期刊
COMPUTER AIDED GEOMETRIC DESIGN
卷 43, 期 -, 页码 39-52出版社
ELSEVIER
DOI: 10.1016/j.cagd.2016.02.015
关键词
3D shapes; Segmentation; Co-segmentation; Deep learning; High-level features
资金
- National Natural Science Foundation of China [11226328, 61222206, 11526212, 61300168, 61273332]
- Natural Science Foundation of Zhejiang Province [LY13F020018]
- Opening Foundation of Zhejiang Provincial Top Key Discipline [XKXL1406]
- Natural Science Foundation of Ningbo City Grant [2015A610123]
- One Hundred Talent Project of the Chinese Academy of Sciences
- Ningbo Sc. & Tech. (Innovation Team) Plan Project [2014B82015]
In this paper, we propose a novel unsupervised algorithm for automatically segmenting a single 3D shape or co-segmenting a family of 3D shapes using deep learning. The algorithm consists of three stages. In the first stage, we pre-decompose each 3D shape of interest into primitive patches to generate over-segmentation and compute various signatures as low-level shape features. In the second stage, high-level features are learned, in an unsupervised style, from the low-level ones based on deep learning. Finally, either segmentation or co-segmentation results can be quickly reported by patch clustering in the high-level feature space. The experimental results on the Princeton Segmentation Benchmark and the Shape COSEG Dataset exhibit superior segmentation performance of the proposed method over the previous state-of-the-art approaches. (C) 2016 Elsevier B.V. All rights reserved.
作者
我是这篇论文的作者
点击您的名字以认领此论文并将其添加到您的个人资料中。
推荐
暂无数据