MPEG motion picture coding with long-term constraint on distortion variation

Today the rapid evolution of storage-based video coding is driven by emergent cable network standards and novel storage media technologies. Next-generation disk technology such as HD-DVD is now being developed. This has aroused both industry and researchers to start seeking improvements for standard video coding technique, including MPEG and H.26x.; A highly desirable feature in medium-to-high rate storage video applications is uniform decoded video quality. Variable bit rate (VBR) coding has the potential to produce nearly constant visual quality throughout the entire movie by distributing available bits across video segments of different coding complexities, This can be defined as a bit allocation problem with a long-term constraint on distortion variation.; We first consider the optimal bit allocation problem with multiple constraints including disk capacity. Additional distortion bounds are imposed on the individual decoded frames. We find the optimality conditions for this problem and propose a Lagrangian based iterative solution. We then address computational complexity of the exact rate-distortion (R-D) functions for real movies and propose a practical statistical rate production model. The rate production model is used to estimate the coded bits for a given quantizer step size, the so-called rate-quantization (R-Q) function. Correspondingly we develop a distortion model to estimate distortion-quantization ( D-Q) function. Both models are then used in a novel two-pass MPEG-2 VBR encoder to solve the coding problem with long-term near-constant quality. Experimental results are promising and the two-pass encoder effectively avoids disk overflow and controls objective quality variation.; We also address constant subjective quality, which is important for medium bit rates. A human visual system (HVS) based subjective two-pass VBR coder is then proposed. This encoder employs subjective weighting and macroblock-level adaptive quantization. Preliminary subjective tests conclude the improvement is observable and the encoder is still able to avoid disk overflow.; Future research topics are also suggested for possible extensions to the new H.264 coding standard.