RU2794224C1 - Encoding and decoding method and device for it - Google Patents

Abstract

FIELD: image encoding and decoding.

SUBSTANCE: determining the predicted quantization parameter of the current quantization group determined according to at least one of the block split information and the block size information, determining the difference quantization parameter of the current quantization group, determining the quantization parameter of the current quantization group based on the predicted quantization parameter and a difference quantization parameter of the current quantization group, and inverse quantization of the current block included in the current quantization group according to the quantization parameter of the current quantization group.

EFFECT: creating methods for encoding and decoding image data with a minimum deterioration in image quality and an increased compression ratio.

3 cl, 32 dwg

Description

FIELD OF TECHNOLOGY TO WHICH THE INVENTION RELATES

[0001] The disclosure relates to a method for encoding and a method for decoding images, and in particular, to a method for efficiently encoding and decoding motion vector information.

BACKGROUND OF THE INVENTION

[0002] When a high quality image is encoded, a large amount of data is required. However, since the bandwidth available for transmitting image data is limited, the data rate applied to transmitting image data may be limited. Thus, in order to efficiently transmit image data, there is a need for methods for encoding and decoding image data with minimal deterioration in image quality and an increased compression ratio.

[0003] Image data can be compressed by removing spatial redundancies and temporal redundancies between pixels. Since neighboring pixels generally have common characteristics, encoding information of a data element composed of pixels is transmitted to remove redundancies between neighboring pixels.

[0004] The pixel values of the pixels included in the data item are not directly transmitted, but information on how the pixel values are obtained is transmitted. A prediction method in which a pixel value that is similar to the original value is predicted is determined for each data element, and encoding information about the prediction method is transmitted from an encoder to a decoder. Since the prediction value is not completely equal to the original value, residual difference data between the original value and the prediction value is transmitted from the encoder to the decoder.

[0005] When the prediction is accurate, the size of the encoding information for determining the prediction method is increased, but the size of the residual data is reduced. Thus, the prediction method is determined in view of the sizes of the encoding information and the residual data. In particular, the data elements into which a frame is divided have different sizes, and in this regard, as the data element size increases, there is an increased probability that the prediction accuracy decreases while the size of the coding information decreases. Thus, the block size is determined according to the characteristics of the frame.

[0006] The prediction method includes intra-frame (intra) prediction and inter-frame (inter) prediction. Intra-frame prediction is a method for predicting the pixels of a block from pixels adjacent to the block. Inter-frame prediction is a method for predicting pixels by referring to pixels of another referenced frame for a frame including a block. Thus, spatial redundancy is removed by intra prediction, and temporal redundancy is removed by inter prediction.

[0007] When the number of prediction methods is increased, the amount of encoding information for indicating the prediction method is increased. Thus, the amount of coding information can be reduced by predicting, from another block, coding information to apply to the block.

[0008] Since the loss of image data is allowed to the extent that the human eye cannot recognize the loss, residual data can be lossy compressed according to transformation and quantization processes, and thus the amount of residual data can be reduced.

DESCRIPTION OF EMBODIMENTS

TECHNICAL PROBLEM

[0009] An image encoding method and an image encoding apparatus for determining a quantization parameter of a quantization group based on block split information and block size information are provided. An image decoding method and an image decoding apparatus for determining a quantization parameter of a quantization group based on block split information and block size information are provided.

[0010] An image encoding method and an image encoding apparatus for mapping a current block to a current quantization parameter element based on at least one of the position and size of the current block are provided. An image decoding method and an image decoding device are proposed to match the current block with the current element of the quantization parameter based on at least one of the position and size of the current block.

[0011] Further provided is a computer-readable storage medium having a program recorded thereon for executing, on a computer, an image encoding method and an image decoding method according to one embodiment of the disclosure.

SOLUTION

[0012] A method for decoding pictures is provided, comprising: determining a predicted quantization parameter of a current quantization group determined according to at least one of block split information and block size information; determining a difference quantization parameter of the current quantization group; determining a quantization parameter of the current quantization group based on the predicted quantization parameter and the difference quantization parameter of the current quantization group; and performing inverse quantization of the current block included in the current quantization group according to the quantization parameter of the current quantization group.

[0013] An image decoding apparatus is provided, including a processor configured to determine a predicted quantization parameter of a current quantization group determined according to at least one of block split information and block size information, determine a difference quantization parameter of a current quantization group, determine a quantization parameter of the current quantization group based on the predicted quantization parameter and the difference quantization parameter of the current quantization group, and inverse quantize the current block included in the current quantization group according to the quantization parameter of the current quantization group.

[0014] A method for decoding images is provided, including the steps of: mapping a current block to a current quantization parameter element based on at least one of the position and size of the current block; obtaining a predicted quantization parameter with respect to the current element of the quantization parameter; obtaining a difference quantization parameter with respect to the current quantization parameter element; determining a quantization parameter of the current quantization parameter element based on the predicted quantization parameter and the difference quantization parameter; and performing inverse quantization of the current block according to the quantization parameter of the current quantization parameter element.

[0015] An image decoding apparatus is provided, including a processor configured to map a current block to a current quantization parameter element based on at least one of the position and size of the current block, obtain a predicted quantization parameter with respect to the current quantization parameter element, obtain a difference quantization parameter with respect to the current quantization parameter element, determine the quantization parameter of the current quantization parameter element based on the predicted quantization parameter and the difference quantization parameter, and inverse quantize the current block according to the quantization parameter of the current quantization parameter element.

[0016] A computer-readable storage medium is provided, having a program recorded thereon for executing an image encoding method and an image decoding method.

[0017] The technical objectives to be achieved by the present disclosure are not limited to the technical features described above, and other technical objectives can be inferred from the embodiments below.

BENEFITS OF DISCLOSURE

[0018] The block quantization parameter is determined according to the quantization group or the quantization parameter element, so that the information necessary to determine the quantization parameter can be compressed efficiently.

BRIEF DESCRIPTION OF THE DRAWINGS

[0019] FIG. 1A shows a block diagram of an image encoding device based on coding elements according to a tree structure according to one embodiment of the disclosure.

[0020] FIG. 1B shows a block diagram of an image decoding apparatus based on tree structure coding elements according to one embodiment.

[0021] FIG. 2 depicts a process by which an image decoding apparatus determines at least one coding unit by splitting the current coding unit, according to one embodiment.

[0022] FIG. 3 depicts a process for determining at least one coding unit by splitting a non-square coding unit, according to one embodiment.

[0023] FIG. 4 depicts a process of splitting a coding unit based on at least one of block shape information and split shape information according to one embodiment.

[0024] FIG. 5 shows a method for determining a preset coding unit from an odd number of coding units, according to one embodiment.

[0025] FIG. 6 depicts a processing order of a plurality of coding units when a plurality of coding units is determined by splitting the current coding unit, according to one embodiment.

[0026] FIG. 7 depicts a process for determining that the current coding unit should be split into an odd number of coding units when coding blocks are not pre-processed, according to one embodiment.

[0027] FIG. 8 depicts a process for determining at least one coding element by splitting the first coding element, according to one embodiment.

[0028] FIG. 9 shows that the shape into which the second coding unit is splittable is limited when the second coding unit having a non-square shape, which is determined by splitting the first coding unit, satisfies a predetermined condition, according to one embodiment.

[0029] FIG. 10 shows a square coding unit splitting process when splitting shape information indicates that the square coding unit should not be split into four square coding units, according to one embodiment.

[0030] FIG. 11 shows that the processing order between multiple coding units may be changed depending on the coding unit splitting process, according to one embodiment.

[0031] FIG. 12 depicts a process for determining the depth of a coding unit as the shape and size of the coding unit change when the coding unit is recursively split such that a plurality of coding units is determined, according to one embodiment.

[0032] FIG. 13 depicts depths that can be determined based on coding unit shapes and sizes and part indices (PIDs) that are intended to distinguish coding units, according to one embodiment.

[0033] FIG. 14 shows that a plurality of coding elements is determined based on a plurality of preset data elements included in a frame, according to one embodiment.

[0034] FIG. 15 shows a processing unit serving as an element for determining an order of determination of reference coding units included in a frame, according to one embodiment.

[0035] FIG. 16 shows an image decoding apparatus for determining a block quantization parameter and decoding block residual data according to the determined quantization parameter.

[0036] FIGS. 17A-17D are diagrams of embodiments in which a quantization group is determined according to the number of times of a quadtree split.

[0037] FIGS. 18A-18C depict an embodiment of a method for determining a quantization group in a largest coding block to which non-quadtree partitioning is applied.

[0038] FIG. 19 shows a syntax structure with respect to a method for decoding a difference quantization parameter included in a bitstream when quadtree splitting and non-quadtree splitting are enabled.

[0039] FIG. 20 shows a method for decoding pictures by determining a block quantization parameter according to a quantization group, and decoding block residual data according to the determined quantization parameter.

[0040] FIG. 21 shows an embodiment of a quantization parameter element structure and a coding block tree structure.

[0041] FIGS. 22A and 22B show a method for determining a quantization parameter element corresponding to the current block.

[0042] FIGS. 23A and 23B depict the correlation between a block and a bin of a quantization parameter.

[0043] FIG. 24 shows a method for decoding pictures by determining a block quantization parameter according to a quantization parameter element and decoding block residual data according to the determined quantization parameter.

BEST IMPLEMENTATION OPTION

[0044] A method for decoding images is provided, including the steps of: determining a predicted quantization parameter of a current quantization group determined according to at least one of block split information and block size information; determining a difference quantization parameter of the current quantization group; determining a quantization parameter of the current quantization group based on the predicted quantization parameter and the difference quantization parameter of the current quantization group; and performing inverse quantization of the current block included in the current quantization group according to the quantization parameter of the current quantization group. An image decoding apparatus is also provided, including a process for performing an image decoding method.

[0045] A method for decoding images is provided, including the steps of: mapping a current block to a current quantization parameter element based on at least one of the position and size of the current block; obtaining a predicted quantization parameter with respect to the current element of the quantization parameter; getting a difference quantization parameter with respect to the current quantization parameter element; determining a quantization parameter of the current quantization parameter element based on the predicted quantization parameter and the difference quantization parameter; and performing inverse quantization of the current block according to the quantization parameter of the current quantization parameter element. An image decoding apparatus is also provided, including a process for performing an image decoding method.

DISCLOSURE OPTION

[0046] The advantages and features of the embodiments and methods for achieving them can be more easily understood by reference to the embodiments and the accompanying drawings. In this regard, the disclosure may take various forms and should not be construed as being limited to the embodiments set forth herein. Indeed, these embodiments are provided so that this disclosure is comprehensive and complete, and will fully convey the concept of the disclosure to those of ordinary skill in the art.

[0047] Terms used in the technical description will be briefly defined, and embodiments will be described in detail.

[0048] All terms, including descriptive or technical terms that are used in the technical description, should be construed as having meanings that are obvious to one of ordinary skill in the art. However, the terms may have other meanings according to the intention of one of ordinary skill in the art, precedent cases, or the advent of new technologies. Also, some terms may be arbitrarily chosen by the applicant, in which case the meaning of the chosen terms will be described in detail in the detailed description of the disclosure. Thus, the terms used in the disclosure should not be interpreted on the basis of their names alone, but should be defined based on the meanings of the terms along with the descriptions throughout the technical specification.

[0049] In the following technical description, singular forms include plural forms unless the context clearly indicates otherwise.

[0050] When an item "includes" or "comprises" an element, unless specifically described to the contrary, the item may additionally include other elements without excluding other elements. In the following descriptions, terms such as "element" indicate a software or hardware component, such as a field programmable gate array (FPGA) or application-specific semiconductor (ASIC), and "element" performs specific functions. However, "element" is not limited to software or hardware. An "element" may be configured to reside in an addressable storage medium, or may be configured to operate on one or more processors. Thus, for example, the term "element" may refer to components such as software components, object-oriented software components, class components, and task components, and may include processes, functions, attributes, procedures, routines, program code segments, drivers, firmware, microcode, circuits, data, databases, data structures, tables, arrays or variables. A function provided by components and "elements" may be associated with fewer components and "elements", or may be split into additional components and "elements".

[0051] The term "current block" indicates one of the current coding element to be encoded or decoded, a prediction element, and a transform element. For convenience of description, when it is desired to distinguish between other types of blocks such as a prediction element, a transformation element or the like, the terms "current coding block", "current prediction block", "current transformation block" may be used. In addition, "lower block" denotes a data element obtained by splitting the "current block". "Top-level block" means a data element including the "current block".

[0052] Hereinafter, "sample (sample)" refers to data assigned to a sampling position of an image, i. e. data to be processed. For example, the image pixel values in the spatial domain and the transform coefficients in the transform domain may be discrete samples. An element including at least one such discrete sample may be defined as a block.

[0053] Hereinafter, the disclosure will be described more fully with reference to the accompanying drawings, so that one of ordinary skill in the art will be able to carry out the embodiments without any difficulty. Additionally, parts irrelevant to the description will be omitted from the drawings for a clear description of the disclosure.

[0054] FIG. 1A shows a block diagram of an image encoding device 100 based on coding elements according to a tree structure according to one embodiment of the disclosure.

[0055] The image encoding device 100 may include an encoder 110 and a bitstream generator 120.

[0056] Encoder 110 partitions a frame or slice (slice) included in a frame into a plurality of largest coding units according to the sizes of the largest coding units. The largest coding units may be data units having a size of 32×32, 64×64, 128×128, 256×256, or the like, and each may be a square data element having a width and length equal to powers of 2. Encoder 110 may provide the bitstream generator 120 with largest bin size information indicating the size of the largest coding unit. The bitstream generator 120 may add size information of the largest coding unit to the bitstream.

[0057] Encoder 110 determines coding units by splitting the largest coding unit. Whether to split a coding unit is determined according to whether it is efficient to split a coding unit by optimizing the ratio of rate to distortion. Additionally, split information indicating whether the coding unit is split may be generated. The split information may be in the form of a flag.

[0058] The coding element can be split in various ways. For example, a square coding unit can be broken into four square coding units that are half the width and height of the square coding unit. The square coding unit may be split into two rectangular coding units having a width half that of the square coding unit. The square coding element may be split into two rectangular coding elements having a height less than half that of the square coding element. A square coding element can be split into three coding elements by splitting its width or height in a ratio of 1:2:1.

[0059] A rectangular coding unit having a width twice its height can be split into two square coding units. A rectangular coding unit having a width twice its height may be split into two square coding units having a width four times its height. A rectangular coding unit having a width twice its height can be split into two rectangular coding units and one square coding unit by splitting the width of the rectangular coding unit in a ratio of 1:2:1.

[0060] Likewise, a rectangular coding unit having a height twice its width can be split into two square coding units. A rectangular coding unit having a height twice its width may be split into two rectangular coding units having a height four times their width. Similarly, a rectangular coding unit having a height twice its width can be split into two rectangular coding units and one square coding unit by splitting the height of the rectangular coding unit in a ratio of 1:2:1.

[0061] When two or more partitioning methods are applicable to the image encoding apparatus 100, information about the partitioning method applicable to the coding element from among the partitioning methods applicable to the image encoding apparatus 100 can be determined for each frame. Thus, only specific partitioning methods can be defined to be used for each frame. When the image encoding apparatus 100 employs only one partitioning method, information about the partitioning method applicable to the coding element is not further determined.

[0062] An encoding element of a particular size may be split by a particular splitting method. For example, when the coding unit size is 256×265, the coding unit can be set to split into only four square coding units whose width and height are half that of the coding unit.

[0063] When the coding element split information indicates that the coding element is to be split, split shape information indicating how the coding element is split may be generated. When there is only one split method applicable to the frame to which the coding element belongs, split shape information may not be generated. When the partitioning method is adaptively determined based on the coding information about the proximity of the coding element, the partitioning shape information may not be generated.

[0064] As described above, the image data of the current frame is divided into the largest bins according to the maximum bin size. The largest coding unit may include coding units that are hierarchically broken down from the largest coding unit. The shape and position of the lower level encoding element may be determined based on the split shape of the upper level encoding element. A minimum coding unit size that limits splitting of the coding unit may be preset.

[0065] The encoder 110 compares the coding gain when the coding unit is hierarchically split with the coding gain when the coding unit is not split. Then, the encoder 110 determines whether to split the coding unit according to the result of the comparison. When it is determined that it is more efficient to split the coding unit, the encoder 110 splits the coding unit hierarchically. The encoding element is not split when the result of the comparison demonstrates that it is more efficient not to split the encoding element. Whether to split an encoding element can be determined independently of whether to split other encoding elements adjacent to that encoding element.

[0066] The coding unit that is split last may be predicted due to intra-frame prediction or inter-frame prediction. Intra-picture prediction is a method for predicting prediction element samples by reference samples around the prediction element. Inter-picture prediction is a method for predicting prediction element samples by obtaining a reference sample from a reference frame referenced for the current frame.

[0067] For intra prediction, encoder 110 can select the most efficient intra prediction method by applying a plurality of intra prediction methods to a prediction element. The intra prediction methods include DC mode, planar mode, directional mode such as vertical mode and horizontal mode, or the like.

[0068] Intra-frame prediction may be performed for each prediction element when the reconstructed sample around the coding element is used as a reference sample. However, when the reconstructed sample in the coding element is used as the reference sample, the reference sample in the coding element must be reconstructed first, and thus the prediction order of the prediction element can obey the transformation order of the transformation element. Thus, when a reconstructed sample in a bin is used as a reference bin, only the intra prediction method for transform elements corresponding to the prediction element is determined for the prediction element, and intra prediction can be performed on essentially each transform element.

[0069] Encoder 110 may select the most efficient inter-frame prediction method by determining the optimal motion vector and reference frame. For inter-frame prediction, encoder 110 may determine a set of potentially suitable motion vectors from coding units spatially and temporally adjacent to the current coding unit and determine the most efficient motion vector as a motion vector among the set of potentially suitable motion vectors. Likewise, a set of potentially suitable reference frames may be determined from among the number of coding units spatially and temporally adjacent to the current coding unit, and the most efficient reference frame may be determined among the set of potentially suitable reference frames. In one embodiment, the reference frame may be determined from a list of reference frames defined in advance for the current frame. In one embodiment, for accurate prediction, the most efficient motion vector from a set of potentially suitable motion vectors may be determined as the predicted motion vector, and the motion vector may be determined by correcting the predicted motion vector. Inter prediction may be performed in parallel for each prediction element included in a coding element.

[0070] The encoder 110 may recover the bin by obtaining only the information representing the motion vector and the reference frame according to the skip mode. According to the skip mode, all encoding information including the residual signal is omitted except for information representing a motion vector and a reference frame. Since the residual signal is omitted, the skip mode is applicable when the prediction accuracy is very high.

[0071] The division mode to be used may be limited according to the prediction method for the prediction element. For example, only the split mode for prediction elements having sizes of 2Nx2N and NxN can be applied to intra prediction, while the split mode for prediction tiles having sizes of 2Nx2N, 2NxN, Nx2N, and NxN can be applied to inter prediction. Further, only a division mode for a prediction element having a size of 2Nx2N can be applied to the skip mode from inter prediction. The division mode allowed for each prediction method in the image encoding apparatus 100 may be changed according to the encoding efficiency.

[0072] The image encoding device 100 may perform the transformation based on the coding element. The image encoding apparatus 100 can convert the residual data, which is the difference between the original pixel values included in the coding unit and the prediction values thereof, by a predetermined process. For example, the image encoding apparatus 100 may perform lossy compression on the residual data by quantization and discrete cosine transform (DCT)/discrete sine transform (DST). Alternatively, the image encoding apparatus 100 may perform lossless compression on the residual data without quantization.

[0073] Finally, encoder 110 determines the most efficient prediction method for the current coding element from a plurality of intra-prediction methods and inter-prediction methods. Next, the encoder 110 determines a method for predicting the current bin based on the coding gain according to the prediction result. Likewise, the encoder 110 may determine a transform method based on the coding gain according to the prediction result. Based on the most efficient prediction method and transform method determining the scheme with respect to the coding element, the coding efficiency of the coding element is finally determined. The encoder 110 determines the hierarchical structure of the largest coding unit according to the coding efficiency of the finally broken coding unit.

[0074] The encoder 110 can measure the coding efficiency of the bins, the prediction efficiency of the prediction methods, etc. by a rate-to-distortion optimization technique based on Lagrange multipliers.

[0075] The encoder 110 may generate split information indicating whether the coding unit is split based on the determined hierarchical structure of the largest coding unit. The encoder 110 may generate split mode information to determine the prediction bin and transform the split bin information to determine the transform bin with respect to the split coding bin. When there are two or more ways to split a coding unit, the encoder 110 may generate split shape information indicating the split method together with the split information. Then, the encoder 110 may generate information about the prediction method and the transform method used for the prediction element and the transform element.

[0076] The bitstream generator 120 may output the information generated by the encoder 110 in the form of a bitstream based on the hierarchical structure of the largest coding unit.

[0077] A method for determining a coding element, a prediction element, and a transformation element according to the tree structure of the largest coding element according to one embodiment will be described in detail with reference to FIGS. 3-12 below.

[0078] FIG. 1B shows a block diagram of an image decoding apparatus 150 based on tree structure coding elements, according to one embodiment.

[0079] The image decoding device 150 includes a receiver 160 and a decoder 170.

[0080] Various terms such as "coding element", "prediction element", "transformation element" and various "split information" related to the decoding operation of the image decoding apparatus 150 according to one embodiment are as described above with references in Fig.1 and the device 100 encoding images. Further, the image decoding apparatus 150 is configured to recover image data, and thus various encoding methods used on the image encoding apparatus 100 are applicable to the image decoding apparatus 150.

[0081] The receiver 160 receives and parses the encoded image bitstream. Receiver 160 extracts the information necessary to decode each largest coding unit from the parsed bitstream and provides the information to decoder 170. Receiver 160 may extract information about the current frame's maximum coding unit size from a header, sequence parameter set, or frame parameter set for the current frame. .

[0082] The receiver 160 extracts, from the parsed bitstream, splitting information of tree structure coding units for each largest coding unit. The extracted split information is output to the decoder 170. The decoder 170 may determine the tree structure of the largest coding unit by splitting the largest coding unit according to the extracted split information.

[0083] The split information extracted by the decoder 170 is split information of a tree structure determined by the image encoding apparatus 100 to generate a minimum encoding error. Thus, the image decoding apparatus 150 can recover an image by decoding data according to an encoding method that generates a minimum encoding error.

[0084] Decoder 170 may extract split information about a data element, such as a prediction element and a transform element, included in the coding element. For example, decoder 170 may extract information about the most efficient partitioning mode for a prediction element. The decoder 170 may extract the transform split information of the tree structure that is most efficient in the transform element.

[0085] The decoder 170 may obtain prediction method information that is most efficient in the split prediction bins from the coding unit. The decoder 170 may obtain information about the transformation method that is most efficient in the transformation elements obtained by splitting from the coding element.

[0086] The decoder 170 extracts information from the bitstream according to the method of configuring the bitstream by the bitstream generator 120 of the image encoding device 100 .

[0087] The decoder 170 may split the largest coding unit into the coding units of the most efficient tree structure based on the split information. The decoder 170 may split the bin into prediction bins according to the split mode information. The decoder 170 may split the coding unit into transform units according to the transform split information.

[0088] The decoder 170 may predict the prediction element according to the prediction method information. The decoder 170 may perform inverse quantization and inverse transform on the residual data corresponding to the difference between the original value and the pixel prediction value, based on the transform element transform method information. Further, the decoder 170 may reconstruct the pixels of the coding element according to the prediction result of the prediction element and the transformation result of the transformation element.

[0089] FIG. 2 depicts a process by which the image decoding apparatus 150 determines at least one coding unit by splitting the current coding unit, according to one embodiment.

[0090] According to one embodiment, the image decoding apparatus 150 can determine the shape of the coding unit by the block shape information, and can determine the shape into which the coding unit is to be split by the split shape information. That is, the splitting method of the coding element, where the method is indicated by splitting type information, can be determined based on the block shape indicated by the block shape information used by the image decoding apparatus 150 .

[0091] According to one embodiment, the image decoding apparatus 150 may use block shape information indicating that the current coding unit is square. For example, the image decoding apparatus 150 can determine whether to split a square coding unit, whether to vertically split a square coding unit, whether to split a square coding unit horizontally, or whether to split a square coding unit into four coding units, based on split shape information. Referring to FIG. 2, when the block shape information of the current coding unit 200 indicates a square shape, the decoder 180 may determine that the coding unit 210a having the same size as the current coding unit 200 is not split based on the split shape information , indicating not to perform splitting, or may determine the coding units 210b, 210c, or 210d to be split based on splitting shape information indicating a preset splitting method.

[0092] Referring to FIG. 2, according to one embodiment, the image decoding apparatus 150 can determine two coding units 210b obtained by splitting the current coding unit 200 in the vertical direction based on split shape information indicating to split in the vertical direction. The image decoding apparatus 150 may determine two coding units 210c obtained by splitting the current coding unit 200 in the horizontal direction based on the split shape information indicating splitting in the horizontal direction. The image decoding apparatus 150 may determine the four coding units 210d obtained by splitting the current coding unit 200 in the vertical and horizontal direction based on the split shape information indicating splitting in the vertical and horizontal directions. However, methods for splitting a square coding unit are not limited to the methods described above, and the split shape information may indicate various methods. The preset split shapes into which the square coding unit is to be split will be described in detail below with respect to various embodiments.

[0093] FIG. 3 depicts a process performed by the image decoding apparatus 150 for determining at least one coding unit by splitting a non-square coding unit, according to one embodiment.

[0094] According to one embodiment, the image decoding apparatus 150 may use block shape information indicating that the current coding unit has a non-square shape. The image decoding apparatus 150 may determine whether to split the non-square current bin or split the non-square current coding unit by a preset splitting method based on the split shape information. Referring to FIG. 3, when the block shape information of the current coding unit 300 or 350 indicates a non-square shape, the image decoding apparatus 150 may determine that the coding unit 310 or 360 having the same size as the current coding unit 300 or 350, is not split based on the split shape information indicating no split, or to determine the encoding elements 320a and 320b, 330a-330c, 370a and 370b, or 380a-380c to be split based on the split shape information indicating a preset splitting method. The preset splitting methods for splitting a non-square coding unit will be described in detail below with respect to various embodiments.

[0095] According to one embodiment, the image decoding apparatus 150 may determine the splitting method of the coding unit by splitting shape information, in which case the splitting shape information may indicate the number of one or more coding units generated by splitting the coding unit. Referring to FIG. 3, when the split shape information indicates to split the current coding unit 300 or 350 into two coding units, the image decoding apparatus 150 may determine two coding units 320a and 320b or 370a and 370b included in the current coding unit 300 or 350, by splitting the current coding element 300 or 350 based on the split shape information.

[0096] According to one embodiment, when the image decoding apparatus 150 splits the non-square current coding unit 300 or 350 based on split shape information, the current coding unit can be split considering the long side position of the non-square current coding unit 300 or 350. For example, the image decoding apparatus 150 may determine a plurality of coding units by dividing the long side of the current coding unit 300 or 350 based on the shape of the current coding unit 300 or 350.

[0097] According to one embodiment, when the split shape information indicates to split the coding unit into an odd number of blocks, the image decoding apparatus 150 may determine an odd number of coding units included in the current coding unit 300 or 350. For example, when the split shape information indicates to split the current coding unit 300 or 350 into three coding units, the image decoding apparatus 150 may split the current coding unit 300 or 350 into three coding units 330a, 330b, and 330c or 380a, 380b, and 380c. According to one embodiment, the image decoding apparatus 150 may determine an odd number of coding units included in the current coding unit 300 or 350, and not all of the determined coding units may be of the same size. For example, a preset coding element 330b or 380b of a certain odd number of coding elements 330a, 330b and 330c or 380a, 380b and 380c may have a different size than other coding elements 330a and 330c or 380a and 380c. That is, coding units that can be determined by splitting the current coding unit 300 or 350 may have multiple sizes, and, in some cases, all of an odd number of coding units 330a, 330b and 330c or 380a, 380b and 380c may have different sizes. .

[0098] According to one embodiment, when the split shape information indicates to split the coding unit into an odd number of blocks, the image decoding apparatus 150 may determine an odd number of coding units included in the current coding unit 300 or 350, and may impose a predetermined limit on at least at least one coding unit from an odd number of coding units generated by splitting the current coding unit 300 or 350. Referring to FIG. 3, the image decoding apparatus 150 may allow the decoding method of the coding element 330b or 380b to be different from the decoding method of other coding elements 330a and 330c or 380a and 380c where the coding element 330b or 380b is at the center position of the three elements 330a , 330b and 330c or 380a, 380b and 380c coding generated by splitting the current coding element 300 or 350. For example, the image decoding apparatus 150 may limit the encoding element 330b or 380b at the center position so that it no longer splits or only splits a predetermined number of times, unlike other encoding elements 330a and 330c or 380a and 380c.

[0099] FIG. 4 shows a process performed by the image decoding apparatus 150 for splitting a coding unit based on at least one of block shape information and split shape information according to one embodiment.

[00100] According to one embodiment, the image decoding apparatus 150 may determine whether or not to split the square first coding unit 400 into coding units based on at least one of the block shape information and the split shape information. According to one embodiment, when the split shape information indicates to split the first coding unit 400 in the horizontal direction, the image decoding apparatus 150 may determine the second coding unit 410 by splitting the first coding unit 400 in the horizontal direction. The first coding element, the second coding element, and the third coding element used according to one embodiment are terms used to understand the relationship before and after splitting the coding element. For example, the second coding element may be determined by splitting the first coding element, and the third coding element may be determined by splitting the second coding element. It will be appreciated that the relationship between the first coding element, the second coding element, and the third coding element follows the descriptions above.

[00101] According to one embodiment, the image decoding apparatus 150 may determine whether or not to split the determined second coding unit 410 into coding units based on at least one of the block shape information and the split shape information. Referring to FIG. 4, the image decoding apparatus 150 may or may not split the non-square second coding unit 410, which is determined by splitting the first coding unit 400, into one or more third coding units 420a or 420b, 420c, and 420d based on at least one of the block shape information and the split shape information. The image decoding apparatus 150 can obtain at least one of the block shape information and the split shape information, and determine a plurality of second bins of different shapes (for example, 410) by splitting the first coding bin 400 based on the obtained at least one of the block shape information and split shape information, and the second coding unit 410 may be split by the split method of the first encoding unit 400 based on at least one of the block shape information and the split shape information. According to one embodiment, when the first coding unit 400 is split into second coding units 410 based on at least one of the block shape information and split shape information of the first coding unit 400, the second coding unit 410 may also be split into third units 420a or 420b, 420c and 420d encoding based on at least one of the block shape information and split shape information of the second encoding element 410 . That is, the coding unit can be recursively split based on at least one of the block shape information and split shape information of each coding unit. A method that can be used to recursively split an coding unit will be described below with respect to various embodiments.

[00102] According to one embodiment, the image decoding apparatus 150 may decide to split each of the third coding units 420a or 420b, 420c, and 420d into coding units or not to split the second coding unit 410 based on at least one of the block shape information and information splitting forms. According to one embodiment, the image decoding apparatus 150 may partition the non-square second coding unit 410 into an odd number of third coding units 420b, 420c, and 420d. The image decoding apparatus 150 may impose a preset restriction on a preset third coding unit from an odd number of third coding units 420b, 420c, and 420d. For example, the image decoding apparatus 150 may limit the third coding unit 420c at the center position of an odd number of third coding units 420b, 420c, and 420d so that it no longer splits or splits a settable number of times. Referring to FIG. 4, the image decoding apparatus 150 may restrict the third coding unit 420c that is at the center position of an odd number of third coding units 420b, 420c, and 420d included in the non-square second coding unit 410 so that it no longer breaks, breaks by a predetermined splitting method (eg, only split into four coding units, or split by the splitting method of the second coding unit 410) or split only a predetermined number of times (eg, split only n times (where n>0)). However, restrictions on the third coding element 420c at the center position are not limited to the examples described above and may include various restrictions for decoding the third coding element 420c at the center position differently from other third coding elements 420b and 420d.

[00103] According to one embodiment, the image decoding apparatus 150 can obtain at least one of block shape information and split shape information that is used to split the current coding unit from a preset position in the current coding unit.

[00104] According to one embodiment, when the current bin is split into a predetermined number of bins, the image decoding apparatus 150 may select one of the bins. Various methods may be used to select one of a plurality of coding units, as will be described below in relation to various embodiments.

[00105] According to one embodiment, the image decoding apparatus 150 may split the current coding unit into a plurality of coding units, and may determine the coding unit at a preset position.

[00106] FIG. 5 shows a method performed by the image decoding apparatus 150 for determining a preset position coding unit from an odd number of coding units, according to one embodiment.

[00107] According to one embodiment, the image decoding apparatus 150 may use information indicative of positions of an odd number of coding units to determine a coding unit at a center position from an odd number of coding units. Referring to FIG. 5, the image decoding apparatus 150 may determine an odd number of coding units 520a, 520b, and 520c by splitting the current coding unit 500. The image decoding apparatus 150 may determine the coding unit 520b at the center position by position information of an odd number of coding units 520a, 520b, and 520c. For example, the image decoding apparatus 150 may determine center position encoding element 520b by determining the positions of encoding elements 520a, 520b, and 520c based on information indicative of the positions of preset samples included in encoding elements 520a, 520b, and 520c. In detail, the image decoding apparatus 150 can determine the coding element 520b at the center position by determining the positions of the coding elements 520a, 520b, and 520c based on information indicative of the positions of the top-left samples 530a, 530b, and 530c of the coding elements 520a, 520b, and 520c.

[00108] According to one embodiment, information indicative of the positions of the top left discrete samples 530a, 530b, and 530c that are included in coding elements 520a, 520b, and 520c, respectively, may include information about positions or coordinates of elements 520a, 520b, and 520c frame encoding. According to one embodiment, information indicative of the positions of the top left discrete samples 530a, 530b, and 530c that are included in coding units 520a, 520b, and 520c, respectively, may include information indicative of the widths or heights of coding units 520a, 520b, and 520c, included in the current coding unit 500, and the width or height may correspond to information indicating differences between the coordinates of coding units 520a, 520b, and 520c in a frame. That is, the image decoding apparatus 150 can determine the coding element 520b at the center position by directly using information about the positions or coordinates of the coding elements 520a, 520b, and 520c in the frame, or by information about the widths or heights of the coding elements that correspond to the value difference between the coordinates.

[00109] According to one embodiment, information indicating the position of the upper left sample 530a of the upper layer coding element 520a may include coordinates (xa, ya), information indicating the position of the upper left sample 530b of the middle coding element 520b may include (xb, yb) coordinates, and information indicating the position of the upper left sample 530c of the lower coding element 520c may include (xc, yc) coordinates. The image decoding apparatus 150 may determine the middle coding unit 520b by the coordinates of the upper left discrete samples 530a, 530b, and 530c that are included in the coding units 520a, 520b, and 520c, respectively. For example, when the coordinates of the top left samples 530a, 530b, and 530c are sorted in ascending or descending order, the coding unit 520b including the coordinates (xb, yb) of the sample 530b at the center position may be determined as the coding unit at the center position among coding elements 520a, 520b, and 520c determined by splitting the current coding element 500. However, the coordinates indicating the positions of the top left discrete samples 530a, 530b, and 530c may include coordinates indicating absolute positions in the frame, or may use coordinates (dxb, dyb) indicating the relative position of the top left discrete sample 530b of the middle coding element 520b, and coordinates (dxc, dyc) indicating the relative position of the upper left sample 530c of the lower coding element 520c with reference to the position of the upper left sample 530a of the upper coding element 520a. The method of determining the coding element at the preset position by the coordinates of the sample included in the coding element as information indicating the position of the sample is not limited to the above-described method, and may include various arithmetic methods with the possibility of using the coordinates of the sample.

[00110] According to one embodiment, the image decoding apparatus 150 may partition the current coding unit 500 into a plurality of coding units 520a, 520b, and 520c, and may select one of the coding units 520a, 520b, and 520c based on a predetermined criterion. For example, the image decoding apparatus 150 may select a coding unit 520b that has a size different from the size of the others from the coding units 520a, 520b, and 520c.

[00111] According to one embodiment, the image decoding apparatus 150 may determine the widths or heights of the coding units 520a, 520b, and 520c by means of coordinates (xa, ya) indicating the position of the upper left sample 530a of the upper coding unit 520a, coordinates (xb, yb) , indicating the position of the upper left sample 530b of the middle bin 520b, and coordinates (xc, yc) indicating the position of the upper left bin 530c of the lower coding element 520c. The image decoding apparatus 150 may determine the respective sizes of encoding elements 520a, 520b, and 520c by means of (xa, ya), (xb, yb), and (xc, yc) coordinates indicating positions of encoding elements 520a, 520b, and 520c.

[00112] According to one embodiment, the image decoding apparatus 150 may determine the width of the top encoding element 520a as xb-xa and determine its height as yb-ya. According to one embodiment, the image decoding apparatus 150 may determine the width of the middle coding element 520b as xc-xb and determine its height as yc-yb. According to one embodiment, the image decoding apparatus 150 may determine the width or height of the bottom coding unit 520c by the width or height of the current coding unit 500 or the widths or heights of the top and middle coding units 520a and 520b. The image decoding apparatus 150 may determine a coding unit that has a different size than the others based on the determined widths and heights of the coding units 520a, 520b, and 520c. Referring to FIG. 5, the image decoding apparatus 150 may determine the middle coding unit 520b, which has a size different from that of the top and bottom coding units 520a and 520c, as the preset position coding unit. However, the above-described method performed by the image decoding apparatus 150 of determining a coding unit having a size different from the size of other coding units only corresponds to an example of determining a coding unit at a preset position by coding unit sizes that are determined based on sample coordinates, and thus Thus, various methods of determining the coding element at a preset position by comparing the sizes of the coding elements, which are determined based on the coordinates of the preset discrete samples, can be used.

[00113] However, the positions of the samples considered to determine the positions of the bins are not limited to the upper left positions described above, and information about the random positions of the bins included in the bins can be used.

[00114] According to one embodiment, the image decoding apparatus 150 may select a coding unit at a preset position from an odd number of coding units determined by splitting the current coding unit based on the shape of the current coding unit. For example, when the current coding unit has a non-square shape whose width is longer than the height, the image decoding apparatus 150 may determine the coding unit at a preset position in the horizontal direction. That is, the image decoding apparatus 150 can determine one of the coding units at various positions in the horizontal direction and impose a constraint on the coding unit. When the current coding unit has a non-square shape whose height is longer than the width, the image decoding apparatus 150 may determine the coding unit at a preset position in the vertical direction. That is, the image decoding apparatus 150 may determine one of the coding units at various positions in the vertical direction, and may impose a constraint on the coding unit.

[00115] According to one embodiment, the image decoding apparatus 150 can use information indicating the respective positions of an even number of coding units to determine a coding unit at a preset position from an even number of coding units. The image decoding apparatus 150 may determine an even number of coding units by splitting the current coding unit, and may determine a coding unit at a preset position by position information of an even number of coding units. The operation associated with this may correspond to the operation of determining the coding element at a preset position (eg, the center position) from an odd number of coding elements, which was described in detail above with respect to Fig. 5, and thus its detailed descriptions are not provided here. .

[00116] According to one embodiment, when the non-square current coding unit is split into a plurality of coding units, preset information about the coding unit at the preset position may be used in a split operation to determine the coding unit at the preset position from the plurality of coding units. For example, the image decoding apparatus 150 may use at least one of the block shape information and the split shape information that is stored in the sample included in the coding unit at the center position in a split operation to determine the coding unit at the center position from a plurality of coding units. , determined by splitting the current encoding element.

[00117] Referring to FIG. 5, the image decoding apparatus 150 may partition the current coding unit 500 into a plurality of coding units 520a, 520b, and 520c based on at least one of the block shape information and the split shape information, and may determine the coding unit 520b at the center position from a plurality of coding elements 520a, 520b, and 520c. In addition, the image decoding apparatus 150 can determine the coding unit 520b at the center position in view of the position from which at least one of the block shape information and the split shape information is obtained. That is, at least one of the block shape information and split shape information of the current coding unit 500 can be obtained from a sample 540 at the center position of the current coding unit 500, and when the current coding unit 500 is split into a plurality of coding units 520a, 520b, and 520c based on at least one of the block shape information and the split shape information, the coding unit 520b including the discrete sample 540 can be determined as the coding unit at the center position. However, the information used to determine the coding element at the center position is not limited to at least one of the block shape information and the split shape information, and various types of information can be used to determine the coding element at the center position.

[00118] According to one embodiment, preset information for identifying a coding element at a preset position can be obtained from a preset sample included in a coding element to be determined. Referring to FIG. 5, the image decoding apparatus 150 may use at least one of the block shape information and the split shape information that is obtained from the sample at a preset position in the current coding element 500 (for example, the sample at the center position of the current element 500 coding) to determine the coding unit at a preset position from the plurality of coding units 520a, 520b, and 520c determined by splitting the current coding unit 500 (eg, the coding unit at the center position of the set of split coding units). That is, the image decoding apparatus 150 can determine the sample at a preset position considering the block shape of the current coding unit 500, determine the coding unit 520b including the sample from which the preset information (for example, at least one of the block shape information and split shape information) may be obtained from a plurality of coding units 520a, 520b, and 520c determined by splitting the current coding unit 500, and may impose a predetermined restriction on coding unit 520b. Referring to FIG. 5, according to one embodiment, the image decoding apparatus 150 may determine a sample 540 at the center position of the current coding element 500 as a sample from which preset information can be obtained, and may impose a preset constraint on the element 520b encoding, including discrete sample 540, in the decoding operation. However, the sample position from which the preset information can be obtained is not limited to the above-described position, and may include arbitrary sample positions included in the encoding element 520b to be determined for restriction.

[00119] According to one embodiment, the sample position from which preset information can be obtained can be determined based on the shape of the current coding element 500. According to one embodiment, the block shape information may indicate whether the current coding unit is square or non-square, and the sample position from which preset information may be obtained may be determined based on the shape. For example, the image decoding apparatus 150 may determine a sample located at the boundary for dividing at least one of the width and height of the current coding unit in half as a sample from which preset information can be obtained by at least one of the information about the width of the current encoding element and information about the height of the current encoding element. As another example, when the block shape information of the current coding unit indicates a non-square shape, the image decoding apparatus 150 may determine one of the discrete samples adjacent to the boundary for bisecting the long side of the current coding unit as a discrete sample from which preset information can be received.

[00120] According to one embodiment, when the current coding unit is partitioned into a plurality of coding units, the image decoding apparatus 150 may use at least one of the block shape information and split shape information to determine the coding unit at a preset position from the plurality of coding units. According to one embodiment, the image decoding apparatus 150 may obtain at least one of the block shape information and the split shape information from a sample at a preset position in a coding unit, and may split a plurality of bins that are generated by splitting the current coding unit by at least one of split shape information and block shape information that is obtained from a sample of a preset position in each of the plurality of coding units. That is, a coding unit can be recursively split based on at least one of the block shape information and the split shape information that is obtained from a sample at a preset position in each coding unit. The operation of recursively splitting the coding element has been described above with respect to FIG. 4, and thus its detailed descriptions will not be given here.

[00121] According to one embodiment, the image decoding apparatus 150 may determine one or more coding units by splitting the current coding unit, and may determine the decoding order of one or more coding units based on a preset block (eg, the current coding unit).

[00122] FIG. 6 shows a processing order of a plurality of coding units when the image decoding apparatus 150 determines a plurality of coding units by splitting a current coding unit according to one embodiment.

[00123] According to one embodiment, the image decoding apparatus 150 may determine the second coding units 610a and 610b by splitting the first coding unit 600 in the vertical direction, may determine the second coding units 630a and 630b by splitting the first coding unit 600 in the horizontal direction, or may determine second encoding units 650a, 650b, 650c, and 650d by splitting the first encoding unit 600 in the vertical and horizontal directions based on the block shape information and the split shape information.

[00124] Referring to FIG. 6, the image decoding apparatus 150 may determine to process second coding units 610A and 610b, which are determined by splitting the first coding unit 600 in the vertical direction, in horizontal direction order 610c. The image decoding apparatus 150 may decide to process the second coding units 630A and 630b, which are determined by splitting the first coding unit 600 in the horizontal direction, in the vertical direction order 630c. The image decoding apparatus 150 may decide to process the second coding units 650a, 650b, 650c, and 650d, which are determined by splitting the first coding unit 600 in the vertical and horizontal directions according to a predetermined order (e.g., raster scan order or Z-scan order 650e), whereby the encoding elements in a line are processed, and then the encoding elements in the next line are processed.

[00125] According to one embodiment, the image decoding apparatus 150 may recursively split coding units. Referring to FIG. 6, the image decoding apparatus 150 may determine a plurality of encoding elements 610a, 610b, 630a, 630b, 650a, 650b, 650c, and 650d by splitting the first encoding element 600, and may recursively split each of the determined plurality of encoding elements 610a, 610b, 630a, 630b, 650a, 650b, 650c and 650d encoding. The partitioning method of the plurality of coding elements 610a, 610b, 630a, 630b, 650a, 650b, 650c, and 650d may correspond to the partitioning method of the first encoding element 600. Thus, each of the plurality of coding units 610a, 610b, 630a, 630b, 650a, 650b, 650c, and 650d can be independently divided into a plurality of coding units. Referring to FIG. 6, the image decoding apparatus 150 may determine the second coding units 610A and 610b by splitting the first coding unit 600 in the vertical direction, and may determine whether or not to split each of the second coding units 610a and 610b independently.

[00126] According to one embodiment, the image decoding apparatus 150 may determine the third coding units 620a and 620b by splitting the left second coding unit 610a in the horizontal direction, and may not split the right second coding unit 610b.

[00127] According to one embodiment, the processing order of the coding units can be determined based on the coding unit split operation. In other words, the processing order of split coding units can be determined based on the processing order of coding units just before splitting. The image decoding apparatus 150 may determine the processing order of the third coding units 620a and 620b determined by splitting the left second coding unit 610a independently of the right second coding unit 610b. Since the third coding units 620A and 620b are determined by splitting the left second coding unit 610a in the horizontal direction, the third coding units 620a and 620b can be processed in the order 620c of the vertical direction. Since the left and right second encoding elements 610a and 610b are processed in the horizontal direction order 610c, the right second encoding element 610b can be processed after the third encoding elements 620a and 620b included in the left second encoding element 610a are processed in the vertical direction order 620c. The operation of determining the processing order of coding units based on the coding unit before splitting is not limited to the above-described example, and various methods can be used to independently process coding units that are split and determined with different shapes in a predetermined order.

[00128] Fig. 7 depicts a process performed by the image decoding apparatus 150 for determining that the current coding unit should be split into an odd number of coding units when the coding units are unable to be processed in a preset order, according to one embodiment.

[00129] According to one embodiment, the image decoding apparatus 150 can determine whether the current coding unit should be split into an odd number of coding units based on the received block shape information and split shape information. Referring to FIG. 7, the square first coding unit 700 may be split into non-square second coding units 710a and 710b, and the second coding units 710a and 710b may be independently divided into third coding units 720a and 720b, and 720c-720e. According to one embodiment, the image decoding apparatus 150 may determine a plurality of third coding units 720a and 720b by splitting the left second coding unit 710a in the horizontal direction, and may split the right second coding unit 710b into an odd number of third coding units 720c-720e.

[00130] According to one embodiment, the image decoding apparatus 150 may determine whether any coding unit should be split into an odd number of coding units by determining whether it is possible to process the third coding units 720a and 720b, and 720c-720e in advance. in the prescribed manner. Referring to FIG. 7, the image decoding apparatus 150 may determine third coding units 720a and 720b, and 720c-720e by recursively splitting the first coding unit 700. The image decoding apparatus 150 may determine whether any of the first coding unit 700, second coding units 710a and 710b and third coding units 720a and 720b, and 720c, 720d and 720e coding be split into an odd number of coding units based on at least one of block shape information and split shape information. For example, the second coding unit to the right of the second coding units 710a and 710b may be split into an odd number of third coding units 720c, 720d and 720e. The processing order of the plurality of coding elements included in the first coding element 700 may be a predetermined order (for example, the order of the Z-scan 730), and the image decoding apparatus 70 may determine whether the third coding elements 720c, 720d, and 720e are satisfied, which are determined by splitting the right second coding unit 710B into an odd number of coding units, a condition for processing in a predetermined order.

[00131] According to one embodiment, the image decoding apparatus 150 can determine whether the third encoding elements 720a and 720b and 720c, 720d and 720e included in the first encoding element 700 satisfy a condition for processing in a preset order, and the condition refers to whether at least one of the width and height of the second coding elements 710a and 710b is to be bisected along the boundary of the third coding elements 720a and 720b and 720c, 720d and 720e. For example, the third coding units 720a and 720b determined by dividing the height of the non-square left second coding unit 710a in half satisfy the condition. However, since the boundaries of the third encoding elements 720c, 720d, and 720e determined by dividing the right second encoding element 710b into three encoding elements do not bisect the width or height of the right second encoding element 710b, it can be determined that the third encoding elements 720c, 720d, and 720e do not satisfy the condition. When the condition is not satisfied as described above, the image decoding apparatus 150 may decide to decouple the scan order and determine that the right second coding unit 710b should be split into an odd number of coding units based on the decision result. According to one embodiment, when a coding unit is partitioned into an odd number of coding units, the image decoding apparatus 150 may impose a preset constraint on the coding unit at a preset position of the coding units. The limitation or preset position has been described above with respect to various embodiments, and thus, detailed descriptions thereof will not be given here.

[00132] FIG. 8 shows a process performed by the image decoding apparatus 150 to determine at least one coding unit by splitting the first coding unit 800, according to one embodiment. According to one embodiment, the image decoding apparatus 150 may split the first coding unit 800 based on at least one of the block shape information and the split shape information that is obtained by the receiver 160. The square first coding unit 800 may be split into four square coding units, or may be split into many non-square coding units. For example, with reference to FIG. 8, when the block shape information indicates that the first coding unit 800 has a square shape, and the partition shape information indicates to split the first coding unit 800 into non-square coding units, the image decoding apparatus 150 may split the first coding unit 800 into set of non-square coding elements. In more detail, when the split shape information indicates to determine an odd number of coding units by splitting the first coding unit 800 in the horizontal direction or the vertical direction, the image decoding apparatus 150 may split the square first coding unit 800 into an odd number of coding units, such as the second units 810a, 810b and 810c coding determined by splitting the square first coding unit 800 in the vertical direction, or second coding units 820a, 820b, and 820c determined by splitting the square first coding unit 800 in the horizontal direction.

[00133] According to one embodiment, the image decoding apparatus 150 can determine whether the second coding units 810a, 810b, 810c, 820a, 820b, and 820c included in the first coding unit 800 satisfy a condition for processing in a preset order, and the condition applies to whether at least one of the width and height of the first coding unit 800 is to be bisected along the boundary of the second coding units 810a, 810b, 810c, 820a, 820b, and 820c. Referring to FIG. 8, since the boundaries of the second coding units 810a, 810b, and 810c determined by splitting the square first coding unit 800 in the vertical direction do not bisect the height of the first coding unit 800, it can be determined that the first coding unit 800 does not satisfy condition for processing in a predetermined order. Further, since the boundaries of the second coding units 820a, 820b, and 820c determined by splitting the square first coding unit 800 in the horizontal direction do not bisect the width of the first coding unit 800, it can be determined that the first coding unit 800 does not satisfy the condition for pre-processing. in the prescribed manner. When the condition is not satisfied as described above, the image decoding apparatus 150 may decide to decouple the scan order, and may determine that the first coding unit 800 should be split into an odd number of coding units based on the decision result. According to one embodiment, when a coding unit is split into an odd number of coding units, the image decoding apparatus 150 may impose a preset constraint on the coding unit at a preset position of the split coding units. The limitation or preset position has been described above with respect to various embodiments, and thus, detailed descriptions thereof will not be given here.

[00134] According to one embodiment, the image decoding apparatus 150 can determine coding units of various shapes by splitting the first coding unit.

[00135] Referring to FIG. 8, the image decoding apparatus 150 may partition the square first coding unit 800 or the non-square first coding unit 830 or 850 into coding units of various shapes.

[00136] FIG. 9 shows that the shape into which the second coding unit is splittable by the image decoding device 150 is limited when the second coding unit having a non-square shape, which is determined by splitting the first coding unit 900, satisfies a predetermined condition according to one embodiment.

[00137] According to one embodiment, the image decoding apparatus 150 may decide to split the square first coding unit 900 into non-square second coding units 910a, 910b, 920a, and 920b based on at least one of the block shape information and the split shape information that is obtained receiver 160. The second coding units 910a, 910b, 920a, and 920b may be independently split. Thus, the image decoding apparatus 150 can decide whether or not to split the first coding unit 900 into a plurality of coding units based on at least one of the block shape information and the split shape information of each of the second coding units 910a, 910b, 920a, and 920b. . According to one embodiment, the image decoding apparatus 150 can determine the third coding units 912a and 912b by splitting the non-square left second coding unit 910a, which is determined by splitting the first coding unit 900 in the vertical direction, in the horizontal direction. However, when the left second coding unit 910a is split in the horizontal direction, the image decoding apparatus 150 may restrict the right second coding unit 910b not to split in the horizontal direction in which the left second coding unit 910a is split. When the third coding units 914a and 914b are determined by splitting the right second coding unit 910b in the same direction, since the left and right second coding units 910a and 910b are independently split in the horizontal direction, the third coding units 912a, 912b, 914a, and 914b may be defined. However, this case also serves as a case in which the image decoding apparatus 150 partitions the first coding unit 900 into four square second coding units 930a, 930b, 930c, and 930d based on at least one of the block shape information and the split shape information, and may be inefficient in terms of image decoding.

[00138] According to one embodiment, the image decoding apparatus 150 may determine the third coding units 922a, 922b, 924a, and 924b by splitting the non-square second coding unit 920a or 920b, which is determined by splitting the first coding unit 900 in the horizontal direction, in the vertical direction. However, when the second coding unit (for example, the upper second coding unit 920a) is split in the vertical direction, for the reason described above, the image decoding apparatus 150 may restrict the other second coding unit (for example, the lower second coding unit 920b) so that it does not split in the vertical direction. , in which the upper second coding element 920a is split.

[00139] FIG. 18 shows a process performed by the image decoding apparatus 150 for splitting a square coding unit when the split shape information indicates that the square coding unit should not be split into four square coding units, according to one embodiment.

[00140] According to one embodiment, the image decoding apparatus 150 may determine the second coding units 1010a, 1010b, 1020a, 1020b, etc. by splitting the first encoding unit 1000 based on at least one of the block shape information and the split shape information. The partition shape information may include information on various partitioning methods of a coding element, but information on various partitioning methods may not include information for partitioning a coding element into four square coding elements. According to such partition shape information, the image decoding apparatus 150 may not partition the first square coding unit 1000 into four square second coding units 1030a, 1030b, 1030c, and 1030d. The image decoding apparatus 150 may determine the non-square second coding units 1010a, 1010b, 1020a, 1020b, etc. based on the partition shape information.

[00141] According to one embodiment, the image decoding apparatus 150 can independently split non-square second coding units 1010a, 1010b, 1020a, 1020b, etc.. Each of the second elements 1010a, 1010b, 1020a, 1020b, etc. The encoding may be recursively split in a predetermined order, and this splitting method may correspond to the splitting method of the first encoding element 1000 based on at least one of the block shape information and the split shape information.

[00142] For example, the image decoding apparatus 150 may determine the square third coding units 1012a and 1012b by splitting the left second coding unit 1010a in the horizontal direction, and may determine the square third coding units 1014a and 1014b by splitting the right second coding unit 1010b in the horizontal direction. In addition, the image decoding apparatus 150 can determine the square third coding units 1016a, 1016b, 1016c, and 1016d by splitting both of the left and right second coding units 1010a and 1010b in the horizontal direction. In this case, coding units having the same shape as the four square second coding units 1030a, 1030b, 1030c, and 1030d split from the first coding unit 1000 can be determined.

[00143] As another example, the image decoding apparatus 150 may determine the square third coding units 1022a and 1022b by splitting the upper second coding unit 1020a in the vertical direction, and may determine the square third coding units 1024a and 1024b by splitting the lower second coding unit 1020b in the vertical direction. direction. In addition, the image decoding apparatus 150 can determine the square third coding units 1022a, 1022b, 1024a, and 1024b by splitting both of the upper and lower second coding units 1020a and 1020b in the vertical direction. In this case, coding units having the same shape as the four square second coding units 1030a, 1030b, 1030c, and 1030d split from the first coding unit 1000 can be determined.

[00144] FIG. 11 shows that the processing order between multiple coding units may be changed depending on the coding unit splitting process according to one embodiment.

[00145] According to one embodiment, the image decoding apparatus 150 may split the first coding element 1100 based on the block shape information and the split shape information. When the block shape information indicates a square shape and the split shape information indicates to split the first coding unit 1100 in at least one of the horizontal and vertical directions, the image decoding apparatus 150 may determine the second coding units 1110a, 1110b, 1120a, and 1120b by splitting the first coding unit 1100 . Referring to FIG. 11, the non-square second coding units 1110a, 1110b, 1120a, and 1120b determined by splitting the first coding unit 1100 only in the horizontal direction or the vertical direction can be independently split based on the block shape information and split shape information of each coding unit. For example, the image decoding apparatus 150 may determine the third coding units 1116a, 1116b, 1116c, and 1116d by splitting the second coding units 1110a and 1110b, which are generated by splitting the first coding unit 1100 in the vertical direction, in the horizontal direction, and may determine the third units 1126a, 1126b, 1126c, and 1126d encoding by splitting the second coding units 1120a and 1120b, which are generated by splitting the first coding unit 1100 in the horizontal direction, in the vertical direction. The operation of splitting the second coding units 1110a, 1110b, 1120a, and 1120b has been described above with respect to FIG. 9, and thus their detailed descriptions will not be given here.

[00146] According to one embodiment, the image decoding apparatus 150 may process coding units in a predetermined order. The operation of processing coding elements in a preset order has been described above with respect to FIG. 6, and thus its detailed descriptions will not be given here. Referring to FIG. 11, the image decoding apparatus 150 may determine four square third coding units 1116a, 1116b, 1116c, and 1116d, and 1126a, 1126b, 1126c, and 1126d coding by splitting the square first coding unit 1100. According to one embodiment, the image decoding apparatus 150 may determine the processing orders of the third encoding elements 1116a, 1116b, 1116c, and 1116d, and 1126a, 1126b, 1126c, and 1126d based on the partitioning method of the first encoding element 1100.

[00147] According to one embodiment, the image decoding apparatus 150 may determine the third coding units 1116a, 1116b, 1116c, and 1116d by splitting the second coding units 1110a and 1110b generated by splitting the first coding unit 1100 in the vertical direction, in the horizontal direction, and may process third coding units 1116a, 1116b, 1116c and 1116d in the processing order 1117 to initially process the third coding units 1116a that are included in the left second coding unit 1110a in the vertical direction and then process the third coding units 1116b and 1116d that are included in the right second coding unit 1110b encoding, in the vertical direction.

[00148] According to one embodiment, the image decoding apparatus 150 may determine the third coding units 1126a, 1126b, 1126c, and 1126d by splitting the second coding units 1120a and 1120b generated by splitting the first coding unit 1100 in the horizontal direction, in the vertical direction, and may process third encoding elements 1126a, 1126b, 1126c and 1126d in the processing order 1127 to initially process the third encoding elements 1126a and 1126b that are included in the upper second encoding element 1120a in the horizontal direction and then to process the third encoding elements 1126c and 1126d that are included in the lower second coding element 1120b, in the horizontal direction.

[00149] With reference to FIG. 11, square third encoding elements 1116a, 1116b, 1116c, and 1116d, and 1126a, 1126b, 1126c, and 1126d may be determined by splitting second encoding elements 1110a, 1110b, 1120a, and 1120b, respectively. Although the second coding units 1110a and 1110b are determined by splitting the first coding unit 1100 in the vertical direction in a different manner from the second coding units 1120a and 1120b, which are determined by splitting the first coding unit 1100 in the horizontal direction, the third units 1116a, 1116b, 1116c, and 1116d, and 1126a, 1126b, 1126c, and 1126d encodings broken down from them end up showing the same shape coding units broken down from the first encoding element 1100. Thus, by recursively splitting the coding unit in various ways based on at least one of the block shape information and the split shape information, the image decoding apparatus 150 can process a plurality of coding units in different orders, even when the coding units are ultimately determined to have the same and the same shape.

[00150] FIG. 12 depicts a process of determining the depth of an coding unit as the shape and size of the coding unit change when the coding unit is recursively split such that a plurality of coding units is determined, according to one embodiment.

[00151] According to one embodiment, the image decoding apparatus 150 may determine the coding unit depth based on a predetermined criterion. For example, the preset criterion may be the length of the long side of the coding element. When the length of the long side of the coding unit before splitting is 2n times (n>0) greater than the length of the long side of the split current coding unit, the image decoding apparatus 150 may determine that the depth of the current coding unit is increased relative to the depth of the coding unit before splitting by n. In the following description, a coding element having increased depth is expressed as a coding element of deeper depth.

[00152] Referring to FIG. 12, according to one embodiment, the image decoding apparatus 150 may determine the second coding unit 1202 and the third deeper-depth coding unit 1204 by splitting the square first coding unit 1200 based on block shape information indicating a square shape ( for example, block shape information can be expressed as "0: SQUARE"). Assuming that the size of the square first coding unit 1200 is 2N×2N, the second coding unit 1202, determined by dividing the width and height of the first coding unit 1200 by 2, may have a size of N×N. In addition, the third coding element 1204, determined by dividing the width and height of the second coding element 1202 by 2, may have a size of N/2×N/2. In this case, the width and height of the third encoding element 1204 are equal to 1/2 of the width and height of the first encoding element 1200. When the depth of the first encoding element 1200 is D, the depth of the second encoding element 1202, whose width and height are 1/2 of the width and height of the first encoding element 1200, may be D+1, and the depth of the third encoding element 1204, whose width and height are equal to 1/2 of the width and height of the first coding element 1200, may be equal to D+2.

[00153] According to one embodiment, the image decoding apparatus 150 may determine the second coding element 1212 or 1222 and the third deeper-depth coding element 1214 or 1224 by splitting the non-square first coding element 1210 or 1220 based on block shape information indicating a non-square shape (e.g. , the block shape information can be expressed as "1: NS_VER", which indicates a non-square shape whose height is longer than the width, or as "2: NS_HOR", which indicates a non-square shape whose width is longer than the height).

[00154] The image decoding apparatus 150 may determine the second coding unit 1202, 1212, or 1222 by dividing at least one of the width and height of the first coding unit 1210 having a size of N×2N. That is, the image decoding apparatus 150 may determine a second coding unit 1202 having a size of N×N or a second coding unit 1222 having a size of N×N/2 by splitting the first coding unit 1210 in a horizontal direction, or may determine a second coding unit 1212, having a size of N/2×N by splitting the first coding unit 1210 in the horizontal and vertical directions.

[00155] According to one embodiment, the image decoding apparatus 150 may determine the second coding unit 1202, 1212, or 1222 by dividing at least one of the width and height of the first coding unit 1220 having a size of 2N×N. That is, the image decoding apparatus 150 may determine a second coding unit 1202 having a size of N×N or a second coding unit 1212 having a size of N/2×N by splitting the first coding unit 1220 in the vertical direction, or may determine the second coding unit 1222, having a size of N×N/2 by splitting the first coding unit 1220 in the horizontal and vertical directions.

[00156] According to one embodiment, the image decoding apparatus 150 may determine the third coding unit 1204, 1214, or 1224 by separating at least one of the width and height of the second coding unit 1202 having a size of N×N. That is, the image decoding apparatus 150 may determine a third coding unit 1204 having a size of N/2×N/2, a third coding unit 1214 having a size of N/2×N/2, or a third coding unit 1224 having a size of N/2× N/2, by splitting the second coding element 1202 in the vertical and horizontal directions.

[00157] According to one embodiment, the image decoding apparatus 150 may determine the third coding unit 1204, 1214, or 1224 by separating at least one of the width and height of the second coding unit 1212 having a size of N/2×N. That is, the image decoding apparatus 150 can determine a third coding unit 1204 having a size of N/2xN/2 or a third coding unit 1224 having a size of N/2xN/2 by splitting the second coding unit 1212 in the horizontal direction, or may determine a third coding element 1214 having a size of N/2×N/2 by splitting the second coding element 1212 in vertical and horizontal directions.

[00158] According to one embodiment, the image decoding apparatus 150 may determine the third coding unit 1204, 1214, or 1224 by dividing at least one of the width and height of the second coding unit 1212 having a size of N×N/2. That is, the image decoding apparatus 150 may determine a third coding unit 1204 having a size of N/2×N/2 or a third coding unit 1214 having a size of N/2×N/2 by splitting the second coding unit 1222 in the vertical direction, or may determine a third coding unit 1224 having a size of N/2×N/2 by splitting the second coding unit 1222 in vertical and horizontal directions.

[00159] According to one embodiment, the image decoding apparatus 150 may split the square encoding element 1200, 1202, or 1204 in the horizontal or vertical direction. For example, the image decoding apparatus 150 may determine the first coding unit 1210 having a size of N×2N by splitting the first coding unit 1200 having a size of 2N×2N in the vertical direction, or may determine the first coding unit 1220 having a size of 2N×N, by splitting the first coding element 1200 in the horizontal direction. According to one embodiment, when the depth is determined based on the length of the longest side of the coding element, the depth of the coding element determined by splitting the first 2N×2N coding element 1200, 1202, or 1204 in the horizontal or vertical direction may be the same , which is the depth of the first coding element 1200, 1202, or 1204.

[00160] According to one embodiment, the width and height of the third coding element 1214 or 1224 may be 1/2 of the width and height of the first coding element 1210 or 1220. When the depth of the first encoding element 1210 or 1220 is D, the depth of the second encoding element 1212 or 1222 whose width and height are 1/2 of the width and height of the first encoding element 1210 or 1220 may be D+1, and the depth of the third encoding element 1214 or An encoding 1224 whose width and height are equal to 1/2 of the width and height of the first encoding element 1210 or 1220 may be D+2.

[00161] FIG. 13 shows depths that can be determined based on coding unit shapes and sizes, and part indices (PIDs) that are present to distinguish between coding units, according to one embodiment.

[00162] According to one embodiment, the image decoding apparatus 150 can determine second coding units of various shapes by splitting the square first coding unit 1300. Referring to FIG. 13, the image decoding apparatus 150 may determine the second encoding units 1302a and 1302b, 1304a and 1304b, and 1306a, 1306b, 1306c, and 1306d by splitting the first encoding unit 1300 in at least one of the vertical and horizontal directions based on partition shape information. That is, the image decoding apparatus 150 may determine the second encoding elements 1302a and 1302b, 1304a and 1304b, and 1306a, 1306b, 1306c, and 1306d based on the partition shape information of the first encoding element 1300.

[00163] According to one embodiment, the depth of the second encoding elements 1302a and 1302b, 1304a and 1304b, and 1306a, 1306b, 1306c and 1306d, which are determined based on the split shape information of the square first encoding element 1300, can be determined based on the length of their long sides. For example, since the side length of the square first coding element 1300 is equal to the length of the long side of the non-square second coding elements 1302a and 1302b and 1304a and 1304b coding, the first coding element 1300 and the non-square second coding elements 1302a and 1302b and 1304a and 1304b coding may have the same depth, such as D. However, when the image decoding apparatus 150 splits the first coding unit 1300 into four square second coding units 1306a, 1306b, 1306c, and 1306d based on the split shape information, since the side length of the square second coding units 1306a, 1306b, 1306c, and 1306d is 1/2 of the side length of the first encoding element 1300, the depth of the second encoding elements 1306a, 1306b, 1306c, and 1306d may be D+1, which is 1 deeper than the depth D of the first encoding element 1300.

[00164] According to one embodiment, the image decoding apparatus 150 may determine a plurality of second coding units 1312a and 1312b and 1314a, 1314b and 1314c by splitting the first coding unit 1310 whose height is longer than the width in the horizontal direction based on split shape information. According to one embodiment, the image decoding apparatus 150 may determine a plurality of second coding units 1322a and 1322b and 1324a, 1324b and 1324c by splitting the first coding unit 1320 whose width is longer than the height in the vertical direction based on the split shape information.

[00165] According to one embodiment, the depth of the second encoding elements 1312a and 1312b, 1314a, 1314b and 1314c, 1322a and 1322b and 1324a, 1324b and 1324c encoding, which are determined based on the split shape information of the non-square first encoding element 1310 or 1320, can be determined based on the length of their long side. For example, since the length of the side of the square second encoding elements 1312a and 1312b is equal to 1/2 of the length of the long side of the first encoding element 1310 having a non-square shape whose height is longer than the width, the depth of the square second encoding elements 1312a and 1312b is D+1, which is deeper than the depth D non-square first coding element 1310 by 1.

[00166] In addition, the image decoding apparatus 150 may split the non-square first coding unit 1310 into an odd number of second coding units 1314a, 1314b, and 1314c based on the split shape information. An odd number of second coding units 1314a, 1314b, and 1314c may include non-square second coding units 1314a and 1314c and a square second coding unit 1314b. In this case, since the length of the long side of the non-square second encoding elements 1314a and 1314c and the length of the side of the square second encoding element 1314b are equal to 1/2 of the length of the long side of the first encoding element 1310, the depth of the second encoding elements 1314a, 1314b and 1314c may be equal to D+ 1, which is deeper than the D depth of the non-square first coding unit 1310 by 1. The image decoding apparatus 150 can determine the depths of coding units split from the first non-square-shaped coding unit 1320 whose width is longer than the height, by means of the above-described method of determining the depths of coding units split from the first element 1310 coding.

[00167] According to one embodiment, the image decoding apparatus 150 may determine a PID for identifying broken coding units based on a size ratio between coding units when an odd number of broken coding units does not have equal sizes. Referring to FIG. 13, the center position encoding element 1314b of an odd number of broken encoding elements 1314a, 1314b, and 1314c may have a width equal to the width of the other encoding elements 1314a and 1314c, and a height that is twice the height of the other encoding elements 1314a and 1314c. That is, in this case, the encoding element 1314b at the center position may include two of the other encoding elements 1314a or 1314c. Thus, when the PID of the encoding element 1314b at the center position is 1 based on the scan order, the PID of the encoding element 1314c located next to the encoding element 1314b may be incremented by 2, and thus may be equal to 3. That is, discontinuity in PID values may be present. According to one embodiment, the image decoding apparatus 150 can determine whether an odd number of broken bins have equal sizes based on whether there is a discontinuity in the PID to identify the broken bins.

[00168] According to one embodiment, the image decoding apparatus 150 can determine whether to use a particular splitting method based on PID values to identify a plurality of coding units determined by splitting the current coding unit. Referring to FIG. 13, the image decoding apparatus 150 can determine an even number of coding units 1312a and 1312b or an odd number of coding units 1314a, 1314b and 1314c by splitting the first coding unit 1310 having a rectangular shape whose height is longer than the width. The image decoding apparatus 150 may use the PID to identify the respective coding units. According to one embodiment, the PID may be obtained from a sample of the preset position of each coding element (eg, the top left sample).

[00169] According to one embodiment, the image decoding apparatus 150 may determine the coding unit at a preset position from the broken coding units by PID to distinguish the coding units. According to one embodiment, when the partition shape information of the first coding unit 1310 having a rectangular shape whose height is longer than the width indicates to split the coding unit into three coding units, the image decoding apparatus 150 may split the first coding unit 1310 into three units 1314a, 1314b, and 1314c coding. The image decoding device 150 may assign a PID to each of the three encoding elements 1314a 1314b and 1314c. The image decoding apparatus 150 may compare the PIDs of an odd number of broken coding units to determine the coding unit in a central position among the coding units. The image decoding apparatus 150 may determine the coding element 1314b having a PID corresponding to the average of the coding element PIDs as the coding element at the center position among the coding elements determined by splitting the first coding element 1310. According to one embodiment, the image decoding apparatus 150 may determine a PID to distinguish split coding units based on a size ratio between coding units when split coding units are not of equal size. Referring to FIG. 13, encoding element 1314b generated by splitting first encoding element 1310 may have a width equal to the width of other encoding elements 1314a and 1314c and a height that is twice the height of other encoding elements 1314a and 1314c. In this case, when the PID of the encoding element 1314b at the center position is 1, the PID of the encoding element 1314c located next to the encoding element 1314b may be incremented by 2, and thus may be equal to 3. When the PID is incrementally increased as described above , the image decoding apparatus 150 may determine that the coding unit is split into a plurality of coding units including a coding unit having a size different from that of other coding units. According to one embodiment, when the split shape information indicates to split a coding unit into an odd number of coding units, the image decoding apparatus 150 may split the current coding unit such that the coding unit at a preset position of an odd number of coding units (for example, the coding unit in the center positions) has a size different from the size of other encoding elements. In this case, the image decoding apparatus 150 can determine the coding unit at the center position, which has a different size, by means of the PID of the coding units. However, the PID and the size or position of the coding element at the preset position are not limited to the above-described examples, and other PIDs and other positions and sizes of the coding elements can be used.

[00170] According to one embodiment, the image decoding apparatus 150 may use a preset data element where the coding element starts to break recursively.

[00171] FIG. 14 shows that a plurality of coding elements is determined based on a plurality of preset data elements included in a frame, according to one embodiment.

[00172] According to one embodiment, a preset data item may be defined as a data item where the coding item starts to recursively split by at least one of the block shape information and the split shape information. That is, the preset data element may correspond to the uppermost depth coding element that is used to determine the plurality of coding elements split from the current frame. In the following descriptions, for convenience of explanation, the preset data item is referred to as the reference data item.

[00173] According to one embodiment, the reference data element may have a preset size and a preset size shape. According to one embodiment, a reference coding element may include M×N discrete samples. Here, M and N may be equal to each other and may be integer values expressed as powers of 2. That is, the reference data element may be square or non-square and may be split into an integer number of coding units.

[00174] According to one embodiment, the image decoding apparatus 150 may partition the current frame into a plurality of reference data elements. According to one embodiment, the image decoding apparatus 150 may partition a plurality of reference data elements into which the current frame is partitioned by splitting information about each reference data element. The split operation of the reference data element may correspond to the split operation using a quadtree structure.

[00175] According to one embodiment, the image decoding apparatus 150 may previously determine the minimum size allowed for the reference data elements included in the current frame. Accordingly, the image decoding apparatus 150 can determine various reference data elements having sizes greater than or equal to the minimum size, and can determine one or more coding elements by block shape information and split shape information with references to the specific reference data element.

[00176] Referring to FIG. 14, the image decoding apparatus 150 may use a square reference coding element 1400 or a non-square reference coding element 1402. According to one embodiment, the shape and size of the reference coding units may be determined based on various data elements, with the ability to include one or more reference coding units (eg, sequences, frames, slices, slice segments, largest coding units, or the like).

[00177] According to one embodiment, the receiver 160 of the image decoding apparatus 150 can obtain, from the bitstream, at least one of the reference coding unit shape information and the reference coding unit size information with respect to each of the various data units. The operation of splitting a square reference coding element 1400 into one or more coding units has been described above with respect to the operation of splitting the current coding unit 1000 of FIG. current element 1100 or 1150 coding from Fig.11. Thus, their detailed descriptions will not be given here.

[00178] According to one embodiment, the image decoding apparatus 150 may use the PID to identify the size and shape of the reference coding units to determine the size and shape of the reference coding units according to some data elements previously determined based on a predetermined condition. That is, receiver 160 can only obtain a PID from the bitstream to identify the size and shape of the reference coding units with respect to each slice, slice segment, or largest coding unit that is a data unit that satisfies a predetermined condition (e.g., a data unit having a size that less than or equal to slice), from various data elements (eg, sequences, frames, slices, slice segments, largest coding units, or the like). The image decoding apparatus 150 may determine the size and shape of the reference data items with respect to each data item that satisfies a predetermined condition by the PID. When the reference coding unit shape information and the reference coding unit size information are obtained and used from the bitstream according to each data unit having a relatively small size, the utilization efficiency of the bitstream may not be high, and thus only the PID can be obtained and used instead. directly obtaining the shape information of the reference encoding element and the size information of the reference encoding element. In this case, at least one of the size and shape of the reference coding units corresponding to the PID for identifying the size and shape of the reference coding units may be previously determined. That is, the image decoding apparatus 150 can determine at least one of the size and shape of the reference coding units included in the data element serving as the item for obtaining the PID by selecting the previously determined at least one of the size and shape of the reference coding units based on the PID. .

[00179] According to one embodiment, the image decoding apparatus 150 may use one or more reference coding units included in the largest coding unit. That is, the largest coding unit obtained by splitting from a frame may include one or more reference coding units, and the coding units may be determined by recursively splitting each reference coding unit. According to one embodiment, at least one of the width and height of the largest coding unit may be equal to, multiplied by an integer, at least one of the width and height of the reference coding units. According to one embodiment, the size of the reference coding units can be obtained by splitting the largest coding unit n times based on a quadtree structure. That is, the image decoding apparatus 150 can determine the reference coding units by splitting the largest coding unit n times based on the quadtree structure, and may split the reference coding unit based on at least one of the block shape information and the split shape information according to various embodiments.

[00180] FIG. 15 shows a processing unit that functions as an element for determining the order of determination of reference coding units included in frame 1500, according to one embodiment.

[00181] According to one embodiment, the image decoding apparatus 150 may determine one or more processing units split from a frame. A processing unit is a data element including one or more reference coding units split from a frame, and one or more reference coding units included in the processing unit may be determined according to a particular order. That is, the order for determining one or more reference coding units defined in each processing unit may correspond to one of various types of orders for determining reference coding units, and may vary depending on the processing unit. The determination order of the reference coding units, which is determined with respect to each processing unit, may be one of various orders, such as raster scan order, Z-scan, N-scan, top-right diagonal scan, horizontal scan, and vertical scan, but is not limited to the above orders. scanning.

[00182] According to one embodiment, the image decoding apparatus 150 can obtain processing block size information and can determine the size of one or more processing blocks included in a frame. The image decoding apparatus 150 may obtain processing block size information from the bitstream, and may determine the size of one or more processing blocks included in a frame. The size of the processing units may be a predetermined data element size, which is indicated by the size information of the processing unit.

[00183] According to one embodiment, the receiver 160 of the image decoding apparatus 150 can obtain processing block size information from the bitstream according to each specific data element. For example, the processing block size information may be obtained from a bitstream in a data element such as a picture, sequence, frame, slice, or slice segment. That is, the receiver 160 can obtain processing block size information from the bitstream according to each of the various data items, and the image decoding apparatus 150 can determine the size of one or more processing blocks into which a frame is divided by the obtained processing block size information. The size of the processing blocks may be an integer times the size of the reference coding units.

[00184] According to one embodiment, image decoding apparatus 150 may determine the size of processing units 1502 and 1512 included in frame 1500. For example, image decoding apparatus 150 may determine the size of processing units based on processing unit size information obtained from the bitstream. Referring to FIG. 15, according to one embodiment, the image decoding apparatus 150 may determine the width of the processing units 1502 and 1512 to be four times the width of the reference coding units, and may determine the height of the processing units 1502 and 1512 to be four times the height of the reference units. coding. The image decoding apparatus 150 may determine the order in which one or more reference coding units are determined in one or more processing units.

[00185] According to one embodiment, image decoding apparatus 150 may determine processing blocks 1502 and 1512 that are included in frame 1500 based on the size of processing blocks, and may determine the order in which one or more reference coding units are determined in processing blocks 1502 and 1512. According to one embodiment, determining the reference coding units may include determining the size of the reference coding units.

[00186] According to one embodiment, the image decoding apparatus 150 may obtain determination order information of one or more reference coding units included in one or more processing units from the bitstream, and may determine the determination order with respect to one or more reference coding units based on received information of the order of determination. The determination order information may be defined as an order or a direction for determining reference coding units in a processing unit. That is, the order in which the reference coding units are determined can be independently determined with respect to each processing unit.

[00187] According to one embodiment, the image decoding apparatus 150 can obtain reference coding order determination information from the bitstream according to each specific data element. For example, receiver 160 may obtain reference coding unit determination order information from the bitstream according to each data element such as a picture, sequence, frame, slice, slice segment, or processing unit. Since the reference coding unit determination order information indicates an order for determining the reference coding units in a processing unit, determination order information can be obtained with respect to each specific data unit including an integer number of processing units.

[00188] According to one embodiment, the image decoding apparatus 150 may determine one or more reference coding units based on a certain determination order.

[00189] According to one embodiment, the receiver 160 may obtain reference coding unit determination order information from the bitstream as information related to the processing of blocks 1502 and 1512, and the image decoding apparatus 150 may determine the determination order of one or more reference coding units included to processing blocks 1502 and 1512, and determine one or more reference coding units to be included in frame 1500 based on the order of determination. With reference to FIG. 15, image decoding apparatus 150 may determine determination orders 1504 and 1514 of one or more reference coding units in processing blocks 1502 and 1512, respectively. For example, when reference coding unit determination order information is obtained with respect to each processing unit, various types of reference coding unit determination order information may be obtained for processing units 1502 and 1512. When the order 1504 of determining the reference coding units in the processing block 1502 is a raster scan order, the reference coding units included in the processing block 1502 may be determined according to the raster scan order. Conversely, when the reference coding unit determination order 1514 in another processing block 1512 is the reverse order of the bitmap scan, the reference encoding elements included in the processing block 1512 may be determined according to the reverse order of the bitmap scan.

[00190] Figures 1-15 show a method for splitting an image into a largest coding unit and splitting a largest coding unit into coding units of a hierarchical tree structure. Fig.16-24 depict a method for determining the quantization parameter of the current block.

[00191] The image encoding apparatus 100 of FIG. 1 can transform, by a predetermined procedure, the residual data, which is the difference between the original pixel values included in the coding element and their prediction values. In this regard, the image encoding apparatus 100 can reduce the size of the residual data instead of losing the residual data by quantizing the transformed residual data.

[00192] The quantization of the residual data is performed based on the quantization parameter. The quantization parameter denotes an index used to find the scaling matrix needed to quantize the residual data of the current block. When the quantization parameter is large, the scaling matrix where the elements are relatively large is found. Thus, when the quantization parameter is large, the residual data is lost in large amounts, but the compression ratio of the residual data is increased. Conversely, when the quantization parameter is small, the scaling matrix where the elements are relatively small is found. Thus, when the quantization parameter is small, the residual data is lost in small amounts, but the compression ratio of the residual data is reduced.

[00193] That is, in the case where the subjective deterioration in image quality is small even when the compression ratio of the residual data is increased, a large quantization parameter can be used. However, in the case where a subjective image quality deterioration is detected when the compression ratio of the residual data is increased, a small quantization parameter should be used. Thus, different quantization parameters must be used for blocks of the same frame to account for deterioration in image quality.

[00194] FIG. 16 shows an image decoding apparatus for determining a block quantization parameter and decoding block residual data according to the determined quantization parameter.

[00195] The image decoding apparatus 1600 includes a quantization parameter determiner 1610 and an inverse quantizer 1620. In FIG. quantizer 1620 may be combined into one component.

[00196] In FIG. 16, quantization parameter determiner 1610 and inverse quantizer 1620 are illustrated as being included in the same device, but devices performing the respective functions of quantization parameter determiner 1610 and inverse quantizer 1620 may not necessarily be physically adjacent to each other. Thus, in another embodiment, the quantization parameter determiner 1610 and the inverse quantizer 1620 may be separated.

[00197] The quantization parameter determiner 1610 and the inverse quantizer 1620 may be implemented by the same processor according to one embodiment. In another embodiment, the quantization parameter determiner 1610 and the inverse quantizer 1620 may be implemented by multiple processors.

[00198] The image decoding apparatus 1600 may perform inverse quantization based on a quantization group including one or more blocks. Next, the inverse quantization method based on the quantization group will be described.

[00199] When the quantization parameter is varied in each block, the quantization parameter information is increased. Thus, when a quantization parameter is determined with respect to a block element, coding efficiency may decrease. Thus, in order to increase the coding efficiency, a method for determining the same quantization parameter with respect to a plurality of blocks is considered.

[00200] In general, adjacent blocks have the same or similar quantization parameters. Thus, the image decoding apparatus 1600 can use the same quantization parameter for adjacent blocks. A plurality of blocks that are adjacent to each other and use the same quantization parameter is called a quantization group.

[00201] The quantization group may be determined based on the largest coding unit. For example, a quantization group may be set to blocks split from the largest coding unit a predetermined number of times. When the block for which the quantization group is set is not further split, the quantization group parameter of the quantization group is applied only to the block for which the quantization group is set. Conversely, when a block corresponding to a quantization group is further split, the quantization group parameter of the quantization group may be applied to all sub-blocks that are generated by splitting the block for which the quantization group is set.

[00202] Alternatively, the quantization group may be determined based on the size. For example, when the block size is less than or equal to the reference quantization group size, the quantization group may be set for the block. When the block for which the quantization group is set is not further partitioned, the quantization parameter of the quantization group is applied to only one block for which the quantization group is set. Conversely, the block corresponding to the quantization group is further split, the quantization group quantization parameter may be applied to all sub-blocks that are generated by splitting the block for which the quantization group is set. Accordingly, the block quantization parameter is determined based on the quantization block so that information about the quantization parameter is reduced.

[00203] The quantization parameter determiner 1610 may obtain a difference quantization parameter enable flag with respect to the top-level data element of the current quantization group. When the difference quantization parameter enable flag indicates that it is allowed to determine the quantization parameter according to the difference quantization parameter, the quantization parameter determining means 1610 may obtain the difference quantization parameter of the current block.

[00204] The top data element may be one of a Video Parameter Set (VPS), a Sequence Parameter Set (SPS), and a Frame Parameter Set (PPS). Thus, the quantization parameter determiner 1610 may apply the quantization group-based quantization parameter determination method to all blocks included in the top-level data element.

[00205] The quantization parameter determiner 1610 may obtain quantization group information regarding the top-level data element of the current quantization group. The quantization group information indicates a method for determining the quantization group. For example, the quantization group information may include block split information or block size information. When the difference quantization parameter enable flag enables the difference quantization parameter, the quantization parameter determiner 1610 can obtain quantization group information.

[00206] The quantization parameter determining means 1610 may determine a predicted quantization parameter of the current quantization group, wherein the predicted quantization parameter is determined according to at least one of the block split information and the block size information.

[00207] The block split information may include the number of split times with a quadtree and the number of split times without a quadtree. The number of times of quadtree splitting indicates how many times quadtree splitting is performed to obtain the current quantization group from the largest coding unit. For example, blocking 210d of FIG. 2 corresponds to quadtree partitioning.

[00208] The number of times of splitting without a quadtree indicates how many times a split that is not a split with a quadtree is performed to obtain the current quantization group from the largest coding unit. For example, the partitioning method depicted in Fig. 3 corresponds to a partition without a quadtree.

[00209] The block size information may include a block area or a binary logarithm (log) value of the block area. In addition, the block height and width, or the binary logarithm values of the block height and width, can be included here.

[00210] According to one embodiment, the quantization parameter determiner 1610 may determine the current quantization group according to the number of times of quadtree splitting. When only quadtree splitting is used to split the largest coding unit, a quantization group may be set on a block having at least a preset size according to the number of times of quadtree splitting. For example, when the size of the largest coding unit is 256×256 and the number of times of quadtree splitting is 2, a quantization group may be set for a block whose size is 64×64 or more.

[00211] FIGS. 17A-17D are diagrams of embodiments in which a quantization group is determined according to the number of times of a quadtree split.

[00212] Referring to FIG. 17A, the largest coding block 1700 is partitioned into four blocks 1702, 1704, 1706, and 1708 according to quadtree partitioning. The number of times of quadtree splitting of blocks 1702, 1704, 1706, and 1708 is set to 1. Block 1704 is split into four blocks 1710, 1712, 1714, and 1716 according to quadtree splitting. The number of times of quadtree splitting of blocks 1710, 1712, 1714 and 1716 is set to 2. Block 1716 is split into four blocks 1718, 1720, 1722 and 1724 according to quadtree splitting. The number of times of quadtree splitting of blocks 1718, 1720, 1722, and 1724 is set to 3. Based on blocks 1702, 1706, 1708, 1710, 1712, 1714, 1718, 1720, 1722, and 1724 determined when splitting of the largest coding block 1700 is completed, encoding and decoding with prediction and transformation may be performed.

[00213] As shown in FIG. 17A, when the number of times of splitting with a quadtree is increased by 1, the size of the split block is halved. Thus, only when quadtree splitting is enabled, the block size can be determined according to the number of times of quadtree splitting.

[00214] FIG. 17B depicts an embodiment in which a quantization group is determined with respect to a block for which the number of quadtree split times is 1. Referring to FIG. 17B, quantization groups are set with respect to four blocks 1702, 1704, 1706, and 1708 , for which the number of times of splitting with a quadtree is 1.

[00215] Each of blocks 1702, 1706, and 1708 is singly included in each of the quantization groups for blocks 1702, 1706, and 1708. However, the quantization group of block 1704 includes subblocks 1710, 1712, 1714, 1718, 1720, 1722, and 1724 of block 1704 Thus, quantization and inverse quantization according to the same quantization parameter may be applied to subblocks 1710, 1712, 1714, 1718, 1720, 1722, and 1724 of block 1704.

[00216] FIG. 17C shows an embodiment in which a quantization group is determined with respect to a block for which the number of quadtree split times is 2. With reference to FIG. 17C, quantization groups are set with respect to blocks 1702, 1706, 1708, 1710, 1712, 1714, and 1716, for which the number of quadtree split times is less than or equal to 2. For blocks 1702, 1706, and 1708, the quadtree split times is 1, but blocks 1702, 1706, and 1708 are not further split, so that the quantization group is set with respect to blocks 1702, 1706 and 1708.

[00217] Each of the blocks 1702, 1706, 1708, 1710, 1712, and 1714 is singly included in each of the quantization groups for it. However, the quantization group of block 1716 includes subblocks 1718, 1720, 1722, and 1724 of block 1716. Thus, quantization and inverse quantization according to the same quantization parameter can be applied to subblocks 1718, 1720, 1722, and 1724 of block 1716.

[00218] Fig. 17D depicts an embodiment in which a quantization group is defined with respect to a block for which the number of times of quadtree splitting is 3. Since the block for which the number of times of quadtree splitting is 4 is not present in Fig. 17D, the group quantization is set for all blocks 1702, 1706, 1708, 1710, 1712, 1714, 1718, 1720, 1722, and 1724.

[00219] With reference to FIGS. 17A to 17D, when the number of times of splitting with a quadtree of the block splitting information is increased, the size of the quantization group is reduced. Conversely, when the number of split times with a quadtree of the block split information decreases, the size of the quantization group increases. Thus, the size of the quantization parameter information can be increased or decreased based on the number of times of quadtree splitting from the block splitting information.

[00220] The quantization parameter determiner 1610 may determine the current quantization group according to the number of times of splitting with a quadtree and the number of times of splitting without a quadtree. When both of quadtree partitioning and non-quadtree partitioning are applied to block partitioning, the quantization group determination method shown in FIGS. 17A to 17D is not used. Thus, the quantization group can be determined further taking into account the number of times of splitting without a quadtree, or a method for determining the quantization group based on the size of the quantization group can be applied to this. 18A-18C show an embodiment of a method for determining a quantization group in a largest coding block to which non-quadtree partitioning is applied.

[00221] FIG. 18A shows how the largest coding block 1800 is split. The number marked in each block indicates the number of times the split is performed on the largest coding block 1800 .

[00222] The largest coding block 1800 is a quadtree divided into four blocks 1802, 1804, 1806, and 1808. Since the block 1802 is not further split, the number of times the block 1802 is split is 1. Further, it is assumed that the size of the largest coding block 1800 is 4Nx4N.

[00223] Block 1804 is split into two blocks 1810 and 1812 2NxN. Block 1810 is then split into two NxN blocks 1814 and 1816, and block 1812 is split into two N/2xN blocks 1818 and 1822 and one NxN block 1820. The number of times the subblocks 1814, 1816, 1818, 1820, 1822 of the block 1804 are split is 3 in total.

[00224] Block 1806 is split into two Nx2N blocks 1824 and 1826. Block 1824 is then split into two NxN blocks 1828 and 1830, and block 1826 is split into two NxN/2 blocks 1840 and 1844 and one NxN block 1842. Block 1828 is split into two blocks 1832 and 1834 N/2xN. Block 1834 is split into two blocks 1836 and 1838 N/2xN/2. The split times of subblocks 1828, 1830, 1840, 1842 and 1844 of block 1806 is 3. The split times of block 1832 split from block 1828 is 4, and the split times of blocks 1836 and 1838 is 5.

[00225] Block 1808 is split into four NxN blocks 1846, 1848, 1850, and 1852. Block 1846 is split into four blocks 1854, 1856, 1858 and 1860 NxN. Also, block 1848 is split into two Nx2N blocks 1862 and 1864, and block 1862 is split into two NxN blocks 1866 and 1868. The number of block split times 1850 and 1852 is 2, and the number of block split times 1854, 1856, 1858, 1860, and 1864 is 3. The number of block split times 1866 and 1868 is 4.

[00226] When the quantization group is determined according to the number of times the block is split, the size of the quantization group may not be uniform. In detail, with reference to FIG. 18B, the heterogeneity in quantization group size will now be described.

[00227] FIG. 18B shows an embodiment in which a quantization group is set on blocks whose split times is 3. Referring to FIG. 18B, a quantization group is set on blocks 1802, 1814, 1816, 1818, 1820, 1822 , 1828, 1830, 1840, 1842, 1844, 1850, 1852, 1854, 1856, 1858, 1860, 1862, and 1864, the number of partition times of which is 3.

[00228] However, the number of split times of block 1814 and the number of split times of block 1854 are equal, but the size of block 1814 is four times the size of block 1854. Although size 1836 is equal to size 1854, the quantization parameter of the quantization group corresponding to block 1828 is applied, while as the quantization parameter of the quantization group corresponding to block 1854 is applied to block 1854.

[00229] Only when quadtree partitioning is performed, as in the embodiments of FIGS. 17A-17D, are the quantization group sizes the same. However, as described above, in the case where the quantization groups are set according to the number of times of splitting when non-quadtree splitting is performed, the sizes of the quantization groups are different.

[00230] Fig.18C depicts methods for solving the problem. For example, the quantization parameter determiner 1610 may determine the current quantization group according to a weighted sum of the number of split times with a quadtree and the number of split times without a quadtree. The quadtree partitioning is the same, since vertical partitioning and horizontal partitioning are applied sequentially. Thus, one partition with a quadtree is essentially the same as two partitions without a quadtree.

[00231] Thus, the quantization parameter determining means 1610 subdivides the number of split times into the number of split times with a quadtree and the number of split times without a quadtree, and can set a quantization group based on the weighted sum of the number of split times with a quadtree and the number of split times without a quad tree according to weight 2 :1.

[00232] For example, block 1814 is generated due to one split with a quadtree and two splits without a quadtree from the largest coding element 1800. Thus, the weighted sum of the number of split times with a quadtree and the number of split times without a quadtree according to the 2:1 weight in relation to block 1814 is 4. Block 1846 is generated in view of two splits of the quadtree from the largest coding unit 1800. Thus, the weighted sum of the number of split times with a quadtree and the number of times of a split without a quadtree according to the 2:1 weight with respect to block 1846 is 4. Thus, when a quantization group is set with respect to a block whose weighted sum is 4, in contrast to FIG. 18B, in FIG. 18C, block 1854 obtains the quantization parameter from the quantization group set for block 1846.

[00233] According to another embodiment, the quantization parameter determiner 1610 may determine the current quantization group based on the sum of the block height and width, or the average of the block height and width. For example, when a quantization group is set with respect to an NxN block, quantization groups are set with respect to block 1814 and block 1846. Thus, unlike FIG. 18B, in FIG. 1846. Because the top-level blocks 1812 and 1826 are larger than NxN size and thus there is no quantization group corresponding to them, even when blocks 1818, 1822, 1840 and 1844 are smaller than NxN size, quantization groups are set for them.

[00234] Similarly, the quantization parameter determiner 1610 may determine the current quantization group based on the sum of the block height and width logarithm values, or the average of the block height and width logarithm values. Alternatively, the quantization parameter determiner 1610 may determine the current quantization group based on the block area or the area logarithm value.

[00235] The quantization parameter determiner 1610 may determine the predicted quantization parameter of the current block, based on the quantization parameter of the top adjacent block of the current quantization group, the quantization parameter of the left adjacent block of the current quantization group, and the quantization parameter of the quantization group that was decoded just before the current quantization group.

[00236] For example, the means 1610 for determining the quantization parameter may determine, as the quantization parameter of the current quantization group, the average value of the quantization parameter of the upper adjacent block and the quantization parameter of the left adjacent block. When the upper adjacent block quantization parameter does not exist, the quantization parameter determiner 1610 may use, instead of the upper adjacent block quantization parameter, the quantization parameter of the quantization group that was decoded immediately before the current quantization group to determine the quantization parameter of the current quantization group. Similarly, when the left adjacent block quantization parameter does not exist, the quantization parameter determiner 1610 may use, instead of the left adjacent block quantization parameter, the quantization parameter of the quantization group that was decoded immediately before the current quantization group to determine the quantization parameter of the current quantization group.

[00237] Also, the quantization parameter determiner 1610 may determine the slice or frame default quantization parameter as the predicted quantization parameter. For example, when the top adjacent block quantization parameter, the left adjacent block quantization parameter, and the quantization parameter of the quantization group that was decoded immediately before the current quantization group referenced by the current quantization group do not exist, the default quantization parameter can be used.

[00238] The quantization parameter determiner 1610 determines the difference quantization parameter of the current quantization group. The quantization parameter determining means 1610 can obtain, from the bitstream, absolute value information of the difference quantization parameter and sign information of the difference quantization parameter. The quantization parameter determining means 1610 may determine the difference quantization parameter of the current quantization group based on the absolute value information of the difference quantization parameter and the sign information of the difference quantization parameter.

[00239] When the current quantization group includes two or more blocks, the quantization parameter determiner 1610 can obtain absolute value information of the difference quantization parameter and sign information of the difference quantization parameter with respect to the block to be first decoded in the scan order. Then, the quantization parameter determining means 1610 does not obtain the difference quantization parameter absolute value information and the difference quantization parameter sign information with respect to the remaining blocks of the current quantization group, and applies the quantization parameter to the remaining blocks, the quantization parameter being determined with respect to the block to be decoded first. Thus, as a result, the quantization parameter determiner 1610 applies the same quantization parameter to all blocks of the current quantization group.

[00240] When the quantization parameter determiner 1610 decodes all blocks of the current quantization group and then decodes the block of the new quantization group, the quantization parameter determiner 1610 may trigger the difference quantization parameter and the difference quantization parameter related information. The difference quantization parameter related information may include difference quantization parameter decoding information indicating whether the difference quantization parameter has already been decoded, and quantization group position information indicating a quantization group position.

[00241] The quantization parameter determining means 1610 can trigger the difference quantization parameter and the difference quantization parameter related information, and can obtain new absolute value information of the difference quantization parameter and new sign information of the difference quantization parameter from the bitstream.

[00242] The quantization parameter determiner 1610 determines the quantization parameter of the current quantization group based on the predicted quantization parameter and the difference quantization parameter of the current quantization group. In detail, the quantization parameter determiner 1610 may determine the quantization parameter based on the sum of the predicted quantization parameter and the difference quantization parameter of the current quantization group. According to one embodiment, the quantization parameter determiner 1610 may obtain quantization parameter offset information from the bitstream, and may adjust the determined quantization parameter according to the quantization parameter offset information.

[00243] The inverse quantizer 1620 inversely quantizes the current block included in the current quantization group based on the quantization parameter of the current quantization group.

[00244] FIG. 19 shows a syntax structure with respect to a method for decoding a difference quantization parameter included in a bitstream when both quadtree splitting and non-quadtree splitting are enabled.

[00245] The table at the top of FIG. 19 shows the coding_quadtree splitting syntax structure. The quadtree partition syntax of FIG. 19 provides a configuration for determining whether to trigger a difference quantization parameter and difference quantization parameter related information before determining whether to perform quadtree partitioning.

[00246] In the quadtree split syntax structure of FIG. 19, "cu_qp_delta_enabled_flag" indicates a difference quantization parameter enabling flag, "cqtDepth" indicates the number of times of quadtree split, and "diff_cu_qp_delta_depth" indicates block split information. "CuQpDeltaVal" indicates a difference quantization parameter, "IsCuQpDeltaCoded" indicates difference quantization parameter decoding information, and "CuQgTopLeftX" and "CuQgTopLeftY" indicate quantization group position information.

[00247] Referring to FIG. 19, when "cu_qp_delta_enabled_flag" indicates 1 and "cqtDepth" is less than or equal to "diff_cu_qp_delta_depth", "CuQpDeltaVal" and "IsCuQpDeltaCoded" are defined as 0, and "CuQgTopLeftX" and "CuQgTopLeftY" are defined as x0 and y0, which indicate the position of the top left discrete sample of the current block.

[00248] When "cu_qp_delta_enabled_flag" indicates 1, it means that it is possible to obtain a difference quantization parameter.

[00249] When "cqtDepth" is less than or equal to "diff_cu_qp_delta_depth", it means that the number of split times with the quadtree of the current block is less than or equal to the number of split times, which is the quantization group reference indicated by the block split information. The indication that the number of split times with the quadtree of the current block is less than or equal to the number of split times, which is a quantization group reference, means that the current block is not included in the quantization group of the block decoded before the current block.

[00250] When the above conditions are satisfied, "CuQpDeltaVal" and "IsCuQpDeltaCoded" are determined to be 0, and a new difference quantization parameter with respect to the quantization group located in "CuQgTopLeftX" and "CuQgTopLeftY" is obtained based on the difference quantization parameter information that has just obtained from the bitstream.

[00251] The table in the middle of Fig. 19 depicts the syntax structure of a partition without a quadtree. The non-quadtree splitting syntax of FIG. 19 provides a configuration for determining whether to trigger a difference quantization parameter and difference quantization parameter related information before determining whether to perform non-quadtree splitting.

[00252] In the non-quadtree split syntax structure of FIG. 19, "cu_qp_delta_enabled_flag" indicates a difference quantization parameter enable flag, "cqtDepth" indicates the number of times of split with quadtree, "mttDepth" indicates the number of times of split without quadtree, and "diff_cu_qp_delta_depth" indicates information block splits. "CuQpDeltaVal" indicates a difference quantization parameter, "IsCuQpDeltaCoded" indicates decoding information of a difference quantization parameter, and "CuQgTopLeftX" and "CuQgTopLeftY" indicate quantization group position information.

[00253] Referring to FIG. 19, when "cu_qp_delta_enabled_flag" indicates 1 and the sum of "cqtDepth" and "mttDepth" is less than or equal to "diff_cu_qp_delta_depth", "CuQpDeltaVal" and "IsCuQpDeltaCoded" are defined as 0, and "CuQgTopLeftX" and "CuQgTopLeftY" " are defined as x0 and y0, which indicate the position of the top left discrete sample of the current block.

[00254] Similar to the splitting syntax structure with a quadtree, even in a splitting syntax structure without a quadtree, the difference quantization parameter and the difference quantization parameter related information are triggered. However, unlike the quadtree partitioning syntax, in the non-quadtree partitioning syntax, the sum of "cqtDepth" and "mttDepth", instead of "cqtDepth", is compared to "diff_cu_qp_delta_depth". 19, the sum of "cqtDepth" and "mttDepth" is compared to "diff_cu_qp_delta_depth", but in one embodiment, the weighted sum of "cqtDepth" and "mttDepth" may be compared to "diff_cu_qp_delta_depth".

[00255] The table at the bottom of Fig. 19 shows the syntax structure of the transform block. tu_cbf_luma[x0][y0] indicates whether the current luma block located at (x0, y0) has residual data. Then, tu_cbf_cb[x0][y0] and tu_cbf_cr[x0][y0] indicate whether the current block Cb and the current block Cr, respectively located at (x0, y0), have residual data. When the current luminance block, the current Cb block, and the current Cr block have no residual data, difference quantization parameter information is not obtained.

[00256] Conversely, when at least one of the current luma block, the current Cb block, and the current Cr block includes residual data, "cu_qp_delta_abs" indicating absolute value information of the difference quantization parameter, and "cu_qp_delta_sign_flag" indicating parameter sign information difference quantization are obtained from the bitstream. Then, "CuQpDeltaVal" indicating a difference quantization parameter is determined from "cu_qp_delta_abs" and "cu_qp_delta_sign_flag". In addition, "IsCuQpDeltaCoded" indicating whether a difference quantization parameter exists is defined as 1.

[00257] When a block to be decoded after the current block is included in the same quantization group of the current block (i.e., when "cqtDepth" or the (weighted) sum of "cqtDepth" and "mttDepth" is greater than "diff_cu_qp_delta_depth"), "CuQpDeltaVal" and "IsCuQpDeltaCoded" are not triggered, and thus a block to be decoded after the current block is inverse quantized according to the "CuQpDeltaVal" used in the decoding procedure for the current block.

[00258] In Fig. 19, the difference quantization parameter information acquisition configuration is carried out in the syntax structure of the transformation block, but according to one embodiment, the configuration may be carried out in another syntax.

[00259] FIG. 20 shows a method for decoding pictures, determining a block quantization parameter according to a quantization group, and decoding block residual data according to the determined quantization parameter.

[00260] In step 2010, a predicted quantization parameter of the current quantization group determined according to at least one of the block split information and the block size information is determined.

[00261] The current quantization group may be determined according to the number of split times with a quadtree and the number of split times without a quadtree. In detail, the current quantization group can be determined according to the weighted sum of the number of split times with a quadtree and the number of split times without a quadtree.

[00262] The current quantization group may be determined based on the sum of the block height and width, or the average of the block height and width. Alternatively, the current quantization group may be determined based on the sum of the blog values of the block height and width, or the average of the blog values of the block height and width. Alternatively, the current quantization group may be determined based on the block area or the area logarithm value.

[00263] The predicted quantization parameter of the current block may be determined based on the quantization parameter of the top adjacent block of the current quantization group, the quantization parameter of the left adjacent block of the current quantization group, and the quantization parameter of the quantization group that was decoded just before the current quantization group.

[00264] In operation 2020, the difference quantization parameter of the current quantization group is determined. In detail, absolute value information of the difference quantization parameter and sign information of the difference quantization parameter can be obtained from the bitstream. Then, the difference quantization parameter of the current quantization group can be determined based on the absolute value information of the difference quantization parameter and the sign information of the difference quantization parameter.

[00265] According to one embodiment, when the flag provided by the difference quantization parameter indicates that it is possible to determine the quantization parameter according to the difference quantization parameter, the difference quantization parameter of the current block can be obtained.

[00266] At operation 2030, the quantization parameter of the current quantization group is determined based on the predicted quantization parameter and the difference quantization parameter of the current quantization group. For example, the quantization parameter of the current quantization group may be determined based on the sum of the predicted quantization parameter and the difference quantization parameter.

[00267] In operation 2040, the current block included in the current quantization group is inversely quantized according to the quantization parameter of the current quantization group.

[00268] The image decoding method of FIG. 20 may include various embodiments with respect to a method for determining a quantization parameter according to the quantization group of the image decoding apparatus of FIG.

[00269] The image decoding apparatus 1600 may perform inverse quantization based on a quantization parameter element indicating an area where the same quantization parameter is used. Next, the inverse quantization method based on the quantization parameter element will be described.

[00270] FIG. 21 shows an embodiment of a quantization parameter element structure and a coding block tree structure.

[00271] A frame or slice may differ in the subjective degradation of image quality in its parts. Thus, in order to optimize the coding efficiency, it is necessary to set various quantization parameters according to the characteristics of the respective portions of a frame or slice. The distribution of quantization parameters is not equal to the tree structure of the coding unit, which is the basic coding unit. Thus, the quantization parameter element map is determined independently of the tree structure of the coding element.

[00272] In FIG. 21, quantization parameter element 2110 may be an MxN size rectangle. Here, a frame is represented as a quantization parameter map 2120 composed of a plurality of quantization parameter elements. Each of the quantization parameter elements in the quantization parameter map 2120 has a quantization parameter. 21, quantization parameter element 2110 is illustrated as a rectangle, but according to one embodiment, quantization parameter element 2110 may be illustrated as an irregular shape rather than a rectangle.

[00273] The quantization parameter of the quantization parameter element 2110 may be determined according to the characteristic of the part of the corresponding frame. The quantization parameter map 2120 and the quantization parameter of the quantization parameter element 2110 are encoded and decoded independently of the prediction encoding information according to the coding block structure 2140 . When the residual data of the coding unit 2130 is encoded and decoded, the quantization parameter can be obtained from the quantization parameter element 2110 that corresponds to the position of the coding element 2130.

[00274] The quantization parameter determiner 1610 may map the current block to the current quantization parameter element based on at least one of the position and size of the current block.

[00275] For example, the quantization parameter determiner 1610 may determine, as the current quantization parameter element of the current block, a quantization parameter element including the coordinate values of the upper left sample of the current block.

[00276] As another example, when the current block includes a plurality of quantization parameter elements, the quantization parameter determiner 1610 may determine the plurality of quantization parameter elements as the current quantization parameter elements of the current block. In this regard, the means 1610 for determining the quantization parameter may determine, as the quantization parameter of the current block, the average value of the plurality of quantization parameters of the current elements of the quantization parameter.

[00277] FIGS. 22A and 22B show a method for determining a quantization parameter element corresponding to the current block.

[00278] FIG. 22A depicts an embodiment in which a quantization parameter element 2200 corresponds to a plurality of coding elements 2202-2224. For blocks 2202, 2204, 2206, 2210, 2212, and 2214, all of which are included in the quantization parameter element 2200, inverse quantization is performed according to the quantization parameter corresponding to the quantization parameter element 2200.

[00279] Then, the blocks, parts of which are included in the quantization parameter element 2200, can be determined as to whether the quantization parameter of the quantization parameter element 2200 should be applied to it, based on the upper left sample of the block. Thus, for blocks 2208, 2216, 2218, 2220, 2222, and 2224 whose upper left discrete samples are included in quantization parameter element 2200, inverse quantization can be performed according to the quantization parameter corresponding to quantization parameter element 2200.

[00280] In FIG. 22A, an embodiment in which the quantization parameter element is determined based on the upper left sample of the block has been described, but according to one embodiment, the element of the block quantization parameter may be determined based on the center sample, the upper right sample , bottom left sample, bottom right sample, or similar for the block.

[00281] FIG. 22B depicts an embodiment in which a plurality of quantization parameter elements 2252-2274 corresponds to block 2250.

[00282] Quantization parameter elements 2252, 2254, 2258, 2260, 2264, and 2266 are fully included in block 2250. Thus, block 2250 can be inverse quantized according to at least one quantization parameter of elements 2252, 2254, 2258, 2260 , 2264 and 2266 quantization parameters. For example, the block quantization parameter 2250 may be defined as the average of the quantization parameters of quantization parameter elements 2252, 2254, 2258, 2260, 2264, and 2266.

[00283] Alternatively, quantization parameter elements 2256, 2262, 2268, 2270, 2272, and 2274, which partially overlap block 2250, can be used to determine the quantization parameter of block 2250. Thus, the reverse can be done for block 2250. quantization according to a quantization parameter determined according to at least one of the quantization parameter elements 2252-2274.

[00284] FIGS. 23A and 23B depict the correlation between a block and a quantization parameter element.

[00285] FIG. 23A shows a block tree structure and a quantization parameter map according to one embodiment. According to one embodiment, the quantization parameter element corresponding to the upper left discrete sample of the block corresponds to the block. Thus, block 2308 corresponds to quantization parameter element 2300, block 2310 corresponds to quantization parameter element 2302, block 2312 corresponds to quantization parameter element 2304, and block 2314 corresponds to quantization parameter element 2306. When the block and quantization parameter element match references are different, other quantization parameter elements 2302, 2304, and 2306 may correspond to block 2308.

[00286] FIG. 23B shows a block tree structure and a quantization parameter map according to one embodiment. As in FIG. 23A, when the quantization parameter element corresponding to the upper left sample of the block corresponds to a block, all of the blocks 2328, 2330, 2332, and 2334 correspond to the quantization parameter element 2334. Because the quantization parameter of quantization parameter element 2334 is applied to all of blocks 2328, 2330, 2332, and 2334, the quantization parameter is first calculated with respect to block 2328 having the earliest decoding order. The quantization parameter used for block 2328 can then be used unchanged for blocks 2330, 2332, and 2334.

[00287] The quantization parameter is not determined for a block that has no residual data. For example, when block 2328 has no residual data, inverse quantization is not necessary for block 2328, and thus the quantization parameter of block 2328 is not determined. When a block 2330 to be decoded after block 2328 has residual data, a quantization parameter of block 2330 may be determined. The quantization parameter used for block 2330 can then be used unchanged for blocks 2332 and 2334.

[00288] The quantization parameter determiner 1610 may obtain a predicted quantization parameter with respect to the current quantization parameter element.

[00289] The quantization parameter determiner 1610 can obtain a predicted quantization parameter from at least one of the left quantization parameter element of the current quantization parameter element, the top quantization parameter element of the current quantization parameter element, and the block decoded immediately before the current block.

[00290] Alternatively, the quantization parameter determiner 1610 may determine, as a predicted quantization parameter with respect to the current quantization parameter element, a predicted quantization parameter with respect to the frame or slice including the current quantization parameter element.

[00291] The quantization parameter determiner 1610 may obtain a difference quantization parameter with respect to the current quantization parameter element.

[00292] The quantization parameter determiner 1610 may determine the quantization parameter of the current quantization parameter element based on the predicted quantization parameter and the difference quantization parameter.

[00293] An inverse quantizer 1620 may inversely quantize the current block according to the quantization parameter of the current quantization parameter element.

[00294] FIG. 24 shows a method for decoding pictures, determining a block quantization parameter according to a quantization parameter element, and decoding block residual data according to the determined quantization parameter.

[00295] At operation 2410, the current block is mapped to the current quantization parameter element based on at least one of the position and size of the current block.

[00296] According to one embodiment, a quantization parameter element including the coordinate values of the upper left discrete sample of the current block may be determined as the current quantization parameter element of the current block.

[00297] According to one embodiment, when the current block includes a plurality of quantization parameter elements, the plurality of quantization parameter elements may be determined as the current quantization parameter element of the current block. In this regard, the current quantization parameter of the current block may be determined from at least one of the plurality of quantization parameter elements.

[00298] At operation 2420, a predicted quantization parameter is obtained with respect to the current quantization parameter element.

[00299] According to one embodiment, the predicted quantization parameter may be derived from at least one of the left quantization parameter element of the current quantization parameter element, the top quantization parameter element of the current quantization parameter element, and the block decoded just before the current block.

[00300] Alternatively, a predicted quantization parameter with respect to a frame or slice including the current quantization parameter element may be determined as a predicted quantization parameter with respect to the current quantization parameter element.

[00301] At operation 2430, a difference quantization parameter is obtained with respect to the current quantization parameter element.

[00302] At operation 2440, the quantization parameter of the current quantization parameter element is determined based on the predicted quantization parameter and the difference quantization parameter.

[00303] At operation 2450, inverse quantization is performed for the current block according to the quantization parameter of the current quantization parameter element.

[00304] The image decoding method of FIG. 24 may include various embodiments with respect to the method for determining a quantization parameter according to the quantization group of the image decoding apparatus of FIG.

[00305] The spatial image domain data may be encoded for each of the tree structure coding units by a tree structure coding unit based image coding technique, and may be recovered by decoding each of the largest coding units by a tree structure coding unit based image decoding technique as described above with reference to Figs. 1-24, so that a frame and an image, which is a sequence of frames, can be restored. The recovered video may be played by a playback device, may be stored in a storage medium, or may be transmitted over a network.

[00306] The above described embodiments of the disclosure may be implemented as a computer executable program and executed by a computer readable storage medium by a general purpose digital computer to execute the program.

[00307] While the disclosure has been described above in connection with specific best embodiments, other disclosures resulting from alternatives, modifications, and changes to the disclosure will be apparent to one of ordinary skill in the art in view of the above description. That is, the appended claims are to be understood as covering all such alternatives, modifications, and alterations. Thus, all provisions described in this technical specification and illustrated in the drawings are to be interpreted in an illustrative and non-limiting sense.

Statement 1. An image decoding method, comprising the steps of:

determining a predicted quantization parameter of the current quantization group determined according to at least one of the block split information and the block size information;

determining a difference quantization parameter of the current quantization group;

determining a quantization parameter of the current quantization group based on the predicted quantization parameter and the difference quantization parameter of the current quantization group; And

performing inverse quantization of the current block contained in the current quantization group according to the quantization parameter of the current quantization group.

Statement 2. The image decoding method according to Statement 1, wherein, in said determination of the predicted quantization parameter of the current quantization group, the predicted quantization parameter of the current block is determined based on the quantization parameter of the top adjacent block of the current quantization group, the quantization parameter of the left adjacent block of the current quantization group, and the group quantization parameter quantization group decoded just before the current quantization group.

Statement 3. The image decoding method according to Statement 1, wherein the current quantization group contains one or more blocks containing the current block, and the blocks contained in the current quantization group are inversely quantized according to the same quantization parameter.

Statement 4: The image decoding method according to Statement 1, wherein

the block split information contains the number of split times with a quadtree and the number of split times without a quadtree,

the current quantization group is determined according to the number of times of splitting with a quadtree and the number of times of splitting without a quadtree, and

the number of quadtree split times indicates the number of times that a quadtree split is performed to obtain the current quantization group from the largest coding block, and the non-quadtree split times indicates the number of times that a non-quadtree split is performed to obtain the current quantization group from the largest coding block.

Statement 5: The image decoding method according to Statement 4, wherein the current quantization group is determined according to the weighted sum of the number of split times with quadtree and the number of times of split without quadtree.

Statement 6: The image decoding method according to Statement 1, in which the block size information contains the block height and width, and the current quantization group is determined based on the sum of the block height and width or the average of the block height and width.

Statement 7: The image decoding method according to Statement 1, wherein the block size information contains log-binary (log) values of the block height and width, and the current quantization group is determined based on the sum of the log-binary values of the block height and width, or the average of the log-binary values of the height and block width.

Statement 8: The image decoding method according to Statement 1, wherein the block size information contains a block area or an area log value, and the current quantization group is determined based on the block area or area log value.

Statement 9. The image decoding method according to Statement 1, wherein said determination of a difference quantization parameter of a current quantization group comprises the steps of:

obtaining absolute value information of the difference quantization parameter and sign information of the difference quantization parameter from the bitstream; And

determining a difference quantization parameter of the current quantization group according to absolute value information of the difference quantization parameter and sign information of the difference quantization parameter.

Statement 10. The image decoding method according to Statement 1, further comprising the step of obtaining a flag provided by the difference quantization parameter with respect to the upper data element of the current quantization group, wherein said determination of the difference quantization parameter of the current quantization group comprises the step of, when provided by the parameter difference quantization flag indicates that it is possible to determine the quantization parameter according to the difference quantization parameter, the difference quantization parameter of the current block is obtained.

Statement 11. An image decoding device comprising a processor configured to

determine a predicted quantization parameter of the current quantization group determined according to at least one of the block split information and the block size information,

determine the difference quantization parameter of the current quantization group,

determine the quantization parameter of the current quantization group based on the predicted quantization parameter and the difference quantization parameter of the current quantization group, and

perform inverse quantization of the current block contained in the current quantization group according to the quantization parameter of the current quantization group.

Statement 12. An image decoding method, comprising the steps of:

matching the current block with the current element of the quantization parameter based on at least one of the position and size of the current block;

obtaining a predicted quantization parameter with respect to the current element of the quantization parameter;

getting a difference quantization parameter with respect to the current quantization parameter element;

determining a quantization parameter of the current quantization parameter element based on the predicted quantization parameter and the difference quantization parameter; And

performing inverse quantization of the current block according to the quantization parameter of the current element of the quantization parameter.

Statement 13. The image decoding method according to Statement 12, wherein all elements of a quantization parameter of a frame or a slice containing the current element of the quantization parameter are rectangles, each of which has a size of M×N.

Statement 14. The image decoding method according to Statement 12, wherein obtaining the predicted quantization parameter with respect to the current quantization parameter element comprises obtaining the predicted quantization parameter from at least one of the left quantization parameter element of the current quantization parameter element, the upper quantization parameter element the current quantization parameter element and the block decoded just before the current block.

Statement 15. The image decoding method according to Statement 12, wherein obtaining a predicted quantization parameter with respect to a current quantization parameter element comprises determining a predicted quantization parameter with respect to a frame or slice containing the current quantization parameter element as a predicted quantization parameter with respect to the current element of the quantization parameter.

Statement 16. The method for decoding images according to Statement 12, wherein said mapping of the current block to the current quantization parameter element based on at least one of the position and size of the current block comprises determining a quantization parameter element containing coordinate values of the upper left sample of the current block, as the current element of the quantization parameter of the current block.

Statement 17. A method for decoding pictures according to Statement 12, wherein

said mapping of the current block to the current quantization parameter element based on at least one of the position and size of the current block comprises the step of determining the plurality of quantization parameter elements as the current quantization parameter elements of the current block when the current block contains a plurality of quantization parameter elements ,

said obtaining a predicted quantization parameter with respect to the current quantization parameter element, obtaining a difference quantization parameter with respect to the current quantization parameter element, and determining the quantization parameter of the current quantization parameter element based on the predicted quantization parameter and the difference quantization parameter are performed with respect to each of the current quantization parameter elements, and

said inverse quantization of the current block according to the quantization parameter of the current quantization parameter element comprises the step of inverse quantization of the current block according to the average value of the quantization parameters of the current quantization parameter element.

Statement 18. An image decoding device, comprising a processor configured to:

map the current block to the current element of the quantization parameter based on at least one of the position and size of the current block,

get the predicted quantization parameter with respect to the current element of the quantization parameter,

get the difference quantization parameter with respect to the current quantization parameter element,

determine the quantization parameter of the current quantization parameter element based on the predicted quantization parameter and the difference quantization parameter, and

perform inverse quantization of the current block according to the quantization parameter of the current element of the quantization parameter.

Claims (26)

1. A method for decoding an image, comprising the steps of: splitting the upper layer coding block into a plurality of lower layer coding blocks according to the split information; determining a current quantization group based on a split amount of a current coding block among the plurality of lower layer coding units and a split amount of the current quantization group; obtaining a quantization parameter of the current quantization group based on the predicted quantization parameter and the difference quantization parameter; And perform inverse quantization of the transform coefficients in the current transform block in the current coding block using this quantization parameter, wherein the predicted quantization parameter is obtained using at least one of the quantization parameter of the top adjacent block of the current quantization group and the quantization parameter of the left adjacent block of the current quantization group, wherein the split value of the current coding block is incremented by 1 if the current coding block is obtained by splitting the upper layer coding block into two lower layer coding blocks according to non-quadtree partitioning, and the splitting value of the current coding block is increased by 2 if the current coding block is obtained by splitting the upper layer coding block into four lower layer coding blocks according to quadtree partitioning. 2. An image decoding device comprising at least one processor configured to: split the upper layer coding block into a plurality of lower layer coding blocks according to the split information; determine the current quantization group based on the split amount of the current coding block among the plurality of lower layer coding units and the split amount of the current quantization group; obtain a quantization parameter of the current quantization group based on the predicted quantization parameter and the difference quantization parameter; And perform inverse quantization of the transform coefficients in the current transform block in the current coding block using this quantization parameter, wherein the predicted quantization parameter is obtained using at least one of the quantization parameter of the top adjacent block of the current quantization group and the quantization parameter of the left adjacent block of the current quantization group, wherein the split value of the current coding block is increased by 1 if the current coding block is obtained by splitting the upper layer coding block into two lower layer coding blocks according to non-quadtree partitioning, and the split value of the current coding block is increased by 2 if the current coding block is obtained by splitting the upper layer coding block into four lower layer coding blocks according to quadtree partitioning. 3. A method for encoding an image, comprising the steps of: splitting the upper layer coding block into a plurality of lower layer coding blocks; determining a current quantization group based on a split amount of a current coding block among the plurality of lower layer coding units and a split amount of the current quantization group; getting a quantization parameter of the current quantization group; And performing quantization of the coefficients in the current transform block in the current coding block using this quantization parameter, moreover, based on the quantization parameter of the current quantization group and the predicted quantization parameter, the difference quantization parameter is obtained, wherein the predicted quantization parameter is obtained using at least one of the quantization parameter of the top adjacent block of the current quantization group and the quantization parameter of the left adjacent block of the current quantization group, and the split value of the current coding block is incremented by 1 if the current coding block is obtained by splitting the upper layer coding block into two lower layer coding blocks according to non-quadtree partitioning, and the split value of the current coding block is increased by 2 if the current coding block is obtained by splitting the upper layer coding block into four lower layer coding blocks according to quadtree partitioning.

Images