Pulse or digital communications – Bandwidth reduction or expansion – Television or motion video signal
Reexamination Certificate
1999-03-05
2001-08-07
Kelley, Chris (Department: 2713)
Pulse or digital communications
Bandwidth reduction or expansion
Television or motion video signal
Reexamination Certificate
active
06272179
ABSTRACT:
FIELD OF THE INVENTION
The present invention relates to an image coding apparatus, an image decoding apparatus, an image coding method, an image decoding method, and a data storage medium. More particularly, the present invention relates to an apparatus and a method for coding a motion vector (shape motion vector) of a shape signal indicating a shape of an object included in an image on a frame, an apparatus and a method for decoding a coded signal obtained by coding the shape motion vector, and a data storage medium for storing an image processing program for coding or decoding the shape motion vector by software.
BACKGROUND OF THE INVENTION
In recent years, we have greeted the age of multimedia in which audio, video and other data are integrally handled, and the conventional information media (i.e., means for transmitting information media), such as newspapers, magazines, radios, and telephones, have been adopted as the subjects of multimedia. Generally, “multimedia” means to represent, not only characters, but also diagrams, speeches, and especially images simultaneously in relation with each other. In order to handle the conventional information media as the subjects of multimedia, it is necessary to transform the information into a digital format.
When the quantity of data possessed by each information medium described above is estimated as the quantity of digital data, in case of characters, the data quantity for each character is 1~2 byte. However, in case of speech, the data quantity is 64 Kbits per second (quality for telecommunication) and, in case of moving picture, it is more than 100 Mbits per second (quality for current television broadcasting). So, as for the information media described above, it is not practical to handle such massive data as it is in the digital format. For example, although visual telephones have already been put to practical use by ISDN (Integrated Services Digital Network) having a transmission rate of 64 Kbps-1.5 Mbps, it is impossible to transmit an image of a television camera as it is by the ISDN.
So, data compression technologies are demanded. In case of visual telephones, the moving picture compression technologies standardized as H.261 and H.263 by ITU-T (International Telecommunication Union-Telecommunication Sector) are employed. Further, according to the data compression technology based on MPEG1, it is possible to record image data, together with audio data, in an ordinary music CD (compact disk).
MPEG (Moving Picture Experts Group) is an international standard of data compression for data of a moving picture (an image signal of a moving picture). In MPEG1, data of a moving picture is compressed to 1.5 Mbps, i.e., data of a television signal is compressed to about 1/100. Since the transmission rate to which MPEG1 is directed is limited to about 1.5 Mbps, MPEG2 is standardized to meet the demand for higher image quality. In MPEG2, data of a moving picture is compressed to 2~15 Mbps.
Under the existing circumstances, standardization of MPEG4 is now proceeded by the working group for standardization of MPEG1 and MPEG2 (ISO/IEC JTC1/SC29/WG11). MPEG4 enables coding and signal operation in object units, and thereby realizes new functions required in the age of multimedia. MPEG4 has originally aimed at standardization of a coding method at a low bit rate, but the aim of standardization is now extended to a more versatile coding process at a high bit rate adaptable to an interlaced image.
FIGS.
8
(
a
)-
8
(
f
) are diagrams for explaining an object-by-object (object-based) coding process.
Shown in FIG.
8
(
a
) is an image space Ts obtained from a color signal comprising a luminance signal and a chrominance signal and including an object To corresponding to an image. Shown in FIG.
8
(
b
) is an image space Ss obtained from a shape signal indicating a shape of the object. In the image space Ss shown in FIG.
8
(
b
), a portion painted in black is an inside region of an object So. The fish image (object) To in the image space Ts is represented by a color signal corresponding to pixels composing the inside region (pixels inside the object).
According to MPEG4, an image signal comprising the shape signal and the color signal corresponding to an object is coded for each unit region (macroblock) composed of a predetermined number of pixels. While MPEG1 and MPEG2 have introduced a technique for coding the color signal for each macroblock, MPEG4 adopts a technique for coding the color signal and the shape signal for each macroblock.
FIG.
8
(
c
) schematically shows a blocking process in which the color signal is divided into signals corresponding to respective macroblocks Tmb composing the image space Ts. FIG.
8
(
d
) schematically shows a blocking process in which the shape signal is divided into signals corresponding to respective macroblocks Smb composing the image space Ss.
With a view to performing an efficient coding process for a frame (image space)including an object, according to MPEG4, only the inside of a rectangular region including the object in the frame rather than an image signal corresponding to the entire frame is coded. The rectangular region is called a “bounding box”.
To be specific, for the color signal, as shown in FIG.
8
(
e
), the bounding box Tb is divided into portions corresponding to respective macroblocks (blocking process). Also, for the shape signal, as shown in FIG.
8
(
f
), a bounding box Sb is divided into portions corresponding to macroblocks (blocking process).
The image signal comprising the color signal and the shape signal is subjected to a motion compensation coding process for each macroblock like a motion compensation coding process according to MPEG1 and MPEG2.
In this motion compensation coding process, there should be a match of motion change from frame to frame between the color and shape signals. In actuality, however, there is no match between real motion of the object and motion of the object of respective signals which improves coding efficiency.
The motion compensation coding process is to perform coding to a motion compensation error of the color or shape signal. If motion information giving the motion compensation error smaller than that of motion information (motion vector) indicating real motion of the object is obtained in the motion compensation coding process, it is more desirable to perform the motion compensation process based on the motion information giving the smaller error, because this increases coding efficiency. Since the color signal is a multi-valued signal, and the shape signal is a binary signal for distinguishing between pixels inside the object and pixels outside the object, their qualities differ from each other a little, and correspondingly the motion vectors thereof giving smaller errors differ from each other a little.
The motion compensation error is a difference value between a color or shape signal corresponding to a macroblock to-be-processed (target macroblock)i.e., macroblock to be coded, in a current frame (frame being coded) and a color or shape signal corresponding to a region equal to the macroblock to-be-processed in size (prediction macroblock) in a previous (past) frame (coded frame serving as a reference frame). The motion vector of each macroblock is position information indicating the positioning relationship between the macroblock to-be-processed on the frame and the prediction macroblock.
According to MPEG4, the color signal and the shape signal are subjected to the motion compensation coding process based on the motion vectors (hereinafter referred to as a color motion vector and a shape motion vector) obtained from the corresponding color and shape signals, respectively.
Although a match might not be always found between the color motion vector and the corresponding shape motion vector with respect to size or direction, there is high correlation between them. For this reason, in MPEG4, according to the shape motion vector of the coded macroblock in a current frame and the corresponding color motion vector, a prediction value of the shape mot
Kelley Chris
Matsushita Electric Industrial Company Limited
Vo Tung
Wenderoth , Lind & Ponack, L.L.P.
LandOfFree
Image coding apparatus, image decoding apparatus, image... does not yet have a rating. At this time, there are no reviews or comments for this patent.
If you have personal experience with Image coding apparatus, image decoding apparatus, image..., we encourage you to share that experience with our LandOfFree.com community. Your opinion is very important and Image coding apparatus, image decoding apparatus, image... will most certainly appreciate the feedback.
Profile ID: LFUS-PAI-O-2524879