WO2012050301A2

WO2012050301A2 - Method for encoding and decoding image and device using same

Info

Publication number: WO2012050301A2
Application number: PCT/KR2011/006538
Authority: WO
Inventors: 박준영; 김정선; 박승욱; 성재원; 전병문; 임재현; 전용준; 최영희
Original assignee: 엘지전자 주식회사
Priority date: 2010-10-14
Filing date: 2011-09-02
Publication date: 2012-04-19
Also published as: WO2012050301A3

Abstract

The present invention relates to a method for encoding/decoding image information, and according to the present invention, current block information is induced from collocated block information, thereby reducing the amount of transmitted information. According to the present invention, a method for encoding image information comprises the steps of: determining to induce current block information from collocated block information; encoding the current block information by using the collocated block information; and transmitting indicative information on the induction of said current block information from said collocated block information.

Description

Image coding and decoding method and apparatus using same

The present invention relates to image compression technology, and more particularly, to a method and apparatus for inter picture prediction.

Recently, the demand for high resolution and high quality images is increasing in various application fields. However, as the resolution and quality of an image become higher, the amount of information on the image also increases. Therefore, when image information is transmitted using a medium such as a conventional wired / wireless broadband line or when image information is stored using an existing storage medium, information transmission cost and storage cost increase.

High-efficiency image compression technology can be used to effectively transmit, store, and reproduce high-resolution, high-quality video information.

In order to increase the efficiency of image compression, inter prediction and intra prediction may be used. In the inter prediction method, a pixel value of a current picture is predicted by referring to information of another picture, and in an intra prediction method, a pixel value is predicted by using a correlation between pixels in the same picture.

An object of the present invention is to provide a method for deriving information of a current block from information of a collocated block.

An object of the present invention is to provide a parameter set and syntax structure for deriving information of a current block by using information of a co-located block.

An object of the present invention is to provide a method for variously selecting a target picture to which a same position block belongs.

An object of the present invention is to provide a method of considering a temporal level between a target picture and a current picture in a process of deriving information of a current block by using information of the same position block.

An object of the present invention is to provide a method for adaptive use, such as using some of the information of the same position block or modifying it.

An object of the present invention is to provide a method for adaptively changing the order of information transmitted from an encoder to a decoder.

(1) An embodiment of the present invention is to determine to derive the information of the current block from the information of the same position block, encoding the information of the current block using the information of the same position block and the same position And transmitting indication information about deriving the information of the current block from the information of the block.

(2) In (1), the information of the current block derived from the same position block may be partition information of the current block.

(3) In (1), the indication information may be a flag indicating to derive information of the current block from information of the same position block.

(4) In (1), the indication information may be transmitted only when the current picture to which the current block belongs is in the inter prediction mode.

(5) In (1), the target picture to which the co-located block belongs is a picture coded immediately before the current picture to which the current block belongs, a picture that has the same temporal level and has the closest temporal distance from the current picture among the already-encoded pictures, The most recently coded picture that has the same temporal level as the current picture among pictures that are already coded, the one that has the smaller temporal level and has the shortest temporal distance than the current picture among pictures that are already coded, or the picture that is currently being coded among pictures that are already coded. The temporal level may be smaller and one of the most recently coded pictures.

(6) In (1), the indication information indicates that, when the temporal level of the current picture to which the current block belongs and the target picture to which the same position block belongs is different, the depth of the coding unit having the maximum depth is increased by a predetermined size. The dividing state of the co-located block may be adjusted so as to decrease the instruction to use the dividing state of the current block.

(7) In (1), the instruction information instructs to use the division state of the same position block as the division information of the current block up to a predetermined depth, and separate for the current block for the division exceeding the predetermined depth. The split information may be transmitted.

(8) In (1), the indication information instructs to use the division state of the same position block as the division information of the current block, and for the additional division performed based on the division state of the same position block, The split information can be transmitted.

(9) Another embodiment of the present invention includes the steps of constructing information for predicting a pixel value of a current block and transmitting the configured information, wherein the configured information is information about a prediction mode of the current block. Is an image information encoding method configured to be transmitted before segmentation information of the current block.

(10) Another embodiment of the present invention provides a method for deriving information of a current block and information of the derived current block according to the indication information about deriving information of a current block using information of a co-located block. Predicting a pixel value of the current block based on the image information decoding method.

(11) In (10), the indication information may be a flag indicating to derive information of the current block from information of the same location block.

(12) In (10), the information of the current block derived from the same position block may be partition information of the current block.

(13) In (10), the target picture to which the co-located block belongs is a picture that is coded immediately before the current picture to which the current block belongs, a picture that has the same temporal level and has the closest temporal distance from the current picture among the already coded pictures, and is already The most recent coded picture that has the same temporal level as the current picture among the coded pictures, the picture that has a smaller temporal level and has the closest temporal distance than the current picture among the previously coded pictures, or the picture that is currently coded among the pictures that are already coded. The temporal level may be one of the smallest and most recently coded pictures.

(14) In (10), when the temporal level of the current picture to which the current block belongs and the target picture to which the same position block belongs is different, the same position so that the depth of the coding unit having the maximum depth is reduced by a predetermined size. The partitioned state of the block can be adjusted and used as the partitioned state of the current block.

(15) In (10), the partitioning state of the same position block is used as the partitioning information of the current block up to a predetermined depth, and the partitioning beyond the predetermined depth is based on separately received partitioning information of the current block. Division can be performed.

(16) In (15), the predetermined depth may be one step lower than the maximum depth of the current block.

(17) In (10), the partitioning state of the same location block is used as the partitioning information of the current block, and the additional partitioning performed from the partitioning state of the same location block may be performed based on the separately received partitioning information. .

(18) In (10), the information of the current block can be derived from the information of the same block only when the current picture to which the current block belongs and the temporal level of the target picture to which the same location block belongs are the same.

(19) Another embodiment of the present invention includes receiving information for predicting a pixel value of a current block and predicting a pixel value of the current block according to the received information, wherein the current block In the step of predicting a pixel value of, the image information encoding method does not apply the split information of the current block when the received information indicates that the current block is predicted to be a skip mode.

(20) In another embodiment of the present invention, there is provided a method of determining information derived from information of a base co-located block, encoding information of the current layer using information of the base layer, and And transmitting a parameter set for deriving information of the current layer from information of a base layer, wherein the parameter set is a flag and a base indicating to derive information of the current layer using information of the base layer. Information on how to use the direction of the layer and / or the information of the base layer, and information on how to use the information of the base layer indicates any one of full use, partial use, and modified use of the base layer information. And a method of using the information of the base layer If the information derived according to the information is different from the information of the current layer, the video information encoding method is characterized by transmitting additional information for deriving the information of the current layer.

(21) In another embodiment of the present invention, the method comprises: receiving first information for deriving information of a current layer using information of a base layer, and based on the first information, the current layer from information of the base layer. Deriving information of the current layer and predicting pixel values of the current layer based on the derived information of the current layer, wherein the first information includes information of the current layer using information of the base layer. A flag indicating to induce and information on the direction of the base layer and / or a method of using the information of the base layer, wherein the information on the method of using the information of the base layer includes full use, partial use of the base layer information. And using one of the modifications, and using the information of the base layer. If the derived information and the information of the current layer are different according to the related information, the information of the current layer is derived based on additional information for deriving information of the current layer and information derived from the base layer. The video information decoding method.

According to the present invention, since the information of the current block can be derived using the information of the same position block, the amount of information transmitted can be reduced.

According to the present invention, since the target picture to which the same position block belongs can be variously selected, the same position block suitable for deriving the information of the current block can be effectively selected.

According to the present invention, since the information of the same position block is used in consideration of the temporal level of the target picture and the current picture, the information of the current block can be derived more accurately.

According to the present invention, since the order of information transmitted from the encoder to the decoder can be adaptively changed, the amount of information and the amount of computation processed by the decoder can be reduced.

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

2 is a conceptual diagram schematically illustrating a prediction unit according to an embodiment of the present invention.

3 is a diagram schematically illustrating an example of a quad tree structure of a processing unit in a system to which the present invention is applied.

4 is a block diagram schematically illustrating an image decoder according to an embodiment of the present invention.

5 is a conceptual diagram schematically illustrating a prediction unit of an image decoder according to an embodiment of the present invention.

FIG. 6 is a diagram schematically illustrating candidate blocks that may be used in inter prediction of a current block.

FIG. 7 is a diagram schematically illustrating division of a current LCU by using a split state of the same location LCU.

FIG. 8 is a flowchart schematically illustrating an example of a method of encoding using a flag indicating whether to use information of an identical location LCU in a system to which the present invention is applied.

9 is a flowchart schematically illustrating a method of decoding information of a current LCU using information of a co-located LCU in a system to which the present invention is applied.

10 is a flowchart schematically illustrating an example of a method for transmitting information by configuring a parameter set in an encoder to which the present invention is applied.

[Correction under Rule 91 14.10.2011]
11 is a flowchart schematically illustrating an example of a method of performing decoding by using a parameter set received by a decoder to which the present invention is applied.

12 is a flowchart schematically illustrating an example of a method for configuring and transmitting information about a current LCU in an encoder to which the present invention is applied.

FIG. 13 is a flowchart schematically illustrating decoding of a coding unit using a parameter set received by a decoder to which the present invention is applied.

14 is a conceptual diagram schematically illustrating a hierarchical B structure.

15 is a view schematically showing a divided state of the same position LCU.

FIG. 16 is a diagram schematically showing a derivation state of a current LCU using information of the same location LCU.

17 schematically illustrates an example of a split state of the same position LCU.

18 shows an example of a split state of a current LCU derived using information of the same location LCU.

19 is a flowchart schematically illustrating an example of a method for configuring a parameter set in an encoder to which the present invention is applied and transmitting the same to a decoder.

20 is a flowchart schematically illustrating an example of a method of decoding a coding unit by receiving a parameter set in a decoder to which the present invention is applied.

21 schematically illustrates an example of a split state with respect to the same position LCU.

FIG. 22 schematically illustrates an example in which the split state for the current LCU is further divided after being derived from the same position LCU.

23 schematically illustrates an example of a split state for the same position LCU.

FIG. 24 schematically illustrates an example in which the split state of the current LCU is finally divided by additional split information after being derived from the split state of the same position LCU up to a predetermined depth.

FIG. 25 is a conceptual diagram schematically illustrating a method for additionally transmitting additional partition information by using split information of the same location LCU in a system to which the present invention is applied.

FIG. 26 is a flowchart schematically illustrating an example of a method for configuring a parameter set in an encoder to which the present invention is applied and transmitting the same to a decoder.

27 is a flowchart schematically illustrating an example of a method of decoding a coding unit by receiving a parameter set in a decoder to which the present invention is applied.

28 is a block diagram schematically illustrating an operation of an encoder in a system to which SVC is applied.

29 is a conceptual diagram schematically illustrating an example of inter-layer prediction.

30 is a flowchart schematically illustrating an example of a method for configuring a parameter set in an encoder to which the present invention is applied and transmitting the same to a decoder.

31 is a flowchart schematically illustrating an example of a method of receiving and decoding a parameter set in a decoder to which the present invention is applied.

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

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

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

1 is a block diagram schematically illustrating an image encoding apparatus (encoder) according to an embodiment of the present invention. Referring to FIG. 1, the image encoding apparatus 100 may include a picture splitter 105, a predictor 110, a transformer 115, a quantizer 120, a realigner 125, and an entropy encoder 130. , An inverse quantization unit 135, an inverse transform unit 140, a filter unit 145, and a memory 150.

The picture dividing unit 105 may divide the input picture into at least one processing unit. In this case, the processing unit may be a prediction unit (PU), a transform unit (TU), or a coding unit (CU).

The predictor 110 includes an inter prediction unit for performing inter prediction and an intra prediction unit for performing intra prediction. The prediction unit 110 generates a prediction block by performing prediction on the processing unit of the picture in the picture division unit 105. The processing unit of the picture in the prediction unit 110 may be a coding unit, a transformation unit, or a prediction unit. In addition, it is possible to determine whether the prediction performed on the processing unit is inter-screen prediction or intra-screen prediction, and determine specific contents (eg, prediction mode, etc.) of each prediction method. In this case, the processing unit in which the prediction is performed may differ from the processing unit in which the prediction method and the details are determined. For example, the method of prediction and the prediction mode are determined in units of prediction units, and the performance of prediction may be performed in units of transform units. The residual value (residual block or residual signal) between the generated prediction block and the original block is input to the converter 115. In addition, prediction mode information and motion vector information used for prediction are encoded by the entropy encoder 130 along with the residual value and transmitted to the decoder.

The transform unit 115 performs a transform on the residual block in transform units and generates transform coefficients. The transform unit in the transform unit 115 may be a transform unit and may have a quad tree structure. In this case, the size of the transform unit may be determined within a range of a predetermined maximum and minimum size. The transform unit 115 may transform the residual block using a discrete cosine transform (DCT) and / or a discrete sine transform (DST).

The quantization unit 120 may generate quantization coefficients by quantizing the residual values transformed by the transformation unit 115. The value calculated by the quantization unit 120 is provided to the inverse quantization unit 135 and the reordering unit 125.

The reordering unit 125 rearranges the quantization coefficients provided from the quantization unit 120. By rearranging the quantization coefficients, the efficiency of encoding in the entropy encoder 130 may be increased. The reordering unit 125 may rearrange the quantization coefficients in the form of a two-dimensional block into a one-dimensional vector form through a coefficient scanning method. The reordering unit 125 may increase the entropy coding efficiency of the entropy encoder 130 by changing the order of coefficient scanning based on probabilistic statistics of coefficients transmitted from the quantization unit.

The entropy encoder 130 may perform entropy encoding on the quantized coefficients rearranged by the reordering unit 125. Entropy encoding may use, for example, an encoding method such as Exponential Golomb, Context-Adaptive Variable Length Coding (CAVLC), or Context-Adaptive Binary Arithmetic Coding (CABAC). The entropy encoder 130 may include quantization coefficient information, block type information, prediction mode information, division unit information, prediction unit information, transmission unit information, and motion vector of the coding unit received from the reordering unit 125 and the prediction unit 110. Various information such as information, reference picture information, interpolation information of a block, and filtering information may be encoded.

In addition, if necessary, the entropy encoder 130 may apply a constant change to a transmitted parameter set or syntax.

The inverse quantization unit 135 inverse quantizes the quantized values in the quantization unit 120, and the inverse transformer 140 inversely transforms the inverse quantized values in the inverse quantization unit 135. The residual value generated by the inverse quantization unit 135 and the inverse transformer 140 may be combined with the prediction block predicted by the prediction unit 110 to generate a reconstructed block.

The filter unit 145 may apply a deblocking filter and / or an adaptive loop filter (ALF) to the reconstructed picture.

The deblocking filter may remove block distortion generated at the boundary between blocks in the reconstructed picture. The adaptive loop filter (ALF) may perform filtering based on a value obtained by comparing the reconstructed image with the original image after the block is filtered through the deblocking filter. ALF may be performed only when high efficiency is applied.

Meanwhile, the filter unit 145 may not apply filtering to the reconstructed block used for inter prediction.

The memory 150 may store the reconstructed block or the picture calculated by the filter unit 145. The reconstructed block or picture stored in the memory 150 may be provided to the predictor 110 that performs inter prediction.

A coding unit (CU) is a unit in which coding / decoding of a picture is performed and may be divided with a depth based on a quad tree structure. Unlike the existing macro block, the coding unit may have various sizes such as 64x64, 32x32, 16x16, and 8x8. The largest coding unit is called a largest coding unit (LCU).

In order to indicate various sizes of coding unit sizes, an LCU size, a maximum depth at which a coding unit may be divided, and the like may be defined in a sequence parameter set (SPS) transmitted from an encoder to a decoder.

The encoder may transmit information about a largest coding unit (LCU) and a minimum coding unit (SCU) to the decoder. Information (depth information) regarding the number of splittable times together with information about the maximum coding unit and / or the minimum coding unit may be transmitted to the decoder. Information on whether the coding unit is split based on the quad tree structure may be transmitted from the encoder to the decoder through flag information such as a split flag.

Referring to FIG. 2, the predictor 200 may include an inter prediction unit 210 and an intra prediction unit 220.

The inter prediction unit 210 may generate a prediction block by performing prediction based on information of at least one picture of a previous picture and / or a subsequent picture of the current picture. In addition, the intra prediction unit 220 may generate a prediction block by performing prediction based on pixel information in the current picture. The inter prediction unit 210 may select a reference picture with respect to the prediction unit and select a reference block having the same size as that of the prediction unit in units of integer pixel samples. Next, the inter prediction unit 210 generates a prediction block in which the residual signal with the current prediction unit is minimized and the size of the motion vector is also minimized. The prediction block may be generated in sub-integer sample units such as 1/2 pixel sample unit and 1/4 pixel sample unit.

In this case, the motion vector may be expressed in units of integer pixels or less. For example, the motion vector may be expressed in units of 1/4 pixels for luminance pixels and in units of 1/8 pixels for chrominance pixels.

Information such as an index, a motion vector (ex. Motion Vector Predictor), and a residual signal of the reference picture selected by the inter prediction unit 210 is encoded and transmitted to the decoder.

The maximum coding unit (LCU) 300 may have a hierarchical structure composed of smaller coding units 310 through division, and the hierarchical structure of the coding unit may be based on size information, depth information, split flag information, and the like. Can be specified. The size information of the largest coding unit, the split depth information, and whether the current coding unit is split may be included in a sequence parameter set (SPS) on the bit stream and transmitted to the image decoder. However, since the minimum coding unit is no longer divided into small coding units, the split flag of the coding unit may not be transmitted to the minimum coding unit.

Meanwhile, which prediction between inter prediction and intra prediction may be performed may be determined in units of coding units. When performing inter prediction, inter prediction may be performed in units of prediction units. When performing intra prediction, a prediction mode may be determined in units of prediction units, and prediction may be performed in units of prediction units. In this case, a prediction mode may be determined in units of prediction units, and intra prediction may be performed in units of transformation units.

Referring to FIG. 3, in the case of intra prediction, the prediction unit 320 may be 2N × 2N or N × N (N is an integer), and in case of inter prediction, the prediction unit 330 may be 2N ×. 2N, 2N × N, N × 2N, or N × N. In this case, for example, N × N may be determined to be applied only to the minimum size coding unit or to be applied only to intra prediction. In addition to the size of the illustrated prediction block, N × mN, mN × N, 2N × mN, or mN × 2N (m <1) may be further defined and used.

4 is a block diagram schematically illustrating an image decoder according to an embodiment of the present invention. Referring to FIG. 4, the image decoder 400 includes an entropy decoder 410, a reordering unit 415, an inverse quantizer 420, an inverse transform unit 425, a predictor 430, and a filter unit 435. Memory 440 may be included.

When an image bit stream is input from the image encoder, the input bit stream may be decoded according to a procedure in which image information is processed by the image encoder.

For example, when variable length coding (VLC, hereinafter called 'VLC') is used to perform entropy encoding in the image encoder, the entropy decoder 410 is also identical to the VLC table used in the encoder. It can be implemented as a VLC table to perform entropy decoding. Even when CABAC is used to perform entropy encoding in the image encoder, the entropy decoding unit 410 may perform entropy decoding using CABAC correspondingly.

Information for generating a prediction block among the information decoded by the entropy decoder 410 may be provided to the predictor 430, and a residual value of which entropy decoding is performed by the entropy decoder may be input to the reordering unit 415.

The reordering unit 415 may reorder the bit stream deentropy decoded by the entropy decoding unit 410 based on a method of reordering the image encoder. The reordering unit 415 may reorder the coefficients expressed in the form of a one-dimensional vector by restoring the coefficients in the form of a two-dimensional block. The reordering unit 415 may receive the information related to the coefficient scanning performed by the encoder and perform the rearrangement by performing a reverse scanning method based on the scanning order performed by the corresponding encoder.

The inverse quantization unit 420 may perform inverse quantization based on the quantization parameter provided by the encoder and the coefficient values of the rearranged block.

The inverse transform unit 425 may perform inverse DCT and / or inverse DST on DCT and DST performed by the transform unit of the encoder with respect to the quantization result performed by the image encoder. The inverse transform may be performed based on a transmission unit determined by the encoder or a division unit of an image. In the encoder, the DCT and / or DST may be selectively performed according to a plurality of pieces of information, such as a prediction method, a size and a prediction direction of the current block, and the inverse transformer 425 of the decoder is performed by the transformer of the encoder. Inverse transformation may be performed based on the transformation information.

The prediction unit 430 may generate the prediction block based on the prediction block generation related information provided by the entropy decoding unit 410 and previously decoded block and / or picture information provided by the memory 440. The reconstruction block may be generated using the prediction block generated by the predictor 430 and the residual block provided by the inverse transform unit 425.

The reconstructed block and / or picture may be provided to the filter unit 435. The filter unit 435 applies deblocking filtering, sample adaptive offset (SAO), and / or adaptive loop filtering to the reconstructed blocks and / or pictures.

The memory 440 may store the reconstructed picture or block to use as a reference picture or reference block, and provide the reconstructed picture to the output unit.

Referring to FIG. 5, the predictor 500 may include an intra prediction unit 510 and an inter prediction unit 520.

The intra prediction unit 510 may generate a prediction block based on pixel information in the current picture when the prediction mode for the prediction unit is an intra prediction mode (intra prediction mode).

The inter screen prediction unit 520 may include motion information required for inter screen prediction of the current prediction unit provided by the image encoder when the prediction mode for the corresponding prediction unit is an inter prediction mode (inter prediction mode). An inter-screen prediction of the current prediction unit may be performed based on information included in at least one of a previous picture or a subsequent picture of the current picture including the current prediction unit by using information about a motion vector, a reference picture index, and the like. . In this case, the motion information may be determined in response to the skip flag, the merge flag, and the like of the received coding unit.

In FIG. 5, for convenience of description, the prediction unit 500 has been described as including a configuration for each function. However, the present invention is not limited thereto, and the prediction unit may be implemented as a single configuration for performing the above-described functions.

Referring to FIG. 6, when the inter prediction is applied, the current block 600 may select an available block from blocks of the upper region B and the left region A. FIG. The motion information of the current block may be predicted using the motion information of the selected block. A block that is an intra prediction mode among neighboring blocks of the current block may be excluded from the available block.

In the case of selecting a candidate block from the neighboring blocks of the current block 600, the available candidate blocks may be selected while searching each area A and B in a predetermined order, and a block located at a specific position of each area may be selected. It can also be fixedly used as a candidate block. For example, at least one of

blocks

650, 660, and 670 located at each corner of the current block 600 may be selected as a candidate block. In addition, the left block 630 and / or the right block 640 except for the corner blocks 660 and 670 among the blocks of the upper area B may be selected as candidate blocks, and the corner blocks among the blocks of the left area A may be selected as candidate blocks. Except for 650, the upper block 610 and / or the lower block 620 may be selected as candidate blocks. The motion information of the current block may be predicted using a block selected by a predetermined criterion among candidate blocks selected in each area (corner, left, top).

In this case, in addition to the picture in which the current block 600 is located, a block having the same position as the current block (co-located block 680, hereinafter, referred to as 'same location block' for convenience of description) in another picture is included in the candidate block. It can be used to predict the motion information of the current block. For convenience of description, a picture to which the same location block belongs is referred to as a " same location picture. &Quot;

On the other hand, not only the motion information of the current block but also a method of using the information of the same position block to derive various information of the current block may be considered.

For example, as described above, the current LCU may be recursively divided according to the split flag included in the sequence parameter set. When the split flag (eg, split_coding_unit_flag) of the coding unit has a value of 1, the current coding unit may be split. If the split flag value is 0, the current coding unit may not be split. The division may be divided into four symmetric subunits or asymmetric subunits. Therefore, a split flag is required for every LCU or every coding unit, and this syntax may affect the performance of encoding / decoding.

In the present invention, by using the partition information of the same position block, it is possible to reduce the number of transmission of the partition flag and improve coding performance. Hereinafter, unless otherwise stated, the term "transfer" in the present specification means to be transmitted from the encoder to the decoder.

<동일 위치 블록의 분할 정보를 이용할 것인지를 지시하는 플래그><Flag indicating whether to use split information of the same location block>

In order to use the split state of the LCU among the split information of the same position LCU, a flag indicating whether to use the split state of the same position LCU is defined as it is. As an example of this flag, a flag such as use_coloc_LCU_split_info_flag may be defined and used.

Table 1 schematically shows an example of a syntax structure indicating whether a block is divided using a split flag in a conventional system. In Table 1, the division of the coding unit is indicated by a flag as an example.

Referring to Table 1, when the unit size of the current coding unit is used at the position of the current coding unit, a split flag (split_coding_unit_flag) is exceeded when the size of the current coding unit is larger than the minimum coding unit size. Through, it may be indicated whether to split the coding unit. Here, x0 and y0 may indicate the positions of the top left samples of the coding unit with respect to the top-left samples of the current picture.

Meanwhile, the split information may be transmitted in a syntax of a coding tree level instead of the coding unit level. Table 2 schematically shows an example of an encoding tree syntax structure indicating whether a block is divided using a split flag in a conventional system.

In the example of Table 2, splitting information (split_coding_unit_flag) relating to a coding unit is transmitted as the splitting target block as the syntax of the coding tree level.

On the other hand, as described above, instead of directly transmitting the split information for the current LCU, it is possible to transmit a flag indicating to use the split information (divided state) of the same location LCU. For example, a flag indicating whether to use information of the same location LCU may be defined, such as use_coloc_LCU_split_info_flag.

Table 3 schematically illustrates an example of a coding unit level syntax structure indicating whether a split state (division information) of the same location LCU is currently used for the LCU in a system to which the present invention is applied.

In addition, a flag indicating whether to use split information of the same position LCU may be transmitted in a syntax of a coding tree level. Table 4 schematically illustrates an example of a coded tree level syntax structure indicating, via a flag, whether to use a split state of a co-located LCU for a current LCU in a system to which the present invention is applied.

In the case of indicating through the flag that the division state of the same position LCU is used as it is in the current LCU, the division information for the coding unit in the current LCU may be derived from the division information (division state) of the same position LCU. For example, referring to Table 3, when the value of use_coloc_LCU_split_info_flag is 1, split information for a coding unit in the current LCU, for example, a value of split_coding_unit_flag for each coding unit in the current LCU, may be derived.

That is, when a flag indicates that the division state of the same position LCU is to be used as it is in the current LCU, each sub-combination unit in the current LCU may be divided in the same manner as the coding unit in the same position LCU.

In this case, the co-location LCU from which the split information of the current LCU is derived may be specified as the co-location LCU of the picture closest to the current picture in time among the pictures that are already encoded.

FIG. 7 is a diagram schematically illustrating division of a current LCU by using a split state of the same location LCU. FIG. 7A schematically illustrates a coding unit division state of a picture that is closest in time to a current picture among encoded pictures. 7B schematically illustrates a coding unit division state of a current picture.

Referring to FIG. 7, it can be seen that the split state of the first LCU is the same between the same position picture and the current picture. Therefore, the division state of the same position LCU can be used as it is through the flag, and the division state of the second LCU of the same position picture can be used as the division state of the second LCU of the current picture.

In FIG. 7, to indicate a split state of the first LCU of the current picture, assuming that the maximum dividable depth is greater than 2 in the current LCU, whether the depth 0 is divided, whether the depth 1 is divided, and whether the depth 2 is divided Nine flags should be sent to indicate. However, when using the split state of the same position LCU as the current LCU, only one flag (use_coloc_LCU_split_info_flag) needs to be transmitted.

In the case of the 0th and 2nd LCUs of the current picture, one flag (use_coloc_LCU_split_info_flag) indicates that the split state of the co-located LCU is not applied to the current LCU, and then indicates information indicating the split state of the current LCU. For example, a split flag (split_coding_unit_flag) for a coding unit may be transmitted. In addition, when division information (divided state) of the same position block is not applied to the current block like the 0th and 2nd LCUs of the current picture, the flag indicates that the division state of the same position LCU is not applied to the current LCU. In addition, the split information on the coding unit of the current LCU may be transmitted.

Referring to FIG. 8, the encoder divides the coding unit starting from the LCU (S810). The splitting of the coding unit may be performed by the picture splitter in the encoder. The encoder may split the coding unit and may split the prediction unit and the transform unit.

The encoder may configure a syntax to be transmitted to the decoder, that is, a parameter set, based on the split information of the current LCU (S820). The encoder may determine whether to transmit information about the split state of the current LCU using the split state of the same location LCU. When the split state information of the current LCU is delivered using the split information of the same position LCU, the encoder may configure a parameter set including a flag indicating to use the split information of the same position LCU.

The encoder transmits the configured parameter set to the decoder (830). The information in the parameter set may be encoded and transmitted.

Referring to FIG. 9, the decoder receives a parameter set from an encoder (S910). The parameter set may be encoded and transmitted to the decoder in a bitstream. The decoder decodes the parameter set to obtain information necessary to decode the information about the image of the current LCU.

The received parameter set includes a flag indicating whether to derive the information of the current LCU using the information of the same location LCU. For example, the parameter set may include a flag indicating to use the split state of the same position LCU as it is.

The decoder decodes information about the current CU (S920). For example, when a flag indicating to use the split state of the same position LCU is transmitted, the decoder configures the split state of the current LCU to be the same as the split state of the same position LCU, and decodes based on the split state of the current LCU. Necessary procedures, such as predictions, can be performed.

<분할을 지시하는 플래그와 다른 신택스들의 위치> <Location of flags and other syntaxes indicating partitioning>

On the other hand, with respect to the division of the current block, conventionally, when recursively splitting the LCU through the division flag, other syntaxes for the coding unit are transmitted only when a syntax is transmitted that no longer divides the current coding block. It became. In this case, the syntaxes transmitted after indicating that the current coding unit is no longer divided include a skip flag (skip_flag) indicating whether to apply a skip mode, a syntax (pred_mode) indicating a prediction mode, and the like.

If the skip flag indicates 1, the current coding unit may be encoded using the skip mode. For example, the prediction unit of the current coding unit may predict the information of the reference block indicated by the index transmitted later by using the information of the current coding unit and not transmitting the residual signal. In this case, the prediction unit of the current coding unit may be a partition obtained by dividing the current coding unit. The index indicating the reference block may be a merge index. In the case of using the inter prediction, the skip mode may not transmit the residual signal by using the motion information of the block indicated by the merge index as the motion information of the current block and the pixel value of the prediction block as the pixel value of the current block. In the merge mode, the motion information of the block indicated by the merge index may be used as the motion information of the current block, and the residual signals of the prediction block and the current block may be transmitted.

A syntax indicating a prediction mode may indicate whether a prediction mode of a current block is an intra prediction mode or an inter prediction mode. In addition, the syntax pred_mode may indicate information on the partition state of the current block together with the prediction mode.

Table 5 schematically shows a conventional syntax structure indicating other information about a current block after indicating whether a current coding block is to be split through a split flag. In Table 5, the skip flag is described as an example of information other than division.

Referring to Table 5, when the split information about the current coding unit is informed to split_coding_unit_flag and no further split is performed (split_coding_unit_flag = 0), the slice (slice_type! = I) to which the inter prediction mode is applied Whether to apply the skip mode is indicated in the parameter set for the prediction unit.

In this regard, before indicating with a flag whether a current coding block is to be split, a method of first transmitting other information about the current coding block may be considered. For example, before indicating whether the current coding unit is split by the flag split_coding_unit_flag, a skip flag (skip_flag) may be used to indicate whether the current coding unit is to be coded in the skip mode.

Table 6 schematically shows an example of a syntax structure for first transmitting other information about a current coding block before indicating with a flag whether a current coding block is to be divided in a system to which the present invention is applied.

Referring to Table 6, a skip flag (skip_flag) first indicates whether a skip mode is applied to an image slice (slice_type! = I) to which inter prediction is applied, and then a split flag (split_coding_unit_flag) indicates whether a current coding unit is split. Let me know.

Therefore, when the skip mode is applied, the split information may not be transmitted unnecessarily with the split flag, thereby reducing the total amount of information.

Furthermore, according to the characteristics of the input image and the environment of encoding / decoding, a splitting flag (for example, split_coding_unit_flag) and a splitting flag for transmitting split information in order to increase the efficiency of encoding / decoding and to reduce information processing cost and / or information amount, etc. It may be considered to adaptively change the order of syntaxes for transmitting other information (eg, a skip flag indicating whether to apply ski mode or skip_flag).

For example, if the splitting probability of the current block is high, the splitting flag may be transmitted first and then the skip flag is transmitted. If the splitting probability is low, the skipping flag may be transmitted first, followed by the splitting flag. In more detail, the split flag and the skip flag may be adaptively determined according to the following conditions.

(1) Determining the depth of the current coding unit as a condition

If the depth of the coding unit is shallow, it is likely to be divided. For example, it may be said that a coding unit having a depth of zero has a greater possibility of splitting than a coding unit having a depth of one.

Therefore, if a predetermined reference depth (depth_th) is determined and the depth of the current coding unit is shallower than the reference depth (the depth of the current coding unit <the reference depth), the division flag is first transmitted in the transmission of the parameter set, and the If the depth is equal to or deeper than the reference depth (the depth of the current coding unit ≥ the reference depth), the skip flag may be transmitted first. For example, when the depth of the current coding unit is 0, the split flag (split_coding_unit_flag) is sent first when transmitting the parameter set, and when the depth of the current coding unit is equal to or greater than 1, the skip flag (skip_flag) is sent first. can do.

(2) Determine the size of the current coding unit as a condition

Similarly to the depth of the coding unit, the size of the coding unit is also likely to be divided if the size of the coding unit is large. For example, it may be said that an LCU is more likely to be split than a coding unit having a size of 32x32.

Therefore, if a predetermined reference size (size_th) is determined, and if the size of the current coding unit is larger than the reference size (size of the current coding unit <reference size), the division flag is transmitted first in the transmission of the parameter set, and the size of the current coding unit If is equal to or smaller than the reference size (size of the current coding unit ≥ reference size), the skip flag may be transmitted first. For example, if the size of the current coding unit is 32x32 or larger than 32x32, the split flag (split_coding_unit_flag) is sent first in the transmission of the parameter set, and if the size of the current coding unit is smaller than 32x32, the skip flag (skip_flag) is sent first. can do.

(3) Determine the condition of the quantization parameter of the current image slice

If the quantization parameter (QP) is small, more detailed information can be transmitted, and the possibility of partitioning the current block is high.

Therefore, if the predetermined reference quantization parameter QP_th is determined and the quantization parameter of the current coding unit is smaller than the quantization parameter (quantization parameter <reference quantization parameter of the current coding unit), the division flag is first transmitted in the transmission of the parameter set, If the quantization parameter of the current coding unit is equal to or greater than the reference quantization parameter (quantization parameter ≥ reference quantization parameter of the current coding unit), the skip flag may be transmitted first. For example, if the quantization parameter of the current coding unit is less than 30, a split flag (split_coding_unit_flag) is sent first in the transmission of the parameter set, and if the quantization parameter of the current coding unit is equal to or greater than 30, the skip flag (skip_flag) is transmitted. ) Can be sent first.

(4) a combination of the above conditions

A combination of depth, size, and quantization parameter of the coding unit may be combined to determine whether to send a split flag or skip flag first for the current coding unit.

For example, when at least two or more of the above (1) to (3) correspond to the condition, the division flag may be transmitted before the skip flag.

For example, when the depth of the current coding unit is smaller than the reference depth and the size of the current coding unit is larger than the reference size, the split flag may be transmitted before the skip flag. The split flag may be transmitted before the skip flag when the depth of the current coding unit is smaller than the reference depth and the quantization parameter of the current coding unit is smaller than the reference quantization parameter. When the size of the current coding unit is larger than the reference size and the quantization parameter of the current coding unit is smaller than the reference quantization parameter, the split flag may be transmitted before the skip flag.

In the case of using a combination of two of the three conditions (1) to (3) above, the other one may be used as a condition for acknowledging an exception. For example, even when the depth of the current coding unit is smaller than the reference depth and the size of the current coding unit is larger than the reference size, the division flag may be transmitted later than the skip flag if the quantization parameter of the current coding unit is larger than the reference quantization parameter. Even when the depth of the current coding unit is smaller than the reference depth and the quantization parameter of the current coding unit is smaller than the reference quantization parameter, the split flag may be transmitted later than the skip flag if the size of the current coding unit is smaller than the reference size. Even when the size of the current coding unit is larger than the reference size and the quantization parameter of the current coding unit is smaller than the reference quantization parameter, the split flag may be transmitted later than the skip flag if the depth of the current coding unit is larger than the reference depth.

In addition, when all of the conditions (1) to (3) are satisfied, the division flag may be transmitted before the skip flag. For example, when the depth of the current coding unit is smaller than the reference depth, the size of the current coding unit is larger than the reference size, and the quantization parameter of the current coding unit is smaller than the reference quantization parameter, the split flag may be transmitted before the skip flag.

Herein, the information on whether to apply the skip mode as the information to be determined after the split information as the information on the current coding unit other than the split information (split flag) has been described as an example, but the present invention is not limited thereto. As described above, the above-described method may be similarly applied to other information about the current coding unit, such as a syntax indicating a prediction mode.

In addition, although it has been described herein that the transmission order of the split information and the skip information is determined for the current coding unit, the present invention is not limited thereto, and the above-described method is similarly applied to other processing units (prediction unit, transformation unit). Applicable

Referring to FIG. 10, the encoder first configures a parameter set (S1010). In this case, the encoder may adaptively determine the location of the split information for the current LCU and the information about the prediction mode in the parameter set.

Methods and conditions for determining the location of split information (eg, split_coding_unit_flag) and information about a prediction mode (pred_mode) for the current LCU are as described above.

The encoder transmits the configured parameter set to the decoder (S1020). The parameter set may be encoded and transmitted along with other information.

11 is a flowchart schematically illustrating an example of a method of performing decoding by using a parameter set received by a decoder to which the present invention is applied.

Referring to FIG. 11, the decoder receives a parameter set from an encoder (S1110). The parameter set may be encoded and transmitted on the bitstream.

In the parameter set, the splitting information (for example, split_soding-unit_flag) for the current LCU and information (pred_mode) indicating which prediction mode to apply may vary depending on the situation as described above.

The decoder may decode the coding unit according to the information and / or the indication of the received parameter set (S1120). The decoder may first apply the transmitted information first, and sequentially decode the current LCU using the following information. For example, when split information of the current LCU is delivered, the current LCU is split as indicated by the information. Subsequently, when a prediction mode is transmitted for each LCU or each partition (CU / PU / TU), prediction may be performed in a corresponding region by applying the prediction mode.

On the contrary, when the prediction mode is delivered first, the prediction mode may be applied first. For example, when the prediction mode is transmitted that the skip mode is applied as inter-screen prediction (slice_type! = I & & skip_flag = 1), the encoding is no longer parsed by parsing the split information of the coding unit. The decoding may be performed by applying a skip mode by inter prediction to the unit.

<동일 위치 LCU의 대상 픽처 및 이용 가능한 정보><Target picture and available information of the same location LCU>

Meanwhile, in the above description of the present invention, the same position LCU used to derive the split state of the current LCU has been described as a picture closest in time to the current picture among previously encoded pictures, but other pictures are also the same position LCU. We can consider how to use as target of.

(1) Target picture of the same position LCU

As a picture that is a target of a co-located LCU capable of deriving information about the current LCU, the following pictures may be used in addition to the picture closest in time to the current picture among pictures that are already encoded. At this time, as described above, the LCU in the same position as the current LCU spatially in the target picture becomes the same position LCU.

1) The picture encoded immediately before the picture currently being encoded (the current picture).

2) Among pictures that are already encoded, a picture having the same temporal level and the closest temporal distance from the picture currently being encoded.

3) Among pictures that are already encoded, a picture that is currently encoded has the same temporal level and the most recently encoded picture.

4) Among pictures that are already encoded, a picture having a temporal level smaller than the picture currently being encoded and having the closest temporal distance.

5) Of the pictures that are already encoded, the temporal level is smaller than the picture currently being encoded and the most recently encoded picture.

Here, when the temporal level is encoded using a hierarchical B picture structure, the temporal level refers to the temporal layer of each picture. The temporal level is assigned an ID to a temporal layer corresponding to each temporal level. It can also be expressed as a temporal id. In the hierarchical B picture structure, the current picture does not refer to a picture of a higher level than itself. 12 is a conceptual diagram schematically illustrating a hierarchical B structure. Referring to FIG. 12, each picture 1200 to 1280 may make predictions by referring to a picture having a temporal level lower than itself or a picture at a same temporal level (pictures having a time ID less than or equal to itself). No prediction is made by referring to a picture at a higher temporal level.

On the other hand, the method of using the same position LCU to derive the split state of the current LCU is not used when the slice type of the current picture is an intra slice (Intra Slice) of the intra prediction mode, and the slice type of the current picture is inter-screen. It may be used only when a slice (inter slice) of a prediction mode is used.

In addition, the method of using the co-located LCU to derive the split state of the current LCU may be used only when the slice type of the current picture and the slice type of the co-located LCU are both slices of an inter prediction mode. have.

(2) Information available from the same location LCU

As described above, when the current LCU is encoded, the information of the current LCU may be derived / predicted from the information of the same location LCU, or the information of the same location LCU may be used as the information of the current LCU. Hereinafter, for convenience of description, the expression 'use information of the same location LCU' is used herein to mean both including deriving / predicting information of the current LCU and using information of the same location LCU as the current LCU. do. When the encoder uses information of the same position LCU, a flag indicating this may be transmitted to the decoder.

As an example of a flag indicating that information of the same location LCU is used, use_coloc_LCU_info_flag may be defined. When the information currently used in the LCU as information on the same location LCU is limited to the split information of the LCU, use_coloc_LCU_split_info_flag and use_coloc_LCU_info_flag described above may play the same role.

Hereinafter, 'use_coloc_LCU_info_flag' is used as an example of a flag indicating that information of the same location LCU is used for convenience of description.

When the value of use_coloc_LCU_info_flag is 1, the information of the current LCU, which uses the information of the same location LCU, is not transmitted separately. This is because the information of the same location LCU can be inferred or the corresponding information of the same location LUC can be used as it is. For example, if the information of the same location LCU used through use_coloc_LCU_info_flag is split information of the LCU, the split flag, for example split_coding_unit_flag, may not be transmitted.

Various information of the same location LCU as well as the split information of the LCU can be used as the information of the current LCU. In this case, the information of the same position LCU may be information of the coding unit CU in the same position LCU. The information of the coding unit in the same position LCU may be information of a specific coding unit or may be information of each coding unit. Information of a coding unit in the same position LCU may be used for a corresponding coding unit in the current LCU, or may be used for each coding unit or a specific coding unit in the current LCU. The information available is as follows:

1) Split information on a current LCU and / or a coding unit in a current LCU (hereinafter, for convenience of description, 'the current LCU and / or a coding unit in the current LCU' is referred to as a 'current coding unit'). For example, all splitting flags (split_coding_unit_flag) in a parameter set related to the current coding unit.

2) Skip information for the current coding unit. For example, all skip flags (skip_flag) in the parameter set for the current coding unit.

3) Prediction mode for the current coding unit. For example, all prediction mode syntax values (pred_mode) values in the parameter set for the current coding unit.

4) Mode for the current coding unit. For example, all mode values in the parameter set for the current coding unit. The decoder converts the received codeword into a mode table index through a variable length coding (VLC) table and converts the mode table index into a mode value through a mode table (ranking table). . The mode value includes a prediction mode (pred_mode), a flag indicating division of a prediction unit (intra_split_flag), an indicator indicating partitioning of inter prediction, an inter_partitioning_idc, a reference picture list (ref_idx_l0), an inter prediction prediction indicator (inter_pred_idx), and a transformation. It may be used to calculate a value such as a split flag (split_transform_unit_flag) for a unit. In this case, the mode value of the same unit LCU and / or the coding unit within the same position LCU (hereinafter, for convenience of description, the term 'the same position LCU and / or the coding unit within the same position LCU' is referred to as the same position coding unit) Can be used as a mode value for the current coding unit.

5) Information about the splitting of the prediction unit for the current coding unit. For example, all prediction unit splitting flags (intra_split_flag) in the parameter set for the current coding unit.

6) Partitioning information of the coding unit for inter prediction in the current coding unit. For example, in the parameter set for the current coding unit, the indicator for specifying the type of partitioning the current coding unit for the purpose of inter-picture prediction (inter_partitioning_idc)

7) All kinds of information that can be transmitted in the parameter set related to the current coding unit in addition to the above-described information.

8) Combination of the information of 1) to 7)

As described above, the information of 1) to 8) can be obtained by using the information of the same position coding unit.

When the information of 1) to 8) is applied to a coding unit other than the current LCU, it may be indicated to which coding unit in the current LCU to apply the information. Table 7 schematically illustrates an example of a syntax structure for applying the information of 1) to 8) in units of coding units in the LCU rather than the LCU level.

Referring to Table 7, the target coding unit may be designated (x0, y0) to transmit information on whether to use the information of the same position coding unit.

In the example of Table 7, a flag using information of the same position coding unit has been described at the coding unit level. However, when a coding unit unit indicates that specific information is to be used, a flag for designating target information may be used. For example, when using split information of the same position block, a flag (use_coloc_CU_split_info_flag) that specifically indicates that the information indicating whether to use is split information may be used.

On the other hand, it is also possible to specifically specify the current coding unit that can use the information of the same position coding unit. For example, the information of the same position coding unit may be used only for the coding unit that is no longer split among the current coding units. Table 8 schematically shows an example of a syntax structure indicating whether to use information of the same position coding unit for a coding unit that is no longer split among coding units in the current LCU.

In the example of Table 8, the SCU represents a coding unit that is no longer split.

13 is a flowchart schematically illustrating an example of a method for configuring and transmitting information about a current coding unit in an encoder to which the present invention is applied. In the description of FIG. 13, the coding unit is a meaning including an LCU.

Referring to FIG. 13, the encoder determines whether to use information of the same position coding unit with respect to the current coding unit (S1310). In this case, the target picture of the same position coding unit may be selected from various pictures as well as the picture encoded immediately before the current picture. Various information may be derived from the same position coding unit as well as information on the split among the information about the current coding unit. The information on the method of selecting the target picture and the information available from the same position coding unit are as described above.

Meanwhile, the method of deriving the current information from the information of the same position coding unit may be applied only when the current picture is an inter slice of the inter prediction mode.

The encoder configures a parameter set including information of the current coding unit (S1320). The encoder may include, in the parameter set, a flag indicating whether to derive information of the current coding unit by using the information of the same position coding unit. As described above, the information used from the same position coding unit is not limited to the split information. Therefore, not only the splitting information but also flags (use_coloc_LCU_info_flag and / or use_coloc_CU_info_flag) indicating whether to use more various information of the same position coding unit can be used. When the value of the flag is 1, the decoder may induce the information of the current coding unit by using the information of the same position coding unit, and when the value of the flag is 0, separate information may be transmitted to the decoder. In the case of using information of the same position coding unit, separate information about the current coding unit may not be transmitted with respect to the corresponding information. The method and syntax structure for signaling whether to use information of the same position block for the LCU and the coding unit in the LCU have been described above.

When the flag (use_coloc_LCU_info_flag and / or use_coloc_CU_info_flag) indicates to use the information of the same location coding unit, all the same location coding unit except for information transmitted separately for the LCU or a specific partition (CU / PU / TU) You can also use the information. No additional information is transmitted with respect to the information derived from the same position coding unit.

Meanwhile, even when the value of the flag (use_coloc_LCU_info_flag and / or use_coloc_CU_info_flag) is 1, the parameter set may be configured to include a flag specifically indicating information to be used.

The encoder transmits the configured parameter set to the decoder (S1330). The parameter set may be encoded and transmitted along with other information.

14 is a flowchart schematically illustrating decoding of a coding unit using a parameter set received by a decoder to which the present invention is applied. In the description of FIG. 14, the coding unit is a meaning including an LCU.

Referring to FIG. 14, the decoder may receive a parameter set from an encoder (S1410). The parameter set may be encoded and received through the bitstream. The method and syntax structure for signaling whether to use information of the same position block for the LCU and the coding unit in the LCU have been described above.

The decoder can derive information of the current coding unit from the received parameter set (S1420). The parameter set may include a flag indicating whether to use the information of the same position coding unit to derive the information of the current coding unit. When the value of the flag is 1, the decoder may derive the information of the current coding unit from the information of the same position coding unit.

The parameter set may further include a flag that specifically indicates what information of the same position coding unit to use. In addition, the parameter set may include a flag indicating to use information of the same position coding unit only when the current picture is an inter slice.

Regarding information derived from the same position coding unit, separate information may not be transmitted through a parameter set. For example, when the information derived from the information of the same position coding unit is split information, separate split information (split_coding_unit_flag) for the current coding unit may not be included in the received parameter set.

Whether the information derived from the information of the same position encoding unit is specifically indicated or whether the information is derived from the information of the same position encoding unit unless separately transmitted may be predetermined between the encoder and the decoder.

The decoder may decode the coding unit based on the derived information (S1430).

<시간적 레벨의 고려>Consideration of temporal level

On the other hand, when an image is encoded using a hierarchical B structure as described above, as shown in FIG. 12, a temporal level between a current picture and a target picture of the same position LCU may be different. .

For example, if the current picture is the picture 1220 and the target picture of the same position LCU is the picture 1240, the temporal level of the current picture is 2 and the temporal level of the target picture is 1. Also, if the current picture is the picture 1230 and the target picture of the same position LCU is the picture 1240, the temporal level of the current picture is 3 and the temporal level of the target picture is 1.

When the temporal level is different between the current picture and the target picture of the same position LCU, the statistical probability that the current LCU and the same position LCU have different coding characteristics increases. The greater the difference in temporal levels between LCUs, the greater the probability of having different coding characteristics. Therefore, when the temporal level of the same LCU and the current LCU are different, the coding efficiency of the current LCU may be lowered when the method of using the information of the same LCU is applied.

Therefore, when the temporal level of the same position LCU and the current LCU are different, there is a need for a method of using the information of the same position LCU in consideration of the temporal level.

(1) how to consider only the same temporal level

When there is a difference in temporal level between the current picture and the target picture of the same position LCU, a flag indicating whether to use information of the same position LCU may not be transmitted. If a flag indicating whether to use information of the same location LCU is not transmitted, the value of the flag may not be inferred.

Therefore, a method of transmitting a flag indicating whether to use information of the same location LCU may be used only when the temporal level of the current picture and the target picture are the same.

(2) How to modify and use the information of the same location LCU

If there is a difference in temporal level between the current picture and the target picture of the same position LCU, the information of the same position LCU can be used without any change in the information of the same position LCU.

For example, when the current picture and the target picture of the same position LCU have different temporal levels and use the information of the same position LCU (use_coloc_LCU_info_flag = 1), the following information is modified without using the information of the same position LCU as it is. It is available.

A case of using split information (for example, split_coding_unit_flag) of a coding unit among information of the same position LCU will be described. In the above-described methods, the split information of the same location LCU is used / applied to the current LCU as it is, but in this method, the information may be modified in consideration of the temporal level of pictures.

When using the split information of the coding unit among the information of the co-located LCU, if the target picture of the co-located LCU and the picture of the current LCU have different temporal levels, the depth of the highest CU depth region of the co-located LCU ( The segmentation information can be derived so that the depth is reduced by a predetermined depth. For example, split_coding_unit_flag may be derived such that the depth of the deepest region in the same position LCU is reduced by one. For example, it may be regarded as clipping about a depth available as split information of the current LCU among split information of the same position LCU.

However, when the depth of the highest CU depth region of the same position LCU is 0, the depth of the coding unit can no longer be reduced, thereby maintaining the depth of the coding unit.

15 is a view schematically showing a divided state of the same position LCU. FIG. 16 is a diagram schematically showing a derivation state of a current LCU using information of the same location LCU.

In the examples of FIGS. 15 and 16, it is assumed that the temporal level of the target picture and the current picture of the same position LCU are different from each other. For example, it is assumed that the temporal level of the target picture is 1, and the temporal level of the current picture is 3.

In FIG. 15 and FIG. 16, the numbers indicated in the blocks represent the depths of the corresponding coding units.

15 and 16, the region having the deepest depth among the co-located LCUs 1510 is the

coding units

1520 and 1530 and has a depth of three.

As described above, the temporal level of the target picture of the same position LCU shown in FIG. 15 and the current picture shown in FIG. 16 are different from each other. Accordingly, the split information of the current LCU may be derived by reducing the depth of the deepest region in the same location LCU by a predetermined size, for example, 1, without using the split information of the same location LCU as it is.

Therefore, in the co-located LCU of FIG. 15, the split state of the coding unit is induced to use the split state of the current LCU so that the depth of the region having the depth is 2, and the split state of the remaining portion of the co-located LCU is the current LCU. Use it as it is. Referring to FIG. 16, it can be seen that the

regions

1620 and 1630 of the current LCU 1610 corresponding to the deepest regions in the same location LCU have a depth adjusted to 2 to induce a split state.

(3) How to adaptively use the information of the same location LCU

If there is a difference in temporal level between the current picture and the target picture of the same position LCU, the information of the same position LCU can be adaptively used according to the degree of the difference in temporal level without using the information of the same position LCU as it is. have.

For example, when the current picture and the target picture of the same position LCU have different temporal levels and use the information of the same position LCU (use_coloc_LCU_info_flag = 1), the information of the same position LCU is not used as it is and is adaptive as follows. Can be used as A case of using split information (for example, split_coding_unit_flag) of a coding unit among information of the same position LCU will be described.

If the depth of one region (e.g., coding unit) of each partition of the co-located LCU is greater than or equal to 'highest CU depth of the same region LCU-difference in temporal level + 1', i.e., 'current picture and target' If the temporal level difference of the picture is greater than the difference between the depth of the region and the maximum depth of the same position LCU in the same position LCU, then the depth for that region in the same position LCU is the maximum depth of the same position LCU. It is adjusted to be a 'temporal level difference between the current picture and the target picture' to derive the split state of the same position LCU. For example, when the split information of the same position LCU is used as the split information of the current LCU, it can be regarded as clipping to what extent a depth difference between the same position LCU and the current LCU is allowed.

The partitioned state of the co-located LCU derived may be used as the partitioned state of an area corresponding to the corresponding area in the current LCU. In this case, when the 'maximum depth of the same location LCU'-'the temporal level difference between the current picture and the target picture' is less than or equal to zero, the depth of the corresponding area may be zero.

In the same area LCU, the divided state of the area having a depth smaller than the 'highest CU depth of the same area LCU-difference in temporal level + 1' may be used as the divided state of the current LCU.

17 schematically illustrates an example of a split state of the same position LCU. In the example of FIG. 17, it is assumed that the target picture of the same position LCU has a temporal level of 1. FIG.

18 shows an example of a split state of a current LCU derived using information of the same location LCU. In FIG. 18, it is assumed that the temporal level of the current picture is three.

17 and 18, the number of each partition represents the depth of the corresponding area.

Referring to FIG. 17, it can be seen that the maximum coding unit depth is 3 in the same position LCU 1710. Since the temporal level of the current picture is 3 and the temporal level of the target picture is 1, as described above, the maximum coding unit depth of the same position LCU is minus the temporal level difference between the current picture and the target picture. Becomes two.

Therefore, in the partitions of the co-located LCU (for example, the coding unit) in FIG. 17, the

areas

1720 and 1730 having a depth greater than or equal to 2 have a depth of 'maximum depth of the co-located LCU'-'the current picture and the target picture. Temporal level difference '. In the case of the above-described example, an area greater than or equal to 2 (area greater than 1) is adjusted to have a depth of 1.

Referring to FIG. 18, it can be seen that

regions

1820 and 1830 corresponding to 1720 and 1730 of FIG. 17 are derived to have a depth of 1.

Referring to FIG. 19, the encoder first configures a parameter set (S1910). The encoder may configure a parameter set including a flag indicating whether to derive information of the current LCU using the information of the same position LCU. The encoder may determine in various ways whether to use information of the same position LCU. For example, it may be determined whether to use the information of the same location LCU based on the state of the current LCU, or an appropriate method may be selected by comparing the case of using the information of the same location LCU and the case of transmitting the information of the current LUC as it is. When it is determined to use the split information of the same position LCU based on the split state of the current LCU, the value of the flag may indicate 1.

In this case, the encoder may consider the temporal level between the current picture to which the current LCU belongs and the target picture to which the same position LCU belongs. For example, the encoder may transmit a flag indicating whether to derive the information of the current LCU using the information of the same location LCU only when the temporal level of the current picture and the target picture are the same.

In addition, the encoder may configure a parameter set to modify or adaptively use information of the same position LCU in consideration of the temporal level of the current picture and the target picture. How the information of the same location LCU is modified or applied is as described above.

In the case of transforming or adaptively using the information of the same position LCU, information indicating a method of transforming or adaptively using the information may be included in the parameter set or may be predetermined between the encoder and the decoder.

The encoder transmits the configured parameter set (S1920). The parameter set may be encoded along with other information and transmitted in the bitstream.

Referring to FIG. 20, the decoder receives a parameter set from an encoder (S2010). The parameter set may be encoded and received on the bitstream.

The decoder may derive information on the current LCU from the parameter set (S2020). The parameter set may directly include information on the current LCU or may include a flag indicating to derive information of the current LCU from information of the same location LCU.

In addition, when it is indicated through the flag to derive information of the current LCU from the same location LCU, the information of the same location LCU may be used as it is, or the information may be modified or adaptively used in a predetermined manner.

Information about a method of modifying or adaptively using the information of the same location LCU may be included in the received parameter set or may be predetermined between the encoder and the decoder. The decoder may derive the information of the current LCU using the information of the co-located LCU according to the method indicated by the parameter set or a predetermined method. Details of a method of modifying or adaptively using the information of the same location LCU are as described above.

The decoder may decode the coding unit based on the derived information (S2030).

<동일 위치 슬라이스의 방향 정보와 동일 위치 LCU 정보의 선택적 이용><Optional use of the same position LCU information as the direction information of the same position slice>

(1) How to tell the direction of the same position slice

Meanwhile, when using the information of the same location LCU, a flag indicating in which direction a co-located slice belongs to the same location LCU may be included in the slice header. The flag indicating the position of the same position slice may indicate whether the direction of the same position slice is L0 or L1.

Here, L0 and L1 are lists of reference pictures, L0 is a list in which the reference picture is assigned a lower index as the reference picture is closer to the current picture in the forward direction (past direction on the time axis), and L1 is the reverse direction of the current picture (the future direction on the time axis). The closer the reference picture is, the lower the index is given.

Therefore, the flag indicating the position of the same position slice may indicate which list direction picture is in the same position slice.

Table 9 shows an example of a syntax table for transmitting information according to a slice type in a conventional slice header part.

In this case, when the flag indicates that the slice type is B and the information of the same position LCU is used, the flag may indicate which list direction is the same position slice. mv_competetion_flag may indicate that a motion vector is selected by competition of the motion vectors.

Table 10 and Table 11 show examples of indicating in which direction the same position slice exists in the list direction when the flag indicates that the information of the same position LCU is used in the system to which the present invention is applied. It is shown.

In the examples of Table 10 and Table 11, collocated_from_l0_flag is a flag indicating which list direction the co-location slice is in. For example, if the value of collocated_from_l0_flag is 1, the same position slice may be indicated to be located in the L0 direction, and if the value of collocated_from_l0_flag is 0, the same position slice may be indicated to be positioned to the L1 direction.

In the examples of Tables 10 and 11, coloc_LCU_info_flag is a flag indicating whether to use information of the same location LCU, and may be transmitted in a sequence parameter set (SPS) or the like.

(2) How to selectively use the information of the same location LCU

Instead of always applying a method of deriving the information of the current LCU using the information of the same location LCU, in some cases, a method of directly transmitting the information of the current LCU may be used.

For example, a flag indicating whether to use the split information of the same location LCU as it is, may not be transmitted under the following conditions, and the information of the current LCU may be directly encoded and transmitted.

When the size of the smallest coding unit of the same position LCU, that is, the size of the smallest coding unit (SCU) is equal to the size of the same position LCU-this case corresponds to the case where the same position LCU is not divided. Therefore, since the split state (split_coding_unit_flag) transmitted for the first LCU is expressed whether or not to use the split state of the same position LCU, it is not necessary to inform whether the information of the same position LCU is used as a flag (use_coloc_LCU_info_flag). .

(3) How to use limited information on the same location LCU

Instead of using all of the same position LCU information, for example, the partition information, it is possible to limit the use of the partition information of the same position LCU when coding the current LCU. For example, rather than copying the split state of the co-located LCU as it is and using it for the current LCU, the split information (split_coding_unit_flag) for the co-located LCU is limitedly used.

Even when information of the same location LCU is used for the current LCU (use_coloc_LCU_info_flag = 1), partly, split information about the current LCU may be further transmitted. For example, the split_coding_unit_flag for the current LCU may be further transmitted so that the size of the SCU in the same location LCU and the size of the SCU in the current LCU may be different.

The conditions for further sending split_coding_unit_flag partially for the current LCU can be determined variously as follows:

1) When the depth of the same position SCU is smaller than the maximum depth

If the depth of the co-located SCU is less than the maximum depth allowed in the current LCU, then additional partitioning is allowed in that area of the current LCU.

At this time, the number of divisions may be limited to one. It may also allow to continue dividing until the maximum depth of the current LCU is reached. According to the number of splits performed further for the current LCU than the split state of the same location LCU, the number (transmission amount) of split information (split_coding_unit_flag) transmitted for the current LCU is changed.

FIG. 22 schematically illustrates an example in which the split state for the current LCU is further divided after being derived from the same position LCU. In FIG. 22, the same position LCU from which the current LCUdml split state is derived is the LCU of FIG. 21, and the maximum depth of the current LCU is 3.

Thus, comparing FIG. 21 and FIG. 22, the current LCU may be split at least once and at least twice in the split state of the same location LCU 2100. For example, for the coding units 2210 and 2220 of the current LCU 2200 corresponding to the

coding units

2110 and 2120 of the same position LCU 2100, two times in the split state of the same

position coding units

2110 and 2120. It can be seen that it can be further divided.

The portion indicated by the solid line in the LCU 2200 of FIG. 22 is a portion divided by using the information of the same location LCU without transmitting additional split information (split_coding_unit_flag) to the current LCU. In addition, the portion indicated by a dotted line in the LCU 2200 of FIG. 22 indicates that it is divided by split information (split_coding_unit_flag) additionally transmitted for the current LCU 2200.

In the example of FIG. 22, the coding unit 2210 at the upper left is further divided in the split state of the same position LCU by the additionally transmitted split information (split_coding_unit_flag), and other coding units are split in the same state as the split state of the same position LCU. It is a state.

Meanwhile, in the LCU of FIG. 22, it may be necessary to transmit a flag (split_coding_unit_flag) indicating whether to be split even for each coding unit indicated by a solid line and no longer split, to indicate that no split is performed.

2) Use the split information of the same position LCU only up to a predetermined depth

When the split information of the same position LCU is used within a range where the depth of the coding unit in the current LCU becomes a predetermined depth, and is divided deeper than the predetermined depth, the split information (split_coding_unit_flag) for the current LCU is transmitted. Can be.

For example, the split information (divided state) of the same position LCU is used until the depth of the coding unit in the current LCU becomes just before the maximum depth allowed in the current LCU (maximum depth-1), and thereafter, separate split information. (split_coding_unit_flag) may be used to determine whether to split the corresponding coding unit.

23 schematically illustrates an example of a split state for the same position LCU 2300.

FIG. 24 schematically illustrates an example in which the split state of the current LCU is finally divided by additional split information after being derived from the split state of the same position LCU up to a predetermined depth. In the example of FIG. 24, it is assumed that the same position LCU used to derive the split state of the current LCU is the LCU of FIG. 23, and the maximum depth allowed in the current LCU is 3.

Therefore, until the coding unit becomes a predetermined depth in the current LCU, it may be divided according to the split state of the same position LCU. As described above, if the depth of the co-located LCU is one to a depth smaller than the maximum depth allowed in the current LCU, the maximum depth of the current LCU is 3 in the example of FIG. Until it becomes 2, it can divide according to the division state of the same position LCU shown in FIG. Subsequently, for a coding unit of the current LCU having a depth of 2, a flag (split_coding_unit_flag) indicating whether to split is transmitted to indicate additional splitting.

In FIG. 23, the

SCUs

2310 and 2320 of the same location LCU have a depth of 3, but only a split state up to depth 2 may be used for dividing the current LCU. Thus, even in the

areas

2410 and 2420 corresponding to the

SCUs

2310 and 2320 of the same position LCU, the coding unit of the current LCU is no longer divided according to the split state of the same position LCU at depth 2.

Subsequently, according to the split flag further transmitted, the coding units in the current LCU may be further split in the split state of the same position LCU. The two

regions

2430 and 2440 of FIG. 24 are further divided in the split state of the same position LCU by the split flag split_coding_unit_flag.

The tree structure of FIG. 25 conceptually illustrates the split state of the current LCU. Referring to FIG. 25, in the case of using part of the split information of the same location LCU, for example, only using split information one step before the maximum depth of the current LCU in the split state of the same location LCU, a depth of maximum depth-1 (2530) ) May be divided according to split information of the same location LCU, and additional split may be performed by separate split information 2510.

In addition, when all the split information of the same position LCU is used, the split is performed according to the split state of the same position LCU up to the depth corresponding to the SCU of the same position LCU, and then further division is performed by the separate partition information 2520. Can be.

Referring to FIG. 26, the encoder determines whether to derive information of the current LCU using the information of the same location LCU (S2610). The encoder may determine whether to derive the information of the current LCU using the information of the co-located LCU based on the current LCU state. For example, the encoder may determine whether to derive the split information of the current LCU using the split information of the same location LCU based on the split state of the current LCU.

The encoder configures a parameter set to be transmitted to the decoder (S2620).

The parameter set may include a flag indicating whether to derive information of the current LCU using the information of the same location LCU. If the information on the current LCU is to be derived using the same location LCU in step S2610, the value of the flag may indicate 1.

Also, the parameter set may include information indicating which direction the collocated slice is in, for example, a flag indicating whether the direction is L0 or L1. The flag indicating the direction of the co-location slice may be included in a parameter set regarding the slice header.

Meanwhile, the parameter set may be configured to restrict the use of the information of the same location LCU. For example, when the same position LCU is not divided, it is possible not to transmit a flag using the split information of the same position LCU. In this case, the split information for the current LCU may be transmitted separately as in the prior art.

In addition, the parameter set may be configured to use only part of the information of the same location LCU without using it as it is, and the parameter set may be configured to transmit additional information instead of determining the current LCU state using only the information of the same location LCU. It may be. The encoder may configure a parameter set to include additionally transmitted information. The encoder may indicate whether to use all or part of the information of the same location LCU through a parameter set or may be predetermined between the encoder and the decoder.

For example, the split information of the same location LCU may be used, but the current LCU may be split along the split information of the same location LCU only until the depth of the coding unit in the current LCU becomes a predetermined depth. In this case, in the case of being further divided after being divided according to the split information of the same location LCU, separate split information (split_coding_unit_flag) may be delivered to the corresponding coding unit. Even if all of the same location LCU partitioning information is used, the partitioning information may be further transmitted after partitioning according to the partitioning information of the same location LCU.

In this case, the encoder configures a parameter set to include additional segmentation information. Whether to use only part or all of the information of the same location LCU may be transmitted by including a parameter indicating the information in the parameter set, or may be predetermined between the encoder and the decoder.

The encoder transmits the parameter set to the decoder (S2630). The parameter set may be coded along with other information and transmitted to the decoder in a bitstream.

Referring to FIG. 27, the decoder receives a parameter set from an encoder (S2710).

The parameter set includes a parameter, for example, a flag indicating whether to derive information of the current LCU using the information of the same position LCU.

When the flag indicates to use information of the same position LCU, a flag indicating which direction of the target slice to which the same position LCU belongs belongs may be further included. In addition, additional information may be included in the parameter set even when information of the same location LCU is used. In this case, the flag and the additional information indicating the direction of the target slice may be included in the same parameter set or separate parameter set, respectively. For example, a flag indicating a direction of the target slice, separately from a flag indicating whether to use information of the same location LCU, may be included in a parameter set regarding the slice header.

The decoder derives information of the current LCU from the received parameter set (S2720).

If instructed to derive information of the current LCU from the co-located LCU, the decoder derives that information from the co-located LCU.

For example, according to a flag indicating a direction of the target slice, the target slice may be found in the corresponding direction.

In addition, predetermined information specified for the current LCU can be derived using the information of the same position LCU. Alternatively, for the current LCU, necessary information may be derived using information of the same location LCU for information on which no separate parameter is transmitted.

On the other hand, even when using the information of the same position LCU, only a part of the information may be used. In this case, additionally necessary information can be obtained from separate information received from the decoder. The separate information may belong to a different parameter set from the information indicating whether to use the information of the same location LCU. For example, the partitioning up to a predetermined depth may be performed in the same manner as the partitioning information of the same location LCU with respect to the current LCU.

In addition, even when all of the information of the same position LCU, information may be further transmitted from the decoder. In this case, the decoder may apply all the information of the same position LCU to the current LCU, and additional information may also be applied to the current LCU. For example, the current LCU may be divided in the same manner as the split state of the same location LCU, and further divided according to the divided information transmitted separately.

The decoder may decode the coding unit based on the derived information (S2730).

<스캐일러블(scalable) 비디오 부호화><Scalable Video Coding>

There is scalable video coding (SVC, hereinafter referred to as SVC) in a manner that allows the decoder to selectively decode a portion of the bitstream. The encoded bitstream may be arranged in a base layer and one or more enhancement layers.

28 is a block diagram schematically illustrating an operation of an encoder in a system to which SVC is applied. Referring to FIG. 28, the encoder encodes a plurality of layers (layer 0 and layer 1) in parallel and transmits a scalable bitstream to the decoder.

The decoder that does not support SVC among the decoders may receive and decode the base layer (layer 0), and the decoder that supports SVC may receive and decode all layers.

SVC includes various scalable coding modes. For example, spatial scalability provides a layer of spatial resolution, while temporal scalability provides a layer of frame rate. In addition, quality scalability and complexity scalability provide a layer of visual quality of image and a complexity of decoding method.

The base layer of the image with spatial scalability includes coded frames of reduced resolution. When only the base layer is decoded, an output image having a low resolution can be obtained. By decoding the enhancement layer together with the base layer, a high resolution output image can be obtained.

The base layer of an image having temporal scalability is encoded at a low video frame rate. If only the base layer is decoded, the frame rate is low, but the frame rate can be increased by decoding the base layer and the enhancement layer together. The enhancement layer includes an I-VOP encoded without prediction, a P-VOP predicted from a previous layer VOP and a subsequent base layer VOP, and a B-VOP predicted from a previous and subsequent layer VOP.

As described above, the input signal used in the SVC may have a different resolution, frame rate, bit-depth, color format, aspect ratio, and the like. Accordingly, by performing prediction between layers in consideration of this point, redundancy may be reduced and encoding performance may be improved compared to simulcast.

Various inter-layer prediction methods may be used, including prediction of inter-layer coding units / prediction units / transform units in order to reduce the amount of information about a coding unit, a prediction unit, and a transform unit transmitted in an enhancement layer. Can be performed.

29 is a conceptual diagram schematically illustrating an example of inter-layer prediction. In FIG. 29, it is assumed that the resolutions of layer 0 and layer 1 are two times different.

Referring to FIG. 29, the LCU 2910 of layer 0 is divided into various CUs and PUs (or TUs). When upscaling coding unit information CU info from the split information of layer 0 in layer 1, the same information as that of block 2920 may be obtained. In this case, since the use of the coding unit / prediction unit / transform unit information of the layer 0 in layer 1 may or may not help coding efficiency, it may be adaptive to use the layer 0 information in layer 1. I can tell you.

CU unit information of the base layer may be quad tree structure or information about partitioning / partitioning, and information about prediction (eg, information about a skip / non-skip mode, a prediction direction, a motion vector, a reference index, etc.) It may be information about both.

Therefore, the methods used to derive the information of the current LCU based on the information of the co-located LCU described above can also be applied to the layers of the SVC. For example, the information on the base layer may be selectively used by using a flag indicating whether to use the information on the base layer as it is. As an example of a flag indicating whether to use the information of the base layer as it is, as a flag indicating whether to use the split information of the same position LCU as it is, indicating whether to use the tree structure of the base layer as it is or to skip it. A flag bl_tree_info_skip_flag may be defined.

In addition, when the partition information of the same location LCU is used only up to a predetermined depth and is further divided than the corresponding depth, a syntax for indicating whether there is information to be used in the base layer may be used similarly to transmitting additional partition information. At this time, it is also possible to specify what information is available. For example, if there is information to be used for partitioning in the base layer, a syntax indicating use of the partitioning information may be used.

If the corresponding information of the layer 0 (base layer) is used for tree / partitioning and / or prediction of the layer 1 (enhancement layer), a syntax structure reflecting this may be configured as shown in Table 12.

Table 12 schematically shows an example of a slice data syntax structure for a scalable extension layer in a system to which the present invention is applied.

Table 13 schematically shows an example of coding tree syntax for a scalable extension layer in a system to which the present invention is applied.

In Table 12 and Table 13, if bl_tree_info_skip_flag is 1, it means that tree information of a base layer (also referred to as a base layer, a reference layer, or layer 0) is used as it is. If bl_tree_info_skip_flag is 0, this means that tree structure information of the base layer is not used.

For example, when the value of bl_tree_info_skip_flag is 1, the tree structure information of the base layer is referred to as an enhancement layer, a current layer, or layer 1 by upscaling of coding unit information of FIG. 29. It is upscaled to match the resolution.

LCU 02930 in FIG. 29 illustrates a case in which tree structure information of a base layer is used for an enhancement layer (current layer). As shown, split_coding_unit_flag values of the current LCU are derived so that the split information of the current LCU has the same value as the split information of the upscaled base layer.

BLSplitInfo [x0] [y0] has a value of 1 when split information exists in upscaled coding unit information and 0 when split information does not exist in upscaled coding unit information. . For example, if the resolution difference between the base layer and the enhancement layer is twice and the depths of the coding units are the same, the split information of the upscaled base layer is one step lower than the split information of the enhancement layer. Becomes

If the split information does not exist in the base layer, set the value of BLSplitInfo to 0 and send additional split information (split_coding_unit_flag) only for the enhancement layer to inform more detailed split information than the split information predicted from the upscaled base layer. Can be.

A similar method can be applied to the transform unit (TU), and using the flag (bl_tu_info_skip_flag) indicating whether to use the information of the base layer, omitting transmission of additional partition information for the current layer (enhanced layer) as shown in Table 14 You may.

Table 14 schematically shows an example of coding unit syntax for a scalable extension layer in a system to which the present invention is applied.

In addition, Table 15 schematically shows an example of a transform tree syntax for a scalable extension layer in a system to which the present invention is applied.

As described above, when the value of bl_tu_info_skip_flag is 1, the upscaled transform unit split information of the base layer is used as it is for the current layer (enhanced layer). The upscaled tu split info of the base layer is used as is in the current layer. If the value of bl_tu_info_skip_flag is 0, transform unit splitting information for the current layer (enhanced layer) is separately transmitted.

Meanwhile, as described above, when the information of the current layer is derived using the information of the base layer in the SVC, some limitations / adjustments may be added to the syntax as in the case of using the same position LCU.

For example, in case of using the information of the base layer, a flag indicating which direction the base layer is to be used may be further added.

It is also possible to adaptively select the position of a flag indicating whether to use the information of the base layer. For example, similarly to the case of using the information of the same position LCU, the information about whether to use the split information of the base layer and the information about the prediction mode may be adaptively determined according to the situation. For example, criteria such as size, depth, QP, etc. may be determined for the LCU of the current layer, and at least one criterion may be applied to determine what information should be transmitted first. For example, when the block size is large or the depth is low, the information about the partition may be transmitted before the information about the prediction mode. In this case, the information about the division may be a flag indicating whether to use the division information of the base layer.

As described above, more information may be used from the base layer in addition to the split information. Herein, the split information and the information about the prediction mode have been described as an example, but the present invention is not limited thereto, and various types of information may be derived. In this case, the syntax may be configured for each information to be used and a region to be used, and if the syntax value is 1, the corresponding information of the base layer may be applied to the corresponding region.

In addition, even in the case of the partitioning information, as in the method of using the information of the same position LCU, the partitioning information of the base layer is to be used as it is, and if the partitioning state is further divided than the partitioning state of the base layer, separate information should be transmitted. Can be. The division state of the base layer may be used only to a predetermined depth in the current layer, and the division state thereafter may be divided by transmitting division information separately. For example, in consideration of the split state of the base layer and the maximum depth of the current layer, the split state of the base layer is used only to one level lower than the maximum depth of the current layer. You can also send it separately. Of the divided states of the base layer, only the divided state corresponding to the depth of the previous stage of the maximum depth of the current layer may be used for the current layer, and the division of the subsequent stages may be performed by transmitting separate segmentation information.

Referring to FIG. 30, the encoder determines whether to induce information of the current layer by using information of the base layer (S3010).

The encoder configures a parameter set to be transmitted to the decoder (S3020).

The encoder may configure the parameter set to include information indicating whether to derive information of the current layer using information of the base layer, for example, a flag.

In the case where it is determined to derive the information of the current layer by using the information of the base layer, in addition to the flag, information indicating whether the base layer is to be used in all directions, information on the base layer, whether to use part of it, or modify it. Can be included in the parameter set.

In addition, even when using the information of the base layer, additional information to be delivered may be included in the parameter set. For example, when the split information of the current layer is derived using the split information of the base layer, the current layer may be split according to the split information of the base layer only up to a certain depth, and then the split layer may be split according to the separate split information. have. In addition, after dividing the current layer as the base layer is divided, the current layer may be further divided according to separate partitioning information. In this case, information for additional division may be transmitted as separate division information.

Direction information and / or additional information of the base layer may be included in a parameter set different from a parameter set including information indicating whether to use information of the base layer.

The encoder transmits the parameter set to the decoder (S3030). The parameter set may be coded along with other information and transmitted to the decoder in a bitstream.

Referring to FIG. 31, the decoder receives a parameter set from an encoder (S3110).

The parameter set includes information indicating whether to derive information of the current layer using information of the base layer, for example, a flag.

In the case where it is instructed to derive information of the current layer using the information of the base layer, in addition to the above flag, it is indicated whether information on the direction of the base layer, information on the base layer is to be used, partially or modified. Information and the like can be received. In addition, even when using the information of the base layer, additional information to be delivered may be received.

The decoder derives information of the current layer from the received parameter set (S3120).

If instructed to derive information of the current layer from the base layer, the decoder derives the information from the base layer.

For example, the base layer may be found in the corresponding direction according to a flag indicating the direction of the base layer. In addition, only predetermined information may be derived using the information of the base layer, or necessary information may be derived using information of the base layer with respect to information to which no additional parameter is transmitted.

On the other hand, even when using the information of the base layer, only a part of the information may be used. In this case, additionally necessary information can be obtained from separate information received from the decoder. For example, the partitioning up to a predetermined depth may be performed with respect to the current layer in the same manner as the partition information of the base layer, and thereafter, the partition may be partitioned according to the partition information transmitted separately.

In addition, even when all the information of the base layer is used, information may be additionally transmitted from the decoder. In this case, the decoder may apply all the information of the base layer to the current layer, and additional information may also be applied to the current layer. For example, the current layer may be divided in the same manner as the split state of the base layer and further divided according to the separately transmitted split information.

The decoder may decode the current layer based on the derived information (S3130).

In the exemplary system described above, the methods are described based on a flowchart as a series of steps or blocks, but the invention is not limited to the order of steps, and certain steps may occur in a different order or concurrently with other steps than those described above. Can be. In addition, the above-described embodiments include examples of various aspects. Accordingly, the invention is intended to embrace all other replacements, modifications and variations that fall within the scope of the following claims.

So far in the description of the present invention, when one component is referred to as being "connected" or "connected" to another component, the other component is directly connected to or connected to the other component. It may be, but it should be understood that other components may exist between the two components. On the other hand, when one component is referred to as being "directly connected" or "directly connected" to another component, it should be understood that no other component exists between the two components.

Claims

Determining to derive information of the current block from information of the same position block;
Encoding information of the current block by using the information of the same position block; And
And transmitting indication information relating to deriving the information of the current block from the information of the same position block.
The method of claim 1, wherein the information of the current block derived from the same position block is split information of the current block.
The video information encoding method of claim 1, wherein the indication information is a flag indicating to derive information of the current block from information of the same position block.
The method of claim 1, wherein the indication information is transmitted only when the current picture to which the current block belongs is in the inter prediction mode.
The target picture to which the co-located block belongs is a picture encoded immediately before the current picture to which the current block belongs, a picture that has the same temporal level and has the closest temporal distance from the current picture among the already encoded pictures, and is already encoded. The most recently coded picture with the same temporal level as the current picture among pictures that have been coded, the picture that has a smaller temporal level than the current picture with the nearest temporal distance among the pictures that have already been encoded, or the picture that is currently coded among the pictures that are already coded. The video information encoding method, characterized in that the level is any one of the most recently encoded pictures.
The method of claim 1, wherein the indication information,
When the temporal level of the current picture to which the current block belongs and the target picture to which the same position block belongs is different, the dividing state of the same position block is adjusted to reduce the depth of the coding unit having the maximum depth by a predetermined size. And instructing to use the segmented state of the current block.
The method according to claim 1, wherein the instruction information instructs to use the division state of the same position block as the division information of the current block to a predetermined depth,
And dividing information for the current block is transmitted for the division exceeding the predetermined depth.
The method of claim 1, wherein the indication information indicates to use the division state of the same block as the partition information of the current block.
And additional partition information is transmitted for the additional partitioning performed on the basis of the partition state of the same location block.
Constructing information for predicting pixel values of the current block; And
Transmitting the configured information;
And the configured information is configured such that information about a prediction mode of the current block is transmitted before segmentation information of the current block.
Deriving information of the current block according to the indication information about deriving information of the current block using information of the same position block; And
And predicting a pixel value of the current block based on the derived information of the current block.
The method of claim 10, wherein the indication information is a flag indicating to induce information of the current block from information of the same position block.
The method of claim 10, wherein the information of the current block derived from the same position block is split information of the current block.
The target picture to which the co-located block belongs is a picture encoded immediately before the current picture to which the current block belongs, a picture that has the same temporal level and has the closest temporal distance from the current picture among the already encoded pictures, and is already encoded. Of the pictures, the temporal level is the same as the current picture, the most recently coded picture, the previously-encoded picture, the temporal level is smaller than the current picture, the closest temporal distance, or the temporal level of the currently coded picture from among the already-encoded pictures. And the smallest and most recently coded picture.
12. The method of claim 10, wherein when the temporal level of the current picture to which the current block belongs and the target picture to which the same position block belongs is different, the depth of the coding unit having the maximum depth is reduced by a predetermined size. And adjusting the division state to use the division state of the current block.
The method of claim 10, wherein the divided state of the same position block is used as the split information of the current block up to a predetermined depth.
And dividing the data beyond the predetermined depth based on separately received split information of the current block.
The method of claim 15, wherein the predetermined depth is one step lower than the maximum depth of the current block.
The method of claim 10, wherein the split state of the same location block is used as split information of the current block.
The method of claim 1, wherein the additional division performed from the division state of the same location block is performed based on the separately received partition information.
The image of claim 10, wherein the information of the current block is derived from the information of the same location block only when the temporal level of the current picture to which the current block belongs and the target picture to which the same location block belongs is the same. Information decryption method.
Receiving information for predicting pixel values of a current block; And
Predicting pixel values of the current block according to the received information;
In the step of predicting the pixel value of the current block,
And when the received information indicates that the current block is predicted to be a skip mode, split information of the current block is not applied.
Determining to derive information of the current layer from information of the base co-located block;
Encoding information of the current layer using information of the base layer; And
Transmitting a parameter set for deriving information of the current layer from information of the base layer,
The parameter set is
A flag indicating to induce information of the current layer by using information of the base layer; And
Includes information about the direction of the base layer and / or how to use the information of the base layer,
The information on the method of using the information of the base layer indicates any one of full use, partial use, and modified use of the base layer information.
And when the derived information and the information of the current layer are different according to the information on the method of using the information of the base layer, additional information for deriving the information of the current layer is transmitted.
Receiving first information for deriving information of a current layer using information of a base layer;
Deriving information of the current layer from information of the base layer based on the first information; And
Predicting pixel values of the current layer based on the derived information of the current layer,
The first information may include a flag indicating to induce information of the current layer by using information of the base layer; And
Includes information about the direction of the base layer and / or how to use the information of the base layer,
The information on the method of using the information of the base layer indicates any one of full use, partial use, and modified use of the base layer information.
If the information derived from the information on the method of using the information of the base layer is different from the information of the current layer, the current layer based on the additional information for deriving the information of the current layer and the information derived from the base layer. Image information decoding method characterized in that to derive the information.