Um estudo comparativo das estruturas de codificação de videos dos padrões H.264/AVC E H.266/VVC

Abstract

This work aims to bring the concepts covered in the encoding standards H.264/AVC and H.266/VVC, in order to study their structures and main differences in their encoders. In terms of services such as video streaming, teleconferencing and videos with UHD resolution (Ultra High Definition), studies on video encoders are considered of vital importance. The H.264/AVC standard brought in its module new tools in the interframe prediction stage called motion estimation and compensation, which were responsible for most of the encoding efficiency in this standard. With the new 8k-16k resolution videos, the H.266/VVC standard achieves an efficiency estimated at 50% above its predecessor, due to new tools implemented, such as flexible partitioning structures, new intra-block prediction tools and a in loop filters module. However, the great advance of coding techniques introduced great algorithmic complexity on the structures, resulting in a great increase in computational cost. From this, this study highlights steps that can be explored by techniques that aim to reduce the computational complexity of these steps and thus reduce the coding time in the new VVC standard. Keywords: Images, Video, Data Compression, Encoding Pattern.