Multimedia Database Management Research Articles

An MPEG-7 Compatible Video Retrieval System with Integrated Support for Complex Multimodal Queries

Early prototype multimedia database management systems used the query-by-example paradigm to respond to user queries. Users needed to formulate their queries by providing examples or sketches. The query-by-keyword paradigm, on the other hand, has emerged due to the desire to search multimedia content in terms of semantic concepts using keywords or sentences rather than low-level multimedia descriptors. After all, it's much easier to formulate some queries by using keywords. However, some queries are still easier to formulate by examples or sketches-for example, the trajectory of a moving object.

Enhancing the retrieval performance in content based image retrieval using meta-heuristic approach

Multimedia and digital image database management have undergone major transformation significantly in recent years in domains like data mining, medical imaging, weather forecasting and remote sensing, etc. CBIR is an important tool, useful to retrieve the required image precisely and effectively from large databases. This paper proposes the use of low level visual content features viz, mean value and colour histogram, to retrieve the colour feature and gray level co-occurrence matrix is proposed for retrieval of the texture feature. IGA is integrated with the above said low level features to derive refined results for query image matching from the database and to differentiate retrieval performance among the image features. The proposed method yields an accurate and faster retrieval from a Corel image database.

Social media in the classroom: examining the effects of social influence mechanism on peer learning

Social media is popular in college students' life. However, there are only a few studies exploring the impact of social media on student relationship, and none of them studied the prevalence of social media by dynamic network analysis. This study developed OpenSlide, a teaching system combining the social function of SlideShare and Youtube as an after-school auxiliary learning tool and analysed the impact of this tool on peer interaction in a classroom by using the advantage of multimedia database management. One of the most widely discussed issues in social network analysis is that both the interaction between individual and network formation comprises two processes: social selection and social influence. This study used the stochastic actor-based model to model the network dynamics of students interact recorded in the system and their face-to-face knowledge discussion networks. Our findings demonstrated that similar social media tools can enhance learning effectiveness, allow knowledge construction across the boundary of class and improve students' academic achievement while identifying successful strategies for social media development at an early stage.

Social media in the classroom: examining the effects of social influence mechanism on peer learning

Social media is popular in college students' life. However, there are only a few studies exploring the impact of social media on student relationship, and none of them studied the prevalence of social media by dynamic network analysis. This study developed OpenSlide, a teaching system combining the social function of SlideShare and Youtube as an after-school auxiliary learning tool and analysed the impact of this tool on peer interaction in a classroom by using the advantage of multimedia database management. One of the most widely discussed issues in social network analysis is that both the interaction between individual and network formation comprises two processes: social selection and social influence. This study used the stochastic actor-based model to model the network dynamics of students interact recorded in the system and their face-to-face knowledge discussion networks. Our findings demonstrated that similar social media tools can enhance learning effectiveness, allow knowledge construction across the boundary of class and improve students' academic achievement while identifying successful strategies for social media development at an early stage.

Application of Multimedia Technology in Database and IT Service Management

Multimedia is a blend of two or more media such as text, sound, graphics, animation and video, to effectually converse ideas to the users and this could be in linear or non-linear form [7]. The application of multimedia technology to database and IT service management has improved enormously in recent times. Multimedia system is a distinctive application which is of time-critical in its use in the various areas of computing. The spatial, secular, storage, processing, recovery, grouping and management requirements of data vary momentously from those that are applied for traditional data [2]. Therefore, the purpose of multimedia technology in database management system and in proper management of IT services to clients is to permit for an efficient way of performing these tasks in all its varied forms in an efficient form. A multimedia database management system affords an efficient storage and manoeuvring of multimedia technologies in all its diverse types. Database technology has offered means to store and recover high volumes of data in the various business domains [8]. Although, database systems have always been planned for the administration of alphanumeric data such as names and numbers, the basic nature of multimedia data are also considered and thus highlights the need for multimedia enhanced database management systems, and present the various obligations and issues required for developing such systems. The various areas of applications consist of, but not limited to digital libraries (text documents, images, sound, video etc), art and entertainment, content management, journalism etc [1]. For technology purposes, multimedia is a computer-based systems that apply associative relationships to allow the users of such systems to navigate and retrieve various information that are stored in a location which could be a combination of text, sounds, graphics, video, and other media formats [12].

A COMPUTATIONAL MODEL FOR CONTEXT-BASED IMAGE CATEGORIZATION AND DESCRIPTION

Automatic image categorization and description are key components for many applications, i.e., multimedia database management, web content analysis, human–computer interactions, and biometrics. In general, image description is a difficult task because of the wide variety of objects potentially to be recognized and the complexity and variety of backgrounds. This paper introduces a computational model for context-based image categorization and description. First, for a given image, a classifier is trained by the associated text features using advanced concepts, so that it can assign the image to a specific category. Then, a similarity matching with that category's annotated templates is performed for images in every other category. The proposed model uses novel text and image features that allow it to differentiate between geometrical images (GIs) and ordinary images. The experimental results show that the model is able to categorize correctly images with an expected increase in similarity matching as larger datasets and neural document classifier (NDC) are used. An important feature of the proposed model is that its specific matching techniques, suitable for a particular category, can be easily integrated and developed for other categories.

Bilvideo-7: an MPEG-7- compatible video indexing and retrieval system

Early prototype multimedia database management systems used the query-by-example paradigm to respond to user queries. Users needed to formulate their queries by providing examples or sketches. The query-by-keyword paradigm, on the other hand, has emerged due to the desire to search multimedia content in terms of semantic concepts using keywords or sentences rather than low-level multimedia descriptors. After all, it's much easier to formulate some queries by using keywords. However, some queries are still easier to formulate by examples or sketches-for example, the trajectory of a moving object.

Address generators for linear systolic array

Systolic arrays (SAs) are very efficient architectures for multimedia processing, database management, and scientific computing applications that are characterized by a high number of data access. However, in these data transfer and storage intensive applications, memory access is often the limiting factor to the computation speed. Since the memory subsystem dominates the cost (area), performance and power consumption of the SA, we have to pay a special attention to how memory subsystem can benefit from customization. In this paper we consider memory organization of linear systolic array with bi-directional links (called BLSA) suitable for implementation of broad class of algorithms. We assume that memory is organized into distributed smaller physical memory modules. In order to provide high bandwidth in data access we have designed special hardware, called address generator unit (AGU). The function of AGU is threefold. First, during the initialization, it transforms host address space into BLSA address space. Second, provides efficient memory data access during BLSA operation. Third, performs fast data transfer between BLSA and host at the end of the computation. In this article, we examine the impact on area and performance of memory access related circuity in eliminating computational intensive offset address calculations performed in software by implementing the needed address transformations with the AGUs. By involving hardware AGUs we achieved a speedup of approximately two, compared to the software implementation of address calculation, with a hardware overhead of only 7.6% in the worst case.

Using editing operations to improve searching by color in multimedia database systems

AbstractSince multimedia database management systems determine similarity by comparing sets of image features, relevant images in the database can be missed if their features do not match those extracted from the query image. Many failed matches can be avoided if modified versions of the missed relevant images are also stored in the underlying database. To minimize the storage cost associated with adding extra images to the database, the modified versions can be stored as sequences of editing operations instead of as large, binary objects. This article presents a technique for processing color‐based queries in this environment that accesses the sequences of editing operations directly. It also presents a methodology that can be used to speed up the query processing just as ordered indices speed up the processing of traditional queries. In addition, this article provides a performance illustrating the technique's strengths and weaknesses when compared with the traditional approach to processing color‐based queries. The results indicate that with low similarity thresholds, the proposed technique processes similarity searches more accurately than the traditional approach while using less database storage space since the modified versions are kept as editing operation sequences. © 2008 Wiley Periodicals, Inc. Int J Imaging Syst Technol, 18, 182–194, 2008

Security and privacy for multimedia database management systems

This paper describes security and privacy issues for multimedia database management systems. Multimedia data includes text, images, audio and video. It describes access control for multimedia database management systems and describes security policies and security architectures for such systems. Privacy problems that result from multimedia data mining are also discussed.

A Video Content Independent Mining Algorithm for Evolved Rule-based Detection of Scene Boundaries

Segmentation of video data stream aims to divide the stream into temporally shorter, meaningful and manageable segments. This is the first step towards content-based multimedia database management, contented-based retrieval and browsing, and is very important to many other applications that aim to work with content. This paper presents a novel video mining algorithm that uses genetic programming for evolved rule-based scene boundary detection and whose key advantage is that is video content independent. Hence, the algorithm can be applied without modification to different video sequences just by feeding it different training data.

The life cycle of multimedia metadata

During its lifetime, multimedia content undergoes different stages or cycles from production to consumption. Content is created, processed or modified in a postproduction stage, delivered to users, and finally, consumed. Metadata, or descriptive data about the multimedia content, pass through similar stages but with different time lines. Metadata may be produced, modified, and consumed by all actors involved in the content production-consumption chain. At each step of the chain, different kinds of metadata may be produced by highly different methods and of substantially different semantic value. Different metadata let us tie the different multimedia processes in a life cycle together. However, to employ these metadata, they must be appropriately generated. The CODAC Project, led by Harald Kosch, implements different multimedia processes and ties them together in the life cycle. CODAC uses distributed systems to implement multimedia processes. The project's core component is a multimedia database management system (MMDBMS) which stores content and MPEG-7-based metadata. It communicates with a streaming server for data delivery. The database is realized in the multimedia data cartridge (MDC) - which is an extension of the Oracle database management system - to handle multimedia content and MPEG-7 metadata.

Integrated use of different content derivation techniques within a multimedia database management system

As amounts of publicly available video data grow, the need to automatically infer semantics from raw video data becomes significant. In this paper, we address the use of three different techniques that support that task, namely, spatio-temporal rule-based method, hidden Markov models, and dynamic Bayesian networks. First, the application of these techniques for detection and recognition of diverse events is briefly described using two case studies (Tennis and Formula 1). We explain the relationships and differences of the three approaches, as well as benefits of their integrated use. Then the focus is moved to the main point of the paper, which is the integration of the aforementioned techniques within a database management system, which provides efficient, flexible, scalable, and domain independent content-based video retrieval. We identify and consider the most important issues when extending a traditional database management system with content-based video retrieval functionality, namely issues concerning video data models, dynamic feature extraction, and extensions of different layers of database architecture. The advantages of the integrated system are demonstrated on examples from our two case studies.

Workshop report

The motivation for organizing a workshop on “Multimedia Databases” at ACM International Conference on Information and Knowledge Management (ACM CIKM 2003) was the fact that more and more multimedia data are now available and effective management of these multimedia data is becoming a critical need. As a consequence, multimedia databases have emerged as an important research area. A multimedia database management system must provide the support for managing text, video, audio and image data, and it must also manage multimedia data types. In addition, the relationships between the objects and the representation of the spatial and/or temporal relationships may involve the extensions to the database management system.

An efficient query optimization strategy for spatio-temporal queries in video databases

The interest for multimedia database management systems has grown rapidly due to the need for the storage of huge volumes of multimedia data in computer systems. An important building block of a multimedia database system is the query processor, and a query optimizer embedded to the query processor is needed to answer user queries efficiently. Query optimization problem has been widely studied for conventional database systems however it is a new research area for multimedia database systems. Due to the differences in query processing strategies, query optimization techniques used in multimedia database systems are different from those used in traditional databases. In this paper, a query optimization strategy is proposed for processing spatio-temporal queries in video database systems. The proposed strategy includes reordering algorithms to be applied on query execution tree. The performance results obtained by testing the reordering algorithms on different query sets are also presented.

An efficient query optimization strategy for spatio-temporal queries in video databases

The interest for multimedia database management systems has grown rapidly due to the need for the storage of huge volumes of multimedia data in computer systems. An important building block of a multimedia database system is the query processor, and a query optimizer embedded to the query processor is needed to answer user queries efficiently. Query optimization problem has been widely studied for conventional database systems; however it is a new research area for multimedia database systems. Due to the differences in query processing strategies, query optimization techniques used in multimedia database systems are different from those used in traditional databases. In this paper, a query optimization strategy is proposed for processing spatio-temporal queries in video database systems. The proposed strategy includes reordering algorithms to be applied on query execution tree. The performance results obtained by testing the reordering algorithms on different query sets are also presented.

A multi-paradigm querying approach for a generic multimedia database management system

To truly meet the requirements of multimedia database (MMDB) management, an integrated framework for modeling, managing and retrieving various kinds of media data in a uniform way is necessary. MediaLand is an experimental MMDB platform being developed at Microsoft Research Asia for users with different levels of experiences and expertise to manage and search multimedia repositories easily, efficiently, and cooperatively. Key features of MediaLand include a uniform data model for describing all kinds of media objects and their relationships, and a 4-tier architecture based on this data model. In this paper, a multi-paradigm querying approach of MediaLand is presented, in which multimedia queries are processed based on a seamless integration of various existing search approaches. In doing so, MediaLand also offers the feature of "media independence" which is analogous to the notion of "data independence" from the classic ANSI SPARC standard. By incorporating a rich set of facilities and techniques, MediaLand lays down a good foundation for addressing further research issues, such as multimedia query rewriting, optimization, and presentation.

Image retrieval by examples

A currently important research field in information sciences is the management of nontraditional distributed multimedia databases. Two related key issues are to achieve an efficient content-based query by example retrieval and a fast response time. This paper presents the architecture of a distributed image retrieval system which provides novel solutions to these key issues. In particular, a method to quantify the effectiveness of low level visual descriptors in database query tasks is presented. The results are also used to improve the system response time, which is an important issue when querying very large databases. A new mechanism to adapt system query strategies to user behavior is also introduced in order to improve the effectiveness of relevance feedback and overall system response time. Finally, the issue of browsing multiple distributed databases is considered and a solution is proposed using multidimensional scaling techniques.

B-SPID: an object-relational database architecture to store, retrieve, and manipulate neuroimaging data.

We propose a hardware and software architecture to respond to crucial problems in the neuroimaging field: storage, retrieval, and processing of large datasets. The B-SPID project, here discussed, concerns the processing of neuroimages and attached components stored in an object-relational multimedia database management system (DBMS). Advanced bioinformation concepts are exploited in this project such as large scale data storage, high level graphical user interfaces and 3D graphical processing and display of data. Our database implementation is based on standard programming components, runs on several UNIX platforms and is written to be evolutive. Queries on this database are designed to obtain and display from neuroimaging data several types of results (pictures, text, or 3D graphical shapes) on heterogeneous systems.

Tree-Based Indexes for Image Data

As in conventional database management systems (DBMSs), to allow users to efficiently access and retrieve data objects, a multimedia database management system (MMDBMS) must employ an effective access method such as indexing and hashing. This paper provides a survey of tree-based multidimensional indexing techniques for MMDBMSs that maintain image data represented as feature vectors. These techniques support such data while maintaining desirable characteristics of a B-tree, an index structure most commonly used in traditional DBMSs. In this survey, we provide descriptions of each tree as well as give examples of the different data organization schemes. We also describe the advantages and disadvantages of using each technique. In addition, we provide classifications of the trees using several different properties. These classifications should assist researchers in identifying the strengths and weaknesses of any new indexing technique they develop as well as help users determine the most appropriate data structure for their applications.