Image analysis – Image compression or coding – Shape – icon – or feature-based compression
Reexamination Certificate
1997-07-08
2001-11-06
Bella, Matthew C. (Department: 2621)
Image analysis
Image compression or coding
Shape, icon, or feature-based compression
C382S232000, C382S284000
Reexamination Certificate
active
06314209
ABSTRACT:
BACKGROUND OF THE INVENTION
1. Field of the Invention
The present invention relates to a video information coding method using an object boundary block merging/splitting technique, and in particular, to an improved video information coding method using an object boundary block merging/splitting technique which makes it possible to more effectively merge and code an object boundary block using a space surplus of a video or between videos existing in a block or between blocks and split the video through an object block splitting technique for a reconstruction of a recovering video and a display.
2. Description of the Conventional Art
Recently, the MPEG (Motion Picture Experts Group) which introduced an MPEG-I and MPEG-II which are an international standard recommendation with respect to a video and audio coding technique and system construction plans to introduce in 1998 an MPEG-IV to be used as a new international standard recommendation with respect to a new generation video and audio coding technique and system construction.
The MPEG-IV is directed to supporting a next generation video and audio application which can not be supported by the previous MPEGs.
In addition, the MPEG-IV includes a new technique for a communication and connection of a video and audio data, for example, an object-based interactive function and connection function through a network having a different characteristic.
Furthermore, it is possible to provide a characteristic by which the system operates under a predetermined communication environment such as an erroneous communication environment and a low transmission ratio.
Moreover, it is possible to code and control a natural video and audio and an artificial video and audio by using a computer graphic technique.
Namely, the MPEG-IV is capable of supporting many functions required in various applicable fields.
Therefore, the MPEG-IV may provide an open structure which is capable of supporting functions required for low cost and high-tech applicable fields as the multimedia technique advances and expandable functions.
Among the functions, there is known a function of enhancing a compression efficiency for a coding effect which is needed for a transmission and storing function and cost reduction.
As applicable fields with respect to the MPEG-IV, there are known a NDB (Networked Database Service) using an IMM (Internet Multimedia), an IVG (Interactive Video Games), an IPC (Interpersonal Communications), an ISM (Interactive Storage Media), an MMM (Multimedia Mailing), a WMM (Wireless Multimedia), an ATM, etc., an RES (Remote Emergency System), and an RVS (Remote Video Surveillance, etc.
In order to support an existing application and an next generation application, a video coding technique is needed so that a user communicates with a desired destination in a predetermined video region, searches and accesses a desired destination and edits the same.
Namely, the MPEG-IV which is to be completed in 1999 as a new video and audio coding technique is basically directed to satisfying the above-described desires.
FIG. 1
illustrates the construction of a video coding unit for an MPEG-IV, which has different construction from H.261, H.263, MPEG-I, and MPEG-II which are international standards for a video coding technique.
Namely, the video coding unit of the MPEG-IV is directed to estimating a motion by the macro block unit as a VOP with respect to an object video formed by a VOP formation unit
10
is inputted into a motion estimation unit
11
.
In addition, the motion information estimated by the motion estimation unit
11
is inputted into a motion compensation unit
12
, thus compensating the motion.
The VOP the motion of which is compensated by the motion compensation unit
12
is inputted into a subtractor
13
together with a VOP formed by the VOP formation unit
10
, thus measuring a difference value therebetween, and the thusly measured difference value by the subtractor
13
is inputted into a video signal coding unit
14
, so that a video signal of an object is coded by the sub-block unit of a macro block.
For example, the video signal coding unit
14
splits the X-axis and Y-axis of a macro block into 8×8 sub-blocks each having 8 pixels, thus coding the video signal of the object.
In addition, the VOP the motion of which is compensated by the motion compensation unit
12
and an internal information of an object coded by the video signal coding unit
14
are inputted into an adder
15
, and an output signal from the adder
15
is inputted into a previously reconstructed VOP
16
, thus detecting a VOP of a previous display.
The VOP of the previous display detected by the VOP detection unit
16
is inputted into the motion estimation unit
11
and the motion compensation unit
12
and is used for a motion estimation and motion compensation.
The VOP formed by the VOP formation unit
10
is inputted into a shape information coding unit
17
, and then the shape information is coded.
Here, the output signal from the shape information coding unit
17
is determined based on the applicable field of the VOP coding unit. The output signal from the shape information coding unit
17
is inputted into the motion estimation unit
11
, the motion compensation unit
12
, and the video signal coding unit
14
, respectively, and then is used for coding the internal information of the motion compensation and object.
In addition, the motion information estimated by the motion estimation unit
11
, the internal information of an object coded by the video signal coding unit
14
and the shape information coded by the shape information coding unit
17
are multiplexed by a multiplexor
18
, and then are transmitted in a bit stream through a buffer
19
.
In the video coding unit for an MPEG-IV, a shape coding unit and a video object plane are used.
Here, the VOP denotes an object on a time-based axis of a content having a predetermined shape that a user can accesses and edit. The VOP is coded by the VOP for supporting a content-based functionality.
FIG. 2
illustrates a video formed of an object, which video is reconstructed by macro blocks formed of 16×16 pixels for a video coding.
When the block is thusly reconstructed into macro blocks, there exist three kinds of macro blocks as shown in FIG.
3
. Namely, there are an intra-object macro block formed of an intra-object information, an extra-object not having an information of the object, and an object boundary macro block partially having an intra-object information.
FIG. 4
illustrates an arrangement of a luminance block and a color block. The intra-object macro block is formed of blocks of the object, and the extra-object macro block is formed of extra-object blocks.
However, in the object boundary macro block, there may exist three kinds of blocks each being formed of 8×8 pixels having the same characteristic as the macro block as shown in
FIG. 5
such as an intra-object sub-block, an extra-object sub-block, and an object boundary sub-block.
FIG. 6
illustrates a conventional video signal coding unit
14
(VM5.0). In the coding unit
14
, there are provided an extra-object region padding unit
14
a
for receiving an inter-video information or an original information from the motion compensation unit
12
and the VOP formation unit
10
as shown in
FIG. 1 and a
padding shape information (for example, an original video shape information and a recovered shape information) from the shape information coding unit
17
and padding the extra-object region, a DCT and coding unit
14
b
for DCT-coding the signal from the extra-object region padding unit
14
a
and outputting a signal information, and an IDCT and decoding unit
14
c
for IDCT-decoding the signal from the DCT and decoding unit
14
b
and outputting a decoded information to the previous VOP detection unit
16
.
In order to decode the intra-video and inter-video information of the object boundary block, the pixel value of pixels in the extra-object region is first determined. Here, this intra-video signal (intra-texture data) and an error signal
Kweon Ji-Heon
Moon Joo-Hee
Bella Matthew C.
Bryan Cave LLP
Hyundai Electronics Industries Co,. Ltd.
LandOfFree
Video information coding method using object boundary block... 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 information coding method using object boundary block..., we encourage you to share that experience with our LandOfFree.com community. Your opinion is very important and Video information coding method using object boundary block... will most certainly appreciate the feedback.
Profile ID: LFUS-PAI-O-2614646