Multimedia Retrieval

Аннотация

Organization of the Book The book consists of thirteen chapters. After the Introduction we treat three languages to describe metadata (Dublin Core, RDF, MPEG-7). Chapter 3 gives an overview of important mathematical techniques in Pattern Recog-nition like Support Vector Machines, Hidden Markov Models, and Dynamic Bayesian Networks. The next chapter summarizes the state-of-the-art Infor-mation Retrieval techniques. Chapter 5 is a long chapter dealing with images and concentrating on image analysis. It gives a classification of objects occur-ring in images and explains detection algorithms. Chapter 6 offers a math-ematical approach to detect features by combining evidence from text and images. Chapter 7 is devoted to audio processing, in this case mainly speech recognition. The Chapters 8, 9, and 10 cover video applications. Evidence for high level features in video may come from more than one source. The chap-ters describe approaches to combine evidence. Chapter 11 gives a description of interactions with a system and the presentation of results. Digital rights management is the topic of Chapter 12. The final chapter handles the difficult topic of evaluation of multimedia systems. Besides theoretical considerations, many evaluation efforts are described ranging from important test sets to eval-uation procedures as pursued by, for instance, the Text Retrieval Conference.