Use of metadata and context for enriching adaptation and search and retrieval in governed distributed multimedia applications
- Autores: Anna Carreras Coch
- Directores de la Tesis: Jaime Delgado Mercé (dir. tes.)
- Lectura: En la Universitat Politècnica de Catalunya (UPC) ( España ) en 2010
- Idioma: español
- Tribunal Calificador de la Tesis: Lluís Torres Urgell (presid.), Rubén Tous Liesa (secret.), Maria Teresa Magalhaes Da Silva Pinto de Andrade (voc.), José María Martínez Sánchez (voc.), Safak Dogan (voc.)
- Materias:
- Texto completo no disponible (Saber más ...)
- Resumen
It is largely recognised that multimedia applications have become an important part of our everyday life. Currently, users consume multimedia contents at any time and from a growing number of different usage environments. Not only the rising tide of available content nowadays is creating the need of new tools for guiding users to be able to search and find what is of their interest, but the increasing heterogeneity and the augmented capabilities of terminals and networks are creating new opportunities for multimedia applications developers. In particular, a lot of research work is being conducted on the maximisation of the quality of users' multimedia experiences. Indeed, nowadays, it is not enough to identify and retrieve multimedia content, but an optimised version of that content is usually required. Nevertheless, the possibility of having all the necessary versions of the same content available is not usually feasible, and thus, multimedia adaptation services such as transcoding or personalisation of the content have become essential. Moreover, due to the increasing diversity of multimedia usage environments, these adaptation services need to be always evolving and thus, they are becoming more and more complex and varied. Within the research community, the Universal Multimedia Access (UMA) concept includes all the technologies and tools contributing to the access of multimedia content anytime and anywhere. But even though a substantial amount of work has been done in this area, it is possible to identify serious limitations in terms of interoperability and extensibility of the systems.
Another key factor in the delivery of multimedia content today is the growing number of intellectual property owners and their increasing eagerness to establish, control and protect their rights. Indeed, the affordability of consumer electronic devices such as recordable players, digital cameras, etc., allow users to also become content producers (processors, distributors, publishers, etc.) as well as consumers, creating new and very interesting challenges from the perspective of enforcing Digital Rights Management (DRM) during content access and consumption. Not only new business models have appeared with their subsequent need of being implemented, but also more flexible DRM mechanisms are required in order to support current pervasive application scenarios. Finally, but not less important, as context-aware applications are becoming more and more popular as a mean of enriching users' multimedia experiences, there is a consequent growing concern on the illegitimate use of contextual information. While more and more users are willing to share their contents on the internet, there are also a growing number of them who are already conscious of its effects on their privacy, and thus, require the necessary mechanism to protect it. Indeed, even though a lot of work has been done in this area, there is not a common model for privacy and neither a serious standardisation initiative trying to address this problem for generic multimedia scenarios. But of course, these limitations in terms of context protection may be understandable if we consider that nowadays there is still a lack of a solid framework on the acquisition and representation of contextual information. The general goal of this research project is to use context, metadata, and digital rights for enriching distributed multimedia applications. The starting point of this thesis is the work developed within the Xarxa IP Audiovisual de Catalunya (XAC) project in 2005-2006, in which a group of content producers (mainly local televisions (TVs)) wanted to have the mean of exchanging their multimedia contents. From the obtained results, we could identify a number of important issues that still needed to be addressed. In particular, three main issues needed to be addressed: First, there was a clear need of integrating context-aware content adaptation functionalities in the search and retrieval and DRM-enabled platform for maximising the quality of users' experiences. Second, more flexible and user-friendly DRM tools would need to be defined. And third, a more interoperable and extensible multimedia search and retrieval application was required.
Furthermore, our contributions should be extensible to a generic distributed multimedia scenario. Thus, our foremost requirement was to use open standards. Finally, our obtained results should be evaluated within a number of solid frameworks, and thus, their application in a number of European and Spanish projects should also be detailed.
In line with the open issues previously identified, this thesis presents five independent but related contributions.
Our first Contribution presents a first attempt of an adaptation authorisation based on MPEG-7 descriptors.
Our second Contribution presents the identification of a complete standardised set of context descriptors, and an innovative approach for modelling them based on profiles.
Our third Contribution presents a novel adaptation authorisation process based on the context profiles specified in Contribution 2 which improves the efficiency of the adaptation authorisation presented in Contribution 1.
Contribution 4 presents a number of extensions to Contribution 3 in order to integrate the implemented Adaptation Authoriser within a complete DRM scenario in which, not only content adaptation needs to be governed.
And finally, Contribution 5 specifies the implementation of a privacy model based on MPEG-21 Rights Expressions Language (REL) aligned with the rest of our work.
With the results obtained throughout this thesis we can confirm the importance of the use of context, metadata, and digital rights within current multimedia scenarios. And we have presented five independent but related contributions which have contributed on the enrichment of multimedia systems through the use of contextual information, metadata and digital rights.
