Extracting Cycle-aware Feature Curve Networks from 3D Models

Zhengda Lu,Jianwei Guo,Jun Xiao,Ying Wang,Xiaopeng Zhang,Dong-Ming Yan

DOI: https://doi.org/10.1016/j.cad.2020.102949

2021-02-01

Abstract:We provide an efficient method to filter out the noise curve from the initial curve network to preserve meaningful feature curves and extract complete FCNs from 3D models.We simultaneously find the curve cycles in the network and generate highquality patch layouts compared with the state-of-the-art segmentation approach.We obtain a compact high-level representation of 3D models that can be used for surface reconstruction.Meaningful feature curves provide high-level shape representation of the geometrical shapes and are useful in various applications. In this paper, we propose an automatic method on the basis of the quadric surface fitting technique to extract complete feature curve networks (FCNs) from 3D surface meshes, as well as finding cycles and generating a high-quality segmentation. In the initial collection of noisy and fragmented feature curves, we first fit the quadric surfaces of each curve and the corresponding neighbor vertices to filter out non-salient or noisy feature curves. Then we conduct a feature extension step to address the curve intersections and form a closed FCN. Finally, we regard circle curves as cycles in the complete FCN and segment the mesh into patches to reveal a highly structured representation of the input geometry. Experimental results demonstrate that our algorithm is more robust for FCN extraction from complex input meshes and achieves higher quality patch layouts compared with the state-of-the-art approaches. We also verify the validity of extracted feature curve cycles by applying them to surface reconstruction.

computer science, software engineering