A rate adaptation approach for streaming multiview plus depth content

Three-dimensional video has recently gained great popularity as it enhances the viewing experience. Streaming of 3D video over IP networks will soon become a common means of multimedia consumption, thanks to the recent standardization efforts such as MPEG-DASH and 3D-HEVC. To this end, rate adaptation techniques are required to overcome the unreliable nature of IP networks. In this paper, we propose a rate adaptation approach for streaming multiview plus depth (MVD) video by exploring the effects of the number of views and the quality of views on the quality of experience. Specifically, we show that concurrently decreasing the distance between virtual camera positions and reducing the number of views help maintain an acceptable perceptual quality level when the network resources are scarce.