Adaptive Multiview Video Streaming: Challenges and Opportunities
Delivering multiview video content over present packet networks poses multiple challenges. First, the best effort nature of the Internet exposes media packets to variable bandwidth, loss, and delay as they traverse the network. Second, the prediction dependencies employed to maximize compression efficiency make the reconstruction process at the client extremely vulnerable to missing data. Third, the heterogeneity of client devices in terms of computing power, display capabilities, and access link capacity necessitates customizing the streaming process per user. My article reviews existing opportunities for addressing these challenges from within each of the three main stages of the content delivery pipeline (i.e., encoding, transmission, and reconstruction). Concretely, I first describe adaptive source coding techniques that construct a compressed representation of the multiview video source that exhibits resilience to network bandwidth variations and client view selection uncertainty. Then I discuss intelligent methods for error protection, caching, and packet scheduling that organize the transmission of multiview data in a bandwidth-effective way. Here, I also review prospective multipath and cloud-assisted techniques for multiview video streaming. Finally, I identify robust client-side content reconstruction schemes and adaptive media playout methods that can minimize the impact of missing data and enhance the user's interactive experience. Then I proceed to describe community-driven streaming techniques for delivering interactive multiview content over a population of social peers. The article concludes with an outline of approaches for synergistic exploitation of the techniques I will present theretofore, jointly across the different layers of the network protocol stack at which they individually operate. Here, I also highlight the main deployment challenges for some of these techniques, and how their design should be addressed accordingly, to overcome them.
- View record in Web of Science
Record created on 2013-10-01, modified on 2016-08-09