Learning Spectral Transform Network on 3D Surface for Non-rigid Shape Analysis

TL;DR

This paper introduces a novel spectral transform network designed for non-rigid 3D shape analysis, achieving state-of-the-art accuracy in shape retrieval and classification tasks.

Contribution

It proposes a simple, shallow spectral transform network architecture with four stages specifically for non-rigid shape analysis on 3D surfaces.

Findings

Achieved highest accuracy on SHREC14 and 15 datasets

Effective for non-rigid shape retrieval and classification

Network is simple and shallow, yet effective

Abstract

Designing a network on 3D surface for non-rigid shape analysis is a challenging task. In this work, we propose a novel spectral transform network on 3D surface to learn shape descriptors. The proposed network architecture consists of four stages: raw descriptor extraction, surface second-order pooling, mixture of power function-based spectral transform, and metric learning. The proposed network is simple and shallow. Quantitative experiments on challenging benchmarks show its effectiveness for non-rigid shape retrieval and classification, e.g., it achieved the highest accuracies on SHREC14, 15 datasets as well as the Range subset of SHREC17 dataset.