WO2018128239A1

WO2018128239A1 - Method and device for decoding image according to block division structure in image coding system

Info

Publication number: WO2018128239A1
Application number: PCT/KR2017/008717
Authority: WO
Inventors: 남정학; 임재현; 장형문
Original assignee: 엘지전자 주식회사
Priority date: 2017-01-03
Filing date: 2017-08-11
Publication date: 2018-07-12

Abstract

An image decoding method performed by a decoding device according to the present invention comprises: a step of acquiring first division information of a first target block; a step of dividing the first target block into first sub-blocks if the first division information indicates that the first target block is to divided; a step of acquiring additional division information and second division information of a second target block which is one of the first sub-blocks of the first target block; a step of dividing the second target block into second sub-blocks on the basis of the additional division information if the second target block is not divided based on the second division information of the second target block; and a step of decoding the second sub-blocks, wherein the second sub-blocks are nonsquare blocks.

Description

Image Decoding Method and Apparatus According to Block Partition Structure in Image Coding System

The present invention relates to an image coding technique, and more particularly, to an image decoding method and apparatus according to a block division structure in an image coding system.

Recently, the demand for high resolution and high quality images such as high definition (HD) images and ultra high definition (UHD) images is increasing in various fields. The higher the resolution and the higher quality of the image data, the more information or bit rate is transmitted than the existing image data. Therefore, the image data can be transmitted by using a medium such as a conventional wired / wireless broadband line or by using a conventional storage medium. In the case of storage, the transmission cost and the storage cost are increased.

Accordingly, a high efficiency image compression technique is required to effectively transmit, store, and reproduce high resolution, high quality image information.

An object of the present invention is to provide a method and apparatus for improving image coding efficiency.

Another technical problem of the present invention is to provide an inter prediction method and apparatus for dividing a picture through various partitioning structures.

Another technical problem of the present invention is to provide a method and apparatus for dividing a picture into non-square blocks, and decoding based on each non-square block.

According to an embodiment of the present invention, there is provided an image decoding method performed by a decoding apparatus. The method may include obtaining first split information about a first target block, and when the first split information indicates that the first target block is split, the first target block may be assigned to the first subblocks. Dividing into; obtaining second splitting information and additional splitting information for a second target block, which is one of the first sub-blocks of the first target block, wherein the second target block is the second target block; Splitting the second target block into second subblocks based on the additional partitioning information, if not split based on the second partitioning information for, and decoding the second subblocks; The second subblocks may be non-square blocks.

According to another embodiment of the present invention, a decoding apparatus for performing image decoding is provided. The decoding apparatus obtains first split information about a first target block through a bitstream, second splitting information about a second target block which is one of first subblocks of the first target block, and An entropy decoding unit for acquiring additional splitting information, when the first splitting information for the first target block indicates that the first target block is to be divided, split the first target block into the first subblocks, If the second split information on the second target block indicates that the second target block is not split, the picture splitter divides the second target block into second subblocks based on the additional split information. And a prediction unit to decode the second sub blocks, wherein the second sub blocks are non-square blocks.

According to another embodiment of the present invention, a video encoding method performed by an encoding apparatus is provided. The method includes dividing a first target block into first subblocks, dividing a second target block that is one of the first subblocks into second subblocks, and decoding the second subblocks. And generating, encoding, and outputting first partitioning information for the first target block, second partitioning information for the second target block, and additional partitioning information, wherein the second subblocks are non-existent. It is characterized in that the square blocks.

According to another embodiment of the present invention, a video encoding apparatus is provided. The encoding apparatus divides a first target block into first subblocks, a picture divider that splits a second target block, which is one of the first subblocks, into second subblocks, and the second subblocks. A prediction unit to decode, and an entropy encoding unit to generate, encode, and output first split information of the first target block, second split information of the second target block, and additional split information, wherein the second split information is generated. The sub blocks are characterized in that they are non-square blocks.

According to the present invention, a picture may be divided into various types of blocks, and thus, prediction efficiency may be improved and overall coding efficiency may be improved.

According to the present invention, a picture may be divided into various types of blocks, and accordingly, transform efficiency may be improved and overall coding efficiency may be improved.

1 is a diagram schematically illustrating a configuration of a video encoding apparatus to which the present invention may be applied.

2 is a diagram schematically illustrating a configuration of a video decoding apparatus to which the present invention may be applied.

3 exemplarily illustrates a CU partitioned through a quad tree binary tree (QTBT) structure and a syntax of the QTBT structure.

4 shows an example in which syntaxes of the QTBT structure for a target CU are transmitted.

5 exemplarily shows a syntax of a CU partitioned through the QTGP structure and the QTGP structure.

6a to 6d exemplarily show a segmentation boundary derived based on information about a segmentation angle and / or a distance from a midpoint of a CU.

7 shows an example in which syntaxes of the QTGP structure for a target CU are transmitted.

8 exemplarily shows a syntax of a CU partitioned through the QTBTGP structure and the QTBTGP structure.

9 shows an example in which syntaxes of the QTBTGP structure for a target CU are transmitted.

10 exemplarily shows a syntax of a CU partitioned through the QTGPBT structure and the QTGPBT structure.

11 shows an example in which syntaxes of the QTGPBT structure for a target CU are transmitted.

12 exemplarily shows a syntax of a CU partitioned through the QTBTGP structure and the QTBTGP structure.

13 shows an example in which syntaxes of the QTBTGP structure for a target CU are transmitted.

14 exemplarily shows a syntax of a CU partitioned through the QTBTGP structure and the QTBTGP structure.

15 shows an example in which syntaxes of the QTBTGP structure for a target CU are transmitted.

16 schematically shows a video encoding method by an encoding device according to the present invention.

17 schematically illustrates a video decoding method by a decoding apparatus according to the present invention.

As the present invention allows for various changes and numerous embodiments, particular embodiments will be illustrated in the drawings and described in detail in the written description. However, this is not intended to limit the invention to the specific embodiments. The terminology used herein is for the purpose of describing particular example embodiments only and is not intended to be limiting of the spirit of the present invention. Singular expressions include plural expressions unless the context clearly indicates otherwise. The terms "comprise" or "having" in this specification are intended to indicate that there is a feature, number, step, operation, component, part, or combination thereof described in the specification, and one or more other features It is to be understood that the numbers, steps, operations, components, parts or figures do not exclude in advance the presence or possibility of adding them.

On the other hand, each configuration in the drawings described in the present invention are shown independently for the convenience of description of the different characteristic functions, it does not mean that each configuration is implemented by separate hardware or separate software. For example, two or more of each configuration may be combined to form one configuration, or one configuration may be divided into a plurality of configurations. Embodiments in which each configuration is integrated and / or separated are also included in the scope of the present invention without departing from the spirit of the present invention.

Hereinafter, with reference to the accompanying drawings, it will be described in detail a preferred embodiment of the present invention. Hereinafter, the same reference numerals are used for the same components in the drawings, and redundant description of the same components is omitted.

In the present specification, a picture generally refers to a unit representing one image of a specific time zone, and a slice is a unit constituting a part of a picture in coding. One picture may be composed of a plurality of slices, and if necessary, the picture and the slice may be mixed with each other.

A pixel or a pel may refer to a minimum unit constituting one picture (or image). Also, 'sample' may be used as a term corresponding to a pixel. A sample may generally represent a pixel or a value of a pixel, and may only represent pixel / pixel values of the luma component, or only pixel / pixel values of the chroma component.

A unit represents the basic unit of image processing. The unit may include at least one of a specific region of the picture and information related to the region. The unit may be used interchangeably with terms such as block or area in some cases. In a general case, an M × N block may represent a set of samples or transform coefficients composed of M columns and N rows.

Referring to FIG. 1, the video encoding apparatus 100 may include a picture divider 105, a predictor 110, a subtractor 115, a transformer 120, a quantizer 125, a reordering unit 130, The entropy encoding unit 135, the residual processing unit 140, the adding unit 150, the filter unit 155, and the memory 160 may be included. The residual processor 140 may include an inverse quantizer 141 and an inverse transform unit 142.

The picture divider 105 may divide the input picture into at least one processing unit.

As an example, the processing unit may be called a coding unit (CU). In this case, the coding unit may be recursively split from the largest coding unit (LCU) according to a quad-tree binary-tree (QTBT) structure. For example, one coding unit may be divided into a plurality of coding units of a deeper depth based on a quad tree structure and / or a binary tree structure. In this case, for example, the quad tree structure may be applied first and the binary tree structure may be applied later. Alternatively, the binary tree structure may be applied first. The coding procedure according to the present invention may be performed based on the final coding unit that is no longer split. In this case, the maximum coding unit may be used as the final coding unit immediately based on coding efficiency according to the image characteristic, or if necessary, the coding unit is recursively divided into coding units of lower depths and optimized. A coding unit of size may be used as the final coding unit. Here, the coding procedure may include a procedure of prediction, transform, and reconstruction, which will be described later.

As another example, the processing unit may include a coding unit (CU) prediction unit (PU) or a transform unit (TU). The coding unit may be split from the largest coding unit (LCU) into coding units of deeper depths along the quad tree structure. In this case, the maximum coding unit may be used as the final coding unit immediately based on coding efficiency according to the image characteristic, or if necessary, the coding unit is recursively divided into coding units of lower depths and optimized. A coding unit of size may be used as the final coding unit. If a smallest coding unit (SCU) is set, the coding unit may not be split into smaller coding units than the minimum coding unit. Here, the final coding unit refers to a coding unit that is the basis of partitioning or partitioning into a prediction unit or a transform unit. The prediction unit is a unit partitioning from the coding unit and may be a unit of sample prediction. In this case, the prediction unit may be divided into sub blocks. The transform unit may be divided along the quad tree structure from the coding unit, and may be a unit for deriving a transform coefficient and / or a unit for deriving a residual signal from the transform coefficient. Hereinafter, a coding unit may be called a coding block (CB), a prediction unit is a prediction block (PB), and a transform unit may be called a transform block (TB). A prediction block or prediction unit may mean a specific area in the form of a block within a picture, and may include an array of prediction samples. In addition, a transform block or a transform unit may mean a specific area in a block form within a picture, and may include an array of transform coefficients or residual samples.

The prediction unit 110 may perform a prediction on a block to be processed (hereinafter, referred to as a current block) and generate a predicted block including prediction samples of the current block. The unit of prediction performed by the prediction unit 110 may be a coding block, a transform block, or a prediction block.

The prediction unit 110 may determine whether intra prediction or inter prediction is applied to the current block. As an example, the prediction unit 110 may determine whether intra prediction or inter prediction is applied on a CU basis.

In the case of intra prediction, the prediction unit 110 may derive a prediction sample for the current block based on reference samples outside the current block in the picture to which the current block belongs (hereinafter, referred to as the current picture). In this case, the prediction unit 110 may (i) derive the prediction sample based on the average or interpolation of neighboring reference samples of the current block, and (ii) the neighbor reference of the current block. The prediction sample may be derived based on a reference sample present in a specific (prediction) direction with respect to the prediction sample among the samples. In case of (i), it may be called non-directional mode or non-angle mode, and in case of (ii), it may be called directional mode or angular mode. In intra prediction, the prediction mode may have, for example, 33 directional prediction modes and at least two non-directional modes. The non-directional mode may include a DC prediction mode and a planner mode (Planar mode). The prediction unit 110 may determine the prediction mode applied to the current block by using the prediction mode applied to the neighboring block.

In the case of inter prediction, the prediction unit 110 may derive the prediction sample for the current block based on the sample specified by the motion vector on the reference picture. The prediction unit 110 may apply one of a skip mode, a merge mode, and a motion vector prediction (MVP) mode to derive a prediction sample for the current block. In the skip mode and the merge mode, the prediction unit 110 may use the motion information of the neighboring block as the motion information of the current block. In the skip mode, unlike the merge mode, the difference (residual) between the prediction sample and the original sample is not transmitted. In the MVP mode, the motion vector of the current block may be derived using the motion vector of the neighboring block as a motion vector predictor.

In the case of inter prediction, the neighboring block may include a spatial neighboring block existing in the current picture and a temporal neighboring block present in the reference picture. A reference picture including the temporal neighboring block may be called a collocated picture (colPic). The motion information may include a motion vector and a reference picture index. Information such as prediction mode information and motion information may be encoded (entropy) and output in the form of a bitstream.

When the motion information of the temporal neighboring block is used in the skip mode and the merge mode, the highest picture on the reference picture list may be used as the reference picture. Reference pictures included in a reference picture list may be sorted based on a difference in a picture order count (POC) between a current picture and a corresponding reference picture. The POC corresponds to the display order of the pictures and may be distinguished from the coding order.

The subtraction unit 115 generates a residual sample which is a difference between the original sample and the prediction sample. When the skip mode is applied, residual samples may not be generated as described above.

The transform unit 120 generates a transform coefficient by transforming the residual sample in units of transform blocks. The transform unit 120 may perform the transformation according to the size of the transform block and the prediction mode applied to the coding block or the prediction block that spatially overlaps the transform block. For example, if intra prediction is applied to the coding block or the prediction block that overlaps the transform block, and the transform block is a 4 × 4 residual array, the residual sample uses a discrete sine transform (DST). In other cases, the residual sample may be transformed by using a discrete cosine transform (DCT).

The quantization unit 125 may quantize the transform coefficients to generate quantized transform coefficients.

The reordering unit 130 rearranges the quantized transform coefficients. The reordering unit 130 may reorder the quantized transform coefficients in the form of a block into a one-dimensional vector form through a coefficient scanning method. Although the reordering unit 130 has been described in a separate configuration, the reordering unit 130 may be part of the quantization unit 125.

The entropy encoding unit 135 may perform entropy encoding on the quantized transform coefficients. Entropy encoding may include, for example, encoding methods such as exponential Golomb, context-adaptive variable length coding (CAVLC), context-adaptive binary arithmetic coding (CABAC), and the like. The entropy encoding unit 135 may encode information necessary for video reconstruction other than the quantized transform coefficients (for example, a value of a syntax element) together or separately. Entropy encoded information may be transmitted or stored in units of network abstraction layer (NAL) units in the form of bitstreams.

The inverse quantization unit 141 inverse quantizes the quantized values (quantized transform coefficients) in the quantization unit 125, and the inverse transform unit 142 inverse transforms the inverse quantized values in the inverse quantization unit 141 to generate a residual sample. Create

The adder 150 reconstructs the picture by combining the residual sample and the predictive sample. The residual sample and the predictive sample may be added in units of blocks to generate a reconstructed block. Although the adder 150 has been described in a separate configuration, the adder 150 may be part of the predictor 110. On the other hand, the adder 150 may be called a restoration unit or a restoration block generation unit.

The filter unit 155 may apply a deblocking filter and / or a sample adaptive offset to the reconstructed picture. Through deblocking filtering and / or sample adaptive offset, the artifacts of the block boundaries in the reconstructed picture or the distortion in the quantization process can be corrected. The sample adaptive offset may be applied on a sample basis and may be applied after the process of deblocking filtering is completed. The filter unit 155 may apply an adaptive loop filter (ALF) to the reconstructed picture. ALF may be applied to the reconstructed picture after the deblocking filter and / or sample adaptive offset is applied.

The memory 160 may store reconstructed pictures (decoded pictures) or information necessary for encoding / decoding. Here, the reconstructed picture may be a reconstructed picture after the filtering process is completed by the filter unit 155. The stored reconstructed picture may be used as a reference picture for (inter) prediction of another picture. For example, the memory 160 may store (reference) pictures used for inter prediction. In this case, pictures used for inter prediction may be designated by a reference picture set or a reference picture list.

Referring to FIG. 2, the video decoding apparatus 200 may include an entropy decoding unit 210, a residual processor 220, a predictor 230, an adder 240, a filter 250, and a memory 260. It may include. Here, the residual processor 220 may include a reordering unit 221, an inverse quantization unit 222, and an inverse transform unit 223.

When a bitstream including video information is input, the video decoding apparatus 200 may reconstruct the video in response to a process in which the video information is processed in the video encoding apparatus.

For example, the video decoding apparatus 200 may perform video decoding using a processing unit applied in the video encoding apparatus. Thus, the processing unit block of video decoding may be, for example, a coding unit, and in another example, a coding unit, a prediction unit, or a transform unit. The coding unit may be split along the quad tree structure and / or binary tree structure from the largest coding unit.

The prediction unit and the transform unit may be further used in some cases, in which case the prediction block is a block derived or partitioned from the coding unit and may be a unit of sample prediction. At this point, the prediction unit may be divided into subblocks. The transform unit may be divided along the quad tree structure from the coding unit, and may be a unit for deriving a transform coefficient or a unit for deriving a residual signal from the transform coefficient.

The entropy decoding unit 210 may parse the bitstream and output information necessary for video reconstruction or picture reconstruction. For example, the entropy decoding unit 210 decodes information in a bitstream based on a coding method such as exponential Golomb coding, CAVLC, or CABAC, quantized values of syntax elements necessary for video reconstruction, and residual coefficients. Can be output.

More specifically, the CABAC entropy decoding method receives a bin corresponding to each syntax element in a bitstream, and decodes syntax element information and decoding information of neighboring and decoding target blocks or information of symbols / bins decoded in a previous step. The context model may be determined using the context model, the probability of occurrence of a bin may be predicted according to the determined context model, and arithmetic decoding of the bin may be performed to generate a symbol corresponding to the value of each syntax element. have. In this case, the CABAC entropy decoding method may update the context model by using the information of the decoded symbol / bin for the context model of the next symbol / bean after determining the context model.

The information related to the prediction among the information decoded by the entropy decoding unit 210 is provided to the prediction unit 230, and the residual value on which the entropy decoding has been performed by the entropy decoding unit 210, that is, the quantized transform coefficient, is used as a reordering unit ( 221 may be input.

The reordering unit 221 may rearrange the quantized transform coefficients in a two-dimensional block form. The reordering unit 221 may perform reordering in response to coefficient scanning performed by the encoding apparatus. Here, the rearrangement unit 221 has been described in a separate configuration, but the rearrangement unit 221 may be part of the inverse quantization unit 222.

The inverse quantization unit 222 may dequantize the quantized transform coefficients based on the (inverse) quantization parameter and output the transform coefficients. In this case, information for deriving a quantization parameter may be signaled from the encoding apparatus.

The inverse transform unit 223 may inversely transform transform coefficients to derive residual samples.

The prediction unit 230 may perform prediction on the current block and generate a predicted block including prediction samples for the current block. The unit of prediction performed by the prediction unit 230 may be a coding block, a transform block, or a prediction block.

The prediction unit 230 may determine whether to apply intra prediction or inter prediction based on the information about the prediction. In this case, a unit for determining which of intra prediction and inter prediction is to be applied and a unit for generating a prediction sample may be different. In addition, the unit for generating a prediction sample in inter prediction and intra prediction may also be different. For example, whether to apply inter prediction or intra prediction may be determined in units of CUs. In addition, for example, in inter prediction, a prediction mode may be determined and a prediction sample may be generated in PU units, and in intra prediction, a prediction mode may be determined in PU units and a prediction sample may be generated in TU units.

In the case of intra prediction, the prediction unit 230 may derive the prediction sample for the current block based on the neighbor reference samples in the current picture. The prediction unit 230 may derive the prediction sample for the current block by applying the directional mode or the non-directional mode based on the neighbor reference samples of the current block. In this case, the prediction mode to be applied to the current block may be determined using the intra prediction mode of the neighboring block.

In the case of inter prediction, the prediction unit 230 may derive the prediction sample for the current block based on the sample specified on the reference picture by the motion vector on the reference picture. The prediction unit 230 may apply any one of a skip mode, a merge mode, and an MVP mode to derive a prediction sample for the current block. In this case, motion information required for inter prediction of the current block provided by the video encoding apparatus, for example, information about a motion vector, a reference picture index, and the like may be obtained or derived based on the prediction information.

In the skip mode and the merge mode, the motion information of the neighboring block may be used as the motion information of the current block. In this case, the neighboring block may include a spatial neighboring block and a temporal neighboring block.

The prediction unit 230 may construct a merge candidate list using motion information of available neighboring blocks, and may use information indicated by the merge index on the merge candidate list as a motion vector of the current block. The merge index may be signaled from the encoding device. The motion information may include a motion vector and a reference picture. When the motion information of the temporal neighboring block is used in the skip mode and the merge mode, the highest picture on the reference picture list may be used as the reference picture.

In the skip mode, unlike the merge mode, the difference (residual) between the prediction sample and the original sample is not transmitted.

In the MVP mode, the motion vector of the current block may be derived using the motion vector of the neighboring block as a motion vector predictor. In this case, the neighboring block may include a spatial neighboring block and a temporal neighboring block.

For example, when the merge mode is applied, a merge candidate list may be generated by using a motion vector of a reconstructed spatial neighboring block and / or a motion vector corresponding to a Col block, which is a temporal neighboring block. In the merge mode, the motion vector of the candidate block selected from the merge candidate list is used as the motion vector of the current block. The information about the prediction may include a merge index indicating a candidate block having an optimal motion vector selected from candidate blocks included in the merge candidate list. In this case, the prediction unit 230 may derive the motion vector of the current block by using the merge index.

As another example, when the Motion Vector Prediction (MVP) mode is applied, a motion vector predictor candidate list may be generated using a motion vector of a reconstructed spatial neighboring block and / or a motion vector corresponding to a Col block which is a temporal neighboring block. Can be. That is, the motion vector of the reconstructed spatial neighboring block and / or the Col vector, which is a temporal neighboring block, may be used as a motion vector candidate. The prediction information may include a prediction motion vector index indicating an optimal motion vector selected from the motion vector candidates included in the list. In this case, the prediction unit 230 may select the predicted motion vector of the current block from the motion vector candidates included in the motion vector candidate list using the motion vector index. The prediction unit of the encoding apparatus may obtain a motion vector difference (MVD) between the motion vector of the current block and the motion vector predictor, and may encode the output vector in a bitstream form. That is, MVD may be obtained by subtracting the motion vector predictor from the motion vector of the current block. In this case, the prediction unit 230 may obtain a motion vector difference included in the information about the prediction, and derive the motion vector of the current block by adding the motion vector difference and the motion vector predictor. The prediction unit may also obtain or derive a reference picture index or the like indicating a reference picture from the information about the prediction.

The adder 240 may reconstruct the current block or the current picture by adding the residual sample and the predictive sample. The adder 240 may reconstruct the current picture by adding the residual sample and the predictive sample in block units. Since the residual is not transmitted when the skip mode is applied, the prediction sample may be a reconstruction sample. Although the adder 240 has been described in a separate configuration, the adder 240 may be part of the predictor 230. On the other hand, the adder 240 may be called a restoration unit or a restoration block generation unit.

The filter unit 250 may apply the deblocking filtering sample adaptive offset, and / or ALF to the reconstructed picture. In this case, the sample adaptive offset may be applied in units of samples and may be applied after deblocking filtering. ALF may be applied after deblocking filtering and / or sample adaptive offset.

The memory 260 may store reconstructed pictures (decoded pictures) or information necessary for decoding. Here, the reconstructed picture may be a reconstructed picture after the filtering process is completed by the filter unit 250. For example, the memory 260 may store pictures used for inter prediction. In this case, pictures used for inter prediction may be designated by a reference picture set or a reference picture list. The reconstructed picture can be used as a reference picture for another picture. In addition, the memory 260 may output the reconstructed picture in an output order.

When coding is performed on an input picture as described above, the coding may be performed based on one processing unit. The processing unit may be referred to as a coding unit (CU). Meanwhile, as coding is performed in units of regions including similar information in the picture, conversion efficiency may be improved, thereby improving overall coding efficiency. In addition, as coding is performed in units of regions including similar information in the picture, prediction accuracy may be improved, thereby improving overall coding efficiency. However, when only a quad tree (QT) structure is applied and the picture is split into only square CUs, splitting the CUs to include exactly similar information may be limited. For example, information representing a specific object in the picture may be widely located in a diagonal direction. In this case, when a single CU includes information representing the specific object, a lot of information other than the information representing the specific object may be included. If the information is included in the plurality of square CUs and includes the information indicating the specific object, coding should be performed on each of the plurality of CUs, thereby reducing coding efficiency. In this case, dividing the picture into non-square CUs including information representing the specific object may improve coding efficiency. Accordingly, the present invention proposes a method of dividing an input picture into a square CU and a non-square CU by applying another division structure together with the quad tree structure. Through this, the picture may be divided into various types of CUs according to the information in the picture, and coding may be performed more efficiently.

The QTBT structure may indicate a structure in which a CU (or CTU) is divided through a QT structure and divided through a binary tree (BT) structure. That is, the QTBT may represent a partition structure formed by combining the QT structure and the BT structure, and when a picture is coded in units of CTU, the CTU may be split through the QT structure, Leaf nodes may be additionally partitioned through the BT structure. Here, the leaf node may represent a CU that is no longer split in the QT structure, and the leaf node may be called an end node. In addition, the QT structure may indicate a structure in which a 2N × 2N size CU (or CTU) is divided into four N × N size sub-CUs, and the BT structure indicates that a 2N × 2N size CU is two N × 2N size CUs or In this case, the structure may be divided into two 2N × N sized sub-CUs. Referring to (a) of FIG. 3, a CU may be divided into square CUs having a lower depth through a QT structure, and in addition, a specific CU among the square CUs may have non-square CUs having a lower depth through a BT structure. It can be divided into

3B illustrates an example in which the syntax of the QTBT structure is transmitted. The solid line shown in (b) of FIG. 3 may represent a QT structure, and the dotted line may represent a BT structure. Also, the syntax for the CU of the lower depth may be indicated in the upper depth from the top to the bottom. In addition, the syntax for the upper left, upper right, lower left and lower right CUs may be indicated in a left to right direction. Specifically, the uppermost number may represent the syntax for a CU of n depth, the numbers in the second position from the top are CUs of n + 1 depth, the numbers in the third position from the CUs of n + 2 depth, and the fourth from the top. The numbers in the location may indicate the syntax for CUs of n + 3 depth. Also, numbers indicated in bold may indicate values of syntaxes for the QT structure, and numbers not indicated in bold may indicate values of syntaxes for the BT structure.

Referring to FIG. 3B, a QT segmentation flag indicating whether a CU is partitioned through the QT structure may be transmitted. That is, a flag indicating whether the 2N × 2N size CU is divided into four N × N size CUs may be transmitted. QT_split_flag may indicate a syntax element for the QT split flag. For example, when the value of the QT splitting flag for the CU is 1, the CU may be divided into 4 sub-CUs, and when the value of the QT splitting flag for the CU is 0, the CU May not be divided. In addition, information about a maximum CU size, a minimum CU size, a maximum depth, etc. in the QT structure may be transmitted to adjust the QT structure for the input image. Information about the above-described QT structure may be transmitted for each slice type, or may be transmitted for each of image components (luminance component, chroma component, etc.).

Referring to (b) of FIG. 3, information about the BT structure may be transmitted for an end node that is no longer split in the QT structure. That is, information about the BT structure for the CU corresponding to the end node in the QT structure may be transmitted. Here, the information including the information on the BT structure may be referred to as additional partition information. For example, a BT partition flag indicating whether the CU is partitioned through the BT structure, that is, whether the BT structure is applied to the CU may be transmitted. BT_split_flag may indicate a syntax element for the BT split flag. Specifically, when the value for the BT split flag is 1, the CU may be split into two sub-CUs, and when the value for the BT split flag is 0, the CU may not be split. In addition, information about a maximum CU size, a minimum CU size, a maximum depth, etc. in the BT structure may be transmitted to adjust the BT structure for the input image. Information about the above-described BT structure may be transmitted for each slice type or may be transmitted for each of the image components. When the CU is divided through the BT structure, the CU may be divided in a horizontal or vertical direction. In other words, the CU of 2N × 2N size may be divided into subCUs of 2N × N size, or the CU of 2N × 2N size may be divided into subCUs of N × 2N size. In which direction the CU is divided, that is, a BT partition mode index indicating the partition type of the CU may be transmitted. BT_split_mode may indicate a syntax element for the BT split mode index. Specifically, for example, when the value of the BT split mode index is 1, the CU may be split into sub-CUs in a vertical direction, that is, Nx2N size, and when the value of the BT split mode index is 0, The CU may be divided into sub-CUs having a horizontal direction, that is, 2N × N size.

Referring to FIG. 4, QT_split_flag for a target CU may be transmitted. The QT_split_flag may indicate whether the target CU is split through the QT structure as described above. That is, the QT_split_flag may indicate whether the target CU is divided into sub-CUs having a size of half height and half width of the target CU.

Specifically, for example, when the value of the QT_split_flag of the target CU is 1, that is, the QT_split_flag indicates that the target CU is divided into sub-CUs having a size of half height and half width of the target CU. The target CU may be divided into the sub CUs. In this case, the QT_split_flag for the sub-CUs may be transmitted. That is, the target CU is divided into CUs of lower depths rather than coding recursively so that CUs of end nodes that are no longer split can be derived.

On the other hand, when the value of the QT_split_flag for the target CU of the terminal node is 0, that is, the QT_split_flag indicates that the target CU is not divided into sub-CUs having the size of half the height and half the width of the target CU The BT_split_flag for the target CU may be transmitted. The BT_split_flag may indicate whether the target CU is divided through the BT structure as described above. That is, the BT_split_flag may indicate, for example, whether the target CU having a 2Nx2N size is divided into sub CUs having an Nx2N size or a 2NxN size. When the BT structure is applied to the target CU, shapes of CUs divided from the target CU may be determined according to the BT_split_flag and BT_split_mode values.

Specifically, when the value of the BT_split_flag is 1, the target CU may be divided into sub-CUs having an Nx2N size or 2NxN size, and when the value of the BT_split_flag is 0, the target CU may not be split. . When the BT_split_flag indicates that the target CU is split through the BT structure, BT_split_mode for the target CU may be transmitted. The BT_split_mode may indicate in which direction the target CU is split, that is, the split type of the target CU. For example, when the value of the BT_split_mode is 1, the CU may be divided into sub-CUs in a vertical direction, that is, an Nx2N size, and when the value of the BT_split_mode is 0, the CU is in a horizontal direction, that is, 2NxN It can be divided into sub-CUs of size.

In addition, the syntax of the QTBT structure can be represented as the following table.

Here, QT_split_flag may indicate a syntax element of the above-described QT splitting flag, BT_split_flag may indicate a syntax element of the above-mentioned BT splitting flag, and BT_split_mode may indicate a syntax element of the above-described BT splitting mode index.

Meanwhile, the CU in the picture may be partitioned through the QTBT structure described above, or may be partitioned through a quad tree geometry partition (QTGP) structure.

The QTGP structure may indicate a structure in which a CU (or CTU) is divided through a QT structure and divided through a geometry partition (GP) structure. The GP structure may be referred to as a geometry tree (GT) structure. That is, the QTGP structure may represent a partition structure formed by combining the QT structure and the GP structure. When a picture is coded in units of CTU, the CTU may be split through the QT structure, and the QT structure The leaf node of may be additionally divided through the GP structure. Here, the GP structure may indicate a structure in which the CU is divided into various non-square sub-CUs. That is, as shown in (a) of FIG. 5, various types of sub-CUs may be derived in addition to the N × 2N or 2N × N sized sub-CUs derived through the BT structure. Referring to FIG. 5A, a CU may be divided into square CUs having a lower depth through a QT structure, and in addition, a specific CU among the square CUs may have a non-square CU having a lower depth through the GP structure. Can be divided into

5B illustrates an example in which the syntax of the QTGP structure is transmitted. The solid line shown in (b) of FIG. 5 may represent a QT structure, and the dotted line may represent a GP structure. Also, the syntax for the CU of the lower depth may be indicated in the upper depth from the top to the bottom. In addition, the syntax for the upper left, upper right, lower left and lower right CUs may be indicated in a left to right direction. Specifically, the uppermost number may represent the syntax for a CU of n depth, the numbers in the second position from the top are CUs of n + 1 depth, the numbers in the third position from the CUs of n + 2 depth, and the fourth from the top. The numbers in the location may indicate the syntax for CUs of n + 3 depth. Also, numbers indicated in bold may indicate values of syntaxes for the QT structure, and numbers not indicated in bold may indicate values of syntaxes for the GP structure.

Referring to FIG. 5B, a QT segmentation flag indicating whether a CU is partitioned through the QT structure may be transmitted. That is, a flag indicating whether the 2N × 2N size CU is divided into four N × N size CUs may be transmitted. QT_split_flag may indicate a syntax element for the QT split flag. For example, when the value of the QT splitting flag for the CU is 1, the CU may be divided into 4 sub-CUs, and when the value of the QT splitting flag for the CU is 0, the CU May not be divided. In addition, information about a maximum CU size, a minimum CU size, a maximum depth, etc. in the QT structure may be transmitted to adjust the QT structure for the input image. Information about the above-described QT structure may be transmitted for each slice type, or may be transmitted for each of image components (luminance component, chroma component, etc.).

Referring to FIG. 5B, information about a GP structure may be transmitted for an end node that is no longer split in the QT structure. That is, the information on the GP structure for the CU corresponding to the end node in the QT structure can be transmitted. Here, the information including the information on the GP structure may be referred to as additional partition information. For example, a GP splitting flag indicating whether the CU is split through the GP structure, that is, whether the GP structure is applied to the CU may be transmitted. GP_split_flag (or GT_split_flag) may indicate a syntax element for the GP split flag. Specifically, when the value for the GP splitting flag is 1, the CU may be split into two sub-CUs, and when the value for the GP splitting flag is 0, the CU may not be split.

On the other hand, if the GP splitting flag for the CU indicates that the GP structure is applied to the CU, the splitting angle and / or from the center of the CU to derive the split type through the GP structure. Information about the distance may be transmitted. That is, information about the division boundary line for the CU may be transmitted, and the CU may be divided based on the division boundary line derived based on the information.

6A to 6D exemplarily show a division boundary derived based on the information on the division angle and / or the distance from the center of the CU. Referring to FIG. 6A, an angle of a direction (or boundary line) in which the CU is divided may be derived based on the information on the split angle, and the boundary of the boundary line in which the CU is divided based on information about a distance from the midpoint. The location can be derived.

For example, when the value of the dividing angle derived based on the information on the dividing angle is 0 degrees and the distance from the midpoint derived from the information on the distance from the midpoint is 0, the dividing boundary line is the The midpoint of the CU may pass vertically, and the CU of 2N × 2N size may be divided in the same way as the type divided into sub-CUs of N × 2N size in the BT structure. Further, when the value of the dividing angle derived based on the information on the dividing angle is 90 degrees, and the distance from the midpoint derived from the information on the distance from the midpoint is 0, the dividing boundary line is defined by the CU. The midpoint may pass horizontally, and the CU of 2N × 2N size may be divided in the same manner as the type divided into subCUs of 2N × N size in the BT structure. In addition, referring to (a) to (e) of FIGS. 6B to 6D, the division angle for the CU may be selectively used depending on the degree of division, such as 11.25 degrees, 25 degrees, 45 degrees, or 90 degrees, and 360 degrees. It may be divided evenly in the angular range or unevenly about a specific angle.

In addition, one unit, two samples, four samples, or the like may be selectively used as a unit of distance from the midpoint, or a distance unit adaptively derived according to the size of the CU may be used. For example, when the size of the CU is 4x4 size, the unit of the distance from the midpoint for the CU can be derived by one sample, and when the size of the CU is 8x8 size, the midpoint for the CU The unit of distance from can be derived with two samples. As a result, a distance from the midpoint for the CU can be derived. For example, when the value of the split information representing the distance from the midpoint obtained through the bitstream is x, the unit of the distance is 1 If the sample, the distance from the midpoint may be derived as x. If the unit of the distance is two samples, the distance from the midpoint may be derived as 2x. In addition, when the size of the CU is 16x16 size, the unit of the distance from the midpoint can be derived with 4 samples. Meanwhile, the greater the distance from the center of the CU, the greater the unit may be applied. Specifically, for example, when the size of the CU is 32x32 size, when the value of the distance from the midpoint is 4 or less, a distance of one sample unit may be used, and the value of the distance from the midpoint is If more than 4 or less than 8, a distance of 2 sample units may be used, and if the value of the distance from the midpoint is more than 8, a distance of 4 sample units may be used.

A syntax element for the information representing the division angle and / or the distance from the midpoint may be transmitted to the decoding apparatus through a bitstream to indicate the division information of the CU.

In addition, specific partition types through the GP structure may be preset, and an index indicating one of the specific partition types may be transmitted. The syntax element representing the index may be referred to as GP_mode (or GT_mode). The index may be referred to as a GP split index. In this case, the partition type of the CU may be derived based on the index, and the CU may be divided into non-square sub-CUs based on the partition type. For example, (N / 4) x2N type, (N / 2) x2N type, Nx2N type, 2NxN type and 2Nx (N / 2), 2Nx (N / 4) types may be preset to the specific division types. The index may indicate one of the types. When the index indicates an nNx2N type, the CU may be divided into an nNx2N size subCU and a (2-n) Nx2N size subCU. In addition, when the index indicates a 2NxnN type, the CU may be divided into a 2NxnN sized sub CU and a 2Nx (2-n) N sized CU.

Referring to FIG. 7, QT_split_flag for a target CU may be transmitted. The QT_split_flag may indicate whether the target CU is split through the QT structure as described above. That is, the QT_split_flag may indicate whether the target CU is divided into sub-CUs having a size of half height and half width of the target CU.

On the other hand, when the value of the QT_split_flag for the target CU of the terminal node is 0, that is, the QT_split_flag indicates that the target CU is not divided into sub-CUs having the size of half the height and half the width of the target CU GP_split_flag for the target CU may be transmitted. The GP_split_flag may indicate whether the target CU is split through the GP structure as described above. That is, the GP_split_flag may indicate, for example, whether the target CU having a 2N × 2N size is divided into various non-square sub-CUs. When the GP structure is applied to the target CU, shapes of CUs divided from the target CU may be determined according to the GP_split_flag and GP_mode values.

Specifically, when the value of the GP_split_flag is 1, the target CU may be split into a split type indicated by the GP_mode, and when the value of the GP_split_flag is 0, the split type of the target CU may be derived as a 2N × 2N type. have. In other words, when the value of the GP_split_flag is 0, the target CU having a size of 2N × 2N may not be divided. When the GP_split_flag indicates that the target CU is split through the GP structure, the GP_mode for the target CU may be transmitted. The GP_mode may indicate in which direction the target CU is divided, that is, the division type of the target CU.

In addition, the syntax of the QTGP structure can be represented as the following table.

Here, QT_split_flag may indicate a syntax element of the above-described QT splitting flag, GP_split_flag may indicate a syntax element of the above-mentioned GP splitting flag, and GP_mode may indicate a syntax element of the above-mentioned GP splitting mode index.

In addition to the above-described structures, a CU in a picture may be partitioned through a quad tree binary tree geometry partition (QTBTGP) structure.

The QTBTGP structure may indicate a structure in which a CU (or CTU) is divided through a QT structure and divided through a BT structure and a geometry partition (GP) structure. That is, the QTBTGP structure may represent a partition structure formed of a combination of the QT structure, the BT structure, and the GP structure. For example, when a picture is coded in units of CTU, the CTU may be split through the QT structure, and the leaf node of the QT structure may be further split through the BT structure or the GP structure. Referring to FIG. 8A, a CU may be divided into square CUs having a lower depth through a QT structure, and in addition, a specific CU among the square CUs may have a lower depth through the BT structure or the GP structure. It can be divided into non-square CU of.

8B illustrates an example in which the syntax of the QTBTGP structure is transmitted. The solid line shown in FIG. 8B may represent a QT structure, and the dotted line may represent a BT structure and a GP structure. Also, the syntax for the CU of the lower depth may be indicated in the upper depth from the top to the bottom. In addition, the syntax for the upper left, upper right, lower left and lower right CUs may be indicated in a left to right direction. Specifically, the uppermost number may represent the syntax for a CU of n depth, the numbers in the second position from the top are CUs of n + 1 depth, the numbers in the third position from the CUs of n + 2 depth, and the fourth from the top. The numbers in the location may indicate the syntax for CUs of n + 3 depth. Further, numbers indicated in bold may indicate values of syntaxes for the QT structure, and numbers not indicated in bold may indicate values of syntaxes for the BT structure and the GP structure.

Referring to FIG. 8B, a QT splitting flag indicating whether a CU is split through the QT structure may be transmitted. That is, a flag indicating whether the 2N × 2N size CU is divided into four N × N size CUs may be transmitted. QT_split_flag may indicate a syntax element for the QT split flag. For example, when the value of the QT splitting flag for the CU is 1, the CU may be divided into 4 sub-CUs, and when the value of the QT splitting flag for the CU is 0, the CU May not be divided. In addition, information about a maximum CU size, a minimum CU size, a maximum depth, etc. in the QT structure may be transmitted to adjust the QT structure for the input image. Information about the above-described QT structure may be transmitted for each slice type, or may be transmitted for each of image components (luminance component, chroma component, etc.).

Referring to (b) of FIG. 8, the information on the BT structure or the information on the GP structure may be transmitted to an end node that is no longer split in the QT structure. That is, information about the BT structure or information about the GP structure may be transmitted for the CU corresponding to the end node in the QT structure. Here, the information including the information on the BT structure or the information on the GP structure may be referred to as additional partition information. For example, a BTGP partition flag indicating whether the CU is partitioned through the BT structure or the GP structure, that is, whether the BT structure or the GP structure is applied to the CU may be transmitted. BTGP_split_flag may indicate a syntax element for the BTGP split flag. Specifically, when the value for the BTGP split flag is 1, the CU may be split into two sub-CUs, and when the value for the BTGP split flag is 0, the CU may not be split.

Meanwhile, when the BTGP partition flag for the CU indicates that the BT structure or the GP structure is applied to the CU, information indicating whether the CU is divided according to which of the BT structure and the GP structure may be transmitted. have. Information representing one of the BT structure and the GP structure may be referred to as BTGP mode information. The syntax element for the BTGP mode information may be referred to as BTGP_mode. In this case, the partition structure of the CU may be derived based on the BTGP mode information, and the CU may be divided into non-square sub-CUs through the derived partition structure. For example, when the value of the BTGP mode information is 1, the CU may be split into the BT structure, and when the value of the BTGP mode information is 0, the CU may be split through the GP structure. . Also, as another example, when the value of the BTGP mode information is 0, the CU may be split into the BT structure, and when the value of the BTGP mode information is 1, the CU may be split through the GP structure. have.

Meanwhile, in order to adjust the BT structure for the input image, information on the maximum CU size, the minimum CU size, the maximum depth, etc. in the BT structure may be transmitted. Information about the above-described BT structure may be transmitted for each slice type or may be transmitted for each of the image components.

When the BTGP mode information for the CU indicates a BT structure, that is, when the CU is split through the BT structure, the CU may be split in a horizontal or vertical direction. In other words, the CU of 2N × 2N size may be divided into subCUs of 2N × N size, and the CU of 2N × 2N size may be divided into subCUs of N × 2N size. Accordingly, in which direction the CU is divided, that is, a BT partition mode index indicating the partition type of the CU may be transmitted. BT_split_mode may indicate a syntax element for the BT split mode index. Specifically, for example, when the value of the BT split mode index is 1, the CU may be split into sub-CUs in a vertical direction, that is, Nx2N size, and when the value of the BT split mode index is 0, The CU may be divided into sub-CUs having a horizontal direction, that is, 2N × N size.

Meanwhile, when the BTGP mode information for the CU indicates a GP structure, that is, when the CU is split through the GP structure, a split angle and / or to derive a split type through the GP structure. Information about the distance from the center of the CU may be transmitted. An angle in a direction in which the CU is split may be derived based on the information on the split angle, and a position of a boundary line at which the CU is split may be derived based on information about a distance from the midpoint. For example, when the value of the dividing angle derived based on the information on the dividing angle is 0 degrees and the distance from the midpoint derived from the information on the distance from the midpoint is 0, the CU of 2N × 2N size In the BT structure, N may be partitioned in the same manner as a type partitioned into N × 2N sub-CUs. Further, when the value of the split angle derived based on the information on the split angle is 90 degrees, and the distance from the midpoint derived from the information on the distance from the midpoint is 0, the CU of 2N × 2N size is determined. In the BT structure, it may be partitioned in the same manner as the type divided into sub CUs having a size of 2N × N.

In addition, when the BTGP mode information for the CU indicates a GP structure, that is, when the CU is partitioned through the GP structure, specific partition types through the GP structure may be preset, and the specific partition type An index pointing to one of these may be transmitted. The syntax element representing the index may be referred to as GP_mode. In this case, the partition type of the CU may be derived based on the index, and the CU may be divided into non-square sub-CUs based on the partition type. The index may be referred to as a GP split index. For example, (N / 4) x2N type, (N / 2) x2N type, Nx2N type, 2NxN type and 2Nx (N / 2), 2Nx (N / 4) types may be preset to the specific division types. The index may indicate one of the types. When the index indicates an nNx2N type, the CU may be divided into an nNx2N size subCU and a (2-n) Nx2N size subCU. In addition, when the index indicates a 2NxnN type, the CU may be divided into a 2NxnN sized sub CU and a 2Nx (2-n) N sized CU.

Referring to FIG. 9, QT_split_flag for a target CU may be transmitted. The QT_split_flag may indicate whether the target CU is split through the QT structure as described above. That is, the QT_split_flag may indicate whether the target CU is divided into sub-CUs having a size of half height and half width of the target CU.

On the other hand, when the value of the QT_split_flag for the target CU of the terminal node is 0, that is, the QT_split_flag indicates that the target CU is not divided into sub-CUs having the size of half the height and half the width of the target CU The BTGP_split_flag for the target CU may be transmitted. The BTGP_split_flag may indicate whether the target CU is split through the BT structure or the GP structure as described above. That is, the BTGP_split_flag may indicate, for example, whether the target CU having a 2N × 2N size is divided into non-square sub-CUs.

Specifically, when the value of the BTGP_split_flag is 1, the BTGP_mode for the target CU may be transmitted. The BTGP_mode may indicate a partition structure applied to the target CU among the BT structure and the GP structure. For example, when the value of the BTGP_mode is 1, the partition structure of the target CU may be derived as the BT structure, and when the value of the BTGP_mode is 0, the partition structure of the target CU is derived as the GP structure. Can be.

When the value of the BTGP_mode is 0, that is, when the BTGP_mode indicates that the target CU is divided through the GP structure, the GP_mode for the target CU may be transmitted. The GP_mode may indicate in which direction the target CU is divided, that is, the division type of the target CU.

In addition, when the value of the BTGP_mode is 1, that is, when the BTGP_mode indicates that the target CU is divided through the BT structure, the BT_split_mode for the target CU may be transmitted. The BT_split_mode may indicate a split type of the target CU among an Nx2N type and a 2NxN type. Here, the Nx2N type may be referred to as a vertical partition, and the 2NxN type may be referred to as a horizontal partition. For example, when the value of the BT_split_mode is 1, the CU may be divided into sub-CUs in a vertical direction, that is, an Nx2N size, and when the value of the BT_split_mode is 0, the CU is in a horizontal direction, that is, 2NxN It can be divided into sub-CUs of size. In addition, when the target CU is split through the BT structure, the target CU may be split into CUs of lower depths more recursively through the BT structure.

In addition, the syntax of the QTBTGP structure can be represented as the following table.

Here, QT_split_flag may represent a syntax element of the above-described QT splitting flag, BTGP_split_flag may represent a syntax element of whether the above-described BT structure or GP structure is split, and BTGP_mode is a syntax element of the above-described BT structure or GP structure recognition splitting. Can be represented.

In addition to the above-described structures, a CU in a picture may be partitioned through a quad tree geometry partition binary tree (QTGPBT) structure.

The QTGPBT structure may indicate a partition structure in which the QT structure, the BT structure, and the GP structure are combined. That is, a CU (or CTU) may be divided through a QT structure, some of the derived CUs may be split through a GP structure, and a CU that is not split into a GP structure may be split through a BT structure. For example, when a picture is coded in units of CTU, the CTU may be split through the QT structure, and the leaf node of the QT structure may be further split through the GP structure, and the GP structure Leaf nodes not partitioned through may be additionally partitioned through the BT structure. Referring to FIG. 10A, a CU may be divided into square CUs having a lower depth through a QT structure, and a specific CU among the square CUs may be divided into the GP structure. CUs other than the CU may be divided into non-square CUs of a lower depth through the BT structure.

10B illustrates an example in which the syntax of the QTGPBT structure is transmitted. The solid line shown in (b) of FIG. 10 may represent a QT structure, and the dotted line may represent a BT structure and a GP structure. Also, the syntax for the CU of the lower depth may be indicated in the upper depth from the top to the bottom. In addition, the syntax for the upper left, upper right, lower left and lower right CUs may be indicated in a left to right direction. Specifically, the uppermost number may represent the syntax for a CU of n depth, the numbers in the second position from the top are CUs of n + 1 depth, the numbers in the third position from the CUs of n + 2 depth, and the fourth from the top. The numbers in the location may indicate the syntax for CUs of n + 3 depth. Further, numbers indicated in bold may indicate values of syntaxes for the QT structure, and numbers not indicated in bold may indicate values of syntaxes for the BT structure and the GP structure.

Referring to FIG. 10B, a QT segmentation flag indicating whether a CU is partitioned through the QT structure may be transmitted. That is, a flag indicating whether the 2N × 2N size CU is divided into four N × N size CUs may be transmitted. QT_split_flag may indicate a syntax element for the QT split flag. For example, when the value of the QT splitting flag for the CU is 1, the CU may be divided into 4 sub-CUs, and when the value of the QT splitting flag for the CU is 0, the CU May not be divided. In addition, information about a maximum CU size, a minimum CU size, a maximum depth, etc. in the QT structure may be transmitted to adjust the QT structure for the input image. Information about the above-described QT structure may be transmitted for each slice type, or may be transmitted for each of image components (luminance component, chroma component, etc.).

Referring to FIG. 10B, information on a GP structure may be transmitted for an end node that is no longer divided in the QT structure. That is, the information on the GP structure for the CU corresponding to the end node in the QT structure can be transmitted. Here, the information including the information on the GP structure may be referred to as additional partition information. For example, a GP splitting flag indicating whether the CU is split through the GP structure, that is, whether the GP structure is applied to the CU may be transmitted. GP_split_flag may indicate a syntax element for the GP split flag. Specifically, when the value for the GP splitting flag is 1, the CU may be split into two sub-CUs, and when the value for the GP splitting flag is 0, the CU is split through the BT structure. Can be.

On the other hand, when the GP splitting flag for the CU indicates that the GP structure is applied to the CU, a split angle and / or midpoint of the CU to derive a split type through the GP structure as described above Information about the distance from the center may be transmitted. An angle in a direction in which the CU is split may be derived based on the information on the split angle, and a position of a boundary line at which the CU is split may be derived based on information about a distance from the midpoint. In addition, as described above, specific partition types through the GP structure may be preset, and an index indicating one of the specific partition types may be transmitted. The syntax element representing the index may be referred to as GP_mode. In this case, the partition type of the CU may be derived based on the index, and the CU may be divided into non-square sub-CUs based on the partition type. The index may be referred to as a GP split index.

On the other hand, when the value for the GP partition flag is 0, that is, when the GP partition flag for the CU indicates that the GP structure is not applied to the CU, it indicates whether the BT structure is applied to the CU. The BT split flag can be sent. BT_split_flag may indicate a syntax element for the BT split flag. Specifically, when the value for the BT split flag is 1, the CU may be split into two sub-CUs, and when the value for the BT split flag is 0, the CU may not be split. In addition, information about a maximum CU size, a minimum CU size, a maximum depth, etc. in the BT structure may be transmitted to adjust the BT structure for the input image. Information about the above-described BT structure may be transmitted for each slice type or may be transmitted for each of the image components. When the CU is divided through the BT structure, the CU may be divided in a horizontal or vertical direction. In other words, the CU of 2N × 2N size may be divided into subCUs of 2N × N size, or the CU of 2N × 2N size may be divided into subCUs of N × 2N size. In which direction the CU is divided, that is, a BT partition mode index indicating the partition type of the CU may be transmitted. BT_split_mode may indicate a syntax element for the BT split mode index. Specifically, for example, when the value of the BT split mode index is 1, the CU may be split into sub-CUs in a vertical direction, that is, Nx2N size, and when the value of the BT split mode index is 0, The CU may be divided into sub-CUs having a horizontal direction, that is, 2N × N size.

Referring to FIG. 11, QT_split_flag for a target CU may be transmitted. The QT_split_flag may indicate whether the target CU is split through the QT structure as described above. That is, the QT_split_flag may indicate whether the target CU is divided into sub-CUs having a size of half height and half width of the target CU.

In addition, referring to FIG. 11, when the value of QT_split_flag for the target CU of the terminal node is 0, that is, the QT_split_flag is divided into sub-CUs in which the target CU has a size of half height and half width of the target CU If not, GP_split_flag for the target CU may be transmitted. The GP_split_flag may indicate whether the target CU is split through the GP structure as described above. That is, the GP_split_flag may indicate, for example, whether the target CU having a 2N × 2N size is divided into various non-square sub-CUs. When the GP structure is applied to the target CU, the GP_mode for the target CU may be transmitted, and the shapes of CUs divided from the target CU may be determined according to the GP_split_flag and GP_mode values as described above.

In addition, referring to FIG. 11, when the value of the GP_split_flag is 0, that is, when the GP_split_flag indicates that the target CU is not divided through the GP structure, BT_split_flag for the target CU may be transmitted. The BT_split_flag may indicate whether the target CU is divided through the BT structure as described above. That is, the BT_split_flag may indicate, for example, whether the target CU having a 2Nx2N size is divided into sub CUs having an Nx2N size or a 2NxN size. When the BT structure is applied to the target CU, shapes of CUs divided from the target CU may be determined according to the BT_split_flag and BT_split_mode values.

Specifically, when the value of the BT_split_flag is 1, the target CU may be divided into sub-CUs having an Nx2N size or 2NxN size, and when the value of the BT_split_flag is 0, the target CU may not be split. . When the BT_split_flag indicates that the target CU is split through the BT structure, BT_split_mode for the target CU may be transmitted. The BT_split_mode may indicate in which direction the target CU is split, that is, the split type of the target CU. For example, when the value of the BT_split_mode is 1, the CU may be divided into sub-CUs in a vertical direction, that is, an Nx2N size, and when the value of the BT_split_mode is 0, the CU is in a horizontal direction, that is, 2NxN It can be divided into sub-CUs of size. Meanwhile, when the target CU is split through the BT structure, BT_split_flag for the sub CU of the target CU may be transmitted and split recursively.

In addition, the syntax of the QTGPBT structure can be represented as the following table.

Here, QT_split_flag may indicate a syntax element of the above-described QT splitting flag, GP_split_flag may indicate a syntax element of the above-mentioned GP splitting flag, and GP_mode may indicate a syntax element of the above-mentioned GP splitting mode index. Referring to Table 4, when GP_split_flag of the target CU to which the QTGPBT structure is applied is 0, the sub CU of the lower depth of the target CU may have the BT structure described above.

In addition to the above-described structures, a CU in a picture may be partitioned through a quad tree binary tree geometry partition (QTBTGP) structure. The QTBTGP structure described below may represent an embodiment different from the QTBTGP structure described above.

The QTBTGP structure may indicate a partition structure in which the QT structure, the BT structure, and the GP structure are combined. That is, a CU (or CTU) may be divided through a QT structure, some of the derived CUs may be split through a BT structure, and a CU not split into a BT structure may be divided through a GP structure. For example, when a picture is coded in units of CTU, the CTU may be divided through the QT structure, and the leaf node of the QT structure may be additionally divided through the BT structure, and the BT structure Leaf nodes that are not partitioned through may be additionally partitioned through the GP structure. Referring to (a) of FIG. 12, a CU may be divided into square CUs having a lower depth through a QT structure, and a specific CU among the square CUs may be divided into the BT structure, and the specific CU may be divided into the BT structure. CUs other than the CU may be divided into non-square CUs of a lower depth through the GP structure.

12B illustrates an example in which the syntax of the QTBTGP structure is transmitted. The solid line shown in (b) of FIG. 12 may represent a QT structure, and the dotted line may represent a BT structure and a GP structure. Also, the syntax for the CU of the lower depth may be indicated in the upper depth from the top to the bottom. In addition, the syntax for the upper left, upper right, lower left and lower right CUs may be indicated in a left to right direction. Specifically, the uppermost number may represent the syntax for a CU of n depth, the numbers in the second position from the top are CUs of n + 1 depth, the numbers in the third position from the CUs of n + 2 depth, and the fourth from the top. The numbers in the location may indicate the syntax for CUs of n + 3 depth. Further, numbers indicated in bold may indicate values of syntaxes for the QT structure, and numbers not indicated in bold may indicate values of syntaxes for the BT structure and the GP structure.

Referring to FIG. 12B, a QT partition flag indicating whether a CU is partitioned through the QT structure may be transmitted. That is, a flag indicating whether the 2N × 2N size CU is divided into four N × N size CUs may be transmitted. QT_split_flag may indicate a syntax element for the QT split flag. For example, when the value of the QT splitting flag for the CU is 1, the CU may be divided into 4 sub-CUs, and when the value of the QT splitting flag for the CU is 0, the CU May not be divided. In addition, information about a maximum CU size, a minimum CU size, a maximum depth, etc. in the QT structure may be transmitted to adjust the QT structure for the input image. Information about the above-described QT structure may be transmitted for each slice type, or may be transmitted for each of image components (luminance component, chroma component, etc.).

Referring to FIG. 12B, information about the BT structure may be transmitted for an end node that is no longer split in the QT structure. That is, information about the BT structure for the CU corresponding to the end node in the QT structure may be transmitted. Here, the information including the information on the BT structure may be referred to as additional partition information. For example, a BT partition flag indicating whether the CU is partitioned through the BT structure, that is, whether the BT structure is applied to the CU may be transmitted. BT_split_flag may indicate a syntax element for the BT split flag. Specifically, when the value for the BT partition flag is 1, the CU may be divided into two sub-CUs, and when the value for the BT partition flag is 0, the CU is partitioned through the GP structure. Can be. In addition, information about a maximum CU size, a minimum CU size, a maximum depth, etc. in the BT structure may be transmitted to adjust the BT structure for the input image. Information about the above-described BT structure may be transmitted for each slice type or may be transmitted for each of the image components. When the CU is divided through the BT structure, the CU may be divided in a horizontal or vertical direction. In other words, the CU of 2N × 2N size may be divided into subCUs of 2N × N size, or the CU of 2N × 2N size may be divided into subCUs of N × 2N size. In which direction the CU is divided, that is, a BT partition mode index indicating the partition type of the CU may be transmitted. BT_split_mode may indicate a syntax element for the BT split mode index. Specifically, for example, when the value of the BT split mode index is 1, the CU may be split into sub-CUs in a vertical direction, that is, Nx2N size, and when the value of the BT split mode index is 0, The CU may be divided into sub-CUs having a horizontal direction, that is, 2N × N size.

Meanwhile, referring to FIG. 12B, when the value for the BT split flag is 0, that is, when the BT split flag for the CU indicates that the BT structure is not applied to the CU, the CU split flag is applied to the CU. A GP splitting flag indicating whether the GP structure is applied to the GP structure may be transmitted. GP_split_flag may indicate a syntax element for the GP split flag. Specifically, when the value for the GP splitting flag is 1, the CU may be split into two sub-CUs, and when the value for the GP splitting flag is 0, the CU may not be split. When the GP splitting flag for the CU indicates that the GP structure is applied to the CU, a split angle and / or a midpoint of the CU may be used to derive a split type through the GP structure as described above. Information about the distance from the center may be transmitted. An angle in a direction in which the CU is split may be derived based on the information on the split angle, and a position of a boundary line at which the CU is split may be derived based on information about a distance from the midpoint. In addition, as described above, specific partition types through the GP structure may be preset, and an index indicating one of the specific partition types may be transmitted. The syntax element representing the index may be referred to as GP_mode. In this case, the partition type of the CU may be derived based on the index, and the CU may be divided into non-square sub-CUs based on the partition type. The index may be referred to as a GP split index.

Referring to FIG. 13, QT_split_flag for a target CU may be transmitted. The QT_split_flag may indicate whether the target CU is split through the QT structure as described above. That is, the QT_split_flag may indicate whether the target CU is divided into sub-CUs having a size of half height and half width of the target CU.

In addition, referring to FIG. 13, when the value of QT_split_flag for the target CU of the terminal node is 0, that is, the QT_split_flag is divided into sub-CUs in which the target CU has a size of half height and half width of the target CU If not, BT_split_flag for the target CU may be transmitted. The BT_split_flag may indicate whether the target CU is divided through the BT structure as described above. That is, the BT_split_flag may indicate, for example, whether the target CU having a 2N × 2N size is divided into non-square sub-CUs. When the BT structure is applied to the target CU, shapes of CUs divided from the target CU may be determined according to the BT_split_flag and BT_split_mode values as described above. The BT_split_mode may indicate a split type of the target CU among an Nx2N type and a 2NxN type. Here, the Nx2N type may be referred to as a vertical partition, and the 2NxN type may be referred to as a horizontal partition. For example, when the value of the BT_split_mode is 1, the CU may be divided into sub-CUs in a vertical direction, that is, an Nx2N size, and when the value of the BT_split_mode is 0, the CU is in a horizontal direction, that is, 2NxN It can be divided into sub-CUs of size. In addition, when the target CU is split through the BT structure, the target CU may be split into CUs of lower depths more recursively through the BT structure.

Meanwhile, referring to FIG. 13, when the value of the BT_split_flag is 0, that is, when the BT_split_flag indicates that the target CU is not split through the BT structure, the GP_split_flag for the target CU may be transmitted. The GP_split_flag may indicate whether the target CU is split through the GP structure as described above. That is, the GP_split_flag may indicate, for example, whether the target CU having a 2N × 2N size is divided into various non-square sub-CUs. When the GP structure is applied to the target CU, the GP_mode for the target CU may be transmitted, and the shapes of CUs divided from the target CU may be determined according to the GP_split_flag and GP_mode values as described above.

Here, QT_split_flag may indicate a syntax element of the above-described QT splitting flag, BT_split_flag may indicate a syntax element of the above-mentioned BT splitting flag, and BT_split_mode may indicate a syntax element of the above-described BT splitting mode index. GP_split_flag may indicate a syntax element of the above-described GP splitting flag, and GP_mode may indicate a syntax element of the above-described GP splitting mode index. Referring to Table 5 above, when the above-described QTBTGP structure is applied, the leaf node of the QT structure may be applied to the GP structure.

On the other hand, the CU in the picture may be divided through the QTBTGP structure of the embodiment different from the above-described QTBTGP structure.

The QTBTGP structure may indicate a partition structure in which the QT structure, the BT structure, and the GP structure are combined. That is, a CU (or CTU) may be divided through a QT structure, some of the derived CUs may be split through a BT structure, and a CU not split into a BT structure may be divided through a GP structure. For example, when a picture is coded in units of CTU, the CTU may be divided through the QT structure, and the leaf node of the QT structure may be additionally divided through the BT structure, and the BT structure Leaf nodes that are not partitioned through may be additionally partitioned through the GP structure. Referring to (a) of FIG. 14, a CU may be divided into square CUs having a lower depth through a QT structure, and a particular CU among the square CUs may be divided into the BT structure, and the specific CU may be divided into the BT structure. CUs other than the CU may be divided into non-square CUs of a lower depth through the GP structure.

14B illustrates an example in which the syntax of the QTBTGP structure is transmitted. The solid line shown in FIG. 14B may represent a QT structure, and the dotted line may represent a BT structure and a GP structure. Also, the syntax for the CU of the lower depth may be indicated in the upper depth from the top to the bottom. In addition, the syntax for the upper left, upper right, lower left and lower right CUs may be indicated in a left to right direction. Specifically, the uppermost number may represent the syntax for a CU of n depth, the numbers in the second position from the top are CUs of n + 1 depth, the numbers in the third position from the CUs of n + 2 depth, and the fourth from the top. The numbers in the location may indicate the syntax for CUs of n + 3 depth. Further, numbers indicated in bold may indicate values of syntaxes for the QT structure, and numbers not indicated in bold may indicate values of syntaxes for the BT structure and the GP structure.

Referring to FIG. 14B, a QT splitting flag indicating whether a CU is split through the QT structure may be transmitted. That is, a flag indicating whether the 2N × 2N size CU is divided into four N × N size CUs may be transmitted. QT_split_flag may indicate a syntax element for the QT split flag. For example, when the value of the QT splitting flag for the CU is 1, the CU may be divided into 4 sub-CUs, and when the value of the QT splitting flag for the CU is 0, the CU May not be divided. In addition, information about a maximum CU size, a minimum CU size, a maximum depth, etc. in the QT structure may be transmitted to adjust the QT structure for the input image. Information about the above-described QT structure may be transmitted for each slice type, or may be transmitted for each of image components (luminance component, chroma component, etc.).

Referring to (b) of FIG. 14, information on the BT structure may be transmitted for an end node that is no longer split in the QT structure. That is, information about the BT structure for the CU corresponding to the end node in the QT structure may be transmitted. Here, the information including the information on the BT structure may be referred to as additional partition information. For example, a BT partition flag indicating whether the CU is partitioned through the BT structure, that is, whether the BT structure is applied to the CU may be transmitted. BT_split_flag may indicate a syntax element for the BT split flag. Specifically, when the value for the BT partition flag is 1, the CU may be divided into two sub-CUs, and when the value for the BT partition flag is 0, the CU is partitioned through the GP structure. Can be. In addition, information about a maximum CU size, a minimum CU size, a maximum depth, etc. in the BT structure may be transmitted to adjust the BT structure for the input image. Information about the above-described BT structure may be transmitted for each slice type or may be transmitted for each of the image components. When the CU is divided through the BT structure, the CU may be divided in a horizontal or vertical direction. In other words, the CU of 2N × 2N size may be divided into subCUs of 2N × N size, or the CU of 2N × 2N size may be divided into subCUs of N × 2N size. In which direction the CU is divided, that is, a BT partition mode index indicating the partition type of the CU may be transmitted. BT_split_mode may indicate a syntax element for the BT split mode index. Specifically, for example, when the value of the BT split mode index is 1, the CU may be split into sub-CUs in a vertical direction, that is, Nx2N size, and when the value of the BT split mode index is 0, The CU may be divided into sub-CUs having a horizontal direction, that is, 2N × N size. On the other hand, when the CU is split through the BT structure, BT_split_flag for the sub-CU of the CU can be transmitted, the BT_split_flag is determined whether the sub-CU is split through the BT structure or through the GP structure It may indicate whether it is divided. That is, when the CU is split through the BT structure, BT_split_flag for the sub CU indicating whether the sub CU is split through the BT structure or the GP structure may be additionally transmitted. It may be recursively partitioned through the BT structure or the GP structure.

Meanwhile, referring to FIG. 14B, when the value for the BT partition flag is 0, that is, when the BT partition flag for the CU indicates that the BT structure is not applied to the CU, the CU partition flag is applied to the CU. A GP splitting flag indicating whether the GP structure is applied to the GP structure may be transmitted. GP_split_flag may indicate a syntax element for the GP split flag. Specifically, when the value for the GP splitting flag is 1, the CU may be split into two sub-CUs, and when the value for the GP splitting flag is 0, the split type of the target CU is 2Nx2N type. Can be derived. That is, when the value of the GP_split_flag is 0, the target CU may be split into a 2N × 2N size, and thus, the target CU having a 2N × 2N size may not be split. When the GP splitting flag for the CU indicates that the GP structure is applied to the CU, a split angle and / or a midpoint of the CU may be used to derive a split type through the GP structure as described above. Information about the distance from the center may be transmitted. An angle in a direction in which the CU is split may be derived based on the information on the split angle, and a position of a boundary line at which the CU is split may be derived based on information about a distance from the midpoint. In addition, as described above, specific partition types through the GP structure may be preset, and an index indicating one of the specific partition types may be transmitted. The syntax element representing the index may be referred to as GP_mode. In this case, the partition type of the CU may be derived based on the index, and the CU may be divided into non-square sub-CUs based on the partition type. The index may be referred to as a GP split index.

Referring to FIG. 15, QT_split_flag for a target CU may be transmitted. The QT_split_flag may indicate whether the target CU is split through the QT structure as described above. That is, the QT_split_flag may indicate whether the target CU is divided into sub-CUs having a size of half height and half width of the target CU.

Also, referring to FIG. 15, when the value of the QT_split_flag for the target CU of the terminal node is 0, that is, the QT_split_flag is divided into sub-CUs in which the target CU has a size of half height and half width of the target CU If not, BT_split_flag for the target CU may be transmitted. The BT_split_flag may indicate whether the target CU is divided through the BT structure as described above. That is, the BT_split_flag may indicate, for example, whether the target CU having a 2N × 2N size is divided into non-square sub-CUs. When the BT structure is applied to the target CU, shapes of CUs divided from the target CU may be determined according to the BT_split_flag and BT_split_mode values as described above. The BT_split_mode may indicate a split type of the target CU among an Nx2N type and a 2NxN type. Here, the Nx2N type may be referred to as a vertical partition, and the 2NxN type may be referred to as a horizontal partition. For example, when the value of the BT_split_mode is 1, the CU may be divided into sub-CUs in a vertical direction, that is, an Nx2N size, and when the value of the BT_split_mode is 0, the CU is in a horizontal direction, that is, 2NxN It can be divided into sub-CUs of size. In addition, when the target CU is split through the BT structure, the target CU may be split into CUs of lower depths more recursively through the BT structure.

Meanwhile, referring to FIG. 15, when the value of the BT_split_flag is 0, that is, when the BT_split_flag indicates that the target CU is not split through the BT structure, the GP_split_flag for the target CU may be transmitted. The GP_split_flag may indicate whether the target CU is split through the GP structure as described above. That is, the GP_split_flag may indicate, for example, whether the target CU having a 2N × 2N size is divided into various non-square sub-CUs. When the GP structure is applied to the target CU, the GP_mode for the target CU may be transmitted, and the shapes of CUs divided from the target CU may be determined according to the GP_split_flag and GP_mode values as described above. On the other hand, if the target CU is split through the BT structure, BT_split_flag for an additional sub-CU is additionally transmitted and can be split into CUs of lower depth recursively through the BT structure or the GP structure. .

Here, QT_split_flag may indicate a syntax element of the above-described QT splitting flag, BT_split_flag may indicate a syntax element of the above-mentioned BT splitting flag, and BT_split_mode may indicate a syntax element of the above-described BT splitting mode index. GP_split_flag may indicate a syntax element of the above-described GP splitting flag, and GP_mode may indicate a syntax element of the above-described GP splitting mode index. Referring to Table 5 above, when the above-described QTBTGP structure is applied, the GP structure may be applied to a leaf node to which the BT structure is not applied.

16 schematically shows a video encoding method by an encoding device according to the present invention. The method disclosed in FIG. 16 may be performed by the encoding apparatus disclosed in FIG. 1. Specifically, for example, S1600 to S1610 of FIG. 16 may be performed by the picture division unit of the encoding apparatus, S1620 may be performed by the prediction unit of the encoding apparatus, and S1630 may be entropy encoding of the encoding apparatus. Can be performed by wealth.

The encoding apparatus divides the first target block into first subblocks (S1600). The encoding apparatus may split the first target block into the first subblocks through a quad tree (QT) structure. In detail, for example, the encoding apparatus may divide the first target block into four first subblocks. The first sub blocks may have sizes of half height and half width of the target block. Meanwhile, the encoding apparatus may generate a quad tree (QT) split flag for the first target block. The QT splitting flag may indicate whether a block is divided into sub-blocks having a size of half height and half width of the block.

The encoding apparatus divides the second target block, which is one of the first subblocks, into second subblocks (S1610). The second target block may not be split through the QT structure. If the second target block is not split through the QT structure, the encoding apparatus may split the second target block into second subblocks. The second sub blocks may be non-square blocks.

For example, the second target block may be divided into the second sub blocks through a binary tree (BT) structure. Specifically, for example, when the size of the second target block is 2N × 2N size, the second target block may be divided into the second subblocks of the 2N × N size. Alternatively, the second target block may be divided into the second sub blocks of the N × 2N size.

As another example, the second target block may be divided into the second sub blocks through a geometry partition (GP) structure. In this case, the second target block may be divided into the second sub blocks, which are various square blocks. That is, the second target block may be divided into the second sub blocks having different sizes. Specifically, for example, the second target block may be divided into the second sub-blocks based on one partition type among preset partition types. The division types may include, for example, (N / 4) x2N type, (N / 2) x2N type, Nx2N type, 2NxN type, and 2Nx (N / 2), 2Nx (N / 4) type. When nNx2N type is applied to the second target block, the second target block may be divided into a second subblock of size nNx2N and a second subblock of size (2-n) Nx2N. Alternatively, when the 2NxnN type is applied to the second target block, the second target block may be divided into a second subblock having a 2NxnN size and a second subblock having a 2Nx (2-n) N size.

For example, the second target block may be divided into the second sub blocks based on a specific boundary line. In this case, information representing an angle of the specific boundary line or information representing a distance from a midpoint of the second target block of the specific boundary line may be generated.

The encoding apparatus decodes the second subblocks (S1620). The encoding apparatus may perform a procedure such as transform intra / inter prediction for the second sub block, and generate a reconstruction sample for the second sub block, based on the procedure. The reconstructed picture may be generated.

The encoding apparatus generates, encodes, and outputs first partition information on the first target block, second partition information on the second target block, and additional partition information (S1630). The encoding apparatus may generate the first partitioning information for the first target block and the second partitioning information for the second target block. The first partitioning information may include a quad tree (QT) partition flag for the first target block, and the second partitioning information may include a QT partitioning flag for the second target block. The QT splitting flag may indicate whether a block is divided into sub-blocks having a size of half height and half width of the block.

In addition, the encoding apparatus may generate the additional partition information for the second target block. The additional split information may be generated when the second target block is not split based on the QT split flag for the second target block. That is, the second target block may be generated when the second target block is not divided through the QT structure.

For example, the additional partition information may include a BT partition flag for the second target block. The BT split flag may indicate whether the second target block is divided into the second subblocks having one of 2NxN size and Nx2N size when the size of the second target block is 2Nx2N. When the value of the BT split flag is 1, that is, the BT split flag indicates that the second target block is split into the second sub blocks having one of 2NxN size and Nx2N size. The information may include a BT partition index for the second target block, and the BT partition index may indicate one of the 2NxN size and the Nx2N size as the sizes of the second subblocks.

In addition, when the value of the BT split flag is 0, that is, the BT split flag indicates that the second target block is not split into the second sub blocks having one of 2NxN size and Nx2N size. The additional partition information may include a geometry partition (GP) partition flag for the second target block, and the GP partition flag is partitioned into the second sub blocks having different sizes. It can indicate whether or not. When the value of the GP splitting flag is 1, that is, when the GP splitting flag indicates that the second target block is split into the second subblocks having different sizes, the additional splitting information is divided into a plurality of preset pieces. It may include a GP partition index indicating one of the partition types. For example, the division types include (N / 4) x2N type, (N / 2) x2N type, Nx2N type, 2NxN type and 2Nx (N / 2), 2Nx (N / 4) type, for example. The GP partition index may indicate one of the types.

Alternatively, when the value of the GP splitting flag is 1, that is, when the GP splitting flag indicates that the second target block is split into the second subblocks having different sizes, the additional splitting information is a specific boundary line. Information indicating an angle of or a distance from the midpoint of the second target block of the specific boundary line may be included.

As another example, the additional partition information may include a geometry partition (GP) partition flag for the second target block, and the GP partition flag may include the second sub blocks having different sizes. It may indicate whether or not divided into. If the value of the GP splitting flag is 1, that is, the GP splitting flag indicates that the second target block is split into the second subblocks having different sizes, the additional splitting information is an angle of a specific boundary line. information indicating an angle or information indicating a distance from a midpoint of the second target block of the specific boundary line. That is, based on the GP splitting flag (when the value of the splitting flag is 1), the second target block is not perpendicular to any boundary of the second target block, but specific across the second target block. It can be divided along the boundary line. The specific boundary line may include a straight line or a curve. Alternatively, when the value of the GP splitting flag is 1, that is, when the GP splitting flag indicates that the second target block is split into the second subblocks having different sizes, the additional splitting information is preset. It may include a GP partition index indicating one of the plurality of partition types.

If the value of the GP splitting flag is 0, that is, the GP splitting flag indicates that the second target block is not split into the second subblocks having different sizes, the additional splitting information is determined by the first splitting information. It may include a binary tree (BT) splitting flag for two target blocks. The BT split flag may indicate whether the second target block is divided into second subblocks having one of a 2NxN size and an Nx2N size when the size of the second target block is 2Nx2N. When the value of the BT partition flag is 1, the additional partition information may include a BT partition index for the second target block, and the BT partition index is one of the 2N × N size and the Nx2N size. It can be represented by the size of the sub-blocks.

As another example, the additional partition information may include a binary tree geometry partition (BTGP) partition flag for the second target block, and the BTGP partition flag may include the second sub block in which the second target block is non-square blocks. It can indicate whether or not divided into. When the value of the BTGP partition flag is 1, the additional partition information may include BTGP mode information for the second target block. The BTGP mode information indicates whether the second target block is divided into the second subblocks having one of 2NxN size and Nx2N size, or divided into the second subblocks having different sizes. Can be. If the BTGP mode information indicates that the second target block is partitioned into the second subblocks having one of a 2NxN size and an Nx2N size, the additional partition information indicates a BT partitioning index for the second target block. It may include. Further, when the BTGP mode information indicates that the second target block is divided into the second subblocks having different sizes, the additional division information may include information indicating an angle of a specific boundary line or information of the specific boundary line. Information indicating the distance from the midpoint of the second target block may be included. Alternatively, when the BTGP mode information indicates that the second target block is divided into the second subblocks having different sizes, the additional partition information indicates a GP partition index indicating one of a plurality of preset partition types. It may include.

On the other hand, the second sub-blocks may be divided into sub-blocks of a lower depth than the recursively through the BT structure or the GP structure. In this case, a BT partition flag for the second sub blocks may be generated and encoded, the BT partition index for the second sub blocks, information representing an angle of the specific boundary line, and the specific boundary line. Information indicating a distance from the midpoint of the second target block of, or the GP partition index, may be generated and encoded.

17 schematically illustrates a video decoding method by a decoding apparatus according to the present invention. The method disclosed in FIG. 17 may be performed by the decoding apparatus disclosed in FIG. 2. Specifically, for example, S1700 and S1720 of FIG. 17 may be performed by an entropy decoding unit of the decoding apparatus, S1710 and 1730 may be performed by a picture divider of the decoding apparatus, and S1740 may be the decoding apparatus. It may be performed by the prediction unit of.

The decoding apparatus obtains first split information about the first target block through the bitstream (S1700). The decoding apparatus may obtain the first partitioning information for the first target block through a bitstream. The first information may include a quad tree (QT) split flag for the first target block. The QT splitting flag may indicate whether a block is divided into sub-blocks having a size of half height and half width of the block.

If the first splitting information indicates that the first target block is split, the decoding apparatus splits the first target block into the first subblocks (S1710). When the QT splitting flag included in the first splitting information indicates that the first target block is split, the decoding apparatus may split the first target block into the first subblocks. For example, the first target block may be divided into four first subblocks, and the first subblock may have a half height and a half width of the first target block. It may be subblocks of size.

The decoding apparatus obtains second partitioning information and additional partitioning information on a second target block, which is one of the first subblocks of the first target block, through the bitstream (S1720). The decoding apparatus may obtain the second partitioning information and the additional partitioning information for the second target block through a bitstream. The second partitioning information may include a QT partitioning flag for the second target block. The additional splitting information may include: splitting the QT splitting flag for the second target block into subblocks having a size of half height and half width of the second target block; May be obtained in the case of not indicating. That is, the additional split information may be obtained when the second target block is not split based on the QT split flag for the second target block.

On the other hand, for example, the additional partition information may include a BT partition flag for the second target block. The BT split flag may indicate whether the second target block is divided into the second subblocks having one of 2NxN size and Nx2N size when the size of the second target block is 2Nx2N. When the value of the BT split flag is 1, that is, the BT split flag indicates that the second target block is split into the second sub blocks having one of 2NxN size and Nx2N size. The information may include a BT partition index for the second target block, and the BT partition index may indicate one of the 2NxN size and the Nx2N size as the sizes of the second subblocks.

In addition, when the value of the BT split flag is 0, that is, the BT split flag indicates that the second target block is not split into the second sub blocks having one of 2NxN size and Nx2N size. The additional partition information may include a geometry partition (GP) partition flag for the second target block, and the GP partition flag is partitioned into the second sub blocks having different sizes. It can indicate whether or not. When the value of the GP splitting flag is 1, that is, when the GP splitting flag indicates that the second target block is split into the second subblocks having different sizes, the additional splitting information is divided into a plurality of preset pieces. And a GP partition index indicating one of partition types, and the second target block may be partitioned into the second sub blocks based on the partition type indicated by the GP partition index. For example, the division types include (N / 4) x2N type, (N / 2) x2N type, Nx2N type, 2NxN type and 2Nx (N / 2), 2Nx (N / 4) type, for example. The GP partition index may indicate one of the types. When the GP partition index indicates an nNx2N type, the second target block may be divided into a second subblock of nNx2N size and a second subblock of (2-n) Nx2N size. Alternatively, when the GP partition index indicates a 2NxnN type, the second target block may be divided into a 2NxnN sized second subblock and a 2Nx (2-n) N sized second subblock.

Alternatively, when the value of the GP splitting flag is 1, that is, when the GP splitting flag indicates that the second target block is split into the second subblocks having different sizes, the additional splitting information is a specific boundary line. Information indicating an angle of or a distance from the midpoint of the second target block of the specific boundary line may be included. The specific boundary line may be derived based on the information representing the angle or the information representing the distance from the midpoint, and the second target block may be divided into the second sub blocks based on the specific boundary line.

Meanwhile, a BT splitting flag for the second subblocks may be transmitted, whereby the second subblocks may include the BT splitting index, information representing an angle of the specific boundary, and the first boundary of the specific boundary. The information may be divided into subblocks of a lower depth than the recursively based on the information indicating the distance from the midpoint of the target block, or the GP split index.

If the second target block is not divided based on the QT splitting flag for the second target block, the decoding apparatus divides the second target block into second subblocks based on the additional partitioning information ( S1730).

For example, the additional partition information may include a binary tree (BT) partition flag for the second target block. Specifically, for example, when the value of the BT partition flag is 1, the additional partition information may include a BT partition index for the second target block. When the BT partition index indicates the 2N × N size, the second target block may be divided into the second subblocks having the 2N × N size. In addition, when the BT partition index indicates the Nx2N size, the second target block may be divided into the second subblocks of the Nx2N size.

In addition, when the value of the BT partition flag is 0, the additional partition information may include a geometry partition (GP) partition flag for the second target block. When the value of the GP splitting flag is 1, the second target block may be divided into the second sub blocks, which are various square blocks. The second sub blocks may be non-square blocks of different sizes. When the value of the GP splitting flag is 0, the second target block may not be split.

Specifically, for example, when the value of the GP splitting flag is 1, the additional splitting information may include a GP splitting index indicating one of a plurality of preset splitting types, and the second target block may be used. It may be divided into the second subblocks based on the partition type indicated by the GP partition index. The division types may include, for example, (N / 4) x2N type, (N / 2) x2N type, Nx2N type, 2NxN type and 2Nx (N / 2), 2Nx (N / 4) type, The GP partition index may indicate one of the types. When the GP partition index indicates an nNx2N type, the second target block may be divided into a second subblock of nNx2N size and a second subblock of (2-n) Nx2N size. Alternatively, when the GP partition index indicates a 2NxnN type, the second target block may be divided into a 2NxnN sized second subblock and a 2Nx (2-n) N sized second subblock.

Also, for example, when the value of the GP splitting flag is 1, the additional splitting information indicates information indicating an angle of the specific boundary line or a distance from a midpoint of the second target block of the specific boundary line. May contain information. The specific boundary line may be derived based on the information representing the angle or the information representing the distance from the midpoint, and the second target block may be divided into the second sub blocks based on the specific boundary line.

As another example, the additional partition information may include a geometry partition (GP) partition flag for the second target block. If the value of the GP splitting flag is 1, that is, the GP splitting flag indicates that the second target block is split into the second subblocks having different sizes, the additional splitting information is an angle of a specific boundary line. information indicating an angle or information indicating a distance from a midpoint of the second target block of the specific boundary line. In this case, the specific boundary line may be derived based on the information representing the angle or the information representing the distance from the midpoint, and the second target block may be divided into the second sub-blocks based on the specific boundary line. Can be.

Alternatively, when the value of the GP splitting flag is 1, that is, when the GP splitting flag indicates that the second target block is split into the second subblocks having different sizes, the additional splitting information is preset. It may include a GP partition index indicating one of the plurality of partition types. The second target block may be divided into the second sub blocks based on the partition type indicated by the GP partition index.

Further, when the value of the GP split flag is 0, that is, when the GP split flag indicates that the second target block is not split into the second sub blocks having different sizes, the additional split information is It may include a binary tree (BT) partition flag for the second target block. When the value of the BT partition flag is 1, the additional partition information may include a BT partition index for the second target block. When the BT partition index indicates the 2N × N size, the second target block may be divided into the second subblocks having the 2N × N size. In addition, when the BT partition index indicates the Nx2N size, the second target block may be divided into the second subblocks of the Nx2N size.

As another example, the additional partition information may include a binary tree geometry partition (BTGP) partition flag for the second target block. When the value of the BTGP partition flag is 1, the additional partition information may include BTGP mode information for the second target block. If the BTGP mode information indicates that the second target block is partitioned into the second subblocks having one of a 2NxN size and an Nx2N size, the additional partition information indicates a BT partitioning index for the second target block. It may include. When the BT partition index indicates the 2N × N size, the second target block may be divided into the second subblocks having the 2N × N size. In addition, when the BT partition index indicates the Nx2N size, the second target block may be divided into the second subblocks of the Nx2N size.

Further, when the BTGP mode information indicates that the second target block is divided into the second subblocks having different sizes, the additional division information may include information indicating an angle of a specific boundary line or information of the specific boundary line. Information indicating the distance from the midpoint of the second target block may be included. In this case, the specific boundary line may be derived based on the information representing the angle or the information representing the distance from the midpoint, and the second target block may be divided into the second sub-blocks based on the specific boundary line. Can be.

Alternatively, when the BTGP mode information indicates that the second target block is divided into the second subblocks having different sizes, the additional partition information indicates a GP partition index indicating one of a plurality of preset partition types. It may include. The second target block may be divided into the second sub blocks based on the partition type indicated by the GP partition index.

The decoding device decodes the second subblocks (S1740). The decoding apparatus may decode the second subblocks. In detail, the decoding apparatus may generate an intra or inter prediction on the second sub block to generate a prediction sample of the second sub block, and based on the prediction sample, the second sub block may be generated. A reconstructed sample may be generated for the block, and a reconstructed picture may be generated based on the reconstructed sample.

Although not shown in the drawing, the decoding apparatus may directly use the prediction sample as a reconstruction sample according to a prediction mode, or generate a reconstruction sample by adding a residual sample to the prediction sample. When there is a residual sample for the target block, the decoding apparatus may receive information about the residual for the target block, and the information about the residual may be included in the information about the face. The information about the residual may include transform coefficients regarding the residual sample. The decoding apparatus may derive the residual sample (or residual sample array) for the target block based on the residual information. The decoding apparatus may generate a reconstructed sample based on the prediction sample and the residual sample, and may derive a reconstructed block or a reconstructed picture based on the reconstructed sample. Thereafter, as described above, the decoding apparatus may apply an in-loop filtering procedure, such as a deblocking filtering and / or SAO procedure, to the reconstructed picture in order to improve subjective / objective picture quality as necessary.

According to the present invention described above it is possible to divide the picture into various types of blocks, through which the prediction efficiency can be improved and the overall coding efficiency can be improved.

In addition, according to the present invention, a picture may be divided into various types of blocks, and accordingly, transform efficiency may be improved and overall coding efficiency may be improved.

In the above-described embodiment, the methods are described based on a flowchart as a series of steps or blocks, but the present invention is not limited to the order of steps, and any steps may occur in a different order or simultaneously from other steps as described above. have. In addition, those skilled in the art will appreciate that the steps shown in the flowcharts are not exclusive and that other steps may be included or one or more steps in the flowcharts may be deleted without affecting the scope of the present invention.

The above-described method according to the present invention may be implemented in software, and the encoding device and / or the decoding device according to the present invention may perform image processing of, for example, a TV, a computer, a smartphone, a set-top box, a display device, and the like. It can be included in the device.

When embodiments of the present invention are implemented in software, the above-described method may be implemented as a module (process, function, etc.) for performing the above-described function. The module may be stored in memory and executed by a processor. The memory may be internal or external to the processor and may be coupled to the processor by various well known means. The processor may include application-specific integrated circuits (ASICs), other chipsets, logic circuits, and / or data processing devices. The memory may include read-only memory (ROM), random access memory (RAM), flash memory, memory card, storage medium and / or other storage device.

Claims

In the video decoding method performed by the decoding apparatus,

Obtaining first split information for the first target block through the bitstream;

Dividing the first target block into the first sub blocks when the first division flag indicates that the first target block is to be divided;

Obtaining second splitting information and additional splitting information for a second target block which is one of the first subblocks of the first target block through the bitstream;

Dividing the second target block into second subblocks based on the additional partitioning information when the second target block is not divided based on the second partitioning information for the second target block; And

And decoding the second sub-blocks,

And the second sub blocks are non-square blocks.
The method of claim 1,

The additional partition information includes a binary tree (BT) partition flag for the second target block,

The BT split flag indicates whether the second target block is divided into the second subblocks having one of 2NxN size and Nx2N size when the size of the second target block is 2Nx2N size. Video decoding method.
The method of claim 2,

When the value of the BT partition flag is 1, the additional partition information includes a BT partition index for the second target block.

The BT split index indicates one of the 2NxN size and the Nx2N size as the size of the second subblocks.
The method of claim 2,

When the value of the BT partition flag is 0, the additional partition information includes a geometry partition (GP) partition flag for the second target block.

And the GP splitting flag indicates whether the second target block is divided into the second subblocks having different sizes.
The method of claim 4, wherein

When the value of the GP partition flag is 1, the additional partition information includes a GP partition index indicating one of a plurality of preset partition types,

And the second target block is divided into the second sub blocks based on a partition type indicated by the GP partition index.
The method of claim 1,

The additional partition information includes a geometry partition (GP) partition flag for the second target block,

And the GP splitting flag indicates whether the second target block is divided into the second subblocks having different sizes.
The method of claim 6,

When the value of the GP splitting flag is 1, the additional splitting information includes information indicating an angle of a specific boundary line or information indicating a distance from a midpoint of the second target block of the specific boundary line,

The specific boundary line is derived based on the information representing the angle or the information representing the distance from the midpoint,

And the second target block is divided into the second sub-blocks based on the specific boundary line.
The method of claim 6,

When the value of the GP partition flag is 1, the additional partition information includes a GP partition index indicating one of a plurality of preset partition types,

And the second target block is divided into the second sub blocks based on a partition type indicated by the GP partition index.
The method of claim 6,

When the value of the GP partition flag is 0, the additional partition information includes a binary tree (BT) partition flag for the second target block.

The BT split flag indicates whether the second target block is divided into second subblocks having one of a 2NxN size and an Nx2N size when the size of the second target block is 2Nx2N. Video decoding method.
The method of claim 9,

When the value of the BT partition flag is 1, the additional partition information includes a BT partition index for the second target block.

The BT split index indicates one of the 2NxN size and the Nx2N size as the size of the second subblocks.
The method of claim 1,

The additional partition information includes a binary tree geometry partition (BTGP) partition flag for the second target block,

The BTGP split flag indicates whether the second target block is split into the second sub blocks, which are non-square blocks.
The method of claim 11,

When the value of the BTGP partition flag is 1, the additional partition information includes BTGP mode information for the second target block.

The BTGP mode information indicates whether the second target block is divided into the second subblocks having one of 2NxN size and Nx2N size, or divided into the second subblocks having different sizes. Image decoding method characterized in that.
In the decoding apparatus for performing image decoding,

Obtaining first split information about the first target block through the bitstream, and receiving the second split information and the additional split information on the second target block which is one of the first sub-blocks of the first target block. An entropy decoding unit;

When the first partitioning information for the first target block indicates that the first target block is divided, the first target block is divided into the first subblocks, and the second target block is divided into the second subblocks. A picture dividing unit dividing the second target block into second sub-blocks based on the additional splitting information when the splitting information indicates that the second target block is not divided; And

A prediction unit for decoding the second sub-blocks,

And the second sub blocks are non-square blocks.
The method of claim 13,

The additional partition information includes a binary tree (BT) partition flag for the second target block,

The BT split flag indicates whether the second target block is divided into the second subblocks having one of 2NxN size and Nx2N size when the size of the second target block is 2Nx2N size. Decoding device.
The method of claim 13,

The additional partition information includes a geometry partition (GP) partition flag for the second target block,

And the GP splitting flag indicates whether the second target block is divided into the second sub blocks having different sizes.