A bilevel flow matching model that first samples a shape code and then generates points conditioned on it matches or improves on prior point cloud generators with far fewer sampling steps.
Learning representations and generative models for 3d point clouds
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
2026 1verdicts
CONDITIONAL 1roles
background 1polarities
unclear 1representative citing papers
citing papers explorer
-
Hierarchical Flow Matching for 3D Point Cloud Generation
A bilevel flow matching model that first samples a shape code and then generates points conditioned on it matches or improves on prior point cloud generators with far fewer sampling steps.