Performing Image-like Convolution on Triangular Meshes

Simple item page
dc.contributor.authorTortorici, Claudioen_US
dc.contributor.authorWerghi, Naoufelen_US
dc.contributor.authorBerretti, Stefanoen_US
dc.contributor.editorTelea, Alex and Theoharis, Theoharis and Veltkamp, Remcoen_US
dc.date.accessioned2018-04-14T18:28:44Z
dc.date.available2018-04-14T18:28:44Z
dc.date.issued2018
dc.description.abstractImage convolution with a filtering mask is at the base of several image analysis operations. This is motivated by Mathematical foundations and by the straightforward way the discrete convolution can be computed on a grid-like domain. Extending the convolution operation to the mesh manifold support is a challenging task due to the irregular structure of the mesh connections. In this paper, we propose a computational framework that allows convolutional operations on the mesh. This relies on the idea of ordering the facets of the mesh so that a shift-like operation can be derived. Experiments have been performed with several filter masks (Sobel, Gabor, etc.) showing state-of-the-art results in 3D relief patterns retrieval on the SHREC'17 dataset. We also provide evidence that the proposed framework can enable convolution and pooling-like operations as can be needed for extending Convolutional Neural Networks to 3D meshes.en_US
dc.description.sectionheadersPosters
dc.description.seriesinformationEurographics Workshop on 3D Object Retrieval
dc.identifier.doi10.2312/3dor.20181060
dc.identifier.isbn978-3-03868-053-6
dc.identifier.issn1997-0471
dc.identifier.pages111-114
dc.identifier.urihttps://doi.org/10.2312/3dor.20181060
dc.identifier.urihttps://diglib.eg.org:443/handle/10.2312/3dor20181060
dc.publisherThe Eurographics Associationen_US
dc.subjectComputing methodologies
dc.subjectBiometrics
dc.subject3D imaging
dc.subjectComputer vision representations
dc.subjectNeural networks
dc.subjectMesh geometry models
dc.subjectShape analysis
dc.titlePerforming Image-like Convolution on Triangular Meshesen_US
Files
Original bundle
Now showing 1 - 1 of 1
Thumbnail Image
Name:
111-114.pdf
Size:
1.34 MB
Format:
Adobe Portable Document Format
Download
Collections
3DOR 18