized 3D point cloud semantic segmentation by about 1-2 mIoU on adverse-weather benchmarks.
Deep learning for 3d point clouds: A survey
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.CV 1years
2025 1verdicts
CONDITIONAL 1representative citing papers
citing papers explorer
-
Domain-aware Category-level Geometry Learning Segmentation for 3D Point Clouds
A category-level geometry embedding and consistency loss improve domain-generalized 3D point cloud semantic segmentation by about 1-2 mIoU on adverse-weather benchmarks.