Pulse or digital communications – Bandwidth reduction or expansion – Television or motion video signal
Reexamination Certificate
2000-01-21
2004-05-18
Rao, Andy (Department: 2613)
Pulse or digital communications
Bandwidth reduction or expansion
Television or motion video signal
C375S240160, C375S240220
Reexamination Certificate
active
06738423
ABSTRACT:
The present invention relates to video coding. In particular, it relates to compression of video information using motion compensated prediction.
BACKGROUND OF THE INVENTION
A video sequence consists of a large number video frames, which are formed of a large number of pixels each of which is represented by a set of digital bits. Because of the large number of pixels in a video frame and the large number of video frames even in a typical video sequence, the amount of data required to represent the video sequence quickly becomes large. For instance, a video frame may include an array of 640 by 480 pixels, each pixel having an RGB (red, green, blue) color representation of eight bits per color component, totaling 7,372,800 bits per frame. Video sequences comprise a sequence of still images, which are recorded/displayed at a rate of typically 15-30 frames per second. The amount of data needed to transmit information about each pixel in each frame separately would thus be enormous.
Video coding tackles the problem of reducing the amount of information that needs to be transmitted in order to present the video sequence with an acceptable image quality. For example, in videotelephony the encoded video information is transmitted using conventional telephone networks, where transmission bit rates are typically multiples of 64 kilobits/s. In mobile videotelephony, where transmission takes place at least in part over a radio communications link the available transmission bit rates can be as low as 20 kilobits/s.
In typical video sequences the change of the content of successive frames is to a great extent the result of the motion in the scene. This motion may be due to camera motion or due to motion of the objects present in the scene. Therefore typical video sequences are characterized by significant temporal correlation, which is highest along the trajectory of the motion. Efficient compression of video sequences usually takes advantage of this property of video sequences. Motion compensated prediction is a widely recognized technique for compression of video. It utilizes the fact that in a typical video sequence, image intensity/chrominance value in a particular frame segment can be predicted using image intensity/chrominance values of some other already coded and transmitted frame, given the motion trajectory between these two frames. Occasionally it is advisable to transmit a whole frame, to prevent the deterioration of image quality due to accumulation of errors and to provide additional functionalities, for example, random access to the video sequence).
A schematic diagram of an example video coding system using motion compensated prediction is shown in
FIGS. 1 and 2
of the accompanying drawings.
FIG. 1
illustrates an encoder
10
employing motion compensation and
FIG. 2
illustrates a corresponding decoder
20
. The operating principle of video coders using motion compensation is to minimize the prediction error frame E
n
(x,y), which is the difference between the current fame I
n
(x,y) being coded and a prediction frame P
n
(x,y). The prediction error frame is thus
E
n
(
x,y
)=
I
n
(
x,y
)−
P
n
(
x,y
). (1)
The prediction frame is built using pixel values of a reference frame R
n
(x,y), which is one of the previously coded and transmitted frames (for example, a frame preceding the current frame), and the motion of pixels between the current frame and the reference frame. The motion of the pixels may be presented as the values of horizontal and vertical displacements &Dgr;x(x,y) and &Dgr;y(x,y) of a pixel at location (x,y) in the current frame I
n
(x,y). The pair of numbers [&Dgr;x(x,y), &Dgr;y(x,y)] is called the motion vector of this pixel. The motion vectors are typically represented using some known functions (called basis functions) and coefficients (this is discussed in more detail below), and an approximate motion vector field ({tilde over (&Dgr;)}x(x,y), {tilde over (&Dgr;)}x(x,y)) can be constructed using the coefficients and the basis functions.
The prediction frame is given by
P
n
(
x,y
)=
R
n
[x+{tilde over (&Dgr;)}x
(
x,y
),
y+{tilde over (&Dgr;)}y
(
x,t
)], (2)
where the reference frame R
n
(x,y) is available in the Frame Memory
17
of the encoder
10
and in the Frame memory
24
of the decoder
20
at a given instant. An information stream
2
carrying information about the motion vectors is combined with information about the prediction error (1) in the multiplexer
16
and an information stream (3) containing typically at least those two types of information is sent to the decoder
20
.
In the Prediction Error Coding block
14
, the prediction error frame E
n
(x,y) is typically compressed by representing it as a finite series (transform) of some 2-dimensional functions. For example, a 2-dimensional Discrete Cosine Transform (DCT) can be used. The transform coefficients related to each function are quantized and entropy coded before they are transmitted to the decoder (information stream
1
in FIG.
1
). Because of the error introduced by quantization, this operation usually produces some degradation in the prediction error frame E
n
(x,y).
In the Frame Memory
24
of the decoder
20
there is a previously reconstructed reference frame R
n
(x,y). Using the decoded motion information ({tilde over (&Dgr;)}x(x,y), {tilde over (&Dgr;)}y(x,y)) and R
n
(x,y) it is possible to reconstruct the prediction frame P
n
(x,y) in the Motion Compensated Prediction block
21
of the decoder
20
. The transmitted transform coefficients of the prediction error frame E
n
(x,y) are used in the Prediction Error Decoding block
22
to construct the decoded prediction error frame {tilde over (E)}
n
(x,y). The pixels of the decoded current frame Ĩ
n
(x,y) are reconstructed by adding the prediction frame P
n
(x,y) and the decoded prediction error frame {tilde over (E)}
n
(x,y)
Ĩ
n
(
x,y
)=
P
n
(
x,y
)+
{tilde over (E)}
n
(
x,y
)=
R
n
[x+{tilde over (&Dgr;)}r
(
x,y
),
y+{tilde over (&Dgr;)}y
(
x,y
)]+
{tilde over (E)}
n
(
x,y
). (3)
This decoded current frame may be stored in the Frame Memory
24
as the next reference frame R
n+l
(x,y).
Let us next discuss in more detail the motion compensation and transmission of motion information. The construction of the prediction frame P
n
(x,y) in the Motion Compensated Prediction block
13
of the encoder
10
requires information about the motion in the current frame I
n
(x,y). Motion vectors [&Dgr;x(·)(x,y), &Dgr;y(x,y)] are calculated in the Motion Field Estimation block
11
in the encoder
10
. The set of motion vectors of all pixels of the current frame [&Dgr;x(·), &Dgr;y(·)] is called the motion vector field. Due to the very large number of pixels in a frame it is not efficient to transmit a separate motion vector for each pixel to the decoder. Instead, in most video coding schemes the current frame is divided into larger image segments and information about the segments is transmitted to the decoder.
The motion vector field is coded in the Motion Field Coding block
12
of the encoder
10
. Motion Field Coding refers to representing the motion in a frame using some predetermined functions or, in other words, representing it with a model. Almost all of the motion vector field models commonly used are additive motion models. Motion compensated video coding schemes may define the motion vectors of image segments by the following general formula:
Δ
x
(
x
,
y
)
=
∑
i
=
0
N
-
1
a
i
f
i
(
x
,
y
)
(
4
)
Δ
y
(
x
,
y
)
=
∑
i
=
0
M
-
1
b
i
g
i
(
x
,
y
)
(
5
)
where coefficients a
i
and b
i
are called motion coefficients. They are transmitted to the decoder. Functions f
i
and g
i
are called motion field basis functions, and they are known both to the encoder and decoder.
In order to minimize the amount of information needed in sending the motion coefficients
Karczewicz Marta
Lainema Jani
Nokia Mobile Phones Ltd.
Parsons Charles E
Perman & Green LLP
Rao Andy
LandOfFree
Method for encoding and decoding video information, a motion... does not yet have a rating. At this time, there are no reviews or comments for this patent.
If you have personal experience with Method for encoding and decoding video information, a motion..., we encourage you to share that experience with our LandOfFree.com community. Your opinion is very important and Method for encoding and decoding video information, a motion... will most certainly appreciate the feedback.
Profile ID: LFUS-PAI-O-3251684