Wavelet-based polygon soup consolidation. (February 2018)

• Record Type: Journal Article
• Authors: Hu, Ling; Li, Qinsong; Liu, Shengjun; Liu, Xinru; Wang, Zheng
• Is Part Of: Computers & graphics. Volume 70(2018)
• Page Start: 39
• Page End: 50
• Publication Date: 2018-02
• Subjects: D4 wavelets -- Polygon soup -- Consolidation -- Indicator function -- Geometry processing -- Implicit surface; Computer graphics -- Periodicals; 006.6
• DOI: 10.1016/j.cag.2017.07.028
• Languages: English
• ISSNs: 0097-8493

Abstract

Highlights: A robust consolidation method for the defective mesh even polygon soup is presented. The indicator function of a model is represented with D4 Daubechies wavelets. The wavelet coefficients are computed by surface integrals over triangles. The wavelet representation can be used in various geometric operations in 3D. Graphical abstract: Abstract: In this paper, a robust consolidation method based on wavelets is presented for the defective polygonal mesh even polygon soup. The indicator function of a model is represented with wavelets, where the wavelet coefficients are computed by surface integrals over polygons. The hierarchical wavelet representation from defective inputs guarantees a consolidated surface extracted from it where orthogonal and compactly supported D4 Daubechies wavelets are employed to balance the efficiency and surface quality. We investigate the usage of the wavelet representation in various geometric operations in 3D, such as data compression, repairing, Boolean operation, denoising, etc. Experiments demonstrate that the method is effective in polygon soup consolidation.