U.S. patent application number 12/574376 was filed with the patent office on 2010-04-08 for method and an apparatus for processing a video signal.
This patent application is currently assigned to LG ELECTRONICS INC.. Invention is credited to Young Hee CHOI, Byeong Moon JEON, Jung Sun KIM, Joon Young PARK, Seung Wook PARK.
Application Number | 20100086051 12/574376 |
Document ID | / |
Family ID | 42075799 |
Filed Date | 2010-04-08 |
United States Patent
Application |
20100086051 |
Kind Code |
A1 |
PARK; Seung Wook ; et
al. |
April 8, 2010 |
METHOD AND AN APPARATUS FOR PROCESSING A VIDEO SIGNAL
Abstract
A method of processing a video signal is disclosed. The present
invention includes determining whether a partition of a current
macroblock according to a current macroblock type is coded in a
skip mode, deriving a motion vector of the partition if the
partition of the current macroblock is coded in the skip mode, and
predicting a pixel value of the partition based on the derived
motion vector. Accordingly, the present invention obtains a pixel
value of a reference block close to an original video using a
motion vector by deriving the motion vector by a block unit smaller
than a size of the macroblock, thereby enhancing a reconstruction
ratio of the macroblock.
Inventors: |
PARK; Seung Wook; (Seoul,
KR) ; KIM; Jung Sun; (Seoul, KR) ; CHOI; Young
Hee; (Seoul, KR) ; JEON; Byeong Moon; (Seoul,
KR) ; PARK; Joon Young; (Seoul, KR) |
Correspondence
Address: |
FISH & RICHARDSON P.C.
P.O. BOX 1022
MINNEAPOLIS
MN
55440-1022
US
|
Assignee: |
LG ELECTRONICS INC.
Seoul
KR
|
Family ID: |
42075799 |
Appl. No.: |
12/574376 |
Filed: |
October 6, 2009 |
Related U.S. Patent Documents
|
|
|
|
|
|
Application
Number |
Filing Date |
Patent Number |
|
|
61103221 |
Oct 6, 2008 |
|
|
|
61119697 |
Dec 3, 2008 |
|
|
|
61120060 |
Dec 5, 2008 |
|
|
|
61143813 |
Jan 11, 2009 |
|
|
|
61154420 |
Feb 22, 2009 |
|
|
|
61159058 |
Mar 10, 2009 |
|
|
|
Current U.S.
Class: |
375/240.16 ;
375/E7.123 |
Current CPC
Class: |
H04N 19/132 20141101;
H04N 19/70 20141101; H04N 19/573 20141101; H04N 19/44 20141101;
H04N 19/176 20141101; H04N 19/513 20141101; H04N 19/61 20141101;
H04N 19/46 20141101 |
Class at
Publication: |
375/240.16 ;
375/E07.123 |
International
Class: |
H04N 7/32 20060101
H04N007/32 |
Foreign Application Data
Date |
Code |
Application Number |
Oct 6, 2009 |
KR |
10-2009-0094486 |
Claims
1. A method of processing a video signal, comprising the steps of:
determining whether a partition of a current macroblock according
to a current macroblock type is coded in a skip mode; deriving a
motion vector of the partition if the partition of the current
macroblock is coded in the skip mode; and predicting a pixel value
of the partition based on the derived motion vector.
2. The method of claim 1, the determining step further comprising
the step of obtaining the current macroblock type.
3. The method of claim 2, further comprising the step of obtaining
skip mode indication information on the partition of the current
macroblock.
4. The method of claim 3, wherein the partition of the current
macroblock is obtained based on size information of a predicted
block according to the current macroblock type.
5. The method of claim 3, wherein the skip mode indication
information comprises information indicating whether the partition
of the current macroblock is coded in the skip mode or not.
6. The method of claim 1, wherein the motion vector of the
partition is derived from a medium value of a motion vector of a
block neighbor to the partition of the current macroblock.
7. The method of claim 6, wherein the block neighbor to the
partition of the current macroblock includes left, top and right
top blocks of the partition.
8. The method of claim 1, wherein the pixel value of the partition
comprises a pixel value of a reference block referred to by the
motion vector.
Description
CROSS-REFERENCE TO RELATED APPLICATION
[0001] This application claims the benefit of U.S. Provisional
Application No. 61/103,221 filed on Oct. 6, 2008, U.S. Provisional
Application No. 61/119,697, filed on Dec. 3, 2008, U.S. Provisional
Application No. 61/120,060, filed on Dec. 5, 2008, U.S. Provisional
Application No. 61/143,813, filed on Jan. 11, 2009, U.S.
Provisional Application No. 61/154,420 filed on Feb. 22, 2009, U.S.
Provisional Application No. 61/159,058 filed on Mar. 10, 2009,
Korean Application No. 10-2009-0094486 filed on Oct. 6, 2009, which
are hereby incorporated by reference.
BACKGROUND OF THE INVENTION
[0002] 1. Field of the Invention
[0003] The present invention relates to an apparatus for processing
a video signal and method thereof, for which inter picture
prediction is used.
[0004] 2. Discussion of the Related Art
[0005] Generally, in transmitting an encoded video signal to a
decoder, a method of removing spatial redundancy and temporal
redundancy, i.e., intra or inter picture prediction is performed to
enhance a compression ratio of the video signal.
SUMMARY OF THE INVENTION
[0006] Accordingly, the present invention is directed to an
apparatus for processing a video signal and method thereof that
substantially obviate one or more of the problems due to
limitations and disadvantages of the related art.
[0007] An object of the present invention is to provide an
apparatus for processing a video signal and method thereof, by
which a video signal can be efficiently processed using inter
picture prediction.
[0008] Another object of the present invention is to provide an
apparatus for processing a video signal and method thereof, by
which a video signal can be efficiently processed in a manner of
performing inter picture prediction using a skip mode.
[0009] A further object of the present invention is to provide an
apparatus for processing a video signal and method thereof, by
which a video signal can be efficiently processed using a direct
mode.
[0010] To achieve these and other advantages, the present invention
proposes a video signal processing method characterized in
configuring various motion vector candidates in performing inter
picture prediction and obtaining a precise motion vector from the
various motion vector candidates.
[0011] The present invention proposes a video signal processing
method characterized in indicating whether a region having a
predetermined size within a macroblock is coded in a skip mode.
[0012] The present invention proposes a video signal processing
method characterized in indicating whether a partition of a
macroblock according to macroblock type information is coded in a
skip mode.
[0013] The present invention proposes a video signal processing
method characterized in proposing a macroblock type for indicating
a skip mode or a direct mode in part within one macroblock and
performing inter picture prediction using the skip mode or the
direct mode based on the macroblock type.
[0014] The present invention proposes a video signal processing
method characterized in deriving a motion vector by a block unit
smaller than a size of a macroblock if the macroblock is coded in a
skip mode.
[0015] And, the present invention proposes a video signal
processing method characterized in defining skip mode indication
information or direct mode indication information, indicating
whether a macroblock is coded in a skip mode, and determining
whether it is a skip mode or a direct mode using CBP (coded block
pattern) information.
[0016] Accordingly, the present invention provides the following
effects and/or advantages.
[0017] First of all, a precise motion vector is obtained from
motion vector candidates for inter picture prediction. Hence, a bit
size of a motion vector difference value to be coded can be reduced
and accuracy in reconstructing an original video can be enhanced.
In case that a skip mode or a direct mode is used for a macroblock
greater than 16.times.16, by means of using a skip mode or a direct
mode by a block unit smaller than a size of the macroblock, error
of inter picture prediction using a skip mode or a direct mode can
be reduced and coding efficiency can be enhanced. By means of using
a macroblock type for indicating a skip mode or a direct mode
within one macroblock partially, it is not necessary to code
information indicating a skip mode or a direct mode for a
predetermined-size region within the macroblock and it is also able
to reduce a bit size to be coded by coding the macroblock type
information only. In case of coding a macroblock in a skip mode, by
a motion vector is derived by a block unit smaller than a size of
the macroblock, it is able to obtain a pixel value of a reference
block close to an original video using the motion vector.
Accordingly, it is able to improve a reconstruction ratio of a
macroblock. As a macroblock uses a skip mode or a direct mode, it
is not necessary to separately code skip mode indication
information and direct mode indication information separately for a
macroblock. Since a skip mode or a direct mode is determined using
already-coded CBP information, it is able to reduce an information
size to be coded. Therefore, efficiency of video signal processing
can be enhanced.
BRIEF DESCRIPTION OF THE DRAWINGS
[0018] The accompanying drawings, which are included to provide a
further understanding of the invention and are incorporated in and
constitute a part of this specification, illustrate embodiments of
the invention and together with the description serve to explain
the principles of the invention.
[0019] In the drawings:
[0020] FIG. 1 is a schematic block diagram of a video signal
decoding apparatus according to the present invention;
[0021] FIG. 2 is a diagram for a method of applying a skip mode by
a block unit smaller than a size of a macroblock according to one
embodiment of the present invention;
[0022] FIG. 3 is a block diagram for a method of indicating a
portion that uses a skip mode within a macroblock based on
macroblock type information;
[0023] FIG. 4 is a diagram for a method of indicating whether a
current macroblock is partially coded in a skip mode at a
sub-macroblock level according to one embodiment of the present
invention;
[0024] FIG. 5 is a diagram for a method of indicating whether a
current macroblock is partially coded in a skip mode according to
another embodiment of the present invention;
[0025] FIG. 6 and FIG. 7 are diagrams of a macroblock type for
indicating a partial skip mode according to one embodiment of the
present invention;
[0026] FIG. 8 is a diagram for a method of deriving a motion vector
in case that a macroblock type is a P skip macroblock according to
one embodiment of the present invention;
[0027] FIG. 9 is a schematic internal block diagram of an inter
prediction unit 700 having a spatial direction prediction mode
applied thereto; and
[0028] FIG. 10 is a block diagram of a process for decoding slice
data using skip or direct mode indication information.
DETAILED DESCRIPTION OF THE INVENTION
[0029] Additional features and advantages of the invention will be
set forth in the description which follows, and in part will be
apparent from the description, or may be learned by practice of the
invention. The objectives and other advantages of the invention
will be realized and attained by the structure particularly pointed
out in the written description and claims thereof as well as the
appended drawings. [0030] To achieve these and other advantages and
in accordance with the purpose of the present invention, as
embodied and broadly described, a method of processing a video
signal includes the steps of determining whether a partition of a
current macroblock according to a current macroblock type is coded
in a skip mode, deriving a motion vector of the partition if the
partition of the current macroblock is coded in the skip mode, and
predicting a pixel value of the partition based on the derived
motion vector. [0031] Preferably, the present invention is
characterized in that the determining step further includes the
step of obtaining the current macroblock type. [0032] More
preferably, the present invention is characterized in that the
method further includes the step of obtaining skip mode indication
information on the partition of the current macroblock. [0033] And,
the present invention is characterized in that the partition of the
current macroblock is obtained based on size information of a
predicted block according to the current macroblock type. [0034]
Moreover, the present invention is characterized in that the skip
mode indication information includes information indicating whether
the partition of the current macroblock is coded in the skip mode
or not. [0035] Preferably, the present invention is characterized
in that the motion vector of the partition is derived from a medium
value of a motion vector of a block neighbor to the partition of
the current macroblock. [0036] More preferably, the present
invention is characterized in that the block neighbor to the
partition of the current macroblock includes left, top and right
top blocks of the partition. [0037] Preferably, the present
invention is characterized in that the pixel value of the partition
includes a pixel value of a reference block referred to by the
motion vector.
[0038] It is to be understood that both the foregoing general
description and the following detailed description are exemplary
and explanatory and are intended to provide further explanation of
the invention as claimed.
[0039] Reference will now be made in detail to the preferred
embodiments of the present invention, examples of which are
illustrated in the accompanying drawings. First of all,
terminologies or words used in this specification and claims are
not construed as limited to the general or dictionary meanings and
should be construed as the meanings and concepts matching the
technical idea of the present invention based on the principle that
an inventor is able to appropriately define the concepts of the
terminologies to describe the inventor's invention in best way. The
embodiment disclosed in this disclosure and configurations shown in
the accompanying drawings are just one preferred embodiment and do
not represent all technical idea of the present invention.
Therefore, it is understood that the present invention covers the
modifications and variations of this invention provided they come
within the scope of the appended claims and their equivalents at
the timing point of filing this application.
[0040] First of all, a macroblock is a basic unit for processing a
video frame and can be defined as a region constructed with
16.times.16 pixels. A macroblock greater than 16.times.16 can
becomes a basic unit for processing a video frame having resolution
higher than that of the related art. For instance, this macroblock
can be defined as a region constructed with 32.times.32 pixels or
64.times.64 pixels for examples.
[0041] A macroblock greater than 16.times.16 can be adaptively used
according to resolution of a video image. For instance, if
resolution of a video image is equal to or smaller than VGA, it is
able to set a size of a macroblock to 16.times.16. Accordingly, one
of 16.times.8, 8.times.16, 4.times.8, 8.times.4 and 4.times.4
blocks is usable as a partition of a macroblock. If resolution of a
video image is equal to or greater than VGA and equal to or smaller
than 1080 P, it is able to set a size of a macroblock to
32.times.32. Accordingly, one of 32.times.16, 16.times.32,
16.times.16, 16.times.8, 8.times.16 and 8.times.8 blocks is usable
as a partition of a macroblock. If resolution of a video image is
equal to or greater than 1080 P and equal to or smaller than 4
k.times.2 k, it is able to set a size of a macroblock to
64.times.64. Likewise, one of 64.times.32, 32.times.64,
32.times.32, 32.times.16, 16.times.32 and 16.times.16 blocks is
usable as a partition of a macroblock. Yet, an encoder is able to
determine an optimal size of one of a 16.times.16 macroblock and
other macroblocks greater than 16.times.16 according to resolution
of a video image. The determined macroblock size information can be
transmitted to a decoder at a slice or sequence level.
[0042] A macroblock type can be defined according to a prediction
mode of a macroblock, i.e., according to an intra picture
prediction mode or an inter picture prediction mode. This can
include size information of a predicted block. And, a macroblock
type for performing inter picture prediction can include
information on a prediction direction as well.
[0043] A size of a predicted block coded in an intra picture
prediction mode can be set to one of 16.times.16 pixels, 8.times.8
pixels and 4.times.4 pixels. The predicted block can means a unit
of a block for performing intra picture prediction or inter picture
prediction within a picture. For the motion compensation of a
predicted block coded in the inter picture prediction mode, one of
16.times.16 pixels, 16.times.8 pixels, 8.times.16 pixels and
8.times.8 pixels is usable as a size of the predicted block. In
case that 8.times.8 pixels are used as a size of a predicted block,
it is able to perform inter picture prediction by a unit of
partition (e.g., 8.times.4 pixels, 4.times.8 pixels and 4.times.4
pixels) of the 8.times.8 pixels predicted block.
[0044] In case that resolution is equal to or greater than VGA and
equal to or smaller than 1080 P, a size of a predicted block coded
in an intra picture prediction mode can be set to 16.times.16
pixels or 8.times.8 pixels. And, a size of a predicted block coded
in an inter picture prediction mode can be set to one of
16.times.16 pixels, 16.times.8 pixels, 8.times.16 pixels and
8.times.8 pixels. In this case, inter picture prediction may not be
performed by a unit of a partition (e.g., 8.times.4 pixels,
4.times.8 pixels and 4.times.4 pixels) of an 8.times.8 pixels
predicted block. In case that resolution is equal to or greater
than 1080 P and equal to or smaller than 4 k.times.2 k, a size of a
predicted block coded in an intra picture prediction mode or an
inter picture prediction mode can be set to 16.times.16 pixels.
[0045] Likewise, an encoder is able to determine an optimal
macroblock type according to resolution of a video image. This
determined macroblock type information can be sent to a decoder at
a slice level or a sequence level. Accordingly, it is able to
adaptively use the determined macroblock type.
[0046] FIG. 1 is a schematic block diagram of a video signal
decoding apparatus according to the present invention.
[0047] Referring to FIG. 1, the decoding apparatus mainly includes
a parsing unit 100, an entropy decoding unit 200, a inverse
quantizing/inverse transform unit 300, an intra prediction unit
400, a deblocking filtering unit 500, a decoded picture buffering
unit 600, an inter prediction unit 700 and the like.
[0048] The parsing unit 100 performs parsing by NAL unit to decode
a received video image. Generally, at least one or more sequence
parameter sets and picture parameter sets are sent to a decoder
before a slice header and data are decoded. In this case, various
kinds of attribute informations can be included in a NAL header
region or an extension region of a NAL header. A parsed bitstream
is entropy-decoded via the entropy decoding unit 200 and a
coefficient of each macroblock, a motion vector and the like are
then extracted. The inverse quantizing/inverse transform unit 300
obtains a coefficient value by multiplying a received quantized
value by a predetermined constant and then reconstructs a pixel
value by inverse-transforming the coefficient value. The intra
prediction unit 400 performs intra picture prediction from a
decoded sample within a current picture using the reconstructed
pixel value. Meanwhile, the deblocking filtering unit 500 is
applied to each coded macroblock to reduce block distortion. A
filter enhances an image quality of a frame by smoothing edges of a
block. Selection for a filtering process depends on a variation of
an image sample. Filtered pictures are outputted or stored in the
decoded picture buffer unit 600 to be used as a reference picture.
The decoded picture buffer unit 600 plays a role in storing or
opening previously coded pictures to perform inter picture
prediction. In this case, in order to store or open them, frame_num
of each picture and POC (picture order count) are used. The inter
prediction unit 700 performs inter picture prediction using the
reference pictures stored in the decoded picture buffer unit 600. A
macroblock coded in an inter mode can be divided into macroblock
partitions. And, each of the macroblock partitions can be predicted
from one or two reference pictures. The inter prediction unit 700
compensates for a motion of a current block using informations
transmitted from the entropy decoding unit 200. In the following
description, it is understood that a block includes a macroblock or
a partition of a macroblock. Motion vectors of blocks neighbor to a
current block are extracted from a video signal. It is then able to
obtain a motion vector predicted value of the current block from
the neighbor block. The neighbor block can include a block at a
left, top or right top side of the current block. For instance, it
is able to set a predicted value of a motion vector of a current
block to be obtained using medium values of horizontal and vertical
components of motion vectors of neighbor blocks. Alternatively, in
case that a left block of a current block has at least one
predicted block coded in an inter picture prediction mode, it is
able to set a predicted value of a motion vector of a current block
to be obtained using a motion vector of a predicted block located
at a top side. In case that a top side block of a current block has
at least one predicted block coded in an inter picture prediction
mode, it is able to set a predicted value of a motion vector of the
current block to be obtained using a motion vector of a predicted
block located at a most left side. In case that blocks located at
top and right sides of a current block among neighbor blocks are
located outside a boundary of a picture or slice, it is able to set
a predicted value of a motion vector of the current block to a
motion vector of a left block. If there exists one block having the
same reference picture index of a current block among neighbor
blocks, it is able to set a motion vector of the corresponding
block to be used for prediction.
[0049] Based on a partition of a macroblock according to a
macroblock type, it is able to obtain a predicted value of a motion
vector. For instance, when a partition of the current macroblock
according to a current macroblock type is 8.times.16, if an index
of the partition of the macroblock is set to 0, it is able to set a
top side block of a current block to be used for prediction. If an
index of the partition of the macroblock is set to 1, it is able to
set a left side block of a current block to be used for
prediction.
[0050] In obtaining a motion vector predicted value of a current
block, it is able to define various kinds of motion vectors
extractable from neighbor blocks as candidates for a motion vector
predicted value. As mentioned in the foregoing description, the
candidates for the motion vector predicted value can include a
motion vector of one of neighbor blocks, a medium value of motion
vectors of neighbor blocks and the like, by which the candidates
are non-limited. Accordingly, an encoder selects a motion vector
having best efficiency as a motion vector predicted value from the
candidates for the motion vector predicted value and is then able
to send the selected motion vector information to a decoder. The
selected motion vector information can be obtained from a slice
level of a macroblock level. It is able to define the candidate of
the motion vector predicted value using an index. And, the selected
motion vector information can include this index.
[0051] By coding a difference vector between a motion vector of a
current block and a predicted value of a motion vector, it is able
to reduce a size of bits to be coded. Using the obtained motion
vector predicted value and the difference vector extracted from a
video signal, the motion of the current block is compensated.
Moreover, this motion compensation can be performed using one
reference picture or a plurality of pictures.
[0052] In case that a size of a macroblock is 16.times.16, total 7
kinds of block sizes including 16.times.16, 16.times.8, 8.times.16,
8.times.8, 8.times.4, 4.times.8, and 4.times.4 are available for
inter picture prediction. They can be hierarchically represented as
macroblock or sub-macroblock types. In particular, a size of a
predicted block can be represented as a macroblock type and one of
16.times.16, 16.times.8, 8.times.16 and 8.times.8 can be
selected.
[0053] A reference picture index and a motion vector can be coded
for each predicted block. Alternatively, a motion vector can be
coded for each predicted block. Of course, a coding scheme for a
sub-macroblock can be determined in the same manner of a
macroblock.
[0054] In reconstructing a current macroblock, a skip macroblock
can be defined as using a pixel value of a macroblock intactly
within a previously coded reference picture instead of having
information on the current macroblock, i.e., a motion vector, a
reference picture index and a residual data. In coding a skip
macroblock, a motion compensation using a motion vector can be
accompanied. This motion vector can be derived using a neighbor
block. The former motion vector deriving method is identically
applicable to a method of deriving the motion vector.
[0055] A process for coding slice data using a skip mode is
explained as follows.
[0056] First of all, it is able to obtain flag information
indicating whether a current macroblock within a slice is a skip
macroblock.
[0057] In case of a macroblock greater than 16.times.16, it is able
to apply a skip mode by a block unit smaller than a size of the
macroblock by indicating a partition of the macroblock that uses a
skip mode. Hence, it is able to enhance efficiency of coding.
[0058] FIG. 2 is a diagram for a method of applying a skip mode by
a block unit smaller than a size of a macroblock according to one
embodiment of the present invention.
[0059] Referring to FIG. 2, in case that a size of a macroblock is
32.times.32, it is able to use flag information indicating whether
it is a skip macroblock by 16.times.16 block unit. If a flag
indicating whether it is the skip macroblock is set to 1, it is
able to define the macroblock is in a skip mode. In case that a
partition of a macroblock located at a left top side within a
macroblock is coded in a skip mode only, four bits `1000` can be
transmitted. If all partitions of 16.times.16 macroblock within a
macroblock are coded in a skip mode, other information such as a
motion vector and the like is not parsed. Yet, if the macroblock
partially includes the partition of the macroblock coded in the
skip mode, a macroblock type will be parsed. In this case, it can
be represented as a macroblock type having a partition of
16.times.16 macroblock.
[0060] FIG. 3 is a block diagram for a method of indicating a
portion that uses a skip mode within a macroblock based on
macroblock type information.
[0061] Referring to FIG. 3, flag information indicating whether a
current block is a skip macroblock at a slice level is parsed
[S300]. As a result of the parsing, if the current macroblock is
not the skip macroblock [`no 1`], current macroblock type
information will be parsed at a macroblock level [S310]. If the
current macroblock is not the skip macroblock, some of partitions
of the current macroblock may be coded in a skip mode. Hence, if
the current macroblock is the skip macroblock [`yes 1`], it is able
to perform decoding without parsing motion information of the
current macroblock and the like [S340]. If the current macroblock
is not the skip macroblock, a current macroblock type is parsed and
partition information of the current macroblock can be then
obtained from the current macroblock type. It is able to parse flag
information indicating whether a partition of the current
macroblock is coded in a skip mode [S320]. When a size of a current
macroblock is 32.times.32, as a result of parsing a current
macroblock type at a macroblock level, if the current macroblock
type is represented as 16.times.16, it is able to parse flag
information indicating whether partitions of the 16.times.16
macroblock are coded in skip mode. According to the flag
information, decoding will be performed on the skip mode coded
partition [`yes 2`] without parsing motion information and the
like. And, motion information and residual data will b parsed for
the partition [`no 2`] that is not coded in the skip mode [S330].
This can be identically applied to a sub-macroblock level.
[0062] FIG. 4 is a diagram for a method of indicating whether a
current macroblock is partially coded in a skip mode at a
sub-macroblock level according to one embodiment of the present
invention. Referring to FIG. 4, if a size of a current macroblock
is 32.times.32, it is able to set a size of a sub-macroblock to
16.times.16. As a result of parsing current macroblock type
information, if a partition of a macroblock according to a current
macroblock type is 16.times.16, it is able to parse flag
information indicating whether four 16.times.16 sub-macroblocks are
coded in skip mode [S400]. As a result of this parsing, it is able
to perform decoding on a skip-mode coded sub-macroblock without
parsing separate information [S410]. Hence, it is able to parse a
motion vector and reference picture index of this sub-macroblock
[S420]. Yet, the motion vector of the sub-macroblock can be parsed
based on a partition off the sub-macroblock according to a
sub-macroblock type.
[0063] FIG. 5 is a diagram for a method of indicating whether a
current macroblock is partially coded in a skip mode according to
another embodiment of the present invention.
[0064] Referring to FIG. 5, it is able to parse flag information
indicating whether a current macroblock within a slice is a skip
macroblock [S500]. If the current macroblock is the skip macroblock
according to the flag information [`yes 1`], it is able to perform
decoding without parsing motion information and the like [S550].
Yet, if the current macroblock is not the skip macroblock [`no 1`],
it is able to parse flag information indicating a presence or
non-presence of a partition of the current macroblock and a
presence or non-presence of skip mode coding of the partition
[S510].
[0065] For instance, in case that a size of a macroblock is
32.times.32, if the flag is set to 1, it is able to define that the
partition of the macroblock is 16.times.16. And, it is also able to
define that at least one of partitions of the macroblock is coded
in a skip mode. On the contrary, if the flag is set to 0, it is
able to define that the partition of the macroblock is not
16.times.16 or define that all partitions of the macroblock are not
coded in the skip mode despite 16.times.16.
[0066] In the following description, assume that the flag is
defined in the above manner. Hence, as a result of parsing the flag
information, if `1` is obtained [`yes 2`], as a partition of the
macroblock is set to 16.times.16, it is not necessary to parse the
macroblock type information. And, it is able to parse flag
information indicating whether the partition of the 16.times.16
macroblock is coded in the skip mode for four 16.times.16
macroblock partitions [S560]. Hence, if the partition of the
16.times.16 macroblock is coded in the skip mode [`yes 3`], it is
able to perform decoding without parsing motion information and the
like [S550]. Moreover, if the partition is coded in the skip mode,
CBP information may not be parsed. Accordingly, by reducing a size
of transmitted information, it is able to enhance efficiency of
signal processing. Yet, if not coded in the skip mode [`no 3`],
motion information and residual are parsed [S540] and it is able to
decode using the parsing result [S550].
[0067] As a result of parsing the flag information, if 0 is
obtained [`no 2`], it is able to parse the current macroblock type
information [S520]. If a partition of the current macroblock
according to a current macroblock type is not 16.times.16, it is
able to parse flag information indicating whether it is coded in a
skip mode [S530]. Yet, if a partition of the current macroblock
according to a current macroblock type is 16.times.16, as all four
partitions are not coded in the skip mode, the flag information
indicating whether the partition is coded in the skip mode may not
be parsed. And, it is able to set the information indicating
whether the partition is coded in the skip mode to be derived into
0.
[0068] If the partition is coded in the skip mode according to the
flag information indicating whether each of the obtained partitions
of the macroblock is coded in the skip mode [`yes 3`], it is able
to perform decoding without parsing motion information and the
like. If the partition is not coded in the skip mode [`no 3`], it
is able to perform decoding by parsing motion information and
residual [S540] [S550].
[0069] In another way, it is able to newly define a macroblock type
for a skip mode indication. FIG. 6 and FIG. 7 are diagrams of a
macroblock type for indicating a partial skip mode according to one
embodiment of the present invention.
[0070] In case that a size of a macroblock is greater than
16.times.16, it is able to define a macroblock type that indicates
whether a skip mode is used by a partition unit of the
macroblock.
[0071] With reference to FIG. 6, explained in the following
description is a case that a skip mode is used by a partition unit
of a macroblock on the assumption that a size of a macroblock is
32.times.32. As mentioned in the foregoing description, in case
that all partitions of the 16.times.16 macroblock within a current
macroblock are coded in the skip mode, it is able to obtain flag
information indicating that a current macroblock is a skip
macroblock at a slice level. On the contrary, in case that all
partitions of the 16.times.16 macroblock are not coded in the skip
mode, flag information indicating that a current macroblock is not
a skip macroblock and the current macroblock type can be then
parsed. In this case, if some of the partitions of the 16.times.16
macroblock are coded in the skip mode in part, a current macroblock
type can be represented as a macroblock type proposed in FIG. 7. In
particular, partitions (partition indexes 0 and 2 of a macroblock)
of the 16.times.16 macroblock are coded in the skip mode and
partitions (partition indexes 1 and 3 of a macroblock) of the
16.times.16 macroblock are not coded in the skip mode, a current
macroblock type will become 5 [S600]. In this case, the partitions
(partition indexes 0 and 2 of a macroblock) of the 16.times.16
macroblock can be decoded without parsing other information such as
motion information and the like. Yet, motion information, reference
picture indexes and the like will be parsed for the partitions
(partition indexes 1 and 3 of a macroblock) of the 16.times.16
macroblock. Each of the partitions (partition indexes 0 and 2 of a
macroblock) of the 16.times.16 macroblock can be set to a size of
16.times.32 predicted block or a size of 16.times.16 predicted
block.
[0072] A method of deriving a motion vector of a skip macroblock is
explained as follows. And, a motion vector of a sub-macroblock can
be derived in the same manner.
[0073] First of all, if a current block is a skip macroblock, it is
able to derive a motion vector using a neighbor block of the
current block.
[0074] For instance, assume that a block at a left side of a
current block, a block at a top side and a block at a right top
side are named a neighbor block A, a neighbor block B and a
neighbor block C, respectively. And, assume that their motion
vectors are set to mvA, mvB and mvC, respectively. A motion vector
of the current block can be derived into medium values of
horizontal and vertical components of the mvA, mvB and mvC or an
average value of the mvA, mvB and mvC. It can be derived into one
of three motion vectors and this can be determined based on a
reference picture index. Moreover, in case of a macroblock greater
than 16.times.16, it is able to obtain a precise motion vector of a
macroblock by deriving a motion vector by a block unit smaller than
a size of the macroblock or the sub-macroblock.
[0075] FIG. 8 is a diagram for a method of deriving a motion vector
in case that a macroblock type is a P skip macroblock according to
one embodiment of the present invention.
[0076] Referring to FIG. 8, in case that a size of a macroblock is
32.times.32, it is able to derive a motion vector by a block unit
smaller than a size of the macroblock, i.e., by 8.times.8 block
unit. In particular, a motion vector of a block 0 can be derived
into an medium value of horizontal and vertical components of mvA0,
mvB0 and mvC0, an average value of mvA0, mvB0 and mvB1, or one
motion vector among mvA0, mvB0 and mvB1. Likewise, a motion vector
of a block 1 can be derived into a medium value of horizontal and
vertical component of mvB0, mvB1 and mvB2. Index of the block may
mean a decoding order.
[0077] Inter picture prediction using a direct mode is explained as
follows. First of all, a direct mode indicates a mode for
predicting motion information of a current block from motion
information of a reconstructed block. The motion information may
include a motion vector and a reference picture index. For
instance, it is able to define a picture having a smallest
reference index among List1 reference pictures as an anchor picture
in a direct mode. And, a block of the anchor picture co-located at
the same spatial position of the current block can be defined as an
anchor block.
[0078] For instance, in a temporal prediction mode, it is able to
predict motion information of a current block using motion
information of the anchor block. It is able to define a motion
vector in a List0 direction of the anchor block as mvCol. In this
case, if a motion vector in the List0 direction of the anchor block
does not exist but a motion vector in List1 direction exists, it is
able set a motion vector in the List1 direction to mvCol. In this
case, in case of a picture B, it is able to two random pictures can
be used as reference pictures irrespective of temporal or spatial
sequence. In doing so, the used prediction is named List0
prediction or List1 prediction. For instance, the List0 prediction
means prediction for a forward direction, while the List1
prediction means prediction for a backward direction.
[0079] In a temporal direct mode, List0 reference picture of a
current block becomes a picture referred to by the mvCol and List1
reference picture becomes an anchor picture. If the anchor block
does not have motion information, a motion vector becomes 0 and a
reference picture index of the List0 reference picture in the
temporal direct mode becomes 0.
[0080] Moreover, it is able to find a motion vector mvL0 of List0
and a motion vector mvL1 of List1 from the motion vector mvCol.
Assuming that a speed of motion within a reference picture, a
current picture and an anchor picture is constant, a size of each
of the mvL0 and the mvL1 becomes proportional to time intervals
among the reference, current and anchor pictures. Hence, using the
interval between the reference picture and the current picture and
the interval between the reference picture and the anchor picture,
it is able to find the mvL0 and the mvL1. In particular, in case
that List0 reference index of the anchor block refers to a
long-term reference picture, it can be found by Formula 1.
mvL0=mvCol
mvL1=0 [Formula 1]
[0081] In case that List0 reference index of the anchor block does
not refers t a long-term reference picture, it can be found by
Formula 2.
mvL0=mvCol*(tb/td)
mvL1=-mvCol*(td-tb)/td=mvL0-mvCol [Formula 2]
[0082] For another example of the direct mode, a spatial direct
mode predicts motion information of a current block using motion
information correlation of a spatial direction. Like the temporal
direct mode, an anchor picture and an anchor block are used in
finding motion information.
[0083] FIG. 9 is a schematic internal block diagram of an inter
prediction unit 700 having a spatial direction prediction mode
applied thereto.
[0084] Referring to FIG. 9, an inter prediction unit 700 mainly
includes a direct mode identifying unit 710, a spatial direct
prediction performing unit 720, etc. And, the spatial direct
prediction performing unit 720 includes a first variable deriving
unit 721, a second variable deriving unit 722 and a motion
information predicting unit 723.
[0085] The direct mode identifying unit 710 identifies a prediction
mode of a current slice. For instance, if a slice type of a current
slice is P slice or B slice, a direct mode is usable. In this case,
it is able to use a direct mode flag indicating whether a temporal
direct mode or a spatial direct mode in a direct mode will be used.
The direct mode flag can be obtained from a slice header. In case
that the spatial direct mode is applied according to the direct
mode flag, it is able to preferentially obtain motion information
of blocks neighbor to a current block. For instance, assuming that
a block at a left side of a current block, a block at a top side
and a block at a right top side are named a neighbor block A, a
neighbor block B and a neighbor block C, respectively, it is able
to obtain motion information of the neighbor blocks A, B and C. The
first variable deriving unit 721 is able to derive a reference
picture index for List0 or List1 direction of the current block
using the motion information of the neighbor blocks. And, it is
then able to derive a first variable based on the reference picture
index of the current block. In this case, the first variable can
mean a variable (directZeroPredictionFlag) used in predicting a
motion vector of the current block as a random value. For instance,
a reference picture index for the List0 or List1 direction of the
current block can be derived into a smallest value of reference
picture indexes of the neighbor blocks. For this, the following
formula (Formula 3) is usable.
refIdxL0=MinPositive(refIdxL0A,MinPositive(refIdxL0B,refIdxL0C))
refIdxL1=MinPositive(refIdxL1A,MinPositive(refIdxL1B,refIdxL1C))
where,MinPositive(x,y)=Min(x,y)(x.gtoreq.0 and y.gtoreq.0)
Max(x,y)(other cases) [Formula 3]
[0086] For example, in detail, it becomes `MinPositive(0,1)=0`. In
particular, in case that two valid indexes exist, it is able to
obtain a smaller value. Alternatively, it becomes
`MinPositive(-1,0)=0`. Namely, in case that one valid index exists,
it is able to obtain a large value that is a valid index value.
Alternatively, it becomes `MinPositive(-1,-1)=-1`. Namely, in case
that two are not valid index at all, e.g., if both neighbor blocks
are blocks coded in an intra picture prediction mode or two
neighbor blocks are not available, a large value `-1` is obtained.
Therefore, in order to obtain an invalid result value, at least one
valid value should not exist.
[0087] First of all, for an initial value of the first variable, it
is able to set the first variable to 0. Moreover, in case that each
reference picture index for the derived List 0 and List1 directions
is smaller than 0, a reference picture index for the List0 or List1
direction of the current block can be set to 0. And, the first
variable can be set to a value for indicating that a reference
picture of the current block does not exist. In this case, the case
that each reference picture index for the derived List0 and List1
directions is smaller than 0 means a case that the neighbor block
is a block coded in an intra picture prediction mode or a case that
the neighbor block is not available, for example. In this case, by
setting the first variable to 1, it is able to set a motion vector
of the current block to 0.
[0088] Since three neighbor blocks are used for the present
embodiment, in case that all three neighbor blocks are coded in the
intra picture prediction mode or not available, a reference picture
index of the current block may have an invalid value of `-1`.
Hence, if all reference picture indexes of the current block are
not available, the first variable can be set to 1. In particular,
if a reference picture to be used by a current block fails to
exist, the first variable can be set to 1.
[0089] The second variable deriving unit 722 is able to derive a
second variable using the motion information of the anchor block
within the anchor picture. In this case, the second variable may
mean a variable that is used to predict a motion vector of a
current block as a random value. For instance, in case that the
motion information of the anchor block meets predetermined
conditions, it is able to set the second variable to 1. If the
second variable is set to 1, it is able to set a motion vector for
the List0 or List1 direction of the current block to 0. The
predetermined conditions can include the following. First of all, a
picture having a smallest reference picture index among reference
pictures for the List1 direction should be a short-term reference
picture. Secondly, a reference index of a referred picture of the
anchor block should be set to 0. Thirdly, each of the horizontal
and vertical components of a motion vector of the anchor block
should be equal to or smaller than .+-.1. Thus, if all the
predetermined conditions are met, it is determined that it is close
to a video having almost no motion. And, the motion vector of the
current block is then set to 0.
[0090] The motion information predicting unit 723 is able to
predict motion information of a current block based on the derived
first and second variables. For instance, if the first variable is
set to 1, it is able to set the motion vector for the List0 or
List1 of the current block to 0. If the second variable is set to
1, it is able to set the motion vector for the List0 or List1 of
the current block to 0. In this case, the setting to 0 or 1 is just
exemplary. And, it is able to set the first or second variable to
another predetermined value to use.
[0091] If the predetermined conditions are not met, it is able to
derive motion information of a current block from motion
information of neighbor blocks within a current picture. For
instance, in case of a motion vector, the motion vector deriving
method of the skip macroblock is identically applicable. A
reference picture index of the current block can be found in a
manner of calculating a minimum value among reference picture
indexes of the motion vectors located at left, top and right top
sides and then setting the minimum value to the reference picture
of the current block.
[0092] A method of indicating whether a current macroblock is coded
in a direct mode is explained as follows. For this, it is
understood that the same method is applicable to a sub-macroblock
as well.
[0093] FIG. 10 is a block diagram of a process for decoding slice
data using skip or direct mode indication information.
[0094] Referring to FIG. 10, it is able to parse flag information
indicating whether a current macroblock within a slice is a skip
macroblock [S1000]. The flag information can be obtained from a
slice layer. As a result of parsing the flag information, if the
current macroblock is the skip macroblock [`yes 1`], it is able to
perform decoding without parsing motion information and residual
data [S1050]. Yet, if the current macroblock is not the skip
macroblock [`no 1`], it is able to parse a block type of the
current macroblock [S1010]. It is able to parse skip or direct mode
indication information for each partition of the current macroblock
according to the block type of the current macroblock [S1020]. In
this case, the skip or direct mode indication information may mean
the information indicating whether the partition of the macroblock
is coded in the skip mode or the direct mode. The following
description is made on the assumption that a size of a current
macroblock is 32.times.32. First of all, as a result of parsing
current macroblock type information, if the current macroblock is
coded in an inter picture prediction mode and a partition of the
current macroblock is 16.times.32, 2 bits will be used as skip or
direct mode indication information. If a partition index 0 of a
macroblock is coded in skip or direct mode but a partition index 1
of the macroblock is not coded in the skip or direct mode, it is
able to obtain `10` as skip or direct mode indication information.
After the skip or direct mode indication information has been
parsed, it is able to parse CBP coded block pattern) information
[S1030]. In this case, the CBP information can mean the information
indicating whether a current block includes residual data.
Therefore, by parsing the CBP information, it is able to know
whether the partition of the current block is coded in the skip
mode or the direct mode. For instance, as a result of parsing the
CBP information on the partition of the current macroblock, if `1`
is obtained, it is able to regard that the partition is coded in
the direct mode. Based on the skip or direct mode indication
information, it is determined whether each partition of the current
macroblock is coded in the skip or direct mode. In case that the
partition is not coded in the skip mode or the direct mode [`no
2`], it is able to perform decoding by parsing motion information
and residual data [S1040] [S1050]. Yet, in case of the partition
coded in the skip mode or the direct mode [`yes 2`], whether the
partition is coded in the skip mode or the direct mode will be
determined based on the parsed CBP information [S1060]. Hence, in
case of coding in the skip mode [`yes 3`], it is able to perform
decoding without parsing the motion information and the like
[S1040]. In case of coding in the direct mode [`no 3`], after the
residual data has been parsed [S1070], it is able to perform
decoding using the residual data [S1050]. In case of coding in the
direct mode, it is able to derive a motion vector in the same
manner of the case of coding in the skip mode. It is able to use
the direct mode if a slice type is a P slice as well as a B
slice.
[0095] If the slice type is the B slice, macroblock type
information can include size information of a predicted block,
prediction direction information and the like. Yet, if it is
indicated that a partition of a current macroblock is coded in a
skip or direct mode according to the skip or direct mode indication
information, a motion vector of the partition is derived. Hence,
the prediction direction information of the partition can be
derived. In this case, it is able to regard that the macroblock
type information includes unnecessary prediction direction
information. Hence, by newly defining a macroblock type, it is
intended to efficiently perform decoding of the slice data using
the above-mentioned skip or direct mode indication information. For
instance, it is able to set a macroblock type for a B slice to one
of B_direct.sub.--32.times.32, B.sub.--32.times.32,
B.sub.--32.times.16, B.sub.--16.times.32 and B.sub.--16.times.16.
Yet, in case that a macroblock type is reset, it is determined
whether the partition of the current macroblock in FIG. 10 is coded
in the skip mode or the direct mode. If it is coded in the skip
mode or the direct mode, decoding will be performed in the same
process as explained in the foregoing description. If it is not
coded in the skip mode or the direct mode, information on a
prediction direction of a macroblock will be parsed from a received
bitstream as well as motion information and residual data.
[0096] Accordingly, the present invention is applicable to video
signal encoding/decoding.
[0097] While the present invention has been described and
illustrated herein with reference to the preferred embodiments
thereof, it will be apparent to those skilled in the art that
various modifications and variations can be made therein without
departing from the spirit and scope of the invention. Thus, it is
intended that the present invention covers the modifications and
variations of this invention that come within the scope of the
appended claims and their equivalents.
* * * * *