KR20080108934A - Method and apparatus for generating header information of stereoscopic image data - Google Patents

Abstract

A stereoscopic image bit stream production apparatus and a method thereof are provided to transmit and store stereoscopic image efficiently using header information. Information about a stereoscopic image is read from a header area of an image bit stream through the parsing of the image bit stream(450). Camera parameter insertion information which shows whether or not the camera parameter about the camera obtaining is inserted about the stereoscopic image is extracted(430). A stereoscopic image is restored using the stereoscopic image data extracted from the image bit stream(470). The stereoscopic image information is used in order to reproduce the restored stereoscopic image as two dimension or 3D playback mode(410).

Description

Method and apparatus for generating stereoscopic image bitstream {Method and apparatus for generating header information of stereoscopic image data}

The present invention relates to a method and apparatus for generating header information of stereoscopic image data, and more particularly, to a method of generating header information in which information necessary for reconstructing and reproducing an original image by decoding a coded stereoscopic image is decoded. And to an apparatus.

Currently, many methods for transmitting stereoscopic images have been proposed. For example, for efficient transmission of stereoscopic video, MPEG-4 depth map transmission method using MPEG-4 Multiple Auxiliary Component (MAC), including MPEG-2 Multi-view Video Profile (MVP), MPEG-4 Standards such as Multi-view Video Coding (MVC) of AVC / H.264 have been enacted. Currently, standards are being established for data on stereoscopic images in the MAF (Multimedia Application Format).

1 illustrates a stereoscopic video format in units of pixels for transmission of a stereoscopic video.

The pixels of the left view image and the right view image are sampled in a grid unit, and the right image is shifted and sampled by one pixel so that the left image and the right image do not overlap, thereby configuring a single image format for a stereoscopic image. A stereoscopic video transmission and reception method uses a conventional two-dimensional encoder and decoder. Lost pixels occur because two left and right images are configured in one stereoscopic image format of the same resolution.

2 illustrates a lost pixel reconstruction apparatus of a stereoscopic image format on a pixel-by-pixel basis.

In order for the display device to reproduce an image having an original resolution, pixels lost in the stereoscopic image format generation step of FIG. 1 must be restored. Therefore, the image sampled in units of pixels is extracted for each direction, and the original left and right images are restored by multiplying each directional pixel value by a predetermined weight and adding them all together.

3 shows a reference structure of an ISO based media file format.

In the conventional ISO-based media file format, actual media data is stored in the media data (mdat) area, and information about characteristics of the media data is stored in the movie data (Movie data, moov) area.

The transmission video format of FIG. 1 uses the fact that there is much information overlapping between the left and right images for transmission efficiency. In other words, by combining the left image and the right image in units of pixels, one mixed image is generated and transmitted, thereby reducing the amount of data to be transmitted, thereby increasing transmission efficiency.

However, in the conventional stereoscopic video format, the camera functions obtained from the scale function, the left image, and the right image used in the process of constructing a mixed image for transmission by combining a left image and a right image in a movie data area (moov). It does not include header information in consideration of information such as whether or not to transmit.

If an arbitrary method of composing a stereoscopic video into a mixed video is defined and implemented according to the video format of FIG. The inverse transform of the mixed image to obtain Therefore, even if lossless transmission is performed when using the conventional stereoscopic video format, the receiver may not reconstruct the correct stereoscopic video because there is no information such as a mixed video composition method and scaling information.

Accordingly, an object of the present invention is to propose a method and apparatus for generating header information of stereoscopic image data including information necessary for reconstructing a stereoscopic image composed of a mixed image in a transmission process in a reception process.

Another object of the present invention is to propose a method and apparatus for generating stereoscopic image data, which efficiently records and transmits a stereoscopic image through essential header information of various stereoscopic images, and restores and reproduces the original image at the receiving end.

Therefore, in order to solve the above problem, the method for generating a stereoscopic video bitstream according to the preferred embodiment of the present invention, the stereoscopic video data reproduced in three dimensions of the video data recorded in the payload region of the video bitstream Recording the 3D playback section information on the header area of the image bitstream, recording camera information on the camera from which the stereoscopic image is obtained, between the reference view image and the additional view image of the stereoscopic image. Recording the parallax information in the header region and recording the image data in the payload region of the bitstream.

One embodiment of the present invention includes the step of recording scaling method information used in a header image when a mixed image including information of a reference view image and an additional view image is included.

One embodiment of the present invention includes the step of recording information on a method of configuring a mixed image having information of a reference view image and an additional view image in a header area.

The 3D reproduction section information of one embodiment includes information on the number of 3D reproduction sections.

The three-dimensional reproduction section information of the preferred embodiment includes start index information indicating the index of the start frame of the three-dimensional playback section recorded in the payload region and end index information indicating the index of the last frame of the three-dimensional playback section recorded in the payload region. It includes.

When the three-dimensional reproduction section of the preferred embodiment is composed of one data stream, the start index information and the end index information indicate an index for one data stream, and when the three-dimensional reproduction section is composed of a plurality of data streams, the start index information And the end index information indicates an index of the main data stream among the plurality of data streams.

The camera information of the embodiment includes recording camera parameter information about the camera in a header area according to the type of camera and whether the stereoscopic image data transmits the camera parameter.

The camera parameter information of one embodiment includes information on the number of camera parameter coefficients and information on the rotation, parallel movement, image center, focal length, and aspect ratio of each of the reference view camera and the additional view camera.

The parallax information of the preferred embodiment includes parallax information of one global disparity vector for the entire stereoscopic image and parallax information of a representative disparity vector for a predetermined object of the stereoscopic image.

The parallax information of the global disparity vector of the preferred embodiment includes information on the number of regions in which the value of the global variation is changed among the frames recorded in the payload region, global variation start index information indicating the index of the frame in which the global variation is changed, and global. Contains vector value information of the variation.

A representative variation vector of a preferred embodiment is a variation vector having a minimum vector value and a variation vector having a maximum vector value, a variation vector of a background and a foreground, a variation vector of each object, and a variation map / depth map. Contains the maximum representable value / minimum value of

The parallax information of the representative variation vector of the preferred embodiment includes the number information of the representative variation vector and the vector value information of the representative variation vector.

The scaling method information of one embodiment is information about using at least one of a sampling method, a linear method, a cubic convolution method, and a variable scaling function as a scaling method.

The variable scaling function of one embodiment is determined by the horizontal and vertical coefficients, the number of horizontal coefficients and the number of vertical coefficients that can be set by an external request.

The mixed image configuration information of another embodiment includes information on arrangement order of the reference view image and the additional view image, and information about the width ratio and the height ratio of the reference view image and the additional view image.

In order to achieve the above object, the apparatus for generating a stereoscopic video bitstream in accordance with a preferred embodiment of the present invention is stereoscopic video data reproduced in three dimensions of the image data recorded in the payload region of the video bitstream A three-dimensional playback section information recording unit for recording three-dimensional playback section information on a header region of the video bitstream, a camera information recording section for recording camera information about a camera from which a stereoscopic image is obtained, and a stereoscopic image And a disparity information recording unit for recording disparity information between the reference view image and the additional view image in the header area, and an image recorder for recording the image data in the payload area of the bitstream.

One embodiment according to the present invention includes a scaling information recording unit for recording scaling method information used in a header area when a mixed image including information of a reference view image and an additional view image is included.

An embodiment according to the present invention includes a mixed image configuration information recording unit for recording information on a method of configuring a mixed image having information of a reference view image and an additional view image in a header area.

In order to achieve the above object, the stereoscopic video bitstream structure according to the preferred embodiment of the present invention, the payload region in which the image data is recorded and the stereoscopic image is reproduced in three dimensions of the image data And a header area including parallax information between the reference view image and the additional view image of the stereoscopic image, the camera information about the obtained 3D reproduction section information, the camera which acquired the stereoscopic image, and the like.

According to an embodiment of the present invention, the header information including scaling method information used when a mixed image including information of a reference view image and an additional view image is used is recorded in the header area.

According to an embodiment of the present invention, header information including information on a method of configuring a mixed image including information of a reference view image and an additional view image is recorded in a header area.

The present invention also includes a computer-readable recording medium having recorded thereon a program for implementing a stereoscopic video bitstream generation method.

The method and apparatus for generating stereoscopic image data according to the present invention can efficiently transmit and store stereoscopic images through header information essential for various stereoscopic image formats, and have universal effects.

In other words, since the parallax information of the stereoscopic image, the camera information, the configuration information of the mixed image, the reproduction section information of the 3D image, and the scaling information are recorded in the header area of the stereoscopic image data of the present invention, the receiver receives the stereoscopic image. There is an effect in which the data required for specifying a three-dimensional reproduction region that can feel a three-dimensional effect when desired and converting it to a stereoscopic image is specifically defined.

Also, by inserting the camera parameter and partial parallax information into the header information of the stereoscopic image data, the camera parameter and parallax information can be easily used in the post-processing process that may exist after the decoding stage.

Hereinafter, a method and apparatus for generating a stereoscopic video bitstream according to an embodiment of the present invention will be described in detail with reference to FIGS. 4A to 12B.

4A and 4B, an apparatus for generating stereoscopic video bitstreams according to an embodiment of the present invention will be described.

4A is a block diagram of an apparatus for generating a stereoscopic video bitstream according to an embodiment of the present invention.

The stereoscopic video bitstream generating apparatus 400 includes a 3D playback section information recording unit 410, a camera information recording unit 430, a parallax information recording unit 450, and an image recording unit 470.

The 3D reproduction section information recording unit 410 records 3D reproduction section information of the stereoscopic image data reproduced in 3D among the image data recorded in the payload region of the stereoscopic image bitstream, in the header region of the image bitstream. The video is output to the video recording unit 470.

The camera information recording unit 430 records camera information on the camera which acquired the stereoscopic image in the header area and outputs the information to the image recording unit 470.

The parallax information recording unit 450 records parallax information between the reference view image and the additional view image of the stereoscopic image in the header area and outputs the parallax information to the image recording unit 470.

After all of the header information is stored in the header area of the stereoscopic video bitstream, the 3D playback section information recording unit 410, the camera information recording unit 430, and the parallax information recording unit 450 are output to the image recording unit 470.

The image recorder 470 records the image data in the payload region of the image bitstream and outputs a stereoscopic image bitstream.

Therefore, the stereoscopic video bitstream generated by the stereoscopic video bitstream generating apparatus 400 according to an embodiment of the present invention stores image data in a payload region, and 3D reproduction section information and camera information in a header region. The header information of the parallax information is stored.

4B is a block diagram of an apparatus for generating a stereoscopic video bitstream according to another embodiment of the present invention.

The stereoscopic video bitstream generating apparatus 405 of the present embodiment includes a three-dimensional reproduction section information recording unit 410, a camera information recording unit 430, a parallax information recording unit 450, and an image of the stereoscopic video bitstream generating device 400. And a recording unit 470, and a mixed image configuration information recording unit 420 and a scaling information recording unit 440.

The mixed image configuration information recording unit 420 records the method information for configuring a mixed image including information of the reference view image and the additional view image in the header area and outputs the information to the image recorder 470.

The scaling information recording unit 440 records scaling method information used when a mixed image including information of a reference view image and an additional view image is recorded in a header area and outputs the information to the image recording unit 470.

Accordingly, the bitstream generated by the stereoscopic video bitstream generation device 405 according to another embodiment of the present invention may include 3D reproduction section information, image configuration information, camera information, and scaling for image data stored in a payload region. The information and parallax information are stored in the header area as header information.

5 to 9B, header information of the stereoscopic video bitstream generated by the stereoscopic video bitstream generating apparatus of the present invention will be described in detail.

FIG. 5 illustrates an embodiment of syntax of 3D reproduction interval information for a stereoscopic video bitstream generating device according to an embodiment of the present invention.

3D_Period represents 3D reproduction section information. In one embodiment of the present invention, three-dimensional reproduction section information is represented by 32 bits (u (32)). The 3D playback section information 3D_Period includes the number information (number_of_3D_period), the start index information (start_position), and the end index information (end_position) of the 3D playback section.

number_of_3D_period represents the number information of the 3D playback section, that is, the number of video data sections that the user can see in 3D.

start_position is start index information and indicates a position index in the payload region of the video bitstream in which the start frame of the 3D playback section in the payload region of the video bitstream is located. In one embodiment of the present invention, three-dimensional reproduction section information is represented by 32 bits (u (32)).

end_position is end index information and indicates a position index in the payload region of the video bitstream in which the last frame of the 3D playback section is located. In one embodiment of the present invention, three-dimensional reproduction section information is represented by 32 bits (u (32)).

Since the start index information (start_position) and the end index information (end_position) are required for each three-dimensional reproduction section, when i is an integer, the start index information (start_position [i]) and the end index information (end_position [i) in a range of 0 ≤ i <number_of_3D_period ]) Must be determined.

In one embodiment, when the three-dimensional reproduction section is composed of one data stream (ES), the start index information (start_position) and the end index information (end_position) indicate an index for one data stream.

Also, in another embodiment, when composed of a plurality of data streams (ES) of the three-dimensional playback section, the start index information (start_position) and the end index information (end_position) for the base data stream of the plurality of data streams Represents an index.

FIG. 6A illustrates an embodiment of syntax of mixed image configuration information for a stereoscopic video bitstream generating device according to an embodiment of the present invention. An embodiment of the present invention selects a base view image and an additional view image as a left view and a right view from among a plurality of view images.

The mixed image configuration information (stereoFormat_params) includes the mixed image format type (Stereo_format_type), the arrangement order of the base view image and the additional view image (LR_first), the width ratio (LR_width_ratio), the height ratio (LR_height_ratio), and the scaling method (scaling_method). ) And a scaling function (scaling_func).

Stereo_format_type is a mixed video format type of stereoscopic video and indicates a format type according to a method of constructing a mixed video using two-dimensional video of a plurality of viewpoints. The mixed image format of the stereoscopic image will be described in detail with reference to FIG. 6B. According to an embodiment, the mixed image format type (Stereo_format_type) is represented by 8 bits (u (8)).

FIG. 6B illustrates an embodiment of a mixed video configuration format used for mixed video configuration information for a stereoscopic video bitstream generation apparatus according to an embodiment of the present invention.

When Stereo_format_type is 0, this indicates a vertical line interleaved format. According to an embodiment of the vertical line interleaved format, lines in the vertical direction of the reference view image and the additional view image are alternately arranged in the lines of the mixed image to form a mixed image.

When Stereo_format_type is 1, it indicates a horizontal line interleaved format. According to an embodiment of the horizontal line interleaved format, horizontal lines of a reference view image and an additional view image are alternately arranged in lines of the mixed image to form a mixed image.

When Stereo_format_type is 2, this indicates a top-down format. In the top-down format, for example, a reference view image and an additional view image of which the resolution is reduced in half in the vertical direction are disposed on each of the top and bottom of the mixed image to form a single mixed image.

If Stereo_format_type is 3, this indicates a side by side format. In the side by side format, for example, a base view image and an additional view image having a resolution reduced in half in the horizontal direction are disposed in each of the left and right regions of the mixed image. Configure

When Stereo_format_type is 4, this indicates a field sequential format. The field sequential format is, for example, an image format according to a method of constructing a mixed image by alternately displaying a reference view video field and an additional view video field on a playback unit.

When Stereo_format_type is 5, it indicates a frame sequential format. The frame sequential format is, for example, an image format according to a method of constructing a mixed image by alternately displaying a reference view image frame and an additional view image frame on a playback unit.

When Stereo_format_type is 6, this indicates a block interleaved format. In the block interleaved format, for example, the reference view image and the additional view image are alternately arranged in blocks of the mixed image in units of blocks to form a mixed image.

When Stereo_format_type is 7, this indicates a disparity map. According to this format, it is an image format according to a method of constructing a mixed image using a reference view image and a disparity map.

If Stereo_format_type is 8, this indicates a depth map. According to this format, a mixed image is generated using a reference view image and a depth map.

When Stereo_format_type is 9, it represents a displacement and motion map. According to this format, a mixed image is generated using a reference view image and a displacement and motion map.

When Stereo_format_type is 10, this indicates a monoscopic image format. According to this format, one short-term view image of either the reference view image or the additional view image is used.

When Stereo_format_type is 11 to 255, this indicates a user private format. Therefore, an image format designated by a user is used to construct a mixed image.

LR_first is a parameter for the arrangement order information of the reference view image and the additional view image, and represents, for example, information on a method of disposing the left and right images. 6C will be described in detail with respect to the arrangement method information of the left and right images. One embodiment of the present invention represents LR_first as one bit u (1).

FIG. 6C illustrates an embodiment of a method of arranging a reference view image and an additional view image used for mixed image configuration information for a stereoscopic video bitstream generation apparatus according to an embodiment of the present invention.

If LR_first is 0, the left image is the reference view image, and the image data of the left image is an odd-numbered line in the vertical line interleaved format, an odd-numbered line in the horizontal line interleaved format, an upper-side in the top-down format, It is disposed in the left region in the side by side format, the odd numbered field in the field sequential format, the odd numbered frame in the frame sequential format, and the odd numbered block in the block interleaved format. Similarly, the right image becomes an additional view image and is disposed in an area opposite to an arrangement area of the left image in each of the mixed image formats described above.

When LR_first is 1, the right picture is the reference view picture, and the arrangement of the right picture and the left picture is opposite to the case where LR_first is 0 described above.

LR_width_ratio is a parameter for the width ratio of the reference view image and the additional view image. The LR_width_ratio parameter adjusts the ratio of the widths of the reference view and the additional view image when the mixed image format is the side-by-side format or the top-down format. One embodiment of the present invention represents LR_width_ratio as 32 bits (u (32)). LR_width_ratio is calculated as in Equation 1.

LR_width_ratio = width _base / width _additional

LR_height_ratio is a parameter for the height ratio of the reference view image and the additional view image, and is a parameter for adjusting the ratio of the height of the reference view and the additional view image when the mixed image format is the side-by-side format or the top-down format. One embodiment of the present invention represents LR_height_ratio as 32 bits (u (32)). LR_height_ratio is calculated as in Equation 2.

LR_height_ratio = height _base / height _additional

scaling_method is a scaling method parameter and indicates a scaling method in a scaling process that occurs when combining left and right images when there is one data stream. It is necessary to transmit the scaling method so that the decoder can correctly reconstruct the left and right images. The scaling method defined by this parameter is described with reference to FIG. 7A. The scaling method parameter scaling_method of one embodiment is represented by 8 bits (u (8)).

FIG. 7A illustrates an embodiment of a scaling function used for scaling method information (scaling_method) for an apparatus for generating a stereoscopic video bitstream, according to an embodiment of the present invention.

When the scaling method parameter value (scaling_method) is 0, it indicates that the stereoscopic image is scaled by the sampling method. The sampling method is a method in which pixels of a predetermined period are extracted and scaled from a short-term view image.

When the scaling method parameter value (scaling_method) is 1, it indicates that the stereoscopic image is scaled in a linear method. The linear method configures pixels of a mixed image using pixel values calculated through a linear equation using at least one pixel value in a short-term image.

When the scaling method parameter value (scaling_method) is 2, it indicates that the stereoscopic image is scaled by cubic convolution. Cubic convolution forms a pixel of a mixed image using an average value calculated in consideration of the distance to 16 surrounding pixels surrounding the current pixel.

When the scaling method parameter value (scaling_method) is 3 to 255, it indicates that the stereoscopic image is scaled by a user private scaling function.

FIG. 7B illustrates an embodiment of syntax of scaling method information (scaling_method) for the apparatus for generating a stereoscopic video bitstream, according to an embodiment of the present invention.

scaling_func is a scaling function set by the user when the scaling method parameter (scaling_method) is 3 or more. One embodiment of the present invention requires two-dimensional scaling function coefficients in the horizontal and vertical directions to set the scaling function.

number_of_hor_coeff represents the number of horizontal coefficients of the scaling function set by the user. One embodiment represents number_of_hor_coeff with 8 bits.

hor_scaling_coeff represents the horizontal coefficient value of the scaling function set by the user. Since the count value is required by the number of number_of_hor_coeff, when i is an integer, hor_scaling_coeff [i] should be set in a range of 0 ≤ i <number_of_hor_coeff. One embodiment represents hor_scaling_coeff [i] in 32 bits.

y _hor [i] is the horizontal pixel value of the scaled image, x _hor [i] is the horizontal pixel value of the original image, h _hor [i] is the horizontal coefficient value of the user-defined scaling function. _{The relation of hor} [i], x _hor [i] and h _hor [i] is as follows.

y _hor [i] = x _hor [i] * h _hor [i]

Where * represents a convolutional operation symbol.

number_of_ver_coeff represents the number of vertical coefficients of the scaling function set by the user. One embodiment represents number_of_ver_coeff with 8 bits.

ver_scaling_coeff represents the vertical coefficient value of the scaling function set by the user. Since the count value is needed by number_of_ver_coeff, ver_scaling_coeff [i] should be set in a range of 0 ≤ i <number_of_ver_coeff when i is an integer. One embodiment represents ver_scaling_coeff [i] in 32 bits.

y _ver [i] denotes the vertical pixel value of the scaled image, x _ver [i] denotes the vertical pixel value of the original image, and h _ver [i] denotes the vertical coefficient value of the scaling function of the user setting. _{The relation of ver} [i], x _ver [i] and h _ver [i] is as follows.

y _ver [i] = x _ver [i] * h _ver [i]

Similarly, where * represents a convolutional operation symbol.

FIG. 8A illustrates an embodiment of syntax of camera information for a stereoscopic video bitstream generating device according to an embodiment of the present invention.

camera_distance represents the distance between stereoscopic cameras. In one embodiment, the camera_distance is represented by 32 bits.

The view_type indicates a type of view of the stereoscopic image, such as a crossed-eye view or a parallel-eye view. The view_type of one embodiment is represented by 8 bits.

View_distance_vs_depth represents a viewing distance to use / validity depth ratio. In one embodiment, View_distance_vs_depth is represented by 32 bits.

camera_type represents a camera type used when a stereoscopic image is obtained. The camera_type of one embodiment is represented by 8 bits.

is_camera_params is a parameter related to information on whether to transmit camera parameters. If it is not 0, is_camera_params transmits parameters of cameras of reference and additional view cameras (Camera_params). Is_camera_params of one embodiment is represented by 1 bit.

is_parallel_info represents whether to transmit parallax information (Parallax_info). If it is not 0, parallax information is transmitted. Is_parallax_info of one embodiment is represented by 1 bit.

8B illustrates an embodiment of syntax of camera parameter information used for camera information for a stereoscopic video bitstream generating device according to an embodiment of the present invention. In this embodiment, the left view and the right view are used as examples of the reference view and the additional view.

The camera parameter information Camera_params includes the number of camera parameter coefficients (number_of_camera_params) and the parameters of left and right cameras (Left_camera_params and Right_camera_params).

number_of_camera_params is the number of camera parameter coefficients. In one embodiment, number_of_camera_params is represented by 8 bits.

Left_camera_params represents a parameter of the left camera. Left_camera_params includes parameters related to intrinsic properties such as rotation information, translation information, and image center, focal length, aspect ratio of the left view camera. Left_camera_params of one embodiment is represented by 32 bits.

Right_camera_params represents a parameter of a right viewpoint camera. The right_camera_params includes parameters related to intrinsic properties such as rotation information, parallel movement information, image center, focal length, aspect ratio, and the like of the right view camera. Right_camera_params of one embodiment is represented by 32 bits.

FIG. 9 illustrates an embodiment of syntax of parallax information Parallax_info for a stereoscopic video bitstream generating device according to an embodiment of the present invention.

The parallax information Parallex_info includes parallax information about a global disparity representing a disparity occurring for the entire stereoscopic image and a representative disparity vector generated for a predetermined object of the stereoscopic image. The target vector of the representative disparity vector includes a disparity vector having a minimum vector value and a disparity vector having a maximum vector value among a plurality of disparity vectors. In addition, certain objects targeted for representative variation include a background, a foreground, and each individual object.

Parallax information (Parallex_info) includes the number of global disparity vectors (num_of_global_disparity), the global disparity start index information (global_disparity_index), the global disparity vector value (global_disparity), the number of representative disparity vectors (num_of_representative_disparity), and the global disparity vector value (representative_disparity). ).

num_of_global_disparity represents the number of times that global variation information has been changed among frames recorded in the payload region. In one embodiment, num_of_global_disparity is represented by 32 bits.

global_disparity_index represents an index of a frame in which a global variation starts to occur among frames in which a global vector occurs. In one embodiment, global_disparity_index is represented by 32 bits.

global_disparity represents a vector value of a global variation. In one embodiment, global_disparity is represented by 16 bits.

i is an integer, and when 0 ≤ i <num_of_global_disparity, the global transition between global_disparity_index [i] and global_disparity_index [i + 1] th frames has a value of global_disparity [i].

num_of_representative_disparity represents information on the number of representative variants. In one embodiment, num_of_representative_disparity is represented by 8 bits.

representative_disparity represents a vector value of representative variation. The representative_disparity of one embodiment is represented by 16 bits.

i is an integer, and when it has a range of 0 ≦ i <num_of_representative_disparity, representative_disparity [i] has a representative variation value. The corresponding information represents the minimum and maximum values of the disparity vector, the disparity vector information of the representative object, and the maximum and minimum values that the disparity / depth vector map can express.

10 illustrates one embodiment of a header region of a stereoscopic video bitstream according to a preferred embodiment of the present invention.

In the header area of the stereoscopic video bitstream generated by the stereoscopic video bitstream generating apparatuses 400 and 405 according to an embodiment of the present invention, a 3D reproduction information area 1010 and a mixed image configuration information area 1030 are provided. And a camera information area 1050.

The 3D reproduction information area 3D period 1010 includes number information (number_of_3D_period, 1011), a start index (start_position) 1012, and an end index (end_position) 1013 of the 3D reproduction period.

The mixed image configuration information region (Stereoscopic format) 1030 includes a format type (Stereo_format_type, 1031) of the stereoscopic image, arrangement order information (LR_first, 1032) of the reference view image and the additional view image, widths of the reference view image and the additional view image. Parameters for the ratio (LR_width_ratio, 1033), parameters for the height ratio of the reference view image and the additional view image (LR_height_ratio, 1034), scaling method (scaling_method, 1035), number of horizontal coefficients of the scaling function (number_of_hor_coeff, 1036) , The horizontal coefficient value of the scaling function (hor_scaling_coeff [i], 1037), the number of vertical coefficients of the scaling function (number_of_ver_coeff [i], 1038), and the vertical coefficient value of the scaling function (ver_scaling_coeff [i], 1039). Include.

Camera information area (Stereoscopic camera, 1050) is the distance between the stereoscopic camera (camera_distance, 1051), the type of viewpoint (view_type, 1052) of the stereoscopic image, the viewing distance to effective depth ratio (View_distance_vs_depth, 1053), when the stereoscopic image is acquired Camera type used (camera_type, 1054), camera parameter transmission information (is_camera_params, 1055), number of camera parameter coefficients (number_of_camera_params, 1056), left camera parameter (Left_camera_params, 1057) and right camera parameter (Right_camera_params, 1058) It includes.

In addition, the camera information area (Stereoscopic camera, 1050) includes the parallax information transmission information (is_parallel_info, 1059), the number of information of the global disparity vector (num_of_global_disparity, 1060), the global disparity start index information (global_disparity_index [i], 1061), and the global disparity. A vector value (global_disparity [i], 1062), information on the number of representative variation vectors (num_of_representative_disparity, 1063), and a representative variation vector value (representative_disparity [i], 1064).

11A is a flowchart of a stereoscopic video bitstream generation method according to a preferred embodiment of the present invention.

In operation 1110, three-dimensional reproduction interval information for stereoscopic image data reproduced in three dimensions among image data to be recorded in the payload region of the image bitstream is recorded in the header region of the image bitstream.

In operation 1130, camera information about a camera that acquires a stereoscopic image is recorded in a header area.

In operation 1150, disparity information between the reference view image and the additional view image of the stereoscopic image is recorded in the header area.

In step 1170, image data is recorded in the payload region of the image bitstream.

11B is a flowchart of a stereoscopic video bitstream generation method according to an embodiment of the present invention.

In operation 1110, three-dimensional reproduction interval information for stereoscopic image data reproduced in three dimensions among image data to be recorded in the payload region of the image bitstream is recorded in the header region of the image bitstream.

In operation 1120, scaling method information used when a mixed image including information of a reference view image and an additional view image is recorded in a header area.

In operation 1130, camera information about a camera that acquires a stereoscopic image is recorded in a header area.

In operation 1140, mixed image configuration information about a method of configuring a mixed image using the reference view image and the additional view image is recorded in the header area.

In operation 1150, disparity information between the reference view image and the additional view image of the stereoscopic image is recorded in the header area.

In step 1170, image data is recorded in the payload region of the image bitstream.

Meanwhile, the above-described embodiments of the present invention can be written as a program that can be executed in a computer, and can be implemented in a general-purpose digital computer that operates the program using a computer-readable recording medium. The computer-readable recording medium may be a magnetic storage medium (for example, a ROM, a floppy disk, a hard disk, etc.), an optical reading medium (for example, a CD-ROM, a DVD, etc.) and a carrier wave (for example, the Internet). Storage medium).

So far I looked at the center of the preferred embodiment for the present invention. Those skilled in the art will appreciate that the present invention can be implemented in a modified form without departing from the essential features of the present invention. Therefore, the disclosed embodiments should be considered in descriptive sense only and not for purposes of limitation. The scope of the present invention is shown in the claims rather than the foregoing description, and all differences within the scope will be construed as being included in the present invention.

1 illustrates a conventional stereoscopic video composition method in pixel units.

2 is a block diagram of a system for recovering lost pixels of a conventional stereoscopic image.

3 shows a reference structure of an ISO based media file format.

4A is a block diagram of an apparatus for generating a stereoscopic video bitstream according to an embodiment of the present invention.

4B is a block diagram of an apparatus for generating a stereoscopic video bitstream according to another embodiment of the present invention.

FIG. 5 illustrates an embodiment of syntax of 3D reproduction interval information for a stereoscopic video bitstream generating device according to an embodiment of the present invention.

FIG. 6A illustrates an embodiment of syntax of mixed image configuration information for a stereoscopic video bitstream generating device according to an embodiment of the present invention.

6B illustrates an embodiment of a stereo video configuration format used for mixed video configuration information for a stereoscopic video bitstream generation apparatus according to an embodiment of the present invention.

FIG. 6C illustrates an embodiment of a method of disposing a reference view image and an additional view image used for mixed image configuration information for a stereoscopic video bitstream generation apparatus according to an embodiment of the present invention.

FIG. 7A illustrates an embodiment of a scaling function used for scaling method information for an apparatus for generating a stereoscopic video bitstream according to an embodiment of the present invention.

FIG. 7B illustrates an embodiment of syntax of scaling method information for a stereoscopic video bitstream generating device according to an embodiment of the present invention.

FIG. 8A illustrates an embodiment of syntax of camera information for a stereoscopic video bitstream generating device according to an embodiment of the present invention.

8B illustrates an embodiment of syntax of camera parameter information used for camera information for a stereoscopic video bitstream generating device according to an embodiment of the present invention.

FIG. 9 illustrates an embodiment of syntax of disparity information for an apparatus for generating a stereoscopic video bitstream, according to an embodiment of the present invention.

10 illustrates one embodiment of a header region of a stereoscopic video bitstream according to a preferred embodiment of the present invention.

11A is a flowchart of a stereoscopic video bitstream generation method according to a preferred embodiment of the present invention.

11B is a flowchart of a stereoscopic video bitstream generation method according to an embodiment of the present invention.

Claims (8)

A method for reconstructing a stereoscopic image from an image bitstream including a payload region into which stereoscopic image data is inserted, Reading attribute information about the stereoscopic image from a header region of the image bitstream through parsing the image bitstream; Extracting camera parameter insertion information indicating whether camera parameters for a camera that obtained the stereoscopic image have been inserted from the attribute information about the stereoscopic image; And Restoring a stereoscopic video using the stereoscopic video data extracted from the video bitstream; The attribute information about the stereoscopic image is used to reproduce the reconstructed stereoscopic image in a two-dimensional or three-dimensional playback mode. The method of claim 1, wherein the stereoscopic image reconstruction method, And reproducing the reconstructed stereoscopic image in a 2D or 3D reproduction mode. The method of claim 2, wherein the stereoscopic video reproduction step comprises: And reconstructing the reconstructed stereoscopic image in a 2D or 3D reproduction mode by using the attribute information about the stereoscopic image. The method of claim 1, wherein the stereoscopic image reconstruction method, And extracting the camera parameter from attribute information about the stereoscopic image based on the camera parameter insertion information. The method of claim 1, wherein the stereoscopic image reconstruction method, Based on the camera parameter insertion information, at least one of image center information of each of the reference view camera and the additional view camera, parallel movement information, focal length information, and aspect ratio information of each of the reference view camera and the additional view camera is selected from the attribute information of the stereoscopic image. Stereoscopic image reconstruction method further comprising the step of extracting. The method of claim 1, wherein the stereoscopic image reconstruction method, And extracting, from the attribute information about the stereoscopic image, mixed image configuration information about a method of composing a mixed image having reference view image data and additional view image data of the stereoscopic image. Stereoscopic video restoration method. The method of claim 6, The mixed image configuration information includes a vertical line interleaved format, a horizontal line interleaved format, a top-down format, and a side-by-side format. by-side format, field sequential format, frame sequential format, block interleaved format, disparity map, depth map, mutation and And at least one of a disparity and motion map and a stereoscopic image format composed of independent single-view images. The method of claim 1, wherein the stereoscopic image reconstruction method, When the video bitstream is an International Standardization Organization (ISO) based media file format, attribute information about the stereoscopic video is extracted from at least one box in the video bitstream, and at least one box in the video bitstream. And extracting the stereoscopic image data from the stereoscopic image reconstruction method.

Images