MGCN: Descriptor Learning using Multiscale GCNs | |
Wang, Yiqun1,2,3; Ren, Jing1; Yan, Dong-Ming2,3; Guo, Jianwei2,3; Zhang, Xiaopeng2,3; Wonka, Peter1 | |
刊名 | ACM TRANSACTIONS ON GRAPHICS |
2020-07-01 | |
卷号 | 39期号:4页码:15 |
关键词 | Multiscale Energy Decomposition Wavelet Convolution Shape Matching |
ISSN号 | 0730-0301 |
DOI | 10.1145/3386569.3392443 |
英文摘要 | We propose a novel framework for computing descriptors for characterizing points on three-dimensional surfaces. First, we present a new non-learned feature that uses graph wavelets to decompose the Dirichlet energy on a surface. We call this new feature Wavelet Energy Decomposition Signature (WEDS). Second, we propose a new Multiscale Graph Convolutional Network (MGCN) to transform a non-learned feature to a more discriminative descriptor. Our results show that the new descriptor WEDS is more discriminative than the current state-of-the-art non-learned descriptors and that the combination of WEDS and MGCN is better than the state-of-the-art learned descriptors. An important design criterion for our descriptor is the robustness to different surface discretizations including triangulations with varying numbers of vertices. Our results demonstrate that previous graph convolutional networks significantly overlit to a particular resolution or even a particular triangulation, but MGCN generalizes well to different surface discretizations. In addition, MGCN is compatible with previous descriptors and it can also be used to improve the performance of other descriptors, such as the heat kernel signature, the wave kernel signature, or the local point signature. |
资助项目 | National Key R&D Program of China[2018YFB2100602] ; National Key R&D Program of China[2019YFB2204104] ; National Natural Science Foundation of China[61620106003] ; National Natural Science Foundation of China[61772523] ; National Natural Science Foundation of China[61802406] ; National Natural Science Foundation of China[61972459] ; Beijing Natural Science Foundation[L182059] ; CCF-Tencent Open Research Fund ; Shenzhen Basic Research Program[JCYJ20180507182222355] ; Alibaba Group through Alibaba Innovative Research Program ; KAUST OSR Award[CRG-2017-3426] |
WOS研究方向 | Computer Science |
语种 | 英语 |
出版者 | ASSOC COMPUTING MACHINERY |
WOS记录号 | WOS:000583700300095 |
资助机构 | National Key R&D Program of China ; National Natural Science Foundation of China ; Beijing Natural Science Foundation ; CCF-Tencent Open Research Fund ; Shenzhen Basic Research Program ; Alibaba Group through Alibaba Innovative Research Program ; KAUST OSR Award |
内容类型 | 期刊论文 |
源URL | [http://ir.ia.ac.cn/handle/173211/41655] |
专题 | 模式识别国家重点实验室_三维可视计算 |
通讯作者 | Wang, Yiqun |
作者单位 | 1.KAUST, Thuwal, Saudi Arabia 2.Chinese Acad Sci, Inst Automat, Beijing, Peoples R China 3.Univ Chinese Acad Sci, Sch AI, Beijing, Peoples R China |
推荐引用方式 GB/T 7714 | Wang, Yiqun,Ren, Jing,Yan, Dong-Ming,et al. MGCN: Descriptor Learning using Multiscale GCNs[J]. ACM TRANSACTIONS ON GRAPHICS,2020,39(4):15. |
APA | Wang, Yiqun,Ren, Jing,Yan, Dong-Ming,Guo, Jianwei,Zhang, Xiaopeng,&Wonka, Peter.(2020).MGCN: Descriptor Learning using Multiscale GCNs.ACM TRANSACTIONS ON GRAPHICS,39(4),15. |
MLA | Wang, Yiqun,et al."MGCN: Descriptor Learning using Multiscale GCNs".ACM TRANSACTIONS ON GRAPHICS 39.4(2020):15. |
个性服务 |
查看访问统计 |
相关权益政策 |
暂无数据 |
收藏/分享 |
除非特别说明,本系统中所有内容都受版权保护,并保留所有权利。
修改评论