协同多媒体著作工具中演示文档的同步模型

多媒体演示文档的协同著作对文档本身的结构与同步模型提出了一定的要求。为了便于面向对象的实现以及灵活多粒度的共享，提出了一个协同多媒体著作工具中分布交互式多媒体文档的同步模型。在此模型中，文档结构分为三层：页面层、对象组层及单媒体对象层，根据各层的特点，其同步分别采用基于跳转、基于事件及基于时间的策略     

基于MPEG-4的多媒体同步模型设计

文中旨在研究基于MPEG4技术的面向对象多媒体同步。所采用的方法是在分析现有多媒体同步模型的基础上，结合MPEG-4系统同步机制和高级同步模型（FlexTime），设计面向对象的多媒体同步模型，以实现MPEG-4中各媒体对象间同步。由此方法设计的层结构同步模型是一种有效可行的多媒体间同步的实现方法。和其它的多媒体间同步方法相比，这种方法在基于对象的多媒体的传输过程中携带较少的时间信息来完成对象间的同步，并且增强了基于对象内容交互的灵活性。     

互动多媒体同步机制设计

研究了基于 MPEG-4的面向对象多媒体同步技术。在分析现有多媒体同步技术的基础上，结合 MPEG-4 系统同步机制和高级同步模型(FlexTime)，设计了层结构同步模型，以实现 MPEG-4 中各媒体对象间同步，是一种有效可行的多媒体间同步的实现方法。     

基于MPEG-4的多媒体同步模型设计

文中旨在研究基于MPEG-4技术的面向对象多媒体同步。所采用的方法是在分析现有多媒体同步模型的基础上,结合MPEG-4系统同步机制和高级同步模型(FlexTime),设计面向对象的多媒体同步模型,以实现MPEG-4中各媒体对象间同步。由此方法设计的层结构同步模型是一种有效可行的多媒体间同步的实现方法。和其它的多媒体间同步方法相比,这种方法在基于对象的多媒体的传输过程中携带较少的时间信息来完成对象间的同步,并且增强了基于对象内容交互的灵活性。     

一种新的多媒体文档存储结构

根据信息系统中多媒体数据的特点，提出了一种能效地组织、存储多媒体数据的文档存储结构。该结构在对象链接与嵌入（OLE）及二进制大对象BLOBs技术的支持下，集成各种类型的多媒体数据，并采用宏文档网络分布结构，构造一种以文档为中心的多媒体数据的组织、存储方式，可以有效地提高信息系统中多媒体数据的存储及处理效率。     

计算机支持的协同设计研究

文中讨论了建立计算机支持的协同设计支撑平台的关键技术。在体系结构上提出一个三层的结构，其中ＣＳＣＷ支持层设计为公共的协作代理。对基于共享对象协作的协同设计建立了超媒体结构的信息模型。按照这种思想建立的原型系统能够支持在地理上分布的多个用户，通过异步和同步的协作方式完成协同的设计工作，在异步协作方式下，可支持多用户通过对共享对象的操作完成协作；在同步协作方式下，可支持多用户通过对共享对象的操作完成协     

一种新的多媒体方案档存储结构

根据信息系统中多媒体数据的特点，提出了一种能有效地组织，存储多媒体数据的文档存储结构，该结构在对象链接与嵌入（ＯＬＥ）及二进制大对象ＢＬＯＢｓ技术的支持下，集成各种类型的多媒体数据，并采用宏文档网络分布结构，构造一种以文档为中心的多媒体数据的组织，存储方式，可以有效地提高信息系统中多媒体数据的存储及处理效率。     

面向多媒体文档的多通道检索系统

提出和实现了一个面向多媒体文档的多通道(对应多种模态,如文本、图像、视频等)检索系统.系统定义了一个新的用来描述多媒体文档内容的框架,该框架不但提取出多媒体文档在各通道下的基于内容的底层特征,而且还记录下多媒体文档中不同多媒体对象间的链接关系.同时,提出一种基于图模型的交叉参照知识库,用来存储从链接关系中挖掘出的多媒体对象间的语义关系,通过一个有效的语义上下文分析算法,在检索过程中计算每个对象与查询的语义相似度.语义上下文分析算法不仅使得基于内容的多媒体信息检索中考虑了多媒体对象的语义信息,同时支持用户通过通道切换的方式进行相关反馈,提供了一种较为灵活的查询模式.实验表明在交叉参照知识库基础上,该系统还能够有效地提高基于内容的多媒体对象的检索性能(如内容覆盖率等).     

基于语义的多代理协同数据库平台

提出了协同数据库平台CDP的体系结构、模型和实现。该平台可支持协同活动和管理多数据源。讨论了包括应用、支持和数据库的3层多Agent体系结构，基于语义的Agent模型以及语义联系的实现，从而屏蔽了异步和同步工作方式。协同用户通过应用语义来进行协同操作和信息共享。     

远程多媒体交互系统中会话记录和文档共享的实现研究

本文分析了基于Internet的远程多媒体交互系统中对于会话记录和文档共享两方面功能的需求．给出了基于SMIL和Active文档技术的解决方案。结合实践对使用SMIL的媒体同步模型进行了深入的探讨．指出了其中的一些问题并提出了改进算法。     

多媒体对象的组织与结构化检索

多媒体的结构化检索具有广泛的应用前景，但多媒体的检索技术尚不成熟。多媒体数据模型的复杂性，连续媒体基于内容检索的低效，以及缺乏适用的查询语言，都使得多媒体检索困难重重。英国肯特大学新近研制的多媒体检索系统在多媒体检索的相关领域取得了一定的突破。这个系统把用户视图中的多媒体对象组织成具有层次结构的虚拟数据库，使用属性来标识数据库中多媒体对象索引的特征；查询代理机允许用户直观地构造查询过程－包括一个结     

Modeling, authoring and formatting hypermedia documents in the HyperProp system

This paper discusses multimedia and hypermedia modeling, authoring and formatting tools, presenting the proposals of the HyperProp system and comparing them to related work. It also highlights several research challenges that still need to be addressed. Moreover, it stresses the importance of document logical structuring and considers the use of compositions in order to represent context relations, synchronization relations, derivation relations and task relations in hypermedia systems. It discusses temporal and spatial synchronization among multimedia objects and briefly presents the HyperProp graphical authoring and formatting tools. Integration between the proposed system and the WWW is also addressed.     

Content-based management of video in a multimedia authoring environment

This paper describes architecture and implementation of a multimedia authoring environment which features content-based management of video.The goal of our work is to provide multimedia application developers with a useful video database, where they gather video materials from various sources, store them in a organized format, and use them in multimedia applications. The principal design concepts are that (1) it should have a flexible means to store video materials in a organized format so that video clips can be retrieved easily, and that (2) it should have a view to each application so that various applications can share the video materials.To meet these requirements we designed a multimedia authoring system based on a video object model. The video object model features (1) a method to attach annotations not only to scenes but also to things shot in the video, and (2) a method to define video clips independently from physical video data.Our experiences in using the system to develop a multimedia CAI application are also described to show the advantages of our system.     

A Scalable Toolkit for Designing Multimedia Authoring Environments

This paper introduces Kaomi, a scalable toolkit for designing authoring environments of multimedia documents. The underlying concept is to provide the designer of multimedia applications with a fast method to get an authoring system based on a set of synchronized views (the presentation view for displaying the document, the scenario view for showing the temporal organization of the document, ...) so that each view is the support of editing actions. Kaomi is flexible enough to support a variety of multimedia documents declarative formats. It is indeed a scalable toolkit since it provides facilities for extending and/or for modifying the resulting authoring environment. In addition, cross-platform portability is provided which allows operation in the heterogenous Internet environment. The use of Kaomi is mainly described through the design of two authoring environments: one for authoring a sub-set of the Smil standard of the W3C and the other one for Madeus, a constraint based multimedia language.     

Interactive hypermedia applications: A model and its implementation

By referring to the object-oriented efforts of the Multimedia and Hypermedia information coding Expert Group (MHEG), and the augmented Petri net model, we present a formal definition of the multimedia hypermedia Petri net (MHPN) model. Through this proposed model, issues of authoring, rendering and synchronization of multimedia information (i.e. structure authoring, spatial temporal relationships, user interaction, etc.) can all be laid to rest. We have implemented a testbed system, the Petri net object information system (POIS), which offers considerable potential for the development of distributed hypermedia applications. The testbed has been implemented on an IBM-PC486 under Windows 3.1. We are now in the process of porting the testbed model to Windows NT.     

Authoring of scalable multimedia documents

With the diversity of usage conditions affecting the consumption of multimedia content, the adaptation of dynamic and interactive multimedia presentations is essential. The challenge consists in allowing a multimedia presentation to adapt in all its dimensions: spatially, temporally and interactively, without lessening its attractiveness and still giving the author the control over adapted versions. Additionally, the authoring of adaptable content should not increase the complexity of authoring. To address this challenge, we propose to transpose the concept of scalability to the world of multimedia documents by introducing the so-called Scalable MSTI model. In this paper, we show the properties of this model and how, from an authoring point of view, scalable multimedia documents can be created to address a wide range of usage conditions.     

A hypermedia authoring tool for augmented and virtual reality

Most existing hypermedia authoring systems are intended for use on desktop computers. These systems are typically designed for the creation of 2D documents and therefore employ 2D authoring mechanisms. In contrast, authoring systems for nontraditional multimedia/hypermedia experiences for 3D virtual or augmented worlds focus mainly on creating separate media objects and embedding them within the user's surroundings. As a result, linking these media objects to create 3D hypermedia is a tedious manual task. To address this issue, we present an authoring tool for creating and editing linked 3D hypermedia narratives that are interwoven with a wearable computer user's surrounding environment. Our system is designed for use by authors who are not programmers, and allows them to preview their results on a desktop workstation, as well as with an augmented or virtual reality system.     

Synchronization in Specification-based Multimedia Presentations

Several information protocols exist for exchange of multimedia information. The {H}yper{T}ext {T}ransfer {P}rotocol (HTTP) of NCSA combines hypertext techniques with multimedia to provide an intuitive, easy-to-use hypermedia interface. However, HTTP does not provide the means to synchronize different multimedia channels. The NCSA hypermedia browser, Mosaic, has been modified to incorporate synchronization constraints. Platform-independent multimedia authoring programs have been combined in a new synchronization scheme. The {H}yper{T}ext {M}arkup {L}anguage (HTML) has been extended to allow the specification of temporal dependencies between different media channels. This extension allows synchronization in the authoring of hypermedia presentations using Mosaic.     

Zodiac: A history-based interactive video authoring system

Easy-to-use audio/video authoring tools play a crucial role in moving multimedia software from research curiosity to mainstream applications. However, research in multimedia authoring systems has rarely been documented in the literature. This paper describes the design and implementation of an interactive video authoring system called Zodiac, which employs an innovative edit history abstraction to support several unique editing features not found in existing commercial and research video editing systems. Zodiac provides users a conceptually clean and semantically powerful branching history model of edit operations to organize the authoring process, and to navigate among versions of authored documents. In addition, by analyzing the edit history, Zodiac is able to reliably detect a composed video stream's shot and scene boundaries, which facilitates interactive video browsing. Zodiac also features a video object annotation capability that allows users to associate annotations to moving objects in a video sequence. The annotations themselves could be text, image, audio, or video. Zodiac is built on top of MMFS, a file system specifically designed for interactive multimedia development environments, and implements an internal buffer manager that supports transparent lossless compression/decompression. Shot/scene detection, video object annotation, and buffer management all exploit the edit history information for performance optimization.