Multiple Descriptions Video Coding Using Coinciding Lattice Vector Quantizer for H.264/AVC and Motion JPEG2000

Widespread use of networked multimedia applications emphasizes the need for robust communication protocols that can provide fault tolerance and error resiliency. Multiple-description (MD) coding is a technique used to mitigate channel failure when retransmission is expensive or impossible. Multiple descriptions lattice vector quantization has become a popular choice for robust data transmission over unreliable network channels. In this paper, a new generic MD video coding scheme is proposed that is based on coinciding similar sublattices of the hexagonal lattice. The proposed video coding scheme has been exploited to form new MD video coding schemes: MDCLVQ-H.264/AVC and MDCLVQ-Motion JPEG2000. The experimental results for several reference videos show that the encoding performance of the scheme improves, while the quality of the videos in terms of PSNR of the central decoder remains acceptable. Thus, the error resiliency of the scheme improves without a significant drop in the quality of the reconstruction.