Orienting Point Clouds with Dipole Propagation

TL;DR

This paper presents a novel method for establishing globally consistent normal orientations in point clouds by combining local neural network predictions with a dipole-based global propagation technique.

Contribution

It introduces a two-phase approach that separates local normal estimation from global orientation propagation using dipole fields, improving robustness and consistency.

Findings

Achieves stable and robust normal orientation across complex point clouds.

Handles noise, holes, and sharp features effectively.

Outperforms existing methods in consistency and accuracy.

Abstract

Establishing a consistent normal orientation for point clouds is a notoriously difficult problem in geometry processing, requiring attention to both local and global shape characteristics. The normal direction of a point is a function of the local surface neighborhood; yet, point clouds do not disclose the full underlying surface structure. Even assuming known geodesic proximity, calculating a consistent normal orientation requires the global context. In this work, we introduce a novel approach for establishing a globally consistent normal orientation for point clouds. Our solution separates the local and global components into two different sub-problems. In the local phase, we train a neural network to learn a coherent normal direction per patch (i.e., consistently oriented normals within a single patch). In the global phase, we propagate the orientation across all coherent patches using a dipole propagation. Our dipole propagation decides to orient each patch using the electric field defined by all previously orientated patches. This gives rise to a global propagation that is stable, as well as being robust to nearby surfaces, holes, sharp features and noise.