Showing papers by "Arif Ghafoor published in 2000"

Models for motion-based video indexing and retrieval

Abstract: With the rapid proliferation of multimedia applications that require video data management, it is becoming more desirable to provide proper video data indexing techniques capable of representing the rich semantics in video data. In real-time applications, the need for efficient query processing is another reason for the use of such techniques. We present models that use the object motion information in order to characterize the events to allow subsequent retrieval. Algorithms for different spatiotemporal search cases in terms of spatial and temporal translation and scale invariance have been developed using various signal and image processing techniques. We have developed a prototype video search engine, PICTURESQUE (pictorial information and content transformation unified retrieval engine for spatiotemporal queries) to verify the proposed methods. Development of such technology will enable true multimedia search engines that will enable indexing and searching of the digital video data based on its true content.

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.

Media synchronization in multimedia Web using a neuro-fuzzy framework

Abstract: We consider the problem of multimedia synchronization in a Web environment. The workload generated by the multimedia server during a Web session exhibits variations that are quite different from the traffic fluctuation offered by a single media stream, e.g., a variable bit rate (VBR) video. We propose a set of parameters that can be used to characterize the workload generated by the multimedia server in a Web-type browsing environment. The workload characterization scheme is subsequently used in designing a server-based synchronization scheme. The problem of scheduling multimedia information to ensure media synchronization in a Web environment is identified as a multicriteria scheduling problem, which is NP-hard. The ability of fuzzy control to deal with multivariables makes it a good alternative for the multicriteria scheduling problem considered. Consequently, we propose a neuro-fuzzy scheduler (NFS) that makes an intelligent compromise among multicriteria by properly combining some scheduling heuristics. Performance of the NFS is compared with several known heuristics and a branch and bound algorithm. The results show that the proposed NFS ran dynamically adjust to the varying workload quite well.

Distributed framework for real-time multimedia object communication

Abstract: We highlight major technical requirements for designing and developing future distributed multimedia information systems using Internet technology. The key requirements of this system are to allow users to access and search and to communicate multimedia documents consisting of text, audio, video and images. We emphasize the role of object-oriented technology for information management and real-time communication protocols to guarantee QoS. We present a reference architecture for a Web-based real-time distributed multimedia system which integrates enabling technologies including real-time streaming, multimedia indexing and searching and distributed object management.

A Petri-net based multilevel security specification model for multimedia documents

Abstract: With the growing need for multimedia data management, security requirements are becoming very crucial. Composing multimedia documents involves bringing together media objects that exist in various formats. These objects may reside in a distributed environment and belong to different security domains. We propose a time-augmented colored Petri net model for multimedia document composition that allows the specification of multilevel security. The model also allows handling of multiple security policies and hierarchical and path based protection schemes.

Synchronized delivery of multimedia information over ATM networks

Abstract: Multimedia communications are expected to have a major impact on scientific and technological development in the near future. This area has spawned new research opportunities in diverse areas such as computer networks, storage systems, semiconductor technology, distributed systems, information theory, databases, real-time operating systems, and computer graphics. Many distributed applications involving information management require presentation of different media types—video, audio, text, images, animation and graphics—to the end user. These applications are emerging in diverse areas such as business, education, medicine, and entertainment. For example, various U.S. government organizations have built multimedia libraries that allow users to interactively author, store, and share multimedia documents (MMDs) [4]. Most of these applications/services use stored information in one form or the other. The structure of the information is generally preorchestrated at the time of storage [2, 3]. Conceptually, an MMD can be logically sub-divided into multimedia objects, which may be either isochronous such as video and audio, or discrete, such as text and images. These objects are temporally related at the time of creation of the MMD, in a process known as MMD preorchestration. An example of temporal relations is depicted in Figure 1a, where presentation starts with a concurrent playout of an NTSC-quality video along with a stereo-quality sound. This is immediately followed by the presentation of a high-resolution image along with an animation sequence that is later joined by the other media objects as depicted in the figure. MMDs can be stored on a single server or at different remote sites distributed over a network. In either case, presentation of an MMD requires real-time delivery and transmission of multiple media streams to the end user. The server needs to control the transmission of these streams in order to guarantee the playout of media objects according to the prespecified temporal structure and without heavily penalizing the resources at the client site. In other words, retrieval of media objects requires efficient and flexible end-to-end transport mechanisms. Integrated networks based on the Asynchronous Transfer Mode (ATM) are the ideal

Design of multimedia Web server using a neuro-fuzzy framework

Abstract: In this paper, we propose a neuro-fuzzy scheduler (NFS) for a multimedia Web server to ensures synchronized delivery of multimedia documents. The problem of scheduling multimedia information to ensure media synchronization in a Web environment is identified as a multicriteria scheduling problem which is NP-hard. The proposed NFS makes an intelligent compromise among multicriteria by properly combining some scheduling heuristics. Performance of the NFS is compared with several known heuristics and a branch and bound algorithm. The results show that the proposed neuro-fuzzy scheduler can dynamically adjust to the varying work-load quite well.

Performance evaluation of an interactive Web-based multimedia document system with streaming media

Abstract: A useful feature in Web-based real-time presentations is to provide interactive functionality to end users. We introduce a new such functionality to seek salient objects in a streaming video hyperlinked to multimedia documents. We propose a non-uniform time interval caching (NUTIC) technique for caching objects arbitrarily located in a streaming video at a user's site. An experimental performance assessment of the proposed NUTIC technique is presented.