| Time: | June 12, 2007 |
|---|---|
| Download as iCal: |
|
A joint event with the Vienna University of Technology, Austria
Stuttgart, Germany, June 12, 2007
University of Stuttgart, Germany
supported by COST 290 and VDE
The seminar was first held last year in Vienna as a joint event of the University of Stuttgart and the Technical University of Vienna. It is an offspring of the cooperation in COST 290. This year's event will be located in Stuttgart at the Vaihingen Campus of the university. We solicit presentations from both academia and industry in the above described or in related fields. If you are interested, please send the title of your presentation along with a brief abstract to video@ikr.uni-stuttgart.de . The deadline for proposals is June 1st, 2007.
We also invite all interested auditors to register for the event. Detailed information can be found here.
Important Dates
Submission Deadline June 1, 2007
Submission
We solicit presentations from both academia and industry in the above described or in related fields. Interested speakers shall send the title of their presentation along with a brief abstract to video@ikr.uni-stuttgart.de . The deadline for proposals is June 1st, 2007.
Registration
If you would like to particiate in the workshop without a presentation, please send an email to video@ikr.uni-stuttgart.de no later than June 8, 1:00 p.m. In the email, please indicate your name and affiliation and whether you would like to participate in the workshop dinner.
Technical Program
We present a cross-layer optimization architecture in order to achieve optimized usage of network resources for multiple users. The objective of our optimization is to improve the perceived quality of the end-users. Our approach is based on a joint optimization of the application layer, the data-link layer and the physical layer of a wireless protocol stack, in which key parameters from each layer are abstracted. We use the Mean Opinion Score (MOS) in the optimization scheme as a common metric to represent the user perceived quality of all applications (e.g. voice calls, video streaming, file transfer). Simulation results show that our MOS-based optimization leads to remarkable improvement in terms of user perceived quality as well as an increased number of users in a cell when compared to the conventional throughput-based optimization.
CELL is a new multimedia-processor which is very suitable for video post-processing like it is needed in streaming applications. We give an introduction into the architecture of the chip including its different levels of parallelism. Afterwards, we present algorithms to enhance video quality, which have been implemented on the architecture.
Video content providers often offer their video streams at three or more different quality levels: low, medium and high quality. These different levels are needed to support different target devices. However, this static kind of adaptation to discrete quality levels cannot meet the requirements of all existing multimedia devices. Because of the increasing heterogeneity of Internet-enabled video devices a more dynamic and individual adaptation to the requirements of the consuming device is needed. One approach for video adaptation is the use of video transcoding techniques, where modifications of the video stream are done in the compressed domain to save processing power. Many transcoding approaches for digital video can be found in the literature, but these focus on adaptation in one single dimension only, such as temporal adaptation, spatial adaptation or quality adaptation. To support as many video-enabled devices with different capabilities as possible, adaptation in more than one dimension is needed. In this talk we present a multidimensional transcoding approach for MPEG-4 encoded video, which smartly combines existing transcoding techniques to enable fine grain adaptation for different video devices.
Necessity of video data rate adaptation to network conditions during a streaming process is undisputed. The common technique is end-to-end rate control. For wireless streaming sessions, the wireless link is usually the bottle-neck. In this case, the end-to-end solution can be too slow to react to highly time-varying and strong changes of channel characteristics.
In this presentation, the network-driven rate control will be introduced. Its advantages and disadvantages are discussed. The evaluation method and results achived by means of IKR video streaming testbed with bit-stream switching support are presented and analyzed.
Provisioning of mobile video streaming is hitting toward to limitations in channel quality and capacity as well as in terminal processing power. These known limitations, network settings, and video content influence the end user quality. In this presentation I explain the estimation of perceived video quality for mobile streaming scenarios. Firstly, I analyze streaming content and usage scenarios. Secondly, I define objective video parameters which reflect the sequence motion character and its content. Finally, video quality estimation methods based on these parameters are developed and compared with common methods.
Mobile multimedia broadcasting, i.e. delivering mass multimedia services to portable devices such as cell phones or PDAs, is a fast growing market. Digital broadcasting networks especially designed for mobile services are currently under deployment, with DVB-H (Digital Video Broadcast ? Handheld) being the most representative technology in Europe. On the other hand, multicast/broadcast extensions to the existing cellular networks are currently under standardization, e.g. MBMS (Multimedia Broadcast Multicast Services). Unfortunately, neither the broadcasting nor cellular systems alone can provide a cost effective service provision in all possible scenarios.
New systems based on interworking/integration of cellular and broadcasting networks are currently under investigation in order to provide affordable mass multimedia services. The goal of this approach is to join the advantages of both networks in order to get a more powerful system. The benefits are manifold, such as reducing the system cost, reducing the overall necessary data rate, improving the perceived area coverage and reducing the electromagnetic exposure, etc.
The presentation deals with introduction in DVB-H and hybrid (UMTS/DVB-H) networks, and describes selected aspects of deployment, planning, and optimization of such networks.
Venue
The 2nd video seminar will take place in Stuttgart in the southern parts of Germany at the Vaihingen campus of the University of Stuttgart. The Vaihingen campus of the University of Stuttgart and all hotels are reachable with the S-Bahn line S1, S2, S3 from the station Flughafen/Airport if you arrive by plane or Hauptbahnhof if you arrive by train.
Map of Stuttgart
Link to Google Maps with the University of Stuttgart as center
Public Transportation
Links
- City of Stuttgart
- All about Stuttgart (Stuttgart Marketing)
- Stuttgart Airport
- VVS - Transit and Tariff Association Stuttgart (Timetable, tickets, etc.)
Contact
If you have any questions, please contact the organizing committee at video@ikr.uni-stuttgart.de.