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.
@article{arxiv.1810.08950,
title = {Learning Spectral Transform Network on 3D Surface for Non-rigid Shape Analysis},
author = {Ruixuan Yu and Jian Sun and Huibin Li},
journal= {arXiv preprint arXiv:1810.08950},
year = {2018}
}