An Overview of AVS2 Standard - Advanced Video Coding Systems - page 48

Fig. 3.11 Slice consists of
an integer number of LCU
lines

3.3.2 Slice and Coding Unit

Picture consists of one or more slices, and slice is composed of an integer number of
LCU ordered consecutively in the raster scan order. LCU lines within a slice shall not
overlap. Slices shall not overlap with each other either. See Fig. 3.11 for an example
of slice structure.
As said before, LCU can be a single CU or can be split into four smaller CUs with a
quad-tree partition structure. And once the splitting of CUhierarchical tree is finished,
the leaf nodeCUs can be further split into PUs.More specifically, both intra- and inter-
prediction partitions can be symmetric or asymmetric. Intra prediction partitions vary
in the set 2N × 2N, N × N, 2N × 0.5N, 0.5N × 2N, while inter-prediction partitions

vary in the set 2N 2N,

where U, D, L and R are the abbreviations of Up, Down, Left and Right respectively,
as shown in Fig. 3.2 .

× 2N, 2N × N, N × 2N, 2N × nU, 2N × nD, nL × 2N, nR ×

3.4 Summary

This chapter gives a brief overview of AVS2 video coding standard, including the
coding framework, coding tools and the syntax structure. AVS2 uses the traditional
prediction/transform hybrid coding scheme, but each module has been designed
delicately for improving the coding efficiency. Especially, for scene video coding,
background picture-based coding shows significant coding efficiency improvement
over preceding standards.