一种新的基于顶点聚类的网格简化算法

A NEW MESH SIMPLIFICATION ALGORITHM BASED ON VERTEX CLUSTERING

在计算机图形学中，经常采用多边形网格来描述物体模型．由于绘制时间和存储量与多边形的数量成正比，过于庞大的物体网格模型通常是不实用的．模型简化在计算机动画、虚拟现实和交互式可视化等计算机图形应用领域有着广阔的应用前景．为此提出一种新的基于顶点聚类的网格简化算法．该算法利用八叉树对网格进行自适应划分，给出了一种基于点到平面距离的有效的误差控制方法，并能在用户指定的误差范围内通过使原始网格中的顶点聚类达到大量简化的目的．该算法实现简单，速度快且能很好地保持边界特征．

In computer graphics,models are often discribed by polygonal meshes.Because the rendering and storage cost is proportional to the number of polygons,too large models are not practical.Model simplification techniques are widely used in computer graphics fields such as computer animation,virtual reality and interactive scientific visualization.In this paper a new algorithm of mesh simplification based on vertex clustering is presented.The algorithm adopts octree structure to subdivide the mesh model adaptively.A new error control method is also presented.The implementation of the algorithm is simple and it runs very fast.Examples illustrate the efficiency of the algorithm.