3D Visual Communications

10.1 Rate Control in Adaptive Streaming

Streaming 3D video over communication networks for real-time application has many challenges. One of these is how to utilize the current limited network resources to achieve the best perceived 3D video quality. Besides, the bandwidth provided by the network may vary from time to time. For the best received quality, the streaming applications need to dynamically perform rate control to adjust the encoding rate at the encoder side to fit in the current network status and maintain the playback smoothness at the decoder side. However, different coding tools and modes can be selected to meet the same target bit rate but result in different perceived quality. Although the encoder can resort to the full search methodology by trying all possibilities for best tradeoff, this kind of solution is often NP hard and/or very time consuming, thus not practical for real-time applications. It is desired to have a fast selection method for near-optimal solutions to reduce the computation complexity at the encoder side. In this section, we will address both the rate control and the mode decision issue in the 3D video streaming applications.

10.1.1 Fundamentals of Rate Control

A generic 3D video streaming framework is shown in Figure 10.1. The encoder located at the transmitter side will adjust the coding parameters to control the bit rate according to the latest estimated network bandwidth and optional error control message through receiver's feedback. The ...