This thesis is concerned with maximizing the coding efficiency, random accessibility and visual performance of scalable compressed video. The unifying theme behind this work is the use of finely embeddedlocalized cod...
详细信息
This thesis is concerned with maximizing the coding efficiency, random accessibility and visual performance of scalable compressed video. The unifying theme behind this work is the use of finely embeddedlocalizedcodingstructures, which govern the extent to which these goals may be jointly achieved. The first aspect of the thesis focuses on scalable volumetric image compression. We investigate 3D transform and coding techniques which exploit inter slice statistical redundancies without compromising slice accessibility. Our study shows that motion-compensated temporal discrete wavelet transform practically achieves an upper bound to the compression efficiency of slice transforms. From a video coding perspective, we evaluate the effectiveness of utilizing context information from neighboring frames. We find that most of the coding gain is attributed to o setting the learning penalty in adaptive arithmetic coding through 3D code block extension, rather than inter-frame context modeling. The second aspect of this thesis examines random accessibility. Accessibil ity refers to the ease with which a region is accessed from a compressed video bitstream. This is analyzed in terms of the smallest subset of subband coe-ficients required for reconstructing a region of interest in a video; subject to the spatiotemporal constraints imposed by code blocks. A key issue that we investi gate is the fundamental implications of motion compensation for random access e ciency and the compression performance of scalable interactive video. We demonstrate that the inclusion of motion compensation operators within the lift ing steps of a temporal subband transform incurs a random access penalty, which varies depending on the characteristics of the motion field. The final aspect of this thesis aims to improve the perceptual quality of scalable wavelet-compressed video. The objective is to make compression artefacts appear as perceptually uniform as possible, so as to minimize the visual imp
暂无评论