Rethinking the compositionality of point clouds through regularization in the hyperbolic spaceDownload PDF

Published: 31 Oct 2022, Last Modified: 03 Jul 2024NeurIPS 2022 AcceptReaders: Everyone
Keywords: Hyperbolic space, Point cloud classification
TL;DR: We exploit the hyperbolic space to regularize point cloud classification through a part-whole hierarchy.
Abstract: Point clouds of 3D objects exhibit an inherent compositional nature where simple parts can be assembled into progressively more complex shapes to form whole objects. Explicitly capturing such part-whole hierarchy is a long-sought objective in order to build effective models, but its tree-like nature has made the task elusive. In this paper, we propose to embed the features of a point cloud classifier into the hyperbolic space and explicitly regularize the space to account for the part-whole hierarchy. The hyperbolic space is the only space that can successfully embed the tree-like nature of the hierarchy. This leads to substantial improvements in the performance of state-of-art supervised models for point cloud classification.
Supplementary Material: pdf
Community Implementations: [![CatalyzeX](/images/catalyzex_icon.svg) 2 code implementations](https://www.catalyzex.com/paper/rethinking-the-compositionality-of-point/code)
21 Replies

Loading