Showing papers on "Multimedia database published in 2000"

MOQL: A Multimedia Object Query Language

Abstract: We describe a general multimedia query language, called MOQL, based on ODMG's Object Query Language (OQL). In contrast to previous multimedia query languages that are either designed for one particular medium (e.g. images) or specialized for a particular application (e.g., medical imaging), MOQL is general in its treatment of multiple media and di erent applications. The language includes constructs to capture the temporal and spatial relationships in multimedia data as well as functions for query presentation. We illustrate the language features by query examples. The language is implemented for a multimedia database built on top of ObjectStore.

High performance clustering based on the similarity join

Abstract: A broad class of algorithms for knowledge discovery in databases (KDD) relies heavily on similarity queries, i.e. range queries or nearest neighbor queries, in multidimensional feature spaces. Many KDD algorithms perform a similarity query for each point stored in the database. This approach causes serious performance degenerations if the considered data set does not fit into main memory. Usual cache strategies such as LRU fail because the locality of KDD algorithms is typically not high enough. In this paper, we propose to replace repeated similarity queries by the similarity join, a database primitive prevalent in multimedia database systems. We present a schema to transform query intensive KDD algorithms into a representation using the similarity join as a basic operation without affecting the correctness of the result of the considered algorithm. In order to perform a comprehensive experimental evaluation of our approach, we apply the proposed transformation to the clustering algorithm DBSCAN and to the hierarchical cluster structure analysis method OPTICS. Our technique allows the application of any similarity join algorithm, which may be based on index structures or not. In our experiments, we use a similarity join algorithm based on a variant of the X-tree. The experiments yield substantial performance improvements of our technique over the original algorithms. The traditional techniques are outperformed by factors of up to 33 for the X-tree and 54 for the R*-tree.

Adaptive multi-stage distance join processing

Abstract: A spatial distance join is a relatively new type of operation introduced for spatial and multimedia database applications. Additional requirements for ranking and stopping cardinality are often combined with the spatial distance join in on-line query processing or internet search environments. These requirements pose new challenges as well as opportunities for more efficient processing of spatial distance join queries. In this paper, we first present an efficient k-distance join algorithm that uses spatial indexes such as R-trees. Bi-directional node expansion and plane-sweeping techniques are used for fast pruning of distant pairs, and the plane-sweeping is further optimized by novel strategies for selecting a sweeping axis and direction. Furthermore, we propose adaptive multi-stage algorithms for k-distance join and incremental distance join operations. Our performance study shows that the proposed adaptive multi-stage algorithms outperform previous work by up to an order of magnitude for both k-distance join and incremental distance join queries, under various operational conditions.

Similarity-based ranking and query processing in multimedia databases

Abstract: Since media-based evaluation yields similarity values, results to a multimedia database query, Q(Y1,…,Yn), is defined as an ordered list SQ of n-tuples of the form 〈X1,…,Xn〉. The query Q itself is composed of a set of fuzzy and crisp predicates, constants, variables, and conjunction, disjunction, and negation operators. Since many multimedia applications require partial matches, SQ includes results which do not satisfy all predicates. Due to the ranking and partial match requirements, traditional query processing techniques do not apply to multimedia databases. In this paper, we first focus on the problem of “given a multimedia query which consists of multiple fuzzy and crisp predicates, providing the user with a meaningful final ranking”. More specifically, we study the problem of merging similarity values in queries with multiple fuzzy predicates. We describe the essential multimedia retrieval semantics, compare these with the known approaches, and propose a semantics which captures the requirements of multimedia retrieval problem. We then build on these results in answering the related problem of “given a multimedia query which consists of multiple fuzzy and crisp predicates, finding an efficient way to process the query.” We develop an algorithm to efficiently process queries with unordered fuzzy predicates (sub-queries). Although this algorithm can work with different fuzzy semantics, it benefits from the statistical properties of the semantics proposed in this paper. We also present experimental results for evaluating the proposed algorithm in terms of quality of results and search space reduction.

Efficient Query Refinement in Multimedia Databases

Abstract: ness of query refinement, there exists no work on how to implement query refinement efficiently in a multimedia database system. We explore such approaches in this paper. The proposed approaches are independent of the refinement model used (e.g., QPM or QEX) and hence work for all models. We assume that each feature is indexed using a multidimensional index structure (called the F-index). A similarity query can then be answered by executing a k-NN query on each F-index and merging the individual feature results to obtain the final results. Our first contribution is to generalize the notion of similarity queries and allow multiple query points in a query (referred to as multipoint queries). This generalization is necessary since refined queries cannot be always expressed as single point queries. We develop a k-NN algorithm that can handle multipoint queries and show that it performs significantly better than the naive approach (i.e. execute several single point queries using the ‘single-point’ kNN algorithm and merge results). The second and the main problem we address is how to evaluate refined queries efficiently. A naive approach is to treat a refined query just like a starting query and execute it from scratch. We observe that the refined queries are not modified drastically from one iteration to another. As a result, most of the execution cost can be saved by appropriately exploiting the information generated during the previous iterations of the query. We propose 3 techniques, namely Reuse (RU), Full Reconstruction (FR) and Selective Reconstruction (SR), that cache certain information during the execution of the previous iterations of the query and use that cached information to save the execution cost (both I/O and CPU costs) during the subsequent iterations. We define notions of I/O optimality and CPU optimality and evaluate the three schemes in terms of these criteria. We find that although RU is significantly better than the naive approach, it is not I/O optimal. So we propose FR and show that it is I/O optimal and is significantly more efficient compared to RU. However, FR is not CPU optimal. We finally propose SR and show that SR, like FR, is I/O optimal, and, under certain conditions, also CPU optimal. Even though SR is not always CPU optimal, it is always better than FR in terms of CPU cost and is hence the best technique. Our experiments show that the above techniques speed up the execution of refined queries by several orders of magnitude compared to the naive technique.

Semantic Models for Multimedia Database Searching and Browsing

Abstract: Semantic Models for Multimedia Database Searching and Browsing begins with the introduction of multimedia information applications, the need for the development of the multimedia database management systems (MDBMSs), and the important issues and challenges of multimedia systems. The temporal relations, the spatial relations, the spatio-temporal relations, and several semantic models for multimedia information systems are also introduced. In addition, this book discusses recent advances in multimedia database searching and multimedia database browsing. More specifically, issues such as image/video segmentation, motion detection, object tracking, object recognition, knowledge-based event modeling, content-based retrieval, and key frame selections are presented for the first time in a single book. Two case studies consisting of two semantic models are included in the book to illustrate how to use semantic models to design multimedia information systems. Semantic Models for Multimedia Database Searching and Browsing is an excellent reference and can be used in advanced level courses for researchers, scientists, industry professionals, software engineers, students, and general readers who are interested in the issues, challenges, and ideas underlying the current practice of multimedia presentation, multimedia database searching, and multimedia browsing in multimedia information systems.

The MLPQ/GIS constraint database system

Abstract: @ for Datalog queries. For example, in Fig. 1 we loaded and displayed a constraint database that represents the midwest United States and loaded two contraint relations describing the movements of two persons. The query icon opened a dialog box into which we entered the query which finds (t, i) pairs such that the two people are in the same state i at the same time t.MLPQ/GIS can animate [2] spatio-temporal objects that are linear constraint relations over and t.Users can also display in discrete color zones (isometric maps) any spatially distributed variable z that is a linear function and For example, Fig. 2 shows the mean annual air temperature Nebraska. Animation and isometric map display can be combined.

Method and apparatus for retrieving multimedia data using shape information

Abstract: A method and apparatus for retrieving multimedia data using shape information. The multimedia data retrieval apparatus comprises a shape information feature extractor for receiving shape information of a query image and extracting a feature of the received shape information using a shape information descriptor based on eigen vectors of a multilayer covariance matrix, an image feature extractor for extracting a feature of each image data in the same manner as the shape information feature extractor, a multimedia database creator for creating a multimedia database on the basis of the features extracted by the image feature extractor, a similarity comparator for comparing the feature of the query image extracted by the shape information feature extractor with the features of the image data in the multimedia database to calculate similarities therebetween, and a compared result output unit for outputting the results calculated by the similarity comparator. Multimedia data can accurately and rapidly be retrieved by defining and using a shape information descriptor with a consistency against rotation, scaling and translation of the shape information.

Imprecision and User Preferences in Multimedia Queries: A Generic Algebraic Approach

Abstract: Specification and efficient processing of similarity queries on multimedia databases have recently attracted several research efforts, even if most of them have considered specific aspects, such as indexing, of this new exciting scenario. In this paper we try to remedy this by presenting an integrated algebraic framework which allows many relevant aspects of similarity query processing to be dealt with. As a starting point, we assume the more general case where "imprecision" is already present at the data level, typically because of the ambiguous nature of multimedia objects' content. We then define a generic similarity algebra, SAMEW, where semantics of operators is deliberately left unspecified in order to better adapt to specific scenarios. A basic feature of SAMEW is that it allows user preferences, expressed in the form of weights, to be specified so as to alter the default behavior of most operators. Finally, we discuss some issues related to "approximation" and to "user evaluation" of query results.

MIMS: a prototype for medical image retrieval

Abstract: With the advent of multimedia information, several problems have bounded traditional database approaches designed strictly for alphanumeric data. One of the many induced problems is the retrieving process. All current proposed systems approach the retrieval problem using one aspect of the concerned media (image, voice and video). Our prototype, called Medical Image Management System (MIMS), is based on a synergy of several approaches and conceived to yield an effective system appropriated to retrieve images in medical domain. The main goals of our system are to acquire significant information associated to medical imaging and answer complex medical queries. This paper presents MIMS prototype and show how the end-user is able to process easily both complex medical image retrieval and storage.

An audio signatures indexing scheme for dynamic content multimedia databases

Abstract: This paper describes a method of content indexing for dynamic multimedia databases using binary signatures of the audio signals. The analysis scheme focuses on the audio objects of the multimedia environment, extracts semi-stationary audio segments and analyzes their key features, These are: the fundamental frequency F0, the first three formants F1, F2, and F3, the associated phoneme and speaker identification information. Binary signatures are associated with the audio segments which they describe. For each audio segment these key features are encoded within the signature and information retrieval can be performed on the multimedia database querying the corresponding signatures of the streamed audio. The signature structure serves as an index of the associated audio objects assisting their retrieval.

Generative user models for adaptive information retrieval

Abstract: For information retrieval (IR) tasks, user models are used to estimate user's true intention and demand. Unfortunately, most user models are constructed in a specialized form that is not applied to other systems or domains. This specialization makes it difficult to share user models as common resources for developing information retrieval systems and for researching cognitive characteristics in various users. In order to solve this problem, we need a general user modeling method. A user model based on a probabilistic framework is proposed. We call this model a generative user model. The generative user model represents user's mental depth by latent (hidden) variables. It also has visible variables that mean word set and qualifier of each word as a subjective probability distribution. The model can handle uncertainty of the user's subjectivity by a probabilistic framework. Recent statistical studies for such latent models give a learning algorithm. Our generative user model can be constructed from a dataset taken by information retrieval tasks. As an example, we also introduce two different kinds of information retrieval systems, ART MUSEUM (Multimedia Database with Sense of Color and Construction upon the Matter of ART) and DSIU (Decision Support for Internet Users). The generative user model is applied to these systems. The properties of the model and interactive learning mechanism are shown.

CoMed: a real-time collaborative medicine system

Abstract: CoMed is a World Wide Web-based real-time collaborative medical system that allows medical specialists to share patient records and to communicate with each other on the Internet. CoMed consists of a multimedia object database containing relevant information about laryngeal diseases and a real-time multimedia collaboration system. The latter includes a teleconferencing system, a whiteboard and chatting system, and a multi-agent system for conference calling. We adopted the object database O/sub 2/ and CORBA technologies for the multimedia database. This provides the flexibility, extensibility and location transparency of the databases. We implemented SeeYou ActiveX control for the teleconferencing and a Java applet for the whiteboard and chatting system, as well as a session manager which integrates these two systems that are based on different technologies. Moreover, our multi-agent system for conference calling makes our system more advantageous by allowing the convener to minimize the effort needed to call a conference.

MRML: A Communication Protocol for Content-Based Image Retrieval

Abstract: In this paper we introduce and describe the Multimedia Retrieval Markup Language (MRML). This XML-based markup language is the basis for an open communication protocol for content-based image retrieval systems (CBIRSs). MRML was initially designed as a means of separating CBIR engines from their user interfaces. It is, however, also extensible as the basis for standardised performance evaluation procedures.Such a tool is essential for the formulation and implementation of common benchmarks for CBIR. A common protocol can also bring new dynamics to the CBIR field -- it makes the development of new systems faster and more efficient, and opens the door of the CBIR research field to other disciplines such as Human-Computer Interaction. The MRML specifications, as well as the first MRML-compliant applications, are freely available and are introduced in this paper.

A probabilistic-based mechanism for video database management systems

Abstract: As more information sources become available in multimedia systems, the development of multimedia database management systems (MDBMSs) to efficiently model and search multimedia data, especially video data, becomes very crucial for multimedia applications. In response to such a demand, a probabilistic-based mechanism called the Markov Model Mediator (MMM) to facilitate an MDBMS for video database systems is presented. In our previous studies, the spatial relations of the semantic objects in the image/video data modeled by the MMM mechanism are assumed given by image processing/computer vision techniques or by human annotations. In this paper, an unsupervised video segmentation method that can identify objects with their corresponding spatial relations automatically is incorporated into the MMM mechanism. Based on the information obtained, users can retrieve video materials from video databases via database queries. Hence, both multimedia data modeling and image processing capabilities are integrated into the MMM mechanism.

Electronic books in digital libraries

Abstract: An electronic book is an application with a multimedia database of instructional resources, which include hyperlinked text, instructor's audio/video clips, slides, animation, still images, etc. As well as content-based information about these data, and metadata such as annotations, tags, and cross-referencing information. Electronic books in the Internet or on CDs today are not easy to learn from. We propose the use of a multimedia database of instructional resources in constructing and delivering multimedia lessons about topics in an electronic book. We introduce an electronic book data model containing (a) topic objects and (b) instructional resources, called instruction module objects, which are multimedia presentations possibly capturing real-life lectures of instructors. We use the notion of topic prerequisites for topics at different detail levels, to allow electronic book users to request/compose multimedia lessons about topics in the electronic book. We present automated construction of the "best" user-tailored lesson (as a multimedia presentation).

Information filtering and personalisation in databases using Gaussian curves

Abstract: Presents an information filtering and adaptive personalisation algorithm for arbitrary information systems based on databases. This algorithm is called GRAS (Gaussian Rating Adaptation Scheme), and it combines content-based and collaborative filtering. The goal is to filter the retrieved documents of a query according to the personal interests of a user and to sort them according to their personal relevance. The algorithm tries to make the benefits of collaborative filtering available to application domains where collaborative filtering could not yet be applied due to lack of the critical mass of users or improper content structure. The algorithm collects background information about the user and the content by implicit and explicit feedback techniques. This information is then used to consecutively adapt user and object profiles according their maturity. The described algorithm is applicable to the personalisation of any kind of application domain, even on multimedia data. GRAS is implemented in the multimedia database MultiMAP as a generic personalisation provider module.

Multimedia databases

Abstract: A multimedia database is a controlled collection of multimedia data items such as text, images, graphic objects, video and audio. A multimedia database management system (DBMS) provides support for the creation, storage, access, querying and control of a multimedia database. The requirements of a multimedia DBMS are: multimedia data modeling; multimedia object storage; multimedia indexing, retrieval and browsing; and multimedia query support. This paper discusses a general framework for multimedia database systems and describes the requirements and architecture for these systems.

A New High-Dimensional Index Structure Using a Cell-Based Filtering Technique

Abstract: In general, multimedia database applications require to support similarity search for content-based retrieval on multimedia data, i.e., image, animation, video, and audio. Since the similarity of two multimedia objects is measured as the distance between their feature vectors, the similarity search corresponds to a search for the nearest neighbors in the feature vector space. In this paper, we propose a new high-dimensional indexing scheme using a cellbased filtering technique which supports the nearest neighbor search efficiently. Our Cell-Based Filtering (CBF) scheme divides a high-dimensional feature vector space into cells, like VA-file. However, in order to make a better effect on filtering, our CBF scheme performs additional filtering based on a distance between an object feature vector and the center of a cell including it, in addition to filtering based on cell signatures before accessing a data file. From our experiment using high-dimensional feature vectors, we show that our CBF scheme achieves better performance on the nearest neighbor search than its competitors, such as VA-File and X-tree.

Distributed multimedia database: a design and application study

Abstract: In this paper, we present the concept, design, and applications of a distributed multimedia database. A standard three-tier client-server architecture is proposed, including (i) database server (ii) middleware and underlying network infrastructure, and (iii) client module. We provide our evaluation of the latest technique developments related to different parts of the configuration, and discuss various options for distributed computing in a heterogeneous environment. We also present a case study/application of concept and configuration discussed in the paper, and review our implementation of an object-relational multimedia database using Informix Universal Server (IUS), CORBA/Java, and a client interface developed by our team.

Fuzzy Signatures for Multimedia Databases

Abstract: In this paper we suggest a new method of information retrieval based on signature files which uses fuzzy sets and fuzzy logic for signature construction and manipulation. We provide the definition of fuzzy signatures and we discuss their possible applications.

Effective access to large audiovisual assets based on user preferences

Abstract: Current multimedia databases contain a wealth of information in the form of audiovisual, as well as text data. Even though efficient search algorithms have been developed for either media, there still exists the need for abstract presentation and summarization of the results of database users' queries. Moreover, multimedia retrieval systems should be capable of providing the user with additional information related to the specific subject of the query, as well as suggest other topics which users with a similar profile are interested in. In this paper, we present a number of solutions to these issues, giving as an example an integrated architecture we have developed, along with notions that support efficient and secure Internet access and easy addition of new material. Segmentation of the video in shots is followed by shot classification in a number of predetermined categories. Generation of users' profiles according to the same categories, enhanced by relevance feedback, permits an efficient presentation of the retrieved video shots or characteristic frames in terms of the user interest in them. Moreover, this clustering scheme assists the notion of “lateral” links that enable the user to continue retrieval with data of similar nature or content to those already returned. Furthermore, user groups are formed and modeled by registering actual preferences and practices; this enables the system to “predict” information that is possibly relevant to specific users and present it along with the returned results. The concepts utilized in this system can be smoothly integrated in MPEG-7 compatible multimedia database systems.

CLIMS - A System for Image Retrieval by Using Colour and Wavelet Features

Abstract: In this paper a system called CLIMS (CLausthal Image Management System) for content based image retrieval as an important sub-system of a general multimedia database is presented. It offers querying by sketch and image example and uses colour and wavelet based features for the comparison of images. Each image in the database is represented by a set of wavelet coefficients and colour attributes, which form the fundament for the retrieval. In order to enable efficient similarity search two index structures, VP-Trees and Lq metric, are introduced and discussed. With the extension of the original VP-tree algorithm a ranking of the n most similar images is possible. The efficiency of the proposed retrieval methods is evaluated on a sample, general image catalogue.

Volume Holographic Optical Correlators

Abstract: Volume holograms have received considerable recent interest for high-capacity data storage [1–3] because of the ability to multiplex many holograms within the same volume. Volume holographic memories may also provide an attractive solution to the massive capacity and transfer rate requirements of multimedia applications [4]. The page-oriented data format of holographic memories is a natural match for the format of certain database objects such as tables of records, images, and video frames. The massive parallelism provides both high capacity and high data rates, which can alleviate the I/O bottleneck of multimedia databases [5].

Extending databases to support image editing

Abstract: In order to understand similarity between images, recent research has focused on adaptable searches (Seidl and Kreigel, 1997) and fuzzy queries (Fagin, 1998). However, one of the best means for determining similarity between images is to know how the image was created (Brown et al., 1999). If the image is a combination of other images in the database, then there is a great deal of similarity between the base images and the created one. This requires extending the database to support image editing operations. We have built a preliminary system which does this by using a Web-based image editor and an image server. The editor and the server understand a logical model language that represents images. This paper then explores the issues of performance for deriving images from a sequence of operations.

A Federated Multimedia Database System.

Abstract: Today, a lot of work is dedicated to the development of systems that store, retrieve, or transport continuous-media data. Much less, however, is investigated to make these systems (i.e. media servers) interoperable, so that they can be used in cooperation with other systems in a single application. In particular, the cooperation of media servers and traditional database management systems (DBMS) would be useful for several reasons: (a) traditional DBMS are in widespread use already and serve very efficiently in the management of structured data, (b) media servers are in use already, too, and they provide efficient support for media data management, and (c) there is a trend to use media data as an add-on to existing information systems that are based on traditional DBMS. An integration of media-data management and structured-data management in one system seems to be a very tedious task and may in fact never work out, since the two types of data are very different and require special treatment, including internal aspect such as buffer management, transaction management, access paths, and so on. Hence, a federation of the two types of data managers seems to be a better way. It is the idea of the thesis, which is presented shortly in this paper, to investigate the various aspects and implications of such a specific kind of federation.

Navigation-dependent visualization of distributed Internet structures

Abstract: One of the main aspects of the Internet is that it is a distributed, multimedia database offering a fast growing amount of valuable information. Searching this database can be done by posing questions to search machines like Alta Vista. On the other hand, it is possible to browse through the space of connected information objects. When browsing, the problem arises that the user usually has no special knowledge of the structure of the visited hypertext area he is looking at. It may happen that he loses orientation and gets 'lost in hyperspace'. This article presents a software architecture that provides a visitor to a limited, distributed Internet area with additional structure information depending on his individual navigation path. Knowledge about the local hypermedia area is made available by a server program and is passed on to a visualization tool working in a frame of the client's Internet browser. This tool is loaded automatically from the server as a Java applet. By communication with the server, the visualization tool remains active throughout the navigation process and can be updated automatically.

A self organizing map (SOM) extended model for information discovery in a digital library context

Abstract: This paper presents the MicroNOMAD Discovering Tool. Its main characteristic is both to provide an user with emergent analyses of a multimedia database content and with querying and browsing guidelines through the use of an advanced topographic interface model. The model also allows the user to dynamically exploit semantic exchanges between multiple viewpoints on the database. The tool basic principles are firstly described. A tool experimentation which is achieved on the multimedia data-base associated to the BIBAN "Art Nouveau" server is then developed. It clearly demonstrates that the combination of both the topographic structures, the textual and iconographic interaction, and the viewpoint exchanges proposed by the MicroNOMAD core model could play an essential role in several discovering and browsing processes.

A parallel algorithm for multi-feature content-based multimedia retrieval

Abstract: In this paper, we propose a parallel, unified model for the indexing and retrieval by content of multimedia data using a weighted cascade. [4] Each multimedia data type can be viewed as k-dimensional (k-d) data in spatiotemporal domain. [3] Each dimension of the data is separated into small blocks and then formed into a multidimensional tree structure, called a k-tree. Using the ktree structure, the retrieval time improves while the retrieval accuracy remains relatively constant. Moreover, since we can realize all types of multimedia data using the same ktree data structure, the data indexing and retrieval algorithms are uniform. In this paper, we demonstrate a parallel algorithm with a weighted cascade that can be used with the uniform model. The parallel processing and weighted cascade algorithm improve the retrieval time. We evaluated the performance results of multimedia database queries using a Beowulf-class cluster of workstations. [1] The experimental results indicate that the retrieval accuracy improves with the utilized tree depth, while the parallel processing minimizes the retrieval time.