Kristl, Seibt & Co. GesmbH entwickelt Prüfstände für den Fahrzeugbau. Am Prüfstand werden realitätsnahe Einflüsse auf Fahrzeuge getestet, die vorher auf einer realen Strecke aufgezeichnet wurden. Um wettbewerbsfähig zu bleiben trat Kristl, Seibt & Co. GesmbH an die Universität Klagenfurt heran, um eine multimediale  Visualisierung der Strecke zu entwickeln. Im Rahmen eines Vorprojekts sollten Machbarkeit und Innovationspotential geprüft die Simulation zusätzlich durch multimediale Visualisierung des Straßenverlaufs anzureichern. Ziel war die Problemidentifikation und Aufwandsabschätzung durch prototypische Aufbauten, sowie Auswertungen der gewonnenen Daten.

1. Die Bereitstellung einer Plattform zur Dokumentation der Geschichte der Alpen-Adria-Universität auf eine einzigartige und offene Weise und

2. die anwendungsorientierte und transdisziplinäre Beforschung der Plattform, die es erlaubt, sowohl Forschung in einem kontrollierten Umfeld durchzuführen als auch einen Mehrwert für die Alpen-Adria Universität Klagenfurt und ide Plattform UniKluHiStories zu erzeugen.

Klaus Zahiragic

Dr. Roland Tusch

The LOOK project aims at developing a content platform for the ASFiNAG Verkehrstelematik GmbH, which provides a broad access to digital content about traffic situations of interest on Austria’s highways and motorways. Thereby, digital content mainly refers to video scenes, images, and correspondingly prepared presentations. Broad access is to be realized in a way that this content may be queried and presented differently, depending on the current usage context. Thus, content should be especially consumable on different (at first ASFiNAG-internal) consumer devices like Desktop PCs, Pocket PCs, and smart phones. The main project tasks include (i) the design of LOOK interfaces to the video and sensor crafts of the ASFiNAG Verkehrstelematik GmbH, (ii) the detection of traffic situations of interest from the data of the sensor craft, (iii) the recording of corresponding video segments from the video craft, (iv) the semi-automatic, standard conform and semantic indexing of these video segments and their publication on the LOOK platform, and finally (v) the search and presentation of these videos in dependence of the current usage context. The LOOK project builds upon some already existing research results in the area of context-aware multimedia services from the scientific partner, and development results in the areas of video detection, video surveillance installations, and sensor data gathering from the industrial partner. The project is proposed to run for 18 months, starting in September 2007. It is organized in seven work packages, which are to be accomplished by five project collaborators. Whereas the months 1 to 16 are devoted to research and development of LOOK platform services, the last two months are dedicated to setup a LOOK test bed at the ASFiNAG Verkehrstelematik GmbH, and to evaluate the platform services there.

The purpose of the CAMUS project is to develop a framework for context-aware multimedia services in stationary and mobile environments, based on existing results obtained from our basic research project ADMITS (Adaptation in Distributed Multimedia IT Systems). The ADMITS project aims at building an experimental distributed multimedia system for investigations into adaptation of both media content and media services. It develops and evaluates a number of algorithms for content and service adaptation by utilizing MPEG-conforming metadata and integrates these algorithms into software components for multimedia servers, proxies, gateways, and clients. The ADMITS project addresses a broad range of scientific questions about adaptation in distributed multimedia systems, but it has no specific orientation in the direction of particular application areas. The CAMUS project aims at translating the results of the ADMITS project and directing the research focus into a multimedia services environment which includes not only stationary clients, but also mobile devices. Experimental results from the ADMITS project have shown that adaptation is a useful technique in networks of stationary devices, but becomes even more promising and essential in mobile environments, where the instantaneous network conditions usually show stronger fluctuations and the capabilities of the client devices are usually much more restricted. Due to the direction towards mobile environments, the context awareness of the ADMITS project has to be extended concerning the location of a client's device. Moreover, due to the much lower presentation capabilities on mobile devices, not only the media data, but also the metadata and presentational data require customization. The major outcome of the project shall be a software framework for context-aware multimedia services, which fully conforms to prevailing (MPEG, W3C, IETF) multimedia coding, communication, and presentation standards, and which can be easily tailored to specific application areas. The project is proposed to run for 24 months and be organized in (1) a sub-project covering the localization and adaptation support in mobile environments, and (2) a sub-project dealing with the efficient management of metadata, contextual data, and presentation data, as well as the development of client applications. Both sub-projects will together design and implement a mobile multimedia test bed, deploy and test the developed framework in this test bed, and evaluate multimedia content adaptation techniques in the test bed. Further evaluations of content adaptation algorithms and the development of a business case conclude the project.

The aim of the project “GeoMT” is the development of a geo-referenced digital tourist guide prototype for selected touristy attractive points of interest in Carinthia. The tourist guide runs either on mobile devices (like PDAs), on stationary info terminals, or is directly accessible via the web. These end devices receive their data from a conjoint geo-referenced multimedia repository, which is accessible through a presentation server. The presentation server automatically adapts all multimedia content (text, images, audio and in particular video) including the presentation GUI itself, according to the capabilities of the actual end device. The adapted data is then transmitted from the server to the end device, by using an available underlying network. The multimedia content is connected to the geographical positions of the selected points of interest by using WinGIS components from the company PROGIS. By coupling location-based services (e.g., by using GPS, RFID, or Bluetooth technology) with this GIS component, multimedia content about adjacent points of interest (POIs) can be presented, and even routes between these POIs can be calculated and drawn into the map. A further essential part of the project is the development of a web-based content management system (CMS), which enables for an easy and distributed editing of POIs, according to the bottom-up approach. The management of the multimedia repository includes the management of each object’s content, metadata and geographical position. The resulting mobile and stationary multi-device prototype will offer the user a personalized and multimedia enriched presentation of selected touristy attractive points of interest in Carinthia. The presentation of the user’s current location will use a map-like interface, where each map may reach a satellite picture-based accuracy of about 2.5 meters.

The project promotes the co-operation of educational and research institutes in area of knowledge technologies and their applications in eLearning, and accelerates the dissemination and application of the research results in the member and associated candidate countries. The current cooperation between the partner institutions will be enhanced by means of visits, working meetings and joint workshops for presenting and analysing their research outcomes, experiments and best practice solutions. The project will increase the societal impact of research results in the field of knowledge technologies, such as semantic web techniques, data and web mining, multiagent technologies, embedded intelligence approaches etc. It will exploit the educational potential of the embedded intelligence applications and the integration of web, digital TV and mobile technologies. The integration of high quality research institutions for exchange of research information, results and personnel in priority IST areas (Semantic-based knowledge systems, Technology-enhanced learning) will be an effective instrument for dissemination and deployment of knowledge intensive applications in the partner ACC in their transition to the Knowledge Society. Furthermore, the project promotes applications of the partners research results in production of the multimedia-based educational materials, permitting interoperability between different eLearning platforms and multilingual re-usability of eLearning content in different language environments.

Thomas Grassauer

Web 2.0 Anwendungen stellen eine besondere Herausforderung an IT-Architekturen dar, da sie zum einen Stabilität und Skalierbarkeit und zum anderen Flexibilität voraussetzen. In diesem Bericht werden die Web 2.0 Plattformen YouTube, PlentyOfFish, Wikimedia, MySpace, LiveJournal  und Twitter bzgl. ihrer IT-Architektur und den verwendeten Technologien untersucht. Weiters wird eine allgemeine Empfehlung für die technische Realisierung von Web 2.0 Plattformen abgegeben.

Thomas Grassauer

Nach Mika (2005) nähert eine Folksonomy eine Ontologie an und integriert dabei die Einzelmeinungen der Benutzer, die Tags Ressourcen zugeordnet haben. Der Prototyp baut aus Tags, Ressourcen und Benutzern ein Netzwerk auf, das für die konzeptuelle Suche genutzt wird. Die Schwierigkeit dabei ist das Netzwerk so einfach wie möglich (Reduktion der Kanten im Netzwerk) aber so ausdrucksstark wie nötig (kein übermäßiges Löschen von Kanten) zu halten. Auf Basis dieses Netzwerks werden dann Analysemethoden angewandt.

This project covers the implementation of a multimedia player for Siemens/BenQ Smartphones and other mobile devices. The player will be able to process and output presentations authored in the new Lightweight Application Scene Representation (LASeR) format (ISO/IEC 14496-20). The resulting software should also be used as a component for further research in adaptation of multimedia presentations.

Thomas Grassauer