Journal of System Simulation ›› 2019, Vol. 31 ›› Issue (5): 853-860.doi: 10.16182/j.issn1004731x.joss.17-0169

Previous Articles     Next Articles

Adaptively Resampling 3D Mesh Models Based on Editable Features

Dai Jiajia1, Fan Lipeng1,2, Pang Mingyong1   

  1. 1. Institute of EduInfo Science & Engineering, Nanjing Normal University, Nanjing 210097, China;
    2. School of Economics and Management, Nanjing University of Science & Technology, Nanjing 210094, China;
  • Received:2017-04-20 Revised:2017-07-04 Online:2019-05-08 Published:2019-11-20

Abstract: We propose a hybrid algorithm for adaptively resampling 3D triangulations by user-defined editable features. The method parameterizes a 3D mesh model into 2D parameter plane, and the geometric properties of the original model is calculated and represented on a planar domain. According to a constructed geometric image of the original model and user-defined editing information, the method creates a global density function for the resampled model. The sampling density function is employed to control distribution of samples in the 2D parameter domain. The method uses centroidal Voronoi tessellation technique to further optimize local distribution of the sampled points. The created samples in 2D domain are mapped to 3D space and the resulted model is obtained with adaptive sampling property. Experiments show that the algorithm can deal with various mesh models efficiently and robustly. The distribution of vertices of resulted model is adaptive and can be controlled by user-defined features.

Key words: adaptive sampling, Ricci flow, model editing, mesh parameterization

CLC Number: