HPNet: Deep Primitive Segmentation Using Hybrid Representations

TL;DR

HPNet is a deep learning model that segments 3D point clouds into primitive parts by using hybrid feature representations and a learned combination strategy, achieving superior results on benchmark datasets.

Contribution

Introduces HPNet, a novel deep primitive segmentation method that combines multiple feature representations with learned weighting for improved accuracy.

Findings

Significant performance improvements over baseline methods.

Effective use of hybrid feature representations and learned combination weights.

Successful application on benchmark datasets ANSI and ABCParts.

Abstract

This paper introduces HPNet, a novel deep-learning approach for segmenting a 3D shape represented as a point cloud into primitive patches. The key to deep primitive segmentation is learning a feature representation that can separate points of different primitives. Unlike utilizing a single feature representation, HPNet leverages hybrid representations that combine one learned semantic descriptor, two spectral descriptors derived from predicted geometric parameters, as well as an adjacency matrix that encodes sharp edges. Moreover, instead of merely concatenating the descriptors, HPNet optimally combines hybrid representations by learning combination weights. This weighting module builds on the entropy of input features. The output primitive segmentation is obtained from a mean-shift clustering module. Experimental results on benchmark datasets ANSI and ABCParts show that HPNet leads to significant performance gains from baseline approaches.