Communications in Information and Systems

Volume 17 (2017)

Number 2

Interactive mesh cutting with Laplace coordinates and gradient

Pages: 65 – 83

DOI: https://dx.doi.org/10.4310/CIS.2017.v17.n2.a1

Authors

Bin Liu (School of Mathematical Sciences, Dalian University of Technology, Dalian City, Liaoning, China)

Weiming Wang (School of Mathematical Sciences, Dalian University of Technology, Dalian City, Liaoning, China)

Junjie Cao (School of Mathematical Sciences, Dalian University of Technology, Dalian City, Liaoning, China)

Xiuping Liu (School of Mathematical Sciences, Dalian University of Technology, Dalian City, Liaoning, China)

Abstract

Sketch-based mesh cutting has been attended by researchers in Computer Graphics, mainly due to its success in dealing with semantic information of mesh with little user interaction. Although most existing approaches have gained great improvement, they depend on the predefined geometric features sensitive to shape and complex mathematical theory, and most methods can not generate a global unique solution for the segmentation problem. In this paper, we propose a novel sketch-based mesh segmentation model with Laplace Coordinates and gradient, which is independent of the complex geometric features and could perceive the differences of mesh parts. Furthermore, our algorithm is easy to implement, mathematically simple, and a global unique solution can be guaranteed because of our convex quadratic model. Benefiting from the Laplace Coordinates and gradient, our method holds an anisotropic behavior and can better fit the cutting boundary. Namely, triangular faces sharing similar attributes are kept closer to each other while big jumps naturally happen on the boundary between mesh parts when our method is applied. A large number of experiments illustrate the enhanced efficacy of our method compared with the state-of-the-art techniques.

Weiming Wang’s research supported in part by NSFC (61702079), 2016M601308 and DUT16RC(3)061.

Junjie Cao’s research supported in part by NSFC (61572099, 61772104.

Xiuping Liu’s research supported in part by NSFC (61370143).

Published 1 February 2018