中国科学院软件研究所机构知识库
Advanced  
ISCAS OpenIR  > 中科院软件所  > 中科院软件所
题名:
虚拟现实中三维复杂几何形体的层次细节模型的研究
作者: 刘学慧
答辩日期: 1998
专业: 计算机软件
授予单位: 中国科学院软件研究所
授予地点: 中国科学院软件研究所
学位: 博士
关键词: 虚拟现实 ; 细节层次模型 ; 网格模型简化 ; 实时连续多分辨率绘制 ; 地表模型 ; 视觉模型 ; 全局光照模型 ; 全局光照网格模型
摘要: 虚拟现实技术是一种高度逼真地模拟人在自然环境中视、听、动等行为的人机界面技术。图形的实时生成是VR成为“现实”的关键技术之一。本论文围绕层次细节模型技术这一有效的图形生成加速方法展开了有益的探索,研究各种LoD模型生成方法及实时绘制技术。研究成果包括以下几个方面:1)提出一种新的渐进网格模型的生成方法和表示形式,并提出新的非规则网格LoD模型的实时绘制方法。2)以焦点以及显示面积二者有效的结合建立较为精确的视觉空间模型,改进基于均匀网格表示物体的LoD模型的实时生成及绘制技术。3)提出一种基于辐射度全局光照网格模型的简化方法。全文共分五章。第一章简单介绍了VR的基本概念、主要特点。对虚拟环境图形生成的各种加速方法作了较为全面的综述,同时阐明了虚拟现实中LoD模型自动生成的绘制技术的研究内容和应用价值。就LoD模型技术的各个相关方面:LoD模型的选择尺度、LoD模型的平滑过渡、LoD模型的选择算法以及LoD模型的自动生成进行了深入的研究,回顾了国内外的研究状况。本论文就各种LoD模型生成方法及实时绘制技术进行了一些尝试和探索,取得了一些成果。下面逐章进行介绍。以几何元素删除实现模型的简化是多细节层次模型自动生成的常用方法。但由于复杂模型庞大的数据量以及模型简化复杂的计算,使得以往的LoD模型生成方法只能预先产生多个间断的简化模型,从而引起实时绘制时图形画面的跳跃。为解决这个问题,人们相继提出了渐进网格模型概念以及三维复杂模型的实时边疆的多分辨率制技术。本文提出一种新的渐进网格生成方法和表示形式。该算法记录下整个模型简化过程中顶点删除的信息,并利用顶点删除与模型面片个数的对应关系,建立起渐进网格的表示形式。利用这种渐进网格表示形式我们实现了三维复杂模型的实时连续多分辨率绘制。地表模型在虚拟现实中有着广泛而重要的应用。非规则网格模型和均匀网格模型作为地表等数字化高度场模型的两种表示形式,在以往的地表模型的层次细节模型生成算法中有着各自独立的特点。Linstorm结合非规则网格表示的高效性和均匀网格模型良好的自适应性,提出一种地表模型的实时连续多分辨率绘制方法。但该方法忽略了人眼对处于不同视觉区域的模型细节的不同分辨能力,对处于不同视觉区域的模型赋予相同的细节,这必然带来图形生成的浪费。本文提出具有聚集因子的基于图象空间误差的地表模型消减算法,以焦点以及显示面积二者有效的结合建立较为精确的视觉空间模型,以此模型衡量人眼对由于面片合并所带来的误差的分辨能力,并以此作为地表模型细节的重要度评价尺度,有效地简化了地表模型的绘制。同时,算法结合均匀网格模型的多分辨率细节层次模型,以“快”作为地表模型大面积简化的空间单位,加速地表模型的简化操作,以实现较为复杂的地表模型的实时绘制。物体表面的光照颜色纹理不仅为人类提供了丰富的视觉效果,而且为人类提供了丰富的形体、邻近及遮挡等视觉效果。但全局光照模型的计算结果往往导致模型复杂度几十甚至几百倍的增加。本文结合辐射度全局光照模型的计算特点,提出一种辐射度全局光照网格模型的简化方法。该方法根据人眼的视觉特点和辐射度计算特点,以最大相对变化值为准则,以面片合并简化操作将最大相对变化在用户定义范围内的面片合并起来,实现辐射度全局光照网络模型的第一步简化,并以原模型顶点的RGB值的变化进一步加大辐射度网格模型的简化,建立起整个环境具有辐射度全局光照效果的多细节层次模型。本算法不仅有效地简化了辐射度全局光照网格模型,而且能较好地保持原光照网格模型的特征。最后,论文对作者的工作作了总结,并根据当前目前的研究结果提出了进一步的工作方向。
英文摘要: Virtual Reality(VR) is an advanced man-machine interface which allows the behaviors of the human being, such as viewing, listening and touching, to be conducted in a natural but virtual environment. Graphics rendering is one of the key techniques for VR. In this thesis, the generation and rendering of models at multiple level of detail(LoD), one of the most efficient methods to accelerate graphics rendering in virtual reality, is studied and archive some new results, which include Chapter 1 firstly describe the basic concepts and primary characters of VR system, and give a systematic survey on various method for accelerating graphics rendering in VR. At the same time, the importance associated with LoD models is emphasized. After addressing the research contents associated with LoD model, including switch between models at different levels of detail, threshold and heuristics for selecting models at multiple level of detail and method to automatic generate models at multiple level of detail, it review the development of methods of automatic generation and rendering of models at multiple level of detail(LoD), covering previous works, key techniques and main result. Here, we concentrate our research on the generation and rendering of models at multiple level of detail(LoD) and some new results are archived. Deleting basic geometric element, such as vertex, edge and polygon, to simplify the original complex mesh is one of the methods to create models at multiple level of detail. Because the extreme complexity of original model and the expensive computation of simplifying process, some key models at different level of detail are generated in advance and stored for rendering, which result in objectionable visual "popping" effect when changing between models at different level of detail in rendering. In order to solve this problem, the concept of progressive mesh(PM) and method of real-time, multi-resolution modeling and rendering of 3D complex virtual environment are proposed. Here, we present a new expression of progressive mesh and new method for constructing PM from an arbitrary mesh. As the initial mesh is simplified into an coarse mesh by applying a sequence of successive vertex removal, a sequence of detail records that indicate how to incrementally refine the coarse mesh exactly back into the original mesh is stored. Using the relationship between number of vertex decimated and number of triangle removed, the new PM representation is get and a new algorithm for generating continuous levels of detail of a given triangle object in real-time is proposed. The experimental results showed that the methods is useful. Terrain surface model has its important and wide-ranged application background in virtual reality. As two basic representation of terrain surface, uniform grid surface triangulation and irregular triangulation network had its own merits. By taking advantage of the two kinds of store formats for terrain surface, Linstrom et al. introduced an algorithm of real-time continuous level of detail rendering of height field that combines the large complexity reduction obtained in TIN construction and the flexibility provided by regular grid. However, the algorithm ignores the different sensibility of human vision to images on different view portion and treats all the points at the same horizontal distance alike, which generate a high constant density and waste a lot of model detail where the user is not looking. Here we present a simplification algorithm which combine size and focus criteria to capture the sensibility of human vision to image quality and provide a finer visual model to estimate the sensibility of human vision to model detail to increase the simplification of terrain surface model. At the same time, our algorithm builds the whole model into block-based multi-resolution representation and uses large scale block-based estimation to accelerate the simplification of whole model. The algorithm has been implemented for approximating and rendering digital terrain models and the experimental results show its high efficiency. The illumination not only give rich visual expression, but also provide the visual effect of shape, vicinity and occlusion etc. But the solution to a global illumination problem consists of an illumination function for each surfaces of the original model, usually represented by an illumination meshes, which result in dramatically increase of model complexity. In this paper, we present an automatic method for simplifying the meshes produced as the solution to global-illumination problem using radiosity method. Based on the character of illumination meshes produced by radiosity and the characteristic of human eyes, maximum relative difference criteria is developed to guide the simplification process. The method consist of three steps: it firstly simplify the original meshes by merging faces and then delete the vertices on the perimeters of the merged face, last triangulate the perimeters. The algorithm not only significantly reduced the complexity of original meshes model, but also performed well to capture the coherent shadow discontinuity feature of illumination meshes. Model simplification samples based on these method is given and good result is archived.
语种: 中文
内容类型: 学位论文
URI标识: http://ir.iscas.ac.cn/handle/311060/6206
Appears in Collections:中科院软件所

Files in This Item:
File Name/ File Size Content Type Version Access License
N98834.pdf(2347KB)----限制开放-- 联系获取全文

Recommended Citation:
刘学慧. 虚拟现实中三维复杂几何形体的层次细节模型的研究[D]. 中国科学院软件研究所. 中国科学院软件研究所. 1998-01-01.
Service
Recommend this item
Sava as my favorate item
Show this item's statistics
Export Endnote File
Google Scholar
Similar articles in Google Scholar
[刘学慧]'s Articles
CSDL cross search
Similar articles in CSDL Cross Search
[刘学慧]‘s Articles
Related Copyright Policies
Null
Social Bookmarking
Add to CiteULike Add to Connotea Add to Del.icio.us Add to Digg Add to Reddit
所有评论 (0)
暂无评论
 
评注功能仅针对注册用户开放,请您登录
您对该条目有什么异议,请填写以下表单,管理员会尽快联系您。
内 容:
Email:  *
单位:
验证码:   刷新
您在IR的使用过程中有什么好的想法或者建议可以反馈给我们。
标 题:
 *
内 容:
Email:  *
验证码:   刷新

Items in IR are protected by copyright, with all rights reserved, unless otherwise indicated.

 

 

Valid XHTML 1.0!
Copyright © 2007-2017  中国科学院软件研究所 - Feedback
Powered by CSpace