基于特征的三维模型简化算法研究

Research of Feature-based 3-D Models Simplification

细节层次模型(Level of Detail,LOD)是模型简化最有效的方法之一。针对以往细节层次模型简化算法对模型的特征与边界处理的不足,提出了一种基于特征的LOD模型简化算法,并给出了一种针对模型特征与模型边界的折叠代价函数,较好的解决了以往算法生成LOD模型时对模型重要特征与边界处理的不足,在对原始模型简化的同时,能够很好的保留模型的边界与特征。实验结果证明了算法的高效、可靠,既保持模型细节特征,又减少了模型简化误差。

Level of detail （LOD） is the most efficiency way to simplify 3D models. But the LOD methods before can＇t deal with the feature and boundary of 3D models very well, they will lost too many important feature in the simplification results. A new LOD method was proposed based on models feature to settle this problem, and a new cost function taking advantages of the sharp features and boundary was given. The method can preserve the important feature and boundary of 3D models efficiently. The results show that the method is highly efficient in terms of both space and time cost, and it can render large models with important model features well preserved.