节点文献
基于体细分的自由变形技术研究
The Free-Form Deformation Technique Base on the Volumetric Subdivision Scheme
【作者】 张华;
【导师】 潘日晶;
【作者基本信息】 福建师范大学 , 计算机应用技术, 2013, 硕士
【摘要】 近年来,随着计算机及互联网技术的迅速发展,三维几何造型技术的应用正越来越广泛,在工业建模、视频处理、游戏制作、影视特技等各行各业都有着及其重要的作用。空间变形是几何造型中一种重要的几何外形编辑技术。传统的自由变形算法控制手段不够灵活,难以进行直观有效的交互操作。另外初始控制网格生成作为自由变形实现的必要前提,已有的方法还存在算法复杂,实现困难,存储量大等问题。本文提出了一种直观、实时的基于体细分技术的自由变形方法,使用任意拓扑的三维控制网格体作为控制手段,来编辑具有大数据量的网格模型外形。本文的主要工作包括:首先,提出了一种新的网格生成算法,该算法克服了传统网格生成算法难以表达任意外形,以及网格数据量增长过快等问题,生成和物体模型形状相似、多分辨率的初始控制网格体;其次,提出了一种新的四面体网格细分方法,该方法克服了传统体细分方法存在的对角线选择或者四面体细分成四面体和八面体的问题,对一个由四面体单元组成的网格实体,体内进行体细分,细分后的体网格全部由四面体单元组成,而实体的表面,采用一种带参数的类Loop曲面细分,用于控制细分后实体的形状;最后,提出了一种新的交互式自由变形方法,克服了传统自由变形方法不够直观,控制手段难以近似任意拓扑的物体形状等问题,利用体细分技术通过任意拓扑的多分辨率三维控制网格来控制物体的变形,根据顶点与控制四面体的距离计算权值,并对所有控制四面体的影响进行加权平均,从而得到光滑的变形效果。与已有的自由变形算法相比,本文提出的算法控制手段灵活、直观,可以实时交互控制,既可以实现全局变形,也可以对局部细节进行良好的编辑。此外,本文还结合了VS.NET和OpenGL技术,设计开发了一个自由变形造型系统,实现本文的各个算法。实验表明,该系统能较好的实现三维曲面模型自由变形,效果良好。
【Abstract】 These years, with the rapid development of computer and internet technology, the three-dimensional geometric modeling technique plays an increasingly important role in many fields, such as industrial modeling, video processing, game production, film and television special effects and so on. Space deformation is an important tool for shape modification in geometric modeling field.The traditional free-form deformation algorithms are not flexible enough, and it is hard to carry out intuitively and effectively. As to the initial control mesh, the necessary precondition of the Free-form deformation, there has been some problems such as too much complexness in algorithm, hard to implement, large storage and so on. A new interactive, real-time deformation scheme based on volumetric subdivision is proposed in this paper, which use three-dimensional volumetric meshes of arbitrary topology as a means of control, to edit the model which has a large amount of data.The main work of this paper include:First, a new algorithm of volumetric meshes generation is proposed, which overcomes the problems of hard to express arbitrary shape, and excessive growth of the amount of data of the traditional algorithm. A multi-resolution volumetric meshes, which has a similar shape with the object, will be generated by the new algorithm. Second, a new subdivision scheme for tetrahedral meshes is proposed, which overcomes the existing problems of the directional preferences, or the generated volumetric meshes composed of tetrahedrons and octahedrons of the current algorithm. For the volumetric meshes of tetrahedron, the inners of the meshes are subdivided by using the volumetric subdivision, and the surfaces of the meshes are subdivided by using a quasi-Loop surface subdivision with a parameter to control the shapes of the solids. Finally, a new interactive free-form deformation method is brought out. The existing method is not intuitive enough by using the parallelepiped as control meshes. The new method carries out the deformation by using a multi-resolution volumetric meshes of arbitrary topology. The weighted value is calculated by the distance of the vertex and the tetrahedrons, and all the affects to the vertex coming from the tetrahedrons are weighted averaged to get the new position of the vertex. When the all new positions of the vertexes of the object are calculated, a smooth deformation is performed. Compared with the existing free-form deformation algorithm, the proposed algorithm is flexible, intuitive, and real-time interactive. Both the global deformation, and the local deformation can be carried out.In addition, a modeling system by the platform of VS.NET and OpenGL, is designed to implement all the algorithms in the paper. Experiments show that, the free-form deformation of the three-dimensional model could be achieved with excellent effect.
【Key words】 geometric modeling; free-form deformation; arbitrary topology; volumetricmeshes; volumetric subdivision; multi-resolution;