English (United Kingdom)

https://curated-unify.zendy.io/wp-json/zendy-region/v1/featured_content/oa?rat=en

https://curated-unify.zendy.io/wp-json/zendy-region/v1/highlighted_journal/

Global: Zendy Tools annual

Presents the keyphrase highlighting as premium feature

Keyphrase Highlighting

Presents the summarisation as premium feature

Summarisation

Insights

Presents the pdf analysis as premium feature

PDF Analysis

Presents the zaia usage as premium feature

ZAIA

Global: Zendy Tools monthly

Global: Zendy Plus monthly

Presents the access of premium content as premium feature

Premium Content

Global: Zendy Plus annual

Global: Zendy Free Plan

We present Picasso, a CUDA-based library comprising novel modules for deeplearning over complex real-world 3D meshes. Hierarchical neural architectureshave proved effective in multi-scale feature extraction which signifies theneed for fast mesh decimation. However, existing methods rely on CPU-basedimplementations to obtain multi-resolution meshes. We design GPU-acceleratedmesh decimation to facilitate network resolution reduction efficientlyon-the-fly. Pooling and unpooling modules are defined on the vertex clustersgathered during decimation. For feature learning over meshes, Picasso containsthree types of novel convolutions namely, facet2vertex, vertex2facet, andfacet2facet convolution. Hence, it treats a mesh as a geometric structurecomprising vertices and facets, rather than a spatial graph with edges asprevious methods do. Picasso also incorporates a fuzzy mechanism in its filtersfor robustness to mesh sampling (vertex density). It exploits Gaussian mixturesto define fuzzy coefficients for the facet2vertex convolution, and barycentricinterpolation to define the coefficients for the remaining two convolutions. Inthis release, we demonstrate the effectiveness of the proposed modules withcompetitive segmentation results on S3DIS. The library will be made publicthrough https://github.com/hlei-ziyan/Picasso.

Picasso: A CUDA-based Library for Deep Learning over 3D Meshes