Pulse or digital communications – Bandwidth reduction or expansion – Television or motion video signal
Reexamination Certificate
2001-08-02
2004-10-05
Vo, Tung T. (Department: 2613)
Pulse or digital communications
Bandwidth reduction or expansion
Television or motion video signal
Reexamination Certificate
active
06801574
ABSTRACT:
BACKGROUND OF THE INVENTION
1. Field of the invention
The present invention relates to an encoding quantity control method and video signal encoding apparatus for spatial scalability profile which is hierarchical coding in a spatial axis direction standardized (ISO/IEC 13818-2) by the ISO Moving Picture Experts Group (MPEG), or for other hierarchical coding techniques, and also relates to the storing and decoding of video signals compressed by the MPEG standard.
2. Description of the Prior Art
Conventional video signal layering-encoding apparatus is based on the spatial scalability profile which is hierarchical coding in a spatial axis direction standardized (ISO/IEC 13818-2) by the ISO Moving Picture Experts Group (MPEG).
The constitution and operation of the conventional video signal layering-encoding apparatus will hereinafter be described in reference to
FIG. 22
, which is a block diagram of the conventional video signal layering-encoding apparatus.
As shown in
FIG. 22
, an input video signal
5000
is converted to a lower resolution video signal than the input video signal
5000
by a resolution converter
5001
.
A motion detector
5002
stores the output of the resolution converter
5001
(the encoded frame of the input video signal
5000
) in image memory
5003
and also detects the motion of a macroblock unit from the encoded frame and the previously encoded and decoded past reference frame in the image memory
5003
. A motion compensator
5004
outputs a differential signal between the video signal of the encoded frame and the reproduced video signal of the reference frame detected by the motion detector
5002
in the unit of a macroblock. A discrete cosine transform (DCT) unit
5005
outputs the output (differential signal) of the motion compensator
5004
as DCT coefficients for each block. A quantizer
5006
quantizes the DCT coefficients by a quantization step (i.e., a divisor employed in quantization) specified by an encoding rate controller
5007
and then outputs the quantized DCT coefficients. An inverse quantizer
5008
inversely quantizes the DCT coefficients quantized at the quantizer
5006
by the quantized quantization step and then outputs the inversely quantized DCT coefficients. An inverse DCT (IDCT) unit
5009
outputs the output of the inverse quantizer
5008
as inverse DCT coefficients. A motion compensator
5010
generates a reproduced video signal by adding the output of the IDCT unit
5009
and the video signal of the reference frame whose motion was compensated at the motion compensator
5004
, and then stores the reproduced video signal in the image memory
5003
. A variable-length encoder
5011
variable-length encodes the output of the quantizer
5006
and a predetermined flag and then outputs the variable-length encoded video signal. A buffer
5012
stores the variable-length encoded video signal temporarily and adjusts the output speed of output data thereof. That is, the buffer
5012
outputs a variable-length encoded video signal with low resolution.
The encoding rate controller
5007
determines the quantization step of the encoded frame for each macroblock, based on the output (differential signal) of the motion compensator
5004
, the encoded length of the encoded video signal, and the residual quantity information on the buffer
5012
.
On the other hand, the resolution converter
5013
resolution converts the output (reproduced video signal) of the motion compensator
5010
, i.e., converts the resolution of the reproduced video signal to the same resolution as the input video signal
5000
and then stores it in image memory
5015
. A motion detector
5014
stores the encoded frame of the input video signal
5000
in the image memory
5015
and also detects the motion of a macroblock unit from the encoded frame, the previously encoded and decoded past reference frame (high resolution video signal) in the image memory
5015
, and from the reference frame that is the video signal of the same time among the low resolution signals which are the outputs of the resolution converter
5013
. That is, in this motion detection, the motion detector
5014
compares the encoded frame of the input video signal
5000
with the above-mentioned past reference frame, the above-mentioned reference frame of the same time, and a reference frame generated by a combination of these, and then detects a reference frame with the highest correlation in the unit of a macroblock. The reference frame with the highest correlation is output from the image memory
5015
to the motion compensator
5016
.
The motion compensator
5016
outputs a differential signal between the video signal of the encoded frame and the reproduced video signal of the reference frame detected by the motion detector
5014
in the unit of a macroblock. A DCT unit
5017
outputs the output (differential signal) of the motion compensator
5016
as DCT coefficients for each block. A quantizer
5018
quantizes the DCT coefficients by a quantization step specified by an encoding rate controller
5019
and then outputs the quantized DCT coefficients. An inverse quantizer
5020
inversely quantizes the DCT coefficients quantized at the quantizer
5018
by the quantized quantization step and then outputs the inversely quantized DCT coefficients. An IDCT unit
5012
outputs the output of the inverse quantizer
5020
as inverse DCT coefficients. A motion compensator
5022
generates a reproduced video signal by adding the output of the IDCT unit
5021
and the reproduced video signal of the reference frame whose motion was compensated at the motion compensator
5016
, and then stores the reproduced video signal in the image memory
5015
. A variable-length encoder
5023
variable-length encodes the output of the quantizer
5018
and a predetermined flag and then outputs the variable-length encoded video signal. A buffer
5024
stores the variable-length encoded video signal temporarily and adjusts the output speed of output data thereof. That is, the buffer
5024
outputs a variable-length encoded video signal with high resolution.
The encoding rate controller
5019
determines the quantization step of the encoded frame for each macroblock, based on the output (differential signal) of the motion compensator
5016
, the encoded length of the encoded video signal, and the residual quantity of the buffer
5024
.
In such a conventional video signal layering-encoding system, however, the two encoding rate controllers
5007
and
5019
control the encoding quantities individually and independently so that the mean bit rates of the outputs become constant, respectively. For that reason, in both the cases where the input video signal
5000
contains and does not almost contain a large quantity of high frequency component, the encoding quantities are controlled in the same manner. More specifically, the control for the encoding quantity between layers is not entirely performed according to the input video signal
5000
.
In addition, although encoding efficiency has been improved in encoding a high resolution video signal by an increase in the number of reference images for motion vector detection, an improvement in the encoding efficiency by using a low resolution signal more effectively has not been performed.
Next, a conventional digital broadcast decoding apparatus will be described in reference to
FIG. 23
, which illustrates a block diagram of the conventional digital broadcast decoding apparatus.
The conventional digital broadcast decoding apparatus shown in
FIG. 23
is a receiver for a digital broadcast that employs video and audio signals compressed and encoded by the MPEG standard.
As shown in
FIG. 23
, a demodulating section
6001
performs a predetermined demodulation on a received digital broadcast
6000
, thereby generating an MPEG stream. A separating section
6002
separates video and audio signal compressed streams on a predetermined channel from the generated MPEG stream. An audio signal decoding section
6003
decodes the audio signal compressed stream and outputs an audio signal
6020
.
Fujiwara Yuji
Nishino Masakazu
Takeuchi Seiichi
Wake Kazuhiro
Matsushita Electric - Industrial Co., Ltd.
Smith , Gambrell & Russell, LLP
Vo Tung T.
LandOfFree
Apparatus of layered picture coding, apparatus of picture... does not yet have a rating. At this time, there are no reviews or comments for this patent.
If you have personal experience with Apparatus of layered picture coding, apparatus of picture..., we encourage you to share that experience with our LandOfFree.com community. Your opinion is very important and Apparatus of layered picture coding, apparatus of picture... will most certainly appreciate the feedback.
Profile ID: LFUS-PAI-O-3290897