Now showing 1 - 10 of 18
No Thumbnail Available
Publication

Large margin methods for structured and interdependent output variables

2005 , Tsochantaridis, I. , Joachims, T. , Hofmann, T. , Altun, Y.

Learning general functional dependencies between arbitrary input and output spaces is one of the key challenges in computational intelligence. While recent progress in machine learning has mainly focused on designing flexible and powerful input representations, this paper addresses the complementary issue of designing classification algorithms that can deal with more complex outputs, such as trees, sequences, or sets. More generally, we consider problems involving multiple dependent output variables, structured output spaces, and classification problems with class attributes. In order to accomplish this, we propose to appropriately generalize the well-known notion of a separation margin and derive a corresponding maximum-margin formulation. While this leads to a quadratic program with a potentially prohibitive, i.e. exponential, number of constraints, we present a cutting plane algorithm that solves the optimization problem in polynomial time for a large class of problems. The proposed method has important applications in areas such as computational biology, natural language processing, information retrieval/extraction, and optical character recognition. Experiments from various domains involving different types of output spaces emphasize the breadth and generality of our approach.

No Thumbnail Available
Publication

Visualization of 3D information with digital holography using laser printers

2001 , Ferri, L.C.

One of the main applications of digital holography is the 3D visualization of information. This paper describes the results obtained in the production of static computer generated holograms, using as output device a low/medium resolution laser printer. In particular, the reconstruction process of 2D/3D intensity objects from Fourier and Fresnel amplitude holograms and an application of the results in the field of optical document security are described.

No Thumbnail Available
Publication

Designing for Semantic Access: A Video Browsing System

2000 , Hollfelder, S. , Everts, A. , Thiel, U.

No Thumbnail Available
Publication

Everything you wanted to know about MPEG-7: Part I

1999 , Nack, F. , Lindsay, A.

No Thumbnail Available
Publication

Saying what it means: Semi-automated (news) media annotation

2004 , Nack, F. , Putz, W.

This paper considers the automated and semi-automated annotation of audiovisual media in a new type of production framework, A4SM (Authoring System for Syntactic, Semantic and Semiotic Modelling). We present the architecture of the framework, describe a prototypical camera, a handheld device for basic semantic annotation, and an editing suite to demonstrate how video material can be annotated in real time and how this information can not only be used for retrieval but also can be used during the different phases of the production process itself. We then outline the underlying XML Schema based content description structures of A4SM and discuss the pros and cons of our approach of evolving semantic networks as the basis for audio-visual content description.

No Thumbnail Available
Publication

Using cryptographic and watermarking algorithms

2001 , Dittmann, J. , Nahrstedt, K. , Wohlmacher, P.

Because of multimedia's structure and complexity, security mechanisms for multimedia data should be specific for each purpose. We introduce the most important security requirements for all types of multimedia systems. We also survey revocation methods for digital certificates and introduce a media-independent classification scheme.

No Thumbnail Available
Publication

Everything you wanted to know about MPEG-7: Part II

1999 , Nack, F. , Lindsay, A.

No Thumbnail Available
Publication

JASMINE: A Java tool for multimedia collaboration on the internet

2003 , Shirmohammadi, S. , Saddik, A. el , Georganas, N.D. , Steinmetz, R.

Although collaboration tools have existed for a long time [8], Internet-based multimedia collaboration has recently received a lot of attention mainly due to easy accessibility of the Internet by ordinary users. The Java platform and programming language has also introduced yet another level of easy access: platform-independent computing. As a result, it is very attractive to use Java to design multimedia collaboration systems for the Internet. Today there are many systems, which use Java for multimedia collaboration. However, most of these systems require the shared Java application to be re-written according to the collaboration system's Application Programming Interface (API)-a task which is sometimes difficult or even impossible. In this paper, we describe a practical approach for transparent collaboration with Java. Our approach is transparent in that the Java application can be shared as is with no modifications. The main idea behind our system is that user events occurring through the interactions with the application can be caught, distributed, and reconstructed, hence allowing Java applications to be shared transparently. Our architecture allows us to make the huge installed base of Java applications collaborative, without any modification to their original code. We also prove the feasibility of our architecture by implementation of the JASMINE 1 prototype.

No Thumbnail Available
Publication

Copyright - Copywrong

2000 , Dittmann, J. , Nack, F.

No Thumbnail Available
Publication

A testbed for mobile multimedia applications

1999 , Pommnitz, J. , Schönfeld, W.