| The existing video compression standards, such as MPEG-2, H.263, MPEG-4 and H.264/AVC provide technological support for network multimedia applications. However, transmitting video data through network environments, such as wireless network and Internet, is a challenging task, as it requires high compression efficiency and network friendly design. Motion compensation, DCT transform, integer transform and entropy coding, etc, have been used in video compression standards. With increasing coding efficiency, video bitstream becomes more and more vulnerable to channel noise. To enhance the error resiliency of compressed video bitstream, a number of error resiliency schemes have been adopted, such as slice structuring, data partitioning, and reversible variable length code (RVLC). Moreover, with the increasing growth of bandwidth and computing power in Internet and wireless networks, video applications such as video on mobile phones, video chatting/instant messaging, and video streaming, scalable video coding techniques are being rapidly developed. However, they are still experimental in nature and often suffer from performance problems.; In this thesis work, a new low-delay motion compensated temporal filtering (MCTF) structure is discussed (Chapter II) based on the residual energy computations, which outperforms the "low-delay MCTF" scheme suggested in the Joint Scalable Video Model (JSVM) in terms of delay, number of badly matched blocks and PSNR performance. In order to evaluate and compare the performance of RVLC codes, we develop two coding measures in this thesis work (Chapter III): coding efficiency and error resiliency measure. Using these measures, different RVLC schemes are compared for different data such as parameterized generalized Gaussian source, English alphabet and MPEG-4 video sequences. The construction of Markov transition matrix for RVLC decoding is also discussed, in order to evaluate its error propagating distance. The effects for propagating and non-propagating errors are discussed as well. We also present (Chapter IV) an improved RVLC decoding scheme for MPEG-4 video that recovers more blocks and sometimes more macroblocks (MBs) from error propagation region of corrupted video packets, as compared to the MPEG-4 scheme. Simulation studies have been carried out to show that the proposed scheme achieves better data recovery, both in terms of PSNR (peak-signal-to-noise ratio) and perceptual quality. |
