Image analysis – Image compression or coding – Shape – icon – or feature-based compression
Reexamination Certificate
1998-10-16
2003-04-22
Chen, Wenpeng (Department: 2724)
Image analysis
Image compression or coding
Shape, icon, or feature-based compression
C382S239000, C375S240080
Reexamination Certificate
active
06553149
ABSTRACT:
BACKGROUND OF THE INVENTION
1. Field of the Invention
The present invention relates to video coding and decoding and more particularly to a shape information coding and decoding apparatus for adaptively bordering and method therefor, wherein bordering is performed with respect to blocks of shape information when forming contexts in order to context-based arithmetic encode a shape in a picture.
2. Description of Related Art
There is a case that only a certain particular object in a frame is intended to be processed to increase coding efficiency or improve picture quality in processing video information. In this case, shape information of the particular object is required to separate the particular object from a background.
A block of a particular object's shape information includes pixels for the particular object, the pixels having a specified value (for example, “1”), and pixels for others except the object, the pixels having a specified value “0”. The shape information of the particular object is divided into blocks of a predetermined size (for example, 16×16 and 8×8) and a specified operation is performed with respect to ambient pixels of the pixels to be coded in order to code the shape information within a block.
FIG. 1
is a block diagram of a conventional object-based video coder.
Concepts of a shape coder and a video object planes (VOP) are introduced here. The VOP indicates an object at a certain point in a time domain of a content having a predetermined shape which can be accessed and edited by a user. The information should be coded by each VOP for support of a content-based functionality.
Primarily, signals of a picture are classified into shape information and texture information and two types of information are respectively inputted to a shape coding unit
11
and motion estimation unit
12
.
The shape coding unit
11
performs lossy coding or lossless coding with respect to the shape information of a relevant frame. Reconstructed shape information is inputted to both motion compensation unit
13
and texture coding unit
17
. Both the motion compensation unit
13
and texture coding unit
17
operate based upon an object. A shape information bit stream, which is another output of the shape coding unit
11
, is inputted to a multiplexer
18
.
The motion estimation unit
12
estimates motion information of a current frame's texture information using input texture information of the current frame and texture information of a previous frame which is stored in a previous reconstructed frame memory
14
. The estimated motion information is inputted to the motion compensation unit
13
while a motion information bit stream is encoded and inputted to the multiplexer
18
. The motion compensation unit
13
performs motion compensation using the motion information obtained through the motion estimation unit
12
and the previous reconstructed frame received from the previous reconstructed frame memory
14
.
The texture coding unit
17
codes a prediction error. The prediction error is a difference between input texture information obtained through a subtracter
15
and motion compensated texture information obtained through the motion compensation unit
13
. A texture bit stream which is generated through the coding at the texture coding unit
17
is inputted to the multiplexer
18
and an error signal of reconstructed texture information is inputted to an adder
16
. The previous reconstructed frame memory
14
stores a previous reconstructed frame signal received from the adder
16
. The previous reconstructed frame signal is obtained by adding the error signal of the reconstructed texture information to the motion compensated signal.
Digital video may be classified into progressive video and interlaced video according to frame constructing methods. For the progressive video, a frame is constructed in such a manner that lines consecutively progresses from the top to the bottom. For the interlaced video, a frame is constructed in such a manner that a field of odd lines is primarily constructed and then even lines in the other field are interlaced with the odd lines of the first field. A height (the number of lines) of the field is a half of the height of the frame. This is illustrated in
FIGS. 2
a
and
2
b.
FIG. 2
a
shows a frame of the progressive video and
FIG. 2
b
shows two fields -a top field and a bottom field- and a frame of the interlaced video. In
FIGS. 2
a
and
2
b,
the top and bottom fields consist of lines (solid arrows in the top field and dotted arrows in the bottom field) and the lines of each field interlace (the solid arrows are interlaced with the dotted arrows) to construct an interlaced frame.
When the top and bottom fields are constructed, as shown in
FIG. 2
b,
there is a time gap between the two fields and the top field precedes the bottom field. In other cases, the bottom field may precede the top field. For the lines forming a frame in an interlaced video, the lines constructing the top field and the lines constructing the bottom field are separately scanned by each field. Because of the time gap between the top field and the bottom field, signal characteristics of neighboring lines in the interlaced frame can be different.
Particularly, in case of a picture having a lot of motion, this feature described above is prominent. When applying video coding functions developed in accordance with properties of the progressive video, such as motion estimation, motion compensation, and discrete cosine transform (DCT), to the coding of the interlaced video, reduction of coding efficiency is caused. Technology, such as field-based motion estimation and compensation and adaptive frame/field DCT, has been developed to overcome this problem. Such technology is disclosed in the standard MPEG-2 established by the ISO/IEC JTC1/SC29/WG11 for applications of digital TV and the like. The technology has been frequently applied to actual application products.
FIGS. 3
a
and
3
b
show interlaced shape information where an object has no motion or a little motion between two fields. As shown in
FIG. 3
a,
correlation between lines in a frame is higher compared with that in each field, so it is better to code the shape information from a frame than from each field in this case.
FIGS. 4
a
and
4
b
show shape information where an object has much motion between two fields. As shown in
FIG. 4
a
where the lines are grouped into each field, variation between shape information of each line is little and correlation between lines is high in the same field. However, as shown in
FIG. 4
a,
when considering a whole frame, the variation between shape information of each line is larger, so the correlation between lines is lower. Therefore, coding efficiency is reduced when coding the shape information from the frame.
It is best to adaptively select one between a field coding mode and a frame coding mode rather than to use only one mode when coding the interlace shape information.
FIG. 5
a
shows a context for performing context-based arithmetic encoding (CAE) in an INTRA mode. A value of a pixel
51
is encoded through a specified operation using pixels C
0
to C
9
neighboring with the pixel
51
to be encoded.
FIG. 5
b
shows a context for performing the CAE in an INTER mode. A value of a pixel
52
is encoded through a specified operation using pixels C
0
to C
3
neighboring with the pixel
52
to be encoded in a current block and a pixel C
6
corresponding to the pixel
52
and its neighboring pixels C
4
, C
5
, C
7
, and C
8
in a previous frame.
When coding shape information of a particular object, the information is divided into binary alpha blocks (BABs) of a predetermined size, for example, 16×16. In this case, bordering is performed to construct contexts of outer pixels of a BAB. As shown in
FIGS. 5
a
and
6
, if the pixel
51
to be encoded is located in the left border of a BAB
61
, values of the pixels C
0
, C
1
, C
5
, C
6
, and C
9
in
FIG. 5
a
cannot be acknowledged, so the BAB
61
is bordered by a left border
63
an
Chun Sung-Moon
Moon Joo-Hee
Shin Dong-Kyoo
Chen Wenpeng
Hyundai Electronics Ind. Co., LTD
Lackenbach Siegel
Nissen J. Harold
LandOfFree
Shape information coding and decoding apparatus for... does not yet have a rating. At this time, there are no reviews or comments for this patent.
If you have personal experience with Shape information coding and decoding apparatus for..., we encourage you to share that experience with our LandOfFree.com community. Your opinion is very important and Shape information coding and decoding apparatus for... will most certainly appreciate the feedback.
Profile ID: LFUS-PAI-O-3029064