Preface. Program Committee. Additional Reviewers. Keynote Papers. Hyperdatabases: Infrastructure for the Information Space H.-J. Schek, et al. Federated Information Systems for Communities D. Abel. Relevance Feedback in CBIR Hong Jian, Zhang, Z. Su. Invited Paper. An Overview of Oracle Chart Builder and Map Viewer L. Molesky, J. Sharma. Papers. SOM-Based K-Nearest Neighbors Search in Large Image Databases Z. Aghbari, et al. Partial Image Retrieval Using Color Regions and Spatial Relationships Y. Cao, et al. High-Dimensional Image Indexing Based on Adaptive Partitioning and Vector Approximation G.-H. Cha. The Segmentation and Classification of Story Boundaries in News Video L. Chaisorn, T.-S. Chua. Architecture and Implementation of an Interactive Tool for the Design and Visualization of Active Capability S. Chakravarthy, S. Yang. Indexing Images in High-Dimensional and Dynamic-Weighted Feature Spaces K. Goh, E. Chang. Modeling Multimedia Displays Using Action Based Temporal Logic G. Gonzalez, et al. An Entropy-based Approach to Visualizing Database Structure D. Groth, E.L. Robertson. Visual Query Processing for GIS with Web Contents R. Lee, et al. Spatiao-temporal Modelling and Querying Video Databases Using High-level Concepts S. Nepal, U. Srinivasan. Perfect Line Simplification for Visualization in Digital Cartography. S. Prasher. Balancing Fidelity and Performance in Virtual Walkthrough Y. Ruan, et al. Aural Interfaces to Databases Based on VoiceXML B. Signer, et al. Visual Querying In Geographic Information Systems V.G. Soares, A.C. Salgado. Automatic Annotation and Retrieval of Images Y. Song, et al. Visualization of Web-Based Australian Export Trading L. Soon, P. Chen. The Metric Histogram: A New and Efficient Approach for Content-based Image Retrieval A.J.M. Traina, et al. Designing Dynamic Web Pages in the WYSIWYG Interface D. Wolber, et al. Improving Image Retrieval with Semantic Classification Using Relevance Feedback H. Wu, et al. Oral-Query-by-Sketch: An XML-based Framework for Speech Access to Image Databases S.-y. Wu, W.-s. Chen. Querying Video Data by Spatio-Temporal Relationships of Moving Object Traces C. Yajima, et al. Multi-View Combined Visual Similarity Retrieval For Face Database Y. Gao, M.K.H. Leung. DVQ: A DTD-driven Visual Query Interface for XML Database Systems L. Zhang, et al. Modelling and Indexing Fuzzy Complex Shapes J. Zhang, et al.
One of the more exciting opportunities enabled by the Internet is sharing of data and software within communities of interest. The early waves of exploitation have provided faster and easier communication, through email and through Web publishing of research results, research materials (including data and software), news and general information. For individual researchers, this has enabled access to a richer set of resources and fostered cooperation within specialist groups. A strengthening trend is for individuals or institutions to share resources (both data and software) through Web services. To search for available information on a gene, for example, a biotechnology researcher can now access a remote genomic database, rather than incur the costs of maintaining her own local database. This paper envisages the establishment of collections of Web services, by a community of interest and for the community of interest. Technically, this form of Federated Information System poses the usual problems of size (the number of services present) and heterogeneity in the services. Although recent advances in the area give some confidence that large-scale, highly-heterogeneous networks of Web services are within reach, the social aspects of establishing shared resources require reconsideration of some fundamental elements of architectural design. This paper examines some informations systems research issues in design of federated information systems for large communities of interest.
In this paper, we investigate the combination of image semantic classification with content-based image retrieval. A flexible scheme is proposed to take advantage of image classification, which may be obtained manually or automatically, to enhance image retrieval. In this scheme, a semantic feature vector is composed for an image based on its class membership information, and is combined with low-level features in image retrieval. Relevance feedback techniques are also used to adjust both the semantic feature and low-level features of the query in order to better reflect the user’s intention. Experimental results on a collection of 10,000 images with manual classification demonstrate the effectiveness of the proposed method.
In this paper, we present a novel approach to retrieve images that contain the query image as a part regardless of the size and position the query image appears. Images in Database are segmented in advance, for each major region obtained, a composite measurement of color, area percentage and position are stored as the feature. While retrieving, the query image is also first segmented, and then the major regions’ colors, their area ratios and spatial relationships are generated for narrowing the searching space. The utilization of multiple regions alleviates the influence of inaccurate segmentation, the presegmentation of images in database allows indexing of features of color regions and enables the fast retrieving. The experiment shows the advantages and weakness of the proposed method.
The segmentation and classification of news video into single-story semantic units is a challenging problem. This research proposes a two-level, multi-modal framework to tackle this problem. The video is analyzed at the shot and story unit (or scene) levels using a variety of features and techniques. At the shot level, we employ a Decision Tree to classify the shot into one of 13 predefined categories. At the scene level, we perform the HMM (Hidden Markov Models) analysis to eliminate shot classification errors and to locate story boundaries. We test the performance of our system using two days of news video obtained from the MediaCorp of Singapore. Our initial results indicate that we could achieve a high accuracy of over 95 % for shot classification. The use of HMM analysis helps to improve the accuracy of the shot classification and achieve over 89% accuracy on story segmentation.
This article presents a framework that defines a visual query language for Geographical Information Systems (GIS). The main objective of the proposed framework is to improve the friendliness of a GIS interface, which is achieved by using pictorial information to build query clauses and by reducing the differences among GIS internal structures. The queries use visual elements and visual spatial operators, which are represented according to a Geographic Visual Query Language, the GeoVisualQL.
In this paper, we propose the LPC + -file for efficient indexing of high-dimensional image data. With the proliferation of multimedia data, there is an increasing need to support the indexing and retrieval of high-dimensional image data. Recently, we developed the LPC-file (Cha et al. 2002) for indexing high-dimensional data based on vector approximation. The LPC-file gives good performance especially when the dataset is uniformly distributed. However, compared with for the uniformly distributed dataset, its performance degrades when the dataset is clustered. Real image datasets are often strongly clustered and the dimensions of the feature vectors are usually correlated. We improve the performance of the LPC-file for the strongly clustered image data-set. The basic idea is to adaptively partition the data space to find subspaces with high-density clusters and to assign more bits to them to increase the discriminatory power of the vector approximation. An empirical evaluation shows that the LPC+-file results in significant performance improvements for real image datasets strongly clustered.
Although a variety of techniques have been developed for content-based image retrieval (CBIR), automatic image retrieval by semantics still remains a challenging problem. We propose a novel approach for semantics-based image annotation and retrieval. Our approach is based on the monotonic tree model. The branches of the monotonic tree of an image, termed as structural elements, are classified and clustered based on their low level features such as color, spatial location, coarseness, and shape. Each cluster corresponds to some semantic feature. The category keywords indicating the semantic features are automatically annotated to the images. Based on the semantic features extracted from images, high-level (semantics-based) querying and browsing of images can be achieved. We apply our scheme to analyze scenery features. Experiments show that semantic features, such as sky, building, trees, water wave, placid water, and ground, can be effectively retrieved and located in images.
E-commerce applications tend to be used by a non-homogenous user population, requiring special attention to the modelling of underlying business processes, in order to make the execution of an activity as conspicuous as possible. Current modelling languages do not provide for concepts and symbols to represent all communicative aspects of a business transaction. Speech act theory offers categories that may be used to supplement the concepts and notations of current modelling languages. We are developing a Visual Business Modelling Language (VBML), for which we propose various speech acts in addition to the common symbols in a modelling language. To explore and demonstrate the expressiveness and logic of VBML, we apply it to the web-based Australian export trade. In particular, we show how an export trader can make a cargo declaration through the web to facilitate a document exchange with the Australian Customs Service. This model benefits application designers, software developers, Australian Customs Service, and the export traders since it delivers a clear view of the trade and the corresponding software application processes.
Modelling and querying video content involves understanding the spatiotemporal relationships of objects1 represented in a video sequence. In this paper, we present a binary representation-based framework for modelling and querying video object relationships at a semantic level. The framework supports spatio-temporal queries using high-level concepts. We propose a set of semantic (qualitative) terms to describe the spatio-temporal relations between objects and then show how such qualitative terms can be calculated within the proposed framework using a fuzzy-logic-based approach. We also report the experimental results performed on a video clip from the MPEG-7 video content archive.
Oracle Chart Builder is a real-time Java charting API. It enables more effective communication and analysis of information for business graphics and performance applications. MapViewer is a programmable tool for rendering maps using spatial data managed by Oracle Spatial. MapViewer provides tools that hide the complexity of spatial data queries and cartographic rendering, while providing customizable options for more advanced users. This paper presents an overview of data visualization features in the 9i Application Server product from Oracle.
We present a metalanguage, named Alan that can be used to model dynamic multimedia displays, particularly those that display multimedia database query results. Alan is an action language that uses temporal logic to model non-Markovian systems. We show how it can be used for specifying the behavior of fairly complex dynamic multimedia display systems, modeling all graphical user interface elements on the display plus the effects of actions and of the passage of time on media such as video and audio.
In many geographic objects such as a travel planning, the use of web information is significantly increasing. For an efficient support of such work, it is very important to combine web information with map semantics. Current web systems usually do not support map semantics. Conversely, conventional Geographic Information Systems (GIS) do not utilize the web resources. The purpose of the research is as follows: (1) to get semantics from the web contents to realize advanced GIS functions on geographic web searches, and (2) to develop a user interface which can utilize web contents and map semantics in an effective integrating way. For such a purpose, we construct two map semantics about geographic characteristics and relationships available on the web. Utilizing semantics, we have developed a prototype system, KyotoSEARCH; its main function is to support users’ information navigations among the web, the map and web-based geographic knowledge, in an integrated way.
As part of a general framework for the development of global information systems, we include support for the development of aural interfaces. The framework uses an object-oriented database for the management of application, document content and presentation data. The access layer is based around an XML server and XSLT for document generation from default and customised templates. Specifically, aural interfaces are supported through a VoiceXML server that provides the speech recognition and synthesis mechanisms, together with XSLT templates for the generation of VoiceXML. In this paper, we describe the implementation of a generic voice browser for application databases as well as the development of a customised aural interface for a community diary managing appointments and events.
We propose an XML-based framework for speech access to multiple image databases without the need for keyboard or pen-based input. The key idea is to orally command the system in drawing an abstract sketch of the target images using simple graphical objects. We define an oral language for image description and query control named SpeechQuel. The abstract sketch is represented in W3C SVG format while all other query constraints are represented using a device and system independent XML language called SpeechQuelX. This facilitates content-based access to multiple and/or heterogeneous image databases with a single query. Preliminary implementation result successfully demonstrates the feasibility and efficiency of our approach.
The amount of stored information is exploding as a consequence of the immense progress in computer and communication technology during the last decades. However tools for accessing relevant information and processing globally distributed information in a convenient manner are under-developed. In short, the infrastructure for the "Information Space" needs much higher attention. In order to improve this situation, we envision the concept of a hyperdatabase that provides database functionality at a much higher level of abstraction, i.e., at the level of complete information components in an n-tier architecture. In analogy to traditional database systems that manage shared data and transactions, a hyperdatabase manages shared information components and transactional processes. It provides "higher-order data independence" by guaranteeing the immunity of applications not only against changes in the data storage and access structures but also against changes in the application components and services, e.g., with respect to the location, implementation, workload, and number of replica of components. In our vision, a hyperdatabase will be the key infrastructure for developing and managing the information space.
As information retrieval systems evolve to deal with multimedia content, we see the dimensions of content feature space increasing, and relevance feedback being employed to provide more accurate query results. In this paper, we propose using Tree-structured Vector Quantization (TSVQ) to index high-dimensional data for supporting efficient similarity searches and effective relevance feedback. To support efficient similarity searches, we first use TSVQ to cluster data and store each cluster in a sequential file. We then model a similarity search as a classification problem—similar objects are much more likely to be found in the clusters into which the query object is classified. When relevance feedback is considered, and thereby features are weighted differently, we show that our approach remains very effective. Our empirical study on both a 51K and a one-million-image dataset shows that tackling indexing as a classification problem and solving the problem with TSVQ is efficient, effective, and scalable with respect to both data dimensions and dataset size.
