Project

Back to overview

YouStream 3D: Immersive Context-Aware Personal Communications

Applicant Cakareski Jakov
Number 143101
Funding scheme Ambizione
Research institution Institut für Informatik Universität Bern
Institution of higher education University of Berne - BE
Main discipline Information Technology
Start/End 01.10.2012 - 31.03.2013
Approved amount 80'307.00
Show all

All Disciplines (5)

Discipline
Information Technology
Mathematics
Electrical Engineering
Sociology
Social work

Keywords (8)

content delivery; decentralized networking; context-driven information processing; online social networks; peer-to-peer systems; cloud computing media; multi-graphs; multiple view video systems

Lay Summary (English)

Lead
Lay summary

The project will investigate intelligent multimedia systems that will address the new engineering and scientific challenges that are introduced by the fusion of community sites, content production tools, the proliferation of multi-camera arrays, and the advent of virtual worlds. Our systems will account for the plethora of contextual information generated in such scenarios to efficiently and reliably deliver immersive multimedia communication to all corners of the worlds. Important applications spanning telecollaboration and telecommuting, environmental monitoring and disaster relief, as well as telemedicine and health care, will all benefit from the advances in networked multimedia that the project will enable. In conjunction, we will study novel techniques for efficient quality-of-experience aware signal processing and communication of decentralized information sources in loose coalitions of networked devices, as frequently encountered in social networking settings and wireless mesh networks, in general. Our distributed protocols will take advantage of the inherent robustness and bandwidth scalability of such environments and will exploit concepts from reinforcement learning, Pareto optimality, and network coding to comprehensively advance the state-of-the-art. They will incorporate energy-aware aspects for further power efficiency and energy conservation.

Direct link to Lay Summary Last update: 21.02.2013

Responsible applicant and co-applicants

Employees

Publications

Publication
Adaptive multi-view video streaming: Challenges and opportunities
Jacob Chakareski, Adaptive multi-view video streaming: Challenges and opportunities, in IEEE Communications Magazine.
Informative State-Based Video Communication
Jacob Chakareski, Informative State-Based Video Communication, in IEEE Transactions on Image Processing, 1-13.
View Selection Policy for Multi-View Video Delivery
Jakov Cakareski, View Selection Policy for Multi-View Video Delivery, in IEEE ICASSP 2013.

Associated projects

Number Title Start Funding scheme
126416 YouStream: Context-Aware Distributed Media Delivery 01.10.2009 Ambizione

Abstract

We are witnessing an integration of computer-communication technologies on an unprecedented scale. People, devices, and computers interact in ways unforeseen before. Future predictions speak of even more fundamental changes lying ahead spurred by the ever increasing amount of digital content pervading our lives. At the heart of these technological advances lies our drive to interconnect and share information, anywhere and anytime. In particular, social networking applications are increasingly becoming a landmark of our online existence. For instance, according to recent statistics, we spend on average 15 minutes on YouTube every day and more content is uploaded there in 60 days than all three major US networks created in 60 years. The unparalleled scale and massive content production of online social networks raise a new set of challenges for the reliable delivery and understanding of content in such environments. Even the most powerful content delivery networks today such as Akamai and Limelight are not capable of serving such large audiences, simultaneously. The anticipated personalization of the streaming process in the future, according to different viewing times, client devices, and viewing angles, e.g., as in online games, will only augment the complexity of the problem further. Equally important, the fusion of online communities and media sharing sites has brought to the forefront unfamiliar phenomena such as audience-content interaction and interdependencies. Finally, the proliferation of smart mobile devices has promoted yet another layer of complex context-driven time-varying interrelations between people, locations, and data.The project will investigate intelligent multimedia systems that will address the new engineering and scientific challenges that are introduced by the fusion of community sites, content production tools, the proliferation of multi-camera arrays, and the advent of virtual worlds. Our systems will account for the plethora of contextual information generated in such scenarios to efficiently and reliably deliver immersive multimedia communication to all corners of the worlds. Important applications spanning telecollaboration and telecommuting, environmental monitoring and disaster relief, as well as telemedicine and health care, will all benefit from the advances in networked multimedia that the project will enable. In conjunction, we will study novel techniques for efficient quality-of-experience aware signal processing and communication of decentralized information sources in loose coalitions of networked devices, as frequently encountered in social networking settings and wireless mesh networks, in general. Our distributed protocols will take advantage of the inherent robustness and bandwidth scalability of such environments and will exploit concepts from reinforcement learning, Pareto optimality, and network coding to comprehensively advance the state-of-the-art. They will incorporate energy-aware aspects for further power efficiency and energy conservation.The present proposal represents a continuation of the ongoing Ambizione project PZ00P2-126416. During the course of our investigations, we have made a number of exciting discoveries that were not anticipated at the time when the original project was formulated. They pointed us to new fruitful directions of research that could lead to further important insights into the context-driven and media-rich information systems of the future. However, due to the limited time span of the present project, we will not be able to explore all of them. Therefore, we feel it is important to continue the present project along the research investigations described in this proposal. To summarize the contributions of the ongoing Ambizione project PZ00P2-126416, we published to date two book chapters, six journal papers, and 20 conference articles. Many of these studies are reviewed in Section 2.2. of the proposal that describes our investigations relevant for the present proposal. The author has been invited to give talks about this research at numerous academic institutions recognized internationally for their excellence. There is an additional number of articles that are either in review or will be submitted this year, based on our ongoing investigations. It should be mentioned that the present project has stimulated very fruitful collaboration relationships with a number of research institutions throughout the world, of which the author is particularly proud. I believe that with the additional funding support anticipated by the present proposal further important research contributions of the same nature will follow.
-