A novel quality-aware 3D video adaptive scheme

As increasing amounts of rich multimedia services, such as 3D video, are being requested by users at higher quality, delivering these services at acceptable quality levels is very challenging. A solution is for service providers to provide adaptive services in dynamic network environments. An ideal adaptive system should be able to make the best choice in terms of the delivered multimedia bit-rate in order to achieve optimum user Quality of Experience (QoE) in the given network conditions. Existing systems make adaptations based on current observations of the stream delivery, without considering the resulting effect of these adaptations on the network and the adaptive stream itself. This paper proposes a novel Quality Aware 3D video (QA3D) adaptive scheme that adapts the 3D video source according to changes in received quality due to network variations, based on the prediction of the consequences of all possible adaptation choices. With the proposed scheme, we achieved up to 52% more effective adaptation in comparison to an alternative approach, with 6.4% higher perceived quality maintained on average across the entire streaming session.