Method · The generator as the distilled artifact
Point Cloud DD
Point Cloud Dataset Distillation
Deyu Bo, Xinchao Wang
ICML 2025 · first public 2025-01
In one paragraph
This paper adapts dataset distillation to unstructured 3D point clouds by theoretically showing that matching rotation-invariant features between real and synthetic data matters for 3D distillation, introducing a plug-and-play point cloud rotator that aligns point clouds to a canonical orientation, and replacing fixed-size synthetic data with a point-wise generator that produces point clouds at multiple resolutions instead of optimizing one fixed set directly.
Where it sits
- The generator as the distilled artifact (Generative priors)
- Setting: Other data types
BibTeX (generated; prefer the venue's official entry)
@article{bo2025point,
title = {Point Cloud Dataset Distillation},
author = {Deyu Bo and Xinchao Wang},
journal = {ICML 2025},
year = {2025}
}Nearby in The generator as the distilled artifact
2023-03