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:
基于图象的建模和绘制研究
作者: 李洪举
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: 基于图象的虚拟现实 ; 基于图象的建模和绘制 ; 表示 ; 算法 ; 全景图 ; 运动分析 ; 纹理映射 ; 图象拼接 ; 图象为形 ; 失焦 ; 景深 ; 弥散圆 ; 势函数
摘要: 基于图象的建模和绘制,或者更一般地称为基于图的虚拟现实,是九十年代中期出现的一种新的人机界面技术。虽然在虚拟现实系统中使用图象的工作可以追溯到八十年代初,但是,真正采用图象技术来构造虚拟现实系统,甚至基于图象(Image-based)这个术语,只是在三、四年前才出现。该技术一经诞生就引起了学术界和工业界的广泛注意,在短短几年内得到了迅速发展,逐渐成为虚拟现实研究的一个重要分支,被广泛应用于仿真模拟、动画等许多领域。本论文对基于图象的建模和绘制技术做了较全面的总结,对其中一些理论和应用问题进行了深入分析和探讨,取得了一些新的成果,提出了一些新的方法和模型。主要的工作包括:1. 提出了一种用于描述封闭环境的新表示:θ-Φ综合图,同时给出了基于新表示的交互算法,包括环视、俯仰视、连续变焦以及漫游等。2. 实现了一个构造室内场景的方案,该方案只需要输入少量照片,不需要苛刻的拍摄条件和专门的设备。3. 建立了一个模拟“失焦(defocus)”和景深效果的新模型和相关算法,该模型具有很强的可调节性和良好的局部性,可用于高真实感特殊效果的生成。全文共分五章:在第一章,我们首先对基于图象的虚拟现实(建模、绘制及其它相关技术)做了较为全面的总结。目前,无论在理论基础还是技术手段方面,该领域都处于迅速发展阶段,因为历史较短,其中还有许多基础问题尚待解决。本章综述了近年来国际国内的基于图象的虚拟现实以及相关领域中的工作,讨论了一些主要问题,诸如:基于图象的虚拟现实的意义和特点,实现思路,建模和绘制所用的表示和算法,以及一些待解决的问题,等等。其中也包括我们自己的一些看法。在第二章,我们探讨了场景表示的问题。该问题是基于图象的建模和绘制的关键。在基于图象的虚拟现实系统中,由于光强度信息非常丰富,而且场景的复杂性和对交互方式的要求相差很大,一般而言很难找到一个普适的表示。针对虚拟现实系统中常见的封闭环境(如建筑物内部),我们提出了一种新的表示:θ-Φ综合图。该表示将场景的信息投影在由经纬度构成的2D平面上;不但表示了场景的光强分布,同时也表示了相应的距离分布。本章还给出了基于这一新表示的交互算法,包括环视、俯仰视、连续变焦以及漫游等。与其它表示(如柱面全景图)相比,θ-Φ-综合图支持各种相机运动,数据量较小,交互算法简单快速,获取原始数据也比较容易。第三章给出了一个构造θ-Φ综合图的实例,其场景对象为建筑物的内景。事实上,建筑物的虚拟漫游问题有着很强的应用背景,但传统的建模方法大都是基于CAD技术的,需要操作者输入场景的几何结构数据,而且由于缺乏光强度信息,产生的视图也不够逼真。本章则实现了一个基于图象的建模方法:从室内场景的一组照片出发,通过运动分析确定场景的几何结构,并采用图象变换和拼接技术获取场景的光强度分布,最手得到场景的θ-Φ综合图。这一方案不需要苛刻的拍摄条件和专门的设备,操作者只需要在图象中识别出一些匹配特征,交互方式自然、方便。在第四章,我们讨论了基于图象的绘制中的另一个重要问题:景深和失焦等特殊效果的生成。特殊效果的生成对于增强视图的真实感很有作用,在传统的基于图形的绘制技术中,这一问题已得到了一些研究。但对于基于图象的绘制,由于实际拍摄的图象缺乏深度值,传统的方法不能实现这此特殊效果。有鉴于此,为了在实拍图象中产生景深和失焦效果,我们建立了一个基于势函数的弥散指数分布模型,并给出了相应的失焦变换算法。新的模型可以对图象中的各区域进行不同程度的模糊化,且模糊程度的调节非常方便。最后,第五章对本论文的工作作了总结,并提出了我们对这一领域工作的一些想法,以及进一步的工作方向。
英文摘要: As an important branch of virtual reality, image-based modeling and rendering (IBM/IBR), also called image-based virtual reality (IBVR), emerged in the mid-1990s. In the past few years, IBM/IBR developed fast and has opened a vista to new and powerful application. On the other hand, there exist a lot of problems to be solved till now. In this thesis, we explore some open questions in IBM/IBR, and achieve some new results: The first is a new representation, called θ-Φ combination image, and its relative algorithms to run in virtual enclosed environments. The second is a hybrid method for constructing virtual indoor environments, based on θ-Φ combination image. The last is a new model for simulating some types of special effects -- defocusing and depth-of-field. This is an important topic in generating more realistic images. The thesis is divided into five chapters: 1. Image-based Virtual Reality: As a prelude to subsequent chapters, this chapter provides a general overview of the IBM/IBR and IBVR techniques. Firstly, a historical perspective on the emerging techniques is given, and the potential approaches to establish photo-realistic environments are explored. Afterwards, the most challenging problem, representations and construction algorithms used in IBVR systems are discussed in detail. Finally, some open questions are addressed. 2. A New Representation for Enclosed Environment Composition: As mentioned in the first chapter, various methods have been proposed in recent years for the development of image-based VR systems where a panoramic view of cylinder surface is typically utilized. However, the solution could be only be only in general used for a constrained view, with the viewing direction rotating around a vertical axis. In order to promote the viewing freedom, to be particularly applied in many indoor environments where a view covering a full solid angle in 3D space is required, a new representation called θ-Φ combination image and a hybrid approach associated with the representation are presented in this chapter. The representation defines intensity-range function on a latitude-longitude plane, and has a viewing capability of three-degree-of-freedom in terms of camera rotating, continuous zooming and camera movement. 3. An Image-based Approach for Composing Virtual Indoor Environment: On the basis of the new representation θ-Φ combination image, an image-based approach is addressed in this chapter for constructing virtual indoor environment. It only requires a sparse set of static photographs and has no strait restriction on taking original photographs. In the approach, two main components are employed: the first is to reconstruct the geometrical model of the scene from a sequence of photographs with structure from motion (SFM) technology, and the second is to extract textures from these input photographs and then stitch them together. Experiment results show that these algorithms are robust. 4. A New Approach for Generating Depth-of-field and Defocus Effects: Depth of field and defocusing are very useful effects in computer graphics and virtual reality to enhance realism. In order to generate more realistic effects in the images taken from reals scenes, a new model and its relative translation algorithm are presented in this chapter. Based on the potential field function, the new model can describe the distribution of defocus degree, so our method can defocus any objects or areas in the input image, and the degree of blurring can be controlled finely by simply adjusting the weight of potential function. Finally, a summary of our work is given in Chapter 5.
语种: 中文
内容类型: 学位论文
URI标识: http://ir.iscas.ac.cn/handle/311060/7012
Appears in Collections:中科院软件所

Files in This Item:
File Name/ File Size Content Type Version Access License
LW002842.pdf(1667KB)----限制开放-- 联系获取全文

Recommended Citation:
李洪举. 基于图象的建模和绘制研究[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
[李洪举]'s Articles
CSDL cross search
Similar articles in CSDL Cross Search
[李洪举]‘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