KR20120137305A - Methods of spliting block and apparatuses for using the same - Google Patents

Abstract

PURPOSE: A block division method and an apparatus using the method are provided to increase the efficiency of encoding by encoding a division structure. CONSTITUTION: An apparatus induces an information about determination of division structure candidate in a reference block(S1800). The apparatus determines whether a division target encoding unit satisfies a predetermined condition based on the induced segmentation structure candidate determining information(S1810). The apparatus transmits the division information through bit stream(S1820). [Reference numerals] (AA) Start; (BB) End; (S1800) Inducing division structure candidate determination information in reference block; (S1810) Determining whether division target encoding unit satisfies predetermined condition based on division structure candidate determination information; (S1820) Storing and transmitting division information in bit stream by encoding same

Description

METHODS OF SPLITING BLOCK AND APPARATUSES FOR USING THE SAME}

The present invention relates to a block division method and an apparatus using the method, and more particularly, to an image encoding / decoding method and an image encoding / decoding device.

Recently, the demand for high resolution and high quality images such as high definition (HD) image and ultra high definition (UHD) image is increasing in various applications. As the video data becomes higher resolution and higher quality, the amount of data increases relative to the existing video data. Therefore, when the video data is transmitted or stored using a medium such as a conventional wired / wireless broadband line, The storage cost will increase. High-efficiency image compression techniques can be utilized to solve such problems as image data becomes high-resolution and high-quality.

An inter picture prediction technique for predicting a pixel value included in a current picture from a previous or a subsequent picture of a current picture using an image compression technique, an intra picture prediction technique for predicting a pixel value included in a current picture using pixel information in the current picture, There are various techniques such as an entropy encoding technique in which a short code is assigned to a value having a high appearance frequency and a long code is assigned to a value having a low appearance frequency. Image data can be effectively compressed and transmitted or stored using such an image compression technique.

An object of the present invention is to provide a block partitioning method for increasing image encoding efficiency.

In addition, another object of the present invention is to provide an apparatus for performing a block partitioning method for increasing image encoding efficiency.

According to an aspect of the present invention, there is provided a method of encoding a partition structure candidate decision information from a reference block, the derived partition structure candidate decision information, and information derived from a split target coding unit. The method may include determining whether to use the split target coding unit as an LCU partition structure candidate using a predetermined condition calculated on the basis of. The predetermined condition includes a split depth of a block included in a call LCU, motion information of a prediction unit included in a call LCU, reference index information of coding units in a call LCU, and transform coefficient information of coding units in a call LCU. It may be a condition for determining whether to use the split target coding unit as an LCU partition structure candidate using at least one information. The predetermined condition is whether a split depth of a coding target to be split is 0, whether a coding unit having the largest split depth exists in a call LCU, and 10 of x and y components of a motion vector of a coding unit included in a call LCU. It may be at least one condition of whether a larger value exists or whether a value larger than 10 among x and y components of a motion vector difference of a coding unit included in a call LCU exists. Determining whether to use the split target coding unit as the LCU split structure candidate using the predetermined condition calculated based on the derived split structure candidate determination information and the information derived from the split target coding unit may be performed. Determining whether the split structure candidate determination information and the information derived from the split target coding unit satisfy a first condition and the derived split structure candidate determination information and the information derived from the split target coding unit satisfy a second condition And determining whether the first condition and the second condition are generated based on the partition depth information of the split target coding unit and the partition structure candidate determination information derived from the reference block.

According to an aspect of the present invention, there is provided a method of encoding a partition structure candidate decision information from an adjacent block, and extracting the partition structure candidate decision information and the information derived from the split target coding unit. The method may include determining whether to use the split target coding unit as an LCU partition structure candidate using a predetermined condition calculated as a basis. The predetermined condition is LCU partitioning the split target coding unit using at least one of depth information of an adjacent block, motion information of an adjacent block, reference index information of an adjacent block, and transform coefficient information of an adjacent block. It may be a condition for determining whether to use as a structure candidate.

In accordance with an aspect of the present invention, there is provided a method of encoding an image, the method comprising: deriving partition structure candidate determination information from a reference block, partition structure candidate determination information derived from the reference block, and a target to be split Determining whether to use the split target coding unit as an LCU partition structure candidate using a first condition calculated based on information derived from the step of deriving partition structure candidate determination information from a neighboring block and the neighboring block; And determining whether to use the split target coding unit as the LCU split structure candidate using the second condition calculated based on the split structure candidate determination information derived from and the information derived from the split target coding unit. have. The first condition includes a split depth of a block included in a call LCU, motion information of a prediction unit included in a call LCU, reference index information of coding units in a call LCU, and transform coefficient information of coding units in a call LCU. It may be a condition for determining whether to use the split target coding unit as an LCU partition structure candidate using at least one information. The second condition is LCU partitioning the split target coding unit using at least one of depth information of an adjacent block, motion information of an adjacent block, reference index information of an adjacent block, and transform coefficient information of an adjacent block. It may be a condition for determining whether to use as a structure candidate.

According to an aspect of the present invention, there is provided a method of decoding a partition structure candidate decision information from a reference block, the derived partition structure candidate decision information, and information derived from a split target coding unit. The method may include determining whether to decode the split information of the split target coding unit using a predetermined condition calculated based on. The predetermined condition includes a split depth of a block included in a call LCU, motion information of a prediction unit included in a call LCU, reference index information of coding units in a call LCU, and transform coefficient information of coding units in a call LCU. It may be a condition for determining whether to use the split target coding unit as an LCU partition structure candidate using at least one information.

According to an aspect of the present invention, there is provided a decoding method for deriving partition structure candidate determination information from an adjacent block, derived partition structure candidate determination information, and information derived from a splitting target coding unit. The method may include determining whether to decode the split information of the split target coding unit by using a predetermined condition calculated as a basis. The predetermined condition is LCU partitioning the split target coding unit using at least one of depth information of an adjacent block, motion information of an adjacent block, reference index information of an adjacent block, and transform coefficient information of an adjacent block. It may be a condition for determining whether to use as a structure candidate.

A decoding method according to an aspect of the present invention for achieving the above object of the present invention comprises deriving the partition structure candidate determination information from the reference block derived from the partition structure candidate determination information and the split target coding unit derived from the reference block Determining whether to decode the split information of the split target coding unit using a first condition calculated based on the obtained information, deriving split structure candidate determination information from an adjacent block, and splitting derived from the adjacent block. The method may include determining whether to decode the split information of the split target coding unit using a second condition calculated based on the structure candidate determination information and the information derived from the split target coding unit. The first condition includes a split depth of a block included in a call LCU, motion information of a prediction unit included in a call LCU, reference index information of coding units in a call LCU, and transform coefficient information of coding units in a call LCU. It may be a condition for determining whether to use the split target coding unit as an LCU partition structure candidate using at least one information. The second condition is LCU partitioning the split target coding unit using at least one of depth information of an adjacent block, motion information of an adjacent block, reference index information of an adjacent block, and transform coefficient information of an adjacent block. It may be a condition for determining whether to use as a structure candidate.

According to an aspect of the present invention, an encoding apparatus according to an aspect of the present invention receives partition structure candidate determination information received from a reference block or an adjacent block and derives the partition structure candidate determination information and information derived from a splitting target coding unit. A determination unit that determines whether to use the split target coding unit as an LCU partition structure candidate using a predetermined condition calculated based on the LCU division, and an LCU partition that determines an LCU partition structure candidate based on a result determined by the determination unit It may include a structure candidate determiner. The predetermined condition may include a partition depth of a block included in a call LCU, motion information of a prediction unit included in a call LCU, reference index information of coding units in a call LCU, transform coefficient information of coding units in a call LCU, or At least one of the depth information of the neighboring block, the motion information of the neighboring block, the reference index (Reference Index) information of the neighboring block, the transform coefficient information of the neighboring block may be at least one condition.

In accordance with an aspect of the present invention, a decoding apparatus according to an aspect of the present invention receives partition structure candidate decision information from a reference block or an adjacent block and derives the partition structure candidate decision information and information derived from a splitting target coding unit. A determination unit that determines whether to decode the split information of the split target coding unit using a predetermined condition calculated based on the C; and a CU partition information decoder that decodes the LCU partition structure based on the result determined by the determination unit. It may include. The predetermined condition may include a partition depth of a block included in a call LCU, motion information of a prediction unit included in a call LCU, reference index information of coding units in a call LCU, transform coefficient information of coding units in a call LCU, or At least one of the depth information of the neighboring block, the motion information of the neighboring block, the reference index (Reference Index) information of the neighboring block, the transform coefficient information of the neighboring block may be at least one condition.

As described above, according to the block division method and the apparatus using the method in the embodiment of the present invention, encoding efficiency is increased by performing encoding only on a partition structure that may be selected as an optimal partition structure in partitioning a block. Can be.

1 is a block diagram illustrating an image encoding apparatus according to an embodiment of the present invention.
2 is a block diagram illustrating a configuration of an image decoding apparatus according to an embodiment of the present invention.
3 is a conceptual diagram illustrating a splitting method of coding units according to an embodiment of the present invention.
4 and 5 are conceptual views illustrating splitting of a 64x64 coding unit according to an embodiment of the present invention.
6 is a conceptual diagram illustrating an LCU partitioning structure according to an embodiment of the present invention.
7 is a result of encoding using random access and low delay as a test condition using reference software according to an embodiment of the present invention.
8 to 11 illustrate information of coding units (intra prediction direction information (IntraDir), split information, skip information, skip, motion merge information (merge), coding mode information (Predic), and block division. A graph showing distribution of quantization parameters of information (Part size), motion prediction information (AMVP), motion difference information (MVD), and prediction direction information (Dir).
12 is a conceptual diagram illustrating a reference block according to an embodiment of the present invention.
13 is a flowchart illustrating a method of generating LCU partition structure information based on reference block information according to an embodiment of the present invention.
14 is a conceptual diagram illustrating adjacent coding units for deriving partition structure candidate determination information in an LCU partition structure candidate determination method according to an embodiment of the present invention.
15 is a flowchart illustrating a method of generating LCU partition structure information based on reference block information according to an embodiment of the present invention.
16 is a conceptual diagram illustrating a reference block and an adjacent block according to an embodiment of the present invention.
17 is a flowchart illustrating a method of generating LCU partition structure information based on reference block information according to an embodiment of the present invention.
18 is a flowchart illustrating a method of encoding split information according to an embodiment of the present invention.
19 is a flowchart illustrating a method of decoding split information according to an embodiment of the present invention.
20 is a flowchart illustrating a method of encoding split information according to an embodiment of the present invention.
21 is a flowchart illustrating a method of decoding partition information according to an embodiment of the present invention.
22 is a flowchart illustrating a method of encoding split information according to an embodiment of the present invention.
23 is a flowchart illustrating a method of decoding partition information according to an embodiment of the present invention.
24 is a conceptual diagram illustrating a part of an image encoding apparatus according to an embodiment of the present invention.
25 is a conceptual diagram illustrating a part of an image encoding apparatus according to an embodiment of the present invention.
26 is a conceptual diagram illustrating a part of an image encoding apparatus according to an embodiment of the present invention.
27 is a conceptual diagram illustrating a part of an image encoding apparatus according to an embodiment of the present invention.
28 is a conceptual diagram illustrating a part of an image decoding apparatus according to an embodiment of the present invention.
29 is a conceptual diagram illustrating a part of an image encoding apparatus according to an embodiment of the present invention.
30 is a conceptual diagram illustrating a part of an image decoding apparatus according to an embodiment of the present invention.
31 is a conceptual diagram illustrating a part of an image encoding apparatus according to an embodiment of the present invention.
32 is a conceptual diagram illustrating a part of an image decoding apparatus according to an embodiment of the present invention.
33 is a flowchart illustrating a method of determining an LCU partition structure candidate according to an embodiment of the present invention.
34 is a flowchart illustrating a method of determining an LCU partition structure candidate according to an embodiment of the present invention.
35 is a flowchart illustrating a method of determining an LCU partition structure candidate according to an embodiment of the present invention.
36 is a flowchart illustrating a method of determining an LCU partition structure candidate according to an embodiment of the present invention.
37 is a conceptual diagram illustrating an image encoder according to an embodiment of the present invention.
38 is a conceptual diagram illustrating an image encoder according to an embodiment of the present invention.
39 is a conceptual diagram illustrating an image encoder according to an embodiment of the present invention.
40 is a conceptual diagram illustrating an image encoder according to an embodiment of the present invention.
41 is a conceptual diagram illustrating an image decoder according to an embodiment of the present invention.
42 is a conceptual diagram illustrating an image encoder according to an embodiment of the present invention.
43 is a conceptual diagram illustrating an image decoder according to an embodiment of the present invention.
44 is a conceptual diagram illustrating an image encoder according to an embodiment of the present invention.
45 is a conceptual diagram illustrating an image decoder according to an embodiment of the present invention.

Hereinafter, embodiments of the present invention will be described in detail with reference to the drawings. In the following description of the embodiments of the present invention, a detailed description of known functions and configurations incorporated herein will be omitted when it may make the subject matter of the present disclosure rather unclear.

It is to be understood that when an element is referred to as being "connected" or "connected" to another element, it may be directly connected or connected to the other element, . In addition, the description of "including" a specific configuration in the present invention does not exclude a configuration other than the configuration, and means that additional configurations can be included in the practice of the present invention or the technical scope of the present invention.

The terms first, second, etc. may be used to describe various components, but the components should not be limited by the terms. The terms are used only for the purpose of distinguishing one component from another. For example, without departing from the scope of the present invention, the first component may be referred to as a second component, and similarly, the second component may also be referred to as a first component.

In addition, the components shown in the embodiments of the present invention are shown independently to represent different characteristic functions, which does not mean that each component is composed of separate hardware or software constituent units. That is, each constituent unit is included in each constituent unit for convenience of explanation, and at least two constituent units of the constituent units may be combined to form one constituent unit, or one constituent unit may be divided into a plurality of constituent units to perform a function. The integrated embodiments and separate embodiments of the components are also included within the scope of the present invention, unless they depart from the essence of the present invention.

In addition, some of the components are not essential components to perform essential functions in the present invention, but may be optional components only to improve performance. The present invention can be implemented only with components essential for realizing the essence of the present invention, except for the components used for the performance improvement, and can be implemented by only including the essential components except the optional components used for performance improvement Are also included in the scope of the present invention.

1 is a block diagram illustrating an image encoding apparatus according to an embodiment of the present invention.

Referring to FIG. 1, the image encoding apparatus 100 may include a motion predictor 111, a motion compensator 112, an intra predictor 120, a switch 115, a subtractor 125, and a converter 130. And a quantization unit 140, an entropy encoding unit 150, an inverse quantization unit 160, an inverse transform unit 170, an adder 175, a filter unit 180, and a reference image buffer 190.

The image encoding apparatus 100 performs encoding in an intra mode or an inter mode with respect to an input image and outputs a bit stream. In the embodiment of the present invention, intra prediction can be used in the same way as inter prediction, and inter prediction can be used in the same meaning as inter prediction. In order to determine an optimal prediction method for the prediction unit, an intra prediction method and an inter prediction method may be selectively used for the prediction unit. The image encoding apparatus 100 generates a prediction block for the original block of the input image, and then encodes the difference between the original block and the prediction block.

In the intra prediction mode, the intra prediction unit 120 (or the intra prediction unit may be used as a term having the same meaning) may perform spatial prediction using pixel values of an already encoded block around the current block. Generate a predictive block.

In the inter prediction mode, the motion predictor 111 finds a motion vector in the reference image stored in the reference image buffer 190 that best matches the input block in the motion prediction process. The motion compensation unit 112 generates a prediction block by performing motion compensation using a motion vector.

The subtracter 125 generates a residual block by a difference between the input block and the generated prediction block. The transforming unit 130 performs a transform on the residual block to output a transform coefficient. The quantization unit 140 quantizes the input transform coefficient according to the quantization parameter and outputs a quantized coefficient. The entropy encoding unit 150 entropy-codes the input quantized coefficients according to a probability distribution to output a bit stream.

Since the HEVC performs inter prediction coding, i.e., inter prediction coding, the currently encoded image needs to be decoded and stored for use as a reference image. Accordingly, the quantized coefficients are inversely quantized in the inverse quantization unit 160 and inversely transformed in the inverse transformation unit 170. The inverse quantized and inverse transformed coefficients are added to the prediction block through the adder 175 and a reconstruction block is generated.

The reconstruction block passes through the filter unit 180, and the filter unit 180 applies at least one or more of a deblocking filter, a sample adaptive offset (SAO), and an adaptive loop filter (ALF) to the reconstruction block or the reconstruction picture. can do. The filter unit 180 may be referred to as an adaptive in-loop filter. The deblocking filter can remove block distortion occurring at the boundary between the blocks. SAO may add an appropriate offset value to pixel values to compensate for coding errors. The ALF may perform filtering based on a value obtained by comparing the reconstructed image with the original image, and may be performed only when high efficiency is applied. The reconstructed block that has passed through the filter unit 180 is stored in the reference image buffer 190.

2 is a block diagram illustrating a configuration of an image decoding apparatus according to another embodiment of the present invention.

2, the image decoding apparatus 200 includes an entropy decoding unit 210, an inverse quantization unit 220, an inverse transform unit 230, an intra prediction unit 240, a motion compensation unit 250, an adder 255, a filter unit 260, and a reference image buffer 270.

The video decoding apparatus 200 receives the bit stream output from the encoder and decodes the video stream into an intra mode or an inter mode, and outputs a reconstructed video, i.e., a reconstructed video. In the intra mode, the prediction block is generated using the intra prediction mode, and in the inter mode, the prediction block is generated using the inter prediction method. The image decoding apparatus 200 obtains a residual block from the input bitstream, generates a prediction block, adds the residual block and the prediction block, and generates a reconstructed block, that is, a reconstruction block.

The entropy decoding unit 210 entropy-decodes the input bitstream according to a probability distribution and outputs a quantized coefficient. The quantized coefficients are inversely quantized in the inverse quantization unit 220 and inversely transformed in the inverse transformation unit 230. As a result of inverse quantization / inverse transformation of the quantized coefficients, a residual block is generated.

In the intra prediction mode, the intra prediction unit 240 (or the intra prediction unit) generates a prediction block by performing spatial prediction using pixel values of blocks that are already encoded around the current block.

In the inter prediction mode, the motion compensator 250 generates a predictive block by performing motion compensation using the motion vector and the reference image stored in the reference image buffer 270.

The residual block and the prediction block are added through the adder 255, and the added block is passed through the filter unit 260. [ The filter unit 260 may apply at least one or more of the deblocking filter, SAO, and ALF to the reconstructed block or the reconstructed picture. The filter unit 260 outputs a reconstructed image, that is, a reconstructed image. The reconstructed picture may be stored in the reference picture buffer 270 to be used for inter prediction.

Methods for improving the prediction performance of the encoding / decoding apparatus include a method of increasing the accuracy of the interpolation image and a method of predicting the difference signal. Here, the difference signal is a signal indicating the difference between the original image and the predicted image. In the present invention, the term " difference signal " may be replaced by a " difference signal ", " residual block ", or " difference block " depending on the context. Those skilled in the art may influence the idea You will be able to distinguish this within the scope of not giving.

As described above, in the embodiment of the present invention, a coding unit (coding unit) is used as a coding unit for convenience of explanation, but it may be a unit for performing not only coding but also decoding.

In the exemplary embodiment of the present invention, the image encoding method and the image decoding method to be described later may be performed by each component included in the image encoder and the image decoder described above with reference to FIGS. 1 and 2. The meaning of the constituent part may include not only a hardware meaning but also a software processing unit which can be performed through an algorithm.

3 is a conceptual diagram illustrating a splitting method of coding units according to an embodiment of the present invention.

Referring to FIG. 3, in order to encode an image, the image is divided into predefined coding units, and then encoding is performed by dividing the image into smaller coding units based on the coding unit split into the predefined units. A predefined coding unit for splitting is called a large gesture coding unit (LCU). Every time a split is performed, starting with the LCU. Four coding units in which the horizontal length and the vertical length of the coding unit are reduced by half may be generated. When one division is performed based on the LCU, four coding units are generated, and a block that performs additional division among the four coding units generated is a block having a horizontal length and a vertical length of 1/4 based on the LCU. Can be divided.

The split depth is the number of times the split is performed based on the LCU, and the split depth value may be increased by 1 each time the split is performed once based on the LCU. For example, if the partition is divided once based on the LCU, the partition depth is 1, and if the partition is further divided, the partition depth is 2. That is, as the division is additionally performed, the division depth may increase.

How deep the coding unit is divided may vary depending on the characteristics of the image. For example, a complex image may be divided into coding units having a relatively larger depth than that of an uncomplicated image since there are many changes in the LCU.

How much division is performed depends on which division depth is the most ideal division in terms of compression efficiency and image quality. For example, when the partition is partitioned to a larger depth than the optimal partition, it may be better in terms of image quality, but encoding efficiency may be degraded because bits used for encoding may increase. On the contrary, in case of dividing with a smaller depth of division than the optimal division, the image quality may be lower than that of dividing into a larger depth, but the coding efficiency may be relatively increased. Therefore, it is important to determine the appropriate segmentation depth in terms of compression efficiency and image quality according to the characteristics of the image.

Information on whether to split the coding unit may be expressed through split information. This splitting information may be present in the remaining coding units except for the coding unit of the smallest unit (coding unit having the largest division depth) that can no longer be split. For example, when split information of a coding unit is 0, hierarchical splitting may be performed by four coding units that are smaller in half horizontally and vertically when split information of a coding unit is 1 without splitting the coding unit. As described above, whenever the splitting of the coding unit occurs, the splitting depth is increased, and the horizontal length and the vertical length of the coding unit having the same splitting depth are the same.

The maximum value and the minimum value of the division depth that the division depth may have may be defined in advance, and the coding unit may not be divided beyond the maximum value of the division depth by dividing the coding unit starting from the minimum value. In the case of FIG. 3, the LCU has a size of 128 × 128 and the splitting depth is from the minimum depth of 0 to the maximum depth of 4.

Referring to the upper part of FIG. 3, the uppermost case is N = 64 and the split depth is 0. On the left side, a split flag is 0, and a coding unit having a size of 128x128 and the right side are split. The case in which the information (Spilt Flag) is 1 is divided into four coding units having a 64x64 size. That is, each of the divided coding units has a split depth value of 1.

3 shows a 64x64 partition unit having a partition depth value of N = 32 and has the same size as the partitioned coding unit at the top of FIG. 3. The coding unit on the left side of FIG. 3 represents a case where split information is 0, and no further splitting occurs at a split depth 1, and the coding unit on the right side of the break corresponds to split information. In this case, additional splitting occurs at split depth 1 to form a coding unit having split depth 2.

The lower part of FIG. 3 shows an 8x8 sized division unit having N = 4 having a division depth value of 4, and shows an additional three divisions in the divided coding units of FIG. 3. When the maximum value of the split depth is predefined as 4, when the split depth becomes 4 and the size of the coding unit becomes 8x8, encoding may be performed without using additional flag information for determining whether to split additionally. .

4 and 5 are conceptual views illustrating splitting of a 64x64 coding unit according to an embodiment of the present invention.

4 and 5, information in which a 64x64 coding unit is divided and the coding unit is divided may be represented through a tree structure. In the case of a 64x64 coding unit, if the split depth is 0, for example, since the 64x64 coding unit is divided into four 32x32 coding units, the split information may be 1 in the 64x64 coding unit. If the coding unit divided in the z scan order is a first coding unit, a second coding unit, a third coding unit, or a fourth coding unit, the first, third, and fourth coding units are not divided, but the second coding unit 400 is used. ) Is further divided to one more division depth. In the tree structure, split information at split depth 1 may be represented as 0100 to indicate split information on coding units. For example, when the maximum split depth is 2, since information on whether to be additionally split in the second coding unit 400 does not need to be encoded, it may be encoded without including split information. However, when the maximum split depth is greater than 2, the second coding unit 400 may not be split further, and the split information at split depth 2 of the second coding unit may be expressed as 0000.

As described above, when encoding the coding unit, hierarchical division is performed based on the LCU unit. The LCU size and the maximum depth value may be important information for determining the number of times of storing segmentation information and the size of a coding unit in compressing an image.

6 is a conceptual diagram illustrating an LCU partitioning structure according to an embodiment of the present invention.

Referring to FIG. 6, it is assumed that encoding is performed from a 64x64 sized coding unit having a minimum depth of 0 to a coding unit having a maximum depth of 8x8 size. The LCU splitting structure is a structural characteristic of the LCU determined according to whether coding units existing for each depth in the LCU are divided. As shown in FIG. 6, LCU splitting structures having various shapes may exist in LCU units depending on whether coding units of each depth are split. Can be. The LCU splitting structure is split information of a coding unit. After generating various LCU partition structures and storing them in the LCU partition structure candidate as described above, one of the LCU partition structure candidates is selected as the optimal LCU partition structure in LCU units in the step of determining an optimal LCU partition structure. . By using such a method, encoding is performed based on an LCU partitioning structure that is adaptive to the characteristics of an image in units of LCUs, and thus, encoding may be efficiently performed in terms of encoding efficiency and image quality.

In the encoder of HEVC, a splitting method of coding units in an LCU may be determined differently according to characteristics of an image. That is, in order to determine the optimal LCU partitioning structure, the partitioning structure in various cases can be encoded. As a method for determining the optimal LCU partitioning structure, the prediction mode (Intra, Inter mode, etc.) of all coding units at each partitioning depth. After encoding using, a method of determining a prediction mode of a corresponding coding unit may be used according to the amount of bits to be encoded and the image quality. For example, as shown in FIG. 6, after encoding each of the 64x64 coding units having a depth of 0 in the Intra mode and the Inter mode, the optimal mode is stored, and each 64x64 size coding unit is divided into 4 pieces to encode each 32x32 size. In units of Intra mode and Inter mode, encoding may be performed recursively. In this case, each of the four coding units having a size of 32x32 may be independently selected as a prediction mode. In addition, even within a 32x32 coding unit, encoding is performed by dividing into four 16x16 coding units. After coding is performed by dividing the coding unit in a recursive manner, the partition structure of the coding unit that is most efficient in terms of bit quantity and quality is determined. For example, when the bit amount and the image quality of four encoded 32x32 coding units are more efficient than those encoded by 64x64 coding units, it is determined that the coding units are divided into four 32x32 coding units. When encoding an image in an encoder (image encoder), an optimal coding unit distribution is found for all cases in which the LCU can be divided, which increases the computational complexity of the encoder. For efficient compression performance, the computational complexity increases as the coding efficiency is determined for a large number of cases.

7 is a result of encoding using random access and low delay as a test condition using reference software according to an embodiment of the present invention.

As an experimental condition, the largest coding unit was set to 128x128 and the comparison target was coded by setting the largest coding unit to 64x64. When the size of the largest coding unit is set to 128x128, encoding is performed from the 128x128 coding unit to the 8x8 coding unit, and the comparison target encodes everything from the 64x64 coding unit to the 8x8 coding unit. It was. As can be seen from the table, when the 64x64 coding unit is the largest coding unit, the coding efficiency (BD-Rate) is deteriorated, but the complexity of the encoder is reduced to about 60%. The difference in computational complexity is the complexity required to encode a 128x128 coding unit. Therefore, when encoding is performed using a 128x128 coding unit, the coding efficiency is improved, but the computational complexity is increased. As described above, in order to increase coding efficiency, a computational complexity generally increases. However, when the computational complexity increases a lot compared to the rate at which the coding efficiency increases, this needs to be improved. Therefore, there is a need for a method of reducing computational complexity without lowering coding efficiency.

As described above, it is understood that determining whether to encode a specific coding unit when encoding is performed in the coding unit affects computation complexity and bit amount. When encoding is performed in coding units, the encoder determines whether to perform encoding by dividing the coding unit and transmits the encoding to a decoder (video decoder), which is transmitted through the split information. This information is closely related to the divided information of the coding unit and is an important factor in determining the bit amount of the bitstream. The following figure shows the distribution of information in a coding unit in a bitstream for each quantization parameter (QP).

8 to 11 illustrate information of coding units (intra prediction direction information (IntraDir), split information, skip information, skip, motion merge information (merge), coding mode information (Predic), and block division. A graph showing distribution of quantization parameters of information (Part size), motion prediction information (AMVP), motion difference information (MVD), and prediction direction information (Dir). In particular, in the case of encoding the coding unit information, it can be seen that the split information, which is information necessary for splitting the coding unit, occupies a large portion in the bitstream. In addition, as the size of the quantization parameter increases, the weight of encoding transform coefficients decreases, so that the weight of split information increases. Therefore, when partitioning is limited to reduce computational complexity, the bit amount may increase or decrease by affecting partitioning information.

In the image encoding method, encoding and decoding are performed in coding units. When an encoding unit is to be encoded, the encoder may determine whether to perform encoding by the size of the coding unit or to encode the size of the coding unit in a smaller unit. In the encoder, for example, all modes are encoded in coding units existing at all division depths, and then the sizes and modes of coding units that are most efficient in terms of coding efficiency and quality are determined. In determining the size of the coding unit, information about whether to split the coding unit is stored in the split information, and the information is transmitted to the decoder through the bitstream. The decoder determines the split structure of the coding unit through the split information of the coding unit through the corresponding information. However, as described above, a method of encoding coding units of all depths may increase computational complexity.

According to an embodiment of the present invention, in order to solve the problem of increasing computational complexity in determining a partitioning structure, in generating an LCU partitioning structure for a coding unit, information of a coding unit and coding units adjacent to each other within a same frame and a reference frame exist. A coding unit having a low probability of being determined as the final LCU split structure when generating an LCU split structure candidate by predicting characteristics of the coding unit through information of a coding unit or a prediction structure of a frame unit may not be included in the LCU split structure candidate. . By using this method, the encoder can reduce computational complexity by not performing encoding when a coding unit satisfies a specific condition, and improve compression efficiency by not sending segmentation information in a specific case. In the case of not only a block for performing inter prediction but also a block for performing only intra prediction, whether to split may be predicted with reference to the call LCU and the split information of the call CU.

Hereinafter, according to an exemplary embodiment of the present invention, a coding unit that is to be determined whether to split is a coding unit to be split, a call LCU existing in a reference frame or a call CU to a neighboring coding unit while being in the same frame as the reference block and the splitting coding unit. Is defined as the term of adjacent coding units.

In the present invention, the information of the coding unit is encoded at the time of high efficiency video encoding.

(1) It is determined whether a coding target coding unit is included in the LCU partition structure candidate through coding unit information (reference block information) corresponding to the current coding unit in the reference frame or coding unit information (adjacent block information) adjacent in the same frame. How to

(2) Whether the encoding target coding unit is included in the LCU partition structure candidate through the coding unit information (reference block information) corresponding to the encoding target coding unit in the reference frame or the coding unit information (adjacent block information) adjacent to the same frame. Disclosed is a method for determining whether to encode and decode split information of a coding unit.

According to an embodiment of the present invention, the splitting target coding unit is assigned to the LCU partition structure candidate through coding unit information (reference block information) corresponding to the splitting target coding unit in the reference frame or coding unit information (adjacent block information) adjacent to the same frame. The method of determining whether to include may be performed as follows. The LCU partition structure candidate refers to a candidate that can be determined as an optimal LCU partition structure. That is, when a specific LCU partition structure does not correspond to the LCU partition structure, since the coding does not need to be performed on the LCU partition structure, the computational complexity may be relatively lower than that when encoding is performed on all LCU partition structures.

In the block partitioning method according to the embodiment of the present invention, the encoder may determine the optimal LCU partitioning structure without performing encoding for all the partition depths in order to find the optimal LCU partitioning structure. For example, the encoder may determine whether to store the split target coding unit in the LCU split structure candidate by predicting a split characteristic of the split target coding unit based on the information of the coding unit corresponding to the split target coding unit in the reference frame. Here, the reference frame is a frame that has been coded before the current frame and is a frame used for inter prediction of a block included in the current frame. The coding unit corresponding to the current coding unit in the reference frame may be as shown in FIG. 12. A block present in the reference frame for determining whether the split target coding unit is included in the LCU split structure candidate is a collocated LCU (hereinafter referred to as a call LCU) that exists at the same position on the reference frame based on the position of the split target coding unit. ) Or a collocated block (call block), and such a block will be defined and used as a reference block in the following embodiments of the present invention.

When the call LCU assumes that the point located at the upper left or the center of the encoding target coding unit is (xP, yP), the LCU including the pixel existing at the position (xP, yP) in the reference picture, the call block is the encoding target encoding When a point located on the upper left of the unit is assumed to be (xP, yP), it may indicate a coding unit including a pixel existing at a position (xP, yP) in the reference picture.

In general, the encoder encodes a coding unit having the smallest division depth in order to encode a coding unit for all division depths, and stores the encoding unit as an LCU partition structure candidate. The coding unit may be divided into four coding units, the depth of which is increased by 1, and each coding unit divided into four units may also be encoded and stored in the LCU partition structure candidate. That is, these processes may be recursively performed until the maximum division depth is reached, and the coding units split and performed may be stored in the LCU partition structure candidate. By repeating the above process, among the stored LCU partition structure candidates, the most efficient LCU partition structure candidate may be selected in terms of encoding efficiency and quality. However, in the block partitioning method according to an embodiment of the present invention, only a part of a block partitioning structure that satisfies a predetermined condition is selected as an LCU partitioning structure candidate without performing encoding for all partitioning depths, thereby calculating the computational complexity that occurs in encoding. Can be reduced.

In the above-described block partitioning method, even in the case of not only a block for performing inter prediction but also a block for performing only intra prediction, it is possible to predict whether to split by referring to the split information of the call LCU and the call CU.

13 is a flowchart illustrating a method of generating LCU partition structure information based on reference block information according to an embodiment of the present invention.

If there is a coding unit having the largest split depth among the reference blocks of the reference frame, if the difference in the motion information of the coding unit included in the reference block is large or the motion information is large, the split depth value of the split target coding unit is referenced. When a condition is satisfied by setting a predetermined condition such as a case where the block is smaller than the largest division depth value, the encoding may be performed on the split target coding unit to be included in the LCU partition structure candidate.

For example, when a coding unit having a maximum split depth value exists in a reference block of a reference frame, an optimal coding unit of a split target coding unit may also be a coding unit having a large split depth value. Therefore, since coding units having a relatively small split depth which are relatively unlikely to occur are not included in the LCU partition structure candidate, unnecessary calculation processes required for encoding may be reduced in the encoder, thereby reducing computation complexity.

Referring to FIG. 13, reference block information is derived (step S1300).

As described above, the method for determining the LCU partitioning structure according to the embodiment of the present invention may divide the partitioning structure in which the encoding is performed on the encoding target coding unit based on the information of the reference block (for example, the call LCU or the call block). You can decide whether to use it as a rescue candidate. Information for determining whether a split target coding unit is used as a partition structure candidate is defined and used as a term of partition structure candidate determination information.

That is, in an embodiment of the present invention, the coding unit to be used as the LCU partition structure candidate among the split target coding units may be determined using the split structure candidate determination information derived from the call LCU information or the reference block such as the call block.

It is determined whether the partition structure candidate determination information of the reference frame satisfies a predetermined condition (step S1310).

On the basis of the partition structure candidate determination information derived from the reference block information of the reference frame, it is determined whether the split target coding unit satisfies a predetermined condition, and the encoding is performed on the split target coding unit to be used as an LCU partition structure candidate. Whether to split the encoding target coding unit into one larger segmentation depth or to split the encoding target coding unit into one larger segmentation depth without performing encoding on the encoding target coding unit to be used as an LCU segmentation structure candidate. Can be.

The above predetermined condition may be, for example, the following condition when a reference block is a call LCU.

1) Whether the coding unit having the largest division depth exists in the call LCU

2) whether the difference in the motion information in the prediction unit included in the call LCU is greater than a predetermined criterion (for example, when the difference in the motion information between CUs or PUs in the call LCU is large).

3) Whether the motion information is greater than a certain criterion in the prediction unit included in the call LCU (eg, when the component of the motion vector (x, y) is large).

4) The reference indexes of the coding units in the call LCU of the reference frame are different.

5) When there is transform coefficient information of coding units in a call LCU of a reference frame.

In addition, a method of obtaining information of a coding unit corresponding to a split target coding unit in a reference frame may be variously used. Partition structure candidate determination information may be obtained based on one of the above conditions, or partition structure candidate determination information may be obtained based on a plurality of conditions.

Based on the split structure candidate determination information, the split target coding unit may be used as the LCU split structure candidate to determine whether to perform encoding.

For some of the above predetermined conditions, for example, when the current division depth of the division target coding unit is 0 and a coding unit having 4, the largest division depth, exists in the call LCU of the reference frame, the division target coding unit Is likely to be further divided. Therefore, since the splitting target coding unit is determined to be additionally divided, the LCU splitting structure corresponding to the splitting depth of the coding target splitting unit 0 may not be used as a candidate for the LCU splitting structure. May not be performed.

As another example, when the difference of the motion information in the prediction unit included in the call LCU is larger than a predetermined criterion or the motion information in the prediction unit included in the call LCU is larger than the predetermined criterion, the likelihood of splitting coding units is further increased. The split target coding unit may not be used as an LCU split structure candidate, and coding for the split target coding unit may not be performed.

If the predetermined condition is not satisfied, the division target coding unit is encoded and included in the LCU partition structure candidate (step S1320).

If the above-described conditions are not satisfied in operation S1310, the split depth of the encoding target coding unit may be included in the LCU split structure candidate. That is, since the split depth of the split target coding unit may also be an optimal LCU split structure, when the predetermined condition is not satisfied, the split target coding unit may be encoded and included in the LCU split structure candidate.

It is determined whether the current split depth of the split target coding unit is the maximum split depth (step S1330).

In order to generate the LCU split structure candidate for the split target coding unit, the coding unit needs to be split up to the largest split depth, so it is determined whether the split target coding unit corresponds to the largest split depth.

If the current split depth of the split target coding unit is not the maximum split depth, the split target coding unit is split (step S1340).

When the current split depth of the split target coding unit is not the maximum split depth, the split target coding unit may be further split.

In addition, when the split target coding unit returns to step S1300 again, derives a reference block corresponding to the split target coding unit and derives partition structure candidate determination information based on the derived reference block, and fails to satisfy a predetermined condition. The determination of step S1310 for determining whether the partition structure candidate determination information of the reference frame satisfies a predetermined condition may be performed again.

However, if the reference block is not changed even when the split target block is divided as in the case where the reference block is the call LCU information, and the split structure candidate determination information is the same, the determination of steps S1300 and S1310 may not be performed again. That is, the flowchart of FIG. 13 may be used as one embodiment for implementing the present invention as long as it does not depart from the essence of the present invention.

For example, when the predetermined condition for performing the determination in step S1310 is whether a coding unit having the largest division depth exists in the call LCU, steps S1300 and S1310 need not be additionally determined. In this case, only the LCU partitioning structure having the largest partitioning depth may be stored as the LCU partitioning structure candidate because the splitting target coding unit is continuously partitioned through only step S1340.

In the LCU partition structure candidate determination method according to an embodiment of the present invention, partition structure candidate determination information may be derived from adjacent coding unit information existing in the same frame instead of a reference block.

Hereinafter, in an embodiment of the present invention, such a neighboring coding unit may be defined and used as a term of a neighboring block.

14 is a conceptual diagram illustrating adjacent coding units for deriving partition structure candidate determination information in an LCU partition structure candidate determination method according to an embodiment of the present invention.

The neighboring coding unit disclosed in FIG. 14 may be used as another neighboring coding unit as one example.

The neighboring coding unit is a coding unit in which encoding is performed before the splitting target coding unit, and may be used as a coding unit for deriving split structure candidate determination information in the splitting target coding unit.

For example, the neighboring coding unit may refer to at least one block among blocks in which encoding is performed in advance among coding units existing around the encoding target coding unit as illustrated in FIG. 14.

15 is a flowchart illustrating a method of deriving an LCU partition structure candidate according to an embodiment of the present invention.

Referring to FIG. 15, LCU partition structure candidate determination information is derived from an adjacent block (step S1500).

The partition structure candidate determination information that may be derived from the neighboring coding units may be depth information of the neighboring coding unit, motion information of the neighboring coding unit, reference index information of the neighboring coding unit, and transform coefficient information of the neighboring coding unit.

It is determined whether the partition structure candidate determination information derived from the adjacent block satisfies a predetermined condition (step S1510).

The predetermined condition in step S1510 may be the following condition.

(1) A method for determining whether a split target coding unit is determined as an LCU partition structure candidate based on depth information of an adjacent coding unit.

 (2) A method of determining whether a split target coding unit is determined as an LCU partition structure candidate based on a motion information difference of adjacent coding units.

(3) A method for determining whether a split target coding unit is determined as an LCU partition structure candidate based on motion information of an adjacent coding unit.

(4) A method for determining whether to determine a split target coding unit as an LCU partition structure candidate based on reference index information among adjacent coding unit information.

(5) A method of determining whether to determine a split target coding unit as an LCU partition structure candidate based on transform coefficient information of a coding unit among adjacent coding unit information.

For example, assuming that split depth information of an adjacent coding unit is 3, a coding unit having split depth information smaller than a split depth of an adjacent coding unit is less likely because a coding unit having split depth information smaller than a neighboring coding unit is less likely to occur. May not be used as an LCU partition structure candidate. That is, when the split depth of the neighboring coding unit is 3, since the split depth of the split target block is also likely to be close to 3, the block having the split depth 1 or 2 may not be used as the LCU split structure candidate.

As another example, when the difference between the motion related information of the adjacent coding unit and the motion related information of the split target coding unit is large, the split target coding unit may not be used as the LCU partition structure candidate.

In the above-described determination method used in step S1510, one or more conditions may be used to determine whether to determine the split target coding unit as the LCU partition structure candidate. In addition, the condition for determining whether to determine the split target coding unit as the LCU partition structure candidate may use various methods as well as the above-described embodiment.

If the condition of step S1510 is not satisfied, the division target coding unit is encoded and stored in the LCU partition structure candidate (step S1520).

If the condition of step S1510 is not satisfied, the split target coding unit may be stored as an LCU partition structure candidate.

It is determined whether the coding unit has the largest depth (step S1530).

When the condition of step S1510 is satisfied, it is determined whether the split target coding unit is a coding unit having the largest splitting depth, and when the splitting target coding unit is a coding unit having the largest splitting depth, the splitting procedure is terminated. If the target coding unit is not the coding unit having the largest division depth, the split coding unit may be additionally divided (step S1540) to derive the split structure candidate determination information from the adjacent coding unit through step S1500.

According to an embodiment of the present invention, in determining the division structure, as described above with reference to FIG. 13, the encoder encodes the coding unit information of the current depth by predicting the characteristics of the encoding target coding unit based on the reference block information and the adjacent block information. It may be determined whether to store the LCU partition structure candidate.

16 is a conceptual diagram illustrating a reference block and an adjacent block according to an embodiment of the present invention.

Referring to FIG. 16, a reference block and an adjacent block may be used to derive split structure candidate determination information of a split target coding unit.

17 is a flowchart illustrating a method of determining an LCU partition structure candidate according to an embodiment of the present invention.

According to an embodiment of the present invention, it is possible to determine whether to use a split target coding unit as an LCU partition structure candidate based on split structure candidate determination information derived from a reference block and an adjacent block.

Referring to FIG. 17, reference block information is derived (step S1700).

Partition structure candidate determination information for determining an LCU partition structure candidate may be derived from a reference block (call coding unit, call LCU).

When the partition structure candidate determination information assumes that the reference block is a call LCU, the partition structure candidate determination information may include split depth information of a coding unit existing in the call LCU, motion information of a coding unit present in the call LCU, and the call LCU. Various values such as a reference picture index of the coding unit and transform coefficient information of the coding unit present in the call LCU may be used as the partition structure candidate determination information.

It is determined whether the division target coding unit satisfies a predetermined condition (first condition) based on the calculated partition structure candidate determination information (step S1710).

For example, the determination may be performed using the following determination method.

1) Whether the coding unit having the largest division depth exists in the call LCU

2) whether the difference of the motion information in the prediction unit included in the call LCU is greater than a predetermined criterion (for example, when the difference in motion information between CUs or PUs in the call LCU is large).

3) whether the motion information is larger than a certain criterion in the prediction unit included in the call LCU (for example, when the component of the motion vector (x, y) is large).

4) The reference indexes of the coding units in the call LCU of the reference frame are different.

5) When there is transform coefficient information of coding units in a call LCU of a reference frame.

When the split target coding unit satisfies the above-described condition (a predetermined condition based on split structure candidate determination information), it is determined whether the split target coding unit is a coding unit having the largest split depth (step S1715).

When the split target coding unit does not satisfy a predetermined condition based on the calculated split structure candidate determination information, the split structure candidate determination information is derived from the adjacent block (step S1720).

It is determined whether the partition structure candidate determination information derived from the adjacent block satisfies a predetermined condition (step S1730).

In operation S1730, the predetermined condition may be determined by the following method.

(1) A method for determining whether to determine a split target coding unit as an LCU partition structure candidate based on depth information of an adjacent block.

(2) A method of determining whether a split target coding unit is determined as an LCU partition structure candidate based on a difference of motion information of an adjacent block.

(3) A method for determining whether to determine a split target coding unit as an LCU partition structure candidate based on motion information of an adjacent block.

(4) A method for determining whether to determine a split target coding unit as an LCU partition structure candidate based on reference index information among adjacent blocks.

 (5) A method for determining whether to determine a split target coding unit as an LCU split structure candidate based on transform coefficient information of a coding unit among adjacent block information.

If the condition of step S1730 is not satisfied, the division target coding unit is stored as the LCU partition structure candidate (step S1740).

When the condition of step S1730 is satisfied, it is determined whether the split target coding unit is a coding unit having the largest division depth (step S1715).

If the splitting target coding unit is a coding unit having the largest division depth, the LCU split structure candidate generation step is terminated.

If the split target coding unit is not the coding unit having the largest split depth, the split depth of the split target coding unit is increased by one (step S1750).

The splitting target coding unit having the increased splitting depth may perform the above-described procedure once again through step S1700.

17 is a flow chart shown in FIG. 17 as an embodiment, when there is a step that does not need to be performed again through steps S1700 to S1750, the step may be omitted and such an embodiment also includes the scope of the present invention. Included in

According to an embodiment of the present invention, by using a method of inferring information on whether or not to perform splitting on a split target coding unit based on split structure candidate determination information of a reference block, encoding without using bits indicating whether to split is performed. And decryption.

18 is a flowchart illustrating a method of encoding split information according to an embodiment of the present invention.

Referring to FIG. 18, partition structure candidate determination information is derived from a reference block (step S1800).

On the basis of the derived partition structure candidate determination information, it is determined whether the split target coding unit satisfies a predetermined condition (step S1810).

For example, as described above, it may be determined whether the following conditions are satisfied.

1) Whether the coding unit having the largest division depth exists in the call LCU

2) whether the difference of the motion information in the prediction unit included in the call LCU is larger than a predetermined criterion (for example, when the difference of the motion information between CUs or PUs in the call LCU is large).

3) whether the motion information is larger than a certain criterion in the prediction unit included in the call LCU (for example, when the component of the motion vector (x, y) is large).

4) The reference indexes of the coding units in the call LCU of the reference frame are different.

5) When there is transform coefficient information of coding units in a call LCU of a reference frame.

On the basis of the derived partition structure candidate determination information, only the case where the split target coding unit does not satisfy a predetermined condition is performed, and the partition information is encoded and stored and transmitted in the bitstream (step S1820).

If the splitting target coding unit does not satisfy the above-described predetermined condition based on the derived split structure candidate determination information, encoding may be performed by indicating whether to split the splitting coding unit as bitstream information.

That is, the split information encoding method according to the embodiment of the present invention determines whether the split target coding unit satisfies a predetermined condition and selectively expresses split information by selectively using bits representing split information of the split target coding unit. The bit used can be reduced.

19 is a flowchart illustrating a method of decoding split information according to an embodiment of the present invention.

Referring to FIG. 19, partition structure candidate determination information is derived from a reference block (step S1900).

In the decoding step, the partition structure candidate determination information may be derived from the reference block by using the same method as the encoding step of FIG. 18 in order to know the split information of the split target coding unit.

On the basis of the derived partition structure candidate determination information, it is determined whether the split target coding unit satisfies a predetermined condition (step S1910).

That is, it is possible to determine whether to decode the split information of the split target coding unit from the bitstream based on the split structure candidate determination information of the reference block of the split target coding unit.

If the determination result in step S1910 is satisfied, the decoding target coding unit may have a high probability of being split, and thus the decoding target coding unit may be split without determining whether the decoding target coding unit is split. That is, in the block partitioning method according to the embodiment of the present invention, when the probability of splitting the decoding target coding unit is high, the block partitioning information may be decoded by inferring that the partitioning information is partitioned without decoding the partitioning information.

For example, when there is a coding unit having the largest split depth in the call LCU, or when the difference in the motion information of the coding units in the call LCU is large or the motion information is large, the split target coding unit is considered to be split, and the split target coding is performed. The division information of the unit is not decoded from the bitstream.

If it is determined in step S1910 that the corresponding condition is not satisfied, the split information for the coding unit of the current depth is decoded from the bitstream as before (step S1920).

20 is a flowchart illustrating a method of encoding split information according to an embodiment of the present invention.

Referring to FIG. 20, split structure candidate determination information is derived from a coding unit adjacent to a split target coding unit (step S2000).

As described above with reference to FIG. 14, the partition structure candidate determination information may be derived from a neighboring coding unit adjacent to the split target coding unit.

On the basis of the derived partition structure candidate determination information, it is determined whether the split target coding unit satisfies a predetermined condition (step S2010).

For example, as described above, it may be determined whether the following conditions are satisfied.

(1) A method for determining whether to determine a split target coding unit as an LCU partition structure candidate based on depth information of an adjacent coding unit.

(2) A method of determining whether a split target coding unit is determined as an LCU partition structure candidate based on a motion information difference of adjacent coding units.

(3) A method for determining whether a split target coding unit is determined as an LCU partition structure candidate based on motion information of an adjacent coding unit.

(4) A method for determining whether to determine a split target coding unit as an LCU split structure candidate based on reference index information among adjacent coding unit information.

(5) A method of determining whether to determine a split target coding unit as an LCU split structure candidate based on transform coefficient information of a coding unit among adjacent coding unit information.

If the splitting target coding unit does not satisfy a predetermined condition based on the derived split structure candidate determination information, the splitting information is encoded and stored and transmitted in the bitstream (step S2020).

For example, when the depth information of an adjacent coding unit of the split target coding unit is smaller than the depth information of the split target coding unit, the difference in the motion information is small, or the motion information is small in the same frame, the encoding target coding unit is not divided. In this case, the split information of the split target coding unit is encoded and stored and transmitted in the bitstream. This means that the coding unit is not likely to be split when the above conditions are met. However, when the coding unit information is encoded, when the coding unit is not likely to be split, the splitting information is not sent, thereby reducing the bit amount. have. If the above condition is not satisfied, the split information of the coding unit of the current depth may be encoded, stored, and transmitted in the bitstream.

21 is a flowchart illustrating a method of decoding partition information according to an embodiment of the present invention.

Referring to FIG. 21, partition structure candidate determination information is derived from an adjacent block (step S2100).

In the decoding step, the partition structure candidate determination information may be derived from an adjacent block of the encoding target coding unit using the same method as the encoding step of FIG. 20 in order to know the split information of the coding block.

On the basis of the derived partition structure candidate determination information, it is determined whether the split target coding unit satisfies a predetermined condition (step S2110).

That is, it is possible to determine whether to decode the split information of the split target coding unit from the bitstream through an adjacent block of the split target coding unit. have. For example, when the depth information of the coding unit adjacent to the split target coding unit is smaller than the depth information of the decoding target coding unit, the difference in the motion information is small, or the motion information is small in the same frame, the split target coding unit is not divided. It is assumed that the split information of the split target coding unit is not decoded from the bitstream.

If it is determined in step S2110 that the predetermined condition is not satisfied, the split information of the split target coding unit is decoded from the bitstream as before (step S2120).

22 is a flowchart illustrating a method of encoding split information according to an embodiment of the present invention.

Referring to FIG. 22, reference block information is derived (step S2200).

Split structure candidate determination information for determining an LCU split structure candidate in a reference block (call coding unit, call LCU) and an adjacent block may be derived.

It is determined whether or not the division target coding unit satisfies a predetermined condition based on the calculated partition structure candidate determination information (step S2210).

For example, the determination may be performed using the following determination method.

1) Whether the coding unit having the largest division depth exists in the call LCU

2) whether the difference of the motion information in the prediction unit included in the call LCU is greater than a predetermined criterion (for example, when the difference in motion information between CUs or PUs in the call LCU is large).

3) whether the motion information is larger than a certain criterion in the prediction unit included in the call LCU (for example, when the component of the motion vector (x, y) is large).

4) The reference indexes of the coding units in the call LCU of the reference frame are different.

5) When there is transform coefficient information of coding units in a call LCU of a reference frame.

When the above conditions are satisfied, since the encoding target coding unit is more likely to be further split, split information of the current split target coding unit is not encoded.

When the split target coding unit does not satisfy the predetermined condition based on the calculated split structure candidate determination information, the split structure candidate determination information is derived from the neighboring coding unit of the split target coding unit (step S2220).

It is determined whether the partition structure candidate determination information derived from the neighboring coding unit of the split target coding unit satisfies a predetermined condition (step S2230).

In operation S2230, the predetermined condition may be determined by the following method.

(1) A method for determining whether a split target coding unit is determined as an LCU partition structure candidate based on depth information of an adjacent coding unit.

 (2) A method of determining whether a split target coding unit is determined as an LCU partition structure candidate based on a motion information difference of adjacent coding units.

(3) A method for determining whether a split target coding unit is determined as an LCU partition structure candidate based on motion information of an adjacent coding unit.

(4) A method for determining whether to determine a split target coding unit as an LCU partition structure candidate based on reference index information among adjacent coding unit information.

 (5) A method of determining whether to determine a split target coding unit as an LCU partition structure candidate based on transform coefficient information of a coding unit among adjacent coding unit information.

When the condition of step S2230 is satisfied, since the split target coding unit may not be further split, split information of the split target coding unit is not encoded.

If the condition of step S2230 is not satisfied, split information of the split target coding unit is encoded (step S2240).

That is, according to the embodiment of the present invention, when the partition information is transmitted by performing encoding only on the partition information when there is a possibility that it may be used as the LCU partition structure candidate, the partition information may be transmitted using smaller bits.

23 is a flowchart illustrating a method of decoding partition information according to an embodiment of the present invention.

Referring to FIG. 23, reference block information is derived (step S2300).

Split structure candidate determination information for determining an LCU split structure candidate in a reference block (call coding unit, call LCU) and an adjacent block may be derived.

It is determined whether or not the division target coding unit satisfies a predetermined condition based on the partition structure candidate determination information calculated from the reference block (step S2310).

The above condition is the same as the step S2210 of performing a determination to perform split information encoding in the encoding step.

When the splitting target coding unit satisfies a predetermined condition based on the split structure candidate determination information calculated through the determination process in step S2310, it is determined that the splitting target coding unit is not split and the splitting information of the splitting target coding unit is determined. Do not decrypt.

If the splitting target coding unit does not satisfy a predetermined condition based on the calculated split structure candidate determination information through the determination process in step S2310, splitting structure candidate determination information is derived from an adjacent block of the splitting target coding unit ( Step S2320).

It is determined whether the partition structure candidate determination information derived from the neighboring block of the split target coding unit satisfies a predetermined condition (step S2330).

In operation S2330, it may be determined whether the partition structure candidate determination information derived from the neighboring block of the split target coding unit satisfies a predetermined condition using the same determination method as in operation S2230.

When the partition structure candidate determination information derived from the neighboring coding unit of the split target coding unit satisfies a predetermined condition, the split target coding unit may not be further split, and thus the split state of the split target coding unit is not decoded.

When the partition structure candidate determination information derived from the neighboring coding unit of the split target coding unit satisfies a predetermined condition, split information of the split target coding unit is decoded (step S2340).

When the partition structure candidate determination information derived from the neighboring coding unit of the split target coding unit satisfies a predetermined condition, information on whether to split the split target coding unit may be decoded, and may be stored and transmitted in the bitstream.

Hereinafter, an embodiment of the present invention is a conceptual diagram illustrating a case where the aforementioned split information encoding method and decoding method are implemented in an apparatus.

24 is a conceptual diagram illustrating a part of an image encoding apparatus according to an embodiment of the present invention.

Referring to FIG. 24, the apparatus for encoding an image may include a determiner 2400, a CU encoding and LCU partition structure candidate registerer 2420, and an LCU partition structure determiner 2440.

The determination unit 2400 receives the reference block information and the split target coding unit information as an input and determines whether to encode the split target coding unit to include in the LCU partition structure candidate. The determination unit 2400 may go through the determination process described above with reference to FIG. 13.

If the determination unit 2400 determines to encode the split target coding unit, the split target coding unit may be encoded and included in the LCU split structure candidate through the CU encoding and the LCU split structure candidate register 2420. If the determination unit 2400 determines not to encode the encoding target coding unit, the determination unit 2400 may not pass through the CU encoding and the LCU partition structure candidate registration unit 2420.

If the coding target coding unit is not the coding unit having the largest splitting depth, the splitting coding unit may be divided into four coding units having a depth increased by one, and the determination unit 2400 may determine that the splitting coding unit is an LCU splitting structure. It may be determined whether to be included in the candidate.

The LCU split structure determiner 2440 may select an optimal LCU split structure among the LCU split structure candidates registered by the CU encoding and the LCU split structure candidate register 2420.

25 is a conceptual diagram illustrating a part of an image encoding apparatus according to an embodiment of the present invention.

Referring to FIG. 25, the determination unit 2500 receives input of coding unit information and split target coding unit information adjacent to a split target coding unit in the same frame, and determines whether to split the split target coding unit and include the split target coding unit in the LCU partition structure candidate. Decide The determination unit 2500 may go through the determination process described above with reference to FIG. 15.

If the determination unit 2500 determines to encode the split target coding unit, the split target coding unit is encoded by the CU encoding and the LCU split structure candidate registration unit 2520 to be included in the LCU split structure candidate. If the determination unit 2500 determines not to encode the split target coding unit, the CU encoding and the LCU split structure candidate registration unit 2520 are not passed through. Thereafter, if the splitting target coding unit is not the largest depth, the splitting unit may be divided into four coding units having the increased depth by 1, and the determination unit 2500 may determine whether the split coding unit is included in the candidate LCU partitioning structure. have. The LCU partition structure determination unit 2540 selects an optimal LCU partition structure among the LCU partition structure candidates registered in the CU encoding and the LCU partition structure candidate registration unit 2520.

26 is a conceptual diagram illustrating a part of an image encoding apparatus according to an embodiment of the present invention.

Referring to FIG. 26, the determination unit 2600 receives, as input, information of a reference block that is coding unit information corresponding to a split target coding unit, coding unit information adjacent to a split target coding unit, and split target coding unit information in a reference frame. It may be determined whether the split target coding unit is encoded and included in the LCU partition structure candidate. The determination unit 2600 may go through the determination process described above with reference to FIG. 17.

If the determination unit 2600 determines to encode the split target coding unit, the split target coding unit is encoded by the CU encoding and the LCU split structure candidate registration unit 2620 and included in the LCU split structure candidate. If the determination unit 2600 determines not to encode the split target coding unit, the CU encoding and the LCU split structure candidate registration unit 2620 are not passed through.

After that, if the splitting target coding unit is not the largest splitting depth, the splitting target coding unit is split into four coding units having the splitting depth increased by one, and the above process is performed again. The LCU partition structure determiner 2640 selects an optimal LCU partition structure among registered LCU partition structure candidates.

27 is a conceptual diagram illustrating a part of an image encoding apparatus according to an embodiment of the present invention.

Referring to FIG. 27, the determination unit 2700 determines whether to encode split information of a split target coding unit by receiving information about a coding unit corresponding to a split target coding unit and a split target coding unit in a reference frame. The determination unit 2700 may determine whether to encode the split information of the split target coding unit based on the determination procedure described above with reference to FIG. 18.

If the determination unit 2700 determines to encode the split information of the split target coding unit, the CU split information encoder 2720 encodes the split information of the split target coding unit, and stores and transmits the split information in the bitstream. If the determination unit 2700 decides not to encode the split information of the split target coding unit, the split information of the split target coding unit is not stored and transmitted in the bitstream without passing through the CU split information encoder 2720.

28 is a conceptual diagram illustrating a part of an image decoding apparatus according to an embodiment of the present invention.

Referring to FIG. 28, the image decoding apparatus may include a determiner 2800 and a CU partition information decoder 2820.

The determination unit 2800 receives input information of the reference block corresponding to the split target coding unit and split target coding unit information in the reference frame, and determines whether to decode the split information of the split target coding unit from the bitstream. The determination unit 2800 may determine whether to encode the split information of the split target coding unit based on the determination procedure described above with reference to FIG. 19.

If the determination unit 2800 decides to decode the split information of the split target coding unit, the CU split information decoder 2820 decodes the split information of the split target coding unit. If the determination unit 2800 decides not to decode the split information of the split target coding unit, the determination unit 2800 may not pass through the CU split information decoder 2820. The generated partition information may be stored in the CU information.

29 is a conceptual diagram illustrating a part of an image encoding apparatus according to an embodiment of the present invention.

Referring to FIG. 29, an apparatus for encoding an image may include a determiner 2900 and a CU partition information decoder 2920.

The determination unit 2900 determines whether to decode the split information of the split target coding unit by receiving the coding unit information adjacent to the split target coding unit and the split target coding unit information in the same frame.

The determiner 2900 may determine whether to encode split information of the split target coding unit based on the above-described determination procedure of FIG. 20.

If the determination unit 2900 determines to encode the split information of the split target coding unit, the CU split information encoder 2920 encodes the split information of the split target coding unit, and stores and transmits the split information in the bitstream. If the determination unit 2900 determines not to encode the split information of the split target coding unit, the split information of the corresponding CU is not stored and transmitted in the bitstream without passing through the CU split information encoder 2920.

30 is a conceptual diagram illustrating a part of an image decoding apparatus according to an embodiment of the present invention.

Referring to FIG. 30, the image decoding apparatus may include a determiner 3000 and a CU partition information decoder 3020.

The determination unit 3000 receives the coding unit information adjacent to the split target coding unit and the split target coding unit information in the same frame, and determines whether to decode the split information of the split target coding unit from the bitstream.

The determination unit 3000 may determine whether to encode the split information of the split target coding unit based on the determination procedure described above with reference to FIG. 21.

If the determination unit 3000 determines to decode the split information of the split target coding unit, the CU split information decoder 3020 decodes the split information of the split target coding unit. If the determination unit 3000 determines not to decode the split information of the split target coding unit, the CU split information decoder 3020 is not passed through. Thereafter, the partition information may be stored in the CU information.

31 is a conceptual diagram illustrating a part of an image encoding apparatus according to an embodiment of the present invention.

Referring to FIG. 31, the apparatus for encoding an image may include a determiner 3100 and a CU partition information decoder 3120.

The determination unit 3100 receives input block information, coding unit information adjacent to the split target coding unit, and split target coding unit information as input, and determines whether to encode split information of the split target coding unit.

The determination unit 3100 may determine whether to encode the split information of the split target coding unit based on the determination procedure described above with reference to FIG. 22.

If the determination unit 3100 determines to encode the split information of the split target coding unit, the CU split information encoder 3120 encodes split information of the split target coding unit, and stores and transmits the split information in the bitstream. If the determination unit 3100 determines not to encode the split information of the split target coding unit, the split information of the split target coding unit is not stored and transmitted in the bitstream without passing through the CU split information encoder 3120.

32 is a conceptual diagram illustrating a part of an image decoding apparatus according to an embodiment of the present invention.

Referring to FIG. 32, the image decoding apparatus may include a determination unit 3100 and a CU partitioning information decoding unit 3120.

The determination unit 3100 receives, as input, coding unit information adjacent to the split target coding unit and information of the decoding target coding unit in a frame such as CU information corresponding to the split target coding unit, from the bitstream. It is determined whether to decode the partition information.

The determination unit 3100 may determine whether to encode split information of the split target coding unit based on the determination procedure described above with reference to FIG. 23.

If the determination unit 3100 determines to decode the split information of the split target coding unit, the CU split information decoder 3120 decodes the split information of the split target coding unit. If the determination unit 3100 determines not to decode the split information of the split target coding unit, the CU split information decoder 3120 is not passed. Thereafter, the partition information is stored in the CU information.

Hereinafter, an embodiment of the present invention discloses a method for determining an additional LCU partition structure candidate.

33 is a flowchart illustrating a method of determining an LCU partition structure candidate according to an embodiment of the present invention.

Referring to FIG. 33, partition structure candidate determination information is derived from a reference block (step S3300).

Information about whether there is a block having the largest partition depth information among the reference blocks and whether there is a value greater than 10 of the x component or the y component of the reference block for use as the partition structure candidate determination information may be derived. Can be.

It is determined whether a predetermined condition derived from the division target coding unit and the reference block is satisfied (step S3310).

The following conditions may be used as predetermined conditions for determining whether to use a split target coding unit as an LCU partition structure candidate.

1) Whether the split depth of the split target coding unit is 0

2) whether the coding unit having the largest partition depth exists in the call LCU

3) whether there is a value greater than 10 among the x and y components of the motion vector of the coding unit included in the call LCU,

Etc. can be used to determine whether the condition is satisfied.

As the above conditions are arbitrary, the split depth information of the split target coding unit or the information about the values of the motion vector components of the split target coding unit may use different values, and such an embodiment is also included in the scope of the present invention.

If the above-described condition is satisfied, it is determined whether the split target coding unit is the largest division depth (step S3320).

If the splitting target coding unit has the largest division depth, the LCU division structure candidate determination procedure is terminated without performing an additional division process.

If the split target coding unit is not the largest split depth, the split target coding unit is further split (step S3330).

On the basis of the split coding unit, a determination may be made on whether to use the split target coding unit as an LCU split structure candidate through step S3310.

If the above condition is not satisfied, the division target coding unit is encoded and stored as an LCU partition structure candidate (step S3340).

33 is an embodiment for generating an LCU partition structure candidate, and when there are unnecessary additional steps, the determination may be performed except for the corresponding steps, and such embodiments are also included in the scope of the present invention.

34 is a flowchart illustrating a method of determining an LCU partition structure candidate according to an embodiment of the present invention.

Referring to FIG. 34, partition structure candidate determination information is derived from a reference block (step S3400).

Whether there is a block having the largest partition depth information among the reference blocks for use as the partition structure candidate determination information, and whether there is a value greater than 10 of the x component or the y component of the motion vector difference of the reference block. Information on whether or not can be derived.

It is determined whether a predetermined condition derived from the division target coding unit and the reference block is satisfied (step S3410).

The following conditions may be used as a predetermined condition for determining whether to use a split target coding unit as an LCU partition structure candidate.

1) Whether the split depth of the split target coding unit is 0

2) whether the coding unit having the largest partition depth exists in the call LCU

3) whether there is a value greater than 10 among the x and y components of the motion vector difference of the coding unit included in the call LCU,

Etc. can be used to determine whether the condition is satisfied.

As the above conditions are arbitrary, the split depth information of the split target coding unit or the information about the values of the motion vector components of the split target coding unit may use different values, and such an embodiment is also included in the scope of the present invention.

If the above condition is satisfied, it is determined whether the split target coding unit is the largest division depth (step S3420).

If the splitting target coding unit has the largest division depth, the LCU division structure candidate determination procedure is terminated without performing an additional division process.

If the split target coding unit is not the largest split depth, the split target coding unit is further split (step S3430).

On the basis of the split coding unit, a determination may be made on whether to use the split target coding unit as the LCU split structure candidate through step S3410.

If the above condition is not satisfied, the division target coding unit is encoded and stored as an LCU partition structure candidate (step S3440).

FIG. 34 is an embodiment for generating an LCU partition structure candidate, and when there is an unnecessary additional step, the determination may be performed by excluding the corresponding step, and such an embodiment is also included in the scope of the present invention.

35 is a flowchart illustrating a method of determining an LCU partition structure candidate according to an embodiment of the present invention.

Referring to FIG. 35, partition structure candidate determination information is derived from a reference block (step S3500).

It is determined whether to use the split target coding unit as the LCU split structure candidate based on the split structure candidate determination information derived from the reference block and the split depth information of the split target coding unit (step S3510).

To determine whether to use a split target coding unit as an LCU partition structure candidate

1) Whether the depth information of the split target coding unit is 0

2) It should be determined whether a coding unit having the largest division depth among the reference blocks exists.

Also, a method for determining whether a coding unit having a predetermined split depth value among the reference blocks may be used as the depth information of the split target coding unit among the above conditions to determine the LCU split structure candidate. May be used and such embodiments are also within the scope of the present invention.

If the determination is made in step S3510 and the corresponding condition is not satisfied, the split target coding unit may be an optimal LCU split structure, and thus the split target coding unit may be stored in the LCU split unit structure candidate (step S3520). ).

When the determination is made in step S3510, when the corresponding condition is satisfied, it is determined whether the split target coding unit is a coding unit having the largest division depth (step S3530).

When the determination is performed in step S3510, when the corresponding condition is satisfied, the split target coding unit may be further split, and thus may not be stored in the LCU partition structure candidate.

As a result of the determination in step S3530, if the splitting target coding unit is a coding unit having the largest splitting depth, the procedure for generating the LCU split structure candidate is terminated. If not, additional splitting may be performed on the split target coding unit (step S3540).

In addition, the split target coding unit may perform a determination procedure to determine whether to include the split target coding unit in the LCU split structure.

36 is a flowchart illustrating a method of determining an LCU partition structure candidate according to an embodiment of the present invention.

Referring to FIG. 36, partition structure candidate determination information is derived from a reference block (step S3600).

It is determined whether a predetermined condition (first condition) is satisfied as to whether to use the division target coding unit as the LCU partition structure candidate (step S3610).

The following conditions may be used as a predetermined condition for determining whether to use a split target coding unit as an LCU partition structure candidate.

1) Whether the split depth of the split target coding unit is 0

2) whether the coding unit having the largest partition depth exists in the call LCU

3) whether there is a value greater than 10 among the x and y components of the motion vector of the coding unit included in the call LCU,

Etc. can be used to determine whether the condition is satisfied.

As the above conditions are arbitrary, the split depth information of the split target coding unit or the information about the values of the motion vector components of the split target coding unit may use different values, and such an embodiment is also included in the scope of the present invention.

If the predetermined condition is not satisfied whether the split target coding unit is used as the LCU partition structure candidate, it is determined whether the split target coding unit and the reference block satisfy the following additional conditions (step S3620).

Judgment values used in the following two conditions may be changed to arbitrary values. For example, it is also possible to determine whether or not the split depth of the split target coding unit is a split depth smaller than the maximum split depth.

1) whether the split depth of the split target coding unit is the maximum split depth;

2) whether the splitting depth of the reference block is less than 2

When at least one of the above two conditions is satisfied, the split target coding unit is less likely to be selected as an optimal LCU partition structure, and thus may not be stored as an LCU partition structure candidate.

If the above two conditions are not satisfied, the division target coding unit is encoded and stored in the LCU partition structure candidate (step S3630).

If the above two conditions are not satisfied, the split target coding unit may be selected as an optimal LCU split structure, and the split target coding unit is encoded and stored in the LCU split structure candidate.

When a predetermined condition is satisfied as to whether to use the split target coding unit as the LCU split structure candidate, it is determined whether the split target coding unit is a coding unit having the largest split depth (step S3640).

As a result of the determination in step S3640, when the splitting target coding unit is a coding unit having the largest splitting depth, the procedure for generating the LCU split structure candidate is ended, and in step S3640, the coding unit having the largest splitting depth is determined. If not, additional splitting may be performed on the split target coding unit (step S3650).

In encoding and decoding the partition structure candidates determined through the method of splitting the LCU partition structure candidates described with reference to FIGS. 33 to 36, a determination condition used to determine whether to store the LCU partition structure candidate may be used. As described above with reference to FIGS. 18 to 23, when the determination condition used to determine whether to store the LCU partition structure candidate is satisfied, the encoding step may not use the coding bit indicating the corresponding LCU partition structure, and may also be encoded in the decoding step. As in the step, the corresponding bit may not be decoded in the bitstream.

If the determination condition used to determine whether to store the LCU partition structure candidate is not satisfied, encoding and decoding may be performed using the same method as the existing LCU partition structure encoding and decoding method.

37 is a conceptual diagram illustrating an image encoder according to an embodiment of the present invention.

Referring to FIG. 37, the image encoder may include a determiner 3700, a CU encoding and LCU partition structure candidate registerer 3720, and an LCU partition structure determiner 3740.

The determination unit 3700 determines whether to include the CU information corresponding to the split target coding unit and the split target coding unit information in the reference frame as an input, to encode the split target coding unit, and to include the split target coding unit in the candidate LCU partition structure. The determination unit 3700 may determine whether to encode the split information of the split target coding unit based on the determination procedure described above with reference to FIG. 33. If the determination unit 3700 decides to encode the split target coding unit, the split target coding unit is encoded by the CU encoding and the LCU split structure candidate registration unit 3720 and included in the LCU split structure candidate. If the determination unit 3700 determines not to encode the splitting target coding unit, the determination unit 3700 may not pass through the CU encoding and the LCU split structure candidate register 3720. Subsequently, if the splitting target coding unit is not the largest depth, the above process may be performed again by further splitting the split target coding unit into four splitting target coding units having a depth increased by one. The LCU partition structure determiner 3740 may select an optimal LCU partition structure from among registered LCU partition structure candidates by encoding coding units having a minimum depth to a maximum depth in LCU units.

38 is a conceptual diagram illustrating an image encoder according to an embodiment of the present invention.

Referring to FIG. 38, an image encoder may include a determiner 3800, a CU encoding and LCU partition structure candidate register 3820, and an LCU partition structure determiner 3840.

The determination unit 3800 may receive the adjacent block adjacent to the split target coding unit and split target coding unit information in the same frame as an input, and determine whether to encode the split target coding unit and include the split target coding unit in an LCU partition structure candidate. The determination unit 3800 may determine whether to encode the split information of the split target coding unit based on the determination procedure described above with reference to FIG. 34.

If the determination unit 3800 determines to encode the split target coding unit, the split target coding unit may be encoded and included in the LCU split structure candidate through the CU encoding and the LCU split structure candidate register 3820. If the determination unit 3800 decides not to encode the split target coding unit, the CU encoding and the LCU split structure candidate registration unit 3820 are not passed through. Thereafter, if the splitting target coding unit is not the largest depth, the splitting process is split into four splitting target coding units having a depth increased by one, and the above process is performed again. The LCU partition structure determination unit 3840 encodes coding units from the minimum depth to the maximum depth in LCU units to select an optimal LCU partition structure among registered LCU partition structure candidates.

39 is a conceptual diagram illustrating an image encoder according to an embodiment of the present invention.

Referring to FIG. 39, the image encoder may include a determiner 3900, a CU encoding and LCU partition structure candidate register 3920, and an LCU partition structure determiner 3940.

The determination unit 3900 inputs block information (adjacent block information) adjacent to the split target coding unit and information of the split target coding unit in a frame, such as CU information (reference block information) corresponding to the split target coding unit, in the reference frame. The method determines whether to divide the encoding target coding unit to include in the LCU partition structure candidate.

The determination unit 3900 may determine whether to encode the split information of the split target coding unit based on the determination procedure described above with reference to FIG. 35.

If the determination unit 3900 () decides to encode the split target coding unit, the split target coding unit is encoded by the CU encoding and the LCU split structure candidate register 3920 to be included in the LCU split structure candidate. If the determination unit 3900 determines not to encode the split target coding unit, the CU encoding and the LCU partition structure candidate registration unit 3920 are not passed through. After that, if the splitting target coding unit is not the largest depth, the splitting process is split into four splitting target coding units having a depth increased by one, and the above process is performed again. The LCU partition structure determiner 3940 selects an optimal LCU partition structure from among registered LCU partition structure candidates by encoding a CU from the minimum depth to the maximum depth in LCU units.

40 is a conceptual diagram illustrating an image encoder according to an embodiment of the present invention.

Referring to FIG. 40, the image encoder may include a determiner 4000 and a CU partition information encoder 4020.

The determination unit 4000 receives reference block information which is coding unit information corresponding to the split target coding unit and split target coding unit information in the reference frame, and determines whether to encode the split information of the current CU.

The determination unit 4000 may determine whether the following conditions are satisfied in order to determine whether to encode the split target coding unit to include in the LCU partition structure candidate.

1) Whether the split depth of the split target coding unit is 0

2) whether the coding unit having the largest partition depth exists in the call LCU

3) Whether there is a value greater than 10 among the x and y components of the motion vector of the coding unit included in the call LCU

As the above conditions are arbitrary, the split depth information of the split target coding unit or the information about the values of the motion vector components of the split target coding unit may use different values, and such an embodiment is also included in the scope of the present invention.

If the determination unit 4000 determines to encode the split information of the split target coding unit, the CU split information encoder 4020 encodes the split information of the split target coding unit, and stores and transmits the split information in the bitstream. If the determination unit 4000 determines not to encode the split information of the split target coding unit, the split information of the split target coding unit is not stored and transmitted in the bitstream without passing through the CU split information encoder 4020.

41 is a conceptual diagram illustrating an image decoder according to an embodiment of the present invention.

Referring to FIG. 41, an image decoder may include a determiner 4100 and a CU partition information decoder 4120.

It is determined whether to decode the split information of the split target coding unit from the bitstream by receiving the reference block information which is the coding unit information corresponding to the split target coding unit and the split target coding unit information in the reference frame. The determination unit 4100 may determine whether to decode the split information of the encoding target coding unit using the same method as the method of determining whether to encode in FIG. 40.

If the determination unit 4100 determines to decode the split information of the split target coding unit, the CU split information decoder 4120 decodes the split information of the split target coding unit. If the determination unit 4100 determines not to decode the split information of the split target coding unit, the CU split information decoder 4120 is not passed. Thereafter, the partition information may be stored in the CU information.

42 is a conceptual diagram illustrating an image encoder according to an embodiment of the present invention.

Referring to FIG. 42, the image encoder may include a determiner 4200 and a CU partition information encoder 4220.

The determination unit 4200 determines whether to encode the split information of the split target coding unit by receiving adjacent block information and split target coding unit information, which are coding unit information adjacent to the coding target coding unit, in the same frame.

The determination unit 4200 may determine whether the following conditions are satisfied in order to determine whether to encode the split target coding unit to include in the LCU partition structure candidate.

1) Whether the split depth of the split target coding unit is 0

2) whether the coding unit having the largest partition depth exists in the call LCU

3) whether there is a value greater than 10 among the x and y components of the motion vector difference of the coding unit included in the call LCU,

If the determination unit 4200 determines to encode the split information of the split target coding unit, the CU split information encoder 4220 may encode split information of the split target coding unit, and store and transmit the split information in the bitstream. If the determination unit decides not to encode the split information of the split target coding unit, the split information of the split target coding unit is not stored and transmitted in the bitstream without passing through the CU split information encoder 4220.

43 is a conceptual diagram illustrating an image decoder according to an embodiment of the present invention.

Referring to FIG. 43, an image decoder may include a determiner 4300 and a CU partition information encoder 4320.

The determination unit 4300 determines whether to decode the split information of the split target coding unit from the bitstream by receiving the coding unit information adjacent to the split target coding unit and the split target coding unit information in the same frame.

The determination unit 4300 may determine whether to decode the split information of the encoding target coding unit using the same method as the method of determining whether to encode in FIG. 42. If the determination unit 4300 determines to decode the split information of the split target coding unit, the CU split information decoder 4320 decodes the split information of the split target coding unit. If the determination unit 4300 decides not to decode the partition information of the current CU, the CU partition information decoder 4320 is not passed through. Thereafter, the partition information may be stored in the CU information.

44 is a conceptual diagram illustrating an image encoder according to an embodiment of the present invention.

Referring to FIG. 44, an image encoder may include a determiner 4400 and a CU partition information encoder 4420.

The determination unit 4400 inputs coding unit information (adjacent block information) adjacent to the split target coding unit and split target coding unit information in a frame such as coding unit information (reference block information) corresponding to the split target coding unit in the reference frame. It determines whether to encode the split information of the split target coding unit.

The determination unit 4400 may determine whether the following conditions are satisfied in order to determine whether to encode the split target coding unit to include in the LCU partition structure candidate.

1) Whether the depth information of the split target coding unit is 0

2) Whether there is a coding unit having the largest division depth among the reference blocks, to determine a candidate candidate for the LCU segmentation structure, depth information of the coding target coding unit among the above conditions may use a value other than 0 and a predetermined value among the reference blocks may be used. A method of determining whether a coding unit having a split depth value of may exist may also be used, and such an embodiment is also included in the scope of the present invention.

In addition, the determination unit 4400 may determine whether the following conditions are satisfied in order to determine whether to encode a split target coding unit to include in the LCU partition structure candidate.

1) whether the split depth of the split target coding unit is the maximum split depth;

2) whether the splitting depth of the reference block is less than 1

When the above two conditions are satisfied, the corresponding split target coding unit is less likely to be selected as an optimal LCU partition structure, and thus may not be stored as an LCU partition structure candidate.

If the determination unit 4400 determines to encode the split information of the split target coding unit, the CU split information encoder 4420 encodes the split information of the split target coding unit, and stores and transmits the split information in the bitstream. If the determination unit 4400 decides not to encode the split information of the split target coding unit, the split information of the split target coding unit is not stored and transmitted in the bitstream without passing through the CU split information encoder 4420.

45 is a conceptual diagram illustrating an image decoder according to an embodiment of the present invention.

Referring to FIG. 45, the image decoder may include a determiner 4500 and a CU partition information encoder 4520.

The determination unit 4500 receives from the bitstream the coding unit information adjacent to the splitting target coding unit and the splitting target coding unit information in a frame such as the coding unit information (reference block information) corresponding to the splitting coding unit in the reference frame. It is determined whether to decode the split information of the split target coding unit. The determination unit () may determine whether to decode the split information of the split target coding unit using the same method as the method of determining whether to encode in FIG. 44.

If the determination unit 4500 determines to decode the split information of the split target coding unit, the CU split information decoder 4520 decodes the split information of the split target coding unit. If the determination unit 4500 determines not to decode the split information of the split target coding unit, the CU split information decoder 4520 is not passed through. Thereafter, the partition information is stored in the CU information.

Although described with reference to the embodiments above, those skilled in the art will understand that the present invention can be variously modified and changed without departing from the spirit and scope of the invention as set forth in the claims below. Could be.

Claims (20)

Deriving partition structure candidate determination information from a reference block; And
And determining whether to use the split target coding unit as the LCU split structure candidate using predetermined conditions calculated based on the derived split structure candidate determination information and the information derived from the split target coding unit. Way. The method according to claim 1, wherein the predetermined condition is
Information on at least one of a partition depth of a block included in a call LCU, motion information of a prediction unit included in a call LCU, reference index information of coding units in a call LCU, and transform coefficient information of coding units in a call LCU And a condition for determining whether to use the split target coding unit as an LCU partition structure candidate. The method according to claim 1, wherein the predetermined condition is
Whether the splitting depth of the coding target to be split is 0, whether a coding unit having the largest splitting depth exists in the call LCU, or a value larger than 10 among the x and y components of the motion vector of the coding unit included in the call LCU And at least one condition of whether there is a value greater than 10 among x and y components of a motion vector difference of a coding unit included in a call LCU. The method of claim 1, wherein whether to use the split target coding unit as an LCU split structure candidate is determined using a predetermined condition calculated based on the derived split structure candidate determination information and information derived from the split target coding unit. The steps are
Determining whether the derived partition structure candidate determination information and information derived from the division target coding unit satisfy a first condition; And
Determining whether the derived partition structure candidate determination information and information derived from the split target coding unit satisfy a second condition,
And the first condition and the second condition are conditions derived based on split depth information of a split target coding unit and split structure candidate determination information derived from a reference block. Deriving partition structure candidate determination information from an adjacent block; And
And determining whether to use the split target coding unit as an LCU split structure candidate using a predetermined condition calculated based on the split structure candidate determination information and the information derived from the split target coding unit. . The method according to claim 5, wherein the predetermined condition is
Whether to use the split target coding unit as an LCU partition structure candidate using at least one of depth information of an adjacent block, motion information of an adjacent block, reference index information of an adjacent block, and transform coefficient information of an adjacent block. An image encoding method which is a condition for determining whether or not. Deriving partition structure candidate determination information from a reference block;
Determining whether to use the split target coding unit as an LCU split structure candidate using a first condition calculated based on split structure candidate determination information derived from the reference block and information derived from split target coding units;
Deriving partition structure candidate determination information from an adjacent block; And
Determining whether to use the split target coding unit as an LCU split structure candidate using a second condition calculated based on split structure candidate determination information derived from the neighboring block and information derived from split target coding units; Image coding method comprising. The method of claim 7, wherein the first condition,
Information on at least one of a partition depth of a block included in a call LCU, motion information of a prediction unit included in a call LCU, reference index information of coding units in a call LCU, and transform coefficient information of coding units in a call LCU And a condition for determining whether to use the split target coding unit as an LCU partition structure candidate. The method of claim 7, wherein the second condition,
Whether to use the split target coding unit as an LCU partition structure candidate using at least one of depth information of an adjacent block, motion information of an adjacent block, reference index information of an adjacent block, and transform coefficient information of an adjacent block. An image encoding method which is a condition for determining whether or not. Deriving partition structure candidate determination information from a reference block; And
And determining whether to decode the split information of the split target coding unit using a predetermined condition calculated based on the derived split structure candidate determination information and the information derived from the split target coding unit. The method of claim 10, wherein the predetermined condition,
Information on at least one of a partition depth of a block included in a call LCU, motion information of a prediction unit included in a call LCU, reference index information of coding units in a call LCU, and transform coefficient information of coding units in a call LCU And a condition for determining whether to use the split target coding unit as an LCU partition structure candidate. Deriving partition structure candidate determination information from an adjacent block; And
And determining whether to decode the split information of the split target coding unit using a predetermined condition calculated based on the derived split structure candidate determination information and the information derived from the split target coding unit. The method of claim 12, wherein the predetermined condition is
Whether to use the split target coding unit as an LCU partition structure candidate using at least one of depth information of an adjacent block, motion information of an adjacent block, reference index information of an adjacent block, and transform coefficient information of an adjacent block. Image decoding method that is a condition for determining whether or not. Deriving partition structure candidate determination information from a reference block;
Determining whether to decode the split information of the split target coding unit using the first condition calculated based on the split structure candidate determination information derived from the reference block and the information derived from the split target coding unit;
Deriving partition structure candidate determination information from an adjacent block; And
Determining whether to decode the split information of the split target coding unit by using the second condition calculated based on the split structure candidate determination information derived from the neighboring block and the information derived from the split target coding unit. Image Decoding Method. The method of claim 14, wherein the first condition,
Information on at least one of a partition depth of a block included in a call LCU, motion information of a prediction unit included in a call LCU, reference index information of coding units in a call LCU, and transform coefficient information of coding units in a call LCU And a condition for determining whether to use the split target coding unit as an LCU partition structure candidate. The method of claim 14, wherein the second condition,
Whether to use the split target coding unit as an LCU partition structure candidate using at least one of depth information of an adjacent block, motion information of an adjacent block, reference index information of an adjacent block, and transform coefficient information of an adjacent block. Image decoding method that is a condition for determining whether or not. The LCU partitioning structure is obtained by receiving the partition structure candidate determination information from a reference block or an adjacent block and using the predetermined condition calculated based on the partition structure candidate determination information derived and information derived from the splitting target coding unit. A determination unit determining whether to use as a candidate; And
And an LCU partition structure candidate determiner that determines an LCU partition structure candidate based on a result determined by the determination unit. The method of claim 17, wherein the predetermined condition is
Split depth of a block included in a call LCU, motion information of a prediction unit included in a call LCU, reference index information of coding units in a call LCU, transform coefficient information of coding units in a call LCU, or depth information of an adjacent block And at least one condition among at least one of motion information of adjacent blocks, reference index information of adjacent blocks, and transform coefficient information of adjacent blocks. The split information of the split target coding unit is obtained by receiving split structure candidate determination information from a reference block or an adjacent block and using predetermined conditions calculated based on the split structure candidate determination information derived and information derived from the split target coding unit. A determination unit determining whether to decode; And
And a CU partitioning information decoder for decoding the LCU partitioning structure based on the result determined by the determination unit. The method of claim 19, wherein the predetermined condition is
Split depth of a block included in a call LCU, motion information of a prediction unit included in a call LCU, reference index information of coding units in a call LCU, transform coefficient information of coding units in a call LCU, or depth information of an adjacent block And at least one condition among at least one of motion information of a neighboring block, reference index information of a neighboring block, and transform coefficient information of the neighboring block.

Images