Method for motion vector coding of MPEG-4

Details Method for motion vector coding of MPEG-4 Method for motion vector coding of MPEG-4

1998-04-24

2002-12-24

Kelley, Chris (Department: 2613)

Pulse or digital communications

Bandwidth reduction or expansion

C375S240000, C375S240160, C375S240240, C382S232000, C382S236000, C382S238000

Reexamination Certificate

active

06498810

ABSTRACT:

BACKGROUND OF THE INVENTION
1. Field of the Invention
The present invention relates to a coding method, and in particular, to a method for motion vector coding of a moving picture experts group-4 (MPEG-4) video verification model (VM).
2. Background of the Related Art
Redundancy in a temporal domain of image data and redundancy in a two dimensional spatial domain of image data must be removed to efficiently compress a video sequence that changes with time. A discrete cosine transform (DCT) is used to remove the redundancy in the two dimensional spatial domain, and a motion compensation method is used to remove the redundancy in the temporal domain in the MPEG.
The DCT is a method of removing the correlation between data through a two dimensional spatial transformation. After a picture is divided into blocks, each block in the picture is spatially transformed using the DCT. Data, which has been spatially transformed, tends to be driven to a certain direction.
Pictures, which are consecutive in the temporal domain, generally form motions of a human being or an object at the center of the frame. This property is used to remove the redundancy of the temporal domain in the motion compensation method. The volume of data to be transmitted can be minimized by taking out the similar region from the preceding picture and filling a region that has not changed (or has very little change) in the present picture with the similar region of the preceding picture. The operation of finding the most similar blocks between pictures is called motion estimation. The displacement representing the degree of motion is called a motion vector. The motion vector is two dimensional vector that is an offset between the coordinates of a reference frame or a reference field and the coordinates of a present picture or field for motion compensation of the picture.
An 8×8 block search for integer pixel motion estimation is performed inside of a search window of ±2½ pixels based on a 16×16 motion vector in the MPEG-4. When
FIG. 1
a particular macroblock is established to an 8×8 mode, respective motion vectors against the four blocks in the macroblock are within a specified bound.
FIG. 1
shows a bound that motion vectors of the four blocks can indicate when a macroblock is established in the 8×8 mode according to the related art. This bound of the four motion vectors greatly contributes to increasing the coding efficiency of the motion vector. Four motion vectors are maximum permissible inside of the macroblock. When transmitting the four motion vectors, the amount of bits is great. The amount of bits can be decreased by performing variable length coding (VLC) to transmit only a motion vector difference from a motion vector of the preceding macroblock. The difference between successive motion vectors is expressed as a motion vector difference (MVD).
As shown in
FIGS. 2
a
to
2
d
, an actual motion vector of a block (e.g., an 8×8 block) is estimated using a difference (MVD
X
, MVD
Y
) between a median of three neighboring motion vector candidates and a motion vector of the relevant block. This can be expressed by the following equation:
P
X
=Median(MV
1
X
, MV
2
X
, MV
3
X
),
P
Y
=Median(MV
1
Y
, MV
2
Y
, MV
3
Y
),
MVD
X
=MV
X
−P
X
, MVD
Y
=MV
Y
−P
Y
.
FIG. 1
illustrates a limit of motion vector estimation. The motion vector estimation limit shown in
FIG. 1
cannot completely support an MPEG-4 VM for the following reasons.
For motion vector estimation candidates in an 8×8 mode macroblock shown in
FIGS. 2
a
-
2
d
, a case of a motion vector used as a median prediction that exceeds the limit of a motion vector in the 8×8 mode is related to MVD
2
corresponding to block
2
. Since only one of three motion vector candidates is inside of the macroblock in the 8×8 mode, the absolute value of MVD
2
, which is obtained from the prediction established by the median, can exceed the limit of 5.0.
Thus, there can be a case that the prediction established by the median is out of the limit. When defining MV of block
2
as M
BLOCK
(a motion vector of a relevant block), MV
1
of block
2
as MV
INSIDE MV PREDICTION
(a motion vector prediction inside a bound), and Median(MV
1
, MV
2
, MV
3
) of block
2
as MV
MEDIAN PREDICTION
.
FIG. 3
illustrates the case that a motion vector median prediction MV
MEDIAN PREDICTION
exceeds a limit.
As described above, the related art method for motion vector coding uses a median of three neighboring motion vector candidates to estimate one motion vector. However, since this method uniformly estimates all pixels based upon the median of the three neighboring motion vector candidates, there is a problem that motion vector median prediction can exceed a predetermined limit. In addition, there is a problem that coding efficiency significantly decreases in more complicated motion sequences (i.e., motion vector median prediction out of bounds).
SUMMARY OF THE INVENTION
An object of the present invention is to provide a method for motion vector coding that substantially obviates one or more of the limitations and disadvantages of the related art.
Another object of the present invention is to provide a method for motion vector coding of an MPEG-4 that decreases an overall bit rate.
A further objective of the present invention is to provide a method of motion vector coding by finding a motion vector having a minimum difference from an actual motion vector among three motion vector candidates and transmitting the value of motion vector candidate found and a mode bit.
To achieve at least these objects and other advantages in a whole or in parts, and in accordance with the purpose of the present invention as embodied and broadly described, a method for motion vector coding of an MPEG-4 includes selecting a candidate that has a minimum bit rate when being coded among three motion vector candidates; sending a motion vector prediction error and prediction mode information to a decoder; estimating a minimum bit rate for coding a motion vector difference; and coding a motion vector against components X and Y, using a minimum rate predictor.
Additional advantages, objects, and features of the invention will be set forth in part in the description which follows and in part will become apparent to those having ordinary skill in the art upon examination of the following or may be learned from practice of the invention. The objects and advantages of the invention may be realized and attained as particularly pointed out in the appended claims.


REFERENCES:
patent: 5608458 (1997-03-01), Chen et al.
patent: 5691770 (1997-11-01), Keesman et al.
patent: 5731840 (1998-03-01), Kikuchi et al.
patent: 5818536 (1998-10-01), Morris et al.
patent: 6047088 (2000-04-01), Van Beck et al.
patent: 6289049 (2001-09-01), Kim et al.
patent: 02-065583 (1990-03-01), None
patent: 08-168054 (1996-06-01), None
Avideh Zakhor, “Iterative Procedures for Reduction of Blocking Effects In Transform Image Coding”, IEEE Transactions On Circuits and Systems For Video Technology, vol. 2, No. 1, pp. 91-95 (Mar. 1992).
Yongyi Yang et al., “Regularized Reconstructions to Reduce Blocking Artifacts of Block Discrete Cosine Transform Compressed Images”, IEEE Transactions On Circuits and Systems For Video Technology, vol. 3, No. 6, pp. 421-432 (Dec. 1993).
Gary J. Sullivan et al., “Motion Compensation for Video Compression Using Control Grid Interpolation”, IEEE International Conference, pp. 2713-2716 (1991).
G. de Haan et al., “IC For Motion-Compensation 100Hz TV With Natural-Motion Movie-Mode”, IEEE Transactions On Consumer Electronics, vol. 42, pp. 165-174 (Feb. 1996).
Taner Özcelik et al., “Image and Video Compression Algorithms Based on Recovery Techniques Using Mean Field Annealing”, Proceedings of the IEEE, vol. 83, No. 2, pp. 304-316, Feb. 1995.
Yasuyuki Nakajima et al., “A PEL Adaptive Reduction of Coding Artifacts for MPEG Video Signals”, pp. 928-932, IEEE, 1994.

Affiliated with

Also associated with

Rating

Method for motion vector coding of MPEG-4 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 motion vector coding of MPEG-4, we encourage you to share that experience with our LandOfFree.com community. Your opinion is very important and Method for motion vector coding of MPEG-4 will most certainly appreciate the feedback.

Rate now

Comments { 0 }

Profile ID: LFUS-PAI-O-2991284