Pulse or digital communications – Bandwidth reduction or expansion – Television or motion video signal
Reexamination Certificate
1998-12-11
2002-04-30
Kelley, Chris (Department: 2613)
Pulse or digital communications
Bandwidth reduction or expansion
Television or motion video signal
Reexamination Certificate
active
06381277
ABSTRACT:
BACKGROUND OF THE INVENTION
1. Field of the Invention
The present invention relates to a shape information coding device for interlaced scanning video and method therefor, and more particularly, to a shape information coding device for interlaced scanning video and method therefor which can detect an amount of motion of object video, on coding of interlaced scanning video and encode the object video in a frame type block unit or in a field type block unit in accordance with the detected result.
2. Discussion of Related Art
With the recent development of the research of next generation video/audio coding technology and system construction, improved effort is tried for video/audio applications which have not been supported by the well known standard plans such as, for example, H.263 and H.261 of ITU-T, and MPEG-1 and MPEG-2 of ISO/IEC. Examples of the accomplished functions are object-based interactive functionality and object-based manipulation. To provide various kinds of new functions, shape information should be of course transmitted. The shape information serves to divide video into an object area and non-object area(that is, background), to allow a signal process in transmitting/receiving terminals to be implemented not for whole video but for object area, and to provide the above new functions.
A general binary shape information has a binary mask type, in which the pixels corresponding to an object have different values from the pixels to non-object. For instance, the pixel corresponding to the object has a logic value of “1”, and that corresponding to the non-object has a logic value of “0”. A method for coding motional video by using the shape information is called. “object-based video coding”.
In other words, the divided object video can be coded and compressed, independently of the background video. In the case where an object to be coded in the video screen exists, an operation for coding the shape information corresponding to the object or area is necessary. The shape information coding method is achieved in a context-based arithmetic encoding(hereinafter, referred to “CAE”) manner, under arithmetic encoding by using a probability table of the context configuration.
FIG. 1
is a block diagram illustrating configuration of a representative object-based video coder. Firstly, each of video signals is divided into shape information and texture information. The shape information is inputted to a shape information coding unit
11
and the texture information to a motion estimating unit
12
.
In the shape information coding unit
11
, a lossy coding or a lossless coding process is performed for the shape information of the corresponding video, and a reconstructed shape information as an output of the shape information coding unit
11
is inputted to the motion estimating unit
12
, a motion compensating unit
13
and a texture information coding unit
17
, respectively, in an object unit.
Meanwhile, shape information bitstream as another output of the shape information coding unit
11
is inputted to a multiplexer
18
. The motion estimating unit
12
serves to estimate motion information for the texture information of current video, by using the texture information inputted in current frame and the texture information of the previous video stored in a previous reconstructed frame memory
14
. The estimated motion information is inputted to the motion compensating unit
13
and the coded motion information is also inputted to the multiplexer
18
. The motion compensating unit
13
serves to execute motion compensation prediction by using the motion information obtained by the motion estimating unit
12
and the previous reconstructed frame stored in the previous reconstructed frame memory
14
. A subtractor
15
obtains a prediction error between the inputted texture information and the motion compensation texture information obtained by the motion compensating unit
13
. A texture information coding unit
17
functions to code the prediction error obtained by the subtractor
15
. The texture information bitstream generated from the texture information coding unit
17
is inputted to the multiplexer
18
, and the error signal of the reconstructed texture information is inputted to an adder
16
. The previous reconstructed frame memory
14
stores the previous reconstructed frame signal which is outputted from the adder
16
in which the motion compensation prediction signal and the reconstructed error signal are added.
On the other hand, a digital video is divided into a progressive scanning video and an interlaced scanning video in accordance with the structure arrangement of frame. In the progressive scanning video, frames are arranged in order by one line and the video is coded, transmitted and displayed in one frame unit. Contrarily, in the interlaced scanning video, two fields are arranged in order by one line, and each frame is formed in the manner where the two fields are inserted therein by one line, so that the height(the number of lines) of each field is half the height of the frame.
Examples of the progressive scanning frame and interlaced scanning frame are shown in
FIGS. 2A and 2B
.
In more detail,
FIG. 2A
shows the progressive scanning frame and
FIG. 2B
shows the interlaced scanning frame in which two fields(top field and bottom field) are inserted by one line. The top field(indicated by a solid line arrow) and bottom field(indicated by a dotted line arrow) are arranged by one line in the interlaced scanning frame, so that the solid line arrow and the dotted line arrow are arranged in turn in the frame. As shown in
FIG. 2B
, a time difference between the top and bottom fields exists, and in this case, the top field precedes the bottom field. However, the bottom field may precede the top field. Because of the time difference between the top and bottom fields, signal characteristics between the lines adjacent to each other within the interlaced scanning frame can be different. More particularly, the different signal characteristic is apparent in the video having a large amount of motion information.
FIGS. 3A and 3B
are views each illustrating a method for determining a motion vector predictor for shape(hereinafter, referred to as “MVPs”) in a conventional shape information coding method.
FIG. 3A
shows a current shape binary alpha block(hereinafter, referred to as “BAB”), the BABs of left side, top side and right top side of the current shape BAB, and the motion vector of each of the BABs adjacent to the current shape BAB. At this time, it is assumed that the size of BAB is 16×16.
FIG. 3B
shows a texture information macroblock(hereinafter, referred to as “MB”) corresponding to the current shape BAB, the MBs of left side, top side and right top side of the MB, and the motion vector of each of the MBs adjacent to the MB. At this time, it is assumed that the size of MB is 16×16, in the same manner as the BAB. Each of the motion vectors MV
1
, MV
2
and MV
3
of the adjacent blocks to the texture information MB indicates the motion vector of the corresponding block, and if the corresponding MB estimates and compensates one motion vector per 16×16 MB, the motion vector of the MB is the same as the above. However, if the corresponding MB estimates and compensates one motion vector per 8×8 MB, the motion vector of the MB positioned is indicated. The motion vector MVs of the current shape BAB is given as follows: MVs=MVDs(motion vector difference value for shape)+MVPs. In other words, the MVPs is first determined and from the determined value, the MVDs is obtained. The MVDs is the information which is transmitted to the receiving terminal from the transmitting terminal. The receiving terminal determines the MVs with the MVDs transmitted from the transmitting terminal and the MVPs obtained in the same manner as the transmitting terminal. Therefore, since the MVPs is obtained by using the same information as the transmitting and receiving terminals, the determination of the MVPs should be made by using only the information w
Chun Sung Moon
Lee Jin Hak
Moom Joo Hee
Shin Dong Kyoo
Hyundai Electronics Ind. Co, Ltd.
Kelley Chris
Lackenbach Siegel
Nissen J. Harold
Wong Allen
LandOfFree
Shaped information coding device for interlaced scanning... does not yet have a rating. At this time, there are no reviews or comments for this patent.
If you have personal experience with Shaped information coding device for interlaced scanning..., we encourage you to share that experience with our LandOfFree.com community. Your opinion is very important and Shaped information coding device for interlaced scanning... will most certainly appreciate the feedback.
Profile ID: LFUS-PAI-O-2859876