KR20050038054A - Apparatus for adaptive multiplexing/demultiplexing for 3d multiview video processing and its method - Google Patents

Abstract

The present invention relates to an adaptive multiplexing / demultiplexing apparatus and method for three-dimensional multiview multimedia processing, and more particularly, to efficiently send a multiview image to a remote receiver as in an application field such as digital broadcasting. The present invention relates to an apparatus and a method for efficiently reconstructing a multiview image.

The adaptive multiplexing method for 3D multi-view multimedia processing of the present invention includes a preprocessing step of correcting a parameter or the like due to a characteristic difference of a camera with respect to a multi-view image input from a predetermined camera and removing noise; A multi-view image compression step of removing and compressing temporal and spatial redundancy between images output from each camera, and adding time and synchronization information to each of the compressed video streams, and generating an SL packet by dividing by AU units. Generating a flex multiplexer packet, which is one stream from the SL packets, separating the flex multiplexer packet into AU units according to the specifications of the MPEG-2 system, and obtaining time and synchronization information to generate a PES packet; The PES packet has a technical feature in that it consists of a multiplexing step including the step of transmitting a TS packet of 188 bytes to the receiver.

Accordingly, the adaptive multiplexing / demultiplexing apparatus and method for three-dimensional multiview multimedia processing of the present invention have an advantage of effectively transmitting an image obtained from a multiview image, and are compatible with a conventional system and efficiently again. The point image can be normalized.

Description

Apparatus for adaptive multiplexing / demultiplexing for 3D multiview video processing and its method

The present invention relates to an adaptive multiplexing / demultiplexing apparatus and method for three-dimensional multiview multimedia processing, and more particularly, to efficiently send a multiview image to a remote receiver as in an application field such as digital broadcasting. The present invention relates to an apparatus and a method for efficiently reconstructing a multiview image.

The demand for three-dimensional images for viewing a three-dimensional space that is perceived by human beings on TV or a movie is increasing rapidly. In particular, since digital broadcasting has been commercialized, there have been many studies on 3D video in fields such as 3D TV and 3D information terminal. Currently, 3D TVs through binocular (two cameras) stereo images using polarized glasses such as human eyes are being implemented. However, the binocular system as described above should always use glasses, problems such as focusing and many people at the same time difficult to see.

1 is a block diagram illustrating a prior art binocular stereo imaging system. Referring to FIG. 1, a system configuration includes a left eye motion compensated DCT encoder 1 and a right eye motion compensated DCT encoder 2, a disparity estimator DE, and a disparity compensator. And a stereoscopic encoder module comprising a motion and floating compensation DCT encoder 10 and a system multiplexer 5 and a left eye motion compensation DCT decoder 6 and a right eye motion compensation DCT decoder 7. And a three-dimensional decoder module including a floating compensation DCT decoder 11 and a system demultiplexer 9.

The operation of the system is as follows. In operation of the system, the left image is selected as the base image, the left eye motion compensation DCT encoder 1 is encoded by the MPEG-2 standard algorithm, and then the encoded stream is transmitted to the system multiplexer 5. Meanwhile, the left eye motion compensation DCT encoder 1 generates a reconstructed image for encoding the next image and the right image, and transmits the reconstructed image to the floating estimator 3 and the floating compensator 4. The right original image is passed to the motion compensation DCT encoder 2 and the floating estimator 3 for the right eye. In the floating estimator block 3, the reconstructed image of the left image and the original image of the right image are compared and evaluated with a disparity vector, DV) and pass it to the right eye compensation DCT encoder block (2) and floating compensator block (4). The DV delivered to the floating compensator block 4 generates a compensation image for the next right image by referring to the left restored image and the DV delivered to the right eye compensation DCT encoder block 2 is the same as the MV of the MPEG-2 standard. Is encoded. In the right eye motion compensation DCT encoder block 2, the right original image is encoded by an MPEG-2 encoding algorithm using DCT for the difference image between the compensation image generated in the floating compensator block 4 and the right original image. The encoded stream for the right picture is separately transmitted to the system multiplexer 5 in the same manner as the left picture. The system multiplexer 5 follows the specifications of the MPEG-2 system. The multiplexed streams from the system multiplexer 5 are passed to the system demultiplexer 9 and demultiplexed into two video encoding streams, each of which is left-eye motion compensation DCT decoder block 6 and right eye motion compensation. It is passed to the DCT decoder block 7. The left eye motion compensation DCT decoder block 6 that obtains the encoded stream of the left image reconstructs the image by an algorithm satisfying the MPEG-2 standard, and transfers the reconstructed image to the output and floating compensator block 8. In addition, the right eye motion compensation DCT decoder block 7 having received the encoded stream for the right picture reconstructs the stream to obtain a difference picture between the original picture and the compensation picture referring to the left picture generated by the DV and the floating compensator block 8. Acquire. The obtained DV is transferred to the floating compensator block 8 to generate a compensation image by referring to the previously restored left image, and the compensation image generated by the floating compensator block 8 is again a right eye motion compensation DCT decoder block 7. ) And the right image is reconstructed by combining with the reconstructed difference image.

2 is a block diagram illustrating a multiplexing / demultiplexing apparatus in a conventional digital broadcasting technique. The operation of the multiplexing device will be described. The video elementary stream (ES) and the audio elementary stream, which are the outputs of the video encoder 12 and the audio encoder 14, are input to the video packetizer 13 and the audio packetizer 15, respectively, to access AU (Access). It is divided into units, and time and synchronization information is added to make a packetized elementary stream (ES), which is output in units of PES packets. The PES packets are input to a transport stream (TS) multiplexer or a program stream (PS) multiplexer. When input to a TS multiplexer, a TS packet of 188 bytes is generated and output. The PS multiplexer outputs multiple video PES and audio PES in one or multiple packs. The demultiplexer reconstructs one or more encoded streams in the reverse order of the multiplexer operation.

However, as described above, the MPEG-2 system of the related art cannot operate a system having a data rate higher than 27 MHz, and cannot process multiple video ESs including multiview images, because the MPEG-2 system operates at a system clock of 27 MHz. There was a problem in that the point image cannot be efficiently restored by the receiver (decoder, terminal).

Accordingly, the present invention is to solve the above-mentioned disadvantages and problems of the prior art, based on a multi-view image transmission system that can enjoy a three-dimensional image at the same time without a pair of glasses in multiple cameras SUMMARY OF THE INVENTION An object of the present invention is to provide an adaptive multiplexing / demultiplexing apparatus capable of effectively transmitting a multiview image obtained and compatible with a conventional digital broadcasting system, and an operation method thereof.

The above object of the present invention is a pre-processing step of correcting a parameter due to the characteristic difference of the camera and the like to remove the noise for the multi-view image input from a predetermined camera; A multi-view image compression step of removing and compressing temporal and spatial redundancy between images output from each camera, and adding time and synchronization information to each of the compressed video streams, and generating an SL packet by dividing by AU units. Generating a flex multiplexer packet, which is one stream from the SL packets, separating the flex multiplexer packet into AU units according to the specifications of the MPEG-2 system, and obtaining time and synchronization information to generate a PES packet; The PES packet is achieved by an adaptive multiplexing method for three-dimensional multi-view multimedia processing, which comprises a multiplexing step comprising transmitting a 188-byte TS packet to a receiving unit.

Another object of the present invention is to extract a TS packet from the TS transmitted from the transmitter to generate a PES, extracting a flex multiplexer packet from the PES in accordance with the specification of the MPEG-4 system, SL packet from the flex multiplexer packet Demultiplexing comprising extracting and restoring a video image stream from the SL packet; It is also achieved by an adaptive demultiplexing method for three-dimensional multi-view multimedia processing, which comprises a multi-view image restoration step of restoring each compressed image and a multi-view image synthesis step of synthesizing the restored images in the multi-view image restoration step. .

In addition, another object of the present invention is to add a time packet and synchronization information to the video image stream and the SL packetizer for generating an SL packet divided by AU unit; A flex multiplexer performing an intermediate multiplexing to generate a multiplexer packet, which is a stream from the SL paddle; 3D multi-view multimedia processing consisting of a PES packetizer for separating the PLEX packet into AU units, obtaining time and synchronization information, generating a PES packet, and a TS multiplexer for transmitting the TS packet into 188 bytes It is also achieved by an adaptive multiplexing device.

Still another object of the present invention is a TS demultiplexer for generating a PES including a video image of each view by extracting a TS packet from the TS; A PES depacketizer for extracting a flex multiplexer packet from the video PES; It is also achieved by an adaptive demultiplexer for three-dimensional multi-view multimedia processing comprising a deplex multiplexer extracting an SL packet from the flex multiplexer packet and an SL depacketizer reconstructing a video image stream from the SL packet.

In addition, another object of the present invention is a pre-processing step for removing noise and correction of parameters due to the difference in characteristics of the camera for the multi-view image input from a predetermined camera and the temporal and spatial redundancy between the images output from each camera In the video multiplexing method using a multiview consisting of a video compression step of removing a and a multiplexing step of generating each independent video stream into a single multiplexed stream, the multiview image is a multiview image system presenter, a multiview image It is also achieved by an adaptive multiplexing method for three-dimensional multi-view multimedia processing, characterized by having a presenter and a camera presenter.

In addition, another object of the present invention is a demultiplexing step of restoring a multiplexed video stream to each independent video stream, a multi-view image restoring step of restoring each compressed image, and an image restored in the multi-view image restoring step. In the multi-view video demultiplexing method using a multi-view image synthesis step of synthesizing the multi-view image, wherein the multi-view image has a multi-view image system presenter, a multi-view image presenter and a camera presenter It is also achieved by an adaptive demultiplexing method for point multimedia processing.

The function of the multiplexing / demultiplexing device is divided into two parts. The first is the multiplexing which physically and logically sends several data to one logical channel, and the second is synchronizing several inputted data. The above functions vary depending on the environment to which they are applied, that is, the network environment and the application field.

Details of the above object and technical configuration of the present invention and the effects thereof according to the present invention will be more clearly understood by the following detailed description with reference to the drawings showing preferred embodiments of the present invention.

3 is a block diagram illustrating a three-dimensional multi-view image transmission system according to the present invention. As shown in FIG. 3, images are input from multiple cameras 115, and the preprocessor 110 corrects parameters due to differences in camera characteristics, and removes the redundancy of the images at each viewpoint as much as possible. The video compressor 105 compresses the video streams into elementary streams (ES), and inputs them to the multi-view video multiplexer 100 to generate each independent video stream as one multiplexed stream.

The receiver demultiplexes the multiplexed multiview image efficiently by the multiview video demultiplexer 120 and inputs each video image stream to a video decoder for each view existing in the multiview video decompressor 125 of the receiver. To restore each image compressed by the compressor 105. The multi-view video synthesizer 130 generates an intermediate image between the restored multi-view images, controls the display control unit 135, and outputs the intermediate image to the display 140.

The preprocessor 110 includes an imbalance reduction filter for correcting luminance and chromaticity, and a noise reduction filter for transmitting a video signal from which the noise is removed to the multi-view video compressor 105. The multi-view video compressor 105 encodes temporal and spatial redundancy according to the homogeneity between the closest camera images at the same time with respect to image frames continuously input from each camera and encodes the MPEG-2 standard. Compress through.

The multi-view video multiplexing apparatus 100 is a device for making an independent video stream compressed by the multi-view video compressor 105 into one multiplexed stream, and adds an identifier, time information, program information, etc. to each independent video stream. do. The detailed configuration of the multi-view video multiplexing device 100 is shown in FIG. 4.

4 is a block diagram illustrating a three-dimensional multiplexing / demultiplexing apparatus according to the present invention. FIG. 4A is a partial block diagram of a multiview video multiplexing device, including the Synchronization Layer (SL) -packetizer 150 of the MPEG-4 system, the Flex multiplexer 155, and the PES of the MPEG-2 system. It consists of a packetizer 160 and a TS multiplexer 165. At first, only video is considered, but it can be extended to audio.

In the video image stream at each time point, the SL packetizer 150 first adds time and synchronization information and divides the data into AU units to generate an SL packet, and performs an intermediate multiplexing process to make several SL packets into one stream. The multiplexer 155 passes through the multiplexer 155 to generate a flex multiplexer packet. The flex multiplexer packet enters the PES packetizer 160 and is separated into AU units according to the specifications of the MPEG-2 system to obtain time / synchronization information to generate a PES packet. The PES packet is generated by the TS multiplexer 165 as a TS packet of 188 bytes and transmitted to the receiver.

4 (b) is a partial block diagram of a multi-view video demultiplexer, the configuration of which is the SL depacketizer 170, the deplex multiplexer 175 of the MPEG-4 system, and the PES depacket of the MPEG-2 system. It is composed of a riser 180 and a TS demultiplexer 185. The receiver extracts TS packets from the TS demultiplexer 185 using the TS transmitted from the transmitter, and includes video PESs including program video, program information, multi-view video information, encryption information, and the like. Obtain Program Specific Information (PSI) information. The PSI is a table having information on a program and is divided into four types: a program association table (PAT), a program map table (PMT), a network information table (NIT), and a conditional access table (CAT). Each table in the PSI has a different PID and transmits only one type of table.

The video PESs are extracted from the PES depacketizer 180 into the flex multiplexer packet according to the specifications of the MPEG-4 system, and the flex multiplexer packets are extracted into the SL packets from the deplex multiplexer 175 to be the SL depacketizer. The video image stream is restored at 170.

As described above, the present invention includes both the MPEG-2 and MPEG-4 systems, and may selectively apply only the MPEG-2 system or the MPEG-4 system according to the video compression method. Of course, the MPEG-2 system and the MPEG-4 system are not applied as they are, and should include the new standard defined in the present invention.

5 is a flowchart illustrating a method of transmitting and receiving a 3D multiview image according to the present invention. First, in order to efficiently transmit and reconstruct a multiview image, it is necessary to define a feature of the multiview image. To this end, the present invention defines a multi_view_system_descriptor viewed from the perspective of a multiview image system, a mutli_view_video_descriptor for each viewpoint image, a camera_descriptor for defining characteristics of each camera, and the like. Descriptors defined above are applicable to both MPEG-2 and MPEG-4 systems and can be used adaptively.

A method of transmitting a 3D multiview image according to the present invention will be described based on the above-described presenters. First, multi-view images input from multiple cameras undergo a preprocessing step of correcting a parameter due to a camera characteristic difference and removing noise. The multiplexing step consists of multiplexing each independent video stream into a single multiplexed stream through a multi-view video compression step of removing and compressing temporal and spatial redundancy between images output from each camera. Here, the multi-view images may be compressed by a compressor for multiview image compression, such as a binocular, a three-eye and a five-eye, and the number of viewpoints of the input multiview image may vary. In the multiplexing step, an SL packet is generated by adding time and synchronization information to each video stream and dividing it by AU units, and generating a multiplexer packet, which is a stream to reduce overhead of a transport stream from multiple SL packets. After separating the AU unit according to the specifications of the MPEG-2 system, the PES packet is generated by acquiring time and synchronization information, and is transmitted to the receiving unit by converting the packet into a TS packet of 188 bytes.

This information is described using fields such as Num_Video for video number of multi_view_system_descriptor of adaptive multiplexing device, Num_Audio for audio number, Coding_Type for compression method, and the like. The information of the multi_view_system_descriptor is used to determine whether the decoding unit is a bundle of multiview images. When the multiview images are compressed, a reference image or a reference image may be used according to the method, and information such as distinguishing a viewpoint of the reference image may be needed. The above information is described using fields such as camera_number for a camera number of multi_view_video_descriptor, ref_camera_number for a camera number including a reference image, and independence for determining whether the current video image is a reference image. When decoding each viewpoint image in the decoder system, each image stream and multiview image reconstruction are used for mapping with the decoder and interconnection between the decoders. When decoding each view image, it is necessary to maintain temporal synchronization between the reference image and the reference image.

To this end, the adaptive multiplexing / demultiplexing apparatus makes this possible by newly defining the frame_number field in the PES packet header and redefining the usage of the AU_sequence_Number field in the SL packet header. In addition, camera information and geometry information for obtaining each view image necessary for efficiently displaying the decoded image are transmitted by using the camera_descriptor in the multiplexing / demultiplexing apparatus.

Next, a method of receiving a 3D multiview image according to the present invention will be described. First, a demultiplexing step of reconstructing a multiplexed video stream into each independent video stream is performed using the TS transmitted from the transmitter. A multi-view image restoration step of restoring each compressed image is performed, followed by a multi-view image synthesis step of synthesizing the restored image in the multi-view image restoration step.

Finally, the image is controlled by the display controller through the multi-view image synthesis step. In the demultiplexing step, the TS packet is extracted from the TS demultiplexer, the flex multiplexer packet is extracted from the PES depacketizer, and the SL packet is extracted from the deplex multiplexer to restore the video image stream from the SL depacketizer.

In detail, it first accepts MPEG-2 TS-based packets and finds a synchronization byte of 0x47. Then, the PAT with PID = 0 × 0000 is found and the program table is mapped with the selected program. The mapped program compares the information described in the multi_view_system_descriptor with the performance of the receiver to determine whether reception and decoding are possible. The PAT is a table indicating information on a program currently being transmitted, and includes a PID of a PMT having a program number and PIDs such as video and audio constituting the program.

After that, the PMT is found and ES_PIDs are obtained for the compressed streams of the respective view images, and at the same time, the decoder and the multiview image streams according to the reconstruction method and the reconstruction method of each view image are obtained. Get information about the multi_view_video_descriptor. This information is used for mapping between the decoding buffer, the transport buffer, and the respective compressed stream which are connected to the multi-view video decoder. In addition, the camera_descriptor describing the geometry information and the camera information of each viewpoint image is obtained to provide information so that the decoded multiview image is correctly displayed on the display device.

Each of the video streams using ES_PIDs is stored as a pure video stream in the decoding buffer through the PES depacket step, the deplex multiplexer step, and the SL depacket step. Then, each video stream synchronizes and decodes each reference video and the reference video by using temporal synchronization information obtained from the frame_number field of the PES or the AU_sequence_Number field of the SL packet in the multi-view video decoder. Each decoded image displays the image efficiently by using the camera_descriptor obtained in the PMT.

Finally, the details of the standardized features for efficiently transmitting a multiview image are described in FIG. 6.

6 is a diagram illustrating syntax in an operation module according to the present invention. FIG. 6A is a diagram illustrating a syntax of a multi_view_system_descriptor, that is, a multiview image system presenter, where descriptor_tag means type information of a descriptor, and descriptor_length means length information of a descriptor. In addition, Num_Video refers to the number information of the multi-view video, that is, the number of cameras, Num_Audio is audio information, reserved is video discrimination information, '0' for the two-dimensional video stream, '' for the three-dimensional video stream Set to 1 '.

FIG. 6B illustrates multi_view_video_descriptor, that is, syntax of a multiview image presenter, where descriptor_tag means type information of a descriptor and descriptor_length means length information of a descriptor. Camera_number is camera number information indicating a spatial view of a multiview image, and ref_Camera_number is camera number information including a reference image by a floating estimator. Independence is reference image information for determining whether the current video image is a reference image, and reserved is image discrimination information.

6 (c) is a diagram illustrating a syntax of a camera_descriptor, that is, a camera descriptor, where descriptor_tag denotes type information of a descriptor and descriptor_length denotes length information of a descriptor. Camera_id refers to camera number information, height_of_image_device refers to height information of an image displayed in a camera by 1/1000 mm when viewing an image through a camera, and vertical_angle_of_view refers to vertical angle information of a field of view to a 1/10000 angle unit. When looking at the image through the camera, if the image seen by the camera is defined as the image plane, image_plane_center_X is the horizontal coordinate information of 1/1000 mm of the center of the image plane in the coordinate system defined by the user, and image_plane_center_Y is defined by the user. The vertical coordinate information of the center of the image plane in the coordinate system, image_plane_center_Z is Z-axis coordinate information of the center of the image plane in the coordinate system defined by the user.

optical_center_X is the X-axis coordinate information of the center point when the image plane image is at one point of the camera lens, optical_center_Y is the Y-axis coordinate information of the center point when the image plane image is at the point of the camera lens, and optical_center_Z is the image plane Z-axis coordinate information of the center point when the lens is attached to a point of the camera lens. Finally, image_plane_vertical_X, Y, and Z are spatial coordinate information forming a vector perpendicular to the axis of the image plane starting from the center of the image plane and parallel to the edge of the image plane.

The present invention can be applied to digital broadcasting using the existing MPEG-2 system, and also to the MPEG-4 system used for internet streaming or other multimedia services. In addition, the existing standard has been extended to transmit, restore, and display three-dimensional images.

Although the present invention has been shown and described with reference to preferred embodiments as described above, it is not limited to the above-described embodiments and those skilled in the art without departing from the spirit of the present invention. Various changes and modifications will be possible.

Accordingly, the adaptive multiplexing / demultiplexing apparatus and method for three-dimensional multiview multimedia processing of the present invention have an advantage of effectively transmitting an image obtained from a multiview image, and are compatible with a conventional system and efficiently again. The point image can be normalized.

1 is a block diagram illustrating a prior art binocular stereo imaging system.

2 is a block diagram illustrating a multiplexing / demultiplexing apparatus in a conventional digital broadcasting technique.

3 is a block diagram illustrating a three-dimensional multi-view image transmission system according to the present invention.

4 is a block diagram illustrating a three-dimensional multiplexing / demultiplexing apparatus according to the present invention.

5 is a flowchart illustrating a method of transmitting and receiving a 3D multiview image according to the present invention.

6 is a diagram illustrating syntax in an operation module according to the present invention.

<Description of the symbols for the main parts of the drawings>

100. Multiview Video Multiplexer 105. Multiview Video Compressor

110. Multiview Video Preprocessor 115. Camera

120. Multiview Video Demultiplexer 125. Multiview Video Restorer

130. Multiview Video Synthesizer 135. Display Controls

140.Display 150.SL Packetizer

155. Flex Multiplexer 160. PES Packetizer

165.TS Multiplexer 170.SL Depacketizer

175. Deplex Multiplexer 180.PES Depacketizer

185.TS Demultiplexer

Claims (12)

In the video multiplexing method using multi-view, A preprocessing step of correcting a parameter or the like due to a characteristic difference of the camera with respect to a multi-view image input from a predetermined camera and removing noise; A multi-view image compression step of removing and compressing temporal and spatial redundancy between images output from each camera; And Adding time and synchronization information to each of the compressed video streams and dividing them by AU to generate an SL packet, generating a flex multiplexer packet as one stream from a plurality of SL packets, and emplexing the flex multiplexer packet -2 a multiplexing step comprising separating each AU unit according to a standard of the system, obtaining time and synchronization information, generating a PES packet, and transmitting the PES packet into a 188-byte TS packet to a receiver; Adaptive multiplexing method for 3D multi-view multimedia processing, characterized in that comprises a. The preprocessing step of removing noise and parameter due to the camera's characteristic difference and the image compression step of removing temporal and spatial redundancy between the images output from each camera, In the video multiplexing method using a multi-view consisting of a multiplexing step of generating an independent video stream as a single multiplexed stream, And said multiview image has a multiview image system presenter, a multiview image presenter and a camera presenter. The method of claim 2, The multiview video system presenter may include presenter type information, presenter length information, video number information, audio number information, and image discrimination information indicating whether the video stream is two-dimensional or three-dimensional video data. Adaptive Multiplexing Method for Dimensional Multiview Multimedia Processing. The method of claim 2, The multiview image presenter may include presenter type information, presenter length information, camera number information, reference camera number information indicating a camera number including a reference image, reference image information and an image stream for determining whether the current video image is a reference image. Adaptive multiplexing method for 3D multi-view multimedia processing, characterized in that it comprises image discrimination information indicating whether the two-dimensional or three-dimensional video data. The method of claim 2, The camera presenter may include presenter type information, presenter length information, camera number information, image height information, vertical angle information of the field of view, horizontal, vertical and Z-axis coordinate information of the center of the image plane, and one of the different camera lenses. The coordinates of the X, Y, and Z axes of the center point, and the X, Y, and Z coordinate information in space, starting from the center of the image plane and forming a vector perpendicular to the center line and parallel to the edges of the image plane. Adaptive Multiplexing Method for 3D Multi-view Multimedia Processing. In the video demultiplexing method using multi-view, Extracting a TS packet from a TS transmitted from a transmitter to generate a PES, extracting a flex multiplexer packet from the PES according to the specification of an MPEG-4 system, extracting an SL packet from the flex multiplexer packet, and the SL A demultiplexing step comprising restoring a video image stream in the packet; A multi-view image reconstruction step of reconstructing each compressed image; And A multiview image synthesis step of synthesizing the restored image in the multiview image restoration step Adaptive demultiplexing method for 3D multi-view multimedia processing, characterized in that comprises a. A demultiplexing step of restoring the multiplexed image streams into respective independent image streams, a multi-view image restoring step of restoring each compressed image, and a multi-view image synthesizing step of synthesizing the restored images in the multi-view image restoring step In the video demultiplexing method using a multi-view made, The multi-view image has a multi-view image system presenter, a multi-view image presenter and a camera presenter, adaptive demultiplexing method for 3D multi-view multimedia processing. The method of claim 7, wherein The multiview video system presenter may include presenter type information, presenter length information, video number information, audio number information, and image discrimination information indicating whether the video stream is two-dimensional or three-dimensional video data. Adaptive Demultiplexing Method for Dimensional Multiview Multimedia Processing. The method of claim 7, wherein The multiview image presenter may include presenter type information, presenter length information, camera number information, reference camera number information indicating a camera number including a reference image, reference image information and an image stream for determining whether the current video image is a reference image. Adaptive demultiplexing method for 3D multi-view multimedia processing, characterized in that it comprises image discrimination information indicating whether the two-dimensional or three-dimensional video data. The method of claim 7, wherein The camera presenter may include presenter type information, presenter length information, camera number information, image height information, vertical angle information of the field of view, horizontal, vertical and Z-axis coordinate information of the center of the image plane, and one of the different camera lenses. The coordinates of the X, Y, and Z axes of the center point, and the X, Y, and Z coordinate information in space, starting from the center of the image plane and forming a vector perpendicular to the center line and parallel to the edges of the image plane. Adaptive demultiplexing method for 3D multi-view multimedia processing. In the video multiplexing apparatus using multi-view, An SL packetizer for adding time and synchronization information to a video image stream and dividing the information into AU units to generate an SL packet; A flex multiplexer performing an intermediate multiplexing to generate a multiplexer packet, which is a stream from the SL paddle; A PES packetizer for splitting the flex multiplexer packet into AU units and generating a PES packet by obtaining time and synchronization information; And TS multiplexer for TS-forming the PES packet into 188 bytes Adaptive multiplexing device for three-dimensional multi-view multimedia processing, characterized in that comprises a. In the video demultiplexing apparatus using multi-view, A TS demultiplexer for extracting TS packets from the TS and generating a PES including a video image of each view; A PES depacketizer for extracting a flex multiplexer packet from the video PES; A deplex multiplexer for extracting an SL packet from the flex multiplexer packet; And SL depacketizer for restoring a video image stream from the SL packet Adaptive demultiplexer for three-dimensional multi-view multimedia processing, characterized in that configured to include.