| Popis: |
The interest in new video technologies is probably greater now than it has ever been, as hybrid block-based motion compensated coding is not expected to deliver further significant performance gains. Therefore this is a challenging time for ground-breaking video compression approaches. Recently, research activities in this area have emphasized the design and development of efficient perception-oriented video coding frameworks. Hence, number of research groups have focused on adapting perceptual, computer graphics and computer vision methods to several video coding platforms. These approaches will be referred to as video coding with texture analysis and synthesis in the following. While there are many promising algorithms available, there is still a lack of theoretical understanding on how such perception-oriented coding strategies should be optimally designed. This chapter aims to achieve three main goals: (1) to give an extensive overview of theoretical considerations, relevant codec architectures and important implementation details; (2) to motivate and guide researchers that are interested in advanced video compression technologies; (3) to provide an easy-to-read reference to readers with some prior knowledge of video coding. |
