基于八叉树的大规模流程工厂模型细节裁剪算法

发布时间：2018-03-01 08:26

本文关键词： 流程工厂模型细节裁剪八叉树体素特征帧率　出处：《计算机应用》2017年09期 　论文类型：期刊论文

【摘要】：针对传统的细节裁剪算法,处理粒度为面片级,无法在规定时间内快速处理顶点、面片规模数以亿计的流程工厂模型的问题,提出一种基于八叉树的大规模流程工厂模型细节裁剪算法,用于解决传统细节裁剪效率低下的问题,以保证较高的交互帧率。该方法依据组成流程工厂模型的构件体素特征以及空间特征,提出用于量化构件大小的占屏值,并结合建立的八叉场景树,依据所述占屏值近似估算一系列构件投影至屏幕所占据的像素点数量上限,从而快速判断是否裁剪。实验结果表明,所提算法简单有效,与现今主流的漫游软件在加载具有10 000根管线的工厂模型后进行对比,其帧率至少提升50%,显著提升了平台的交互流畅度,对流程工厂行业以及图形平台整体的设计水平的提升有积极的意义。
[Abstract]:In view of the traditional detail clipping algorithm, the process factory model, which has a grain size of surface level, can not be quickly processed within a specified time, and has hundreds of millions of face sizes, can not be processed quickly. In this paper, an octree based detail clipping algorithm for large-scale process factory model is proposed, which is used to solve the problem of low efficiency of traditional detail clipping. In order to ensure high interactive frame rate, according to the component voxel feature and spatial feature of the process factory model, this method proposes a screen value to quantify the component size, and combines with the established octa-scene tree. The upper limit of the number of pixels projected by a series of components onto the screen is estimated approximately according to the occupancy value of the screen, so as to quickly judge whether or not to cut. The experimental results show that the proposed algorithm is simple and effective. Compared with today's popular roaming software that loads factory models with 10,000 pipelines, its frame rate increases by at least 50 percent, significantly increasing the platform's interaction fluency. Process factory industry and graphics platform as a whole design level improvement has a positive significance.
【作者单位】：中国科学院计算技术研究所;中国科学院大学计算机与控制学院;北京中科辅龙计算机技术股份有限公司;
【分类号】：F49;TP391.41

【相似文献】