Note: Descriptions are shown in the official language in which they were submitted.
PCTUS2021032924
TECHNIQUES FOR RANDOM ACCESS POINT INDICATION AND PICTURE
OUTPUT IN CODED VIDEO STREAM
CROSS-REFERENCE TO RELATED APPLICATION
[0001] This application claims priority from U.S. Provisional Application
No.
63/037,903, filed on June 11, 2020; U.S. Provisional Application No.
63/036,335, filed on
June 8, 2020; U.S. Provisional Application No. 63/035,274, filed on June 5,
2020; U.S.
Provisional Application No. 63/027,826, filed on May 20, 2020; and U.S.
Application No.
17/320,764, filed on May 14, 2021, the disclosures of which are incorporated
herein by
reference in their entireties.
FIELD
[0002] Embodiments of the present disclosure relate to video coding and
decoding,
and more specifically, to random access picture and its output process in a
coded video
stream with multiple layers.
BACKGROUND
[0003] Video coding and decoding using inter-picture prediction with motion
compensation has been previously used. Uncompressed digital video can include
a series of
pictures, each picture having a spatial dimension of, for example, 1920 x 1080
luminance
samples and associated chrominance samples. The series of pictures can have a
fixed or
variable picture rate (informally also known as frame rate), of, for example
60 pictures per
second or 60 Hz. Uncompressed video has significant bitrate requirements. For
example,
1080p60 4:2:0 video at 8 bit per sample (1920x1080 luminance sample resolution
at 60 Hz
frame rate) requires close to 1.5 Gbit/s bandwidth. An hour of such video
requires more than
600 GByte of storage space.
1
Date recue/date received 2021-10-26
PCTUS2021032924
[0004] One purpose of video coding and decoding can be the reduction of
redundancy
in the input video signal, through compression. Compression can help reduce
aforementioned
bandwidth or storage space requirements, in some cases by two orders of
magnitude or more.
Both lossless and lossy compression, as well as a combination thereof can be
employed.
Lossless compression refers to techniques where an exact copy of the original
signal can be
reconstructed from the compressed original signal. When using lossy
compression, the
reconstructed signal may not be identical to the original signal, but the
distortion between
original and reconstructed signal may be small enough to make the
reconstructed signal
useful for the intended application. In the case of video, lossy compression
is widely
employed. The amount of distortion tolerated depends on the application; for
example, users
of certain consumer streaming applications may tolerate higher distortion than
users of
television contribution applications. The compression ratio achievable can
reflect that: higher
allowable/tolerable distortion can yield higher compression ratios.
[0005] A video encoder and decoder can utilize techniques from several
broad
categories, including, for example, motion compensation, transfoun,
quantization, and
entropy coding, some of which will be introduced below.
[0006] Previously, video encoders and decoders tended to operate on a
given picture
size that was, in most cases, defined and stayed constant for a coded video
sequence (CVS),
Group of Pictures (GOP), or a similar multi-picture timeframe. For example, in
MPEG-2,
system designs were used to change the horizontal resolution (and, thereby,
the picture size)
dependent on factors such as activity of the scene, but only at I pictures,
hence typically for a
GOP. The resampling of reference pictures for use of different resolutions
within a CVS has
been used in, for example, ITU-T Rec. H.263 Annex P. However, here the picture
size does
not change, only the reference pictures are being resampled, resulting
potentially in only parts
of the picture canvas being used (in case of downsampling), or only parts of
the scene being
2
Date recue/date received 2021-10-26
PCTUS2021032924
captured (in case of upsampling). Further, H.263 Annex Q allows the resampling
of an
individual macroblock by a factor of two (in each dimension), upward or
downward. Again,
the picture size remains the same. The size of a macroblock is fixed in H.263,
and therefore
does not need to be signaled.
[0007] Changes of picture size in predicted pictures became more mainstream
in
modern video coding. For example, VP9 allows reference picture resampling and
change of
resolution for a whole picture. Similarly, certain proposals made towards VVC
(including, for
example, Hendry, et. al, "On adaptive resolution change (ARC) for VVC", Joint
Video Team
document JVET-M0135-v1, Jan 9-19, 2019, incorporated herein in its entirety)
allow for
resampling of whole reference pictures to different¨higher or
lower¨resolutions. In such
document, different candidate resolutions are suggested to be coded in the
sequence
parameter set and referred to by per-picture syntax elements in the picture
parameter set.
[0008] Bross, et. al, "Versatile Video Coding (Draft 9)", Joint Video
Experts Team
document JVET-R2001-vA, April 2020, is incorporated herein in its entirety.
SUMMARY
[0009] In a coded video stream. it is widely used to indicate the random
access point
information in high-level syntax structure, such as network abstraction layer
(NAL) unit
header, parameter set, picture header, or slice header. Based on the random
access
information, decoded leading pictures associated with the random access
picture are
managed. In this disclosure, in order to make the decoded picture management
associated
with the random access process clear, some relevant syntax elements and
constraints are
explained.
[0010] When a video bitstream is randomly accessed by a trick mode play, an
intra
random access point (IRAP) picture may enable the random access to an
intermedia point of
the bitstream and successful decoding of the video bitstream at the random
access point. One
3
Date recue/date received 2021-10-26
PCTUS2021032924
of possible methods is the gradual refresh of the scene with some amount of
recovery time. In
VVC and other video codecs, the gradual decoding refresh (GDR) picture and
access unit
(AU) are defined to specify syntax and semantics of the random access
operation with
gradual decoding refresh. In this disclosure, to correctly specify the
signaling and decoding
process of GDR, its syntax, semantics, and constraints are described.
[0011] When one or more reference picture lists are constructed for inter-
prediction in
P or B slices, one or more pictures may be unavailable due to random access or
unintended
picture loss. To avoid any decoder crash or unintended behavior, it is desired
to generate
unavailable pictures with defaults values of pixels and parameters. After
generating
unavailable pictures, it may be needed to check the validation of all
reference pictures in
reference picture list.
[0012] Embodiments of the present disclosure relate to random access
picture and its
output process in a coded video stream with multiple layers. Embodiments of
the present
disclosure related to random access picture and its leading picture output
indication in a
coded video stream with multiple layers. Embodiments of the present disclosure
relate to
signaling random access pictures with gradual decoding refresh and recovery
point in a coded
video stream with multiple layers. Embodiments of the present disclosure
relate to reference
picture list construction and unavailable picture generation in a coded video
stream with
multiple layers. Embodiments of the present disclosure include techniques for
signaling of
adaptive picture size in a video bitstream.
[0013] One or more embodiments of the present disclosure include a method
performed by at least one processor. The method includes: receiving a coded
video stream
that includes an access unit, including a picture; signaling a first flag, in
an access unit
delimiter of the coded video stream, that indicates whether the access unit
includes either or
neither one from among an intra random access point (TRAP) picture and a
gradual decoding
4
Date recue/date received 2021-10-26
PCTU S2021032924
refresh (GDR) picture; signaling a second flag, in a picture header of the
coded video stream,
that indicates whether the picture is the TRAP picture; and decoding the
picture, as a current
picture, based on the signaling of the first flag and the second flag, wherein
a value of the
first flag and a value of the second flag are aligned.
[0014] According to an embodiment, the method further includes: signaling a
third
flag, in the picture header of the coded video stream, that indicates whether
the picture is the
GDR picture, wherein the value of the first flag and a value of the third flag
are aligned.
[0015] According to an embodiment, the third flag is signaled based on the
second
flag indicating that the picture is not the TRAP picture.
[0016] According to an embodiment, the first flag has a value that
indicates that the
picture is either one from among the TRAP picture and the GDR picture, the
second flag has a
value that indicates that the picture is the TRAP picture, and the method
further includes
signaling a third flag, in a slice header of a slice of the picture of the
coded video stream, that
indicates whether any pictures prior to the TRAP picture are outputted.
[0017] According to an embodiment, the method further includes: determining
a
network abstraction layer (NAL) unit type of the slice, wherein the third flag
is signalled
based on the NAL unit type that is determined.
[0018] According to an embodiment, the third flag is signalled based on the
NAL unit
type being determined to be equal to IDR W RADL, IDR N LP, or CRA NUT.
[0019] According to an embodiment, the method further includes: signaling a
fourth
flag, in the picture header of the coded video stream, that indicates whether
the picture is the
GDR picture, wherein the value of the first flag and a value of the fourth
flag are aligned.
[0020] According to an embodiment, the third flag is signaled based on the
second
flag indicating that the picture is not the TRAP picture.
Date recue/date received 2021-10-26
PCTU S2021032924
[0021] According to an embodiment, the decoding includes: constructing a
reference
picture list; generating an unavailable reference picture in the reference
picture list; and
checking bitstream conformance, for references pictures in the reference
picture list, that the
following constraints apply: a number of entries that are indicated to be in
the reference
picture list is not less than a number of active entries that are indicated to
be in the reference
picture list, each picture referred to by an active entry in the reference
picture list is present in
a decoded picture buffer (DPB) and has a temporal identifier value less than
or equal to a
temporal identifier value of the current picture, and each picture referred to
by an entry in the
reference picture list is not the current picture and is indicated by a
picture header flag to
potentially be a reference picture.
[0022] According to an embodiment, the checking bistream conformance is
performed based on determining the current picture is an independent decoder
refresh (IDR)
picture, a clean random access (CRA) picture, or a gradual decoding refresh
(GDR) picture.
[0023] According to one or more embodiments, a system is provided. The
system
includes: at least one processor that is configured to receive a coded video
stream that
includes an access unit, including a picture; and memory storing computer
code, the
computer code including: first signaling code configured to cause the at least
one processor to
signal a first flag, in an access unit delimiter of the coded video stream,
that indicates whether
the access unit includes either or neither one from among an intra random
access point
(TRAP) picture and a gradual decoding refresh (GDR) picture; second signaling
code
configured to cause the at least one processor to signal a second flag, in a
picture header of
the coded video stream, that indicates whether the picture is the TRAP
picture; and decoding
code configured to cause the at least one processor to decode the picture, as
a current picture,
based on the signaling of the first flag and the second flag, wherein a value
of the first flag
and a value of the second flag are aligned.
6
Date recue/date received 2021-10-26
PCTU S2021032924
[0024] According to an embodiment, the computer code further includes:
third signaling code configured to cause the at least one processor to signal
a third
flag, in the picture header of the coded video stream, that indicates whether
the picture is the
GDR picture, wherein the value of the first flag and a value of the third flag
are aligned.
[0025] According to an embodiment, the third flag is signaled based on the
second
flag indicating that the picture is not the TRAP picture.
[0026] According to an embodiment, the first flag has a value that
indicates that the
picture is either one from among the TRAP picture and the GDR picture, the
second flag has a
value that indicates that the picture is the TRAP picture, and the compute
code further
includes third signaling code configured to cause the at least one processor
to signal a third
flag, in a slice header of a slice of the picture of the coded video stream,
that indicates
whether any pictures prior to the TRAP picture are outputted.
[0027] According to an embodiment, the computer code further includes:
determining
code configured to cause the at least one processor to determine a network
abstraction layer
(NAL) unit type of the slice, wherein the third flag is signalled based on the
NAL unit type
that is determined.
[0028] According to an embodiment, the third flag is signalled based on the
NAL unit
type being determined to be equal to IDR W RADL, IDR N LP, or CRA NUT.
[0029] According to an embodiment, the computer code further includes:
fourth
signaling codec configured to cause the at least one processor to signal a
fourth flag, in the
picture header of the coded video stream, that indicates whether the picture
is the GDR
picture, wherein the value of the first flag and a value of the fourth flag
are aligned.
[0030] According to an embodiment, the third flag is signaled based on the
second
flag indicating that the picture is not the TRAP picture.
7
Date recue/date received 2021-10-26
PCTU S2021032924
[0031] According to an embodiment, the decoding code includes: constructing
code
configured to cause the at least one processor to construct a reference
picture list; generating
code configured to cause the at least one processor to generate an unavailable
reference
picture in the reference picture list; and checking code configured to cause
the at least one
processor to check bitstream conformance, for references pictures in the
reference picture list,
that the following constraints apply: a number of entries that are indicated
to be in the
reference picture list is not less than a number of active entries that are
indicated to be in the
reference picture list, each picture referred to by an active entry in the
reference picture list is
present in a decoded picture buffer (DPB) and has a temporal identifier value
less than or
equal to a temporal identifier value of the current picture, and each picture
referred to by an
entry in the reference picture list is not the current picture and is
indicated by a picture header
flag to potentially be a reference picture.
[0032] According to one or more embodiments, a non-transitory computer-
readable
medium storing computer instructions is provided. The computer instructions,
when executed
by at least one processor that receives a coded video stream that includes an
access unit,
including a picture, cause the at least one processor to: signal a first flag,
in an access unit
delimiter of the coded video stream, that indicates whether the access unit
includes either or
neither one from among an intra random access point (TRAP) picture and a
gradual decoding
refresh (GDR) picture; signal a second flag, in a picture header of the coded
video stream,
that indicates whether the picture is the TRAP picture; and decode the
picture, as a current
picture, based on the signaling of the first flag and the second flag, wherein
a value of the
first flag and a value of the second flag are aligned.
8
Date recue/date received 2021-10-26
PCTUS2021032924
BRIEF DESCRIPTION OF THE DRAWINGS
[0033] Further features, the nature, and various advantages of the
disclosed subject
matter will be more apparent from the following detailed description and the
accompanying
drawings in which:
[0034] FIG. 1 is a schematic illustration of a simplified block diagram of
a
communication system in accordance with an embodiment.
[0035] FIG. 2 is a schematic illustration of a simplified block diagram of
a
communication system in accordance with an embodiment.
[0036] FIG. 3 is a schematic illustration of a simplified block diagram of
a decoder in
accordance with an embodiment.
[0037] FIG. 4 is a schematic illustration of a simplified block diagram of
an encoder
in accordance with an embodiment.
[0038] FIG. 5A is a schematic illustration of a first configuration for
signaling ARC
parameters in accordance with a comparative art embodiment.
[0039] FIG. 5B is a schematic illustration of a second configuration for
signaling
ARC parameters in accordance with a comparative art embodiment.
[0040] FIG. 6A is a schematic illustration of a first configuration for
signaling ARC
parameters in accordance with an embodiment.
[0041] FIG. 6B is a schematic illustration of a second configuration for
signaling
ARC parameters in accordance with an embodiment.
[0042] FIG. 6C is a schematic illustration of a third configuration for
signaling ARC
parameters in accordance with an embodiment.
[0043] FIG. 7A is a schematic illustration of an excerpt of a tile group
header in
accordance with an embodiment.
9
Date recue/date received 2021-10-26
PCTUS2021032924
[0044] FIG. 7B is a schematic illustration of an excerpt of a sequence
parameter set in
accordance with an embodiment.
[0045] FIG. 8 is an example of prediction structure for scalability with
adaptive
resolution change.
[0046] FIG. 9A illustrates an example of a syntax table in accordance with
an
embodiment.
[0047] FIG. 9B illustrates an example of a syntax table in accordance with
an
embodiment.
[0048] FIG. 10 is a schematic illustration of a simplified block diagram of
parsing and
decoding POC cycle per access unit and access unit count value in accordance
with an
embodiment.
[0049] FIG. 11 is a schematic illustration of a video bitstream structure
comprising
multi-layered sub-pictures in accordance with an embodiment.
[0050] FIG. 12 is a schematic illustration of a display of a selected sub-
picture with
an enhanced resolution in accordance with an embodiment.
[0051] FIG. 13 is a block diagram of the decoding and display process for a
video
bitstream comprising multi-layered sub-pictures in accordance with an
embodiment.
[0052] FIG. 14 is a schematic illustration of 360 video display with an
enhancement
layer of a sub-picture in accordance with an embodiment.
[0053] FIG. 15A illustrates an example of a layout of divided sub-pictures
in
accordance with an embodiment.
[0054] FIG. 15B illustrates an example of a corresponding sub-picture size
and
position information of one sub-picture in accordance with an embodiment.
[0055] FIG. 16 illustrates a corresponding picture prediction structure of
the sub-
pictures illustrated in FIGS. 15A-B.
Date recue/date received 2021-10-26
PCTUS2021032924
[0056] FIG. 17 illustrates an example of an input picture divided into
multiple sub-
regions that may be coded with one or more layers, in accordance with an
embodiment.
[0057] FIG. 18 illustrates a corresponding layer and picture prediction
structure, with
spatial scalability modality of local region, of the sub-regions illustrated
in FIG. 17.
[0058] FIG. 19A is a schematic illustration of an excerpt of a video
parameter set in
accordance with an embodiment.
[0059] FIG. 19B is a schematic illustration of an excerpt of a sequence
parameter set
in accordance with an embodiment.
[0060] FIG. 20 is an example of a syntax table for sub-picture layout
information in
accordance with an embodiment.
[0061] FIG. 21 is an example of a syntax table to indicate output layers
and
profile/tier/level information for each output layer set according to an
embodiment.
[0062] FIG. 22 is an example of a syntax table to indicate output layer
mode on for
each output layer set according to an embodiment.
[0063] FIG. 23 is an example of a syntax table to indicate the present sub-
picture of
each layer for each output layer set.
[0064] FIG. 24 is an example of a syntax table of video parameter set RBSP.
[0065] FIG. 25 is an example of a syntax table to indicate the output layer
set with
output layer set mode.
[0066] FIG. 26 is an example of a syntax table to indicate the random
access point
associated with IRAP or GDR picture in picture header.
[0067] FIG. 27 is an example of a syntax table to indicate the random
access AU in
AU delimiter.
[0068] FIG. 28 is an example of a syntax table to indicate IRAP picture
presence in
picture header.
11
Date recue/date received 2021-10-26
PCTUS2021032924
[0069] FIG. 29 is an example of a syntax table to indicate output of
leading picture in
slice header.
[0070] FIG. 30 is an example of a syntax table to indicate IRAP picture and
GDR
picture presence in AU delimiter.
[0071] FIG. 31 is an example of a syntax table to indicate a GDR picture
and its
recovery point in picture header.
[0072] FIG. 32 is a schematic illustration of a simplified block diagram of
reference
picture list construction and generation of an unavailable picture.
[0073] FIG. 33 is a block diagram of computer code according to
embodiments.
[0074] FIG. 34 is a diagram of a computer system suitable for implementing
embodiments.
DE TAILED DESCRIPTION
[0075] FIG. 1 illustrates a simplified block diagram of a communication
system (100)
according to an embodiment of the present disclosure. The system (100) may
include at least
two terminals (110, 120) interconnected via a network (150). For
unidirectional transmission
of data, a first terminal (110) may code video data at a local location for
transmission to the
other terminal (120) via the network (150). The second terminal (120) may
receive the coded
video data of the other terminal from the network (150), decode the coded data
and display
the recovered video data. Unidirectional data transmission may be common in
media serving
applications and the like.
[0076] FIG. 1 illustrates a second pair of terminals (130, 140) provided to
support
bidirectional transmission of coded video that may occur, for example, during
videoconferencing. For bidirectional transmission of data, each terminal (130,
140) may code
video data captured at a local location for transmission to the other terminal
via the network
12
Date recue/date received 2021-10-26
PCTUS2021032924
(150). Each terminal (130, 140) also may receive the coded video data
transmitted by the
other terminal, may decode the coded data, and may display the recovered video
data at a
local display device.
[0077] In FIG. 1, the terminals (110-140) may be illustrated as servers,
personal
computers, and smart phones, and/or any other type of terminal. For example,
the terminals
(110-140) may be laptop computers, tablet computers, media players and/or
dedicated video
conferencing equipment. The network (150) represents any number of networks
that convey
coded video data among the terminals (110-140), including for example wireline
and/or
wireless communication networks. The communication network (150) may exchange
data in
circuit-switched and/or packet-switched channels. Representative networks
include
telecommunications networks, local area networks, wide area networks, and/or
the Internet.
For the purposes of the present discussion, the architecture and topology of
the network (150)
may be immaterial to the operation of the present disclosure unless explained
herein below.
[0078] FIG 2 illustrates, as an example for an application for the
disclosed subject
matter, the placement of a video encoder and decoder in a streaming
environment. The
disclosed subject matter can be equally applicable to other video enabled
applications,
including, for example, video conferencing, digital TV, storing of compressed
video on
digital media including CD, DVD, memory stick and the like, and so on.
[0079] As illustrated in FIG. 2, a streaming system (200) may include a
capture
subsystem (213) that can include a video source (201) and an encoder (203).
The video
source (201) may be, for example, a digital camera, and may be configured to
create an
uncompressed video sample stream (202). The uncompressed video sample stream
(202) may
provide a high data volume when compared to encoded video bitstreams, and can
be
processed by the encoder (203) coupled to the camera (201). The encoder (203)
can include
hardware, software, or a combination thereof to enable or implement aspects of
the disclosed
13
Date recue/date received 2021-10-26
PCTUS2021032924
subject matter as described in more detail below. The encoded video bitstream
(204) may
include a lower data volume when compared to the sample stream, and can be
stored on a
streaming server (205) for future use. One or more streaming clients (206) can
access the
streaming server (205) to retrieve video bit streams (209) that may be copies
of the encoded
video bitstream (204).
[0080] In embodiments, the streaming server (205) may also function as a
Media-
Aware Network Element (MANE). For example, the streaming server (205) may be
configured to prune the encoded video bitstream (204) for tailoring
potentially different
bitstreams to one or more of the streaming clients (206). In embodiments, a
MANE may be
separately provided from the streaming server (205) in the streaming system
(200).
[0081] The streaming clients (206) can include a video decoder (210) and a
display
(212). The video decoder (210) can, for example, decode video bitstream (209),
which is an
incoming copy of the encoded video bitstream (204), and create an outgoing
video sample
stream (211) that can be rendered on the display (212) or another rendering
device (not
depicted). In some streaming systems, the video bitstreams (204, 209) can be
encoded
according to certain video coding/compression standards. Examples of such
standards
include, but are not limited to, ITU-T Recommendation H.265. Under development
is a video
coding standard informally known as Versatile Video Coding (VVC). Embodiments
of the
disclosure may be used in the context of VVC.
[0082] FIG. 3 illustrates an example functional block diagram of a video
decoder
(210) that is attached to a display (212) according to an embodiment of the
present disclosure.
[0083] The video decoder (210) may include a channel (312), receiver (310),
a buffer
memory (315), an entropy decoder/parser (320), a scaler/inverse transform unit
(351), an
intra prediction unit (352), a Motion Compensation Prediction unit (353), an
aggregator
(355), a loop filter unit (356), reference picture memory (357), and current
picture memory 0.
14
Date recue/date received 2021-10-26
PCTUS2021032924
In at least one embodiment, the video decoder (210) may include an integrated
circuit, a
series of integrated circuits, and/or other electronic circuitry. The video
decoder (210) may
also be partially or entirely embodied in software running on one or more CPUs
with
associated memories.
[0084] In this embodiment, and other embodiments, the receiver (310) may
receive
one or more coded video sequences to be decoded by the decoder (210) one coded
video
sequence at a time, where the decoding of each coded video sequence is
independent from
other coded video sequences. The coded video sequence may be received from the
channel
(312), which may be a hardware/software link to a storage device which stores
the encoded
video data. The receiver (310) may receive the encoded video data with other
data, for
example, coded audio data and/or ancillary data streams, that may be forwarded
to their
respective using entities (not depicted). The receiver (310) may separate the
coded video
sequence from the other data. To combat network jitter, the buffer memory
(315) may be
coupled in between the receiver (310) and the entropy decoder/parser (320)
("parser"
henceforth). When the receiver (310) is receiving data from a store/forward
device of
sufficient bandwidth and controllability, or from an isosynchronous network,
the buffer (315)
may not be used, or can be small. For use on best effort packet networks such
as the Internet,
the buffer (315) may be required, can be comparatively large, and can be of
adaptive size.
[0085] The video decoder (210) may include a parser (320) to reconstruct
symbols
(321) from the entropy coded video sequence. Categories of those symbols
include, for
example, information used to manage operation of the decoder (210), and
potentially
information to control a rendering device such as a display (212) that may be
coupled to a
decoder as illustrated in Fig. 2. The control information for the rendering
device(s) may be in
the form of, for example, Supplementary Enhancement Information (SET) messages
or Video
Usability Information (VUI) parameter set fragments (not depicted). The parser
(320) may
Date recue/date received 2021-10-26
PCTUS2021032924
parse/entropy-decode the coded video sequence received. The coding of the
coded video
sequence can be in accordance with a video coding technology or standard, and
can follow
principles well known to a person skilled in the art, including variable
length coding,
Huffman coding, arithmetic coding with or without context sensitivity, and so
forth. The
parser (320) may extract from the coded video sequence, a set of subgroup
parameters for at
least one of the subgroups of pixels in the video decoder, based upon at least
one parameters
corresponding to the group. Subgroups can include Groups of Pictures (GOPs),
pictures, tiles,
slices, macroblocks, Coding Units (CUs), blocks, Transform Units (TUs),
Prediction Units
(PUs) and so forth. The parser (320) may also extract from the coded video
sequence
information such as transform coefficients, quantizer parameter values, motion
vectors, and
so forth.
[0086] The parser (320) may perform entropy decoding/parsing operation on
the
video sequence received from the buffer (315), so to create symbols (321).
[0087] Reconstruction of the symbols (321) can involve multiple different
units
depending on the type of the coded video picture or parts thereof (such as:
inter and intra
picture, inter and intra block), and other factors. Which units are involved,
and how they are
involved, can be controlled by the subgroup control information that was
parsed from the
coded video sequence by the parser (320). The flow of such subgroup control
information
between the parser (320) and the multiple units below is not depicted for
clarity.
[0088] Beyond the functional blocks already mentioned, decoder 210 can be
conceptually subdivided into a number of functional units as described below.
In a practical
implementation operating under commercial constraints, many of these units
interact closely
with each other and can, at least partly, be integrated into each other.
However, for the
purpose of describing the disclosed subject matter, the conceptual subdivision
into the
functional units below is appropriate.
16
Date recue/date received 2021-10-26
PCTUS2021032924
[0089] One unit may be the scaler/inverse transform unit (351). The
scaler/inverse
transform unit (351) may receive quantized transform coefficient as well as
control
information, including which transform to use, block size, quantization
factor, quantization
scaling matrices, etc. as symbol(s) (321) from the parser (320). The
scaler/inverse transform
unit (351) can output blocks comprising sample values that can be input into
the aggregator
(355).
[0090] In some cases, the output samples of the scaler/inverse transform
(351) can
pertain to an intra coded block; that is: a block that is not using predictive
information from
previously reconstructed pictures, but can use predictive information from
previously
reconstructed parts of the current picture. Such predictive information can be
provided by an
intra picture prediction unit (352). In some cases, the intra picture
prediction unit (352)
generates a block of the same size and shape of the block under
reconstruction, using
surrounding already reconstructed information fetched from the current (partly
reconstructed)
picture from the current picture memory (358). The aggregator (355), in some
cases, adds, on
a per sample basis, the prediction information the intra prediction unit (352)
has generated to
the output sample information as provided by the scaler/inverse transform unit
(351).
[0091] In other cases, the output samples of the scaler/inverse transform
unit (351)
can pertain to an inter coded, and potentially motion compensated block. In
such a case, a
Motion Compensation Prediction unit (353) can access reference picture memory
(357) to
fetch samples used for prediction. After motion compensating the fetched
samples in
accordance with the symbols (321) pertaining to the block, these samples can
be added by the
aggregator (355) to the output of the scaler/inverse transform unit (351) (in
this case called
the residual samples or residual signal) so to generate output sample
information. The
addresses within the reference picture memory (357), from which the Motion
Compensation
Prediction unit (353) fetches prediction samples, can be controlled by motion
vectors. The
17
Date recue/date received 2021-10-26
PCTUS2021032924
motion vectors may be available to the Motion Compensation Prediction unit
(353) in the
form of symbols (321) that can have, for example, X, Y, and reference picture
components.
Motion compensation also can include interpolation of sample values as fetched
from the
reference picture memory (357) when sub-sample exact motion vectors are in
use, motion
vector prediction mechanisms, and so forth.
[0092] The output samples of the aggregator (355) can be subject to various
loop
filtering techniques in the loop filter unit (356). Video compression
technologies can include
in-loop filter technologies that are controlled by parameters included in the
coded video
bitstream and made available to the loop filter unit (356) as symbols (321)
from the parser
(320), but can also be responsive to meta-information obtained during the
decoding of
previous (in decoding order) parts of the coded picture or coded video
sequence, as well as
responsive to previously reconstructed and loop-filtered sample values.
[0093] The output of the loop filter unit (356) can be a sample stream that
can be
output to a render device such as a display (212), as well as stored in the
reference picture
memory (357) for use in future inter-picture prediction.
[0094] Certain coded pictures, once fully reconstructed, can be used as
reference
pictures for future prediction. Once a coded picture is fully reconstructed
and the coded
picture has been identified as a reference picture (by, for example, parser
(320)), the current
reference picture can become part of the reference picture memory (357), and a
fresh current
picture memory can be reallocated before commencing the reconstruction of the
following
coded picture.
[0095] The video decoder (210) may perform decoding operations according to
a
predetermined video compression technology that may be documented in a
standard, such as
ITU-T Rec. H.265. The coded video sequence may conform to a syntax specified
by the
video compression technology or standard being used, in the sense that it
adheres to the
18
Date recue/date received 2021-10-26
PCTUS2021032924
syntax of the video compression technology or standard, as specified in the
video
compression technology document or standard and specifically in the profiles
document
therein. Also, for compliance with some video compression technologies or
standards, the
complexity of the coded video sequence may be within bounds as defined by the
level of the
video compression technology or standard. In some cases, levels restrict the
maximum
picture size, maximum frame rate, maximum reconstruction sample rate (measured
in, for
example megasamples per second), maximum reference picture size, and so on.
Limits set by
levels can, in some cases, be further restricted through Hypothetical
Reference Decoder
(HRD) specifications and metadata for HRD buffer management signaled in the
coded video
sequence.
[0096] In an embodiment, the receiver (310) may receive additional
(redundant) data
with the encoded video. The additional data may be included as part of the
coded video
sequence(s). The additional data may be used by the video decoder (210) to
properly decode
the data and/or to more accurately reconstruct the original video data.
Additional data can be
in the form of, for example, temporal, spatial, or SNR enhancement layers,
redundant slices,
redundant pictures, forward error correction codes, and so on.
[0097] FIG. 4 illustrates an example functional block diagram of a video
encoder
(203) associated with a video source (201) according to an embodiment of the
present
disclosure.
[0098] The video encoder (203) may include, for example, an encoder that is
a source
coder (430), a coding engine (432), a (local) decoder (433), a reference
picture memory
(434), a predictor (435), a transmitter (440), an entropy coder (445), a
controller (450), and a
channel (460).
[0099] The encoder (203) may receive video samples from a video source (201)
(that is not
part of the encoder) that may capture video image(s) to be coded by the
encoder (203).
19
Date recue/date received 2021-10-26
PCTUS2021032924
[0100] The video source (201) may provide the source video sequence to be
coded by
the encoder (203) in the form of a digital video sample stream that can be of
any suitable bit
depth (for example: 8 bit, 10 bit, 12 bit, ... ), any colorspace (for example,
BT.601 Y CrCB,
RGB, ...) and any suitable sampling structure (for example Y CrCb 4:2:0, Y
CrCb 4:4:4). In
a media serving system, the video source (201) may be a storage device storing
previously
prepared video. In a videoconferencing system, the video source (203) may be a
camera that
captures local image information as a video sequence. Video data may be
provided as a
plurality of individual pictures that impart motion when viewed in sequence.
The pictures
themselves may be organized as a spatial array of pixels, wherein each pixel
can comprise
one or more sample depending on the sampling structure, color space, etc. in
use. A person
skilled in the art can readily understand the relationship between pixels and
samples. The
description below focuses on samples.
[0101] According to an embodiment, the encoder (203) may code and compress
the
pictures of the source video sequence into a coded video sequence (443) in
real time or under
any other time constraints as required by the application. Enforcing
appropriate coding speed
is one function of controller (450). The controller (450) may also control
other functional
units as described below and may be functionally coupled to these units. The
coupling is not
depicted for clarity. Parameters set by the controller (450) can include rate
control related
parameters (picture skip, quantizer, lambda value of rate-distortion
optimization
techniques, ... ), picture size, group of pictures (GOP) layout, maximum
motion vector search
range, and so forth. A person skilled in the art can readily identify other
functions of
controller (450) as they may pertain to video encoder (203) optimized for a
certain system
design.
[0102] Some video encoders operate in what a person skilled in the are
readily
recognizes as a "coding loop". As an oversimplified description, a coding loop
can consist of
Date recue/date received 2021-10-26
PCTUS2021032924
the encoding part of the source coder (430) (responsible for creating symbols
based on an
input picture to be coded, and a reference picture(s)), and the (local)
decoder (433) embedded
in the encoder (203) that reconstructs the symbols to create the sample data
that a (remote)
decoder also would create when a compression between symbols and coded video
bitstream
is lossless in certain video compression technologies. That reconstructed
sample stream may
be input to the reference picture memory (434). As the decoding of a symbol
stream leads to
bit-exact results independent of decoder location (local or remote), the
reference picture
memory content is also bit exact between a local encoder and a remote encoder.
In other
words, the prediction part of an encoder "sees" as reference picture samples
exactly the same
sample values as a decoder would "see" when using prediction during decoding.
This
fundamental principle of reference picture synchronicity (and resulting drift,
if synchronicity
cannot be maintained, for example because of channel errors) is known to a
person skilled in
the art.
[0103] The operation of the "local" decoder (433) can be the same as of a
"remote"
decoder (210), which has already been described in detail above in conjunction
with FIG. 3.
However, as symbols are available and en/decoding of symbols to a coded video
sequence by
the entropy coder (445) and the parser (320) can be lossless, the entropy
decoding parts of
decoder (210), including channel (312), receiver (310), buffer (315), and
parser (320) may
not be fully implemented in the local decoder (433).
[0104] An observation that can be made at this point is that any decoder
technology,
except the parsing/entropy decoding that is present in a decoder, may need to
be present, in
substantially identical functional form in a corresponding encoder. For this
reason, the
disclosed subject matter focuses on decoder operation. The description of
encoder
technologies can be abbreviated as they may be the inverse of the
comprehensively described
21
Date recue/date received 2021-10-26
PCTUS2021032924
decoder technologies. Only in certain areas a more detail description is
required and provided
below.
[0105] As part of its operation, the source coder (430) may perform motion
compensated predictive coding, which codes an input frame predictively with
reference to
one or more previously-coded frames from the video sequence that were
designated as
"reference frames." In this manner, the coding engine (432) codes differences
between pixel
blocks of an input frame and pixel blocks of reference frame(s) that may be
selected as
prediction reference(s) to the input frame.
[0106] The local video decoder (433) may decode coded video data of frames
that
may be designated as reference frames, based on symbols created by the source
coder (430).
Operations of the coding engine (432) may advantageously be lossy processes.
When the
coded video data may be decoded at a video decoder (not shown in FIG. 4), the
reconstructed
video sequence typically may be a replica of the source video sequence with
some errors. The
local video decoder (433) replicates decoding processes that may be performed
by the video
decoder on reference frames and may cause reconstructed reference frames to be
stored in the
reference picture memory (434). In this manner, the encoder (203) may store
copies of
reconstructed reference frames locally that have common content as the
reconstructed
reference frames that will be obtained by a far-end video decoder (absent
transmission
errors).
[0107] The predictor (435) may perform prediction searches for the coding
engine
(432). That is, for a new frame to be coded, the predictor (435) may search
the reference
picture memory (434) for sample data (as candidate reference pixel blocks) or
certain
metadata such as reference picture motion vectors, block shapes, and so on,
that may serve as
an appropriate prediction reference for the new pictures. The predictor (435)
may operate on
a sample block-by-pixel block basis to find appropriate prediction references.
In some cases,
22
Date recue/date received 2021-10-26
PCTUS2021032924
as determined by search results obtained by the predictor (435), an input
picture may have
prediction references drawn from multiple reference pictures stored in the
reference picture
memory (434).
[0108] The controller (450) may manage coding operations of the video coder
(430),
including, for example, setting of parameters and subgroup parameters used for
encoding the
video data.
[0109] Output of all aforementioned functional units may be subjected to
entropy
coding in the entropy coder (445). The entropy coder translates the symbols as
generated by
the various functional units into a coded video sequence, by loss-less
compressing the
symbols according to technologies known to a person skilled in the art as, for
example
Huffman coding, variable length coding, arithmetic coding, and so forth.
[0110] The transmitter (440) may buffer the coded video sequence(s) as
created by
the entropy coder (445) to prepare it for transmission via a communication
channel (460),
which may be a hardware/software link to a storage device which would store
the encoded
video data. The transmitter (440) may merge coded video data from the video
coder (430)
with other data to be transmitted, for example, coded audio data and/or
ancillary data streams
(sources not shown).
[0111] The controller (450) may manage operation of the encoder (203).
During
coding, the controller (450) may assign to each coded picture a certain coded
picture type,
which may affect the coding techniques that may be applied to the respective
picture. For
example, pictures often may be assigned as an Intra Picture (I picture), a
Predictive Picture (P
picture), or a Bi-directionally Predictive Picture (B Picture).
[0112] An Intra Picture (I picture) may be one that may be coded and
decoded
without using any other frame in the sequence as a source of prediction. Some
video codecs
allow for different types of Intra pictures, including, for example
Independent Decoder
23
Date recue/date received 2021-10-26
PCTUS2021032924
Refresh (IDR) Pictures. A person skilled in the art is aware of those variants
of I pictures and
their respective applications and features.
[0113] A Predictive picture (P picture) may be one that may be coded and
decoded
using intra prediction or inter prediction using at most one motion vector and
reference index
to predict the sample values of each block.
[0114] A Bi-directionally Predictive Picture (B Picture) may be one that
may be
coded and decoded using intra prediction or inter prediction using at most two
motion vectors
and reference indices to predict the sample values of each block. Similarly,
multiple-
predictive pictures can use more than two reference pictures and associated
metadata for the
reconstruction of a single block.
[0115] Source pictures commonly may be subdivided spatially into a
plurality of
sample blocks (for example, blocks of 4x4, 8x8, 4x8, or 16x16 samples each)
and coded on a
block-by- block basis. Blocks may be coded predictively with reference to
other (already
coded) blocks as determined by the coding assignment applied to the blocks'
respective
pictures. For example, blocks of I pictures may be coded non-predictively or
they may be
coded predictively with reference to already coded blocks of the same picture
(spatial
prediction or intra prediction). Pixel blocks of P pictures may be coded non-
predictively, via
spatial prediction or via temporal prediction with reference to one previously
coded reference
pictures. Blocks of B pictures may be coded non-predictively, via spatial
prediction or via
temporal prediction with reference to one or two previously coded reference
pictures.
[0116] The video coder (203) may perform coding operations according to a
predetermined video coding technology or standard, such as ITU-T Rec. H.265.
In its
operation, the video coder (203) may perform various compression operations,
including
predictive coding operations that exploit temporal and spatial redundancies in
the input video
24
Date recue/date received 2021-10-26
PCTUS2021032924
sequence. The coded video data, therefore, may conform to a syntax specified
by the video
coding technology or standard being used.
[0117] In an embodiment, the transmitter (440) may transmit additional data
with the
encoded video. The video coder (430) may include such data as part of the
coded video
sequence. Additional data may comprise temporal/spatial/SNR enhancement
layers, other
forms of redundant data such as redundant pictures and slices, Supplementary
Enhancement
Information (SET) messages, Visual Usability Information (VUI) parameter set
fragments,
and so on.
[0118] Before describing certain aspects of embodiments of the disclosure
in more
detail, a few terms are introduced below that are referred to in the remainder
of this
description.
[0119] "Sub-Picture" henceforth refers to, in some cases, a rectangular
arrangement
of samples, blocks, macroblocks, coding units, or similar entities that are
semantically
grouped, and that may be independently coded in changed resolution. One or
more sub-
pictures may form a picture. One or more coded sub-pictures may form a coded
picture. One
or more sub-pictures may be assembled into a picture, and one or more sub
pictures may be
extracted from a picture. In certain environments, one or more coded sub-
pictures may be
assembled in the compressed domain without transcoding to the sample level
into a coded
picture, and in the same or certain other cases, one or more coded sub-
pictures may be
extracted from a coded picture in the compressed domain.
[0120] "Adaptive Resolution Change" (ARC) henceforth refers to mechanisms
that
allow the change of resolution of a picture or sub-picture within a coded
video sequence, by
the means of, for example, reference picture resampling. "ARC parameters"
henceforth refer
to the control information required to perform adaptive resolution change,
that may include,
Date recue/date received 2021-10-26
PCTUS2021032924
for example, filter parameters, scaling factors, resolutions of output and/or
reference pictures,
various control flags, and so forth.
[0121] Above description is focused on coding and decoding a single,
semantically
independent coded video picture. Before describing the implication of
coding/decoding of
multiple sub pictures with independent ARC parameters and its implied
additional
complexity, embodiments for signaling ARC parameters shall be described.
[0122] Referring to FIGS. 6A-C, shown are several novel example embodiments
for
signaling ARC parameters. As noted with each of the embodiments, they have
certain
advantages from a coding efficiency, complexity, and architecture viewpoint. A
video coding
standard or technology may implement one or more of these embodiments, and may
also
include embodiments known from comparative art, for signaling ARC parameters.
Comparative art embodiments include the examples illustrated in FIGS. 5A-B.
The novel
embodiments may not be mutually exclusive, and conceivably may be included in
a standard
or technology that also includes comparative art embodiments so that either
may be used
based on application needs, standards technology involved, or encoder's
choice.
[0123] Classes of ARC parameters may include: (1) up/downsample factors,
separate
or combined in X and Y dimension, or (2) up/downsample factors, with an
addition of a
temporal dimension, indicating constant speed zoom in/out for a given number
of pictures.
Either of the above two may involve the coding or decoding of one or more
syntax elements
that may point into a table containing the factor(s). Such syntax elements may
be short in
length in embodiments.
[0124] "Resolution" may refer to resolution in the X or Y dimension, in
units of
samples, blocks, macroblocks, CUs, or any other suitable granularity, of the
input picture,
output picture, reference picture, coded picture, combined or separately. If
there are more
than one resolution (such as, for example, one for input picture, one for
reference picture)
26
Date recue/date received 2021-10-26
PCTUS2021032924
then, in certain cases, one set of values may be inferred from another set of
values. The
resolution could be gated, for example, by the use of flags. A more detailed
example of
resolution is provided further below.
[0125] "Warping" coordinates, akin to those used in H.263 Annex P, may be
in a
suitable granularity as described above. H.263 Annex P defines one efficient
way to code
such warping coordinates, but other, potentially more efficient ways could
conceivably also
be used. For example, the variable length reversible, "Huffman"-style coding
of warping
coordinates of Annex P could be replaced by a suitable length binary coding,
where the
length of the binary code word could, for example, be derived from a maximum
picture size,
possibly multiplied by a certain factor and offset by a certain value, so to
allow for "warping"
outside of the maximum picture size's boundaries.
[0126] With reference to up or downsample filter parameters, in the
easiest case, there
may be only a single filter for up and/or downsampling. However, in certain
cases, it can be
advantageous to allow more flexibility in filter design, which may be
implemented by
signaling of filter parameters. Such parameters may be selected through an
index in a list of
possible filter designs, the filter may be fully specified (e.g. through a
list of filter coefficients
using suitable entropy coding techniques), and/or the filter may be implicitly
selected through
up/downsample ratios which are signaled according to any of the mechanisms
mentioned
above, and so forth.
[0127] Henceforth, the description assumes an example case where the
coding of a
finite set of up/downsample factors (the same factor to be used in both X and
Y dimension),
that are indicated through a codeword. That codeword can advantageously be
variable length
coded by, for example, using the Ext-Golomb code common for certain syntax
elements in
video coding specifications such as H.264 and H.265. One suitable mapping of
values to
up/downsample factors can, for example, be according to Table 1 below.
27
Date recue/date received 2021-10-26
PCTUS2021032924
TABLE 1
Codeword Ext-Golomb Code Original / Target resolution
0 1 1 / 1
1 010 1 / 1.5 (upscale by 50%)
2 011 1.5 / 1 (downscale by 50%)
3 00100 1 / 2 (upscale by 100%)
4 00101 2 / 1 (downscale by 100%)
[0128] Many similar mappings could be devised according to the needs of an
application and the capabilities of the up and downscale mechanisms available
in a video
compression technology or standard. The table could be extended to more
values. Values
may also be represented by entropy coding mechanisms other than Ext-Golomb
codes (e.g.
using binary coding) that may have certain advantages when the resampling
factors were of
interest outside the video processing engines (encoder and decoder foremost)
themselves, for
example by MANEs. It should be noted that, for the (presumably) most common
case where
no resolution change is required, an Ext-Golomb code can be chosen that is
short (e.g. only a
single bit as, for example, shown in the second row of TABLE 1) that can have
a coding
efficiency advantage over using binary codes for the most common case.
[0129] The number of entries in the table, as well as their semantics, may
be fully or
partially configurable. For example, the basic outline of the table may be
conveyed in a
"high" parameter set such as a sequence or decoder parameter set.
Alternatively or in
addition, one or more such tables may be defined in a video coding technology
or standard,
and may be selected through, for example, a decoder or sequence parameter set.
28
Date recue/date received 2021-10-26
PCTUS2021032924
[0130] Provided below is a description of how an upsample/downsample factor
(ARC
information), coded as described above, may be included in a video coding
technology or
standard syntax. Similar considerations may apply to one or a few codewords
controlling
up/downsample filters. Provided below is also a description regarding when
comparatively
large amounts of data may be required for a filter or other data structures.
[0131] With reference to FIG. 5A, H.263 Annex P includes ARC information
(502) in
the form of four warping coordinates within a picture header (501),
specifically in an H.263
PLUSPTYPE (503) header extension. Such a design may be sensible when (a) there
is a
picture header available, and (b) frequent changes of the ARC information are
expected.
However, the overhead when using H.263-style signaling can be quite high, and
scaling
factors may not pertain to picture boundaries because picture header can be of
transient
nature.
[0132] With reference to FIG. 5B, JVCET-M135-v1 includes ARC reference
information (505), (an index) located in a picture parameter set (504), that
indexes a table
(506) including target resolutions that is located inside a sequence parameter
set (507). The
placement of the possible resolution in the table (506) in the sequence
parameter set (507)
may be justified by using the SPS (507) as an interoperability negotiation
point during
capability exchange. Resolution can change, within the limits set by the
values in the table
(506) from picture to picture by referencing the appropriate picture parameter
set (504).
[0133] With reference to FIGS. 6A-C, the following embodiments of the
present
disclosure may convey ARC information in a video bitstream to, for example, a
decoder of
the present disclosure. Each of those embodiments has certain advantages over
comparative
art described above. The embodiments may be simultaneously present in the same
video
coding technology or standard.
29
Date recue/date received 2021-10-26
PCTUS2021032924
[0134] In an embodiment with reference to FIG. 6A, ARC information (509)
such as
a resampling (zoom) factor may be present in a header (508) such as, for
example, a slice
header, GOB header, tile header, or tile group header. As an example, FIG. 6A
illustrates the
header (508) as a Tile Group header. Such a configuration can be adequate if
the ARC
information is small, such as a single variable length ue(v) or fixed length
codeword of a few
bits, for example as shown in TABLE 1. Having the ARC information directly in
a tile group
header has the additional advantage that the ARC information may be applicable
to a sub
picture represented by, for example, the tile group corresponding to the tile
group header,
rather than the whole picture. In addition, even if the video compression
technology or
standard uses only whole picture adaptive resolution changes (in contrast to,
for example, tile
group based adaptive resolution changes), putting the ARC information into a
tile group
header (e.g. into an H.263-style picture header) has certain advantages from
an error
resilience viewpoint. While the above description describes the ARC
information (509) being
present in a tile group header, it will be understood that the above
description may also
similarly apply in cases where the ARC information (509) is present in, for
example, a slice
header, GOB header, or tile header.
[0135] In the same or another embodiment with reference to FIG. 6B, ARC
information (512) itself may be present in an appropriate parameter set (511)
such as, for
example, a picture parameter set, header parameter set, tile parameter set,
adaptation
parameter set, and so forth. As an example, FIG. 6B illustrates the parameter
set (511) as an
adaptation parameter set (APS). The scope of that parameter set can
advantageously be no
larger than a picture. For example, the scope of the parameter set may be a
tile group. The use
of the ARC information (512) may be implicit through the activation of the
relevant
parameter set. For example, when a video coding technology or standard
contemplates only
Date recue/date received 2021-10-26
PCTUS2021032924
picture-based ARC, then a picture parameter set or equivalent may be
appropriate as the
relevant parameter set.
[0136] In the same or another embodiment with reference to FIG. 6C, ARC
reference
information (513) may be present in a Tile Group header (514) or a similar
data structure.
The ARC reference information (513) can refer to a subset of ARC information
(515)
available in a parameter set (516) with a scope beyond a single picture. For
example, the
parameter set (516) may be a sequence parameter set (SPS) or a decoder
parameter set (DPS).
[0137] The additional level of indirection implied activation of a PPS from
a tile
group header, PPS, or SPS as used in JVET-M0135-v1 may be unnecessary, as
picture
parameter sets, just as sequence parameter sets, can be used for capability
negotiation or
announcements. However, if ARC information should be applicable to a sub
picture that is
also represented by, for example, a tile group(s), a parameter set (e.g. an
adaptation parameter
set or a header parameter set) with an activation scope limited to a tile
group may be the
better choice. Also, if the ARC information is of more than negligible
size¨for example
contains filter control information such as numerous filter coefficients¨then
a parameter
may be a better choice than using a header directly from a coding efficiency
viewpoint, as
those settings may be reusable by future pictures or sub-pictures by
referencing the same
parameter set.
[0138] When using the sequence parameter set or another higher parameter
set with a
scope spanning multiple pictures, certain considerations may apply:
[0139] (1) The parameter set (516) to store the ARC information (515) in a
table can,
in some cases, be a sequence parameter set, but in other cases can
advantageously be a
decoder parameter set. The decoder parameter set can have an activation scope
of multiple
CVSs, namely the coded video stream, i.e. all coded video bits from session
start until session
teardown. Such a scope may be more appropriate because possible ARC factors
may be a
31
Date recue/date received 2021-10-26
PCTUS2021032924
decoder feature, possibly implemented in hardware, and hardware features tend
not to change
with any CVS (which in at least some entertainment systems is a Group of
Pictures, one
second or less in length). Nevertheless, some embodiments may include the ARC
information
table in the sequence parameter set as described herein, in particular in
conjunction with point
(2) below.
[0140] (2) The ARC reference information (513) may advantageously be placed
directly into the header (514) (e.g. picture/slice tile/GOB/tile group header;
tile group header
henceforth) rather than into the picture parameter set as in JVCET-M0135-v1,
The reason is
as follows: when an encoder wants to change a single value in a picture
parameter set, such as
for example the ARC reference information, then the encoder may have to create
a new PPS
and reference that new PPS. In a case that only the ARC reference information
changes, but
other information such as, for example, the quantization matrix information in
the PPS stays,
such information can be of substantial size, and would need to be
retransmitted to make the
new PPS complete. As the ARC reference information may be a single codeword,
such as the
index into an ARC information table, which would be the only value that
changes, it would
be cumbersome and wasteful to retransmit, for example, all the quantization
matrix
information. Accordingly, placing ARC reference information directly into a
header (e.g.
header (514)) may be considerably better from a coding efficiency viewpoint
because
indirection through the PPS, as proposed in JVET-M0135-v1, can be avoided.
Also, putting
the ARC reference information into the PPS has the additional disadvantage
that the ARC
information referenced by ARC reference information necessarily needs to apply
to the whole
picture and not to a sub-picture, as the scope of a picture parameter set
activation is a picture.
[0141] In the same or another embodiment, the signaling of ARC parameters
can
follow a detailed example as outlined in FIGS. 7A-B. FIGS. 7A-B depict syntax
diagrams.
The notation of such syntax diagrams roughly follows C-style programming.
Lines in
32
Date recue/date received 2021-10-26
PCTUS2021032924
boldface indicate syntax elements present in the bitstream, and lines without
boldface often
indicate control flow or the setting of variables.
[0142] As an example syntax structure of a header applicable to a (possibly
rectangular) part of a picture, a tile grouper header (600) can conditionally
contain, a variable
length, Exp-Golomb coded syntax element dec_pic size idx (602) (depicted in
boldface).
The presence of this syntax element in the tile group header (600) can be
gated by the use of
adaptive resolution (603). Here, the value of the adaptive resolution flag is
not depicted in
boldface, which means that the flag is present in the bitstream at the point
where it occurs in
the syntax diagram. Whether or not adaptive resolution is in use for this
picture or parts
thereof can be signaled in any high level syntax structure inside or outside
the bitstream. In
the example illustrated in FIGS. 7A-B, adaptive resolution is signaled in a
sequence
parameter set (610) as outlined below.
[0143] FIG. 7B illustrates an excerpt of the sequence parameter set (610).
The first
syntax element shown is adaptive_pic resolution change flag (611). When true,
such flag
can indicate the use of adaptive resolution which, in turn, may require
certain control
information. In the example, such control information is conditionally present
based on the
value of the flag based on the if() statement (612) in the sequence parameter
set (610) and the
tile group header (600).
[0144] When adaptive resolution is in use, in this example, coded is an
output
resolution (613) in units of samples. The output resolution (613) in this
example embodiment
refers to both of syntax elements output_pic width in luma samples and
output_pic height in luma samples, which together can define the resolution of
the output
picture. Elsewhere in a video coding technology or standard, certain
restrictions to either
value can be defined. For example, a level definition may limit the number of
total output
samples, which could be the product of the value of the above two syntax
elements. Also,
33
Date recue/date received 2021-10-26
PCTUS2021032924
certain video coding technologies or standards, or external technologies or
standards such as,
for example, system standards, may limit the numbering range (for example, one
or both
dimensions must be divisible by a power of 2 number), or the aspect ratio (for
example, the
width and height must be in a relation such as 4:3 or 16:9). Such restrictions
may be
introduced to facilitate hardware implementations or for other reasons.
[0145] In certain applications, it can be advisable that the encoder
instructs the
decoder to use a certain reference picture size rather than implicitly assume
a size to be the
output picture size. In this example, the syntax element reference_pic
size_present flag
(614) gates the conditional presence of reference picture dimensions (615)
(again, the
numeral refers to both width and height in the example embodiment).
[0146] FIG. 7B further illustrates a table of possible decoding picture
width and
heights. Such a table can be expressed, for example, by a table indication
(616) (e.g. syntax
element num dec_pic size in luma samples minusl) . The "minus 1" of the syntax
element
can refer to the interpretation of the value of that syntax element. For
example, if the coded
value of the syntax element is zero, one table entry is present. If the coded
value is five, six
table entries are present. For each "line" in the table, decoded picture width
and height are
then included in syntaxes as table entries (617).
[0147] The table entries (617) presented can be indexed using the syntax
element
dec_pic size idx (602) in the tile group header (600), thereby allowing
different decoded
sizes¨in effect, zoom factors¨per tile group.
[0148] Certain video coding technologies or standards, for example VP9,
support
spatial scalability by implementing certain forms of reference picture
resampling (which may
be signaled quite differently from embodiments of the present disclosure) in
conjunction with
temporal scalability, so to enable spatial scalability. In particular, certain
reference pictures
may be upsampled using ARC-style technologies to a higher resolution to form
the base of a
34
Date recue/date received 2021-10-26
PCTUS2021032924
spatial enhancement layer. Such upsampled pictures could be refined using
normal prediction
mechanisms at the high resolution so to add detail.
[0149] Embodiments of the disclosure can be used in such an environment.
In certain
cases, in the same or another embodiment, a value in the network abstraction
layer (NAL)
unit header, for example the Temporal ID field, can be used to indicate not
only the temporal
but also the spatial layer. Doing so has certain advantages for certain system
designs; for
example, existing Selected Forwarding Units (SFU) created and optimized for
temporal layer
selected forwarding based on the NAL unit header Temporal ID value can be used
without
modification for scalable environments. In order to enable that, embodiments
of the present
disclosure may include a mapping between the coded picture size and the
temporal layer to
be indicated by the temporal ID field in the NAL unit header.
[0150] In some video coding technologies, an Access Unit (AU) can refer to
coded
picture(s), slice(s), tile(s), NAL Unit(s), and so forth, that were captured
and composed into a
respective picture/slice/tile/NAL unit bitstream at a given instance in time.
Such instance in
time can be the composition time.
[0151] In High Efficiency Video Coding (HEVC), and certain other video
coding
technologies, a picture order count (POC) value can be used for indicating a
selected
reference picture among multiple reference picture stored in a decoded picture
buffer (DPB).
When an access unit (AU) comprises one or more pictures, slices, or tiles,
each picture, slice,
or tile belonging to the same AU may carry the same POC value, from which it
can be
derived that they were created from content of the same composition time. In
other words, it
can be determined that two picture/slice/tile belong to the same AU and have
the same
composition time in a scenario where the two pictures/slices/tiles carry the
same given POC
value. Conversely, two pictures/tiles/slices having different POC values can
indicate those
pictures/slices/tiles belong to different AUs and have different composition
times.
Date recue/date received 2021-10-26
PCTUS2021032924
[0152] In an embodiment of the disclosure, the aforementioned rigid
relationship can
be relaxed in that an access unit can comprise pictures, slices, or tiles with
different POC
values. By allowing different POC values within an AU, it becomes possible to
use the POC
value to identify potentially independently decodable pictures/slices/tiles
with identical
presentation time. Accordingly, the embodiment of the present disclosure can
enable support
of multiple scalable layers without a change of reference picture selection
signaling (e.g.
reference picture set signaling or reference picture list signaling), as
described in more detail
below.
[0153] In an embodiment, it is still desirable to be able to identify the
AU in which a
picture/slice/tile belongs to, with respect to other picture/slices/tiles
having different POC
values, from the POC value alone. This can be achieved in embodiments as
described below.
[0154] In the same or other embodiments, an access unit count (AUC) may be
signaled in a high-level syntax structure, such as NAL unit header, slice
header, tile group
header, SET message, parameter set or AU delimiter. The value of AUC may be
used to
identify which NAL units, pictures, slices, or tiles belong to a given AU. The
value of AUC
may be corresponding to a distinct composition time instance. The AUC value
may be equal
to a multiple of the POC value. By dividing the POC value by an integer value,
the AUC
value may be calculated. In certain cases, division operations can place a
certain burden on
decoder implementations. In such cases, small restrictions in the numbering
space of the
AUC values may allow substitution of the division operation by shift
operations performed
by embodiments of the present disclosure. For example, the AUC value may be
equal to a
Most Significant Bit (MSB) value of the POC value range.
[0155] In the same embodiment, a value of POC cycle per AU (e.g. syntax
element
poc cycle au) may be signaled in a high-level syntax structure, such as NAL
unit header,
slice header, tile group header, SET message, parameter set or AU delimiter.
The
36
Date recue/date received 2021-10-26
PCTUS2021032924
poc cycle au syntax elements may indicate how many different and consecutive
POC values
can be associated with the same AU. For example, if the value of poc cycle au
is equal to 4,
the pictures, slices or tiles with the POC value equal to 0 ¨ 3, inclusive,
are associated with
the AU with AUC value equal to 0, and the pictures, slices or tiles with POC
value equal to 4
¨ 7, inclusive, are associated with the AU with AUC value equal to 1. Hence,
the value of
AUC may be inferred by embodiments of the present disclosure by dividing the
POC value
by the value of poc cycle au.
[0156] In the same or another embodiment, the value of poc cycle au may be
derived
from information, located for example in the video parameter set (VPS), that
identifies the
number of spatial or SNR layers in a coded video sequence. Such a possible
relationship is
briefly described below. While the derivation as described above may save a
few bits in the
VPS and hence may improves coding efficiency, it can be advantageous to
explicitly code
poc cycle au in an appropriate high level syntax structure hierarchically
below the video
parameter set, so to be able to minimize poc cycle au for a given small part
of a bitstream
such as a picture. This optimization may save more bits than can be saved
through the
derivation process above because POC values (and/or values of syntax elements
indirectly
referring to POC) may be coded in low level syntax structures.
[0157] In the same or another embodiment, FIG. 9A illustrates an example of
a
syntax table to signal the syntax element of vps_poc cycle au (632) in VPS
(630) or SPS,
which indicates the poc cycle au used for all picture/slices in a coded video
sequence, and
FIG. 9B illustrates an example of a syntax table to signal the syntax element
of
slice_poc cycle au (642), which indicates the poc cycle au of the current
slice in slice
header (640). If the POC value increases uniformly per AU, vps contant_poc
cycle_per au
(634) in VPS (630) is set equal to 1 and vps_poc cycle au (632) is signaled in
VPS (630) . In
this case, slice_poc cycle au (642) is not explicitly signaled, and the value
of AUC for each
37
Date recue/date received 2021-10-26
PCTUS2021032924
AU is calculated by dividing the value of POC by vps_poc cycle au (632). If
the POC value
does not increase uniformly per AU, vps contant_poc cycle_per au (634) in VPS
(630) is
set equal to 0. In this case, vps access unit cnt is not signaled, while slice
access unit cnt is
signaled in slice header for each slice or picture. Each slice or picture may
have a different
value of slice access unit cnt. The value of AUC for each AU is calculated by
dividing the
value of POC by slice_poc cycle au (642).
[0158] FIG. 10 illustrates a block diagram for describing relevant work
flow of the
embodiment. For example, the decoder (or encoder) parses VPS/SPS an identifies
whether
the POC cycle per AU is constant or not (652). Following, the decoder (or
encoder) makes a
decision (654) based on whether the POC cycle per AU is constant within a
coded video
sequence. That is, if the POC cycle per AU is constant, the decoder (or
encoder) calculates
the value of the access unit count from the sequence level poc cycle au value
and POC value
(656). Alternatively, if the POC cycle per AU is not constant, the decoder (or
encoder)
calculates the value of access unit count from the picture level poc cycle au
value and POC
value (658). In either case, the decoder (or encoder) may then repeat the
process by, for
example, parsing a VPS/SPS, and identifying whether the POC cycle per AU is
constant or
not (662).
[0159] In the same or other embodiments, even though the value of POC of a
picture,
slice, or tile may be different, the picture, slice, or tile corresponding to
an AU with the same
AUC value may be associated with the same decoding or output time instance.
Hence,
without any inter-parsing/decoding dependency across pictures, slices, or
tiles in the same
AU, all or a subset of pictures, slices, or tiles associated with the same AU
may be decoded in
parallel, and may be outputted at the same time instance.
[0160] In the same or other embodiments, even though the value of POC of a
picture,
slice, or tile may be different, the picture, slice, or tile corresponding to
an AU with the same
38
Date recue/date received 2021-10-26
PCTUS2021032924
AUC value may be associated with the same composition/display time instance.
When the
composition time is contained in a container format, even though pictures
correspond to
different Ails, if the pictures have the same composition time, the pictures
can be displayed
at the same time instance.
[0161] In the same or other embodiments, each picture, slice, or tile may
have the
same temporal identifier (e.g. syntax element temporal id) in the same AU. All
or subset of
pictures, slices or tiles corresponding to a time instance may be associated
with the same
temporal sub-layer. In the same or other embodiments, each picture, slice, or
tile may have
the same or a different spatial layer id (e.g. sytax element layer id) in the
same AU. All or
subset of pictures, slices or tiles corresponding to a time instance may be
associated with the
same or a different spatial layer.
[0162] FIG. 8 shows an example of a video sequence structure (680) with
combination of temporal id, layer id, and POC and AUC values with adaptive
resolution
change. In this example, a picture, slice, or tile in the first AU with AUC =
0 may have
temporal id = 0 and layer id = 0 or 1, while a picture, slice, or tile in the
second AU with
AUC = 1 may have temporal id = 1 and layer id = 0 or 1, respectively. The
value of POC is
increased by 1 per picture regardless of the values of temporal id and layer
id. In this
example, the value of poc cycle au can be equal to 2. In an embodiment, the
value of
poc cycle au may be set equal to the number of (spatial scalability) layers.
In this example,
the value of POC is increased by 2 while the value of AUC is increased by 1.
As an example,
FIG. 8 illustrates, within the first AU (AUC = 0), an I-slice (681) having a
POC 0, TID 0, and
LID 0, and a B-slice (682) having a POC 1, TID 0, and LID 1. Within the second
AU (AUC
= 1), FIG. 8 illustrates a B-slice (683) having a POC 2, TID 1, and LID 0, and
a B-slice (684)
having a POC 3, TID 1, and LID 1. Within the third AU (AUC = 3), FIG. 8
illustrates a B-
39
Date recue/date received 2021-10-26
PCTUS2021032924
slice (685) having a POC 4, TID 0, and LID 0, and a B-slice (686) having a POC
5, TID 0,
and LID 1.
[0163] In the above embodiments, all or sub-set of inter-picture or inter-
layer
prediction structure and reference picture indication may be supported by
using the existing
reference picture set (RPS) signaling in HEVC or the reference picture list
(RPL) signaling.
In RPS or RPL, the selected reference picture is indicated by signaling the
value of POC or
the delta value of POC between the current picture and the selected reference
picture. In
embodiments of the present disclosure, the RPS and RPL can be used to indicate
the inter-
picture or inter-layer prediction structure without change of signaling, but
with the following
restrictions. If the value of temporal id of a reference picture is greater
than the value of
temporal id of a current picture, the current picture may not use the
reference picture for
motion compensation or other predictions. If the value of layer id of a
reference picture is
greater than the value of layer id of the current picture, the current picture
may not use the
reference picture for motion compensation or other predictions.
[0164] In the same and other embodiments, the motion vector scaling based
on POC
difference for temporal motion vector prediction may be disabled across
multiple pictures
within an access unit. Hence, although each picture may have a different POC
value within
an access unit, the motion vector may not be scaled and used for temporal
motion vector
prediction within an access unit, because a reference picture with a different
POC in the same
AU may be considered a reference picture having the same time instance.
Therefore, in the
embodiment, the motion vector scaling function may return 1 when the reference
picture
belongs to the AU associated with the current picture.
[0165] In the same and other embodiments, the motion vector scaling based
on POC
difference for temporal motion vector prediction may be optionally disabled
across multiple
pictures, when the spatial resolution of the reference picture is different
from the spatial
Date recue/date received 2021-10-26
PCTUS2021032924
resolution of the current picture. When the motion vector scaling is allowed,
the motion
vector may be scaled based on both POC difference and the spatial resolution
ratio between
the current picture and the reference picture.
[0166] In the same or another embodiment, the motion vector may be scaled
based on
AUC difference instead of POC difference for temporal motion vector
prediction, especially
when the poc cycle au has non-uniform value (when vps contant_poc cycle_per au
¨ 0).
Otherwise (when vps contant_poc cycle_per au ¨ 1), the motion vector scaling
based on
AUC difference may be identical to the motion vector scaling based on POC
difference.
[0167] In the same or another embodiment, when the motion vector is scaled
based on
AUC difference, the reference motion vector in the same AU (with the same AUC
value)
with the current picture is not scaled based on AUC difference and used for
motion vector
prediction without scaling or with scaling based on spatial resolution ratio
between the
current picture and the reference picture.
[0168] In the same and other embodiments, the AUC value is used for
identifying the
boundary of AU and used for hypothetical reference decoder (HRD) operation,
which needs
input and output timing with AU granularity. In most cases, the decoded
picture with the
highest layer in an AU may be outputted for display. The AUC value and the
layer id value
can be used for identifying the output picture.
[0169] In an embodiment, a picture may comprise one or more sub-pictures.
Each
sub-picture may cover a local region or the entire region of the picture. The
region supported
by a sub-picture may or may not be overlapped with the region supported by
another sub-
picture. The region composed by one or more sub-pictures may or may not cover
the entire
region of a picture. If a picture consists of a sub-picture, the region
supported by the sub-
picture may be identical to the region supported by the picture.
41
Date recue/date received 2021-10-26
PCTUS2021032924
[0170] In the same embodiment, a sub-picture may be coded by a coding
method
similar to the coding method used for the coded picture. A sub-picture may be
independently
coded or may be coded dependent on another sub-picture or a coded picture. A
sub-picture
may or may not have any parsing dependency from another sub-picture or a coded
picture.
[0171] In the same embodiment, a coded sub-picture may be contained in one
or more
layers. A coded sub-picture in a layer may have a different spatial
resolution. The original
sub-picture may be spatially re-sampled (up-sampled or down-sampled), coded
with different
spatial resolution parameters, and contained in a bitstream corresponding to a
layer.
[0172] In the same or another embodiment, a sub-picture with (W, H), where
W
indicates the width of the sub-picture and H indicates the height of the sub-
picture,
respectively, may be coded and contained in the coded bitstream corresponding
to layer 0,
while the up-sampled (or down-sampled) sub-picture from the sub-picture with
the original
spatial resolution, with (W*S,k, H* Sh,k), may be coded and contained in the
coded bitstream
corresponding to layer k, where Sw,k, Sh,k indicate the resampling ratios,
horizontally and
vertically. If the values of Sw,k, Sh,k are greater than 1, the resampling is
equal to the up-
sampling. Whereas, if the values of Sw,k, Sh,k are smaller than 1, the
resampling is equal to the
down-sampling.
[0173] In the same or another embodiment, a coded sub-picture in a layer
may have a
different visual quality from that of the coded sub-picture in another layer
in the same sub-
picture or different sub-picture. For example, sub-picture i in a layer, n, is
coded with the
quantization parameter, Qi,n, while a sub-picture j in a layer, m, is coded
with the quantization
parameter, Q,..
[0174] In the same or another embodiment, a coded sub-picture in a layer
may be
independently decodable, without any parsing or decoding dependency from a
coded sub-
picture in another layer of the same local region. The sub-picture layer,
which can be
42
Date recue/date received 2021-10-26
PCTU S2021032924
independently decodable without referencing another sub-picture layer of the
same local
region, is the independent sub-picture layer. A coded sub-picture in the
independent sub-
picture layer may or may not have a decoding or parsing dependency from a
previously
coded sub-picture in the same sub-picture layer, but the coded sub-picture may
not have any
dependency from a coded picture in another sub-picture layer.
[0175] In the same or another embodiment, a coded sub-picture in a layer
may be
dependently decodable, with any parsing or decoding dependency from a coded
sub-picture
in another layer of the same local region. The sub-picture layer, which can be
dependently
decodable with referencing another sub-picture layer of the same local region,
is the
dependent sub-picture layer. A coded sub-picture in the dependent sub-picture
may reference
a coded sub-picture belonging to the same sub-picture, a previously coded sub-
picture in the
same sub-picture layer, or both reference sub-pictures.
[0176] In the same or another embodiment, a coded sub-picture comprises one
or
more independent sub-picture layers and one or more dependent sub-picture
layers. However,
at least one independent sub-picture layer may be present for a coded sub-
picture. The
independent sub-picture layer may have the value of the layer identifier (e.g.
syntax element
layer id), which may be present in NAL unit header or another high-level
syntax structure,
equal to 0. The sub-picture layer with the layer id equal to 0 may be the base
sub-picture
layer.
[0177] In the same or another embodiment, a picture may comprise one or
more
foreground sub-pictures and one background sub-picture. The region supported
by a
background sub-picture may be equal to the region of the picture. The region
supported by a
foreground sub-picture may be overlapped with the region supported by a
background sub-
picture. The background sub-picture may be a base sub-picture layer, while the
foreground
sub-picture may be a non-base (enhancement) sub-picture layer. One or more non-
base sub-
43
Date recue/date received 2021-10-26
PCTUS2021032924
picture layers may reference the same base layer for decoding. Each non-base
sub-picture
layer with layer id equal to a may reference a non-base sub-picture layer with
layer id equal
to b, where a is greater than b.
[0178] In the same or another embodiment, a picture may comprise one or
more
foreground sub-pictures with or without a background sub-picture. Each sub-
picture may
have its own base sub-picture layer and one or more non-base (enhancement)
layers. Each
base sub-picture layer may be referenced by one or more non-base sub-picture
layers. Each
non-base sub-picture layer with layer id equal to a may reference a non-base
sub-picture
layer with layer id equal to b, where a is greater than b.
[0179] In the same or another embodiment, a picture may comprise one or
more
foreground sub-pictures with or without a background sub-picture. Each coded
sub-picture in
a (base or non-base) sub-picture layer may be referenced by one or more non-
base layer sub-
pictures belonging to the same sub-picture and one or more non-base layer sub-
pictures,
which are not belonging to the same sub-picture.
[0180] In the same or another embodiment, a picture may comprise one or
more
foreground sub-pictures with or without a background sub-picture. A sub-
picture in a layer a
may be further partitioned into multiple sub-pictures in the same layer. One
or more coded
sub-pictures in a layer b may reference the partitioned sub-picture in a layer
a.
[0181] In the same or another embodiment, a coded video sequence (CVS) may
be a
group of the coded pictures. The CVS may comprise of one or more coded sub-
picture
sequences (CSPS), where the CSPS may be a group of coded sub-pictures covering
the same
local region of the picture. A CSPS may have the same or a different temporal
resolution than
that of the coded video sequence.
[0182] In the same or another embodiment, a CSPS may be coded and contained
in
one or more layers. A CSPS may comprise or consist of one or more CSPS layers.
Decoding
44
Date recue/date received 2021-10-26
PCTUS2021032924
one or more CSPS layers corresponding to a CSPS may reconstruct a sequence of
sub-
pictures corresponding to the same local region.
[0183] In the same or another embodiment, the number of CSPS layers
corresponding
to a CSPS may be identical to or different from the number of CSPS layers
corresponding to
another CSPS.
[0184] In the same or another embodiment, a CSPS layer may have a different
temporal resolution (e.g. frame rate) from another CSPS layer. The original
(uncompressed)
sub-picture sequence may be temporally re-sampled (up-sampled or down-
sampled), coded
with different temporal resolution parameters, and contained in a bitstream
corresponding to
a layer.
[0185] In the same or another embodiment, a sub-picture sequence with the
frame
rate, F, may be coded and contained in the coded bitstream corresponding to
layer 0, while
the temporally up-sampled (or down-sampled) sub-picture sequence from the
original sub-
picture sequence, with F* St,k, may be coded and contained in the coded
bitstream
corresponding to layer k, where St,k indicates the temporal sampling ratio for
layer k. If the
value of St,k is greater than 1, the temporal resampling process is equal to
the frame rate up
conversion. Whereas, if the value of St,k is smaller than 1, the temporal
resampling process is
equal to the frame rate down conversion.
[0186] In the same or another embodiment, when a sub-picture with a CSPS
layer a is
referenced by a sub-picture with a CSPS layer b for motion compensation or any
inter-layer
prediction, if the spatial resolution of the CSPS layer a is different from
the spatial resolution
of the CSPS layer b, decoded pixels in the CSPS layer a are resampled and used
for
reference. The resampling process may need an up-sampling filtering or a down-
sampling
filtering.
Date recue/date received 2021-10-26
PCTUS2021032924
[0187] FIG. 11 shows an example video stream including a background video
CSPS
with layer id equal to 0 and multiple foreground CSPS layers. While a coded
sub-picture
may comprise of one or more enhancement CSPS layers (704), a background
region, which
does not belong to any foreground CSPS layer, may comprise a base layer (702).
The base
layer (702) may contain a background region and foreground regions, while an
enhancement
CSPS layer (704) contains a foreground region. An enhancement CSPS layer (704)
may have
a better visual quality than the base layer (702), at the same region. The
enhancement CSPS
layer (704) may reference the reconstructed pixels and the motion vectors of
the base layer
(702), corresponding to the same region.
[0188] In the same or another embodiment, the video bitstream corresponding
to a
base layer (702) is contained in a track, while the CSPS layers (704)
corresponding to each
sub-picture are contained in a separated track, in a video file.
[0189] In the same or another embodiment, the video bitstream corresponding
to a
base layer (702) is contained in a track, while CSPS layers (704) with the
same layer id are
contained in a separated track. In this example, a track corresponding to a
layer k includes
CSPS layers (704) corresponding to the layer k, only.
[0190] In the same or another embodiment, each CSPS layer (704) of each sub-
picture is stored in a separate track. Each track may or may not have any
parsing or decoding
dependency from one or more other tracks.
[0191] In the same or another embodiment, each track may contain bitstreams
corresponding to layer i to layer j of CSPS layers (704) of all or a subset of
sub-pictures,
where 0<i=<j=<k, k being the highest layer of CSPS.
[0192] In the same or another embodiment, a picture comprises or consists
of one or
more associated media data including depth map, alpha map, 3D geometry data,
occupancy
46
Date recue/date received 2021-10-26
PCTUS2021032924
map, etc. Such associated timed media data can be divided to one or multiple
data sub-stream
each of which corresponding to one sub-picture.
[0193] In the same or another embodiment, FIG. 12 shows an example of a
video
conference based on the multi-layered sub-picture method. In a video stream,
one base layer
video bitstream corresponding to the background picture and one or more
enhancement layer
video bitstreams corresponding to foreground sub-pictures are contained. Each
enhancement
layer video bitstream may correspond to a CSPS layer. In a display, the
picture corresponding
to the base layer (712) is displayed by default. The base layer (712) may
contain one or more
user's picture in a picture (PIP). When a specific user is selected by a
client's control, the
enhancement CSPS layer (714) corresponding to the selected user is decoded and
displayed
with the enhanced quality or spatial resolution.
[0194] FIG. 13 illustrates a diagram for operation of the embodiment. In
the
embodiment, a decoder may decode the video bitstream that includes multiple
layers such as,
for example, one base layer and one or more enhancement CSPS layers (722).
Following, the
decoder may identify the background region and one or more foreground sub-
pictures (724)
and make a decision as to whether a specific sub-picture region is selected
(726). If a specific
sub-picture region corresponding to, for example, a user's PIP is selected
(YES), the decoder
may decode and display the enhanced sub-picture corresponding to the selected
user (728).
For example, the decoder may decode and display the image corresponding to the
enhancement CSPS layer (714). If no specific sub-picture region is selected
(NO), the
decoder may decode and display the background region (730). For example, the
decoder may
decode and display the image corresponding to the base layer (712).
[0195] In the same or another embodiment, a network middle box (such as
router)
may select a subset of layers to send to a user depending on its bandwidth.
The picture/sub-
picture organization may be used for bandwidth adaptation. For instance, if
the user does not
47
Date recue/date received 2021-10-26
PCTUS2021032924
have the bandwidth, the router strips of layers or selects some sub-pictures
due to their
importance or based on used setup. In an embodiment, such processes may be
done
dynamically to adapt to bandwidth.
[0196] FIG. 14 illustrates an example use case of 360 video. When a
spherical 360
picture (742) is projected onto a planar picture, the spherical 360 picture
(742) that is
projected may be partitioned into multiple sub-pictures (745) as a base layer
(744). An
enhancement layer (746) of a specific one of the sub-pictures (745) may be
coded and
transmitted to a client. A decoder may decode both the base layer (744)
including all sub-
pictures (745) and an enhancement layer (746) of a selected one of the sub-
pictures (745).
When the current viewport is identical to the selected one of the sub-pictures
(745), the
displayed picture may have a higher quality with the decoded sub-picture (745)
with the
enhancement layer (746). Otherwise, the decoded picture with the base layer
(744) can be
displayed with a lower quality.
[0197] In the same or another embodiment, any layout information for
display may be
present in a file as supplementary information (such as SEI message or
metadata). One or
more decoded sub-pictures may be relocated and displayed depending on the
signaled layout
information. The layout information may be signaled by a streaming server or a
broadcaster,
or may be regenerated by a network entity or a cloud server, or may be
determined by a
user's customized setting.
[0198] In an embodiment, when an input picture is divided into one or more
(rectangular) sub-region(s), each sub-region may be coded as an independent
layer. Each
independent layer corresponding to a local region may have a unique layer id
value. For each
independent layer, the sub-picture size and location information may be
signaled. For
example, picture size (width, height) and offset information of the left-top
corner (x offset,
y offset) may be signaled. FIG. 15A illustrates an example of the layout of
divided sub-
48
Date recue/date received 2021-10-26
PCTUS2021032924
pictures (752), FIG. 15B illustrates an example of a corresponding sub-picture
size and
position information of one of the sub-pictures (752), and FIG. 16 illustrates
the
corresponding picture prediction structure. The layout information including
the sub-picture
size(s) and the sub-picture position(s) may be signaled in a high-level syntax
structure, such
as parameter set(s), header of slice or tile group, or SEI message.
[0199] In the same embodiment, each sub-picture corresponding to an
independent
layer may have its unique POC value within an AU. When a reference picture
among pictures
stored in DPB is indicated by using syntax element(s) in RPS or RPL structure,
the POC
value(s) of each sub-picture corresponding to a layer may be used.
[0200] In the same or another embodiment, in order to indicate the (inter-
layer)
prediction structure, the layer id may not be used and the POC (delta) value
may be used.
[0201] In the same embodiment, a sub-picture with a POC value equal to N
corresponding to a layer (or a local region) may or may not be used as a
reference picture of a
sub-picture with a POC value equal to K+N, corresponding to the same layer (or
the same
local region) for motion compensated prediction. In most cases, the value of
the number K
may be equal to the maximum number of (independent) layers, which may be
identical to the
number of sub-regions.
[0202] In the same or another embodiment, FIGS. 17-18 illustrate an
extended case of
FIGS. 15A-B and FIG. 16. When an input picture is divided into multiple (e.g.
four) sub-
regions, each local region may be coded with one or more layers. In the case,
the number of
independent layers may be equal to the number of sub-regions, and one or more
layers may
correspond to a sub-region. Thus, each sub-region may be coded with one or
more
independent layer(s) and zero or more dependent layer(s).
[0203] In the same embodiment, with reference to FIG. 17, the input picture
may be
divided into four sub-regions, including a top-left sub-region (762), a top-
right sub-region
49
Date recue/date received 2021-10-26
PCTUS2021032924
(763), a bottom-left sub-region (764), and a bottom-right sub-region (765).
The top-right sub-
region (763) may be coded as two layers, which are layer 1 and layer 4, while
the bottom-
right sub-region (765) may be coded as two layers, which are layer 3 and layer
5. In this case,
the layer 4 may reference the layer 1 for motion compensated prediction, while
the layer 5
may reference the layer 3 for motion compensation.
[0204] In the same or another embodiment, in-loop filtering (such as
deblocking
filtering, adaptive in-loop filtering, reshaper, bilateral filtering or any
deep-learning based
filtering) across layer boundary may be (optionally) disabled.
[0205] In the same or another embodiment, motion compensated prediction or
intra-
block copy across layer boundary may be (optionally) disabled.
[0206] In the same or another embodiment, boundary padding for motion
compensated prediction or in-loop filtering at the boundary of sub-picture may
be processed
optionally. A flag indicating whether the boundary padding is processed or not
may be
signaled in a high-level syntax structure, such as parameter set(s) (VPS, SPS,
PPS, or APS),
slice or tile group header, or SET message.
[0207] In the same or another embodiment, the layout information of sub-
region(s)
(or sub-picture(s)) may be signaled in VPS or SPS. FIG. 19A shows an example
of syntax
elements in a VPS (770), and FIG. 19B shows an example of syntax elements of
an SPS
(780). In this example, vps sub_picture dividing flag (772) is signaled in VPS
(770). The
flag may indicate whether input picture(s) are divided into multiple sub-
regions or not. When
the value of vps sub_picture dividing flag (772) is equal to 0, the input
picture(s) in the
coded video sequence(s) corresponding to the current VPS may not be divided
into multiple
sub-regions. In this case, the input picture size may be equal to the coded
picture size
(pic width in luma samples (786), pic height in luma samples (788)), which is
signaled
in SPS (680). When the value of vps sub_picture dividing flag (772) is equal
to 1, the input
Date recue/date received 2021-10-26
PCTUS2021032924
picture(s) may be divided into multiple sub-regions. In this case, the syntax
elements
vps full_pic width in luma samples (774) and vps full_pic height in luma
samples (776)
are signaled in VPS (770). The values of vps full_pic width in luma samples
(774) and
vps full_pic height in luma samples (776) may be equal to the width and height
of the
input picture(s), respectively.
[0208] In the same embodiment, the values of vps full_pic width in luma
samples
(774) and vps full_pic height in luma samples (776) may not be used for
decoding, but
may be used for composition and display.
[0209] In the same embodiment, when the value of vps sub_picture dividing
flag
(772) is equal to 1, the syntax elements pic offset x (782) and pic offset y
(784)) may be
signaled in SPS (780), which corresponds to a specific layer(s). In this case,
the coded picture
size (pic width in luma samples (786), pic height in luma samples (788))
signaled in SPS
(780) may be equal to the width and height of the sub-region corresponding to
a specific
layer. Also, the position (pic offset x (782), pic offset_y (784)) of the left-
top corner of the
sub-region may be signaled in SPS (780).
[0210] In the same embodiment, the position information (pic offset x
(782),
pic offset_y (784)) of the left-top corner of the sub-region may not be used
for decoding, but
may be used for composition and display.
[0211] In the same or another embodiment, the layout information (size and
position)
of all or sub-set sub-region(s) of (an) input picture(s), and the dependency
information
between layer(s) may be signaled in a parameter set or an SET message. FIG. 20
illustrates an
example of syntax elements that indicate the information of the layout of sub-
regions, the
dependency between layers, and the relation between a sub-region and one or
more layers. In
this example, the syntax element num sub region (791) indicates the number of
(rectangular)
sub-regions in the current coded video sequence. The syntax element num layers
(792)
51
Date recue/date received 2021-10-26
PCTUS2021032924
indicates the number of layers in the current coded video sequence. The value
of num layers
(792) may be equal to or greater than the value of num sub region (791). When
any sub-
region is coded as a single layer, the value of num layers (792) may be equal
to the value of
num sub region (791). When one or more sub-regions are coded as multiple
layers, the value
of num layers (792) may be greater than the value of num sub region (791). The
syntax
element direct dependency flag[ i ][ ii (793) indicates the dependency from
the j-th layer to
the i-th layer. The syntax element num layers for region[ i ] (794) indicates
the number of
layers associated with the i-th sub-region. The syntax element sub region
layer id[ i ][ j ]
(795) indicates the layer id of the j-th layer associated with the i-th sub-
region. The syntax
elements sub region offset x[ i ] (796) and sub region offset y[ i ] (797)
indicate the
horizontal and vertical location of the left-top corner of the i-th sub-
region, respectively. The
syntax elements sub region width [ i ] (798) and sub region height[ i ] (799)
indicate the
width and height of the i-th sub-region, respectively.
[0212] In one embodiment, one or more syntax elements that specify the
output layer
set to indicate one of more layers to be outputted with or without profile
tier level information
may be signaled in a high-level syntax structure (e.g. VPS, DPS, SPS, PPS,
APS, or SET
message). Referring to FIG. 21, the syntax element num output layer sets (804)
indicating
the number of output layer set (OLS) in a coded video sequence referring to a
VPS may be
signaled in the VPS. For each output layer set, the syntax element output
layer flag (810)
may be signaled as many times as the number of output layers.
[0213] In the same embodiment, the syntax element output layer flag (810)
equal to
1 specifies that the i-th layer is output. The syntax element output layer
flag (810) equal to 0
specifies that the i-th layer is not output.
[0214] In the same or another embodiment, one or more syntax elements that
specify
the profile tier level information for each output layer set may be signaled
in a high-level
52
Date recue/date received 2021-10-26
PCTUS2021032924
syntax structure (e.g. VPS, DPS, SPS, PPS, APS, or SET message). Still
referring to FIG. 21,
the syntax element num_profile tier level (806) indicating the number of
profile tier level
information per OLS in the coded vide sequence referring to the VPS may be
signaled in the
VPS. For each output layer set, a set of syntax elements for profile tier
level information or
an index indicating a specific profile tier level information among entries in
the profile tier
level information may be signaled as many times as the number of output
layers.
[0215] In the same embodiment, the syntax element profile tier level idx[ i
1[j ]
(812) specifies the index, into the list of profile tier level( ) (808) syntax
structures in the
VPS, of the profile tier level( ) (808) syntax structure that applies to the j-
th layer of the i-th
OLS.
[0216] Profiles, tiers, and levels (and corresponding information thereof)
may specify
restrictions on bitstreams and, thus, limits on capabilities needed for
decoding the bitstreams.
Profiles, tiers, and levels (and corresponding information thereof) may also
be used to
indicate interoperability points between individual decoder implementations. A
profile may
be a subset of the entire bitstream syntax of, for example, a standard. Each
profile (and
corresponding information thereof) may specify a subset of algorithmic
features and limits
that may be supported by all decoders conforming to the profile. Tiers and
levels may be
specified within each profile, and a level of a tier may be a specified set of
constraints
imposed on values of the syntax elements in the bitstream. Each level of a
tier (and
corresponding information thereof) may specify a set of limits on the values
and/or limits on
arithmetic combinations of values that may be taken by the syntax elements of
the disclosure.
The same set of tier and level definitions may be used with all profiles, but
individual
implementations may support a different tier and within a tier a different
level for each
supported profile. For any given profile, a level of a tier may correspond to
a particular
53
Date recue/date received 2021-10-26
PCTUS2021032924
decoder processing load and memory capability. A level specified for a lower
tier may be
more constrained than a level specified for a higher tier.
[0217] In the same or another embodiment, referring to FIG. 22, the syntax
elements
num_profile tier level (806) and/or num output layer sets (804) may be
signaled when the
number of maximum layers is greater than 1 (vps max layers minusl > 0).
[0218] In the same or another embodiment, referring to FIG. 22, the syntax
element
vps output layers mode[ i ] (822) indicating the mode of output layer
signaling for the i-th
output layer set may be present in VPS.
[0219] In the same embodiment, the syntax element vps output layers mode[ i
]
(822) equal to 0 specifies that only the highest layer is output with the i-th
output layer set.
The syntax element vps output layers mode[ ii (822) equal to 1 specifies that
all layers are
output with the i-th output layer set. The syntax element vps output layers
mode[ i ] (822)
equal to 2 specifies that the layers that are output are the layers with
vps output layer flag[ i ][ j ] equal to 1 with the i-th output layer set.
More values may be
reserved.
[0220] In the same embodiment, the syntax element output layer flag[ i ][ j
] (810)
may or may not be signaled depending on the value of the syntax element
vps output layers mode[ ii (822) for the i-th output layer set.
[0221] In the same or another embodiment, referring to FIG. 22, the flag
vps_ptl signal flag[ i ] (824) may be present for the i-th output layer set.
Depending on the
value of vps_ptl signal flag[ i ] (824), the profile tier level information
for the i-th output
layer set may or may not be signaled.
[0222] In the same or another embodiment, referring to FIG. 23, the number
of sub-
pictures, max subpics minusl, in the current CVS may be signalled in a high-
level syntax
structure( e.g. VPS, DPS, SPS, PPS, APS, or SEI message).
54
Date recue/date received 2021-10-26
PCTUS2021032924
[0223] In the same embodiment, referring to FIG. 23, the sub-picture
identifier,
sub_pic id[i] (821), for the i-th sub-picture may be signalled, when the
number of sub-
pictures is greater than 1 ( max subpics minusl > 0).
[0224] In the same or another embodiment, one or more syntax elements
indicating
the sub-picture identifier belonging to each layer of each output layer set
may be signalled in
VPS. Referring to FIG. 23, the identifier sub_pic id layer[i][j][k] (826)
indicates the k-th
sub-picture present in the j-th layer of the i-th output layer set. By using
the information of
the identifier sub_pic id layer[i][j][k] (826), a decoder may recognize which
sub-picture
may be decoded and outputtted for each layer of a specific output layer set.
[0225] In an embodiment, picture header (PH) is a syntax structure
containing syntax
elements that apply to all slices of a coded picture. A picture unit (PU) is a
set of NAL units
that are associated with each other according to a specified classification
rule, are consecutive
in decoding order, and contain exactly one coded picture. A PU may contain a
picture header
(PH) and one or more video coding layer (VCL) NAL units composing a coded
picture.
[0226] In an embodiment, an SPS (RBSP) may be available to the decoding
process
prior to being referenced, by being included in at least one AU with
Temporalld equal to 0 or
provided through external means.
[0227] In an embodiment, an SPS (RBSP) may be available to the decoding
process
prior to being referenced, by being included in at least one AU with
Temporalld equal to 0 in
the CVS, which contains one or more PPS referring to the SPS, or provided
through external
means.
[0228] In an embodiment, an SPS (RBSP) may be available to the decoding
process
prior to being referenced by one or more PPS, by being included in at least
one PU with
nuh layer id equal to the lowest nuh layer id value of the PPS NAL units that
refer to the
Date recue/date received 2021-10-26
PCTUS2021032924
SPS NAL unit in the CVS, which contains one or more PPS referring to the SPS,
or provided
through external means.
[0229] In an embodiment, an SPS (RBSP) may be available to the decoding
process
prior to being referenced by one or more PPS, by being included in at least
one PU with
TemporalId equal to 0 and nuh layer id equal to the lowest nuh layer id value
of the PPS
NAL units that refer to the SPS NAL unit or provided through external means.
[0230] In an embodiment, an SPS (RBSP) may be available to the decoding
process
prior to being referenced by one or more PPS, by being included in at least
one PU with
Temporalld equal to 0 and nuh layer id equal to the lowest nuh layer id value
of the PPS
NAL units that refer to the SPS NAL unit in the CVS, which contains one or
more PPS
referring to the SPS, or provided through external means or provided through
external means.
[0231] In the same or another embodiment, an identifier pps seq parameter
set id
specifies the value of an identifier sps seq parameter set id for the
referenced SPS. The
value of the identifier pps seq parameter set id may be the same in all PPSs
that are
referred to by coded pictures in a coded layer video sequence (CLVS).
[0232] In the same or another embodiment, all SPS NAL units with a
particular value
of the identifier sps seq parameter set id in a CVS may have the same content.
[0233] In the same or another embodiment, regardless of the nuh layer id
values,
SPS NAL units may share the same value space of the identifier sps seq
parameter set id.
[0234] In the same or another embodiment, the nuh layer id value of an SPS
NAL
unit may be equal to the lowest nuh layer id value of the PPS NAL units that
refer to the
SPS NAL unit.
[0235] In an embodiment, when an SPS with nuh layer id equal to m is
referred to
by one or more PPS with nuh layer id equal to n, the layer with nuh layer id
equal to m
56
Date recue/date received 2021-10-26
PCTUS2021032924
may be the same as the layer with nuh layer id equal to n or a (direct or
indirect) reference
layer of the layer with nuh layer id equal to m.
[0236] In an embodiment, a PPS (RBSP) may be available to the decoding
process
prior to being referenced, by being included in at least one AU with
Temporalld equal to the
Temporalld of the PPS NAL unit or provided through external means.
[0237] In an embodiment, a PPS (RBSP) may be available to the decoding
process
prior to being referenced, by being included in at least one AU with
Temporalld equal to the
Temporalld of the PPS NAL unit in the CVS, which contains one or more PHs (or
coded
slice NAL units) referring to the PPS, or provided through external means.
[0238] In an embodiment, a PPS (RBSP) may be available to the decoding
process
prior to being referenced by one or more PHs (or coded slice NAL units), by
being included
in at least one PU with nuh layer id equal to the lowest nuh layer id value of
the coded slice
NAL units that refer to the PPS NAL unit in the CVS, which contains one or
more PHs (or
coded slice NAL units) referring to the PPS, or provided through external
means.
[0239] In an embodiment, a PPS (RBSP) may be available to the decoding
process
prior to it being referenced by one or more PHs (or coded slice NAL units), by
being included
in at least one PU with Temporalld equal to the Temporalld of the PPS NAL unit
and
nuh layer id equal to the lowest nuh layer id value of the coded slice NAL
units that refer
to the PPS NAL unit in the CVS, which contains one or more PHs (or coded slice
NAL units)
referring to the PPS, or provided through external means.
[0240] In the same or another embodiment, an identifier ph_pic_parameter
set id in
PH specifies the value of an identifier pps_pic_parameter set id for the
referenced PPS in
use. The value of pps seq parameter set id may be the same in all PPSs that
are referred to
by coded pictures in a CLVS.
57
Date recue/date received 2021-10-26
PCTUS2021032924
[0241] In the same or another embodiment, all PPS NAL units with a
particular value
of the identifier pps_pic_parameter set id within a PU may have the same
content.
[0242] In the same or another embodiment, regardless of the nuh layer id
values,
PPS NAL units may share the same value space of the identifier
pps_pic_parameter set id.
[0243] In the same or another embodiment, the nuh layer id value of a PPS
NAL
unit may be equal to the lowest nuh layer id value of the coded slice NAL
units that refer to
the NAL unit that refers to the PPS NAL unit.
[0244] In an embodiment, when a PPS with nuh layer id equal to m is
referred to by
one or more coded slice NAL units with nuh layer id equal to n, the layer with
nuh layer id
equal to m may be the same as the layer with nuh layer id equal to n or a
(direct or indirect)
reference layer of the layer with nuh layer id equal to m.
[0245] In an embodiment, a PPS (RBSP) may be available to the decoding
process
prior to it being referenced, by being included in at least one AU with
Temporalld equal to
the Temporalld of the PPS NAL unit or provided through external means.
[0246] In an embodiment, a PPS (RBSP) may be available to the decoding
process
prior to being referenced, by being included in at least one AU with
Temporalld equal to the
Temporalld of the PPS NAL unit in the CVS, which contains one or more PHs (or
coded
slice NAL units) referring to the PPS, or provided through external means.
[0247] In an embodiment, a PPS (RBSP) may be available to the decoding
process
prior to being referenced by one or more PHs (or coded slice NAL units), by
being included
in at least one PU with nuh layer id equal to the lowest nuh layer id value of
the coded slice
NAL units that refer to the PPS NAL unit in the CVS, which contains one or
more PHs (or
coded slice NAL units) referring to the PPS, or provided through external
means.
[0248] In an embodiment, a PPS (RBSP) may be available to the decoding
process
prior to being referenced by one or more PHs (or coded slice NAL units), by
being included
58
Date recue/date received 2021-10-26
PCTUS2021032924
in at least one PU with TemporalId equal to the TemporalId of the PPS NAL unit
and
nuh layer id equal to the lowest nuh layer id value of the coded slice NAL
units that refer
to the PPS NAL unit in the CVS, which contains one or more PHs (or coded slice
NAL units)
referring to the PPS, or provided through external means.
[0249] In the same or another embodiment, an identifier ph_pic_parameter
set id in
PH specifies the value of an identifier pps_pic_parameter set id for the
referenced PPS in
use. The value of the identifier pps seq parameter set id may be the same in
all PPSs that
are referred to by coded pictures in a CLVS.
[0250] In the same or another embodiment, all PPS NAL units with a
particular value
of pps_pic_parameter set id within a PU may have the same content.
[0251] In the same or another embodiment, regardless of the nuh layer id
values,
PPS NAL units may share the same value space of the identifier
pps_pic_parameter set id.
[0252] In the same or another embodiment, the nuh layer id value of a PPS
NAL
unit may be equal to the lowest nuh layer id value of the coded slice NAL
units that refer to
the NAL unit that refer to the PPS NAL unit.
[0253] In an embodiment, when a PPS with nuh layer id equal to m is
referred to by
one or more coded slice NAL units with nuh layer id equal to n, the layer with
nuh layer id
equal to m may be the same as the layer with nuh layer id equal to n or a
(direct or indirect)
reference layer of the layer with nuh layer id equal to m.
[0254] An output layer may be a layer of an output layer set that is
output. An output
layer set (OLS) may be a set of layers that is specified, where one or more
layers in the set of
layers are specified to be output layers. An output layer set (OLS) layer
index is an index, of
a layer in an OLS, to the list of layers in the OLS.
[0255] A sublayer may be a temporal scalable layer of a temporal scalable
bitstream,
of the sublayer including VCL NAL units with a particular value of the
Temporalld variable
59
Date recue/date received 2021-10-26
PCTUS2021032924
and the associated non-VCL NAL units. A sublayer representation may be a
subset of the
bitstream that includes NAL units of a particular sublayer and the lower
sublayers.
[0256] A VPS RBSP may be available to the decoding process prior to being
referenced, by being included in at least one AU with TemporalId equal to 0 or
provided
through external means. All VPS NAL units with a particular value of
vps video_parameter set id in a CVS may have the same content.
[0257] With reference to FIGS. 24-25, syntax elements of example VPS RBSPs
are
described below.
[0258] The syntax element vps video_parameter set id (842) provides an
identifier
for the VPS for reference by other syntax elements. The value of the syntax
element
vps video_parameter set id (842) may be greater than 0.
[0259] The syntax element vps max layers minusl (802) plus 1 specifies the
maximum allowed number of layers in each CVS referring to the VPS.
[0260] The syntax element vps max sublayers minusl (846) plus 1 specifies
the
maximum number of temporal sublayers that may be present in a layer in each
CVS referring
to the VPS. The value of the syntax element vps max sublayers minusl (846) may
be in the
range of 0 to 6, inclusive.
[0261] The syntax element vps all layers same num sublayers flag (848)
equal to
1 specifies that the number of temporal sublayers is the same for all the
layers in each CVS
referring to the VPS. The syntax element vps all layers same num sublayers
flag (848)
equal to 0 specifies that the layers in each CVS referring to the VPS may or
may not have the
same number of temporal sublayers. When not present, the value of
vps all layers same num sublayers flag (848) may be inferred to be equal to 1.
[0262] The syntax element vps all independent layers flag (850) equal to 1
specifies that all layers in the CVS are independently coded without using
inter-layer
Date recue/date received 2021-10-26
PCTUS2021032924
prediction. The syntax element vps all independent layers flag (850) equal to
0 specifies
that one or more of the layers in the CVS may use inter-layer prediction. When
not present,
the value of vps all independent layers flag (850) may be inferred to be equal
to 1.
[0263] The syntax element vps layer id[ ii (852) specifies the nuh layer
id value of
the i-th layer. For any two non-negative integer values of m and n, when m is
less than n, the
value of vps layer id[ ml may be less than vps layer id[ n ].
[0264] The syntax element vps independent layer flag[ i ] (854) equal to 1
specifies
that the layer with index i does not use inter-layer prediction. The syntax
element
vps independent layer flag[ i ] (854) equal to 0 specifies that the layer with
index i may use
inter-layer prediction and the syntax elements vps direct ref layer flag[ i ][
j ] for j in the
range of 0 to i ¨ 1, inclusive, are present in VPS. When not present, the
value of the syntax
element vps independent layer flag[ i ] (854) may be inferred to be equal to
1.
[0265] The syntax element vps direct ref layer flag[ i ][j 1 (856) equal
to 0
specifies that the layer with index j is not a direct reference layer for the
layer with index i.
The syntax element vps direct ref layer flag [ i ][ j ] (856) equal to 1
specifies that the layer
with index j is a direct reference layer for the layer with index i. When the
syntax element
vps direct ref layer flag[ i ][ j ] (856) is not present for i and j in the
range of 0 to
vps max layers minusl, inclusive, is the syntax element may be inferred to be
equal to 0.
When the syntax element vps independent layer flag[ i ] (854) is equal to 0,
there may be at
least one value of j in the range of 0 to i ¨ 1, inclusive, such that the
value of the syntax
element vps direct ref layer flag[ i ][ j ] (856) is equal to 1.
[0266] The variables NumDirectRefLayers[ i ],
DirectRefLayerIdx[ i ][ d ],
NumRefLayers[ i ], RefLayerIdx[ i ][ r ], and LayerUsedAsRefLayerFlag[ j ] may
be derived
as follows:
61
Date recue/date received 2021-10-26
PCTUS2021032924
for( i = 0; i <= vps max layers minusl; i++) {
for( j = 0; j <= vps max layers minusl; j++) {
dependencyFlag[ i ][ j ] = vps direct ref layer flag[ i ][ j ]
for( k = 0; k < i; k++ )
if( vps direct ref layer flag[ i ][ k ] && dependencyFlag[ k ][ j I)
dependencyFlag[ i ][ j ] = 1
1
LayerUsedAsRefLayerFlag[ ii = 0
I
for( i = 0; i <= vps max layers minusl; i++) {
for( j = 0, d = 0, r = 0;j <= vps max layers minusl; j++ ) { (37)
if( vps direct ref layer flag[ i ][ j ] ) {
DirectRefLayerIdx[ i ][ d++ ] = j
LayerUsedAsRefLayerFlag[ j ] = 1
1
if( dependencyFlag[ i ][ j I)
RefLayerIdx[ i ][ r++ ] =j
1
NumDirectRefLayers[ i ] = d
NumRefLayers[ i ] = r
1
[0267] The variable GeneralLayerIdx[ ii, specifying the layer index of the
layer with
nuh layer id equal to vps layer id[ ii (852), may be derived as follows:
for( i = 0; i <= vps max layers minusl; i++) (38)
GeneralLayerIdx[ vps layer id[ ii ] = i
[0268] For any two different values of i and j, both in the range of 0 to
vps max layers minusl (846), inclusive, when dependencyFlag[ i ][ j ] equal to
1, it may be
a requirement of bitstream conformance that the values of chroma format idc
and
62
Date recue/date received 2021-10-26
PCTUS2021032924
bit depth minus8 that apply to the i-th layer may be equal to the values of
chroma format idc and bit depth minus8, respectively, that apply to the j-th
layer.
[0269] The
syntax element max tid ref_present flag[ ii (858) equal to 1 specifies
that the syntax element max tid il [iiref_pics_plus1 (860) is
present. The syntax element
max tid ref_present flag[ ii (858) equal to 0 specifies that the syntax
element
max tid il [iiref_pics_plusl (860) is not present.
[0270] The syntax element max tid il [iiref_pics_plusl (860)
equal to 0 specifies
that inter-layer prediction is not used by non-TRAP pictures of the i-th
layer. The syntax
element max tid il ref_pics_plusl[ ii (860) greater than 0 specifies that, for
decoding
pictures of the i-th layer, no picture with Temporalld greater than
max tid il [iiref_pics_plusl ¨ 1 is
used as inter-layer reference picture (ILRP). When not
present, the value of the syntax element max tid il ref_pics_plusl[ i ] (860)
may be inferred
to be equal to 7.
[0271] The
syntax element each layer is an ols flag (862) equal to 1 specifies that
each OLS contains only one layer and each layer itself in a CVS referring to
the VPS is an
OLS with the single included layer being the only output layer. The syntax
element
each layer is an ols flag (862) equal to 0 specifies that an OLS may contain
more than one
layer. If the syntax element vps max layers minusl is equal to 0, the value of
the syntax
element each layer is an ols flag (862) may be inferred to be equal to 1.
Otherwise, when
the syntax element vps all independent layers flag (854) is equal to 0, the
value of the
syntax element each layer is an ols flag (862) may be inferred to be equal to
0.
[0272] The
syntax element ols mode idc (864) equal to 0 specifies that the total
number of OLSs specified by the VPS is equal to vps max layers minusl + 1, the
i-th OLS
includes the layers with layer indices from 0 to i, inclusive, and for each
OLS only the
highest layer in the OLS is output.
63
Date recue/date received 2021-10-26
PCTUS2021032924
[0273] The syntax element ols mode idc (864) equal to 1 specifies that the
total
number of OLSs specified by the VPS is equal to vps max layers minusl + 1, the
i-th OLS
includes the layers with layer indices from 0 to i, inclusive, and for each
OLS all layers in the
OLS are output.
[0274] The syntax element ols mode idc (864) equal to 2 specifies that the
total
number of OLSs specified by the VPS is explicitly signalled and for each OLS
the output
layers are explicitly signalled and other layers are the layers that are
direct or indirect
reference layers of the output layers of the OLS.
[0275] The value of the syntax element ols mode idc (864) may be in the
range of 0
to 2, inclusive. The value 3 of the syntax element ols mode idc (864) may be
reserved for
future use by ITU-T 1 ISO/IEC.
[0276] When the syntax element vps all independent layers flag (850) is
equal to 1
and each layer is an ols flag (862) is equal to 0, the value of the syntax
element
ols mode idc (864) may be inferred to be equal to 2.
[0277] The syntax element num output layer sets minusl (866) plus 1
specifies the
total number of OLSs specified by the VPS when the syntax element ols mode idc
(864) is
equal to 2.
[0278] The variable TotalNumOlss, specifying the total number of OLSs
specified by
the VPS, may be derived as follows:
if( vps max layers minusl = = 0)
TotalNumOlss = 1
else if( each layer is an ols flag 11 ols mode idc = = 011 ols mode idc = = 1)
TotalNumOlss = vps max layers minusl + 1
else if( ols mode idc = = 2)
TotalNumOlss = num output layer sets minusl + 1
64
Date recue/date received 2021-10-26
PCTUS2021032924
[0279] The syntax element ols output layer flag[ i 1[j 1(868) equal to 1
specifies that
the layer with nuh layer id equal to vps layer id[ j ] is an output layer of
the i-th OLS when
ols mode idc (864) is equal to 2. The syntax element ols output layer flag[ i
][ j ] (868)
equal to 0 specifies that the layer with nuh layer id equal to vps layer id[ j
] is not an output
layer of the i-th OLS when the syntax element ols mode idc (864) is equal to
2.
[0280] The variable NumOutputLayersInOls[ ii, specifying the number of
output
layers in the i-th OLS, the variable NumSubLayersInLayerInOLS[ i ][ j ],
specifying the
number of sublayers in the j-th layer in the i-th OLS, the variable
OutputLayerIdInOls[ i ][ j ],
specifying the nuh layer id value of the j-th output layer in the i-th OLS,
and the variable
LayerUsedAsOutputLayerFlag[ k ], specifying whether the k-th layer is used as
an output
layer in at least one OLS, may be derived as follows:
NumOutputLayersInOls[ 0 ] = 1
OutputLayerIdInOls[ 0 ][ 0 ] = vps layer id[ 0 ]
NumSubLayersInLayerInOLS[ 0 ][ 0 ] = vps max sub layers minusl + 1
LayerUsedAsOutputLayerFlag[ 0 ] = 1
for( i = 1, i <= vps max layers minusl; i++) {
if( each layer is an ols flag 11 ols mode idc < 2)
LayerUsedAsOutputLayerFlag[ ii = 1
else /*( !each layer is an ols flag && ols mode idc = = 2) */
LayerUsedAsOutputLayerFlag[ ii = 0
I
for( i = 1; i < TotalNumOlss; i++)
if( each layer is an ols flag 11 ols mode idc = = 0) {
NumOutputLayersInOls[ ii = 1
OutputLayerIdInOls[ i ][ 0 ] = vps layer id[ ii
for( j = 0; j <i && ( ols mode idc = = 0 ); j++)
NumSubLayersInLayerInOLS[ i 1[i ] = max tid il ref_pics_plusl[ ii
NumSubLayersInLayerInOLS[ i ][ i ] = vps max sub layers minusl + 1
1 else if( ols mode idc = = 1) {
Date recue/date received 2021-10-26
PCTUS2021032924
NumOutputLayersInOls[ ii = i + 1
for( j = 0; j < NumOutputLayersInOls[ ii; j++ ) {
OutputLayerIdInOls[ i ][ j ] = vps layer id[ j ]
NumSubLayersInLayerInOLS[ i ][ j ] = vps max sub layers minusl + 1
1
1 else if( ols mode idc = = 2) {
for( j = 0; j <= vps max layers minusl; j++) {
layerIncludedInOlsFlag[ i ][ j ] = 0
NumSubLayersInLayerInOLS[ i ][ j ] = 0
I
for( k = 0, j = 0; k <= vps max layers minusl; k++) (40)
if( ols output layer flag[ i ][ k I) {
layerIncludedInOlsFlag[ i ][ k ] = 1
LayerUsedAsOutputLayerFlag[ k ] = 1
OutputLayerIdx[ i ][ j ] = k
OutputLayerIdInOls[ i ][ j++ ] = vps layer id[ k ]
NumSubLayersInLayerInOLS[ i ][ j ] =
vps max sub layers minusl + 1
1
NumOutputLayersInOls[ ii = j
for( j = 0;j < NumOutputLayersInOls[ i ]; j++ ) {
idx = OutputLayerIdx[ i ][ j ]
for( k = 0; k < NumRefLayers[ idx ]; k++) {
layerIncludedInOlsFlag[ i ][ RefLayerIdx[ idx ][ k II = 1
if( NumSubLayersInLayerInOLS[ i ][ RefLayerIdx[ idx ][ k II <
max tid il ref_pics_plusl [ OutputLayerIdInOls[ i ][ j II)
NumSubLayersInLayerInOLS[ i ][ RefLayerIdx[ idx ][ k II =
max tid il ref_pics_plusl [ OutputLayerIdInOls[ i ][ j II
1
1
1
66
Date recue/date received 2021-10-26
PCTUS2021032924
[0281] For each value of i in the range of 0 to vps max layers minusl,
inclusive, the
values of LayerUsedAsRefLayerFlag[ i ] and LayerUsedAsOutputLayerFlag[ ii may
not be
both equal to 0. In other words, there may be no layer that is neither an
output layer of at least
one OLS nor a direct reference layer of any other layer.
[0282] For each OLS, there may be at least one layer that is an output
layer. In other
words, for any value of i in the range of 0 to TotalNumOlss ¨ 1, inclusive,
the value of
NumOutputLayersInOls[ ii may be greater than or equal to 1.
[0283] The variable NumLayersInOls[ ii, specifying the number of layers in
the i-th
OLS, and the variable LayerIdInOls[ i ][ ii, specifying the nuh layer id value
of the j-th
layer in the i-th OLS, may be derived as follows:
NumLayersInOls[ 0 ] = 1
LayerIdInOls[ 0 ][ 0 ] = vps layer id[ 0 ]
for( i = 1; i < TotalNumOlss; i++) {
if( each layer is an ols flag ) {
NumLayersInOls[ ii = 1
LayerIdInOls[ i ][ 0 ] = vps layer id[ ii
1 else if( ols mode idc = = 011 ols mode idc = = 1) {
NumLayersInOls[ ii = i + 1
for( j = 0; j < NumLayersInOls[ ii; j++ )
LayerIdInOls[ i ][ j ] = vps layer id[ j ]
1 else if( ols mode idc = = 2) {
for( k = 0, j = 0; k <= vps max layers minusl; k++)
if( layerIncludedInOlsFlag[ i ][ k I)
LayerIdInOls[ i ][ j++ ] = vps layer id[ k ]
NumLayersInOls[ ii =i
1
1
67
Date recue/date received 2021-10-26
PCTUS2021032924
[0284] The variable 01sLayerIdx[ i ][ j ], specifying the OLS layer index
of the layer
with nuh layer id equal to LayerIdInOls[ i ][ j ], is derived as follows:
for( i = 0; i < TotalNumOlss; i++)
for j = 0; j < NumLayersInOls[ ii; j++ )
01sLayerIdx[ i ][ LayerIdInOls[ i ][ j ] ] = j
[0285] The lowest layer in each OLS may be an independent layer. In other
words,
for each i in the range of 0 to TotalNumOlss ¨ I, inclusive, the value of
vps independent layer flag[ GeneralLayerIdx[ LayerIdInOls[ i ][ 0 ] ] ] may be
equal to 1.
Each layer may be included in at least one OLS specified by the VPS. In other
words, for
each layer with a particular value of nuh layer id, nuhLayerId equal to one of
vps layer id[ k ] for k in the range of 0 to vps max layers minusl, inclusive,
there may be
at least one pair of values of i and j, where i is in the range of 0 to
TotalNumOlss ¨ 1,
inclusive, and j is in the range of NumLayersInOls[ ii ¨ 1, inclusive, such
that the value of
LayerIdInOls[ i 1[i ] is equal to nuhLayerId.
[0286] In an embodiment, a decoding process may operate as follows for a
current
picture (e.g. sytax element CurrPic) to set the syntax element
PictureOutputFlag:
[0287] PictureOutputFlag is set equal to 0 if one of the following
conditions is true:
(1) the current picture is a RASL picture and NoOutputBeforeRecoveryFlag of
the
associated TRAP picture is equal to 1;
(2) gdr enabled flag is equal to 1 and the current picture is a GDR picture
with
NoOutputBeforeRecoveryFlag equal to 1;
(3) gdr enabled flag is equal to 1, the current picture is associated with a
GDR
picture with NoOutputBeforeRecoveryFlag equal to 1, and PicOrderCntVal of the
current
picture is less than RpPicOrderCntVal of the associated GDR picture;
68
Date recue/date received 2021-10-26
PCTUS2021032924
(4) sps video_parameter set id is greater than 0, ols mode idc is equal to 0
and the
current AU contains a picture (e.g. syntax element picA) that satisfies all of
the following
conditions: (a) PicA has PictureOutputFlag equal to 1, (b) PicA has nuh layer
id nuhLid
greater than that of the current picture, (c) PicA belongs to the output layer
of the OLS (i.e.
OutputLayerIdInOls[ TargetOlsIdx ][ 0 ] is equal to nuhLid);
(5) sps video_parameter set id is greater than 0, ols mode idc is equal to 2,
andols output layer flag[ TargetOlsIdx ][ GeneralLayerIdx[ nuh layer id ] ] is
equal to 0.
[0288] If none
of the above conditions are true, the syntax element PictureOutputFlag
may be set equal to the syntax element pic output flag.
[0289] After
all slices of the current picture have been decoded, the current decoded
picture may be marked as "used for short-term reference", and each ILRP entry
in
RefPicList[ 0 ] or RefPicList[ 11 may be marked as "used for short-term
reference".
[0290] In the
same or another embodiment, when each layer is an output layer set, the
syntax element PictureOutputFlag is set equal to pic output flag, regardless
of the value of
the syntax element ols mode idc (864).
[0291] In the
same or another embodiment, the syntax element PictureOutputFlag is
set equal to 0 when sps video_parameter set id is greater than 0, each layer
is an ols flag
(862) is equal to 0, ols mode idc (864) is equal to 0, and the current AU
contains a picture
picA that satisfies all of the following conditions: PicA has
PictureOutputFlag equal to 1,
PicA has nuh layer id nuhLid greater than that of the current picture, and
PicA belongs to
the output layer of the OLS (i.e. OutputLayerIdInOls[ TargetOlsIdx ][ 0 ] is
equal to nuhLid).
[0292] In the
same or another embodiment, the syntax element PictureOutputFlag is
set equal to 0 when sps video_parameter set id is greater than 0, each layer
is an ols flag
is equal to 0, ols mode idc is equal to 2, and
ols output layer flag[ TargetOlsIdx ][ GeneralLayerIdx[ nuh layer id ] ] is
equal to 0.
69
Date recue/date received 2021-10-26
PCTU S2021032924
[0293] An intra random access point (TRAP) picture may be a coded picture
for
random access, which may be an instantaneous decoding refresh (IDR) picture
that supports
closed group of picture structures or a clean random access (CRA) picture that
supports open
group of picture structure. A gradual decoding refresh (GDR) picture may be a
picture for
gradual random access with partial refresh of the picture.
[0294] Embodiments of the present disclosure may include syntax elements
that
indicate an TRAP or GDR picture. For example, with reference to FIG. 26, a
picture header
(1) may be provided. In the picture header (1), a flag ph gdr or irap_pic flag
(2) may be
signaled. The flag indicates the presence of an TRAP or GDR picture in the
current PU
associated with the picture header (1).
[0295] In the same or another embodiment, a flag ph no output of_prior_pics
flag
(3) may be conditionally signaled, only when ph_gdr or irap_pic flag (2) is
equal to 1, as
shown in FIG. 26. The value of ph no output of_prior_pics flag (3) may be used
for the
output and removal process of pictures from the DPB. The value of the flag may
affect the
output of previously-decoded pictures in the DPB after the decoding of a
picture in a CVSS
AU that is not the first AU in the bitstream.
[0296] There is a potential issue that an TRAP picture may have
ph gdr or irap_pic flag equal to 0, because the constraint specified by the
semantics of
ph gdr or irap_pic flag may be "one way", as follows: The flag ph gdr or
irap_pic flag
equal to 1 specifies that the current picture is a GDR or TRAP picture. The
flag
ph gdr or irap_pic flag equal to 0 specifies that the current picture is not a
GDR picture and
may or may not be an TRAP picture. When the value of ph gdr or irap_pic flag
of an TRAP
picture is equal to 0, the value of ph no output of_prior_pics flag may be
used for DPB
operations, without signaling or any inference rule.
Date recue/date received 2021-10-26
PCTUS2021032924
[0297] In order to address the potential issue, in an embodiment, the
constraints of the
semantics of ph gdr or irap_pic flag (2) may be specified as "two way", so
that
ph no output of_prior_pics flag (3) may be required to be signaled when the
current picture
is an TRAP picture, as follows: The flag ph gdr or irap_pic flag (2) equal to
1 specifies that
the current picture is a GDR or TRAP picture. The flag ph gdr or irap_pic flag
(2) equal to
0 specifies that the current picture is neither a GDR or TRAP picture.
[0298] In the same or another embodiment, the inference rule of the
ph no output of_prior_pics flag (3) value, when not present, may be specified
as follows:
The flag ph no output of_prior_pics flag (3) affects the output of previously-
decoded
pictures in the DPB after the decoding of a picture in a CVSS AU that is not
the first AU in
the bitstream. It may be a requirement of bitstream conformance that, when
present, the value
of ph no output of_prior_pics flag (3) shall be the same for all pictures in
an AU.
[0299] When ph no output of_prior_pics flag (3) is present in the picture
headers
(1) of the pictures in an AU, the ph no output of_prior_pics flag (3) value of
the AU is the
ph no output of_prior_pics flag (3) value of the pictures in the AU. When not
present, the
value of ph no output of_prior_pics flag (3) may be inferred to be equal to 0.
[0300] With reference to FIG. 27, an AU delimiter (10) may be used to
indicate the
start of an AU, whether the AU is an TRAP or GDR AU, and the type of slices
present in the
coded pictures in the AU containing the AU delimiter NAL unit. When the
bitstream contains
only one layer, there may not be a normative decoding process associated with
the AU
delimiter (10).
[0301] In the AU delimiter (10), aud irap or_gdr au flag (12) may indicate
the
presence of an TRAP or GDR AU, and may be signaled as shown in FIG. 27. The
flag
aud kap or gdr au flag (12) equal to 1 may specify that the AU containing the
AU
71
Date recue/date received 2021-10-26
PCTUS2021032924
delimiter is an TRAP or GDR AU. The flag aud irap or gdr au flag (12) equal to
0 may
specify that the AU containing the AU delimiter (10) is neither an TRAP or GDR
AU.
[0302] In the same or another embodiment, the flag aud irap or gdr au flag
(12) of
an TRAP or GDR AU may be present, when the bitstream has multiple layers, with
sps video_parameter set id is greater than 0. A video coding technology or
standard may
mandate the presence of an AU delimiter for a multiple layered bitstream.
[0303] In the same or another embodiment, referring to FIGS. 26-27, when
aud irap or gdr au flag (12) is present and the value of aud irap or gdr au
flag (12) is
equal to 1, the value of ph gdr or irap_pic flag (2) may be required to be
equal to 1. It is
because each PU may have a GDR or TRAP picture, when aud irap or gdr au flag
(12) in
AU delimiter (10) is equal to 1.
[0304] In the same or another embodiment, when pps mixed nalu types in_pic
flag
is equal to 1, the value of ph no output of_prior_pics flag may not be
present, When
pps mixed nalu types in_pic flag is determined (e.g. by a decoder) to be
present, the value
of ph no output of_prior_pics flag (3) may be ignored.
[0305] In the same or another embodiment, ph no output of_prior_pics flag
(3) may
affect the output of previously-decoded pictures in the DPB after the decoding
of a picture in
a CVSS AU that is not the first AU in the bitstream. When not present, the
value of
ph no output of_prior_pics flag (3) may be inferred to be equal 1.
[0306] In the same or another embodiment, in order to resolve an issue that
the value
of ph no output of_prior_pics flag (3) is used without an inferrence rule,
when
ph no output of_prior_pics flag (3) is not present and ph_gdr or irap_pic flag
(2) is equal
to 1, ph gdr or irap_pic flag (2) may be replaced in picture header (1) with
ph irap_pic flag (6), as shown in FIG. 28. The flag ph irap_pic flag (6) equal
to 1 may
72
Date recue/date received 2021-10-26
PCTUS2021032924
specify that the current picture is an TRAP picture. The flag ph irap_pic flag
(6) equal to 0
may specify that the current picture is not an TRAP picture.
[0307] In the same or another embodiment, in order to resolve the issue
that the value
of ph no output of_prior_pics flag (3) is used without an inferrence rule,
when
ph no output of_prior_pics flag (3) is not present and ph_gdr or irap_pic flag
(2) is equal
to 1, ph no output of_prior_pics flag (3) in picture header (1) may be
replaced by
sh no output of_prior_pics flag (23) in slice header (20), as shown in FIG.
29.
[0308] In the same embodiment, sh no output of_prior_pics flag (23) may be
conditionally present in slice header (20) only when the NAL unit type of the
current VCL
NAL is equal to IDR W RADL, IDR N LP, or CRA NUT. IDR W RADL is an NAL unit
type that contains a coded slice segment of an IDR picture that does not have
associated
RASL pictures present in the bitstream, but may have associated RADL pictures
in the
bitstream. IDR N LP may be an NAL unit type that contains a coded slice
segment of an
IDR picture that does not have associated leading pictures present in the
bitstream.
CRA NUT is an NAL unit type that contains a coded slice segment of a CRA
picture.
[0309] In the same or another embodiment, sh no output of_prior_pics flag
(23)
may affect the output of previously-decoded pictures in the DPB after the
decoding of a
picture in a CVSS AU that is not the first AU in the bitstream.
[0310] In the same or another embodiment, it may be a requirement of
bitstream
conformance that, when present, the value of sh no output of_prior_pics flag
(23) shall be
the same for all pictures in an AU. When sh no output of_prior_pics flag (23)
is present in
the slice headers (20) of the pictures in an AU, the sh no output
of_prior_pics flag (23)
value of the AU may be the sh no output of_prior_pics flag (23) value of the
pictures in the
AU.
73
Date recue/date received 2021-10-26
PCTU S2021032924
[0311] In the same or another embodiment, when pps mixed nalu types in_pic
flag
in picture parameter set is equal to 1, the value of sh no output
of_prior_pics flag (23) may
not be present. When present, the value of sh no output of_prior_pics flag
(23) may be
ignored.
[0312] In the same or another embodiment, aud irap au flag (16) may be
present in
AU delimiter (10), as shown in FIG. 30. The flag aud irap au flag (16) equal
to 1 may
specify that the AU containing the AU delimiter (10) is an IRAP AU. The flag
aud irap au flag (16) equal to 0 may specify that the AU containing the AU
delimiter (10) is
not an IRAP AU.
[0313] In the same embodiment, when aud irap au flag (16) is present, the
value of
ph irap_pic flag (6) in picture header (1) may be equal to aud irap au flag
(16) in AU
delimiter (10). The flag ph irap_pic flag (7) equal to 1 may specify that
picture associated
with PH (1) is an IRAP picture. The flag ph irap_pic flag (6) equal to 0 may
specify that
picture associated with PH (1) is not an IRAP picture.
[0314] In the same or another embodiment, aud_gdr au flag (17) may be
present in
AU delimiter (10), as shown in FIG. 30. The flag aud_gdr au flag (17) equal to
1 may
specify that the AU containing the AU delimiter (10) is a GDR AU. The flag
aud irap au flag (17) equal to 0 may specify that the AU containing the AU
delimiter (10) is
not a GDR AU.
[0315] In the same embodiment, when aud gdr au flag (17) in AU delimiter
(10) is
present, the value of ph gdr_pic flag (7) in picture header (1) may be equal
to
aud_gdr au flag (17) in AU delimiter (10). The flag ph gdr_pic flag (7) equal
to 1 may
specify that the picture associated with PH (1) is a GDR picture. The flag
ph_gdr_pic flag (7)
equal to 0 may specify that the picture associated with PH (1) is not a GDR
picture.
74
Date recue/date received 2021-10-26
PCTUS2021032924
[0316] Gradual decoding refresh (GDR) may be specified by the following
definitions:
GDR AU: An AU in which there is a PU for each layer specified by the VPS
and the coded picture in each present PU is a GDR picture.
GDR PU: A PU in which the coded picture is a GDR picture.
GDR picture: A picture for which each VCL NAL unit has nal unit type
equal to GDR NUT.
GDR sub-picture: A sub-picture for which each VCL NAL unit has
nal unit type equal to GDR NUT.
GDR NUT: A NAL unit type that includes a coded tile group of a GDR
picture.
[0317] According to embodiments, the first picture in the bitstream in
decoding order
may be an TRAP or GDR picture. Trailing pictures associated with an TRAP or
GDR picture
may also follow the TRAP or GDR picture in decoding order. Pictures that
follow the
associated TRAP picture in output order and precede the associated TRAP
picture in decoding
order may not be allowed.
[0318] In an embodiment, with reference to FIG. 31, syntax elements
indicating GDR
are signaled in PH (1) such as ph gdr_pic flag (7). When not present, the
value of
ph gdr_pic flag (7) may be inferred to be equal to 0. When sps_gdr enabled
flag is equal to
0, the value of ph gdr_pic flag (7) may be inferred to be equal to 0. The
syntax element
ph recovery_poc cnt (32) may specify the recovery point of decoded pictures in
output
order.
[0319] When the current picture is a GDR picture, the variable
recoveryPointPocVal
may be derived as follows:
recoveryPointPocVal = PicOrderCntVal + ph recovery_poc cnt
Date recue/date received 2021-10-26
PCTU S2021032924
[0320] In the same or another embodiment, as shown in FIG 31,
ph recovery_poc cnt (34) is signaled after signaling picture order count (POC)
syntax
elements (e.g. ph_pic order cnt lsb (32) and ph_poc msb cycle val (36))
because
PicOrderCntVal that is used for deriving recoveryPointPocVal is derived from
the values of
ph_pic order cnt lsb (32) and ph_poc msb cycle val (36).
[0321] If the current picture is a GDR picture, and there is a picture picA
that follows
the current GDR picture in decoding order in the CLVS that has PicOrderCntVal
equal to
recoveryPointPocVal, the picture picA may be referred to as the recovery point
picture.
Otherwise, the first picture in output order that has PicOrderCntVal greater
than
recoveryPointPocVal in the CLVS may be referred to as the recovery point
picture. The
recovery point picture may not precede the current GDR picture in decoding
order. The
pictures that are associated with the current GDR picture and have
PicOrderCntVal less than
recoveryPointPocVal may be referred to as the recovering pictures of the GDR
picture. The
value of ph recovery_poc cnt (34) may be in the range of 0 to
MaxPicOrderCntLsb ¨ 1,
inclusive.
[0322] In the same or another embodiment, the recovery point picture may
not
precede the current GDR picture in both decoding and output order.
[0323] In the same or another embodiment, the recovering picture may not
precede
the current GDR picture in both decoding and output order.
[0324] In the same or another embodiment, the recovering picture may
precede the
associated recovery point picture in both decoding and output order.
[0325] In the same or another embodiment, when the current picture is a GDR
picture
or a recovering picture of a GDR picture, and the current picture contains a
non-CTU-aligned
boundary between a "refreshed area" (i.e. an area that has an exact match of
decoded sample
76
Date recue/date received 2021-10-26
PCTUS2021032924
values when starting the decoding process from the GDR picture compared to
starting the
decoding process from the previous TRAP picture in decoding order, when
present) and a
"dirty area" (i.e. an area that might not have an exact match of decoded
sample values when
starting the decoding process from the GDR picture compared to starting the
decoding
process from the previous TRAP picture in decoding order, when present),
chroma residual
scaling of luma mapping with chroma scaling (LMCS) may have to be disabled in
the current
picture to avoid the "dirty area" to affect decoded sample values of the
"refreshed area".
[0326] In the same or another embodiment, the value of recoveryPointPocVal
of a
layer A may be equal to or greater than recoveryPointPocVal of the reference
layer of the
layer A.
[0327] In the same or another embodiment, the value of recoveryPointPocVal
of a
layer A with layerId equal to m may be equal to or greater than
recoveryPointPocVal of
another layer B with layerId equal to n, where m is greater than n.
[0328] In the same or another embodiment, the value of recoveryPointPocVal
of a
layer A with layerId equal to m may be equal to or greater than
recoveryPointPocVal of
another layer B with layerId equal to n, where m is greater than n and the
layers A and B
belong the same output layer set.
[0329] In the same or another embodiment, the value of recoveryPointPocVal
of a
layer A may be equal to recoveryPointPocVal of the reference layer of the
layer A.
[0330] In the same or another embodiment, the value of recoveryPointPocVal
of a
layer A with layerId equal to m may be equal to recoveryPointPocVal of another
layer B with
layerId equal to n, where m is greater than n.
[0331] In the same or another embodiment, the value of recoveryPointPocVal
of a
layer A with layerId equal to m may be equal to recoveryPointPocVal of another
layer B with
77
Date recue/date received 2021-10-26
PCTUS2021032924
layerId equal to n, where m is greater than n and the layers A and B belong
the same output
layer set.
[0332] In the same or another embodiment, when pps mixed nalu types in_pic
flag
is equal to 1, the following may apply:
(1) The picture shall have at least two sub-pictures.
(2) VCL NAL units of the picture shall have two or more different
nal unit type values.
(3) There shall be no VCL NAL unit of the picture that has nal unit type
equal to GDR NUT.
(4) The picture shall not be the recovering or recovery picture associated
with
a GDR picture.
(5) When the VCL NAL units of at least one sub-picture of the picture have a
particular value of nal unit type equal to IDR W RADL, IDR N LP, or
CRA NUT, the VCL NAL units of other sub-pictures in the picture shall all have
nal unit type equal to TRAIL NUT.
[0333] TRAIL NUT is an NAL unit type that contains a coded tile group of
VCL
non-STSA trailing picture.
[0334] At the beginning of the decoding process for each slice of a
picture, the
decoding process for reference picture lists construction may be invoked for
derivation of
reference picture list 0 (RefPicList[ 0 ]) and reference picture list 1
(RefPicList[ 11). After
constructing one or more reference picture lists, the decoding process for
reference picture
marking may be invoked, wherein reference pictures may be marked as "unused
for
reference", "used for short-term reference", or "used for long-term
reference".
[0335] With reference to FIG. 32, according to embodiments, a decoding
process (40)
may be performed by a decoder. In the decoding process (40), one or more
reference picture
78
Date recue/date received 2021-10-26
PCTUS2021032924
lists (RPLs) may be constructed (42). When one or more reference picture lists
are
constructed, one or more pictures may be unavailable due to random access or
unintended
picture loss. The decoder may determine whether a reference picture in the RPL
is available
in the DPB (44). If the reference picture is determined to be unavailable, the
unavailable
reference pictures may be marked as "no reference picture". To avoid any
decoder crash or
unintended behavior, the unavailable reference pictures may be instantly
generated with
defaults values of pixels and parameters (46). After generating the
unavailable reference
pictures (and/or the reference pictures are determined to be available), the
decoder may check
the validation of all reference pictures, including the generated pictures, in
the reference
picture list (48).
[0336] In the same or another embodiment, the one or more reference
picture lists are
constructed (42) by parsing the RPL syntax elements in SPS, PH, and/or SH.
After the
constructing (42), one or more reference pictures in RPL list may not be
available because the
random access skipped leading pictures (RASL) associated with a CRA picture
may be
discarded by a decoder or a system or cannot be successfully decoded when
random access
happens at the CRA picture. The unavailable reference pictures may be
generated with
defaults values of pixels and parameters (46).
[0337] In the same or another embodiment, when the current picture is an
IDR picture
with sps idr rpl_present flag equal to 1 or pps rpl info in_ph flag equal to
1, a CRA
picture with NoOutputBeforeRecoveryFlag equal to 1, or a GDR picture with
NoOutputBeforeRecoveryFlag equal to 1, at least one of the following decoding
processes
for generating unavailable reference pictures (46) is invoked, which may need
to be invoked
only for the first slice of a picture.
[0338] A. General decoding process for generating unavailable reference
pictures
79
Date recue/date received 2021-10-26
PCTUS2021032924
[0339] This process may be invoked once per coded picture when the current
picture
is an IDR picture with sps idr rpl_present flag equal to 1 or pps rpl info
in_ph flag equal
to 1, a CRA picture with NoOutputBeforeRecoveryFlag equal to 1, or a GDR
picture with
NoOutputBeforeRecoveryFlag equal to 1. When this process is invoked, the
following may
apply: For each RefPicList[ i ][ j ], with i in the range of 0 to 1,
inclusive, and j in the range
of 0 to num ref entries[ i ][ RplsIdx[ ill ¨ 1, inclusive, that is equal to
"no reference
picture", a picture is generated as described below in the "Generation of one
unvailable
picture" and the following applies:
(1) The value of nuh layer id for the generated picture is set equal to
nuh layer id of the current picture.
(2) If st ref_pic flag[ i ][ RplsIdx[ ii 1[i ] is equal to 1 and
inter layer ref_pic flag[ i ][ RplsIdx[ ii 1[i ] is equal to 0, the value of
PicOrderCntVal for the generated picture is set equal to RefPicPocList[ i ][ j
] and the
generated picture is marked as "used for short-term reference".
(3) Otherwise, when st ref_pic flag[ i ][ RplsIdx[ i ] ][ j ] is equal to 0
and
inter layer ref_pic flag[ i ][ RplsIdx[ ii 1[i ] is equal to 0, the value of
PicOrderCntVal for the generated picture is set equal to RefPicLtPocList[ i ][
j ], the
value of ph_pic order cnt lsb for the generated picture is inferred to be
equal to
( RefPicLtPocList[ i ][ j ] & ( MaxPicOrderCntLsb ¨ 1 ) ), and the generated
picture is
marked as "used for long-term reference".
(4) The value of PictureOutputFlag for the generated reference picture is set
equal to 0.
(5) RefPicList[ i ][ j ] is set to be the generated reference picture.
(6) The value of Temporalld for the generated picture is set equal to
Temporalld of the current picture.
Date recue/date received 2021-10-26
PCTUS2021032924
(7) The value of ph non ref_pic flag for the generated picture is set equal to
0.
(8) The value of ph_pic_parameter set id for the generated picture is set
equal
to ph_pic_parameter set id of the current picture.
[0340] The flag ph non ref_pic flag equal to 1 may specify that the picture
associated with the PH is never used as a reference picture. The flag ph non
ref_pic flag
equal to 0 may specify the picture associated with the PH may or may not be
used as a
reference picture.
[0341] B. Generation of one unavailable picture
[0342] When this process is invoked, an unavailable picture is generated as
follows:
(1) The value of each element in the sample array SL for the picture is set
equal to 1 << ( BitDepth ¨ 1).
(2) When sps chroma format idc is not equal to 0, the value of each
element in the sample arrays Scb and Scr for the picture is set equal to 1 <<
( BitDepth
¨ 1 ).
(3) The prediction mode CuPredMode[ 0 ][ x ][ y ] is set equal to
MODE INTRA for x ranging from 0 to pps_pic width in luma samples ¨ 1,
inclusive, and y ranging from 0 to pps_pic height in luma samples ¨ 1,
inclusive.
[0343] In the same or another embodiment, after generating the unavailable
reference
picture in RPL list, a bitstream conformances check for all or active
reference pictures in RPL
list may be invoked by, for example, the decoder. For example, the decoder may
check that
the following constraints apply for bitstream conformance:
(1) For each i equal to 0 or 1, num ref entries[ i ][ RplsIdx[ ii
] shall not
be less than NumRefldxActive[ ii.
81
Date recue/date received 2021-10-26
PCTUS2021032924
(2) The picture referred to by each active entry in RefPicList[ 0 ] or
RefPicList[ 11 shall be present in the DPB and shall have TemporalId less than
or
equal to that of the current picture.
(3) The picture referred to by each entry in RefPicList[ 0 ] or
RefPicList[ 11 shall not be the current picture and shall have ph non ref_pic
flag
equal to 0.
(4) A short term reference picture (STRP) entry in RefPicList[ 0 ] or
RefPicList[ 11 of a slice of a picture and a longer term reference picture
(LTRP) entry
in RefPicList[ 0 ] or RefPicList[ 11 of the same slice or a different slice of
the same
picture shall not refer to the same picture.
(5) There shall be no LTRP entry in RefPicList[ 0 ] or RefPicList[ 11 for
which the difference between the PicOrderCntVal of the current picture and the
PicOrderCntVal of the picture referred to by the entry is greater than or
equal to 224.
(6) Let set0fRefPics be the set of unique pictures referred to by all
entries
in RefPicList[ 0 ] that have the same nuh layer id as the current picture and
all
entries in RefPicList[ 11 that have the same nuh layer id as the current
picture. The
number of pictures in set0fRefPics shall be less than or equal to MaxDpbSize ¨
1,
inclusive, where MaxDpbSize is, and set0fRefPics shall be the same for all
slices of a
picture.
(7) When the current slice has nal unit type equal to STSA NUT, there
shall be no active entry in RefPicList[ 0 ] or RefPicList[ 11 that has
Temporalld equal
to that of the current picture and nuh layer id equal to that of the current
picture.
(8) When the current picture is a picture that follows, in decoding order,
a
step-wise temporal sub-layer access (STSA) picture that has Temporalld equal
to that
of the current picture and nuh layer id equal to that of the current picture,
there shall
82
Date recue/date received 2021-10-26
PCTUS2021032924
be no picture that precedes the STSA picture in decoding order, has TemporalId
equal
to that of the current picture, and has nuh layer id equal to that of the
current picture
included as an active entry in RefPicList[ 0 ] or RefPicList[ 11.
(9) When the current sub-picture, with Temporalld equal to a
particular
value tId, nuh layer id equal to a particular value layerId, and sub-picture
index equal
to a particular value subpicIdx, is a sub-picture that follows, in decoding
order, an
STSA sub-picture with Temporalld equal to tId, nuh layer id equal to layerId,
and
sub-picture index equal to subpicIdx, there shall be no picture with
Temporalid equal
to tId and nuh layer id equal to layerId that precedes the picture containing
the STSA
sub-picture in decoding order included as an active entry in RefPicList[ 0 ]
or
RefPicList[ 11.
(10) When the current picture, with nuh layer id equal to a particular value
layerId, is an TRAP picture, there shall be no picture referred to by an entry
in
RefPicList[ 0 ] or RefPicList[ 11 that precedes, in output order or decoding
order, any
preceding TRAP picture with nuh layer id equal to layerId in decoding order
(when
present).
(11) When the current sub-picture, with nuh layer id equal to a particular
value layerId and sub-picture index equal to a particular value subpicIdx, is
an TRAP
sub-picture, there shall be no picture referred to by an entry in RefPicList[
0 ] or
RefPicList[ 11 that precedes, in output order or decoding order, any preceding
picture, in decoding order (when present), containing an TRAP sub-picture with
nuh layer id equal to layerId and sub-picture index equal to subpicIdx.
(12) When the current picture is not a RASL picture associated with a CRA
picture with NoOutputBeforeRecoveryFlag equal to 1, there shall be no picture
referred to by an active entry in RefPicList[ 0 ] or RefPicList[ 11 that was
generated
83
Date recue/date received 2021-10-26
PCTUS2021032924
by the decoding process for generating unavailable reference pictures for the
CRA
picture associated with the current picture.
(13) When the current sub-picture is not a RASL sub-picture associated
with a CRA sub-picture in a CRA picture with NoOutputBeforeRecoveryFlag equal
to 1, there shall be no picture referred to by an active entry in RefPicList[
0 ] or
RefPicList[ 11 that was generated by the decoding process for generating
unavailable
reference pictures for the CRA picture containing the CRA sub-picture
associated
with the current sub-picture.
(14) When the current picture, with nuh layer id equal to a particular value
layerId, is not any of the following, there shall be no picture referred to by
an entry in
RefPicList[ 0 ] or RefPicList[ 11 that was generated by the decoding process
for
generating unavailable reference pictures for the TRAP or GDR picture
associated
with the current picture:
(a) An IDR picture with sps idr rpl_present flag equal to
1 or pps rpl info in_ph flag equal to 1.
(b) A CRA picture with NoOutputBeforeRecoveryFlag
equal to 1.
(c) A picture, associated with a CRA picture with
NoOutputBeforeRecoveryFlag equal to 1, that precedes, in decoding
order, the leading pictures associated with the same CRA picture.
(d) A leading picture associated with a CRA picture with
NoOutputBeforeRecoveryFlag equal to 1
(e) A GDR picture with NoOutputBeforeRecov.eryFlag
equal to 1.
84
Date recue/date received 2021-10-26
PCTUS2021032924
(0 A recovering picture of a GDR picture with
NoOutputBeforeRecoveryFlag equal to 1 and nuh layer id equal to
layerId.
(15) When the current sub-picture, with nuh layer id equal to a particular
value layerId and sub-picture index equal to a particular value subpicIdx, is
not any of
the following, there shall be no picture referred to by an entry in
RefPicList[ 0 ] or
RefPicList[ 11 that was generated by the decoding process for generating
unavailable
reference pictures for the TRAP or GDR picture containing the TRAP or GDR sub-
picture associated with the current sub-picture:
(a) An IDR sub-picture in an IDR picture with
sps idr rpl_present flag equal to 1 or pps rpl info in_ph flag equal
to 1.
(b) A CRA sub-picture in a CRA picture with
NoOutputBeforeRecoveryFlag equal to 1.
(c) A sub-picture, associated with a CRA sub-picture in a
CRA picture with NoOutputBeforeRecoveryFlag equal to 1, that
precedes, in decoding order, the leading pictures associated with the
same CRA picture.
(d) A leading sub-picture associated with a CRA sub-
picture in a CRA picture with NoOutputBeforeRecoveryFlag equal to
1.
(e) A GDR sub-picture in a GDR picture with
NoOutputBeforeRecoveryFlag equal to 1.
Date recue/date received 2021-10-26
PCTUS2021032924
(0 A sub-picture in a recovering picture of a GDR
picture
with NoOutputBeforeRecoveryFlag equal to 1 and nuh layer id equal
to layerId.
(16) When the current picture follows an TRAP picture having the same
value of nuh layer id in both decoding order and output order, there shall be
no
picture referred to by an active entry in RefPicList[ 0 ] or RefPicList[ 11
that precedes
that TRAP picture in output order or decoding order.
(17) When the current sub-picture follows an TRAP sub-picture having the
same value of nuh layer id and the same value of sub-picture index in both
decoding
and output order, there shall be no picture referred to by an active entry in
RefPicList[ 0 ] or RefPicList[ 11 that precedes the picture containing that
TRAP sub-
picture in output order or decoding order.
(18) When the current picture follows an TRAP picture having the same
value of nuh layer id and the leading pictures, if any, associated with that
TRAP
picture in both decoding order and output order, there shall be no picture
referred to
by an entry in RefPicList[ 0 ] or RefPicList[ 11 that precedes that TRAP
picture in
output order or decoding order.
(19) When the current sub-picture follows an TRAP sub-picture having the
same value of nuh layer id and the same value of sub-picture index and the
leading
sub-pictures, if any, associated with that TRAP sub-picture in both decoding
and
output order, there shall be no picture referred to by an entry in RefPicList[
0 ] or
RefPicList[ 11 that precedes the picture containing that TRAP sub-picture in
output
order or decoding order.
86
Date recue/date received 2021-10-26
PCTUS2021032924
(20) When the current picture is a random access decodable leading
(RADL) picture, there shall be no active entry in RefPicList[ 0 ] or
RefPicList[ 11
that is any of the following:
(a) A RASL picture with
pps mixed nalu types in_pic flag is equal to 0. This means that an
active entry of the RPLs of a RADL picture can refer to a RASL
picture with pps mixed nalu types in_pic flag equal to 1. However,
when decoding starts from the associated CRA picture, such a RADL
picture can still be correctly decoded, because the RADL sub-
picture(s) in that referenced RASL picture would be correctly decoded,
as the RADL picture would only refer to the RADL sub-pictures in the
referenced RASL picture, as imposed by the next constraint that
disallows RADL sub-pictures referring to a RASL sub-picture.
(b) A picture that precedes the associated TRAP picture in
decoding order.
(21) When the current sub-picture, with nuh layer id equal to a particular
value layerId and sub-picture index equal to a particular value subpicIdx, is
a RADL
sub-picture, there shall be no active entry in RefPicList[ 0 ] or RefPicList[
11 that is
any of the following:
(a) A picture with nuh layer id equal to layerId containing
a RASL sub-picture with sub-picture index equal to subpicIdx.
(b) A picture that precedes the picture containing the
associated TRAP sub-picture in decoding order.
87
Date recue/date received 2021-10-26
PCTUS2021032924
(22) The following constraints apply for the picture referred to by each
ILRP entry, when present, in RefPicList[ 0 ] or RefPicList[ 11 of a slice of
the current
picture:
(a) The picture shall be in the same AU as the current
picture.
(b) The picture shall be present in the DPB.
(c) The picture shall have nuh layer id refPicLayerId less
than the nuh layer id of the current picture.
(d) Either of the following constraints applies: The picture
shall be an TRAP picture; or the picture shall have Temporalld less
than or equal to
Max( 0, vps max tid il ref_pics_plusl[ currLayerIdx ][ refLayerIdx ]
¨ 1), where currLayerIdx and refLayerIdx are equal to
GeneralLayerIdx[ nuh layer id ] and
GeneralLayerIdx[ refpicLayerId ], respectively.
(23) Each ILRP entry, when present, in RefPicList[ 0 ] or RefPicList[ 11 of
a slice shall be an active entry.
(24) When vps independent layer flag[ GeneralLayerIdx[ nuh layer id ] ]
is equal to 0 and sps num subpics minusl is greater than 0, either of the
following
two conditions (but not both) shall be true:
(a) The picture referred to by each active entry in
RefPicList[ 0 ] or RefPicList[ 11 has the same sub-picture layout as
the current picture (i.e. the SPSs referred to by that picture and the
current picture have the same value of sps num subpics minusl and
88
Date recue/date received 2021-10-26
PCTUS2021032924
the same values of sps subpic ctu top left x[ j ],
sps subpic ctu top left_y[ j ], sps subpic width minus1[ j ], and
sps subpic height minusl[ ii, respectively, for each value of j in the
range of 0 to sps num subpics minusl, inclusive).
(b) The picture referred to by each active entry in
RefPicList[ 0 ] or RefPicList[ 1 ] is an ILRP for which the value of
sps num subpics minusl is equal to 0.
[0344] C. Decoding process for reference picture marking
[0345] In the same or another embodiment, when the reference pictures are
marked,
the following decoding process may be invoked.
[0346] The process may be invoked once per picture, after decoding of a
slice header
and the decoding process for reference picture list construction for the
slice, but prior to the
decoding of the slice data. This process may result in one or more reference
pictures in the
DPB being marked as "unused for reference" or "used for long-term reference".
[0347] A decoded picture in the DPB can be marked as "unused for
reference", "used
for short-term reference" or "used for long-term reference", but only one
among these three at
any given moment during the operation of the decoding process. Assigning one
of these
markings to a picture may implicitly remove another of these markings when
applicable.
When a picture is referred to as being marked as "used for reference", this
collectively refers
to the picture being marked as "used for short-term reference" or "used for
long-term
reference" (but not both).
[0348] STRPs and ILRPs may be identified by their nuh layer id and
PicOrderCntVal values. LTRPs may be identified by their nuh layer id values
and the
Log2( MaxLtPicOrderCntLsb ) LSBs of their PicOrderCntVal values.
89
Date recue/date received 2021-10-26
PCTUS2021032924
[0349] If the current picture is a CLVSS picture, all reference pictures
currently in the
DPB (if any) with the same nuh layer id as the current picture may be marked
as "unused for
reference".
[0350] Otherwise, the following may apply:
(1) For each LTRP entry in RefPicList[ 0 ] or RefPicList[ 11, when the
picture is an LTRP with the same nuh layer id as the current picture, the
picture is
marked as "used for long-term reference".
(2) Each reference picture with the same nuh layer id as the current
picture in the DPB that is not referred to by any entry in RefPicList[ 0 ] or
RefPicList[ 1 ] is marked as "unused for reference".
(3) For each ILRP entry in RefPicList[ 0 ] or RefPicList[ 11, the picture
is
marked as "used for long-term reference".
[0351] In the same or another embodiment, for each LTRP entry in
RefPicList[ 0 ] or
RefPicList[ 11, when the picture with the same nuh layer id as the current
picture is marked
as "used for short-term reference", the picture is marked as "used for long-
term reference".
[0352] In the same or another embodiment, when one or more RPL lists are
constructued, the following may apply:
For each RefPicList[ i ][ j ], with i in the range of 0 to 1, inclusive, and j
in the
range of 0 to num ref entries[ i ][ RplsIdx[ ii ] ¨ 1, inclusive, that is
equal to "no
reference picture", a picture is generated.
[0353] In the same or another embodiment, when one or more RPL lists are
constructed, the following may apply:
Date recue/date received 2021-10-26
PCTUS2021032924
For each RefPicList[ i ][ j ], with i in the range of 0 to 1, inclusive, and j
in the
range of 0 to NumRefldxActive[ ii ¨ 1, inclusive, that is equal to "no
reference
picture", a picture is generated.
[0354] In the decoding process for reference picture list construction,
missing
reference pictures in DPB may be set equal to "no reference picture". The
unavailable
pictures equal to "no reference picture" are generated through the decoding
process for
generating unavailable reference pictures for conformance check purposes. An
issue is that
not only unavailable reference pictures in the same layer but also unavailable
interlayer
reference pictures in the reference layer are set equal to "no reference
picture", as below:
if( !inter layer ref_pic flag[ i ][ RplsIdx[ ii 1[i ] ) {
if( st ref_pic flag[ i ][ RplsIdx[ ii 1[i ] ) {
RefPicPocList[ i 1[i ] =
pocBase + DeltaPocValSt[ i ][ RplsIdx[ ii 1[i ]
if( there is a reference picture picA in the DPB with the same
nuh layer id as the current picture
and PicOrderCntVal equal to RefPicPocList[ i ][ j ] )
RefPicList[ i 1[i ] = picA
else
RefPicList[ i 1[i ] = "no reference picture"
pocBase = RefPicPocList[ i ][ ii
1 else {
if( !delta_poc msb cycle_present flag[ i ][ k I) {
if( there is a reference picA in the DPB with the same nuh layer id
as the current picture and
91
Date recue/date received 2021-10-26
PCTUS2021032924
PicOrderCntVal & ( MaxPicOrderCntLsb ¨ 1) equal to
PocLsbLt[ i ][ k ] )
RefPicList[ i 1[i ] = picA
else
RefPicList[ i ][ j ] = "no reference picture"
RefPicLtPocList[ i ][ j ] = PocLsbLt[ i ][ k ]
1 else {
if( there is a reference picA in the DPB with the same nuh layer id
as the current picture and
PicOrderCntVal equal to FullPocLt[ i ][ k I)
RefPicList[ i ][ j ] = picA
else
RefPicList[ i ][ j ] = "no reference picture"
RefPicLtPocList[ i ][ j ] = FullPocLt[ i ][ k ]
1
k++
1
1 else {
layerIdx =
DirectRefLayerIdx[ GeneralLayerIdx[ nuh layer id ] ][ ilrp idx[ i ][ RplsIdx ]
1111
refPicLayerId = vps layer id[ layerIdx ]
if( there is a reference picture picA in the DPB with nuh layer id equal to
refPicLayerId and
the same PicOrderCntVal as the current picture)
92
Date recue/date received 2021-10-26
PCTUS2021032924
RefPicList[ i 1[i ] = picA
else
RefPicList[ i 1[i ] = "no reference picture"
1
[0355] However, a decoding process for generating unavailable reference
pictures, all
unavailable pictures equal to "no reference picture" may be handled as the
reference pictures
in the same layer, so that the nuh layer id and PicOrderCntVal values of the
unavailable
interlayer reference pictures are set to incorrrect values. Also, the
unavailable interlayer
reference pictures are not correctly marked as "used for long-term reference".
These wrong
values result in errors in decoding process of the following pictures and
bitstream
conformance check. Therefore, according to embodiments, the unavailable
interlayer
reference pictures equal to "no reference picture" should be correctly
generated as follows:
[0356] D. Improved decoding process for generating unavailable reference
pictures
[0357] This process is invoked once per coded picture when the current
picture is an
IDR picture with sps idr rpl_present flag equal to 1 or pps rpl info in_ph
flag equal to 1, a
CRA picture with NoOutputBeforeRecoveryFlag equal to 1, or a GDR picture with
NoOutputBeforeRecoveryFlag equal to 1.
[0358] When this process is invoked, for each RefPicList[ i ][ ii, with i
in the range
of 0 to 1, inclusive, and j in the range of 0 to num ref entries[ i ][
RplsIdx[ ill ¨ 1,
inclusive, that is equal to "no reference picture", a picture may be generated
as previously
described in the present dislcosure and the following may apply:
93
Date recue/date received 2021-10-26
PCTUS2021032924
(1) If inter layer ref_pic flag[ i ][ RplsIdx[ ii 1[i ] is equal to 0, the
value
of nuh layer id for the generated picture is set equal to nuh layer id of the
current
picture.
(2) Otherwise (inter layer ref_pic flag[ i ][ RplsIdx[ ii 1[i ] is equal to
1), value of nuh layer id for the generated picture is set equal to
vps layer id[ DirectRefLayerIdx[ GeneralLayerIdx[ nuh layer id ] ][ ilrp idx[
i ][ R
plsIdx ][ j ] ]]
(3) If st ref_pic flag[ i ][ RplsIdx[ ii 1[i ] is equal to 1 and
inter layer ref_pic flag[ i ][ RplsIdx[ ii 1[i ] is equal to 0, the value of
PicOrderCntVal for the generated picture is set equal to RefPicPocList[ i ][ j
] and the
generated picture is marked as "used for short-term reference".
(4) If st ref_pic flag[ i ][ RplsIdx[ ii ][ ii is equal to 0 and
inter layer ref_pic flag[ i ][ RplsIdx[ ii ][ j ] is equal to 0, the value of
PicOrderCntVal for the generated picture is set equal to RefPicLtPocList[ i ][
j ], the
value of ph_pic order cnt lsb for the generated picture is inferred to be
equal to
( RefPicLtPocList[ i ][ j ] & ( MaxPicOrderCntLsb ¨ 1 ) ), and the generated
picture is
marked as "used for long-term reference".
(5) Otherwise, if inter layer ref_pic flag[ i ][ RplsIdx[ ii ][ j ] is
equal to
1, the value of PicOrderCntVal for the generated picture is set equal to
PicOrderCntVal of the current picture, the value of ph_pic order cnt lsb for
the
generated picture is inferred to be equal to ph_pic order cnt lsb of the
current picture
and the generated picture is marked as "used for long-term reference".
(6) The value of PictureOutputFlag for the generated reference picture is
set equal to 0.
(7) RefPicList[ i ][ j ] is set to be the generated reference picture.
94
Date recue/date received 2021-10-26
PCTUS2021032924
(8) The value of TemporalId for the generated picture is set equal to
Temporalld of the current picture.
(9) The value of ph non ref_pic flag for the generated picture is set equal
to O.
(10) The value of ph_pic_parameter set id for the generated picture is set
equal to ph_pic_parameter set id of the current picture.
[0359] In decoding process for reference picture marking described above,
to clarify
the marking process, the following sentence is modified: For each LTRP entry
in
RefPicList[ 0 ] or RefPicList[ 11, when the picture is an STRP with the same
nuh layer id as
the current picture, the picture is marked as "used for long-term reference".
[0360] As a first option, the original sentence is modified as follows: For
each LTRP
entry in RefPicList[ 0 ] or RefPicList[ 11, when the picture is currently
marked as "used for
short-term reference" with the same nuh layer id as the current picture, the
picture is marked
as "used for long-term reference".
[0361] As a second option, the original sentence is modified as follows:
For each
LTRP entry in RefPicList[ 0 ] or RefPicList[ 11, when the picture is is an
LTRP with the
same nuh layer id as the current picture, the picture is marked as "used for
long-term
reference".
[0362] The meaning of the original sentence is somewhat confused, because
it says
that an STRP is marked as "used for long-term reference". If the original
intention of the
sentence is to mark the LTRP picture of the current picture in RefPicList[ 0 ]
or
RefPicList[ 1 ] as "used for long-term reference", when the LTRP is currently
marked as
"used for short-term reference", option 1 may be used. Otherwise, the option 2
may be used.
[0363] E. Improved decoding process for reference picture lists
construction
Date recue/date received 2021-10-26
PCTUS2021032924
[0364] In HEVC, the reference picture lists are constructed after
generating
unavailable reference pictures in RPS. In VVC (current draft), the reference
picture lists are
constructed before generating unavailable reference pictures in RPS. If the
bitstream
conformance check is performed before generating unavailable reference
pictures as
previously specified, the bitstream cannot obey the constraints specified. To
clarify the order,
the decoding process may be modified as follows to have the following
constraints for
bitstream conformance:
(1) For each i equal to 0 or 1, num ref entries[ i ][ RplsIdx[ ill shall
not
be less than NumRefldxActive[ ii.
(2) The picture referred to by each active entry in RefPicList[ 0 ] or
RefPicList[ 11 shall be present in the DPB and shall have Temporalld less than
or
equal to that of the current picture.
(3) The picture referred to by each entry in RefPicList[ 0 ] or
RefPicList[ 11 shall not be the current picture and shall have ph non ref_pic
flag
equal to 0.
(4) An STRP entry in RefPicList[ 0 ] or RefPicList[ 11 of a slice of a
picture and an LTRP entry in RefPicList[ 0 ] or RefPicList[ 11 of the same
slice or a
different slice of the same picture shall not refer to the same picture.
(5) There shall be no LTRP entry in RefPicList[ 0 ] or RefPicList[ 11 for
which the difference between the PicOrderCntVal of the current picture and the
PicOrderCntVal of the picture referred to by the entry is greater than or
equal to 224.
(6) Let set0fRefPics be the set of unique pictures referred to by all
entries
in RefPicList[ 0 ] that have the same nuh layer id as the current picture and
all
entries in RefPicList[ 11 that have the same nuh layer id as the current
picture. The
number of pictures in set0fRefPics shall be less than or equal to MaxDpbSize ¨
1,
96
Date recue/date received 2021-10-26
PCTUS2021032924
inclusive, where MaxDpbSize, and set0fRefPics shall be the same for all slices
of a
picture.
(7) When the current slice has nal unit type equal to STSA NUT, there
shall be no active entry in RefPicList[ 0 ] or RefPicList[ 11 that has
Temporalld equal
to that of the current picture and nuh layer id equal to that of the current
picture.
(8) When the current picture is a picture that follows, in decoding order,
an
STSA picture that has Temporalld equal to that of the current picture and
nuh layer id equal to that of the current picture, there shall be no picture
that
precedes the STSA picture in decoding order, has Temporalld equal to that of
the
current picture, and has nuh layer id equal to that of the current picture
included as
an active entry in RefPicList[ 0 ] or RefPicList[ 11.
(9) When the current sub-picture, with Temporalld equal to a particular
value tId, nuh layer id equal to a particular value layerId, and sub-picture
index equal
to a particular value subpicIdx, is a sub-picture that follows, in decoding
order, an
STSA sub-picture with Temporalld equal to tId, nuh layer id equal to layerId,
and
sub-picture index equal to subpicIdx, there shall be no picture with Temporand
equal
to tId and nuh layer id equal to layerId that precedes the picture containing
the STSA
sub-picture in decoding order included as an active entry in RefPicList[ 0 ]
or
RefPicList[ 11.
(10) When the current picture, with nuh layer id equal to a particular value
layerId, is an TRAP picture, there shall be no picture referred to by an entry
in
RefPicList[ 0 ] or RefPicList[ 11 that precedes, in output order or decoding
order, any
preceding TRAP picture with nuh layer id equal to layerId in decoding order
(when
present).
97
Date recue/date received 2021-10-26
PCTUS2021032924
(11) When the current sub-picture, with nuh layer id equal to a particular
value layerId and sub-picture index equal to a particular value subpicIdx, is
an TRAP
sub-picture, there shall be no picture referred to by an entry in RefPicList[
0 ] or
RefPicList[ 11 that precedes, in output order or decoding order, any preceding
picture, in decoding order (when present), containing an TRAP sub-picture with
nuh layer id equal to layerId and sub-picture index equal to subpicIdx.
(12) When the current picture is not a RASL picture associated with a CRA
picture with NoOutputBeforeRecoveryFlag equal to 1, there shall be no picture
referred to by an active entry in RefPicList[ 0 ] or RefPicList[ 11 that was
generated
by the decoding process for generating unavailable reference pictures for the
CRA
picture associated with the current picture.
(13) When the current sub-picture is not a RASL sub-picture associated
with a CRA sub-picture in a CRA picture with NoOutputBeforeRecoveryFlag equal
to 1, there shall be no picture referred to by an active entry in RefPicList[
0 ] or
RefPicList[ 11 that was generated by the decoding process for generating
unavailable
reference pictures for the CRA picture containing the CRA sub-picture
associated
with the current sub-picture.
(14) When the current picture, with nuh layer id equal to a particular value
layerId, is not any of the following, there shall be no picture referred to by
an entry in
RefPicList[ 0 ] or RefPicList[ 11 that was generated by the decoding process
for
generating unavailable reference pictures for the TRAP or GDR picture
associated
with the current picture:
(a) An IDR picture with sps idr rpl_present flag equal
to
1 or pps rpl info in_ph flag equal to 1.
98
Date recue/date received 2021-10-26
PCTUS2021032924
(b) A CRA picture with NoOutputBeforeRecoveryFlag
equal to 1.
(c) A picture, associated with a CRA picture with
NoOutputBeforeRecoveryFlag equal to 1, that precedes, in decoding
order, the leading pictures associated with the same CRA picture.
(d) A leading picture associated with a CRA picture with
NoOutputBeforeRecoveryFlag equal to 1.
(e) A GDR picture with NoOutputBeforeRecoveryF lag
equal to 1.
(0 A recovering picture of a GDR picture with
NoOutputBeforeRecoveryFlag equal to 1 and nuh layer id equal to
layerId.
(15) When the current sub-picture, with nuh layer id equal to a particular
value layerId and sub-picture index equal to a particular value subpicIdx, is
not any of
the following, there shall be no picture referred to by an entry in
RefPicList[ 0 ] or
RefPicList[ 11 that was generated by the decoding process for generating
unavailable
reference pictures for the TRAP or GDR picture containing the TRAP or GDR sub-
picture associated with the current sub-picture:
(a) An IDR sub-picture in an IDR picture with
sps idr rpl_present flag equal to 1 or pps rpl info in_ph flag equal
to 1.
(b) A CRA sub-picture in a CRA picture with
NoOutputBeforeRecoveryFlag equal to 1.
99
Date recue/date received 2021-10-26
PCTUS2021032924
(c) A sub-picture, associated with a CRA sub-picture in a
CRA picture with NoOutputBeforeRecoveryFlag equal to 1, that
precedes, in decoding order, the leading pictures associated with the
same CRA picture.
(d) A leading sub-picture associated with a CRA sub-
picture in a CRA picture with NoOutputBeforeRecoveryFlag equal to
1.
(e) A GDR sub-picture in a GDR picture with
NoOutputBeforeRecoveryFlag equal to 1.
(0 A subpicure in a recovering picture of a GDR
picture
with NoOutputBeforeRecoveryFlag equal to 1 and nuh layer id equal
to layerId.
(16) When the current picture follows an TRAP picture having the same
value of nuh layer id in both decoding order and output order, there shall be
no
picture referred to by an active entry in RefPicList[ 0 ] or RefPicList[ 11
that precedes
that TRAP picture in output order or decoding order.
(17) When the current sub-picture follows an TRAP sub-picture having the
same value of nuh layer id and the same value of sub-picture index in both
decoding
and output order, there shall be no picture referred to by an active entry in
RefPicList[ 0 ] or RefPicList[ 11 that precedes the picture containing that
TRAP sub-
picture in output order or decoding order.
(18) When the current picture follows an TRAP picture having the same
value of nuh layer id and the leading pictures, if any, associated with that
TRAP
picture in both decoding order and output order, there shall be no picture
referred to
100
Date recue/date received 2021-10-26
PCTUS2021032924
by an entry in RefPicList[ 0 ] or RefPicList[ 11 that precedes that TRAP
picture in
output order or decoding order.
(19) When the current sub-picture follows an TRAP sub-picture having the
same value of nuh layer id and the same value of sub-picture index and the
leading
sub-pictures, if any, associated with that TRAP sub-picture in both decoding
and
output order, there shall be no picture referred to by an entry in RefPicList[
0 ] or
RefPicList[ 11 that precedes the picture containing that TRAP sub-picture in
output
order or decoding order.
(20) When the current picture is a RADL picture, there shall be no active
entry in RefPicList[ 0 ] or RefPicList[ 11 that is any of the following:
(a) A RASL picture with
pps mixed nalu types in_pic flag is equal to 0. This means that an
active entry of the RPLs of a RADL picture can refer to a RASL
picture with pps mixed nalu types in_pic flag equal to 1. However,
when decoding starts from the associated CRA picture, such a RADL
picture can still be correctly decoded, because the RADL sub-
picture(s) in that referenced RASL picture would be correctly decoded,
as the RADL picture would only refer to the RADL sub-pictures in the
referenced RASL picture, as imposed by the next constraint that
disallows RADL sub-pictures referring to a RASL sub-picture.
(b) A picture that precedes the associated TRAP picture in
decoding order.
(21) When the current sub-picture, with nuh layer id equal to a particular
value layerId and sub-picture index equal to a particular value subpicIdx, is
a RADL
101
Date recue/date received 2021-10-26
PCTUS2021032924
sub-picture, there shall be no active entry in RefPicList[ 0 ] or RefPicList[
11 that is
any of the following:
(a) A picture with nuh layer id equal to layerId containing
a RASL sub-picture with sub-picture index equal to subpicIdx.
(b) A picture that precedes the picture containing the
associated TRAP sub-picture in decoding order.
(22) The following constraints apply for the picture referred to by each
ILRP entry, when present, in RefPicList[ 0 ] or RefPicList[ 11 of a slice of
the current
picture:
(a) The picture shall be in the same AU as the current
picture.
(b) The picture shall be present in the DPB.
(c) The picture shall have nuh layer id refPicLayerId less
than the nuh layer id of the current picture.
(d) Either of the following constraints applies: The picture
shall be an TRAP picture; and the picture shall have Temporalld less
than or equal to
Max( 0, vps max tid il ref_pics_plusl[ currLayerIdx ][ refLayerIdx ]
¨ 1), where currLayerIdx and refLayerIdx are equal to
GeneralLayerIdx[ nuh layer id ] and
GeneralLayerIdx[ refpicLayerId ], respectively.
(23) Each ILRP entry, when present, in RefPicList[ 0 ] or RefPicList[ 11 of
a slice shall be an active entry.
102
Date recue/date received 2021-10-26
PCTUS2021032924
(24) When vps independent layer flag[ GeneralLayerIdx[ nuh layer id ] ]
is equal to 0 and sps num subpics minusl is greater than 0, either of the
following
two conditions (but not both) shall be true:
(a) The picture referred to by each active entry in
RefPicList[ 0 ] or RefPicList[ 11 has the same sub-picture layout as
the current picture (i.e. the SPSs referred to by that picture and the
current picture have the same value of sps num subpics minusl and
the same values of sps subpic ctu top left x[ j ],
sps subpic ctu top left_y[ j ], sps subpic width minus1[ j ], and
sps subpic height minusl[ ii, respectively, for each value of j in the
range of 0 to sps num subpics minusl, inclusive).
(b) The picture referred to by each active entry in
RefPicList[ 0 ] or RefPicList[ 1 ] is an ILRP for which the value of
sps num subpics minusl is equal to 0.
[0365] According to embodiments, the decoder may check whether the above
constraints are met for bitstream conformance. The conformance check may be
done after
invoking decoding process for generating unavailable reference pictures as
described above.
[0366] Embodiments of the present disclosure may comprise at least one
processor
and memory storing computer code. The computer code, when executed by the at
least one
processor, may be configured to cause the at least one processor to perform
the functions of
the embodiments of the present disclosure.
[0367] For example, with reference to FIG. 33, a decoder of the present
disclosure
may comprise at least one processor and memory storing computer code (80). The
decoder
may be configured to receive a bitstream including at least one coded picture
and parameter
103
Date recue/date received 2021-10-26
PCTUS2021032924
sets (e.g., SPS and VPS), headers (e.g. picture headers and slice headers),
and AU delimiters.
The computer code may be configured to cause the at least one processor to
perform any
number of the decoding processes (e.g. constructing a reference picture list,
generating an
unavailable reference picture, and checking bitstream conformance) and aspects
related to
decoding (e.g. signaling of flags and other syntax elements in picture header,
slice header,
and access unit delimiter) as described in the present disclosure. For
example, the computer
code (80) may include a plurality of signaling code (81) and decoding code
(82).
[0368] The plurality of signaling code (81) may include various signaling
code that is
configured to cause the at least one processor to signal (and/or infer) flags
and other syntax
elements in picture header, slice header, and access unit delimiter.
[0369] The decoding code (82) may be configured to cause the at least one
processor
to decode one or more pictures. According to embodiments, the decoding code
(82) may
include constructing code (83), generating code (84), and checking code (85).
The
constructing code (83) may be configured to cause the at least one processor
to construct a
reference picture list(s). The generating code (84) may be configured to cause
the at least one
processor to generate an unavailable reference picture in the reference
picture list(s). The
checking code (85) may be configured to to cause the at least one processor to
check
bitstream conformance, for references pictures in the reference picture
list(s), that the
following constraints apply: (a) a number of entries that are indicated to be
in the reference
picture list is not less than a number of active entries that are indicated to
be in the reference
picture list, (b) each picture referred to by an active entry in the reference
picture list is
present in a decoded picture buffer (DPB) and has a temporal identifier value
less than or
equal to a temporal identifier value of the current picture, and (c) each
picture referred to by
an entry in the reference picture list is not the current picture and is
indicated by a picture
header flag to potentially be a reference picture.
104
Date recue/date received 2021-10-26
PCTUS2021032924
[0370] The techniques described above, can be implemented as computer
software
using computer-readable instructions and physically stored in one or more
computer-readable
media. For example, FIG. 34 shows a computer system (900) suitable for
implementing
embodiments of the disclosed subject matter.
[0371] The computer software can be coded using any suitable machine code
or
computer language, that may be subject to assembly, compilation, linking, or
like
mechanisms to create code comprising instructions that can be executed
directly, or through
interpretation, micro-code execution, and the like, by computer central
processing units
(CPUs), Graphics Processing Units (GPUs), and the like.
[0372] The instructions can be executed on various types of computers or
components
thereof, including, for example, personal computers, tablet computers,
servers, smartphones,
gaming devices, intemet of things devices, and the like.
[0373] The components shown in FIG. 34 for computer system (900) are
exemplary
in nature and are not intended to suggest any limitation as to the scope of
use or functionality
of the computer software implementing embodiments of the present disclosure.
Neither
should the configuration of components be interpreted as having any dependency
or
requirement relating to any one or combination of components illustrated in
the exemplary
embodiment of a computer system (900).
[0374] Computer system (900) may include certain human interface input
devices.
Such a human interface input device may be responsive to input by one or more
human users
through, for example, tactile input (such as: keystrokes, swipes, data glove
movements),
audio input (such as: voice, clapping), visual input (such as: gestures),
olfactory input (not
depicted). The human interface devices can also be used to capture certain
media not
necessarily directly related to conscious input by a human, such as audio
(such as: speech,
music, ambient sound), images (such as: scanned images, photographic images
obtain from a
105
Date recue/date received 2021-10-26
PCTUS2021032924
still image camera), video (such as two-dimensional video, three-dimensional
video including
stereoscopic video).
[0375] Input human interface devices may include one or more of (only one
of each
depicted): keyboard (901), mouse (902), trackpad (903), touch screen (910),
data-glove ,
joystick (905), microphone (906), scanner (907), and camera (908).
[0376] Computer system (900) may also include certain human interface
output
devices. Such human interface output devices may be stimulating the senses of
one or more
human users through, for example, tactile output, sound, light, and
smell/taste. Such human
interface output devices may include tactile output devices (for example
tactile feedback by
the touch-screen (910), data-glove , or joystick (905), but there can also be
tactile feedback
devices that do not serve as input devices). For example, such devices may be
audio output
devices (such as: speakers (909), headphones (not depicted)), visual output
devices (such as
screens (910) to include CRT screens, LCD screens, plasma screens, OLED
screens, each
with or without touch-screen input capability, each with or without tactile
feedback
capability¨some of which may be capable to output two dimensional visual
output or more
than three dimensional output through means such as stereographic output;
virtual-reality
glasses (not depicted), holographic displays and smoke tanks (not depicted)),
and printers
(not depicted).
[0377] Computer system (900) can also include human accessible storage
devices and
their associated media such as optical media including CD/DVD ROM/RW (920)
with
CD/DVD or the like media (921), thumb-drive (922), removable hard drive or
solid state
drive (923), legacy magnetic media such as tape and floppy disc (not
depicted), specialized
ROM/ASIC/PLD based devices such as security dongles (not depicted), and the
like.
106
Date recue/date received 2021-10-26
PCTUS2021032924
[0378] Those skilled in the art should also understand that term "computer
readable
media" as used in connection with the presently disclosed subject matter does
not encompass
transmission media, carrier waves, or other transitory signals.
[0379] Computer system (900) can also include interface to one or more
communication networks. Networks can for example be wireless, wireline,
optical. Networks
can further be local, wide-area, metropolitan, vehicular and industrial, real-
time, delay-
tolerant, and so on. Examples of networks include local area networks such as
Ethernet,
wireless LANs, cellular networks to include GSM, 3G, 4G, 5G, LTE and the like,
TV
wireline or wireless wide area digital networks to include cable TV, satellite
TV, and
terrestrial broadcast TV, vehicular and industrial to include CANBus, and so
forth. Certain
networks commonly require external network interface adapters that attached to
certain
general purpose data ports or peripheral buses (949) (such as, for example USB
ports of the
computer system (900); others are commonly integrated into the core of the
computer system
900 by attachment to a system bus as described below (for example Ethernet
interface into a
PC computer system or cellular network interface into a smartphone computer
system). Using
any of these networks, computer system (900) can communicate with other
entities. Such
communication can be uni-directional, receive only (for example, broadcast
TV), uni-
directional send-only (for example CANbus to certain CANbus devices), or bi-
directional, for
example to other computer systems using local or wide area digital networks.
Such
communication can include communication to a cloud computing environment
(955). Certain
protocols and protocol stacks can be used on each of those networks and
network interfaces
as described above.
[0380] Aforementioned human interface devices, human-accessible storage
devices,
and network interfaces (954) can be attached to a core (940) of the computer
system (900).
107
Date recue/date received 2021-10-26
PCTUS2021032924
[0381] The core (940) can include one or more Central Processing Units
(CPU) (941),
Graphics Processing Units (GPU) (942), specialized programmable processing
units in the
form of Field Programmable Gate Areas (FPGA) (943), hardware accelerators
(944) for
certain tasks , and so forth. These devices, along with Read-only memory (ROM)
(945),
Random-access memory (946), internal mass storage such as internal non-user
accessible
hard drives, SSDs, and the like (947), may be connected through a system bus
(948). In some
computer systems, the system bus (948) can be accessible in the form of one or
more physical
plugs to enable extensions by additional CPUs, GPU, and the like. The
peripheral devices can
be attached either directly to the core's system bus (948), or through a
peripheral bus (949).
Architectures for a peripheral bus include PCI, USB, and the like. A graphics
adapter 950
may be included in the core 940.
[0382] CPUs (941), GPUs (942), FPGAs (943), and accelerators (944) can
execute
certain instructions that, in combination, can make up the aforementioned
computer code.
That computer code can be stored in ROM (945) or RAM (946). Transitional data
can be also
be stored in RAM (946), whereas permanent data can be stored for example, in
the internal
mass storage (947). Fast storage and retrieve to any of the memory devices can
be enabled
through the use of cache memory, that can be closely associated with one or
more CPU (941),
GPU (942), mass storage (947), ROM (945), RAM (946), and the like.
[0383] The computer readable media can have computer code thereon for
performing
various computer-implemented operations. The media and computer code can be
those
specially designed and constructed for the purposes of the present disclosure,
or they can be
of the kind well known and available to those having skill in the computer
software arts.
[0384] As an example and not by way of limitation, the computer system
having
architecture (900), and specifically the core (940) can provide functionality
as a result of
processor(s) (including CPUs, GPUs, FPGA, accelerators, and the like)
executing software
108
Date recue/date received 2021-10-26
PCTUS2021032924
embodied in one or more tangible, computer-readable media. Such computer-
readable media
can be media associated with user-accessible mass storage as introduced above,
as well as
certain storage of the core (940) that are of non-transitory nature, such as
core-internal mass
storage (947) or ROM (945). The software implementing various embodiments of
the present
disclosure can be stored in such devices and executed by core (940). A
computer-readable
medium can include one or more memory devices or chips, according to
particular needs. The
software can cause the core (940) and specifically the processors therein
(including CPU,
GPU, FPGA, and the like) to execute particular processes or particular parts
of particular
processes described herein, including defining data structures stored in RAM
(946) and
modifying such data structures according to the processes defined by the
software. In addition
or as an alternative, the computer system can provide functionality as a
result of logic
hardwired or otherwise embodied in a circuit (for example: accelerator (944)),
which can
operate in place of or together with software to execute particular processes
or particular parts
of particular processes described herein. Reference to software can encompass
logic, and vice
versa, where appropriate. Reference to a computer-readable media can encompass
a circuit
(such as an integrated circuit (IC)) storing software for execution, a circuit
embodying logic
for execution, or both, where appropriate. The present disclosure encompasses
any suitable
combination of hardware and software.
[0385] While
this disclosure has described several non-limiting example
embodiments, there are alterations, permutations, and various substitute
equivalents, which
fall within the scope of the disclosure. It will thus be appreciated that
those skilled in the art
will be able to devise numerous systems and methods which, although not
explicitly shown
or described herein, embody the principles of the disclosure and are thus
within the spirit and
scope thereof.
109
Date recue/date received 2021-10-26
