Pulse or digital communications – Bandwidth reduction or expansion – Television or motion video signal
Reexamination Certificate
1998-11-05
2002-10-15
Lee, Richard (Department: 2613)
Pulse or digital communications
Bandwidth reduction or expansion
Television or motion video signal
C375S240240
Reexamination Certificate
active
06466620
ABSTRACT:
FIELD OF THE INVENTION
The present invention relates to a method and apparatus for adaptively coding video signals; and, more particularly, to a method and apparatus capable of adaptively the coding video signals by way of predicting intra DC and AC coefficients, separately.
DESCRIPTION OF THE PRIOR ART
In digitally televised systems such as video-telephone, teleconference and high definition television systems, a large amount of digital data is needed to define each video frame signal since the video frame signal comprises a sequence of digital data referred to as pixel values. Since, however, an available frequency bandwidth of a conventional transmission channel is limited, in order to transmit the large amount of digital data therethrough, it is necessary to compress or reduce the volume of the data through the use of various data compression techniques, especially in the case of such low bit-rate video signal encoders as video-telephone and teleconference systems.
Among various video compression techniques, the so-called hybrid coding technique, which combines temporal and spatial compression techniques together with a statistical coding technique, is known to be most effective. Most hybrid coding techniques employ a motion compensated DPCM (Differential Pulse Coded Modulation), two-dimensional transform, e.g., DCT (Discrete Cosine Transform), quantization of transform coefficients, and an entropy coding, e.g., VLC (Variable Length Coding).
In the conventional MPEG-4 (Moving Picture Expert Group-4) coding scheme, an intra-mode input video signal is divided into blocks of M×N, e.g., 8×8, pixels and each block is transformed to a set of quantized coefficients, i.e., a DC and a plurality of AC quantized coefficients, through the use of two-dimensional transform, e.g., DCT and quantization thereof, M and N being positive integers. A set of quantized coefficients of a block is then DPCM coded based on a set of quantized coefficients of one of previously coded blocks in the frame.
Referring to
FIG. 1
, there is shown a portion of blocks of an intra-frame. In MPEG-4, a reference block RB of a current block is determined among a left and an upper blocks of the current block based on DC coefficients of the left and the upper and a left-upper blocks of the current block. If a current block is
108
, a reference block RB thereof is determined as:
If |
D
3
−
D
1
|<|<
D
1
−
D
2
|, then
RB is the upper block
104
,
else
RB is the left block
106
(Eq. 1)
wherein D
1
, D
2
and D
3
are DC quantized coefficients of the blocks
102
,
104
and
106
, respectively.
In other words, if the horizontal correlation between the DC quantized coefficients of the previously coded blocks is greater than the vertical correlation, i.e., |D
1
−D
2
|<|D
3
−D
1
|, the previously coded left block
106
is selected as the RB for the current block
108
; and if otherwise, the previously coded upper block
104
is selected. Thereafter, a DPCM DC coefficient for the current block
108
is generated by calculating the difference between the DC quantized coefficient of the RB and that of the current block
108
.
According to the conventional MPEG-4 coding method, the RB determined based on the DC quantized coefficients of the previously coded blocks is also used in predicting the AC quantized coefficients of the current block without considering the correlations between the AC quantized coefficients of the previously coded blocks, which may result in degraded coding efficiency.
SUMMARY OF THE INVENTION
It is, therefore, an object of the present invention to provide a method and apparatus capable of adaptively coding intra-video signals by separately predicting intra DC and AC coefficients.
In accordance with the present invention, there is provided a method for encoding a current block of a video signal based on P previously coded blocks, each of the current block and the previously coded blocks having M×N coefficients of a DC and a plurality of AC coefficients and P being an integer greater than 2, comprising the steps of:
(a) finding a DC reference block among the previously coded blocks based on a correlation of the DC coefficients of the previously coded blocks;
(b) setting a DC coefficient of the DC reference block as a DC reference coefficient;
(c) determining one of the previously coded blocks as an AC reference block based on a correlation of selected AC coefficients of the previously coded blocks;
(d) generating AC reference coefficients based on AC coefficients of the AC reference block; and
(e) encoding coefficients of the current block based on the DC and the AC reference coefficients.
In accordance with the present invention, there is provided an apparatus for encoding a video signal, the video signal is divided into a plurality of blocks of M×N pixels, comprising: a transform block for transforming the video signal on a block-by-block basis to thereby a set of M×N transform coefficients for each block, wherein the transform coefficients includes a DC and a multiplicity of AC transform coefficients; a quantization block for quantizing the set of transform coefficients for each block to provide a set of M×N quantized DC and AC coefficients; a memory for storing the set of quantized coefficients for each block; a prediction block for determining a DC reference block and an AC reference block of a current block based on quantized DC coefficients and quantized AC coefficients of previously coded blocks to thereby provide a set of DPCM (differential Pulse Coded Modulation) DC and AC coefficients of the current block based on quantized DC and AC coefficients of the DC and the AC reference blocks; and an entropy coding block for entropy coding the DPCM DC and the DPCM AC coefficients.
REFERENCES:
patent: 5299025 (1994-03-01), Shirasawa
patent: 5434567 (1995-07-01), Mack et al.
patent: 5533138 (1996-07-01), Kim et al.
patent: 5535013 (1996-07-01), Murata
patent: 5559557 (1996-09-01), Kato
patent: 5933533 (1999-08-01), Fukuda et al.
patent: 6005622 (1999-12-01), Haskell et al.
patent: 6148109 (2000-11-01), Boon et al.
patent: 0559982 (1993-09-01), None
Daewoo Electronics Co. Ltd.
Lee Richard
Pennie & Edmonds LLP
LandOfFree
Method and apparatus for adaptively coding video signals 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 and apparatus for adaptively coding video signals, we encourage you to share that experience with our LandOfFree.com community. Your opinion is very important and Method and apparatus for adaptively coding video signals will most certainly appreciate the feedback.
Profile ID: LFUS-PAI-O-2956066