WO2013157812A1 - 3d video coding/decoding apparatus and 3d video coding/decoding method - Google Patents

Abstract

A 3D video coding/decoding apparatus and a 3D video coding/decoding method are disclosed. The 3D video coding/decoding apparatus can code/decode a video according to existing techniques, such as a prediction technique, an interview prediction technique or a view synthesis technique, and can efficiently transmit a coding mode according to a prediction technique.

Description

3D video encoding / decoding apparatus and encoding / decoding method

The following embodiments are related to an apparatus and method for efficiently encoding / decoding a 3D video and transmitting the same.

The stereoscopic image refers to a 3D image that simultaneously provides shape information about depth and space. In the case of stereo images, images of different viewpoints are provided to the left and right eyes, whereas stereoscopic images provide the same images as viewed from different directions whenever the viewer views different views. Therefore, in order to generate a stereoscopic image, images captured at various viewpoints are required.

Images taken from various viewpoints to generate stereoscopic images have a large amount of data. Therefore, considering the network infrastructure, terrestrial bandwidth, etc. for stereoscopic video, even compression is performed using an encoding device optimized for Single-View Video Coding such as MPEG-2, H.264 / AVC, and HEVC. Realization may be nearly impossible.

However, since images taken at each viewpoint viewed by the observer are related to each other, there is a lot of overlapping information. Accordingly, a smaller amount of data may be transmitted by using an encoding apparatus optimized for a multiview image capable of removing inter-view redundancy.

In particular, since the depth image is an image representing a distance between an object and a viewer in a color image with a value of 0 to 255, characteristics between the depth image and the color image are similar. In general, 3D video includes a color image and a depth image of several views. However, 3D videos have temporal redundancy between successive images in time. In addition, 3D videos have inter-view redundancy among images corresponding to different viewpoints.

Therefore, there is a need for a method of encoding / decoding and transmitting 3D video in a more optimized form by efficiently removing such temporal redundancy and inter-view redundancy.

A decoding method according to an embodiment includes extracting first identification information indicating whether a current block is in view synthesis skip mode; Extracting second identification information indicating whether the current block is in a normal skip mode when the current block is not in the view synthesis skip mode according to the first identification information; And decoding the current block by using the first identification information or the second identification information.

The first identification information may be located before the second identification information when both the left block and the upper block of the current block are in the view synthesis skip mode.

The view synthesis skip mode may use a synthesized image of a virtual view that is the same view as the current view of the current block, and the synthesized image may be determined from encoded texture information and depth information of a neighbor view of the current view.

The normal skip mode may include at least one of an intra skip mode, an inter skip mode, or an interview skip mode.

According to another embodiment, a decoding method includes extracting first identification information indicating whether a current block is in a normal skip mode; Extracting second identification information indicating whether the current block is in view synthesis skip mode when the current block is not in the normal skip mode according to the first identification information; And decoding the current block by using the first identification information or the second identification information.

The first identification information may be located before the second identification information when at least one of the left block and the upper block of the current block is the normal skip mode.

The view synthesis skip mode may use a synthesized image of a virtual view that is the same view as the current view of the current block, and the synthesized image may be determined from texture information and depth information encoded at a neighbor view of the current view.

The normal skip mode may include at least one of an intra skip mode, an inter skip mode, or an interview skip mode.

In another embodiment, a decoding method includes extracting identification information indicating an encoding mode of a current block from a bitstream; Decoding a current block based on the identification information, wherein the identification information includes first identification information indicating whether the current block is a view synthesis skip mode and a second identification indicating whether the current block is a normal skip mode. May contain information.

The order between the first identification information and the second identification information may vary according to an encoding mode of a neighboring block of the current block, and the neighboring block may include an upper block and a left block of the current block.

The first identification information may be located before the second identification information if the upper block and the left block of the current block are both the view synthesis skip mode, and may be located later than the second identification information.

According to an embodiment, an encoding method may include determining an encoding mode of a neighboring block adjacent to a current block; Adaptively placing identification information associated with a skip mode of a current block according to the encoding mode; And encoding the identification information and the current block.

The identification information may include first identification information indicating whether the current block is a view synthesis skip mode and second identification information indicating whether the current block is a normal skip mode.

According to another exemplary embodiment, an encoding method includes determining whether an encoding mode of an upper block and a left block of a current block is a view synthesis skip mode; Determining a location of first identification information indicating whether the current block is in a view synthesis skip mode and second identification information indicating whether the current block is in a normal skip mode according to a determination result; And encoding the first identification information, the second identification information, and the current block.

The decoding apparatus according to an embodiment extracts first identification information indicating whether the current block is in view synthesis skip mode, and if the current block is not in view synthesis skip mode according to the first identification information, the current block is general An identification information extracting unit for extracting second identification information indicating whether the skip mode is present; And a decoder which decodes the current block by using the first identification information or the second identification information.

The first identification information may be located before the second identification information when both the left block and the upper block of the current block are in the view synthesis skip mode.

The view synthesis skip mode may use a synthesized image of a virtual view that is the same view as the current view of the current block, and the synthesized image may be determined from encoded texture information and depth information of a neighbor view of the current view.

The normal skip mode may include at least one of an intra skip mode, an inter skip mode, or an interview skip mode.

The decoding apparatus according to another embodiment extracts first identification information indicating whether the current block is in the normal skip mode, and skips the current block in view synthesis if the current block is not in the normal skip mode according to the first identification information. An identification information extraction unit for extracting second identification information indicating whether the mode is in the mode; And a decoder which decodes the current block by using the first identification information or the second identification information.

The first identification information may be located before the second identification information when at least one of the left block and the upper block of the current block is the normal skip mode.

The view synthesis skip mode may use a synthesized image of a virtual view that is the same view as the current view of the current block, and the synthesized image may be determined from texture information and depth information encoded at a neighbor view of the current view.

The normal skip mode may include at least one of an intra skip mode, an inter skip mode, or an interview skip mode.

In another embodiment, a decoding apparatus includes: an identification information extracting unit configured to extract identification information indicating an encoding mode of a current block from a bitstream; And a decoder which decodes the current block based on the identification information, wherein the identification information includes first identification information indicating whether the current block is a view synthesis skip mode and a second indicating whether the current block is a normal skip mode. It may include identification information.

The order between the first identification information and the second identification information may vary according to an encoding mode of a neighboring block of the current block, and the neighboring block may include an upper block and a left block of the current block.

The first identification information may be located before the second identification information if the upper block and the left block of the current block are both the view synthesis skip mode, and may be located later than the second identification information.

An encoding apparatus according to an embodiment may include an encoding mode determiner configured to determine an encoding mode of a neighboring block adjacent to a current block; An identification information disposition unit for adaptively disposing identification information associated with a skip mode of a current block according to the encoding mode; And an encoding unit encoding the identification information and the current block, wherein the identification information includes first identification information indicating whether the current block is a view synthesis skip mode and second identification information indicating whether the current block is a normal skip mode. It may include.

An encoding apparatus according to another embodiment may include an encoding mode determination unit configured to determine whether encoding modes of an upper block and a left block of a current block are both view synthesis skip modes; An identification information arranging unit configured to determine a position of first identification information indicating whether the current block is in a view synthesis skip mode and second identification information indicating whether the current block is in a normal skip mode according to a determination result; And an encoding unit encoding the first identification information, the second identification information, and the current block.

A computer-readable recording medium recording a bitstream, the bitstream includes: encoded current block, first identification information indicating whether the current block is a view synthesis skip mode, and second identification indicating whether the current block is a normal skip mode. And the first identification information may be located in the bitstream before the second identification information when both the upper block and the left block of the current block are in the view synthesis skip mode.

A decoding apparatus according to an embodiment includes an identification information extracting unit for extracting first identification information and second identification information indicating an encoding mode of a current block in a bitstream; And a decoding unit to decode a current block based on the identification information, wherein the first identification information and the second identification information are selected from a view synthesis skip mode, an interview skip mode, an inter skip mode, and an intra skip mode, and The position allocated to the bitstream may vary according to the encoding mode of the neighboring blocks of the block.

A decoding method according to an embodiment includes extracting first identification information and second identification information indicating an encoding mode of a current block from a bitstream; And decoding the current block based on the identification information, wherein the first identification information and the second identification information are selected from a view synthesis skip mode, an interview skip mode, an inter skip mode, and an intra skip mode, and The position allocated to the bitstream may vary according to the encoding mode of the neighboring blocks of the block.

1 is a diagram illustrating an encoding apparatus and a decoding apparatus, according to an embodiment.

2 is a diagram illustrating a detailed configuration of a decoding apparatus according to an embodiment.

3 is a diagram illustrating a decoding method according to an exemplary embodiment.

4 is a diagram illustrating a structure of a multiview video according to an embodiment.

FIG. 5 is a diagram illustrating an example of a reference picture used to encode a current block according to an embodiment.

6 illustrates a view synthesis skip mode and a view synthesis direct mode according to an embodiment.

7 illustrates a residual signal coding mode of view synthesis according to an embodiment.

8 illustrates a view synthesis according to an embodiment.

9 is a diagram for describing an encoding process using a synthetic image of a virtual view, according to an exemplary embodiment.

10 is a diagram for describing a decoding process using a synthetic image of a virtual view, according to an exemplary embodiment.

11 illustrates a bitstream structure for a macroblock according to an embodiment.

12 illustrates a bitstream structure for a macroblock according to another embodiment.

Hereinafter, embodiments will be described in detail with reference to the accompanying drawings. A 3D video encoding / decoding system method is proposed to efficiently reduce redundancy between viewpoints.

 Existing compression systems perform block-based prediction to remove redundancy between images. In order to efficiently remove redundancy between viewpoints when performing block-based prediction, a view synthesis method may be used in addition to the existing prediction method. The view synthesis method includes (i) a skip (SKIP) mode in which a predicted block is sent without a residual signal between the predicted block and the original block, (ii) a direct mode in which a residual signal is encoded except for motion information, and a motion. It can be classified into a residual coding mode in which both the information and (iii) the residual signal are encoded.

Hereinafter, a skip mode / direct mode based on intra prediction will be expressed as an intra skip mode and an intra direct mode. In addition, a skip mode / direct mode based on inter prediction is expressed as an inter skip mode and an inter direct mode. The skip mode and the direct mode based on the interview prediction will be expressed by the interview skip mode and the interview direct mode. On the other hand, the view synthesis skip mode / direct mode will be expressed as a view synthesis skip mode and a view synthesis direct mode.

1 is a diagram illustrating an encoding apparatus and a decoding apparatus, according to an embodiment.

The encoding apparatus 101 according to an embodiment may encode 3D video and then transmit the encoded data to the decoding apparatus 102 in the form of a bitstream. The encoding apparatus 101 according to an embodiment may improve encoding efficiency by removing redundancy between images as much as possible when encoding 3D video.

Intra, Inter, and Inter-View prediction methods may be used to remove the redundancy between the images. In addition, various coding modes (SKIP, 2NX2N, NXN, 2NxN, NX2N, and intra modes) may be used when predicting a block. According to an embodiment, in addition to the currently defined skip mode, by defining a view synthesis skip mode based on the synthesis prediction of the virtual view, the probability that more blocks constituting the current image may be encoded in the skip mode is increased.

In this case, the encoding apparatus 101 may synthesize the images of the neighboring views, which are already encoded, generate a synthesized image of the virtual view, and encode the image of the current view by using the generated synthesized image. Here, the virtual view may mean a view at the same position as the current view to be encoded. In this case, the images of the neighboring viewpoints may include texture information (color information) and depth information of the neighboring viewpoints.

Since the encoding apparatus 101 transmits only one bit for identifying a skip mode to the decoding apparatus 102 without encoding block information for blocks compressed in a skip (SKIP) mode, many bits are compared with other coding modes. The amount can be reduced.

In addition, since the direct mode is a coding mode in which only the residual signal is compressed and transmitted without motion information, a bit amount related to motion information can be reduced. Therefore, when the encoding apparatus 101 compresses a block, the more the skip mode or the direct modes are selected as the coding mode, the more the compression performance can be improved.

According to an embodiment, the encoding apparatus 101 allows more blocks to be compressed into the skip mode through interview prediction and view synthesis along with a skip mode according to a conventional prediction technique such as intra prediction or inter prediction. In addition, a skip flag (SKIP Flag) for identifying a skip mode using the composite image of the virtual view is required.

In this case, the skip flag of the current block may adaptively determine the positional relationship between the SKIP flag indicating the existing skip mode and the SKIP Flag indicating the skip mode of view synthesis based on the compression information of neighboring blocks adjacent to the current block ( Context-based Adaptive Skip Flag Positioning.

2 is a diagram illustrating a detailed configuration of a decoding apparatus according to an embodiment.

Referring to FIG. 2, the decryption apparatus 102 may include an identification information extractor 201 and a decoder 202.

The identification information extractor 201 may first extract first identification information indicating whether the current block is in a view synthesis skip mode. If the current block is not the view synthesis skip mode according to the first identification information, second identification information indicating whether the current block is the normal skip mode may be extracted. In this case, the identification information may be represented by a flag in the bitstream. For example, the first identification information may be represented by mb_vsskip_flag, and the second identification information may be represented by mb_skip_flag.

In this case, the first identification information means to be located before the second identification information. For example, if both the upper block and the left block of the current block are the view synthesis skip mode, the first identification information may be located before the second identification information. On the contrary, when at least one of the upper block and the left block of the current block is the normal skip mode, the first identification information may be located after the second identification information. The positions of the first identification information and the second identification information may be adaptively changed according to encoding modes of the upper block and the left block, which are neighboring blocks of the current block. Here, the general skip mode may include at least one of an intra skip mode, an inter skip mode, and an interview skip mode except the view synthesis skip mode.

The decoder 202 may decode the current block by using the first identification information and the second identification information. If the current block is encoded according to the view synthesis skip mode according to the first identification information, the decoder 202 may use the synthesized image of the virtual view generated according to the texture information and the depth information of the neighboring view of the current view. The current block can be decrypted. In addition, when the current block is encoded according to the normal skip mode according to the second identification information, the decoder 202 determines any one of an intra skip mode, an inter skip mode, or an interview skip mode, and then considers the corresponding skip mode. The current block can be decrypted.

Meanwhile, when the current block is neither the view synthesis skip mode nor the normal skip mode according to the first identification information and the second identification information, the decoder 202 may use the residual signal for the synthesized image of the virtual view based on the current signal. The block can be decrypted.

3 is a diagram illustrating a decoding method according to an exemplary embodiment.

In operation 301, the decoding apparatus 102 may first extract first identification information indicating whether the current block is in view synthesis skip mode. In step 301, if the current block is not the view synthesis skip mode according to the first identification information, in step 302, the decoding apparatus 102 indicates second identification information indicating whether the current block is the normal skip mode. Can be extracted. In this case, the identification information may be represented by a flag in the bitstream. For example, the first identification information may be represented by mb_vsskip_flag, and the second identification information may be represented by mb_skip_flag.

In this case, the first identification information means to be located before the second identification information. For example, if both the upper block and the left block of the current block are the view synthesis skip mode, the first identification information may be located before the second identification information. On the contrary, when at least one of the upper block and the left block of the current block is the normal skip mode, the first identification information may be located after the second identification information. The positions of the first identification information and the second identification information may be adaptively changed according to encoding modes of the upper block and the left block, which are neighboring blocks of the current block. Here, the general skip mode may include at least one of an intra skip mode, an inter skip mode, and an interview skip mode except the view synthesis skip mode.

In operation 303, the decoding apparatus 102 may decode the current block by using the first identification information and the second identification information. If the current block is encoded according to the view synthesis skip mode according to the first identification information, the decoding apparatus 102 may use the synthesized image of the virtual view generated according to the texture information and the depth information of the neighboring view of the current view. The current block can be decrypted. In addition, when the current block is encoded according to the general skip mode according to the second identification information, the decoding apparatus 102 determines any one of an intra skip mode, an inter skip mode, or an interview skip mode, and then considers the corresponding skip mode. The current block can be decrypted.

On the other hand, when the current block is neither the view synthesis skip mode nor the normal skip mode according to the first identification information and the second identification information, the decoding apparatus 102 uses the residual signal for the synthesized image of the virtual view. The block can be decrypted.

4 is a diagram illustrating a structure of a multiview video according to an embodiment.

Referring to FIG. 4, when a video of three viewpoints (Left, Center, Right) is received, a multiview video coding method of encoding GOP (Group of Picture) '8' is shown. In order to encode a multi-view image, a hierarchical B picture is basically applied to a temporal axis and a view axis, thereby reducing redundancy between images.

According to the structure of a multiview video illustrated in FIG. 4, the multiview video encoding apparatus 101 first encodes a left picture (I-view), and then a right picture (P-view) and a center picture (Center). A picture corresponding to three viewpoints can be encoded by sequentially encoding Picture: B-view.

In this case, the left image may be encoded in such a manner that temporal redundancy is removed by searching for similar regions from previous images through motion estimation. In addition, since the right image is encoded by using the previously encoded left image as a reference image, the right image may be encoded in such a manner that temporal redundancy based on motion estimation and view redundancy based on disparity estimation are removed. have. In addition, since the center image is encoded by using both the left image and the right image, which are already encoded, as a reference image, the inter-view redundancy may be removed according to the estimation of the shift in both directions.

Referring to FIG. 4, in a multi-view video encoding method, an image encoded without using a reference image of another view, such as a left image, may be encoded by predicting and encoding a reference image of another view in one direction, such as an I-View and a right image. An image that is predicted and encoded in both directions, such as a P-View and a center image, is defined as a B-View.

Frames of MVC are largely classified into six groups according to the prediction structure. Specifically, the six groups include an I-view anchor frame for intra coding, an I-view non-anchor frame for inter-time inter-coding, a P-view anchor frame for inter-view unidirectional inter coding, and a unidirectional inter-coding between views. Classified into P-view non-anchor frame for bi-directional inter-coding between time bases, B-view anchor frame for bi-directional inter-coding between views, and B-view non-anchor frame for bi-directional inter coding between time-bases. Can be.

According to an embodiment, the encoding apparatus 101 generates a composite image of a virtual view by synthesizing a first image of a neighboring view, which is a left and right view of a current view, to be encoded, and using the synthesized image, a second image of the current view. Can be encoded. Here, the first image of the neighboring view required for synthesis refers to an image that is already encoded.

In detail, the encoding apparatus 101 may encode the P-View by synthesizing the already encoded I-View. Alternatively, the encoding apparatus 101 may synthesize a previously encoded I-View and a P-View to encode a B-View. As a result, the encoding apparatus 101 may encode a specific image by synthesizing the already encoded image located in the vicinity.

Like the left image, an image that is compressed without prediction from another viewpoint is defined as an I-View. Like the image on the right, an image that is compressed while predicting in one direction from an image from another viewpoint is defined as a P-View. And, the image that is compressed while bidirectionally predicting from the images of left and right view like the center image is defined as B-View.

FIG. 5 is a diagram illustrating an example of a reference picture used to encode a current block according to an embodiment.

When encoding the current block located in the current frame, which is the current picture 501, the encoding apparatus 101 may determine reference pictures 502 and 503 located in the temporal vicinity of the current frame and reference pictures 504 and 505 located in the periphery in view. It is available.

In detail, the encoding apparatus 101 may search the prediction blocks most similar to the current blocks in the reference pictures 502 ˜ 505 to encode a residual signal between the current block and the prediction block.

The encoding apparatus 101 may use reference pictures 502 and 503 that are different in time from the current frame including the current block to search for the prediction block based on the motion vector. In order to search for the prediction block based on the disparity vector, reference pictures 504 and 505 different from the current frame including the current block may be used.

6 illustrates a view synthesis skip mode and a view synthesis direct mode according to an embodiment.

View synthesis prediction is a technique of predicting a block most similar to a current block by using a synthesized image of a virtual view generated based on images of a neighboring view that is already compressed. Therefore, in order to perform view synthesis prediction, a synthesized image is required using Equations 1 to 4 described below. The synthesized image corresponds to a reference picture for view synthesis prediction.

According to the view synthesis prediction, the encoding apparatus 101 may use a zero vector to predict the current block in the synthesized image. The encoding apparatus 101 may divide a coding mode into a skip mode, a direct mode, and a residual signal compression mode (16x16, 16x8, 8x6, and P8x8 modes) according to whether the block information of the current block is compressed.

Referring to FIG. 7, a view synthesis skip mode and a view synthesis direct mode are illustrated. When performing the view synthesis prediction, the encoding apparatus 101 may generate a synthesized image corresponding to the current view that is the virtual view from the reference image of the neighboring view that is already compressed.

The view synthesis skip mode searches for a block pointing to a zero vector in the synthesized image and replaces the block with a zero vector. In this case, the zero vector indicates the same position as the current block in the composite image. The compression of the difference between the prediction block indicated by the zero vector and the current block in the synthesized image may be defined as a view synthesis direct mode.

That is, the encoding apparatus 101 does not encode not only the motion information but also the residual signal that is the difference between the current block and the prediction block, for the current block set to the view synthesis skip mode. However, the encoding apparatus 101 encodes and transmits the residual signal without encoding only motion information on the current block set in the view synthesis direct mode.

7 illustrates a residual signal coding mode of view synthesis according to an embodiment.

In the case of compressing a block according to the residual signal coding mode, the encoding apparatus 101 may perform the synthesis on the synthesized image synthesized from the reference images of the neighboring viewpoints that have been compressed to the current view using Equations 1 to 4 to be described below. The prediction blocks most similar to the current block are found among the search areas set around the prediction blocks indicated by the zero vector. In this case, the zero vector indicates the same position as the current block in the synthesized image.

When searching for a prediction block, the search area uses various block sizes, such as 16x16, 16x8, 8x6, and P8x8 modes. Finally, the difference between the prediction block and the current block may be compressed together with a vector indicating the prediction block and transmitted to the decoding apparatus 102.

According to an embodiment, the encoding apparatus 101 may add and use only a view synthesis skip mode using a synthesized image of a virtual view with respect to an existing coding mode. The encoding apparatus 101 may add and use only the view synthesis direct mode using the synthesized image of the virtual view with respect to the existing coding mode. The encoding apparatus 101 may add and use only the residual signal coding mode using the synthesized image of the virtual view with respect to the existing coding mode. Here, the existing coding mode is a residual signal coding based on intra prediction, inter prediction, intra skip mode, intra direct mode, inter skip mode, inter direct mode, interview skip mode, interview direct mode, intra prediction, inter prediction, or interview prediction. It may include at least one mode.

Alternatively, the encoding apparatus 101 may add and use at least one of a view synthesis skip mode using a synthetic image of a virtual view, a view synthesis direct mode, or a residual signal coding mode of view synthesis with respect to an existing coding mode.

In fact, the encoding apparatus 101 exhibits the best coding performance for at least one of a view synthesis skip mode, a view synthesis direct mode, or a residual signal coding mode of view synthesis using a synthetic image of a virtual view in addition to the existing coding mode. The coding mode of the current block can be selected.

For example, the encoding apparatus 101 may perform coding according to all the above-described prediction techniques for the current block, and then select the final coding mode based on the coding result. In this case, the encoding apparatus 101 may use a rate-distortion cost value. Specifically, the rate-distortion cost function value is determined based on the calculated bit generation amount and the distortion value, and the final coding mode for the current block may be selected as the coding mode having the smallest rate-distortion cost function value. The rate-distortion cost function value is RD Cost =

Can be determined by.

Here, J is a rate-distortion cost function derived according to a prediction technique, and a sum of square difference (SSD) is a squared difference value between the current block s and the prediction block r.

Is the Lagrangian coefficient. R is the number of bits required when encoding a signal obtained by the difference between the predicted image from the original image and the previous image based on the coding mode of the prediction technique.

8 illustrates a view synthesis according to an embodiment.

The synthesized image of the virtual view for the color image and the depth image may be generated using the already encoded color image, the depth image, and camera parameter information. In detail, the synthesized image of the virtual view for the color image and the depth image may be generated according to Equations 1 to 4 below.

D means a pixel value of the pixel position (x, y) in the depth image. Znear and Zfar represent the nearest depth information and the farthest depth information, respectively.

The encoding apparatus 101 obtains the actual depth information Z and then combines the pixels (xr, yr) of the reference viewpoint image as shown in [Equation 2] to synthesize (r) the image of the reference viewpoint into the image of the target viewpoint. You can project in a 3D world coordinate system (u, v, w). In this case, the pixels (xr, yr) represent pixels of the color image when the virtual view synthesis is performed on the color image, and pixels of the depth image when the virtual view synthesis is performed on the depth image.

In Equation 2, A denotes an intrinsic camera matrix, R denotes a camera rotation matrix, T denotes a camera translation vector, and Z denotes depth information.

 Then, the encoding apparatus 101 projects the 3D World coordinate system (u, v, w) into the image coordinate system (xt zt, yt zt, zt) at the target viewpoint. This is done according to equation (3).

Finally, the corresponding pixel in the target viewpoint image becomes (x _t , y _t ).

At this time, depth information (Z _near / Z _far ) and camera parameter information (R / A / T) are additionally required to make a composite image of the virtual view. Therefore, this additional information is encoded in the encoding apparatus, included in the bitstream, and then decoded in the decoding apparatus. For example, the encoding apparatus may selectively determine the transmission method of the depth information and the camera parameter information according to whether the depth information and the camera parameter information are the same for each image to be encoded using the composite image of the virtual view.

Specifically, if additional information such as depth information and camera parameter information is the same for each image to be encoded using the synthesized image of the virtual view, the encoding apparatus may decode the additional information necessary for synthesizing the virtual view only once through the bitstream. Can send to Alternatively, if additional information, such as depth information and camera parameter information, is the same for each image to be encoded using the synthesized video of the virtual view, the encoding apparatus may display additional information necessary for synthesizing the virtual view through a bitstream of a group of picture. ) Can be sent to the decryption device for each.

If the additional information has a different value for each image to be encoded using the composite image of the virtual view, the encoding apparatus decodes the additional information for each image to be encoded using the synthesized image of the virtual view through a bitstream. To the device. Alternatively, if the additional information has a different value for each image to be encoded using the composite image of the virtual view, the encoding apparatus uses only the additional information having a different value for each image to be encoded using the composite image of the virtual view. It can be transmitted to the decoding apparatus through the stream.

According to another embodiment, if the composite image of the virtual view for the color image and the depth image taken by the (1D Parallel arrangement) in the horizontally arranged cameras may be generated according to the following equation (4).

f _x is the horizontal focal length of the camera, t _{x is the} x translation of the camera, and p _x is the horizontal principal point of the camera. d (Disparity) tells us the distance the pixel is shifted horizontally.

Finally, the pixels (xr, yr) in the reference image are mapped to pixels (x _t , y _t ) by d in the image at the target viewpoint.

At this time, depth information (Z _near / Z _far ) and camera parameter information (f _x , t _x , p _x ) are additionally required to create a virtual view image. Therefore, this additional information is encoded in the encoding apparatus, included in the bitstream, and then decoded in the decoding apparatus.

For example, the encoding apparatus may selectively determine the transmission method of the depth information and the camera parameter information according to whether the depth information and the camera parameter information are the same for each image to be encoded using the composite image of the virtual view.

Specifically, if additional information such as depth information and camera parameter information is the same for each image to be encoded using the synthesized image of the virtual view, the encoding apparatus may decode the additional information necessary for synthesizing the virtual view only once through the bitstream. Can send to

Alternatively, if additional information, such as depth information and camera parameter information, is the same for each image to be encoded using the synthesized video of the virtual view, the encoding apparatus may display additional information necessary for synthesizing the virtual view through a bitstream of a group of picture. ) Can be sent to the decryption device for each.

If the additional information has a different value for each image to be encoded using the composite image of the virtual view, the encoding apparatus decodes the additional information for each image to be encoded using the synthesized image of the virtual view through a bitstream. To the device.

Alternatively, if the additional information has a different value for each image to be encoded using the composite image of the virtual view, the encoding apparatus uses only the additional information having a different value for each image to be encoded using the composite image of the virtual view. It can be transmitted to the decoding apparatus through the stream.

In another embodiment, pixel values belonging to an image of a target view may be derived from pixels (x, y) belonging to an image of a reference view using corresponding depth images. In this case, the pixels of the corresponding depth image may be converted into a disparity vector using camera parameter information. The converted disparity vector may indicate which pixel among pixels belonging to an image of a reference view corresponds to a pixel.

 If the corresponding pixel (x, y) in the image at the reference viewpoint is not an integer position (Integer-Pel), the encoding apparatus 101 uses the pixel value of the integer position existing around the pixel (x, y). The pixel values of the pixels (x, y) may be interpolated. However, if the pixel values of the pixels corresponding to the integer positions are the same, the encoding apparatus 101 may use the pixel values of the pixels corresponding to the integer positions without interpolation.

As another example, when the difference between the pixel values of the pixel corresponding to the integer position is smaller than the preset difference, the encoding apparatus 101 does not interpolate and the pixel of the pixel corresponding to the integer position positioned around the pixel (x, y) You can use one of the values.

 If the corresponding pixel (x, y) in the image of the reference view indicates the half position (Half-Pel) instead of the integer position (Integer-Pel), the encoding apparatus 101 corresponds to the half position. A pixel value corresponding to an integer position on the left side of the pixel and a pixel value corresponding to an integer position on the right side may be compared. As a result of the comparison, if the pixel value corresponding to the integer position on the left and the pixel value corresponding to the integer position on the right are the same, the encoding apparatus 101 corresponds to the integer value on the left of the pixel value corresponding to 1/2 position. The pixel value is assigned to either the pixel value or the pixel value corresponding to the right integer position.

As a result of the comparison, if the pixel value corresponding to the left integer position and the pixel value corresponding to the right integer position are different from each other, the encoding device 101 converts the pixel value corresponding to the 1/2 position to the left integer position. After interpolating the pixel value and the pixel value corresponding to the integer position on the right side, the interpolation value may be assigned.

Also, if the corresponding pixel (x, y) in the image of the reference view indicates the quarter position (Quarter-Pel) rather than the integer position (Integer-Pel), the encoding apparatus 101 is located at the 1/4 position. A pixel value corresponding to an integer position on the left side and a pixel value corresponding to an integer position on the right side may be compared with the corresponding pixel.

As a result of the comparison, if the pixel value corresponding to the integer position on the left and the pixel value corresponding to the integer position on the right are the same, the encoding apparatus 101 matches the pixel value corresponding to the quarter position to the integer position on the left. The pixel value is assigned to either the pixel value or the pixel value corresponding to the right integer position.

As a result of the comparison, if the pixel value corresponding to the left integer position and the pixel value corresponding to the right integer position are different from each other, the pixel value corresponding to the 1/2 position and the pixel value corresponding to the left integer position and the right integer An interpolation value may be assigned after interpolating a pixel value corresponding to a position.

If the pixel value corresponding to the 1/4 position is similar to the pixel value corresponding to the left integer position, the left integer position value of the 1/4 position and the interpolation value of the 1/2 position may be compared. As a result of the comparison, when the left integer position value at the 1/4 position and the interpolation value at the 1/2 position are the same as each other, the encoding apparatus 101 may allocate the left integer position value or the interpolation value at the 1/2 position. If, as a result of the comparison, the left integer position value at the 1/4 position and the interpolation value at the 1/2 position are different from each other, the encoding apparatus 101 interpolates the left integer position value and the interpolation value at the 1/2 position again to 1 Used as an interpolation value at the / 4 position.

If the pixel value corresponding to the 1/4 position is close to the right integer position, the encoding apparatus 101 may compare the right integer position value of the 1/4 position with the interpolation value of the 1/2 position. As a result of the comparison, when the right integer position value at the 1/4 position and the interpolation value at the 1/2 position are the same as each other, the encoding apparatus 101 may allocate the right integer position value or the interpolation value at the 1/2 position. When the right integer position value at the 1/4 position and the interpolation value at the 1/2 position are different from each other, the encoding device 101 interpolates the right integer position value and the interpolation value at the 1/2 position again and the 1/4 position. Used as an interpolation value of.

9 is a diagram for describing an encoding process using a synthetic image of a virtual view, according to an exemplary embodiment.

The color image and the depth image constituting the 3D video may be separately encoded and decoded. Referring to FIG. 9, the encoding process is performed by obtaining a residual signal between an original image and a prediction image derived through block-based prediction, and then transforming and quantizing the residual signal. Then, a deblocking filter is performed to accurately predict the next images.

Since the smaller the residual signal, the fewer bits are required for encoding, how important the predicted image is to the original image is very important. According to an embodiment, for block prediction, a view synthesis as well as a skip mode, a direct mode and a residual signal coding mode based on intra prediction, inter prediction, and inter-view prediction Skip mode, view synthesis direct mode, and residual signal coding mode of view synthesis may be used.

Referring to FIG. 9, an additional configuration for synthesizing at the virtual view may be required to generate a synthesized image of the virtual view. Referring to FIG. 9, in order to generate a composite image of a color image of the current view, the encoding apparatus 101 may generate a synthesized image of the color image of the current view by using the color image and the depth image of the neighboring view that are already encoded. Can be generated. In order to generate the composite image of the depth image of the current view, the encoding apparatus 101 may generate the composite image of the depth image of the current view using the depth image of the neighboring view that is already encoded.

10 is a diagram for describing a decoding process using a synthetic image of a virtual view, according to an exemplary embodiment.

Since the decoding apparatus 102 of FIG. 10 performs substantially the same operation as the encoding apparatus 101 of FIG. 9, a detailed description thereof will be omitted.

11 illustrates a bitstream structure for a macroblock according to an embodiment. 11 illustrates the position of a flag that identifies the view synthesis skip mode.

FIG. 11 illustrates a proposed macroblock level Syntax for distinguishing flags of a general normal skip mode and a view synthesis view synthesis skip mode. The normal skip mode means a skip mode other than the view synthesis skip mode, and may include, for example, an intra skip mode, an inter skip mode, or an interview skip mode.

Hereinafter, the normal skip mode and the view synthesis view synthesis skip mode may correspond to First Skip Flag (first_skip_flag) and Second Skip Flag (second_skip_flag) for distinguishing flag positions. Here, first_skip_flag may appear earlier in the bitstream than second_skip_flag. Hereinafter, mb_skip_flag may mean a flag indicating whether the current block is a normal skip mode instead of a view synthesis skip mode, and mb_vsskip_flag may mean a flag indicating whether the current block is a view synthesis skip mode.

<CASE 1>

For example, CASE 1 refers to a case where first_skip_flag is mb_skip_flag and second_skip_flag is mb_vsskip_flag. Then, mb_vsskip_flag is located after mb_skip_flag.

If the coding mode of the current block is the normal skip mode, the encoding apparatus 101 may allocate mb_skip_flag to 1 and transmit the mb_skip_flag to the decoding apparatus 102. For example, the encoding apparatus 101 may send only mb_skip_flag = 1 to the decoding apparatus 102 and may not transmit other encoding information. If the coding mode of the current block is the view synthesis skip mode, the encoding apparatus 101 may allocate mb_skip_flag to 0 and allocate mb_vsskip_flag to 1 and transmit the same to the decoding apparatus 102. For example, the encoding apparatus 101 may transmit only mb_skip_flag = 0 and mb_vsskip_flag = 1 to the decoding apparatus 102, and may not transmit other encoding information.

If the coding mode of the current block is not the normal skip mode and the view synthesis skip mode, the encoding apparatus 101 means that the coding mode of compressing the residual signal, which is a difference between the current block and the prediction block, is determined as the coding mode of the current block. do. Then, the encoding apparatus 101 may transmit the mb_skip_flag and the mb_vsskip_flag to 0, and then transmit the bitstream indicating the type of coding mode, the bitstream about the motion information and the residual signal related to the current block, to the decoding apparatus 102.

<CASE 2>

As another example, CASE 2 refers to a case where first_skip_flag is mb_vsskip_flag and second_skip_flag is mb_skip_flag. Mb_skip_flag is then located after mb_vsskip_flag.

If the coding mode of the current block is the view synthesis skip mode, the encoding apparatus 101 may allocate mb_vsskip_flag to 1. For example, the encoding apparatus 101 may send only mb_vsskip_flag = 1 to the decoding apparatus 102 and may not transmit other encoding information. If the coding mode of the current block is the normal skip mode, the encoding apparatus 101 may allocate mb_vsskip_flag to 0 and assign mb_skip_flag to 1. For example, the encoding apparatus 101 may send only the mb_vsskip_flag = 0 and mb_skip_flag = 1 to the decoding apparatus 102 and may not send other encoding information.

If the mode of the current block is not the normal skip mode and the view synthesis skip mode, since the current block is the residual signal compression mode, the encoding apparatus 101 allocates mb_vsskip_flag and mb_skip_flag to 0 and then selects a bit stream for the type of coding mode. The bit stream of the motion information and the residual signal of the current block is transmitted to the decoding apparatus 102.

Meanwhile, the encoding apparatus 101 may adaptively set (Context-based) first_skip_flag and second_skip_flag by using compression information of neighboring blocks adjacent to the current block (Context-based Adaptive Skip Flag Positioning (CASFP)). Here, the compression information may include coding mode information, motion information, code block pattern (CBP) information, texture information, etc. of neighboring blocks.

The encoding apparatus 101 may set first_skip_flag to mb_vsskip_flag and second_skip_flag to mb_skip_flag based on the compression information of the neighboring block.

For example, when both the upper block and the left block of the current block are in the view synthesis skip mode, the encoding apparatus 101 may set first_skip_flag to mb_vsskip_flag and set second_skip_flag to mb_skip_flag. In other words, when both the above block and the left block of the current block are in view synthesis skip mode, mb_vsskip_flag may be located before mb_skip_flag.

If both the upper block and the left block of the current block are not in the view synthesis skip mode, the encoding apparatus 101 may set first_skip_flag to mb_skip_flag and set second_skip_flag to mb_vsskip_flag. In other words, when only one of the above block and the left block of the current block is in the view synthesis skip mode, or both the above block and the left block of the current block are in the normal skip mode. The encoding apparatus 101 may set first_skip_flag to mb_skip_flag and second_skip_flag to mb_vsskip_flag.

As another example, when one of the upper block and the left block of the current block is the view synthesis skip mode, the encoding apparatus 101 may set first_skip_flag to mb_vsskip_flag and set second_skip_flag to mb_skip_flag. If the upper block and the left block of the current block are not all of the view synthesis skip mode, that is, if the upper and left blocks of the current block are all of the normal skip mode, the encoding apparatus 101 sets first_skip_flag to mb_skip_flag, second_skip_flag may be set to mb_vsskip_flag.

The encoding apparatus 101 may set first_skip_flag to mb_skip_flag and second_skip_flag to mb_vsskip_flag based on the compression information of the neighboring block.

For example, when both the upper block and the left block of the current block are in the normal skip mode, the encoding apparatus 101 may set first_skip_flag to mb_skip_flag and set second_skip_flag to mb_vsskip_flag. In other words, when both the upper block and the left block of the current block are in the normal skip mode, mb_skip_flag may be located before mb_vsskip_flag.

If both the upper block and the left block of the current block are not in the normal skip mode, the encoding apparatus 101 may set first_skip_flag to mb_vsskip_flag and set second_skip_flag to mb_skip_flag. In other words, when only one of the above block and the left block of the current block is in the normal skip mode, or both the above block and the left block of the current block are in view synthesis skip mode. The encoding apparatus 101 may set first_skip_flag to mb_vsskip_flag and set second_skip_flag to mb_skip_flag.

As another example, when one of the upper block and the left block of the current block is the normal skip mode, the encoding apparatus 101 may set first_skip_flag to mb_skip_flag and second_skip_flag to mb_vsskip_flag. When the upper block and the left block of the current block are not all in the normal skip mode, that is, when the upper block and the left block of the current block are all in the view synthesis skip mode, the encoding apparatus 101 sets first_skip_flag to mb_vsskip_flag, second_skip_flag may be set to mb_skip_flag.

Although the description of FIG. 11 illustrates a case in which the view synthesis skip mode is added to the normal skip mode, the same description may also be applied to the case where the general skip mode includes another skip mode of a new type other than the view synthesis skip mode.

Meanwhile, even when the general skip mode is an intra skip mode and an inter skip mode, and the interview skip mode other than the view synthesis skip mode is added as a new type of skip mode, the description of FIG. 11 may be applied. For example, when both the left block and the upper block of the current block are in the interview skip mode, first_skip_flag may be set as a flag indicating an interview skip mode, and second_skip_flag may be set as a flag indicating an intra skip mode or an inter skip mode.

In other words, it is assumed that various types of general skip modes exist and a new skip mode is defined according to the development of an encoding scheme. The position between the identification information indicating the normal skip mode and the identification information indicating the new skip mode may be determined according to the encoding mode of the adjacent neighboring block of the current block. In detail, when the encoding modes of the neighboring blocks are all new skip modes, the identification information indicating the new skip mode may be set before the identification information indicating the normal skip mode for the current block.

12 illustrates a bitstream structure for a macroblock according to another embodiment.

12 illustrates the position of a flag identifying the viewpoint synthesis direct mode. The direct mode, like the skip mode, does not include motion information but refers to a coding mode for encoding the residual signal. Therefore, when the current block is in the direct mode, the encoding apparatus 101 may use a flag to distinguish between the general direct mode and the view synthesis direct mode.

direct_mode_type_flag may be located after the mode type (mb_type) in the bitstream. If the direct_mode_type_flag transmitted to the decoding device 102 is 0, it means a general direct mode. If the direct_mode_type_flag is 1, it means a view synthesis direct mode. In some embodiments, 1 and 0 may be interchanged.

In FIG. 12, first_skip_flag may be mb_skip_flag, and second_skip_flag may be mb_vsskip_flag. Alternatively, first_skip_flag may be mb_vsskip_flag, and second_skip_flag may be mb_skip_flag.

For example, when the current block is in the view synthesis direct mode, the encoding apparatus 101 may set first_skip_flag to 0 and set second_skip_flag to 0. FIG. In addition, the encoding apparatus 101 may allocate a bit indicating that the current block is a direct mode, set direct_mode_type_flag to 1, and transmit the bit to the decoding apparatus 102 together with the compressed bit.

For example, when the current block is in the normal direct mode, the encoding apparatus 101 may set first_skip_flag to 0 and second_skip_flag to 0. FIG. In addition, the encoding apparatus 101 may allocate a bit indicating that the current block is the direct mode, set direct_mode_type_flag to 0, and transmit the bit to the decoding apparatus 102 together with the compressed bit.

For example, when the coding mode of the current block is the residual signal compression mode other than the skip mode or the direct mode, the encoding apparatus 101 may set first_skip_flag to 0 and second_skip_flag to 0. FIG. Thereafter, the encoding apparatus 101 may transmit a bit indicating the residual signal compression mode and a bit obtained by compressing the residual signal to the decoding apparatus 102.

According to an embodiment, a method and apparatus for encoding and decoding 3D video using view synthesis is provided.

 Here, a method and apparatus for synthesizing an image from a viewpoint of an image to be currently compressed using images of neighboring viewpoints already compressed for view synthesis are provided.

 Here, a method and apparatus for using a view synthesis skip mode (View Synthesis SKIP mode) that does not encode and decode block information in view synthesis; A method and apparatus for using a view synthesis direct mode for encoding and decoding a residual signal except motion information among block information in view synthesis; A method and apparatus for using a residual signal compression mode of view synthesis for encoding and decoding block information in view synthesis are provided.

 Here, in the view synthesis skip mode, a method and apparatus for using a block indicated by a zero vector to find a block most similar to a current block in a synthesized image are provided.

 Here, a method and apparatus for placing a flag indicating a view synthesis skip mode (mb_vsskip_flag) before or after a flag (mb_skip_flag) for an existing skip mode are provided.

 Here, (Context-based Adaptive Skip Flag) adaptively located before or after Flag (mb_vsskip_flag) indicating a general skip mode based on compression information of adjacent neighboring blocks for the current block. Positioning: CASFP) method and apparatus are provided.

Here, in view synthesis direct mode, a method and apparatus for using a block indicated by a zero vector to find a block most similar to a current block in a synthesized image are provided.

 Here, a method and apparatus for placing a Flag for distinguishing a direct mode based on an existing prediction from a view synthesis direct mode after a syntax (mb_type) indicating a type of a (direct_mode_type_flag) mode is provided.

 Here, when finding the pixels of the composite image from the image pixels of the reference view, a method and apparatus for converting the pixels of the corresponding depth image into a disparity vector to find the corresponding pixels of the reference view indicated by the disparity vector.

Provided are a method and an apparatus for determining whether to interpolate a corresponding pixel using neighboring pixels when the position of the corresponding pixel at the reference point of view is not an integer position (Interger-Pel).

The following is an embodiment of Syntax based on the Context-based Adaptive Skip Flag Positioning (CASFP) method. In an embodiment, mb_skip_flag represents a normal skip mode and mb_vsskip_flag represents a view synthesis skip mode.

For example, when encoding mb_vsskip_flag of the current block, the encoding apparatus 101 performs entropy encoding with reference to mb_vsskip_flag of the upper block and the left block of the current block.

Slice data syntax may be represented as shown in Table 1 below.

When mb_type is derived to P_Skip using a composite picture as a reference picture when decoding the P slice, and mb_vsskip_flag represents 1 for the current macroblock when the macroblock type is collectively referred to as the P macroblock type. Or, if mb_type is derived as B_Skip using a composite picture as a reference picture when decoding a B slice, and if the macroblock type is collectively referred to as the B macroblock type, then mb_vsskip_flag equals 1 for the current macroblock. Indicates. If the current macroblock is not skipped, mb_vsskip_flag indicates 0. And, if mb_vsskip_flag does not exist, mb_vsskip_flag may be inferred to zero.

The method according to the embodiment may be embodied in the form of program instructions that can be executed by various computer means and recorded in a computer readable medium. The computer readable medium may include program instructions, data files, data structures, etc. alone or in combination. The program instructions recorded on the media may be those specially designed and constructed for the purposes of the embodiments, or they may be of the kind well-known and available to those having skill in the computer software arts. Examples of computer readable recording media include magnetic media such as hard disks, floppy disks and magnetic tape, optical media such as CD-ROMs, DVDs, and magnetic disks such as floppy disks. Magneto-optical media, and hardware devices specifically configured to store and execute program instructions, such as ROM, RAM, flash memory, and the like. Examples of program instructions include not only machine code generated by a compiler, but also high-level language code that can be executed by a computer using an interpreter or the like. The hardware device described above may be configured to operate as one or more software modules to perform the operations of the embodiments, and vice versa.

Although the embodiments have been described by the limited embodiments and the drawings as described above, various modifications and variations are possible to those skilled in the art from the above description. For example, the described techniques may be performed in a different order than the described method, and / or components of the described systems, structures, devices, circuits, etc. may be combined or combined in a different form than the described method, or other components. Or even if replaced or substituted by equivalents, an appropriate result can be achieved.

Therefore, other implementations, other embodiments, and equivalents to the claims are within the scope of the claims that follow.

Claims (29)

1. Extracting first identification information indicating whether the current block is in a view synthesis skip mode;

   Extracting second identification information indicating whether the current block is in a normal skip mode when the current block is not in the view synthesis skip mode according to the first identification information; And,

   Decoding the current block using the first identification information or the second identification information

   Decryption method comprising a.
2. The method of claim 1,

   The first identification information,

   If both the left block and the upper block of the current block are in the view synthesis skip mode, the decoding method is located before the second identification information.
3. The method of claim 1,

   The viewpoint synthesis skip mode,

   By using a composite image of a virtual view that is the same view as the current view of the current block,

   The composite image,

   A decoding method determined from coded texture information and depth information at a peripheral view of the current view.
4. The method of claim 1,

   The normal skip mode,

   A decoding method comprising at least one of an intra skip mode, an inter skip mode, or an interview skip mode.
5. Extracting first identification information indicating whether the current block is in a normal skip mode;

   Extracting second identification information indicating whether the current block is in view synthesis skip mode when the current block is not in the normal skip mode according to the first identification information; And,

   Decoding the current block using the first identification information or the second identification information

   Decryption method comprising a.
6. The method of claim 5,

   The first identification information,

   And at least one of the left block and the upper block of the current block is located before the second identification information.
7. The method of claim 5,

   The viewpoint synthesis skip mode,

   By using a composite image of a virtual view that is the same view as the current view of the current block,

   The composite image,

   A decoding method determined from texture information and depth information encoded at a peripheral view of the current view.
8. The method of claim 5,

   The normal skip mode,

   A decoding method comprising at least one of an intra skip mode, an inter skip mode, or an interview skip mode.
9. Extracting identification information indicating the encoding mode of the current block from the bitstream;

   Decoding the current block based on the identification information

   Including,

   The identification information,

   And first identification information indicating whether the current block is in view synthesis skip mode and second identification information indicating whether the current block is in normal skip mode.
10. The method of claim 9,

    The order between the first identification information and the second identification information,

    Depends on the encoding mode of the neighboring block of the current block,

    The neighboring block includes an upper block and a left block of the current block.
11. The method of claim 9,

    The first identification information,

    If both the upper block and the left block of the current block are located in the view synthesis skip mode, they are positioned before the second identification information;
12. Determining an encoding mode of a neighboring block adjacent to the current block;

    Adaptively placing identification information associated with a skip mode of a current block according to the encoding mode; And

    Encoding the identification information and the current block.

    Including,

    The identification information,

    And first identification information indicating whether the current block is in view synthesis skip mode and second identification information indicating whether the current block is in normal skip mode.
13. Determining whether the encoding modes of the upper block and the left block of the current block are both view synthesis skip modes;

    Determining a location of first identification information indicating whether the current block is in a view synthesis skip mode and second identification information indicating whether the current block is in a normal skip mode according to a determination result;

    Encoding the first identification information, the second identification information, and the current block.

    Encoding method comprising a.
14. Extracting first identification information indicating whether the current block is a view synthesis skip mode, and if the current block is not the view synthesis skip mode according to the first identification information, a second indicating whether the current block is a normal skip mode An identification information extraction unit for extracting identification information; And

    A decoder that decodes the current block by using the first identification information or the second identification information.

    Decoding apparatus comprising a.
15. The method of claim 14,

    The first identification information,

    And the second block is located before the second identification information when both the left block and the upper block of the current block are in the view synthesis skip mode.
16. The method of claim 14,

    The viewpoint synthesis skip mode,

    By using a composite image of a virtual view that is the same view as the current view of the current block,

    The composite image,

    A decoding apparatus that is determined from encoded texture information and depth information at a peripheral view of the current view.
17. The method of claim 14,

    The normal skip mode,

    And at least one of an intra skip mode, an inter skip mode, or an interview skip mode.
18. Extracting first identification information indicating whether the current block is in a normal skip mode, and if the current block is not in the normal skip mode according to the first identification information, a second identification indicating whether the current block is in a view synthesis skip mode An identification information extraction unit for extracting information; And

    A decoder that decodes the current block by using the first identification information or the second identification information.

    Decoding apparatus comprising a.
19. The method of claim 18,

    The first identification information,

    And at least one of the left block and the upper block of the current block is located before the second identification information.
20. The method of claim 18,

    The viewpoint synthesis skip mode,

    By using a composite image of a virtual view that is the same view as the current view of the current block,

    The composite image,

    A decoding apparatus that is determined from texture information and depth information encoded at a peripheral view of the current view.
21. The method of claim 18,

    The normal skip mode,

    And at least one of an intra skip mode, an inter skip mode, or an interview skip mode.
22. An identification information extracting unit which extracts identification information indicating an encoding mode of a current block from the bitstream;

    Decoding unit for decoding the current block based on the identification information

    Including,

    The identification information,

    And first identification information indicating whether the current block is in a view synthesis skip mode and second identification information indicating whether the current block is in a normal skip mode.
23. The method of claim 22,

    The order between the first identification information and the second identification information,

    Depends on the encoding mode of the neighboring block of the current block,

    The neighboring block includes an upper block and a left block of the current block.
24. The method of claim 22,

    The first identification information,

    If both the upper block and the left block of the current block are in the view synthesis skip mode, the decoding apparatus is located before the second identification information;
25. An encoding mode determination unit that determines an encoding mode of a neighboring block adjacent to the current block;

    An identification information disposition unit for adaptively disposing identification information associated with a skip mode of a current block according to the encoding mode; And

    An encoding unit encoding the identification information and the current block

    Including,

    The identification information,

    And encoding first information indicating whether the current block is in the view synthesis skip mode and second identification information indicating whether the current block is in the normal skip mode.
26. An encoding mode determination unit that determines whether the encoding modes of the upper block and the left block of the current block are the view synthesis skip mode;

    An identification information arranging unit configured to determine a position of first identification information indicating whether the current block is in a view synthesis skip mode and second identification information indicating whether the current block is in a normal skip mode according to a determination result;

    An encoder that encodes the first identification information, the second identification information, and the current block.

    Encoding apparatus comprising a.
27. In a computer-readable recording medium that recorded a bitstream,

    The bitstream,

    A current coded current block, first identification information indicating whether the current block is a view synthesis skip mode, and second identification information indicating whether the current block is a normal skip mode,

    The first identification information,

    A recording medium located in the bitstream before the second identification information when both the upper block and the left block of the current block are in a view synthesis skip mode.
28. An identification information extraction unit for extracting first identification information and second identification information indicating an encoding mode of a current block from a bitstream;

    Decoding unit for decoding the current block based on the identification information

    Including,

    The first identification information and the second identification information,

    A decoding apparatus selected from a view synthesis skip mode, an interview skip mode, an inter skip mode, and an intra skip mode, and the position allocated to the bitstream varies according to an encoding mode of a neighboring block of the current block.
29. Extracting first identification information and second identification information indicating the encoding mode of the current block from the bitstream;

    Decoding the current block based on the identification information

    Including,

    The first identification information and the second identification information,

    A decoding method selected from a view synthesis skip mode, an interview skip mode, an inter skip mode, and an intra skip mode, wherein a position allocated to a bitstream varies according to an encoding mode of a neighboring block of a current block.

Images