首页> 外文期刊>ACM Transactions on Graphics >Level-of-Detail Quad Meshing
【24h】

Level-of-Detail Quad Meshing

机译:详细程度四边形网格划分

获取原文
获取原文并翻译 | 示例
           

摘要

The most effective and popular tools for obtaining feature alignedrnquad meshes from triangular input meshes are based on cross fieldrnguided parametrization. These methods are incarnations of a conceptualrnthree-step pipeline: (1) cross field computation, (2) fieldguidedrnsurface parametrization, (3) quad mesh extraction. Whilernin most meshing scenarios the user prescribes a desired target quadrnsize or edge length, this information is typically taken into accountrnfrom step 2 onwards only, but not in the cross field computationrnstep. This turns into a problem in the presence of small scale geometricrnor topological features or noise in the input mesh: closelyrnplaced singularities are induced in the cross field, which are notrnproperly reproducible by vertices in a quad mesh with the prescribedrnedge length, causing severe distortions or even failure of thernmeshing algorithm. We reformulate the construction of cross fieldsrnas well as field-guided parametrizations in a scale-aware mannerrnwhich effectively suppresses densely spaced features and noise ofrngeometric as well as topological kind. Dominant large-scale featuresrnare adequately preserved in the output by relying on the unalteredrninput mesh as the computational domain.
机译:从三角形输入网格中获取特征对齐的四边形网格的最有效,最流行的工具是基于交叉场引导的参数化。这些方法是概念性的三步流水线的化身:(1)跨场计算,(2)场导表面参数化,(3)四边形网格提取。尽管在大多数网格化场景中,用户指定了所需的目标四边形或边长,但通常仅从步骤2开始才考虑此信息,而在跨场计算步骤中则没有考虑。在输入网格中存在小规模几何特征或拓扑特征或噪声的情况下,这将成为一个问题:在交叉场中会产生紧密放置的奇异点,而在具有指定边缘长度的四边形网格中,顶点无法正确复制这些奇点,从而导致严重变形甚至网格划分算法失败。我们以可感知规模的方式重新构造了交叉场的构造以及场引导的参数化,从而有效地抑制了密集分布的特征和几何以及拓扑类型的噪声。通过将不变的输入网格作为计算域,可以在输出中充分保留主要的大规模特征。

著录项

相似文献

  • 外文文献
  • 中文文献
  • 专利
获取原文

客服邮箱:kefu@zhangqiaokeyan.com

京公网安备:11010802029741号 ICP备案号:京ICP备15016152号-6 六维联合信息科技 (北京) 有限公司©版权所有
  • 客服微信

  • 服务号