Graphics Reference
In-Depth Information
Tabl e 2. 2
The 32 HEVC non-VCL NAL unit types
Non-VCL NAL unit types
Parameter sets
32
VPS_NUT
Video parameter set
33
SPS_NUT
Sequence parameter set
34
PPS_NUT
Picture parameter set
Delimiters
35
AUD_NUT
Access unit delimiter
36
EOS_NUT
End of sequence
37
EOB_NUT
End of bitstream
Filler data
38
FD_NUT
Filler data
Supplemental enhancement
information (SEI)
39
PREFIX_SEI_NUT
40
SUFFIX_SEI_NUT
Reserved
41-47
RSV
Unspecified
48-63
UNSPEC
The end of sequence and end of bitstream types are used to indicate the end of
a coded video sequence and the end of a bitstream, respectively. If used, they are
placed last in their access units and must indicate temporal layer 0. They have no
payload so they both consist of only the two byte NAL unit header.
Filler data NAL units have no impact on the decoding process. The payload is a
series of bytes equal to '11111111' followed by a byte equal to '10000000'. It can
be used to fill up a data channel to a desired bit rate in the absence of an adequate
amount of VCL data. Filler data NAL units must signal the same temporal ID as
the coded picture and they are not allowed to precede the first VCL NAL unit in the
access unit.
The Supplemental Enhancement Information (SEI) NAL unit type is explained
in more detail in Sect.
2.5
.
NAL unit types 41-47 are reserved, and types 48-63 are unspecified.
2.3
Parameter Sets
Parameter sets in HEVC are fundamentally similar to the parameter sets in
H.264/AVC, and share the same basic design goals—namely bit rate efficiency,
error resiliency, and providing systems layer interfaces. There is a hierarchy of
parameter sets in HEVC, including the Sequence Parameter Set (SPS) and Picture
Parameter Set (PPS) which are similar to their counterparts in AVC. Additionally,
HEVC introduces a new type of parameter set called the Video Parameter Set (VPS).
Each slice references a single active PPS, SPS and VPS to access information
used for decoding the slice. The PPS contains information which applies to all slices
in a picture, and hence all slices in a picture must refer to the same PPS. The slices
in different pictures are also allowed to refer to the same PPS. Similarly, the SPS
contains information which applies to all pictures in the same coded video sequence.
The VPS contains information which applies to all layers within a coded video
