Visual Media Coding and Transmission

4.10 Conclusions

Research on video coding technologies has been carried out with significant success for several decades since the early days of analog video signal processing. The evolving technologies have been standardized by ITU-T and ISO/IEC for the benefit of economical commercial developments. Still, DVC has been able to successfully mark its place as a promisingly emerging novel approach for video coding. With its flexible architecture, which enables the design of very low-complexity video encoders, DVC is expected to pave the way for a new era of significantly low-cost and miniature video cameras for a number of demanding applications. The potential beneficiaries of DVC include surveillance systems and wireless sensor networks.

DVC is based on the distributed source coding concept, which discusses the independent encoding and joint decoding of statistically-dependent discrete random sequences, as described by the Slepian–Wolf and Wyner–Ziv theorems discussed in Section 4.1. Accordingly, decoding with side information is a feature of DVC, in contrast to conventional video coding techniques. A number of possible DVC implementations have been proposed in the literature. The architecture of the generic pixel domain DVC codec based on turbo coding, which is widely discussed in related literature, was described earlier. Some of the notable components of the DVC codec architecture considered include: quantization, turbo encoding, parity puncturing, side information estimation, ...