基于表面分割的大型三维网格模型多分辨率表示并行构建与绘制

项目名称： 基于表面分割的大型三维网格模型多分辨率表示并行构建与绘制

项目编号： No.61262070

项目类型： 地区科学基金项目

立项/批准年度： 2013

项目学科： 自动化技术、计算机技术

项目作者： 张亚萍

作者单位： 云南师范大学

项目金额： 43万元

中文摘要： 多分辨率技术作为一种非常有效的绘制加速技术，一直得到广泛的应用。但是对于大型三维网格模型而言，这种技术的设计和实现本身存在诸多难点。受限于内存容量，网格模型的几何数据及其辅助数据结构不能完全装载进内存，无法直接应用传统的多分辨率构建与绘制算法；受限于总线带宽和CPU处理能力，大型三维网格模型的多分辨率构建普遍耗时很长。本项目研究基于表面分割的大型三维网格模型多分辨率表示并行构建与绘制算法，提高模型多分辨率表示的构建速度，保证最佳的绘制质量；研究大型三维网格模型的数据存储访问优化技术，提高多分辨率表示的绘制速度。上述研究内容是大型三维网格模型多分辨率表示构建与绘制领域重要的理论与技术基础。研究成果可以直接应用于文物保护、影视娱乐、旅游等行业的虚拟展示。

中文关键词： 大型三维网格模型；多分辨率表示；缓存优化；GPU；基于图像的建模

英文摘要： The multiresolution technique, which is one of the most efficient approaches to improve the rendering performance, has become a hot topic in the field of computer graphics. However, there are many difficulties in the design and implementation of this technique for massive meshes. Firstly, as geometric data and auxiliary data structures of massive meshes can not be completely loaded into memory due to limited memory size, the traditional multiresolution modeling and rendering algorithm can not be directly applied to the massive meshes. Secondly, with limited bus bandwidth and CPU processing power, the construction process of multiresolution representation for massive meshes often takes much time. This project researches a novel approach for parallel constructing and rendering multiresolution representation for massive meshes based on surface segmentation, which provides vertex-grained local refinement and generates the optimal rendering quality. It also researches data memory access optimization of multiresolution representation for massive meshes,which is used to improve the rendering speed of multiresolution representation. These researches are important theoretical and technical basis of constructing and rendering multiresolution representation for massive meshes. The research results can be used to the areas

英文关键词： massive 3D mesh model；multiresolution representation；cache optimization；graphics processing unit；image-based modeling