Pulse or digital communications – Bandwidth reduction or expansion – Television or motion video signal
Reexamination Certificate
2000-02-23
2003-09-23
Lee, Richard (Department: 2613)
Pulse or digital communications
Bandwidth reduction or expansion
Television or motion video signal
C386S349000
Reexamination Certificate
active
06625211
ABSTRACT:
BACKGROUND OF THE INVENTION
1. Field of the Invention
The present invention relates to a method and apparatus for transforming a moving picture coded by a coding system into another moving picture coded by another coding system.
2. Description of the Related Art
With the progress of multimedia techniques for handling voice, image and other data collectively, conventional information media such as a newspaper, magazine, television, radio, and telephone, which are means for conveying information to people, become subjects for the multimedia.
Generally, the multimedia are considered to show not only characters but also other information such as figures, voices, or particularly images, associated with characters at the same time. In order to make the above-mentioned conventional information media the subjects for the multimedia, representing the conventional information in digital format is the essential condition.
However, the estimation of the code amount of each of the conventional information media in digital format results in 64 kb per second for voices (telephone quality) and equal to or more than 100 Mb per second for moving pictures (current television reception quality), while the code amount for a character is 1 to 2 bytes. Therefore, it is not realistic to handle such a massive code amount in digital format without using any transformation.
For example, a video teleconferencing system is already put to practical use with an Integrated Services Digital Networks (ISDN) with a transmission rate of 64 kbps to 1.5 Mbps, it is however impossible to transmit images for a television or camera directly without the transformation with the ISDN.
A data compression technique is consequently required. For example, the video teleconferencing system uses moving picture compression techniques of the H.261 standard and H.263 standard that are international-standardized by the ITU-T (Telecommunication Standardization Sector of International Telecommunication Union).
Further, according to the information compression technique of the MPEG standard, ordinary television broadcast moving picture is compressed to 2 to 15 Mbps.
Herein, the MPEG (Moving Picture Experts Group) means the operation group (ISO/IEC JTC1/SC29/WG11) working for the standardization of speech and moving picture coding in the ISO (International Organization for Standardization), and also means the international standardization of the data compression that this group defines, at the same time. Currently, the MPEG has standardized MPEG-4 enabling the coding and processing by an object basis to achieve the new function needed for the multimedia technique.
A plurality of moving picture coding standards are thus to be used, resulting in the problem to be solved that coded data (hereinafter referred to as bit-stream) in such standards is not compatible with each other.
For example, although a decoder based on MPEG-4 is specified to be capable of decoding H.263 bit-stream, there is generally no compatibility of bit-stream between coding systems of H.261, H.263, MPEG-2 and MPEG-4 (in other words, it is not possible to use a coder and decoder based on different coding systems as a pair).
This condition requires an apparatus that transforms bit-streams based on the different coding systems (hereinafter referred to as transcoder).
FIG. 1
illustrates a conceptual diagram of the transcoder. In
FIG. 1
, different coding systems are A and B, decoder
101
with a system A and coder
103
with a system B are connected through a frame memory and image transformer
102
.
For example, it is assumed that the system A is MPEG-2 (picture size: 704×480 pixels), and that the system B is H.263 (picture size: 360×288 pixels). Although the picture sizes thereof are different from each other, it is possible to perform the transformation by once transforming the bit-stream to an image, and then coding the image again.
In addition, the H.261, H.263, MPEG-2 and MPEG-4 coding systems use Motion-Compensated Discrete Cosine Transformation coding system (hereinafter referred to as MC-DCT coding) as a common method. In the MC-DCT coding system, an image is divided into blocks each of which is called a micro block comprised of 16×16 pixels, a shifted amount (motion vector) that minimizes a differential between successive images (frames) is calculated, and the differential is subjected to Discrete Cosine Transform (DCT) coding.
The processing to obtain the motion vector that minimizes the differential between the frames to obtain a differential image from which the redundancy is removed is called motion compensation.
DCT is used to remove spatial redundancy left in the motion-compensated image (differential image).
In addition, a first frame is coded using only DCT because the differential coding with another frame is not performed. Such a frame is called I picture, and other general picture to be subjected to MC-DCT coding is called P picture.
If the MC-DCT coding system as described above is used in the coding systems A and B, it is possible in decoding in
FIG. 1
to obtain secondary information (hereinafter referred to as side information) generated in the MC-DCT coding such as moved information and DCT coefficient quantization, and it is expected that the use of the side information improves the computation efficiency and coding efficiency (image quality and coded content) in recoding.
A conventional example of the transcoder using the side information is a moving picture coding transformation apparatus disclosed in Japan Unexamined Patent Publication HEI10-271494.
FIG. 2
illustrates the concept of the apparatus.
When it is assumed that the systems A and B in the transcoder illustrated in
FIG. 1
are respectively MPEG-2 and H.263, a configuration of
101
to
103
in
FIG. 1
corresponds to that of
202
to
204
in FIG.
2
.
In the conventional example, since the two coding systems have the MC-DCT coding system as the common method, it is possible to omit motion vector searching in H.263 coding by executing scaling (transforming an absolute value by multiplying a ratio of picture sizes) on the motion vector obtained in MPEG-2 decoding.
As illustrated, for example, in
FIG. 2
, the input image (MPEG 2) and the output image (H.263) have different sizes. Thus, a motion vector scaling block
205
is provided and performs scaling of a motion vector of an input stream (MPEG 2) to transform the same to a size appropriate for the size of the output image and provides the result (H.
263
) to the coder
204
.
In the conventional example illustrated in
FIG. 2
, with the motion vector adopted as the side information, the computation efficiency in recoding is improved using the side information obtained in MPEG-2 decoding.
In composing a practical transcoder, however, there are many subjects to be examined other than the reuse of motion vector.
In the practical coder, a process called rate control is inevitable to adjust the code amount of bit-streams output therefrom.
Generally, an output buffer is installed at an output side to absorb a variation of generated code amount, and corresponding to the buffering data in the output buffer (hereinafter referred to as buffer loaded code amount), the coded content is adjusted.
In the MC-DCT coding system, the coded content adjustment is executed by setting the quantization accuracy for the DCT coefficient to variable (fine accuracy improves the image quality and increases the generated code amount, while rough accuracy results in the opposite effects).
FIG. 3
illustrates a conceptual diagram showing subjects for the transcoder.
An output terminal of output buffer
304
is connected to a communication path or storage medium. When it is assumed that the communication path has a fixed data rate, the bit-stream is output from the output buffer with the constant rate. Coder
303
with the system B adjusts the generated code amount so that data in buffer
304
is not consumed (underflow) or not excessive (overflow). Meanwhile, in order to improve the recoded image quality, it is necessary to use the side i
Etoh Minoru
Iwasaki Osamu
Komiya Daisaku
Uenoyama Tsutomu
Greenblum & Bernstein P.L.C.
Lee Richard
LandOfFree
Method and apparatus for transforming moving picture coding... 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 transforming moving picture coding..., we encourage you to share that experience with our LandOfFree.com community. Your opinion is very important and Method and apparatus for transforming moving picture coding... will most certainly appreciate the feedback.
Profile ID: LFUS-PAI-O-3108389