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 Plus annual

Presents the access of premium content as premium feature

Premium Content

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 Plus monthly

Global: Zendy Tools annual

Global: Zendy Tools monthly

Global: Zendy Free Plan

In this paper, a novel learning-based network, named DeepDT, is proposed toreconstruct the surface from Delaunay triangulation of point cloud. DeepDTlearns to predict inside/outside labels of Delaunay tetrahedrons directly froma point cloud and corresponding Delaunay triangulation. The local geometryfeatures are first extracted from the input point cloud and aggregated into agraph deriving from the Delaunay triangulation. Then a graph filtering isapplied on the aggregated features in order to add structural regularization tothe label prediction of tetrahedrons. Due to the complicated spatial relationsbetween tetrahedrons and the triangles, it is impossible to directly generateground truth labels of tetrahedrons from ground truth surface. Therefore, wepropose a multi-label supervision strategy which votes for the label of atetrahedron with labels of sampling locations inside it. The proposed DeepDTcan maintain abundant geometry details without generating overly complexsurfaces, especially for inner surfaces of open scenes. Meanwhile, thegeneralization ability and time consumption of the proposed method isacceptable and competitive compared with the state-of-the-art methods.Experiments demonstrate the superior performance of the proposed DeepDT.

DeepDT: Learning Geometry From Delaunay Triangulation for Surface Reconstruction