CASIA OpenIR  > 毕业生  > 硕士学位论文
Alternative TitleResearch of Multimedia Adaptation, Browsing and Management in Mobile Environment
Thesis Advisor卢汉清
Degree Grantor中国科学院研究生院
Place of Conferral中国科学院自动化研究所
Degree Discipline模式识别与智能系统
Keyword移动计算 网络多媒体 自适应内容传输 内容供应商 注意力模型 网页结构分析 网页分割 运动视频分析 运动精华片断提取 互联网挖掘 Mobile Computing Network Multimedia Multimedia Adaptation Content Providers Attention Model Webpage Structure Analysis Page Segmentation Sports Video Analysis Sports Highlight Extraction Web Mining
Abstract在当今信息爆炸时代,全世界有大量的数字化内容(Digital Content)以多媒体(例如文本,网页,图像,音频,视频,等等)的形式在生成、存储、传播和转换,并呈现出爆炸性高速增长趋势。互联网的高速发展深刻推动了多媒体服务急剧膨胀。随着互联网持续地向多元化及异构化方向发展,网络客户端环境(包含客户端软硬件设备、网络联接特性和用户个人兴趣偏好等)的多样性与多媒体内容服务的单一性之矛盾愈显突出。例如,通过桌面电脑用户可以方便浏览一副大图片,大多数情况下不需要手工操作,但是如果用户通过一个小屏幕的手持设备(例如手机, 口袋电脑、或者智能手机)来浏览这副图片,大量的手工操作(例如滚动, 放大等)会使情况会变得很糟。 为更加适应移动客户需求,自适应多媒体服务能力的功能应运而生。针对自适应内容传送,需要研究许多新的技术以自动理解多媒体对象。本论文对图像,网页和视频的内容自适应浏览和应用进行了研究,实现了可行的的多媒体内容自适应系统。此外,由于电子图片在因特网上的大量普及,我们提出了网络图片语义信息挖掘的方案,这些图片语义信息可以为移动用户提供更佳的网络图片搜索服务。本论文基于提高多媒体信息内容的可获取性和服务质量,完成了下述一系列研究工作:  介绍了多媒体内容自适应系统的基本概念,对国内外的相关研究现状进行了系统的概述,说明了多媒体内容自适应研究的重要性和必要性。  分别针对客户端能力提出了多媒体内容自适应的有效解决方案,使得图像、网页、运动视频等各种多媒体内容能够在各种手持设备的小屏幕上自适应地显示与浏览。  针对移动设备无线带宽有限和屏幕面积小的特性,我们提出了新的动态网页自适应方案。我们的实验结果表明这种新方案可以全面提高用户的网页浏览和减少无线带宽消耗。  针对运动视频资源在因特网的广泛普及,我们提出一个完整的方案来提高移动用户对运动视频的访问和浏览。我们的方案包括在内容供应端提供运动精华片断提取方案,并在移动客户端设计有效的用户界面来对实现对运动视频精华片段的方便交互浏览。  应用互联网的挖掘技术(Web Mining)来提取网络图片的高级语义信息,例如地点、人物、时间、及描述关键字等。把这些语义信息应用到图片搜索引擎,可以方便的为移动用户提高网络图片的搜索服务质量。
Other AbstractIn this information explosion era, there are a large amount of digital multimedia content (e.g. text, webpage, picture, audio, video, etc) generated, stored, transferred and transformed, which are still having a explosively growing trend. The fast-growing development of Internet has profoundly promoted the explosive increasing of various multimedia services. With the continuous heterogeneity of Internet developments, different characteristics of network clients (including client device hardware, network accessibility and personalized preferences, etc) imposes a more and more salient contradiction with the identity of multimedia services. For example, through desktop computers users can easily browse a large picture without scrolling or zooming operations, however, the thing becomes more difficult when users browse such a large picture on the small screen of a mobile device, e.g. cell-phone, pocket pc, or smart phone. To meet the demands of mobile clients, the self-adaptive multimedia functions have been proposed. To provide adaptive content delivery, it is necessary to conduct research to analyze and understand multimedia content. This paper has studied several kinds of prevalent multimedia content, including picture, webpage, and video. We implement and present practical and feasible multimedia content adaptation systems. Additionally, based on the extensive pervasiveness of digital images on the Web, we proposed a framework for mining semantic knowledge for Web images, aiming to improve Web image search quality. This paper is targeted to improve multimedia content accessibility and service quality, studying a series of research work as follows: • Introduce multimedia content adaptation concepts, systemically summarize related works in current stage, and demonstrate the importance and essence of this research work. • Propose effective and efficient solutions to multimedia content adaptation according to client capacities, aiming to improve mobile media display and browsing on small screens of handheld devices, such as image, webpage and sports video media. • Propose webpage adaptation solution regarding the limited wireless bandwidth and small display screens on mobile devices. Our experimental results demonstrate that this new solution can comprehensively improve mobile users' web browsing and reduce wireless network consumption. • Propose bran-new solutions to improve sports video access and browsing. Our solutions include sports highlights extraction service on content providers, and effective user interfaces on mobile clients to improve sports video interaction and browsing for users. • Develop web mining technologies to extract high-level semantic knowledge for Web images, such as location, people, time and keywords. Integrating mobile users' contextual information, we apply such semantic knowledge to Web image retrieval, aiming to improve web image search quality for the mobile users.
Other Identifier200228014603548
Document Type学位论文
Recommended Citation
GB/T 7714
华志刚. 移动环境下的多媒体自适应、浏览与管理研究[D]. 中国科学院自动化研究所. 中国科学院研究生院,2006.
Files in This Item:
File Name/Size DocType Version Access License
CASIA_20022801460354(5637KB) 暂不开放CC BY-NC-SAApplication Full Text
Related Services
Recommend this item
Usage statistics
Export to Endnote
Google Scholar
Similar articles in Google Scholar
[华志刚]'s Articles
Baidu academic
Similar articles in Baidu academic
[华志刚]'s Articles
Bing Scholar
Similar articles in Bing Scholar
[华志刚]'s Articles
Terms of Use
No data!
Social Bookmark/Share
All comments (0)
No comment.

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