P2P-NET: Bidirectional Point Displacement Net for Shape Transform

TL;DR

P2P-NET is a versatile deep neural network that learns bidirectional geometric transformations between point-based shape representations without requiring point-to-point correspondences, enabling various shape transformation tasks.

Contribution

It introduces a novel bidirectional point displacement network that learns shape transformations from data without relying on explicit point correspondences.

Findings

Effective in transforming shapes across domains

Versatile for multiple shape transformation tasks

Learns without explicit point-to-point matching

Abstract

We introduce P2P-NET, a general-purpose deep neural network which learns geometric transformations between point-based shape representations from two domains, e.g., meso-skeletons and surfaces, partial and complete scans, etc. The architecture of the P2P-NET is that of a bi-directional point displacement network, which transforms a source point set to a target point set with the same cardinality, and vice versa, by applying point-wise displacement vectors learned from data. P2P-NET is trained on paired shapes from the source and target domains, but without relying on point-to-point correspondences between the source and target point sets. The training loss combines two uni-directional geometric losses, each enforcing a shape-wise similarity between the predicted and the target point sets, and a cross-regularization term to encourage consistency between displacement vectors going in opposite directions. We develop and present several different applications enabled by our general-purpose bidirectional P2P-NET to highlight the effectiveness, versatility, and potential of our network in solving a variety of point-based shape transformation problems.