Show simple item record

dc.contributor.authorWang, Yiqun
dc.contributor.authorRen, Jing
dc.contributor.authorYan, Dong Ming
dc.contributor.authorGuo, Jianwei
dc.contributor.authorZhang, Xiaopeng
dc.contributor.authorWonka, Peter
dc.date.accessioned2020-09-15T13:34:48Z
dc.date.available2020-02-25T11:20:27Z
dc.date.available2020-09-15T13:34:48Z
dc.date.issued2020-08-12
dc.identifier.citationWang, Y., Ren, J., Yan, D.-M., Guo, J., Zhang, X., & Wonka, P. (2020). MGCN. ACM Transactions on Graphics, 39(4). doi:10.1145/3386569.3392443
dc.identifier.issn1557-7368
dc.identifier.issn0730-0301
dc.identifier.doi10.1145/3386569.3392443
dc.identifier.urihttp://hdl.handle.net/10754/661684
dc.description.abstractWe 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 overfit 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.
dc.publisherAssociation for Computing Machinery (ACM)
dc.relation.urlhttps://dl.acm.org/doi/10.1145/3386569.3392443
dc.rights© ACM, 2020. This is the author's version of the work. It is posted here by permission of ACM for your personal use. Not for redistribution. The definitive version was published in ACM Transactions on Graphics, {39, 4, (2020-08-12)} http://doi.acm.org/10.1145/3386569.3392443
dc.titleMGCN Descriptor Learning using Multiscale GCNs
dc.typeArticle
dc.contributor.departmentComputer Science Program
dc.contributor.departmentComputer, Electrical and Mathematical Sciences and Engineering (CEMSE) Division
dc.contributor.departmentVisual Computing Center (VCC)
dc.identifier.journalACM Transactions on Graphics
dc.eprint.versionPost-print
dc.contributor.institutionUniversity of Chinese Academy of Sciences
dc.identifier.volume39
dc.identifier.issue4
dc.identifier.arxivid2001.10472
kaust.personWang, Yiqun
kaust.personWonka, Peter
dc.identifier.eid2-s2.0-85090409742
refterms.dateFOA2020-02-25T11:21:21Z
dc.date.published-online2020-08-12
dc.date.published-print2020-07-08
dc.date.posted2020-01-28


Files in this item

Thumbnail
Name:
Preprintfile1.pdf
Size:
8.249Mb
Format:
PDF
Description:
Pre-print

This item appears in the following Collection(s)

Show simple item record

VersionItemEditorDateSummary

*Selected version