基于扩散距离与几何约束的模型内蕴自对称检测  

Shapes intrinsic self-symmetry detection based on diffusion distance and geometric constraints

作　　者：韩丽[1] 刘俣男 李丹 Han Li;Liu Yunan;Li Dan(School of Computer ＆ Information Technology,Liaoning Normal University,Dalian Liaoning 116081,China)

机构地区：[1]辽宁师范大学计算机与信息技术学院,辽宁大连116081

出　　处：《计算机应用研究》2018年第6期1896-1899,1905,共5页Application Research of Computers

基　　金：辽宁省高等学校优秀人才支持项目(LJQ2013110)

摘　　要：针对三维网格模型几何处理中提高算法效率和拓扑噪声不敏感性的要求,提出了基于扩散几何约束的非刚性三维模型内蕴自对称检测方法。通过计算模型的Laplace-Beltrami算子来提取顶点的热核特征描述符,比较描述符之间的扩散距离,与基于谱图理论约束的几何相似性矩阵有效融合,实现形状度量的优化;最后通过线性检测的方式快速获取模型对称点集合,实现模型的自对称形状分析。实验结果进一步验证了该方法不仅能够高效地实现等距非刚性变换模型的内蕴自对称性检测,而且对于残缺模型的对称分析更具有鲁棒性。Aiming at the requirement related to the insensitivity under efficiency and topological noise during the geometry processing of 3 D shapes,this paper proposed an optimized algorithm,which detected the intrinsic self-symmetry among the non-rigid shapes based on heat diffusion geometric constraints. Firstly,it extracted the heat kernel signature descriptor of the vertex by computing the Laplace-Beltrami operator of the model. Compared the diffusion distance between heat kernel signature descriptors,it fused of the geometric similarity matrix based on spectral graph theory,optimized the shape metric. Finally,it obtained a set of symmetric point pairs by searching loop. A series of experimental results show that the proposed algorithm can effectively detect intrinsic self-symmetry for isometric non-rigid transformation model.

关 键 词：Laplace-Beltrami 热核特征 扩散距离 谱图理论 内蕴自对称 

分 类 号：TP391.41[自动化与计算机技术—计算机应用技术]