Abstract
An efficient algorithm for dynamically multiplexing MPEG2 encoded video sources is presented. Sources are grouped into classes regarding different combined levels of spatial detail and amount of movement. Simulations were performed using different associations of sources belonging to distinct classes, different bit rates and GOP structures. The implications associated to a real implementation are analysed and a modular architecture is proposed. Simulation results are presented and discussed, showing that sequences with higher spatial detail and motion are those which exhibit the higher quality improvements.

Abstract
In this paper we present an experimental approach for QoS-aware and bandwidth-efficient transmission of multimedia sources over ATM networks. VBR video sources are statistically multiplexed at the application level and the UPC parameters of a single connection for the aggregate traffic are dynamically renegotiated with the network during the session lifetime. A statistical multiplexer computes the required bandwidth for the aggregate traffic and dynamically assigns bandwidth according to the performances requested and network resources availability. A dynamic UPC manager using feedback information sent by the network, may initiate a renegotiation of traffic parameters upon receiving a request from the statistical multiplexer. Performing the connection admission control procedures for the aggregate traffic, allows significant reductions in the value of the total peak rate when compared to the sum of peak rates of individual connections. It also reduces the burstiness of the flow submitted to the network, thus increasing the lifetime of each set of UPC parameters. © Springer-Verlag Berlin Heidelberg 2000.

Abstract
Very few of the current solutions for content recommendation take into consideration the context of usage when analyzing the preferences of the user and issuing recommendations. Nonetheless, context can be extremely useful to help identify appropriate content for the specific situation or activity the user is in, while consuming the content. In this paper, we present a solution to allow content-based recommendation systems to take full potential of contextual data, by defining a standards-based representation model which accounts for possible relationships among low-level contexts. The MPEG-7 and MPEG-21 standards are used for content description and low-level context representation. OWL/RDF ontologies are used to capture contextual concepts and, together with SWRL to establish relationships and perform reasoning to derive high-level concepts the way humans do. This knowledge is then used to drive the recommendation and content adaptation processes. As a side achievement, an extension to the MPEG-21 specification was developed to accommodate the description of user activities, which we believe have a great impact on the type of content to be recommended.

Abstract
The MPEG-21 standard forms a comprehensive multimedia framework covering the entire multimedia distribution chain. In particular, it provides a flexible approach to represent, process, and transact complex multimedia objects which are referred to as Digital Items (DIs). DIs can be quite generic, independent of the application domain, and can encompass a diversity of media resources and metadata. This flexibility has an impact on the level of interoperability between systems and applications, since not all the functionality needs to be implemented. Furthermore, additional semantic rules may be implemented through the processing of the Digital Item which is possibly driven by proprietary metadata. This jeopardizes interoperability and consequently raises barriers to the successful achievement of augmented and transparent use of multimedia resources. In this context, we have investigated and evaluated the interoperability at the semantic level of Digital Items throughout the automated production, delivery and consumption of complex multimedia resources in heterogeneous environments. This paper describes the studies conducted, the experiments performed, and the conclusions reached towards that goal. © Copyright 2008 ACM.

Abstract
Context-aware applications are fast becoming popular as a means of enriching users' experiences in various multimedia content access and delivery scenarios. Nevertheless, the definition, identification, and representation of contextual information are still open issues that need to be addressed. In this paper, we briefly present our work developed within the VISNET II Network of Excellence (NoE) project on context-based content adaptation in Virtual Collaboration Systems (VCSs). Based on the conducted research, we conclude that MPEG-21 Digital Item Adaptation (DIA) is the most complete standardization initiative to represent context for content adaptation. However, tools defined in MPEG-21 DIA Usage Environment Descriptors (UEDs) are not adequate for Virtual Collaboration application scenarios, and thus, we propose potential extensions to the available UEDs.

Abstract
The use of context is already a reality in different application fields. One such example is the Robotics field, where sensors are used to capture information about the surrounding environment, allowing the system to react to changes. But also in the multimedia cornmunications field, some work has recently emerged to allow combining sensed context and use that combined information to decide how to adapt the content. This paper describes the MULTICAO ontology, which lies at the core of an Adaptation Decision Engine, providing it with a formal description of the surrounding environment in a similar way as humans would describe it. This way the system has higher chances to take adaptation decisions that best meet users' expectations.