基于自适应空间刨分的网格简化算法

提出了一种基于自适应空间刨分的网格简化算法，算法首先对模型中的所有的顶点进行量化赋予一个二次误差阵，并将它们视为一个簇，然后沿坐标轴方向将它们刨分成八个子簇并不断迭代刨分生成新的子簇直至达到指定的精度，将最终的离散点集用适当的方法重新进行三角化，得到简化模型，该算法不仅速度快，能在任意限定的时间内产生一个可显示的结果，而且结果质量也很好.另外，本文还用给出的实例与其他相关算法进行了比较．

Mesh Simplification Algorithm Based on Adaptive Spatial Partition

An algorithm for mesh simplification based on adaptive spatial partition is presented in this paper. The spatial cube of model bounding box is divided in the middle of bounding box. All the vertices in a cube is a cluster and then the approximation error is computed in each of these clusters. Partition the cluster based on the approximation error using octree and iterate the process until all the approximation errors are less than threshold. The approximation of original model is generated using re triangulate the final point set. The algorithm is fast and can generate a displayable result within a given time limit. Results also have good quality. Besides, the algorithm is compared with other algorithm using complex model.