Show simple item record

dc.contributor.authorLudwig, Ingmaren_US
dc.contributor.authorTyson, Danielen_US
dc.contributor.authorCampen, Marcelen_US
dc.contributor.editorMemari, Pooranen_US
dc.contributor.editorSolomon, Justinen_US
dc.date.accessioned2023-06-30T06:18:19Z
dc.date.available2023-06-30T06:18:19Z
dc.date.issued2023
dc.identifier.issn1467-8659
dc.identifier.urihttps://doi.org/10.1111/cgf.14898
dc.identifier.urihttps://diglib.eg.org:443/handle/10.1111/cgf14898
dc.description.abstractWe describe HalfedgeCNN, a collection of modules to build neural networks that operate on triangle meshes. Taking inspiration from the (edge-based) MeshCNN, convolution, pooling, and unpooling layers are consistently defined on the basis of halfedges of the mesh, pairs of oppositely oriented virtual instances of each edge. This provides benefits over alternative definitions on the basis of vertices, edges, or faces. Additional interface layers enable support for feature data associated with such mesh entities in input and output as well. Due to being defined natively on mesh entities and their neighborhoods, lossy resampling or interpolation techniques (to enable the application of operators adopted from image domains) do not need to be employed. The operators have various degrees of freedom that can be exploited to adapt to application-specific needs.en_US
dc.publisherThe Eurographics Association and John Wiley & Sons Ltd.en_US
dc.subjectCCS Concepts: Computing methodologies -> Shape analysis; Mesh models; Neural networks
dc.subjectComputing methodologies
dc.subjectShape analysis
dc.subjectMesh models
dc.subjectNeural networks
dc.titleHalfedgeCNN for Native and Flexible Deep Learning on Triangle Meshesen_US
dc.description.seriesinformationComputer Graphics Forum
dc.description.sectionheadersMeshing
dc.description.volume42
dc.description.number5
dc.identifier.doi10.1111/cgf.14898
dc.identifier.pages10 pages


Files in this item

Thumbnail

This item appears in the following Collection(s)

  • 42-Issue 5
    Geometry Processing 2023 - Symposium Proceedings

Show simple item record