Image analysis – Image compression or coding – Including details of decompression
Reexamination Certificate
1999-03-03
2003-09-16
Mariam, Daniel G. (Department: 2621)
Image analysis
Image compression or coding
Including details of decompression
C382S232000, C382S243000, C382S284000, C382S311000, C348S384100, C348S584000, C358S426010, C345S215000, C375S240000, C375S240080, C375S240250
Reexamination Certificate
active
06621932
ABSTRACT:
FIELD OF THE INVENTION
The present invention relates to a video image decoding and composing method which can realize interactive operation by a user and a video image decoding and composing apparatus, and also a video image composition information coding method.
BACKGROUND OF THE INVENTION
Up to now, video image compression methods such as MPEG1, or MPEG2 have been utilized in the coding of natural video images. Further, there is an object coding method which divides a moving video image into objects in the video image and encodes the video images for each object and each background, as a new coding method.
FIG. 7
is a conceptual view showing composition of object video images. In the figure, numeral
701
represents a background video image, and numerals
702
to
705
designate video images of objects in the video images. Numerals
706
and
707
designate composed images. As the objects, there are rectangular shape object images such as background image
701
and arbitrary shape objects having arbitrary shapes other than a rectangular shape such as objects
702
to
705
. The arbitrary shape object is constituted by a texture video image showing a color signal and a shape video image showing the shape. The objects of the video images
704
,
705
are supposed to be located in the foreground relative to the objects of the video images
702
,
703
.
First of all, the background video image
701
and the texture video image
702
of the object which is the closest to the background are composed using the shape video image
703
, thereby to output the composed video image
706
. Next, the texture video image
704
is composed with the composed video image
706
using the shape video image
705
, thereby to output the composed video image
707
. By these operations, the composed video image is produced. Herein, as shape video images, there is a two-value video image which only indicates whether it is inside or outside of the object, and a video image which indicates the ratio of composition of the pixel values of the background to that of the object by multi-values, thereby enabling semi-transparent composition. In the object coding method, the rectangular shape object video image (
701
), the arbitrary shape object video image (
702
,
703
), and the arbitrary shape object video image (
704
,
705
) can be individually coded for each object. In the MPEG-4 video image coding method, it is possible to perform encoding of such object image having arbitrary shapes other than the rectangular shape.
On the other hand, even in the computer graphics data, the standardization of coding methods has been advancing. As a standard coding method, there is a virtual reality modeling language. In this coding method, information of a top, a line, and a surface, and their materials (such as color, light reflection parameter) can be coded, and it is possible in the decoding apparatus to re-structure the scene of computer graphics by decoding the coded signal of the virtual reality modeling language.
Recently, a coding method comprising combination of the object coding method and the computer graphics data coding method has also attracted attention. When the computer graphics data coding method is extended to the composition of the object video image, changing of the composition positions and the composition with the computer graphics images, of the object images of the object coding method can be performed.
In MPEG-4, there is also realized a coding method comprising a combination of the above-described object coding method and the computer graphics data coding method. By extending the computer graphics data coding method to the composition of the object video image, it is possible to compose the object video image of the object coding method with the computer graphics video image. Thereby, it is possible to realize the computer graphics of higher presentation ability relative to the prior art.
FIG. 8
shows an example of a video image decoding and composing apparatus comprising a combination of the object coding method and the computer graphics data coding method. Hereinafter, a coded signal describing the composition information of the object video image as the above-described extended computer graphics data coded format is called a composition information coded signal.
Numeral
801
designates a composition information coded signal, numeral
802
designates a composition information coded signal decoding means for analyzing the composition information coded signal
801
and outputting the result as composition information, numeral
803
designates a composition information memory storing the composition information as the output of the composition information coded signal decoding means
802
, numeral
804
designates a coded signal of the arbitrary shape object video image, numeral
805
designates an arbitrary shape object decoding means for decoding the coded signal
804
, and numeral
806
designates a shape memory storing a shape video image signal which is decoded by the arbitrary shape object decoding means
805
. Numeral
807
designates a texture memory for storing a texture video image signal which is decoded by the arbitrary shape object decoding means
805
. Numeral
808
designates a coded signal of the rectangular shape object video image and numeral
809
designates a rectangular shape object decoding means for decoding the coded signal
808
. Numeral
810
designates a video image memory for storing the video image signal which is decoded by the rectangular shape object decoding means
809
. Numeral
811
designates a composing means for composing the shape signal stored in the shape memory
806
, the texture signal stored in the texture memory
807
, and the video image signal stored in the video image memory
810
in accordance with the composition information stored in the composition information memory
803
. Numeral
812
designates a composition video image signal which is output from the composing means
811
.
The operation of the video image decoding and composing apparatus constructed as described above will be described with reference to the drawings and the tables which follow.
An example of the composition information coded signal
801
is shown in Table 4. This is described in a format similar to that of the virtual reality modeling language. For the detail of the format of the virtual reality modeling language, please see “VRML2.0-3D cyber space structuring language-” by Kenjiro Miura, Asakura Shoten, 1996. In the format, a node and a field accompanying the node are included. In this example, Group, Shape, Appearance, MovieTexture, and Rectangle objects are nodes. The “children” is a field of Group node, the “appearance” and “geometry” are fields of Shape node, the “texture” is a field of Appearance node, and url is a field of MovieTexture node. The Group node represents a group of nodes, and describes the collection of nodes at the “children” field. The “MovieTexture” node represents a moving video image which is to be texture mapped to the object (in this example, Rectangle node) which is represented by the “geometry” field of the Shape node, and the location of the coded video image signal corresponding to the moving video image is described in the url field.
The composition information coded signal may be compressed in the text format as shown in Table 4, or may be further compressed in the binary format as in MPEG-4.
The arbitrary shape object decoding means
805
inputs and decodes the arbitrary shape coded signal
804
, and the decoded shape video image is stored in the shape memory
806
and the texture video image in the texture memory
807
, respectively. The rectangular shape object decoding means
809
inputs and decodes the rectangular object coded signal
808
, and the decoded video image is stored in the video image memory
810
. The composing means
811
composes the texture video image of the arbitrary shape object stored in the texture memory
807
and the rectangular shape video image stored in the video image memory
810
in accordance with
Hagai Makoto
Matsui Yoshinori
Burr & Brown
Mariam Daniel G.
Matsushita Electric - Industrial Co., Ltd.
LandOfFree
Video image decoding and composing method and video 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 Video image decoding and composing method and video image..., we encourage you to share that experience with our LandOfFree.com community. Your opinion is very important and Video image decoding and composing method and video image... will most certainly appreciate the feedback.
Profile ID: LFUS-PAI-O-3062597