NeAF: Learning Neural Angle Fields for Point Normal Estimation

TL;DR

NeAF introduces an implicit angle field approach for point normal estimation that improves robustness and accuracy by predicting angle offsets rather than direct normals, outperforming existing methods on synthetic and real data.

Contribution

The paper proposes Neural Angle Fields (NeAF), a novel implicit function that predicts angle offsets in spherical coordinates to enhance normal estimation accuracy and robustness.

Findings

Significant improvement over state-of-the-art methods.

Effective on both synthetic data and real scans.

Robust to unseen scenarios and parameter settings.

Abstract

Normal estimation for unstructured point clouds is an important task in 3D computer vision. Current methods achieve encouraging results by mapping local patches to normal vectors or learning local surface fitting using neural networks. However, these methods are not generalized well to unseen scenarios and are sensitive to parameter settings. To resolve these issues, we propose an implicit function to learn an angle field around the normal of each point in the spherical coordinate system, which is dubbed as Neural Angle Fields (NeAF). Instead of directly predicting the normal of an input point, we predict the angle offset between the ground truth normal and a randomly sampled query normal. This strategy pushes the network to observe more diverse samples, which leads to higher prediction accuracy in a more robust manner. To predict normals from the learned angle fields at inference time, we randomly sample query vectors in a unit spherical space and take the vectors with minimal angle values as the predicted normals. To further leverage the prior learned by NeAF, we propose to refine the predicted normal vectors by minimizing the angle offsets. The experimental results with synthetic data and real scans show significant improvements over the state-of-the-art under widely used benchmarks.