DeepDT: Learning Geometry From Delaunay Triangulation for Surface
Reconstruction
Y. Luo, Z. Mi, and W. Tao. (2021)cite arxiv:2101.10353Comment: Accepted by AAAI 2021.
Abstract
In this paper, a novel learning-based network, named DeepDT, is proposed to
reconstruct the surface from Delaunay triangulation of point cloud. DeepDT
learns to predict inside/outside labels of Delaunay tetrahedrons directly from
a point cloud and corresponding Delaunay triangulation. The local geometry
features are first extracted from the input point cloud and aggregated into a
graph deriving from the Delaunay triangulation. Then a graph filtering is
applied on the aggregated features in order to add structural regularization to
the label prediction of tetrahedrons. Due to the complicated spatial relations
between tetrahedrons and the triangles, it is impossible to directly generate
ground truth labels of tetrahedrons from ground truth surface. Therefore, we
propose a multi-label supervision strategy which votes for the label of a
tetrahedron with labels of sampling locations inside it. The proposed DeepDT
can maintain abundant geometry details without generating overly complex
surfaces, especially for inner surfaces of open scenes. Meanwhile, the
generalization ability and time consumption of the proposed method is
acceptable and competitive compared with the state-of-the-art methods.
Experiments demonstrate the superior performance of the proposed DeepDT.
Description
[2101.10353] DeepDT: Learning Geometry From Delaunay Triangulation for Surface Reconstruction
%0 Generic
%1 luo2021deepdt
%A Luo, Yiming
%A Mi, Zhenxing
%A Tao, Wenbing
%D 2021
%K 2021 deep-learning geometry reconstruction triangulation
%T DeepDT: Learning Geometry From Delaunay Triangulation for Surface
Reconstruction
%U http://arxiv.org/abs/2101.10353
%X In this paper, a novel learning-based network, named DeepDT, is proposed to
reconstruct the surface from Delaunay triangulation of point cloud. DeepDT
learns to predict inside/outside labels of Delaunay tetrahedrons directly from
a point cloud and corresponding Delaunay triangulation. The local geometry
features are first extracted from the input point cloud and aggregated into a
graph deriving from the Delaunay triangulation. Then a graph filtering is
applied on the aggregated features in order to add structural regularization to
the label prediction of tetrahedrons. Due to the complicated spatial relations
between tetrahedrons and the triangles, it is impossible to directly generate
ground truth labels of tetrahedrons from ground truth surface. Therefore, we
propose a multi-label supervision strategy which votes for the label of a
tetrahedron with labels of sampling locations inside it. The proposed DeepDT
can maintain abundant geometry details without generating overly complex
surfaces, especially for inner surfaces of open scenes. Meanwhile, the
generalization ability and time consumption of the proposed method is
acceptable and competitive compared with the state-of-the-art methods.
Experiments demonstrate the superior performance of the proposed DeepDT.
@misc{luo2021deepdt,
abstract = {In this paper, a novel learning-based network, named DeepDT, is proposed to
reconstruct the surface from Delaunay triangulation of point cloud. DeepDT
learns to predict inside/outside labels of Delaunay tetrahedrons directly from
a point cloud and corresponding Delaunay triangulation. The local geometry
features are first extracted from the input point cloud and aggregated into a
graph deriving from the Delaunay triangulation. Then a graph filtering is
applied on the aggregated features in order to add structural regularization to
the label prediction of tetrahedrons. Due to the complicated spatial relations
between tetrahedrons and the triangles, it is impossible to directly generate
ground truth labels of tetrahedrons from ground truth surface. Therefore, we
propose a multi-label supervision strategy which votes for the label of a
tetrahedron with labels of sampling locations inside it. The proposed DeepDT
can maintain abundant geometry details without generating overly complex
surfaces, especially for inner surfaces of open scenes. Meanwhile, the
generalization ability and time consumption of the proposed method is
acceptable and competitive compared with the state-of-the-art methods.
Experiments demonstrate the superior performance of the proposed DeepDT.},
added-at = {2021-04-05T06:06:42.000+0200},
author = {Luo, Yiming and Mi, Zhenxing and Tao, Wenbing},
biburl = {https://www.bibsonomy.org/bibtex/264dbb6104ef4af1590bc713005498e33/analyst},
description = {[2101.10353] DeepDT: Learning Geometry From Delaunay Triangulation for Surface Reconstruction},
interhash = {b9616283f3d21af7c80486bf72c0c65e},
intrahash = {64dbb6104ef4af1590bc713005498e33},
keywords = {2021 deep-learning geometry reconstruction triangulation},
note = {cite arxiv:2101.10353Comment: Accepted by AAAI 2021},
timestamp = {2021-04-05T06:06:42.000+0200},
title = {DeepDT: Learning Geometry From Delaunay Triangulation for Surface
Reconstruction},
url = {http://arxiv.org/abs/2101.10353},
year = 2021
}