PIE-NET: Parametric Inference of Point Cloud Edges

TL;DR

PIE-NET is a novel deep learning approach that accurately detects and infers parametric 3D edges in point clouds, outperforming traditional methods and recent deep learning models in both accuracy and robustness.

Contribution

We propose PIE-NET, an end-to-end trainable neural network that infers parametric edge representations in 3D point clouds, introducing a region proposal architecture for improved edge detection.

Findings

Significantly outperforms traditional edge detection pipelines.

Achieves higher accuracy than recent deep learning edge detectors.

Demonstrates robustness across a large CAD dataset.

Abstract

We introduce an end-to-end learnable technique to robustly identify feature edges in 3D point cloud data. We represent these edges as a collection of parametric curves (i.e.,lines, circles, and B-splines). Accordingly, our deep neural network, coined PIE-NET, is trained for parametric inference of edges. The network relies on a "region proposal" architecture, where a first module proposes an over-complete collection of edge and corner points, and a second module ranks each proposal to decide whether it should be considered. We train and evaluate our method on the ABC dataset, a large dataset of CAD models, and compare our results to those produced by traditional (non-learning) processing pipelines, as well as a recent deep learning based edge detector (EC-NET). Our results significantly improve over the state-of-the-art from both a quantitative and qualitative standpoint.