SAMBLE learns shape-specific point sampling strategies using a sparse attention map and per-bin sampling weights to balance local detail and global uniformity.
PRA-Net: Point relation-aware network for 3d point cloud analysis.IEEE Transactions on Image Processing, 30: 4436–4448, 2021
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
-
SAMBLE: Shape-Specific Point Cloud Sampling for an Optimal Trade-Off Between Local Detail and Global Uniformity
SAMBLE learns shape-specific point sampling strategies using a sparse attention map and per-bin sampling weights to balance local detail and global uniformity.