Search (7 results, page 1 of 1)

0.02470427 = product of:
  0.07411281 = sum of:
    0.07411281 = sum of:
      0.044964164 = weight(_text_:indexing in 150) [ClassicSimilarity], result of:
        0.044964164 = score(doc=150,freq=10.0), product of:
          0.19018644 = queryWeight, product of:
            3.8278677 = idf(docFreq=2614, maxDocs=44218)
            0.049684696 = queryNorm
          0.23642151 = fieldWeight in 150, product of:
            3.1622777 = tf(freq=10.0), with freq of:
              10.0 = termFreq=10.0
            3.8278677 = idf(docFreq=2614, maxDocs=44218)
            0.01953125 = fieldNorm(doc=150)
      0.029148644 = weight(_text_:22 in 150) [ClassicSimilarity], result of:
        0.029148644 = score(doc=150,freq=6.0), product of:
          0.17398734 = queryWeight, product of:
            3.5018296 = idf(docFreq=3622, maxDocs=44218)
            0.049684696 = queryNorm
          0.16753313 = fieldWeight in 150, product of:
            2.4494898 = tf(freq=6.0), with freq of:
              6.0 = termFreq=6.0
            3.5018296 = idf(docFreq=3622, maxDocs=44218)
            0.01953125 = fieldNorm(doc=150)
  0.33333334 = coord(1/3)

Classification

006.7 22

Date

7. 3.2007 19:30:22

DDC

006.7 22

Footnote

Rez. in: JASIST 58(2007) no.3, S.457-458 (A.M.A. Ahmad): "The concept of the semantic web has emerged because search engines and text-based searching are no longer adequate, as these approaches involve an extensive information retrieval process. The deployed searching and retrieving descriptors arc naturally subjective and their deployment is often restricted to the specific application domain for which the descriptors were configured. The new era of information technology imposes different kinds of requirements and challenges. Automatic extracted audiovisual features are required, as these features are more objective, domain-independent, and more native to audiovisual content. This book is a useful guide for researchers, experts, students, and practitioners; it is a very valuable reference and can lead them through their exploration and research in multimedia content and the semantic web. The book is well organized, and introduces the concept of the semantic web and multimedia content analysis to the reader through a logical sequence from standards and hypotheses through system examples, presenting relevant tools and methods. But in some chapters readers will need a good technical background to understand some of the details. Readers may attain sufficient knowledge here to start projects or research related to the book's theme; recent results and articles related to the active research area of integrating multimedia with semantic web technologies are included. This book includes full descriptions of approaches to specific problem domains such as content search, indexing, and retrieval. This book will be very useful to researchers in the multimedia content analysis field who wish to explore the benefits of emerging semantic web technologies in applying multimedia content approaches. The first part of the book covers the definition of the two basic terms multimedia content and semantic web. The Moving Picture Experts Group standards MPEG7 and MPEG21 are quoted extensively. In addition, the means of multimedia content description are elaborated upon and schematically drawn. This extensive description is introduced by authors who are actively involved in those standards and have been participating in the work of the International Organization for Standardization (ISO)/MPEG for many years. On the other hand, this results in bias against the ad hoc or nonstandard tools for multimedia description in favor of the standard approaches. This is a general book for multimedia content; more emphasis on the general multimedia description and extraction could be provided.
Semantic web technologies are explained, and ontology representation is emphasized. There is an excellent summary of the fundamental theory behind applying a knowledge-engineering approach to vision problems. This summary represents the concept of the semantic web and multimedia content analysis. A definition of the fuzzy knowledge representation that can be used for realization in multimedia content applications has been provided, with a comprehensive analysis. The second part of the book introduces the multimedia content analysis approaches and applications. In addition, some examples of methods applicable to multimedia content analysis are presented. Multimedia content analysis is a very diverse field and concerns many other research fields at the same time; this creates strong diversity issues, as everything from low-level features (e.g., colors, DCT coefficients, motion vectors, etc.) up to the very high and semantic level (e.g., Object, Events, Tracks, etc.) are involved. The second part includes topics on structure identification (e.g., shot detection for video sequences), and object-based video indexing. These conventional analysis methods are supplemented by results on semantic multimedia analysis, including three detailed chapters on the development and use of knowledge models for automatic multimedia analysis. Starting from object-based indexing and continuing with machine learning, these three chapters are very logically organized. Because of the diversity of this research field, including several chapters of recent research results is not sufficient to cover the state of the art of multimedia. The editors of the book should write an introductory chapter about multimedia content analysis approaches, basic problems, and technical issues and challenges, and try to survey the state of the art of the field and thus introduce the field to the reader.
The final part of the book discusses research in multimedia content management systems and the semantic web, and presents examples and applications for semantic multimedia analysis in search and retrieval systems. These chapters describe example systems in which current projects have been implemented, and include extensive results and real demonstrations. For example, real case scenarios such as ECommerce medical applications and Web services have been introduced. Topics in natural language, speech and image processing techniques and their application for multimedia indexing, and content-based retrieval have been elaborated upon with extensive examples and deployment methods. The editors of the book themselves provide the readers with a chapter about their latest research results on knowledge-based multimedia content indexing and retrieval. Some interesting applications for multimedia content and the semantic web are introduced. Applications that have taken advantage of the metadata provided by MPEG7 in order to realize advance-access services for multimedia content have been provided. The applications discussed in the third part of the book provide useful guidance to researchers and practitioners properly planning to implement semantic multimedia analysis techniques in new research and development projects in both academia and industry. A fourth part should be added to this book: performance measurements for integrated approaches of multimedia analysis and the semantic web. Performance of the semantic approach is a very sophisticated issue and requires extensive elaboration and effort. Measuring the semantic search is an ongoing research area; several chapters concerning performance measurement and analysis would be required to adequately cover this area and introduce it to readers."

0.014988055 = product of:
  0.044964164 = sum of:
    0.044964164 = product of:
      0.08992833 = sum of:
        0.08992833 = weight(_text_:indexing in 2522) [ClassicSimilarity], result of:
          0.08992833 = score(doc=2522,freq=10.0), product of:
            0.19018644 = queryWeight, product of:
              3.8278677 = idf(docFreq=2614, maxDocs=44218)
              0.049684696 = queryNorm
            0.47284302 = fieldWeight in 2522, product of:
              3.1622777 = tf(freq=10.0), with freq of:
                10.0 = termFreq=10.0
              3.8278677 = idf(docFreq=2614, maxDocs=44218)
              0.0390625 = fieldNorm(doc=2522)
      0.5 = coord(1/2)
  0.33333334 = coord(1/3)

Abstract

A model for analysing multimedia information is proposed from the point of view of the theory of communication. After a brief presentation of the complex map of the sciences that deal with multimedia communication in its different aspects, the current multimedia revolution is historically contextualized as a tendency towards messages that are able to build near-reality experiences (virtual reality). After setting the theoretical point of view, an analysis of multimedia messages is substantiated and a model is presented. The first part of the model deals with the different communications channels and tools: still images, movies, sounds, texts, text with illustrations, audiovisuals and interactive multimedia, with an emphasis in nontextual documents. The second part addresses the global properties of the multimedia message, which are of a textual and metatextual nature. The overlapping of media, channels, genres and messages-and the conscious and technical use of such interactions-is precisely one of the main and outstanding characteristics of the multimedia discourse, and requires specific moves in indexing languages development. The multimedia environment has also a great potential to promote a wider theory of knowledge organization, bringing closer distant fields like scientific and fictional indexing or verbal and image indexing. It is stated that such a unified theory requires a closer attention to the pragmatic aspects of indexing and the inclusion of new semantic layers. A simple indexing model is proposed to illustrate who to address these challenges.

0.008975455 = product of:
  0.026926363 = sum of:
    0.026926363 = product of:
      0.053852726 = sum of:
        0.053852726 = weight(_text_:22 in 5627) [ClassicSimilarity], result of:
          0.053852726 = score(doc=5627,freq=2.0), product of:
            0.17398734 = queryWeight, product of:
              3.5018296 = idf(docFreq=3622, maxDocs=44218)
              0.049684696 = queryNorm
            0.30952093 = fieldWeight in 5627, product of:
              1.4142135 = tf(freq=2.0), with freq of:
                2.0 = termFreq=2.0
              3.5018296 = idf(docFreq=3622, maxDocs=44218)
              0.0625 = fieldNorm(doc=5627)
      0.5 = coord(1/2)
  0.33333334 = coord(1/3)

Date

14. 2.2001 19:09:22

0.0080434345 = product of:
  0.024130303 = sum of:
    0.024130303 = product of:
      0.048260607 = sum of:
        0.048260607 = weight(_text_:indexing in 4454) [ClassicSimilarity], result of:
          0.048260607 = score(doc=4454,freq=2.0), product of:
            0.19018644 = queryWeight, product of:
              3.8278677 = idf(docFreq=2614, maxDocs=44218)
              0.049684696 = queryNorm
            0.2537542 = fieldWeight in 4454, product of:
              1.4142135 = tf(freq=2.0), with freq of:
                2.0 = termFreq=2.0
              3.8278677 = idf(docFreq=2614, maxDocs=44218)
              0.046875 = fieldNorm(doc=4454)
      0.5 = coord(1/2)
  0.33333334 = coord(1/3)

Abstract

Multimedia retrieval is a complex and to some extent still unexplored area. Based on a full year of e-mail requests addressed to a large film archive this study analyses what types of information needs real users have and how these needs are expressed. The findings include that the requesters make use of a broad range of need attributes in specifying their information needs. These attributes relate to the production, content, subject, context and screening of films. However, a few attributes - especially title, production year and director - account for the majority of the attribute instances. Further, as much as 43 per cent of the requests contain no information about the context that gives rise to the request. The current indexing of the archived material is restricted to production-related attributes, and access to the material is, thus, frequently dependent on the archivists' extensive knowledge of the archived material and films in general.

0.0067028617 = product of:
  0.020108584 = sum of:
    0.020108584 = product of:
      0.04021717 = sum of:
        0.04021717 = weight(_text_:indexing in 478) [ClassicSimilarity], result of:
          0.04021717 = score(doc=478,freq=2.0), product of:
            0.19018644 = queryWeight, product of:
              3.8278677 = idf(docFreq=2614, maxDocs=44218)
              0.049684696 = queryNorm
            0.21146181 = fieldWeight in 478, product of:
              1.4142135 = tf(freq=2.0), with freq of:
                2.0 = termFreq=2.0
              3.8278677 = idf(docFreq=2614, maxDocs=44218)
              0.0390625 = fieldNorm(doc=478)
      0.5 = coord(1/2)
  0.33333334 = coord(1/3)

Abstract

This article discusses the implementation of MPEG-7 within the Moving Image Collections (MIC) portal. MIC is a union catalog of the world's moving images, as well as a portal to information on the care, management, and use of moving images. The MIC Union Catalog utilizes a core registry schema that is designed to map readily to any metadata schema used to describe moving images. The MIC development team was particularly interested in supporting MPEG-7 for future nontextual digital video indexing applications. An MPEG-7 application profile and Microsoft Access cataloging utility were developed in order to test MPEG-7 within the MIC Union Catalog; 400 science digital videos in the ResearchChannel collection were cataloged in MPEG-7. The MPEG-7 records were mapped to MIC and ingested. Draft MPEG-7 to MIC and MIC to MPEG-7 maps were developed and are available at the MIC Web site. MPEG-7 records are available for viewing for any record in the MIC database via a collections explore search within the Archivists' portal. The MPEG-7 cataloging utility may be downloaded from the MIC project Web site (Moving Image Collections. MIC Cataloging Utility. http://gondolin.rutgers.edu/MIC/text/ how/cataloging_utility.htm). This article also discusses issues with MPEG-7 as a descriptive metadata schema, as well as mapping and implementation issues identified in the project.

0.00536229 = product of:
  0.016086869 = sum of:
    0.016086869 = product of:
      0.032173738 = sum of:
        0.032173738 = weight(_text_:indexing in 500) [ClassicSimilarity], result of:
          0.032173738 = score(doc=500,freq=2.0), product of:
            0.19018644 = queryWeight, product of:
              3.8278677 = idf(docFreq=2614, maxDocs=44218)
              0.049684696 = queryNorm
            0.16916946 = fieldWeight in 500, product of:
              1.4142135 = tf(freq=2.0), with freq of:
                2.0 = termFreq=2.0
              3.8278677 = idf(docFreq=2614, maxDocs=44218)
              0.03125 = fieldNorm(doc=500)
      0.5 = coord(1/2)
  0.33333334 = coord(1/3)

Abstract

Dow Jones & Company (NYSE:DJ) hat am 13. Juni 2007 seine Partnerschaft mit PodZinger bekannt gegeben. PodZinger ist derzeit die einzige Multimedia-Plattform, die es ermöglicht, Audio- und Videoinhalte über Schlagworte zu suchen. Diese Suchfunktion wird über die Nachrichten- und Wirtschaftsinformationsdienste Factiva.com und Factiva iWorks zur Verfügung stehen. So können Anwender ab August 2007 Video- und Audio-Informationen abrufen, darunter Wirtschaftsnachrichten, CEO-Interviews, Vorträge von führenden Managern, Jahreshauptversammlungen, Produktrezensionen und andere einschlägige Inhalte. Laut eMarketer wurden allein im Januar 2007 über 7,2 Milliarden Videostreams von mehr als 123 Millionen Konsumenten betrachtet. AccuStream iMedia erwartet ein jährliches Wachstum an Multimedia-Nutzern von 30 Prozent. Trotzdem ist es noch immer sehr zeitaufwändig, relevante Inhalte zu finden, weil die meisten Multimedia-Seiten nur begrenzt Metadaten anbieten und nur rudimentäre Suchmöglichkeiten für Audio- und Videoinhalte zur Verfügung stellen. Oft resultieren aus der Suche dann große Informationsmengen, die für den Suchenden kaum relevant sind. Das Multimedia-Angebot von Factiva basiert auf den Speech-to-Text und Natural Language Technologien von PodZinger. Genutzt wird bei Factiva Multimedia auch das patentierte Factiva Intelligent Indexing(TM), und Suchergebnisse werden in Form von Charts und Grafiken dargestellt. Der Anwender erhält damit einen Überblick über den Kontext seiner Recherche. Gemäß einer Studie, die PodZinger 2006 in Auftrag gegeben hatte, hören oder schauen sich Menschen nur rund 15 Prozent einer Audio- oder Videodatei an. Dow Jones kombiniert die Stärken von Factiva und PodZinger zu einer einzigartigen Multimedia-Suchmaschine. Relevante Inhalte können mit Hilfe von spezifischen Suchbegriffen herausgefiltert werden. Anschließend können die Anwender selbst entscheiden, ob sie den gesamten Clip oder nur den für ihn relevanten Teil sehen oder hören wollen. Das Multimedia Angebot von Factiva umfasst ergänzend: Kontinuierlich aktualisierte Inhalte aus mehr als 4.000 Nachrichten- und Wirtschaftsquellen mit mehr als 300.000 Clips. Ein 90-Tage-Archiv aus einer Sammlung von Nachrichtenquellen wie dem Wall Street Journal, NPR, CNN, BBC Radio und mehr. Oberfläche in Deutsch, Englisch, Spanisch, Französisch und Russisch

0.0044877273 = product of:
  0.013463181 = sum of:
    0.013463181 = product of:
      0.026926363 = sum of:
        0.026926363 = weight(_text_:22 in 622) [ClassicSimilarity], result of:
          0.026926363 = score(doc=622,freq=2.0), product of:
            0.17398734 = queryWeight, product of:
              3.5018296 = idf(docFreq=3622, maxDocs=44218)
              0.049684696 = queryNorm
            0.15476047 = fieldWeight in 622, product of:
              1.4142135 = tf(freq=2.0), with freq of:
                2.0 = termFreq=2.0
              3.5018296 = idf(docFreq=3622, maxDocs=44218)
              0.03125 = fieldNorm(doc=622)
      0.5 = coord(1/2)
  0.33333334 = coord(1/3)

Date

12. 8.2012 18:05:22