散乱点云的三角网格重构

基于增量扩散法的思想,提出并实现了一个散乱点云的三角网格重构算法,算法首先利用体素网格的散列表对散乱点进行组织,然后在确定了初始种子三角形的基础上,基于活动边扩展规则构造新的三角形,使网格不断向周围扩展直到活动边表空为止,最后算法合并棱边并计算每个三角形的顶点法矢,最终构造出散乱点云的三角网格。     

Robust hashing for multi-view data: Jointly learning low-rank kernelized similarity consensus and hash functions

Learning hash functions/codes for similarity search over multi-view data is attracting increasing attention, where similar hash codes are assigned to the data objects characterizing consistently neighborhood relationship across views. Traditional methods in this category inherently suffer three limitations: 1) they commonly adopt a two-stage scheme where similarity matrix is first constructed, followed by a subsequent hash function learning; 2) these methods are commonly developed on the assumption that data samples with multiple representations are noise-free,which is not practical in real-life applications; and 3) they often incur cumbersome training model caused by the neighborhood graph construction using all N points in the database (O(N)). In this paper, we motivate the problem of jointly and efficiently training the robust hash functions over data objects with multi-feature representations which may be noise corrupted. To achieve both the robustness and training efficiency, we propose an approach to effectively and efficiently learning low-rank kernelized¹ hash functions shared across views. Specifically, we utilize landmark graphs to construct tractable similarity matrices in multi-views to automatically discover neighborhood structure in the data. To learn robust hash functions, a latent low-rank kernel function is used to construct hash functions in order to accommodate linearly inseparable data. In particular, a latent kernelized similarity matrix is recovered by rank minimization on multiple kernel-based similarity matrices. Extensive experiments on real-world multi-view datasets validate the efficacy of our method in the presence of error corruptions.We use kernelized similarity rather than kernel, as it is not a squared symmetric matrix for data-landmark affinity matrix.     

Combinatorial mesh optimization

A new mesh optimization framework for 3D triangular surface meshes is presented, which formulates the task as an energy minimization problem in the same spirit as in Hoppe et al. (SIGGRAPH’93: Proceedings of the 20th Annual Conference on Computer Graphics and Interactive Techniques, 1993). The desired mesh properties are controlled through a global energy function including data attached terms measuring the fidelity to the original mesh, shape potentials favoring high quality triangles, and connectivity as well as budget terms controlling the sampling density. The optimization algorithm modifies mesh connectivity as well as the vertex positions. Solutions for the vertex repositioning step are obtained by a discrete graph cut algorithm examining global combinations of local candidates.     

Sculptured surface machining using triangular mesh slicing

In this paper, an optimized procedure for tool path generation in regional milling is presented. The proposed procedure computes tool paths by slicing a CL-surface (Cutter Location surface), which is a triangular, mesh containing invalid triangles. Tool path generation consists of two steps: firstly, it obtains a set of line segments by slicing the triangular mesh with two-dimensional geometric elements (slicing elements), and, secondly, it extracts a valid tool path from the line segments by removing invalid portions. Two algorithms based on the slicing elements are presented: a ‘line projection’ algorithm based on the plane sweeping paradigm, which works efficiently by using the characteristics of a monotone chain; and a ‘curve projection’ algorithm for the projection of curves, which transforms the curve projection problem into a line projection problem by mapping the XYZ-space of the cylinder surface to the TZ-plane of the unfolded cylinder. The proposed procedure has been implemented and applied to tool path generation in regional milling. Performance tests show the efficiency of the proposed procedure.     

Automatic LEFM crack propagation method based on local Lepp–Delaunay mesh refinement

A numerical method for 2D LEFM crack propagation simulation is presented. This uses a Lepp–Delaunay based mesh refinement algorithm for triangular meshes which allows both the generation of the initial mesh and the local modification of the current mesh as the crack propagates. For any triangle t, Lepp(t) (Longest Edge Propagation Path of t) is a finite, ordered list of increasing longest edge neighbor triangles, that allows to find a pair of triangles over which mesh refinement operations are easily and locally performed. This is particularly useful for fracture mechanics analysis, where high gradients of element size are needed. The crack propagation is simulated by using a finite element model for each crack propagation step, then the mesh near the crack tip is modified to take into account the crack advance. Stress intensify factors are calculated using the displacement extrapolation technique while the crack propagation angle is calculated using the maximum circumferential stress method. Empirical testing shows that the behavior of the method is in complete agreement with experimental results reported in the literature. Good results are obtained in terms of accuracy and mesh element size across the geometry during the process.     

Surface interpolation of meshes by geometric subdivision

Subdivision surfaces are generated by repeated approximation or interpolation from initial control meshes. In this paper, two new non-linear subdivision schemes, face based subdivision scheme and normal based subdivision scheme, are introduced for surface interpolation of triangular meshes. With a given coarse mesh more and more details will be added to the surface when the triangles have been split and refined. Because every intermediate mesh is a piecewise linear approximation to the final surface, the first type of subdivision scheme computes each new vertex as the solution to a least square fitting problem of selected old vertices and their neighboring triangles. Consequently, sharp features as well as smooth regions are generated automatically. For the second type of subdivision, the displacement for every new vertex is computed as a combination of normals at old vertices. By computing the vertex normals adaptively, the limit surface is G¹ smooth. The fairness of the interpolating surface can be improved further by using the neighboring faces. Because the new vertices by either of these two schemes depend on the local geometry, but not the vertex valences, the interpolating surface inherits the shape of the initial control mesh more fairly and naturally. Several examples are also presented to show the efficiency of the new algorithms.     

Multiresolution analysis on irregular surface meshes

Wavelet-based methods have proven their efficiency for visualization at different levels of detail, progressive transmission, and compression of large data sets. The required core of all wavelet-based methods is a hierarchy of meshes that satisfies subdivision-connectivity. This hierarchy has to be the result of a subdivision process starting from a base mesh. Examples include quadtree uniform 2D meshes, octree uniform 3D meshes, or 4-to-1 split triangular meshes. In particular, the necessity of subdivision-connectivity prevents the application of wavelet-based methods on irregular triangular meshes. In this paper, a “wavelet-like” decomposition is introduced that works on piecewise constant data sets over irregular triangular surface meshes. The decomposition/reconstruction algorithms are based on an extension of wavelet-theory allowing hierarchical meshes without property. Among others, this approach has the following features: it allows exact reconstruction of the data set, even for nonregular triangulations, and it extends previous results on Haar-wavelets over 4-to-1 split triangulations     

医学图像三维重建及实时性研究

使用VTK工具包和MC重建算法开发了一套医学影像三维重建系统,并着重对实时性进行研究。由于重建后图像数据的处理时间和存储代价与三角形网格中三角形数量成正比,过于复杂和细节化的网格会给图像数据的存储、传输、计算和实时绘制等带来负担,故采用了顶点合并的三角形网格简化方法来减少三角形数量。另外,网格存储中存在公共顶点的大量重复存储,故提出了三角形网格的哈希映射存储方法,消除了顶点的重复存储。     

散乱数据点的增量快速曲面重建算法

给出了一个新的散乱数据的曲面重建算法.算法充分利用邻近点集反映出的局部拓扑和几何信息,基于二维Delaunay 三角剖分技术快速地实现每个数据点的局部拓扑重建,然后通过自动矫正局部数据点的非法连接关系,以增量扩张的方式把局部三角网拼接成一张标准的整体二维流形网格.该算法在重建过程中能自动进行洞的检测,判断出散乱数据所蕴涵的开或闭的拓扑结构.实验结果表明,该算法高效、稳定,可以快速地直接重构出任意拓扑结构的二维流形三角形网格.     

Uniform supercloseness of Galerkin finite element method for convection–diffusion problems with characteristic layers

In this paper, we consider a singularly perturbed convection–diffusion equation posed on the unit square, where the solution has two characteristic layers and an exponential layer. A Galerkin finite element method on a Shishkin mesh is used to solve this problem. Its bilinear forms in different subdomains are carefully analyzed by means of a series of integral inequalities; a delicate analysis for the characteristic layers is needed. Based on these estimations, we prove supercloseness bounds of order $3∕2$ (up to a logarithmic factor) on triangular meshes and of order $2$ (up to a logarithmic factor) on hybrid meshes respectively. The result implies that the hybrid mesh, which replaces the triangles of the Shishkin mesh by rectangles in the exponential layer region, is superior to the Shishkin triangular mesh. Numerical experiments illustrate these theoretical results.     

Photo Hull regularized stereo

A regularization-based approach to 3D reconstruction from multiple images is proposed. As one of the most widely used multiple-view 3D reconstruction algorithms, Space Carving can produce a Photo Hull of a scene, which is at best a coarse volumetric model. The two-view stereo algorithm, on the other hand, can generate a more accurate reconstruction of the surfaces, provided that a given surface is visible to both views. The proposed method is essentially a data fusion approach to 3D reconstruction, combining the above two algorithms by means of regularization. The process is divided into two steps: (1) computing the Photo Hull from multiple calibrated images and (2) selecting two of the images as input and solving the two-view stereo problem by global optimization, using the Photo Hull as the regularizer. Our dynamic programming implementation of this regularization-based stereo approach potentially provides an efficient and robust way of reconstructing 3D surfaces. The results of an implementation of this theory is presented on real data sets and compared with peer algorithms.     

在机检测中三角网格拓扑重建方法的研究

对STL（Stereo Lithographic）模型中三角面片进行拓扑重建是解决在机检测系统中对零件模型表面进行曲面划分的前提条件之一。在建立点-边拓扑关系的基础上,将原有基于顶点坐标值浮点数比较的半边匹配结构转化为基于索引值的整型数匹配,并提出了基于关联-散列结构三角网格拓扑重建方法。该方法采用散列结构为辅助数据结构来减少STL模型顶点归并和边界归并的计算量,利用STL模型相邻网格单元数据相关性降低哈希函数的构建难度并提高冲突处理效率,在滤除冗余数据的同时完成拓扑关系的建立。计算实例验证了该算法的高效性和有效性。     

Surface reconstruction from scan paths

Given a finite set of points sampled from a surface, a particular important topic of surface reconstruction is to find a triangular mesh connecting the sampling points and approximating the surface. We introduce a new solution for this problem which takes into consideration hidden structural information in form of scan paths. The scan paths, which often are available in sampling data sets, allow us to cope with creases or ridges more reliably than algorithms assuming unstructured point clouds. Moreover, the consideration of scan paths leads to a better reconstruction of surface boundaries than other heuristics.     

基于三角片拼合的STL网格模型重建算法

采用三角片拼合的方法递增建立网格模型,新增三角片被归纳为5种类型.首先通过半边匹配确定新增三角片的类型;然后针对每一种类型使用不同的方法去除冗余顶点,并同时建立拓扑关系;最后通过顶点数组紧缩消除顶点空隙,将半边hash表转化为半边数组,完成网格模型的重建.该模型为改进的半边结构,具有完整的形状和拓扑信息,可以表示非流形边,是法矢调整、网格分块等后续处理的理想起点.实验结果表明,文中算法高效、鲁棒、可扩展.     

基于八叉树和混合搜索树的地质曲面快速求交方法

为处理地质界面之间的空间相交关系,提出一种新的针对三角地质曲面的快速求交方法。该方法融合优化八叉树法和OBB搜索树方法,可以更快速准确地剔除远离交线的其他三角形。求交剩余的三角形得到交线,应用三角网局部重构和网格优化算法修正交线附近的三角网,最终分割交线两侧的地质曲面,完成2个地质曲面的离散化求交过程。与AABB、OBB和空间分解法相比,该方法在大数据量三角曲面求交中效率优势明显,可以快速准确处理地质模型构建和分析中的曲面求交问题,为三维地质模型自动化构建的实现提供有效支撑。     

Evolution of T-spline level sets for meshing non-uniformly sampled and incomplete data

Given a large set of unorganized point sample data, we propose a new framework for computing a triangular mesh representing an approximating piecewise smooth surface. The data may be non-uniformly distributed, noisy, and may contain holes. This framework is based on the combination of two types of surface representations, triangular meshes and T-spline level sets, which are implicit surfaces defined by refinable spline functions allowing T-junctions. Our method contains three main steps. Firstly, we construct an implicit representation of a smooth (C ² in our case) surface, by using an evolution process of T-spline level sets, such that the implicit surface captures the topology and outline of the object to be reconstructed. The initial mesh with high quality is obtained through the marching triangulation of the implicit surface. Secondly, we project each data point to the initial mesh, and get a scalar displacement field. Detailed features will be captured by the displaced mesh. Finally, we present an additional evolution process, which combines data-driven velocities and feature-preserving bilateral filters, in order to reproduce sharp features. We also show that various shape constraints, such as distance field constraints, range constraints and volume constraints can be naturally added to our framework, which is helpful to obtain a desired reconstruction result, especially when the given data contains noise and inaccuracies.     

一种用于表面重建的网格点生成算法

表面重建需要建立表面的多边形(通常为三角形)网格模型,该文给出的算法通过两项参数来控制边缘轮廓上生成的网格点,所生成的网格点不但能够体现边缘的主要特征,而且可以避免在重建时产生斜三角形。该算法具有速度快、适应性强以及可灵活调整重建精度的特点,已在医学影像工作站上的三维重建系统中获得了较好的应用。     

Detail control in line drawings of 3D meshes

We address the problem of rendering a 3D mesh in the style of a line drawing, in which little or no shading is used and instead shape cues are provided by silhouettes and suggestive contours. Our specific goal is to depict shape features at a chosen scale. For example, when mesh triangles project into the image plane at subpixel sizes, both suggestive contours and silhouettes may form dense networks that convey shape poorly. The solution we propose is to convert the input mesh to a multiresolution representation (specifically, a progressive mesh), then view-dependently refine or coarsen the mesh to control the size of its triangles in image space. We thereby control the scale of shape features that are depicted via silhouettes and suggestive contours. We propose a novel refinement criterion that achieves this goal and address the problem of maintaining temporal coherence of silhouette and suggestive contours when extracting them from a changing mesh.     

Fast Recursive Computation of Krawtchouk Polynomials

Krawtchouk polynomials (KPs) and their moments are used widely in the field of signal processing for their superior discriminatory properties. This study proposes a new fast recursive algorithm to compute Krawtchouk polynomial coefficients (KPCs). This algorithm is based on the symmetry property of KPCs along the primary and secondary diagonals of the polynomial array. The \(n-x\) plane of the KP array is partitioned into four triangles, which are symmetrical across the primary and secondary diagonals. The proposed algorithm computes the KPCs for only one triangle (partition), while the coefficients of the other three triangles (partitions) can be computed using the derived symmetry properties of the KP. Therefore, only N / 4 recursion times are required. The proposed algorithm can also be used to compute polynomial coefficients for different values of the parameter p in interval (0, 1). The performance of the proposed algorithm is compared with that in previous literature in terms of image reconstruction error, polynomial size, and computation cost. Moreover, the proposed algorithm is applied in a face recognition system to determine the impact of parameter p on feature extraction ability. Simulation results show that the proposed algorithm has a remarkable advantage over other existing algorithms for a wide range of parameters p and polynomial size N, especially in reducing the computation time and the number of operations utilized.     

针对密集点云的快速曲面重建算法

为了能够快速地从高密度散乱点云生成三角形网格曲面,提出一种针对散乱点云的曲面重建算法.首先通过逐层外扩建立原始点云的近似网格曲面,然后对近似网格曲面进行二次剖分生成最终的精确曲面;为了能够处理噪声点云,在剖分过程中所有网格曲面顶点都通过层次B样条进行了优化.相比于其他曲面重建方法,该算法剖分速度快,且能够保证点云到所生成的三角网格曲面的距离小于预先设定容限.实验结果表明,文中算法能够有效地实现高密度散乱点云的三角剖分,且其剖分速度较已有算法有大幅提高.