APML is a differentiable Sinkhorn-based loss with an automatic temperature schedule that improves structure preservation in point cloud reconstruction, lowering EMD by 15-81% versus Chamfer-style losses.
In: International conference on artificial intelligence and statistics
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
citation-role summary
background 1
citation-polarity summary
fields
cs.CV 1years
2025 1verdicts
CONDITIONAL 1roles
background 1polarities
unclear 1representative citing papers
citing papers explorer
-
APML: Adaptive Probabilistic Matching Loss for Robust 3D Point Cloud Reconstruction
APML is a differentiable Sinkhorn-based loss with an automatic temperature schedule that improves structure preservation in point cloud reconstruction, lowering EMD by 15-81% versus Chamfer-style losses.