WO2018174457A1

WO2018174457A1 - Image processing method and device therefor

Info

Publication number: WO2018174457A1
Application number: PCT/KR2018/002929
Authority: WO
Inventors: 최장원; 허진
Original assignee: 엘지전자(주)
Priority date: 2017-03-22
Filing date: 2018-03-13
Publication date: 2018-09-27

Abstract

An image processing method and a device therefor are disclosed. Particularly, the image processing method comprises: a step of determining whether or not it is possible to apply an adaptive transform kernel to a chroma block; a step of acquiring transform information of the chroma block if it is possible to apply the adaptive transform kernel thereto; and a step of inverse-transforming the chroma block by using the acquired transform information.

Description

Image processing method and apparatus therefor

The present invention relates to a still image or video processing method, and more particularly, to a method for applying an AMT (Adaptive Multiple Core Transform) in the encoding / decoding process of a chrominance image and an apparatus for supporting the same.

Compression coding refers to a series of signal processing techniques for transmitting digitized information through a communication line or for storing in a form suitable for a storage medium. Media such as an image, an image, an audio, and the like may be a target of compression encoding. In particular, a technique of performing compression encoding on an image is called video image compression.

Next-generation video content will be characterized by high spatial resolution, high frame rate and high dimensionality of scene representation. Processing such content would result in a tremendous increase in terms of memory storage, memory access rate, and processing power.

Accordingly, there is a need to design coding tools for more efficiently processing next generation video content.

An object of the present invention is to provide a method and apparatus for efficiently encoding / decoding a still image or a moving image.

Another object of the present invention is to provide a method and apparatus for efficiently transforming a residual signal of a color difference image.

Another object of the present invention is to provide a method and apparatus for applying an adaptive multiple core transform (AMT) to a chroma image.

Another object of the present invention is to provide a method and apparatus for selecting an AMT transform set for a color difference image.

Another object of the present invention is to provide a method and apparatus for setting an AMT transform set for a color difference image.

The technical problems to be achieved in the present invention are not limited to the technical problems mentioned above, and other technical problems not mentioned above will be clearly understood by those skilled in the art from the following description. Could be.

An aspect of the present invention provides a method of decoding a chrominance block of an image, the method comprising: determining whether an adaptive transform kernel is applicable to the chrominance block; Acquiring transform information of the color difference block when the adaptive transform kernel is applicable to the color difference block; And inversely transforming the chrominance block using the obtained transform information.

Preferably, in the adaptive transform kernel, a combination of a first transform applied in a horizontal direction of the chrominance block and a second transform applied in a vertical direction is determined according to the prediction mode of the chrominance block.

Preferably, in the acquiring of the transformation information, the transformation information includes first transformation information indicating whether the adaptive transformation kernel is applied to a luminance block associated with the chrominance block, When the adaptive transform kernel is applied, the apparatus further includes second transform information indicating a combination of a transform applied in the horizontal direction and a transform applied in the vertical direction to the luminance block.

Preferably, when the first transform information indicates that the adaptive transform kernel is applied to the luminance block, inversely transforming the color difference block, using the combination of transforms indicated by the second transform information, the color difference block Inverse transformation of is performed.

Preferably, in the determining whether the adaptive transform kernel is applicable, when the prediction mode of the chrominance block is an intra prediction mode, the size of the chrominance block, or the intra prediction mode of the chrominance block is a specific intra prediction mode. Based on the recognition it is determined whether the adaptive transform kernel is applicable.

Preferably, if the size of the color difference block is 4x4 or more or 64x64 or less, the adaptive transform kernel can be applied to the color difference block.

Preferably, the specific intra prediction mode of the chrominance block uses a first mode for predicting the chrominance pixel value of the chrominance block by using the reconstructed luminance pixel value, or the prediction direction of the luminance block at a position corresponding to the chrominance block. The second mode.

Preferably, in the determining whether the adaptive transform kernel is applicable, when the prediction mode of the color difference block is an inter prediction mode, the size of the color difference block, the prediction block of the color difference block and the prediction block of the luminance block Based on the degree of correlation or the degree of correlation between the predicted block of the chrominance block and the reconstructed block of the luminance block, it is determined whether the adaptive transform kernel is applicable.

Preferably, in the acquiring of the conversion information, when the color difference block and the luminance block have the same shape, the conversion information is obtained from a luminance block existing at a position corresponding to the color difference block. If the shape of the luminance block is different, selecting one of a plurality of luminance blocks as the luminance block; And acquiring the conversion information from the selected one luminance block.

Preferably, in the selecting of one of the plurality of luminance blocks as the luminance block, a luminance block existing at a position corresponding to a specific position of the chrominance block is selected, or an area of an area overlapping with the chrominance block is selected. The largest luminance block is selected, or the luminance block having the largest size is selected among the luminance blocks where the color difference block and the region overlap.

Preferably, in the obtaining of the transform information, the transform information is transmitted from an encoder, and the transform information includes first transform information indicating whether the adaptive transform kernel has been applied to the chrominance block. And when the adaptive transform kernel is applied to the chrominance block, the transform information further includes second transform information indicating a combination of a transform applied in the horizontal direction and a transform applied in the vertical direction to the chrominance block.

Preferably, in the acquiring of the transformation information, if the prediction mode of the chrominance block is a specific intra prediction mode, transformation information of a luminance block associated with the chrominance block is obtained as transformation information of the chrominance block, otherwise, The conversion information is transmitted from the encoder.

An aspect of the present invention provides a device for decoding a color difference block of an image, wherein the device includes an inverse transform unit for inversely transforming inverse quantized transform coefficients of the color difference block, wherein the inverse transform unit is adapted to the color difference block. An adaptive transform kernel application determining unit determining whether or not an adaptive transform kernel is applicable; A transformation information obtaining unit obtaining transformation information of the chrominance block when the adaptive transformation kernel is applicable to the chrominance block; And a color difference block inverse transform unit which inversely transforms the color difference block by using the obtained conversion information.

According to an embodiment of the present invention, data compression efficiency may be improved by applying an adaptive multiple core transform (AMT) in a process of encoding / decoding a color difference image.

In addition, according to an embodiment of the present invention, the number of bits signaled may be reduced by using the AMT information of the luminance block as the AMT information of the chrominance block.

In addition, according to an embodiment of the present invention, whether or not AMT is applicable to the color difference block may be first determined, and a luminance block for obtaining AMT information may be selected through various methods.

In addition, according to an embodiment of the present invention, the conversion efficiency may be improved by using an AMT transform set separately configured according to the characteristics of the chrominance block.

Further, according to an embodiment of the present invention, the number of bits signaled may be reduced by reducing the number of sub transform sets applicable in each prediction mode.

In addition, according to an embodiment of the present invention, data compression efficiency may be further improved by determining whether AMT is applied to each of the Cb block and the Cr block.

The effects obtainable in the present invention are not limited to the above-mentioned effects, and other effects not mentioned will be clearly understood by those skilled in the art from the following description. .

1 is a schematic block diagram of an encoder in which encoding of a still image or video signal is performed according to an embodiment to which the present invention is applied.

2 is a schematic block diagram of a decoder in which encoding of a still image or video signal is performed according to an embodiment to which the present invention is applied.

3 is a diagram for describing a partition structure of a coding unit that may be applied to the present invention.

4 is a diagram for explaining a prediction unit applicable to the present invention.

5 illustrates a prediction direction of an intra prediction mode according to an embodiment of the present invention.

FIG. 6 is a diagram for describing a QTBT (QuadTree BinaryTree) block division structure according to an embodiment of the present invention.

FIG. 7 is a diagram illustrating a block division structure of a QTBT for a luma component and a chroma component according to an embodiment of the present invention.

FIG. 8 illustrates selecting one of a plurality of luminance blocks when the block division structure of the luminance block and the chrominance block is different according to an embodiment of the present invention.

9 is a diagram for describing a linear model (LM) mode according to an embodiment of the present invention.

10 is a block diagram of an inverse transform unit, according to an exemplary embodiment.

11 is a flowchart of a method of decoding a color difference block, according to an embodiment of the present invention.

Hereinafter, exemplary embodiments of the present invention will be described in detail with reference to the accompanying drawings. The detailed description, which will be given below with reference to the accompanying drawings, is intended to explain exemplary embodiments of the present invention and is not intended to represent the only embodiments in which the present invention may be practiced. The following detailed description includes specific details in order to provide a thorough understanding of the present invention. However, one of ordinary skill in the art appreciates that the present invention may be practiced without these specific details.

In some instances, well-known structures and devices may be omitted or shown in block diagram form centering on the core functions of the structures and devices in order to avoid obscuring the concepts of the present invention.

In addition, the terminology used in the present invention was selected as a general term widely used as possible now, in a specific case will be described using terms arbitrarily selected by the applicant. In such a case, since the meaning is clearly described in the detailed description of the part, it should not be interpreted simply by the name of the term used in the description of the present invention, and it should be understood that the meaning of the term should be understood and interpreted. .

Specific terms used in the following description are provided to help the understanding of the present invention, and the use of such specific terms may be changed to other forms without departing from the technical spirit of the present invention. For example, signals, data, samples, pictures, frames, blocks, etc. may be appropriately replaced and interpreted in each coding process.

Hereinafter, the term 'block' or 'unit' refers to a unit in which a process of encoding / decoding such as prediction, transformation, and / or quantization is performed, and may be configured as a multidimensional array of samples (or pixels, pixels).

'Block' or 'unit' may mean a multi-dimensional array of samples for luma components, or may mean a multi-dimensional array of samples for chroma components. In addition, the multi-dimensional arrangement of the sample for the luma component and the multi-dimensional arrangement of the sample for the chroma component may also be included.

For example, 'block' or 'unit' refers to a coding block (CB) that represents an array of samples to be encoded / decoded, and a coding tree block composed of a plurality of coding blocks (CTB). Block), Prediction Block (PB) (or Prediction Unit (PU)), which means an array of samples to which the same prediction is applied, and Transform Block (TB :), which is an array of samples to which the same transformation is applied. It may be interpreted as meaning including all of a transform block (or a transform unit (TU)).

Also, unless stated otherwise in this specification, a 'block' or 'unit' is a syntax structure used in encoding / decoding an array of samples for a luma component and / or a chroma component. can be interpreted to include a sturcture. Here, the syntax structure refers to zero or more syntax elements existing in the bitstream in a specific order, and the syntax element refers to an element of data represented in the bitstream.

For example, a 'block' or 'unit' includes a coding unit (CU) including a coding block (CB) and a syntax structure used for encoding the coding block (CB), and a plurality of coding units. A prediction unit (PU), a transform block (TB) including a coding tree unit (CU), a prediction block (PB), and a syntax structure used for prediction of the prediction block (PB); It may be interpreted as meaning including all transform units (TUs) including a syntax structure used for transformation of the corresponding transform block TB.

In addition, in the present specification, the 'block' or 'unit' is not necessarily limited to an array of square or rectangular samples (or pixels or pixels), and polygonal samples having three or more vertices (or pixels or pixels). It can also mean an array of. In this case, it may also be referred to as a polygon block or a polygon unit.

Referring to FIG. 1, the encoder 100 may include an image divider 110, a subtractor 115, a transform unit 120, a quantizer 130, an inverse quantizer 140, an inverse transform unit 150, and a filtering unit. 160, a decoded picture buffer (DPB) 170, a predictor 180, and an entropy encoder 190. The predictor 180 may include an inter predictor 181 and an intra predictor 182.

The image divider 110 divides an input video signal (or a picture or a frame) input to the encoder 100 into one or more processing units.

The subtractor 115 subtracts the difference from the prediction signal (or prediction block) output from the prediction unit 180 (that is, the inter prediction unit 181 or the intra prediction unit 182) in the input image signal. Generate a residual signal (or difference block). The generated difference signal (or difference block) is transmitted to the converter 120.

The transform unit 120 may convert a differential signal (or a differential block) into a transform scheme (eg, a discrete cosine transform (DCT), a discrete sine transform (DST), a graph-based transform (GBT), and a karhunen-loeve transform (KLT)). Etc.) to generate transform coefficients. In this case, the transform unit 120 may generate transform coefficients by performing a transform using a transform mode determined according to the prediction mode applied to the difference block and the size of the difference block.

The quantization unit 130 quantizes the transform coefficients and transmits the transform coefficients to the entropy encoding unit 190, and the entropy encoding unit 190 entropy codes the quantized signals and outputs them as bit streams.

Meanwhile, the quantized signal output from the quantization unit 130 may be used to generate a prediction signal. For example, the quantized signal may recover the differential signal by applying inverse quantization and inverse transformation through an inverse quantization unit 140 and an inverse transformation unit 150 in a loop. A reconstructed signal may be generated by adding the reconstructed difference signal to a prediction signal output from the inter predictor 181 or the intra predictor 182.

Meanwhile, in the compression process as described above, adjacent blocks are quantized by different quantization parameters, thereby causing deterioration of the block boundary. This phenomenon is called blocking artifacts, which is one of the important factors in evaluating image quality. In order to reduce such deterioration, a filtering process may be performed. Through this filtering process, the image quality can be improved by removing the blocking degradation and reducing the error of the current picture.

The filtering unit 160 applies filtering to the reconstruction signal and outputs it to the reproduction apparatus or transmits the decoded picture buffer to the decoding picture buffer 170. The filtered signal transmitted to the decoded picture buffer 170 may be used as the reference picture in the inter prediction unit 181. As such, by using the filtered picture as a reference picture in the inter prediction mode, not only image quality but also encoding efficiency may be improved.

The decoded picture buffer 170 may store the filtered picture for use as a reference picture in the inter prediction unit 181.

The inter prediction unit 181 performs temporal prediction and / or spatial prediction to remove temporal redundancy and / or spatial redundancy with reference to a reconstructed picture. Here, since the reference picture used to perform the prediction is a transformed signal that has been quantized and dequantized in units of blocks at the time of encoding / decoding, a blocking artifact or a ringing artifact may exist. have.

Accordingly, the inter prediction unit 181 may interpolate the signals between pixels in sub-pixel units by applying a lowpass filter to solve performance degradation due to discontinuity or quantization of such signals. Herein, the subpixel refers to a virtual pixel generated by applying an interpolation filter, and the integer pixel refers to an actual pixel existing in the reconstructed picture. As the interpolation method, linear interpolation, bi-linear interpolation, wiener filter, or the like may be applied.

The interpolation filter may be applied to a reconstructed picture to improve the precision of prediction. For example, the inter prediction unit 181 generates an interpolation pixel by applying an interpolation filter to integer pixels, and uses an interpolated block composed of interpolated pixels as a prediction block. You can make predictions.

The intra predictor 182 predicts the current block by referring to samples in the vicinity of the block to which the current encoding is to be performed. The intra prediction unit 182 may perform the following process to perform intra prediction. First, reference samples necessary for generating a prediction signal may be prepared. The prediction signal may be generated using the prepared reference sample. In addition, the prediction mode is encoded. In this case, the reference sample may be prepared through reference sample padding and / or reference sample filtering. Since the reference sample has been predicted and reconstructed, there may be a quantization error. Accordingly, the reference sample filtering process may be performed for each prediction mode used for intra prediction to reduce such an error.

The prediction signal (or prediction block) generated by the inter prediction unit 181 or the intra prediction unit 182 is used to generate a reconstruction signal (or reconstruction block) or a differential signal (or differential block). It can be used to generate.

2, the decoder 200 includes an entropy decoding unit 210, an inverse quantization unit 220, an inverse transform unit 230, an adder 235, a filtering unit 240, and a decoded picture buffer (DPB). Buffer Unit (250), the prediction unit 260 may be configured. The predictor 260 may include an inter predictor 261 and an intra predictor 262.

The reconstructed video signal output through the decoder 200 may be reproduced through the reproducing apparatus.

The decoder 200 receives a signal (ie, a bit stream) output from the encoder 100 of FIG. 1, and the received signal is entropy decoded through the entropy decoding unit 210.

The inverse quantization unit 220 obtains a transform coefficient from the entropy decoded signal using the quantization step size information.

The inverse transform unit 230 applies an inverse transform scheme to inverse transform the transform coefficients to obtain a residual signal (or a differential block).

The adder 235 outputs the obtained difference signal (or difference block) from the prediction unit 260 (that is, the prediction signal (or prediction block) output from the inter prediction unit 261 or the intra prediction unit 262. ) Generates a reconstructed signal (or a reconstruction block).

The filtering unit 240 applies filtering to the reconstructed signal (or the reconstructed block) and outputs the filtering to the reproduction device or transmits the decoded picture buffer unit 250 to the reproduction device. The filtered signal transmitted to the decoded picture buffer unit 250 may be used as a reference picture in the inter predictor 261.

In the present specification, the embodiments described by the filtering unit 160, the inter prediction unit 181, and the intra prediction unit 182 of the encoder 100 are respectively the filtering unit 240, the inter prediction unit 261, and the decoder of the decoder. The same may be applied to the intra predictor 262.

In general, a still image or video compression technique (eg, HEVC) uses a block-based image compression method. The block-based image compression method is a method of processing an image by dividing the image into specific block units, and may reduce memory usage and calculation amount.

The encoder splits one image (or picture) into units of a coding tree unit (CTU) in a rectangular shape. In addition, one CTU is sequentially encoded according to a raster scan order.

In HEVC, the size of the CTU can be set to any of 64Х64, 32Х32, and 16Х16. The encoder may select and use the size of the CTU according to the resolution of the input video or the characteristics of the input video. The CTU includes a coding tree block (CTB) for luma components and a CTB for two chroma components corresponding thereto.

One CTU may be divided into a quad-tree structure. That is, one CTU has a square shape and is divided into four units having a half horizontal size and a half vertical size to generate a coding unit (CU). have. This partitioning of the quad-tree structure can be performed recursively. That is, a CU is hierarchically divided into quad-tree structures from one CTU.

CU refers to a basic unit of coding in which an input image is processed, for example, intra / inter prediction is performed. The CU includes a coding block (CB) for a luma component and a CB for two chroma components corresponding thereto. In HEVC, the size of a CU may be one of 64Х64, 32Х32, 16Х16, and 8Х8.

Referring to FIG. 3, the root node of the quad-tree is associated with the CTU. The quad-tree is split until it reaches a leaf node, which corresponds to a CU.

More specifically, the CTU corresponds to a root node and has a smallest depth (ie, depth = 0). The CTU may not be divided according to the characteristics of the input image. In this case, the CTU corresponds to a CU.

The CTU may be divided into quad tree shapes, resulting in lower nodes having a depth of 1 (depth = 1). In addition, a node that is no longer divided (ie, a leaf node) in a lower node having a depth of 1 corresponds to a CU. For example, in FIG. 3 (b), CU (a), CU (b), and CU (j) corresponding to nodes a, b, and j are divided once in the CTU and have a depth of one.

At least one of the nodes having a depth of 1 may be divided into quad tree shapes, and as a result, sub nodes having a depth of 1 (ie, depth = 2) are generated. In addition, a node (ie, a leaf node) that is no longer divided in a lower node having a depth of 2 corresponds to a CU. For example, in FIG. 3 (b), CU (c), CU (h) and CU (i) corresponding to nodes c, h and i are divided twice in the CTU and have a depth of two.

In addition, at least one of the nodes having a depth of 2 may be divided into quad tree shapes, resulting in lower nodes having a depth of 3 (ie, depth = 3). And, a node that is no longer partitioned (ie, a leaf node) in a lower node having a depth of 3 corresponds to a CU. For example, in FIG. 3 (b), CU (d), CU (e), CU (f), and CU (g) corresponding to nodes d, e, f, and g are divided three times in the CTU, Has depth.

In the encoder, the maximum size or the minimum size of the CU may be determined according to characteristics (eg, resolution) of the video image or in consideration of encoding efficiency. Information about this or information capable of deriving the information may be included in the bitstream. A CU having a maximum size may be referred to as a largest coding unit (LCU), and a CU having a minimum size may be referred to as a smallest coding unit (SCU).

In addition, a CU having a tree structure may be hierarchically divided with predetermined maximum depth information (or maximum level information). Each partitioned CU may have depth information. Since the depth information indicates the number and / or degree of division of the CU, the depth information may include information about the size of the CU.

Since the LCU is divided into quad tree shapes, the size of the SCU can be obtained by using the size and maximum depth information of the LCU. Or conversely, using the size of the SCU and the maximum depth information of the tree, the size of the LCU can be obtained.

For one CU, information indicating whether the corresponding CU is split (for example, a split CU flag split_cu_flag) may be transmitted to the decoder. This partitioning information is included in all CUs except the SCU. For example, if the flag indicating whether to split or not is '1', the CU is divided into 4 CUs again. If the flag indicating whether to split or not is '0', the CU is not divided further. Processing may be performed.

As described above, a CU is a basic unit of coding in which intra prediction or inter prediction is performed. HEVC divides a CU into prediction units (PUs) in order to code an input image more effectively.

The PU is a basic unit for generating a prediction block, and may generate different prediction blocks in PU units within one CU. However, PUs belonging to one CU are not mixed with intra prediction and inter prediction, and PUs belonging to one CU are coded by the same prediction method (ie, intra prediction or inter prediction).

The PU is not divided into quad-tree structures, but is divided once in a predetermined form in one CU. This will be described with reference to the drawings below.

The PU is divided differently according to whether an intra prediction mode or an inter prediction mode is used as a coding mode of a CU to which the PU belongs.

FIG. 4A illustrates a PU when an intra prediction mode is used, and FIG. 4B illustrates a PU when an inter prediction mode is used.

Referring to FIG. 4 (a), assuming that a size of one CU is 2NХ2N (N = 4,8,16,32), one CU is divided into two types (that is, 2NХ2N or NХN). Can be.

Here, when divided into 2NХ2N type PU, it means that only one PU exists in one CU.

On the other hand, when divided into NХN type PU, one CU is divided into four PUs, and different prediction blocks are generated for each PU unit. However, the division of the PU may be performed only when the size of the CB for the luminance component of the CU is the minimum size (that is, the CU is the SCU).

Referring to FIG. 4 (b), assuming that a size of one CU is 2NХ2N (N = 4,8,16,32), one CU has 8 PU types (ie, 2NХ2N, NХN, 2NХN). , NХ2N, nLХ2N, nRХ2N, 2NХnU, 2NХnD).

Similar to intra prediction, PU splitting in the form of NХN may be performed only when the size of the CB for the luminance component of the CU is the minimum size (that is, the CU is the SCU).

In inter prediction, 2NХN type partitioning in the horizontal direction and NХ2N type PU partitioning in the vertical direction are supported.

In addition, it supports PU partitions of nLХ2N, nRХ2N, 2NХnU, and 2NХnD types, which are Asymmetric Motion Partition (AMP). Here, 'n' means a 1/4 value of 2N. However, AMP cannot be used when the CU to which the PU belongs is a CU of the minimum size.

In order to efficiently encode an input image within one CTU, an optimal partitioning structure of a coding unit (CU), a prediction unit (PU), and a transformation unit (TU) is subjected to the following process to perform a minimum rate-distortion. It can be determined based on the value. For example, looking at the optimal CU partitioning process in the 64Х64 CTU, the rate-distortion cost can be calculated while the partitioning process from the 64Х64 size CU to the 8Х8 size CU. The specific process is as follows.

1) The partition structure of the optimal PU and TU that generates the minimum rate-distortion value is determined by performing inter / intra prediction, transform / quantization, inverse quantization / inverse transform, and entropy encoding on a 64Х64 CU.

2) Divide a 64Х64 CU into four 32Х32 CUs and determine the optimal PU and TU partitioning structure that generates the minimum rate-distortion value for each 32Х32 CU.

3) Subdivide the 32Х32 CU into four CUs of 16Х16 size and determine the optimal PU and TU partitioning structure that generates the minimum rate-distortion value for each 16Х16 CU.

4) The 16Х16 CU is subdivided into four 8Х8 CUs, and the optimal PU and TU partitioning structure is generated to generate the minimum rate-distortion value for each 8Х8 CU.

5) Compare the sum of the rate-distortion values of the 16Х16 CUs calculated in step 3) with the rate-distortion values of the four 8Х8 CUs calculated in step 4) above to find the optimal CU in the 16Х16 block. Determine the partition structure. Do the same for the remaining three 16Х16 CUs.

6) Compare the sum of the rate-distortion values of the 32Х32 CUs calculated in 2) above with the rate-distortion values of the four 16Х16 CUs obtained in the above 5). Determine the partition structure. Do this for the remaining three 32Х32 CUs.

7) Finally, compare the sum of the rate-distortion values of the 64Х64 CUs computed in 1) above with the rate-distortion values of the four 32Х32 CUs obtained in 6) above. Determine the partition structure of the CU.

In the intra prediction mode, a prediction mode is selected in units of PUs, and prediction and reconstruction are performed in units of actual TUs for the selected prediction mode.

TU means a basic unit in which actual prediction and reconstruction are performed. The TU includes a transform block (TB) for a luma component and a TB for two chroma components corresponding thereto.

In the example of FIG. 3, as one CTU is divided into quad-tree structures to generate CUs, the TUs are hierarchically divided into quad-tree structures from one CU to be coded.

Since the TU is divided into quad-tree structures, the TU divided from the CU can be divided into smaller lower TUs. In HEVC, the size of a TU can be set to any of 32Х32, 16Х16, 8Х8, and 4Х4.

Referring again to FIG. 3, it is assumed that a root node of the quad-tree is associated with a CU. The quad-tree is split until it reaches a leaf node, which corresponds to a TU.

In more detail, a CU corresponds to a root node and has a smallest depth (that is, depth = 0). The CU may not be divided according to the characteristics of the input image. In this case, the CU corresponds to a TU.

The CU may be divided into quad tree shapes, resulting in lower nodes having a depth of 1 (depth = 1). In addition, a node (ie, a leaf node) that is no longer divided in a lower node having a depth of 1 corresponds to a TU. For example, in FIG. 3B, TU (a), TU (b), and TU (j) corresponding to nodes a, b, and j are divided once in a CU and have a depth of 1. FIG.

At least one of the nodes having a depth of 1 may be divided into quad tree shapes, and as a result, sub nodes having a depth of 1 (ie, depth = 2) are generated. In addition, a node (ie, a leaf node) that is no longer divided in a lower node having a depth of 2 corresponds to a TU. For example, in FIG. 3B, TU (c), TU (h), and TU (i) corresponding to nodes c, h, and i are divided twice in a CU and have a depth of two.

In addition, at least one of the nodes having a depth of 2 may be divided into quad tree shapes, resulting in lower nodes having a depth of 3 (ie, depth = 3). And, a node that is no longer partitioned (ie, a leaf node) in a lower node having a depth of 3 corresponds to a CU. For example, in FIG. 3 (b), TU (d), TU (e), TU (f), and TU (g) corresponding to nodes d, e, f, and g are divided three times in a CU. Has depth.

A TU having a tree structure may be hierarchically divided with predetermined maximum depth information (or maximum level information). Each divided TU may have depth information. Since the depth information indicates the number and / or degree of division of the TU, it may include information about the size of the TU.

For one TU, information indicating whether the corresponding TU is split (for example, split TU flag split_transform_flag) may be delivered to the decoder. This partitioning information is included in all TUs except the smallest TU. For example, if the value of the flag indicating whether to split is '1', the corresponding TU is divided into four TUs again. If the value of the flag indicating whether to split is '0', the corresponding TU is no longer divided.

예측(prediction)Prediction

The decoded portion of the current picture or other pictures in which the current processing unit is included may be used to reconstruct the current processing unit in which decoding is performed.

Intra picture or I picture (slice), which uses only the current picture for reconstruction, i.e. performs only intra picture prediction, predicts a picture (slice) using at most one motion vector and reference index to predict each unit A picture using a predictive picture or P picture (slice), up to two motion vectors, and a reference index (slice) may be referred to as a bi-predictive picture or a B picture (slice).

Intra prediction means a prediction method that derives the current processing block from data elements (eg, sample values, etc.) of the same decoded picture (or slice). That is, a method of predicting pixel values of the current processing block by referring to reconstructed regions in the current picture.

Inter prediction means a prediction method of deriving a current processing block based on data elements (eg, sample values or motion vectors, etc.) of pictures other than the current picture. That is, a method of predicting pixel values of the current processing block by referring to reconstructed regions in other reconstructed pictures other than the current picture.

인트라Intra 예측( prediction( IntraIntra prediction)(또는 화면 내 예측) prediction (or in-screen prediction)

In intra prediction, the prediction direction may have a prediction direction with respect to the position of a reference sample used for prediction according to a prediction mode. An intra prediction mode having a prediction direction is referred to as an intra directional prediction mode. On the other hand, as an intra prediction mode having no prediction direction, there are an intra planner (INTRA_PLANAR) prediction mode and an intra DC (INTRA_DC) prediction mode.

Table 1 below illustrates the intra prediction mode and related names.

Intra prediction performs prediction on the current processing block based on the derived prediction mode. Since the prediction mode is different from the reference sample used for the prediction according to the prediction mode, when the current block is encoded in the intra prediction mode, the decoder derives the prediction mode of the current block to perform the prediction.

The decoder derives the intra prediction mode of the current processing block. The decoder then checks whether neighboring samples of the current processing block can be used for prediction and constructs reference samples to use for prediction.

In intra prediction, the neighboring samples of the current processing block are samples neighboring the left boundary of the current processing block of size nSХnS and a total of 2ХnS samples neighboring the bottom-left, top of the current processing block. ) Means a total of 2ХnS samples neighboring to the top border and a sample neighboring to the top border and one sample neighboring to the top-left side of the current processing block.

However, some of the surrounding samples of the current processing block may not be decoded yet or may be available. In this case, the decoder can construct reference samples for use in prediction by substituting samples that are not available with the available samples.

The decoder may perform filtering of the reference sample based on the intra prediction mode.

Whether filtering of the reference sample is performed may be determined based on the size of the current processing block. In addition, the filtering method of the reference sample may be determined by the filtering flag transmitted from the encoder.

The decoder generates a predictive block for the current processing block based on the intra prediction mode and the reference samples. That is, the decoder generates a prediction block for the current processing block based on the intra prediction mode derived from the intra prediction mode derivation step, the reference samples obtained through the reference sample construction step and the reference sample filtering step (ie, the current processing block). My prediction sample).

In order to minimize the discontinuity of the boundary between the processing blocks when the current processing block is encoded in INTRA_DC mode, the left boundary sample of the prediction block (ie, the sample in the prediction block adjacent to the left boundary) in the prediction block generation step. ) And a top boundary sample (ie, a sample in a prediction block neighboring the top boundary) may be filtered.

In addition, in the prediction block generation step, filtering may be applied to the left boundary sample or the upper boundary sample in the vertical direction mode and the horizontal mode among the intra directional prediction modes similarly to the INTRA_DC mode.

In more detail, when the current processing block is encoded in the vertical mode or the horizontal mode, the value of the prediction sample may be derived based on a reference sample located in the prediction direction. In this case, a boundary sample which is not located in the prediction direction among the left boundary sample or the upper boundary sample of the prediction block may be adjacent to the reference sample which is not used for prediction. That is, the distance from the reference sample not used for prediction may be much closer than the distance from the reference sample used for prediction.

Thus, the decoder may adaptively apply filtering to left boundary samples or upper boundary samples depending on whether the intra prediction direction is vertical or horizontal. That is, when the intra prediction direction is the vertical direction, the filtering may be applied to the left boundary samples, and when the intra prediction direction is the horizontal direction, the filtering may be applied to the upper boundary samples.

As described above, the prediction block of the current block may be generated by using a total of 35 intra-picture prediction methods. The 35 prediction methods include 33 directional prediction methods and two non-directional prediction methods. For the 33 directional prediction modes, when calculating the prediction sample from the reference samples, the reference sample value is copied to the corresponding prediction sample in consideration of each direction. On the other hand, in the two non-directional prediction methods, DC mode and planar mode, the prediction sample is calculated by the weighted sum and the average value of neighboring reference samples, respectively.

In the following, intra prediction using 67 modes will be described.

Referring to FIG. 5, the intra prediction mode may use 67 modes. In order to improve the accuracy of intra coding and prediction of a high resolution image, 35 intra prediction modes may be extended to 67 intra prediction modes.

Table 2 below shows an example of a related name of the intra prediction mode having 67 modes.

The intra prediction mode with 67 modes includes two non-directional prediction modes and 65 directional prediction modes. The 65 directional prediction modes (modes 2 to 66) include a directional mode in the existing HEVC and a newly added directional mode.

In FIG. 5, the arrows indicated by solid lines represent 33 directional modes of the existing HEVC, and the arrows indicated by dashed lines represent newly added 32 directional modes.

The intra planner (INTRA_PLANAR) mode is the same as the existing intra planner mode, and the intra DC (INTRA_DC) mode is the same as the existing intra DC mode. The newly added 32 directional modes can be applied at all (prediction) block sizes, and can also be applied to both intra-coding of luminance and chrominance components.

쿼드트리와 이진트리(QTBT: Quad-Tree Binary-Tree)Quad-Tree Binary-Tree (QTBT)

QTBT refers to a structure of a coding block in which a quadtree structure and a binarytree structure are combined. In detail, in the QTBT block division structure, an image is coded in units of CTUs, the CTU is divided into quadtrees, and the leaf nodes of the quadtrees are additionally divided into binarytrees.

Hereinafter, the QTBT structure and the split flag syntax supporting the same will be described with reference to FIG. 6.

Referring to FIG. 6, the current block may be divided into a QTBT structure. That is, the CTU may first be hierarchically divided into quadtrees. The leaf nodes of the quadtrees, which are no longer divided into quadtrees, may be hierarchically divided into binary trees.

The encoder may signal a split flag to determine whether to split the quadtree in the QTBT structure. At this time, the quadtree splitting may be adjusted (or limited) by the MinQTLumaISlice, MinQTChromaISlice or MinQTNonISlice values. Here, MinQTLumaISlice represents the minimum size of the luma component quadtree leaf node in the I-slice. MinQTLumaChromaISlice represents the minimum size of a quadtree leaf node of chroma components in an I-slice. MinQTNonISlice represents the minimum size of the quadtree leaf node in non I-slice. The names of the respective flags can be changed.

In the quadtree structure of QTBT, the luma component and the chroma component in the I-slice may have a partition structure that is independent of each other. For example, in the case of I-slice in the QTBT structure, the partition structure of the luma component and the chroma component may be determined differently. In order to support such a partitioning structure, MinQTLumaISlice and MinQTChromaISlice may have different values.

As another example, in the non-I-slice of the QTBT, the quadtree structure may have the same split structure of the luma component and the chroma component. For example, for non I-slices, the quadtree splitting structure of the luma component and the chroma component may be adjusted by the MinQTNonISlice value.

In the QTBT structure, the leaf nodes of the quadtree may be divided into binary trees. In this case, binary tree splitting may be adjusted (or limited) by MaxBTDepth, MaxBTDepthISliceL, and MaxBTDepthISliceC. Where MaxBTDepth represents the maximum depth of binary tree splitting based on leaf nodes of the quadtree in non-I-slices, MaxBTDepthISliceL represents the maximum depth of binary tree splitting of luma components in I-slices, and MaxBTDepthISliceC is I Represents the maximum depth of binary tree splitting of chroma components in slices. The names of the respective flags can be changed.

In addition, since the luma component and the chroma component may have different structures in the I-slice of QTBT, MaxBTDepthISliceL and MaxBTDepthISliceC may have different values in the I-slice.

Referring to FIG. 7, it is assumed that the current slice is an I-slice. Fig. 7 (a) shows the division structure of QTBT for the luma component, and Fig. 7 (b) shows the division structure of QTBT for the chroma component. Leaf nodes of the quadtree divided into quadtree structures may be divided into binary trees. As described above, the luma component and the chroma component in the I-slice may have different partition structures.

In the case of the split structure of QTBT, a quadtree structure and a binary tree structure may be used together. In this case, the following rule may be applied.

First, MaxBTSize is less than or equal to MaxQTSize. Here, MaxBTSize represents the maximum size of the binary tree split and MaxQTSize represents the maximum size of the quadtree split.

Second, the leaf node of QT becomes the root of BT.

Third, once split into BT, can not be split into QT again

Fourth, BT defines a vertical split and a horizontal split.

Fifth, MaxQTDepth and MaxBTDepth are predefined. Here, MaxQTDepth represents the maximum depth of quadtree splitting, and MaxBTDepth represents the maximum depth of binary tree splitting.

Sixth, MaxBTSize and MinQTSize may vary depending on the slice type.

The names of the aforementioned flags may be changed.

As described above, the QTBT structure may be represented as shown in FIG. 7.

Since the YCbCr format divides the image into luma and chroma components, similarities between each other clearly exist, but the chroma component has simpler features than the luma component.

Hereinafter, a block of a luma sample may be referred to as a luma block, and a block of a chroma sample may be referred to as a chroma block.

AMT(Adaptive Multiple Core Transfom)Adaptive Multiple Core Transfom (AMT)

The encoder obtains a residual signal after performing intra prediction or inter prediction. The encoder may then adaptively select (or determine) the transform scheme according to the prediction mode applied to the residual signal (or residual block).

The method in which the encoder adaptively selects / determines a transform scheme to be applied to the residual signal based on the prediction mode may include an adaptive multiple transform (AMT), an enhanced multiple transform (EMT), an adaptive transform kernel, or an adaptive kernel. (Adaptive kernel) and the like. Hereinafter, for convenience, the scheme is referred to as AMT.

AMT may be applied to both a block on which intra prediction is performed and a block on which inter prediction is performed. When the encoder decides to apply the AMT, the encoder generates a transform coefficient by applying the transform technique determined based on the AMT to the residual signal.

In conventional HEVC, only DCT2 and DST7 are used as a transform matrix (or transform technique) to transform the residual signal. In contrast, in AMT, in addition to DST2 and DST7, DCT5, DCT8 and DST1 are additionally used. Therefore, a total of five transformation matrices are used in AMT. In this specification, the name of each transformation matrix may be represented in various forms. For example, DCT2 may be expressed as DCT-Type 2 or DCT-II. Also, for convenience, a transformation matrix or transformation technique may be referred to as a transformation.

Table 3 below shows an example of a basic equation of five transformation matrices used in AMT.

The residual signal obtained through intra prediction may show different statistical characteristics according to the intra prediction mode. In AMT, different transforms may be applied to the residual signal according to the prediction mode.

Table 4 below shows an example of a transform set that can be applied to the current block according to each mode when the prediction mode of the current block is an intra prediction mode.

Table 4 shows both an intra prediction mode having 35 modes and an intra prediction mode having 67 modes. Referring to Table 4, the intra prediction mode having 35 modes and the intra prediction mode having 67 modes are composed of a plurality of mode groups each including one or more modes. For example, Table 4 shows that the intra prediction mode is composed of five mode groups. The number of mode groups and the configuration of modes included in each mode group may be changed.

Different transform sets may be applied to each mode group. The number of transform sets is equal to the number of mode groups. Table 4 discloses five transform sets.

Each transform set includes four combinations of transforms applied in the horizontal direction of the residual block (or residual signal) and transforms applied in the vertical direction. Hereinafter, a transform applied in the horizontal direction of the block may be referred to as a horizontal transform or a row direction transform, and a transform applied in the vertical direction of the block may be referred to as a vertical transform or column. It may be referred to as a column direction transform.

The combination of one horizontal transform and one vertical transform may be referred to as a sub transform set. Each transform set includes four sub transform sets. The four sub transform sets correspond to transform candidates applicable to the current block. That is, in AMT, a transform candidate that can be applied to the residual signal mode-dependent is determined.

The transform set may also be referred to as a set, and the transform subset may also be referred to as a subset. The number of transform subsets included in each transform set, the number of transform sets, or a combination of transform techniques may be changed. The transform set and the sub transform set may be referred to by different names.

In the present specification, one set of sub-transforms (ie, a combination of a horizontal transform and a vertical transform) may be expressed in the form (Hor, Ver). For example, (DCT8, DST7) indicates that the horizontal transform applied to one block is DCT8 and the vertical transform is DST7.

For example, the case where the encoder uses the intra prediction mode having 67 modes and the mode of the current block is 3 will be described. In this case, set 1 is selected according to Table 4. In Set 1, four combinations of DST7 and DST1 are possible. The four combinations are (DST7, DST1), (DST1, DST7), (DST1, DST1) and (DST7, DST7). One of the four combinations is selected by the encoder, and the selected combination can be applied to the current block.

Table 5 below shows an example of a transform set that can be applied to the current block when the prediction mode of the current block is inter prediction.

Referring to Table 5, for inter prediction, four combinations of transform sets are used. The configuration of the transform set can be changed.

First, the encoder determines whether to apply the AMT of the current block. The encoder may compare the results of applying and not applying AMT based on rate-distortion optimization (RDO), and determine whether to apply AMT to the current block according to the comparison result.

When AMT application is selected, the encoder performs the transformation using all four combinations (i.e., four sub-transform sets) that can be applied to the current block according to the prediction mode, and optimizes based on rate-distortion optimization (RDO). Determine one sub-transform set. In this case, Table 4 is used if the prediction mode of the current block is an intra prediction mode, and Table 5 is used if the prediction mode of the current block is inter screen prediction mode.

If no AMT is selected, (DCT2, DCT2) is applied to the residual signal in both intra picture and inter picture prediction. The reason is that the DCT2 is optimal for both the row direction and the column direction due to the characteristics of the residual signal. The encoder may transmit information indicating whether AMT is applied and information indicating a (DCT2, DCT2) transform set to the decoder. In addition, when AMT is not applied, a coding unit may be coded in a transform skip mode. In order to avoid duplication of syntax coding, when the AMT flag is not 0, information indicating a transform skip mode (eg, a transform skip flag) may not be transmitted to the decoder.

For one coding unit (or transform unit), information indicating whether AMT is applied may be transmitted to the decoder. Whether to apply AMT may be controlled by a CU level flag. For example, information indicating whether AMT is applied may be referred to as an AMT flag. The AMT flag may be a 1 bit flag. For example, if a flag value indicating whether AMT is applied is 0, this indicates that AMT is not applied, and if the flag value is 1, AMT may be applied.

When AMT is applied, information indicating an optimal transform set may be additionally transmitted to the decoder. The information indicating the optimal transform set may be one flag indicating one selected sub-transform set, or may be two flags respectively indicating a vertical transform and a horizontal transform.

The tables (or information about transform combinations applicable according to the prediction mode) used in the AMT scheme may be predefined in the encoder and the decoder. When the encoder signals the information of the transform set, the decoder performs the same kind of inverse transform as the transform method indicated by the signaled information. The tables may be defined only in the encoder, in which case the decoder may perform inverse transformation based on the transformation technique indicated by the received information.

For example, AMT may be applied to a coding unit having both a width and a height of 64 or less. In addition, the AMT described above may be applied to a luminance image (block, slice or picture).

Hereinafter, a method of applying the AMT to the encoding / decoding process of the chrominance image proposed in the present specification will be described. Data compression efficiency can be improved by applying AMT to color difference images in addition to luminance images.

In the following description of the embodiments, a method performed by an encoder and a decoder may be described in some cases, or may be described based only on the method performed by the decoder. However, even if only the method performed at the decoder is disclosed, it is obvious that the same or similar manner as the method performed at the decoder may be performed at the encoder.

실시예 1Example 1

According to the first embodiment, the decoder may use the AMT information of the luminance block as the AMT information of the chrominance block. That is, the decoder may determine whether AMT is applied to the color difference block and an AMT transform set to be used for inverse transform based on the information of the luma block. According to the present embodiment, the encoder does not separately transmit AMT information of the chrominance block to the decoder.

The encoding of the chrominance block is performed after all the encoding of the luminance block is finished. In addition, the decoding of the chrominance block is performed after the decoding of the luminance block is completed. Accordingly, the encoder / decoder may know whether AMT is applied to the luminance block at the same position when encoding / decoding the chrominance block. Here, the same position indicates a position corresponding to the position of the corresponding color difference block. In the present embodiment, the same transform as that applied to the luminance block can be used for the color difference block.

First, the decoder determines whether AMT can be applied to a current chroma block before performing inverse transform. If AMT is applicable to the current color difference block, the decoder obtains AMT conversion information of the luminance block as the conversion information to be applied to the color difference block. The encoder may perform the conversion using AMT through the same process.

Hereinafter, a method of determining whether the encoder / decoder can apply AMT to the current color difference block will be described. First, a case in which the prediction mode of the current color difference block is an intra prediction mode will be described, and then a case in which the prediction mode is an inter prediction mode will be described.

Now Color difference AMT is applied to the block Whether it is possible decision

When the prediction mode of the current chrominance block is an intra prediction mode, the encoder / decoder may determine whether AMT can be applied to the current chrominance block by using one of the following three methods.

1) The first method is a method of determining, via a luminance block, whether AMT is applicable to a current color difference block, regardless of a prediction mode and a block size.

The encoder / decoder may determine whether AMT can be applied to all color difference blocks through the AMT information of the luminance block. As an example, when the AMT is applied to the luminance block at a position corresponding to the current color difference block, the encoder / decoder may determine that the AMT is also applicable to the current color difference block.

2) The second method is a method of determining whether AMT can be applied when the intra prediction mode of the current color difference block is a specific mode.

For example, the encoder / decoder may determine that AMT is applicable when the intra prediction mode of the current color difference block is a direct mode (DM) mode or a linear mode (LM) mode. The DM mode is a prediction mode that uses the intra prediction direction of the luminance block at the same position as the current color difference block. In the DM mode, the intra prediction mode of the luminance block at the same position is used as the intra prediction mode of the color difference block. Therefore, in the DM mode, the luminance block and the chrominance block have a high correlation. The LM mode is a method of performing prediction under the assumption that the correlation between the luminance block and the chrominance block is high. Details of the LM mode will be described later.

As another example, when the intra prediction mode of the current color difference block is the DC mode, the encoder / decoder may determine not to apply AMT to the current color difference block.

3) The third method is a method of determining whether AMT is applicable according to the size of a current color difference block.

As an example, the encoder / decoder may determine that AMT is applicable only when the size of the current color difference block is greater than or equal to 4Х4 and less than or equal to 64Х64.

When the prediction mode of the current chrominance block is the inter prediction mode, the encoder / decoder may determine whether AMT is applicable to the current chrominance block by using one of the following two methods.

1) The first method is a method of determining whether AMT is applicable according to the size of a current color difference block.

2) The second method is a method of determining whether to apply AMT based on a correlation between a prediction block of the current color difference block and a prediction block (or reconstructed block) of the luminance block.

According to the encoding / decoding order of the image, the encoder / decoder already acquires the predicted image of the current chrominance block, the predicted image of the luminance block, and the reconstructed image of the luminance block before converting / inversely transforming the current chrominance block. Accordingly, the encoder / decoder may determine that AMT may be applied to the current color difference block when the correlation satisfies a preset condition. As an example, if the correlation between the prediction block of the current color difference block and the prediction block (or reconstruction block) of the luminance block is 0.5 or more, the encoder / decoder may determine that AMT is applicable to the current color difference block.

The decoder can determine whether AMT is applicable to the current chrominance block by using the schemes described above. When AMT is applicable to the current chrominance block, the decoder uses AMT information of one luminance block related to the current chrominance block as information on the current chrominance block.

As described above with reference to FIG. 7, the block division structures of the color difference component and the luminance sample may be the same or different. The decoder may obtain AMT information of the luminance block in different ways based on whether the block division structure of the color difference component and the luminance sample are the same.

If the block division structures of the chrominance component and the luminance component are the same, the shapes of the luminance blocks existing at positions corresponding to the positions of the current chrominance block and the current chrominance block are the same. In this case, the decoder uses the AMT information of the luminance block at the same position as the current color difference block as the AMT information of the current color difference block. AMT information of the luminance block is signaled from the encoder to the decoder. The luminance block at the same position indicates a luminance block existing at a position corresponding to the position of the current color difference block.

When the block division structures of the chrominance component and the luminance component are different, the shapes of the luminance blocks existing at positions corresponding to the positions of the current chrominance block and the current chrominance block may be different. In this case, the decoder first selects one luminance block of the plurality of luminance blocks. Thereafter, the decoder uses the AMT information of the selected one luminance block as the AMT information of the current color difference block. The decoder may determine / select one luminance block by using one of the following three ways.

1) The first method is to select a luminance block existing at a position corresponding to a specific position of a current color difference block. For example, the decoder may obtain AMT information from a luminance block at a position corresponding to the upper left end of the chrominance block or at a position corresponding to the center of the chrominance block.

2) The second method is to select the luminance block having the largest area of the overlapping area with the current color difference block.

3) The third method is to select a luminance block having the largest size among the luminance blocks where the current color difference block and the region overlap.

The decoder obtains AMT information from a luminance block existing at a position corresponding to the current chrominance block or a luminance block selected according to one of the above schemes, and uses the obtained AMT information as AMT information of the current chrominance block. The decoder may determine whether to apply the AMT of the current color difference block and the AMT transform set based on the AMT information of the luminance block. It demonstrates concretely below.

If the AMT is not applied to the luminance block, the decoder does not apply the AMT to the current color difference block. For example, if the AMT flag of the selected luminance block is 0, the decoder may determine not to apply AMT to the current color difference block. In this case, the decoder may apply an inverse transform of (DCT2, DCT2).

If AMT is applied to the luminance block, the decoder also applies AMT to the current color difference block. For example, if the AMT flag of the selected luminance block is 1, the decoder may determine to apply AMT to the current color difference block. At this time, the decoder uses the AMT set (Table 4 and Table 5) of the predefined luminance block. The decoder uses the same sub transform set as the sub transform set applied to the luminance block in the current color difference block. That is, the decoder inversely transforms the current chrominance block based on the sub-conversion set applied to the luminance block.

In Embodiment 1, information indicating whether AMT is applied to a luminance block and information indicating a sub-transform set applied to the luminance block are transmitted from an encoder to a decoder. However, information indicating the sub-conversion set applied to the luminance block is additionally transmitted to the decoder only when AMT is applied to the luminance block. The AMT information of the luminance block is determined by the encoder through the RDO based on Tables 4 and 5 described above. For details, refer to the description of the above-described AMT.

The encoder may adaptively change (or reconstruct) the AMT conversion set (Tables 4 and 5 described above) of the luminance block according to the characteristics of the chrominance block without using the AMT set of the luminance block as it is. Although the chrominance block and the luminance block have similarities, the chrominance block has more flat areas and simpler characteristics than the luminance block. Therefore, when the AMT transform set of the luminance block is used for conversion of the chrominance block as it is, the conversion efficiency can be reduced.

To improve conversion efficiency, the encoder / decoder may adaptively change / reconfigure the AMT transform sets (Tables 4 and 5) applied to the luminance block according to the characteristics of the chrominance block, and use the changed AMT transform set. Specifically, by limiting the types of transforms constituting the AMT transform set of the luminance block, the AMT transform set of the chrominance block may be obtained separately. A more simplified set of AMT transforms can be applied to the chrominance block. The details of the method of limiting the type of transformation will be described later.

The encoder may determine, through the RDO, a sub-transform set to be applied to the chrominance block using an AMT transform set (hereinafter, AMT set of the chrominance block) separately configured for the chrominance block. The AMT transform set of the chrominance block may be stored in advance in the encoder / decoder separately from the AMT transform set (Tables 4 and 5) of the luminance block.

In addition, the encoder may determine whether or not the color difference block is flat based on a correlation between the color difference block and the luminance block, an edge in the color difference block, or a variation of the color difference block. The encoder may reconstruct the AMT transform set of the luminance block only when it determines that the current chrominance block (or slice, picture) is flat and use it for the transform of the chrominance block.

Table 6 below shows the AMT transform set of the luminance block and the AMT transform set of the luminance block when the prediction mode of the current color difference block is an intra prediction mode. An example is shown.

In Table 6 and Table 7 to be described later, a luminance AMT set (Luma AMT set) indicates an AMT transform set that can be applied to a luminance block, and a color difference AMT set (chroma AMT set) indicates an AMT transform set that can be applied to a chrominance block. .

The luminance AMT set disclosed in Table 6 are sub-transform sets of Set 0, which is one of the transform sets of Table 4 described above. Table 6 disclosed color difference AMT set can be obtained by maintaining DST7 and DCT5 changing to DCT2 among the conversions constituting the luminance AMT set. It demonstrates concretely below.

For example, in order to improve conversion efficiency, the AMT transform set of the color difference block may consist of only DST7 and DCT2. That is, in the process of configuring the color difference AMT set by changing the luminance AMT set, all of the DST conversion techniques may be changed to DST7, and all of the DCT conversion techniques may be changed to DCT2. DCT5 and DCT8 may be changed to DCT2, and DST 1 may be changed to DST7. By changing Table 4 in this manner, the AMT transform set used for the chrominance block can be obtained.

Table 7 below shows one of the AMT transform set of the chrominance block obtained by changing the AMT transform set of the luminance block and the AMT transform set of the luminance block when the prediction mode of the current chrominance block is inter prediction. For example.

In Table 7, the luminance AMT set is the same as the transform set (Table 5 described above) that can be applied to the luminance block in the inter prediction mode. The color difference AMT set in Table 7 indicates that in the luminance AMT set, DST7 is not changed and DCT 8 is changed to DCT2.

The encoder can be used for chrominance block conversion by changing the luminance AMT set in a combination consisting of only DCT2 and DST7 in the same manner as described above in the description of Table 6. For details, refer to the description regarding Table 6 above.

The encoder can select one sub-transformation set through the RDO and transform the current color-difference block by using the AMT transform set of the chrominance block. The encoder may transmit information about the sub-transformation set of the color difference AMT set used for the transformation to the decoder. The encoder may send a flag indicating a vertical transform and a flag indicating a horizontal transform. The decoder may inversely transform the current color difference block using a transform technique indicated by the received flags.

Hereinafter, in the present embodiment, it is described whether AMT related information of a current color difference block is signaled from an encoder to a decoder.

1) When the current color difference block does not meet the AMT application condition

AMT is not applied to the current color difference block. In this case, the encoder does not signal the AMT related information of the current color difference block.

2) When the current color difference block satisfies the AMT application condition and uses the AMT information of the luminance block as the AMT information of the color difference block.

In this case, the decoder uses the AMT information of the received luminance block as the AMT information of the chrominance block. Therefore, AMT information of the color difference block is not signaled separately.

3) The encoder compares the result of applying the AMT transform set applied to the luminance block to the current chrominance block and the result of applying the existing transform scheme (DCT2 or DST7) to the current chrominance block based on the RDO, and according to the comparison result, If you decide to apply AMT

If AMT is not applied to the luminance block, AMT is not applied to the chrominance block. Therefore, in this case, AMT information of the color difference block is not signaled.

When AMT is applied to the luminance block, whether the current color difference block is applied to AMT is determined based on the RDO comparison result. In this case, a flag indicating whether AMT is applied to the current color difference block may be signaled. The flag indicating whether AMT is applied to the current color difference block may be a 1-bit flag.

4) The encoder compares the result of applying the AMT transform set applied to the luminance block to the Cb / Cr block and the result of applying the conventional transform method based on the RDO. If you decide independently

If AMT is not applied to the luminance block, AMT is not applied to the Cb block and the Cr block. Therefore, in this case, AMT information is not signaled.

When AMT is applied to the luminance block, it is determined whether AMT is applied to each of the Cb block and the Cr block based on the RDO comparison result. In this case, flags indicating whether to apply AMT to each of the Cb block and the Cr block may be signaled to the decoder. The flags indicating whether the Cb block and the Cr block apply AMT may be 1-bit flags, respectively. Therefore, in this case, a total of 2 bits may be signaled for one color difference block.

As described above, according to the present embodiment, the conversion efficiency can be improved by applying AMT to the color difference block. In addition, the decoder may reduce the number of bits signaled by using the AMT information of the luminance block as the AMT information of the chrominance block.

In FIG. 8, (a) represents a luminance block, and (b) represents chrominance blocks (Cb blocks, Cr blocks) at positions corresponding to the luminance blocks of (a). 8 shows that the block division structures of the luma block and the chroma block are different. Referring to FIG. 8, when the splitting information of the luminance block and the chrominance block is different, the decoder may select one luminance block to obtain AMT information.

When the split information of the luminance block and the chrominance block is different, the decoder may select one luminance block from which the conversion information is to be obtained. For details of the manner in which the decoder selects one luminance block, refer to the description of Embodiment 1 described above.

Among the methods of selecting one luminance block, FIG. 8 illustrates a method of selecting a luminance block existing at a position corresponding to a specific position of the current color difference block. Specifically, in FIG. 8, a luminance block existing at a position corresponding to the upper left end of the color difference block is selected. The decoder uses the AMT information of the selected luminance block as the AMT information of the Cb block and the Cr block.

실시예 2Example 2

According to Embodiment 2, AMT may be applied independently to a chrominance block. According to the present embodiment, the encoder determines whether AMT is applicable to each color difference block, and when it is determined that AMT is applicable, transmits information about an AMT transform set. Unlike the first embodiment, the AMT information of the luminance block is not used in this embodiment. AMT may be independently applied to the color difference block.

First, before performing transformation on the residual signal, the encoder determines whether AMT is applicable to the current color difference block. If AMT is applicable to the current chrominance block, the encoder determines through RDO a transform set to be applied to the chrominance block.

Hereinafter, a method of determining whether or not an encoder can apply AMT to a current color difference block will be described. First, the case where the prediction mode of the current color difference block is the intra prediction mode will be described, and the case where the prediction mode of the current chrominance block is the inter prediction mode will now be described.

When the prediction mode of the current color difference block is intra prediction, the encoder may determine whether AMT is applicable to the current color difference block by using one of the following three methods.

1) The first method is a method of determining whether AMT is applicable regardless of the prediction direction of the current color difference block and the size of the block.

2) The second method is a method of determining whether AMT is applicable only when the intra prediction mode of the current color difference block is a specific mode. As an example, the encoder may determine that AMT is applicable only when the intra prediction mode of the current color difference block is a horizontal mode, a vertical mode, a DC mode, or a PLANAR mode.

3) The third method is a method of determining whether AMT is applicable according to the size of a current color difference block. As an example, the encoder may determine that AMT is applicable only when the size of the current color difference block is greater than or equal to 4Х4 and less than or equal to 64Х64.

When the prediction mode of the current chrominance block is inter prediction, the encoder may determine whether AMT is applicable to the current chrominance block by using one of the following two methods.

1) The first method is a method of determining whether AMT is applicable according to the size of a current color difference block. As an example, the encoder / decoder may determine that AMT is applicable only when the size of the current color difference block is greater than or equal to 4Х4 and less than or equal to 64Х64.

2) The second method is a method of determining whether to apply AMT based on a correlation between a prediction block of the current color difference block and a prediction block (or reconstructed block) of the luminance block. According to the encoding order of the image, the encoder first obtains the predicted image of the current chrominance block, the predicted image of the luminance block, and the reconstructed image of the luminance block before converting the current chrominance block. Accordingly, the encoder may determine that AMT may be applied to the current color difference block when the correlation degree satisfies a preset condition. As an example, if the correlation between the prediction block of the current chrominance block and the prediction block (or reconstruction block) of the luminance block is 0.5 or more, the encoder may determine that AMT is applicable to the current chrominance block.

The encoder can determine whether AMT is applicable to the current chrominance block by using the schemes described above. If AMT is applicable to the current chrominance block, the encoder determines an optimal set of transforms to be applied to the current chrominance block based on the rate-distortion optimization (RDO).

The AMT transform set applied to the current color difference block may be set under the same conditions as the AMT transform set (Tables 4 and 5 described above) of the luminance block. That is, the encoder may use Table 4 or Table 5 according to the prediction mode, and determine an optimal transform set based on the RDO.

In addition, the encoder may use a separate AMT transform set that adaptively reduces the number of AMT transform sets to the chrominance block. The chrominance block has more flat areas and has simpler features than the luminance block. Therefore, when the AMT conversion set of the luminance block is used for the conversion of the chrominance block as it is, the conversion efficiency can be reduced. 1, for improving the conversion efficiency

Table 8 below shows an example of a transform set that can be applied to a current color difference block in an intra prediction mode.

In Table 8, each transform set includes two sub transform sets. In addition, in Table 8, DM_PLANAR, DM_DC, DM_HOR, and DM_VER mean a case where the intra prediction mode of the luminance block is the PLANAR mode, the DC mode, the horizontal mode, and the vertical mode when the mode of the color difference block is the DM mode. DM_ETC represents a case where the intra prediction mode of the luminance block is a mode other than those described above. The DM mode is a mode that uses the mode of the luminance block for the color difference block. In this case, the mode of the luminance block may be a mode that is not defined in the color difference block. In this case, the encoder may apply the AMT transform set applied to the luminance block to the chrominance block. The DM mode and the LM mode will be described later.

For example, when the current prediction mode is the DC mode, the encoder performs transformation using each of the sub-transform sets (DST7, DST7) and (DST1, DST1) included in Set 2, and performs (DST7, DST7) through RDO. And a sub-transform set of one of (DST1, DST1).

Each transform set includes two sub transform sets. The encoder determines one sub transform set based on the RDO of the two sub transform sets. The encoder may then send a flag to the decoder indicating the sub-conversion set applied to the current chrominance block. Since two sub-transform sets are defined for each mode, the flag representing the sub-transform set may be a 1-bit flag. That is, by defining only two sets of sub transforms for each prediction mode, the number of bits signaled can be reduced.

Table 9 below shows an example of a transform set that can be applied to a current color difference block in the case of an inter prediction mode.

Table 9 shows two sets of AMT transformations that can be applied in inter prediction mode. As in the case of the intra prediction mode, the flag indicating the transform set may be a 1 bit flag. This may reduce the number of bits signaled.

Tables 8 and 9 described above may be predefined in the encoder. The configuration of the transform sets defined in Tables 8 and 9 can be changed.

AMT is not applied to the current color difference block. In this case, the encoder does not signal the AMT related information (including information indicating whether the AMT is applied) of the current color difference block to the decoder.

2) When the current color difference block satisfies the AMT application condition and uses the AMT transform set applied to the luminance block in the color difference block.

The encoder may signal information indicating whether AMT is applied to the current color difference block. In addition, when AMT is applied to the current chrominance block, the encoder signals information indicating the applied AMT transform set to the decoder. Information indicating whether AMT is applied may be a 1-bit flag.

In this case, Tables 4 and 5 are used in selecting an AMT transform set. Referring to Tables 4 and 5 described above, four sub-transform sets are defined for one prediction mode in the case of a luminance block. Therefore, in this case, the information representing the AMT conversion set may be a 2-bit flag.

3) When the current chrominance block satisfies the AMT application condition and determines the AMT transform set based on Tables 8 and 9

In this case, unlike in 2), Tables 8 and 9 are used in selecting an AMT transform set. Accordingly, in this case, the encoder signals information indicating whether the current color difference block is applicable to AMT (1 bit flag) and information indicating the AMT transform set to the decoder. Referring to Tables 8 and 9 described above, two sub-transform sets are defined for one prediction mode. Therefore, in this case, the information representing the AMT conversion set may be a 1-bit flag. Compared to 2) described above, the number of bits signaled is reduced.

4) When both the Cb block and the Cr block meet the AMT application condition, and the encoder independently determines whether to apply the AMT to the Cb block and the Cr block, respectively.

Table 10 below shows an example of information signaled from an encoder to a decoder when it is independently determined whether AMT used for a Cb block and a Cr block is independently applied.

In Table 10, the AMT mode indicates whether AMT is applied to each of the Cb block and the Cr block. Referring to Table 10, if the AMT mode is 0, this indicates that AMT is applied to the Cb block and the Cr block. If AMT mode is 1, it indicates that AMT has not been applied to the Cb block and the Cr block. If AMT mode is 2, it indicates that AMT has been applied only to the Cb block. If AMT mode is 3, it indicates that AMT is applied only to the Cr block.

The encoder may signal information indicating whether AMT is applied to the Cb block and the Cr block to the decoder. Referring to Table 10, since there are four cases, information indicating whether AMT is applied may be a 2-bit flag. In addition, the encoder may signal information indicating an AMT transform set used in a block to which AMT is applied among chrominance blocks (Cb / Cr blocks). The information representing the AMT transform set may be a 1-bit flag. When AMT is applied to both the Cb block and the Cr block, the information representing the AMT transform set may be a 2-bit flag.

Through the above-described embodiments, AMT can be applied to a color difference block independently.

실시예 3Example 3

According to the third embodiment, the encoder may use AMT information of a luminance block for a specific prediction mode and apply AMT independently to other prediction modes. Example 3 corresponds to a combination of Example 1 and Example 2 described above.

In the present embodiment, the determination of whether the current color difference block is applicable to the AMT can be performed in the same or similar manner as in the above-described embodiment 1 and embodiment 2, a detailed description thereof will be omitted.

In this embodiment, the AMT transform set is adaptively configured to the current color difference block and used.

Table 11 below shows an example of an AMT transform set that may be applied according to the intra prediction mode when the prediction mode of the current color difference block is the intra prediction mode.

Referring to Table 11, when the prediction mode of the current chrominance block is the DM mode or the LM mode, the decoder uses the AMT transform set applied to the luminance block in the current chrominance block as in the first embodiment. The method of selecting a luminance block to obtain AMT transform set information and the method of selecting an AMT transform set may be performed in the same or similar manner as described in Embodiment 1 described above.

If the prediction mode of the current chrominance block is not the DM mode or the LM mode, the decoder may independently determine the AMT transform set. That is, AMT information of the luminance block is not used. The AMT transform set determination may be performed in the same or similar manner as the contents disclosed in Embodiment 2 described above.

AMT is not applied to the current color difference block. In this case, the encoder does not signal the AMT related information of the current color difference block to the decoder.

2) When the intra prediction mode of the current color difference block is DM mode or LM mode.

The AMT related information may be signaled to the decoder in the same or similar manner as the contents disclosed in Embodiment 1 described above. For details, refer to Embodiment 1 described above.

3) When the intra prediction mode of the current color difference block is not DM mode or LM mode.

The AMT related information may be signaled to the decoder in the same or similar manner as the contents disclosed in Embodiment 2 described above. For details, refer to Embodiment 2 described above.

According to the present embodiment, data compression efficiency may be further improved by using a combination of a method of determining an AMT transform set independently of a method of using an AMT transform set of a luminance block according to a prediction mode.

As one of the intra prediction modes of the chrominance image, a mode for predicting the pixel value of the chrominance image from the pixel value of the restored luminance image is used. This mode may be referred to as a linear model (LM) mode or a cross-component linear model (CCLM) mode. CCLM mode is a method based on the characteristic that the correlation (correlation) between the luminance image and the chrominance image is high.

The prediction of the Cb chrominance image and the Cr chrominance image in the CCLM mode can be performed by using the following equation.

In Equation 1, pred_c (i, j) represents a Cb color difference image or a Cr color difference image to be predicted. recon_L (2i, 2j) represents a restored luminance image. (i, j) represent the coordinates of the pixel.

In the 4: 2: 0 color format, since the size of the luminance image is twice the color difference image, the difference in image size should be taken into account. Therefore, the pixels of the luminance image to be used in the chrominance image pred_c (i, j) may be used considering all the neighboring pixels in addition to recon_L (2i, 2j). Alpha and beta of Equation 1 are calculated through the cross-correlation and average difference between the surrounding template of the color difference block (Cb block or Cr block) and the luminance block surrounding template, as shown in the gray region of FIG. 9.

CCLM methods can be divided into LM, multi-model LM (MMLM), and multi-filter LM (MFLM). The MMLM method calculates luminance pixel values corresponding to chrominance pixels through various filters other than the basic filter used in the LM. The MFLM method classifies alpha and beta of Equation 1 into groups according to pixel values, and predicts chrominance pixels by adaptively applying alpha and beta corresponding to the group.

The decoder includes an inverse quantization unit for obtaining quantized transform coefficients of the chrominance block and inverse quantizing the quantized transform coefficients. The decoder also includes an inverse transform unit for inversely transforming the inverse quantized transform coefficients. The inverse transform unit 230 generates a residual block by inversely transforming an inverse quantized block (transform coefficients).

The inverse transform unit 230 includes an adaptive transform kernel application determiner 10010, a transform information obtainer 10020, and a chrominance block inverse transform unit 10030.

The adaptive transform kernel application determiner 10010 determines whether the adaptive transform kernel can be applied to the color difference block. The adaptive transform kernel is a method of adaptively determining a transform scheme to be applied to the residual signal based on the prediction mode. In other words, the adaptive transform kernel corresponds to the above-described AMT.

In the adaptive transform kernel (ie, AMT), a combination of a first transform applied in the horizontal direction of the chrominance block and a second transform applied in the vertical direction is determined according to the prediction mode of the chrominance block. The combination of the first transform and the second transform corresponds to the transform set described above.

The transform information obtainer 10020 obtains transform information of the color difference block when the adaptive transform kernel can be applied to the color difference block. The conversion information corresponds to the above-described AMT related information.

The conversion information (ie, AMT related information) includes first conversion information (eg, a flag indicating whether to apply AMT) indicating whether the adaptive conversion kernel has been applied to the luminance block associated with the chrominance block. When the adaptive transform kernel is applied to the luminance block, the transformation information may include second transformation information indicating a combination of a transformation applied in the horizontal direction and a transformation applied in the vertical direction (for example, information indicating an optimal transformation set). ) May be further included.

The color difference block inverse transform unit 10030 inversely transforms the color difference block based on the obtained conversion information. When the first transform information indicates that the adaptive transform kernel is applied to the luminance block, the color difference block inverse transform unit may perform inverse transform of the color difference block using a combination of transforms indicated by the second transform information.

When the prediction mode of the chrominance block is an intra prediction mode, the adaptive transform kernel application determining unit 10010 determines whether to apply the adaptive transform kernel based on the size of the chrominance block or whether the intra prediction mode is a specific intra prediction mode. Can be.

When the prediction mode of the color difference block is the inter prediction mode, the adaptive transform kernel application determining unit 10010 may determine the size of the color difference block, the correlation between the prediction block of the color difference block and the prediction block of the luminance block, or the prediction block of the color difference block. Based on the correlation between the reconstructed blocks of the and the luma blocks, it may be determined whether the adaptive transform kernel is applicable.

The adaptive transform kernel application determiner 10010 may determine that the adaptive transform kernel is applicable to the color difference block when the size of the color difference block is 4x4 or more or 64x64 or less.

In addition, the adaptive transform kernel application determining unit 10010 may include a first mode (that is, an LM mode) in which a specific intra prediction mode predicts a color difference pixel value of a color difference block using the reconstructed luminance pixel value, or a color difference block. In the case of the second mode (ie, the DM mode) using the prediction direction of the luminance block at the position corresponding to, it may be determined that the adaptive transform kernel is applicable to the chrominance block.

When the chrominance block and the luminance block have the same shape, the transformation information acquisition unit 10020 may obtain the transformation information from the luminance block existing at the position corresponding to the chrominance block.

When the shape of the chrominance block and the luminance block are different from each other, the transformation information acquisition unit 10020 may select one of the plurality of luminance blocks as the luminance block, and obtain the transformation information from the selected luminance block.

The conversion information obtaining unit 10020 selects a luminance block existing at a position corresponding to a specific position of the chrominance block, selects a luminance block having the largest area of the region overlapping the chrominance block, or the chrominance block and the region Among the overlapping luminance blocks, the luminance block having the largest size may be selected.

Decoding of the chrominance block is performed by the decoder.

The decoder determines whether the adaptive transform kernel is applicable to the color difference block (S11010). The adaptive transform kernel corresponds to the AMT described above. Since this procedure may be performed in the same or similar manner as the procedure for determining whether the AMT described in the above embodiments 1 to 3 can be applied, a detailed description thereof will be omitted.

Subsequently, when the adaptive transform kernel is applicable to the color difference block, the decoder obtains transform information of the color difference block (S11020). The conversion information corresponds to the above-described AMT information. The transformation information includes information indicating whether AMT is applicable to the current color difference block, and further includes information on a transformation set when AMT is applied. Since this procedure may be performed in the same or similar manner as the procedure of acquiring AMT information in the above-described embodiments 1 to 3, detailed description thereof will be omitted.

Thereafter, the decoder inversely transforms the color difference block using the obtained transform information (S11030). This procedure may be performed in the same manner as or similar to the procedure performed by the inverse transform based on the AMT information in Embodiments 1 to 3, and thus, a detailed description thereof will be omitted.

The embodiments described above are the components and features of the present invention are combined in a predetermined form. Each component or feature is to be considered optional unless stated otherwise. Each component or feature may be embodied in a form that is not combined with other components or features. It is also possible to combine some of the components and / or features to form an embodiment of the invention. The order of the operations described in the embodiments of the present invention may be changed. Some components or features of one embodiment may be included in another embodiment or may be replaced with corresponding components or features of another embodiment. It is obvious that the claims may be combined to form an embodiment by combining claims that do not have an explicit citation relationship in the claims or as new claims by post-application correction.

Embodiments according to the present invention may be implemented by various means, for example, hardware, firmware, software, or a combination thereof. In the case of a hardware implementation, an embodiment of the present invention may include one or more application specific integrated circuits (ASICs), digital signal processors (DSPs), digital signal processing devices (DSPDs), programmable logic devices (PLDs), FPGAs ( field programmable gate arrays), processors, controllers, microcontrollers, microprocessors, and the like.

In the case of implementation by firmware or software, an embodiment of the present invention may be implemented in the form of a module, procedure, function, etc. that performs the functions or operations described above. The software code may be stored in memory and driven by the processor. The memory may be located inside or outside the processor, and may exchange data with the processor by various known means.

It will be apparent to those skilled in the art that the present invention may be embodied in other specific forms without departing from the essential features of the present invention. Accordingly, the above detailed description should not be construed as limiting in all aspects and should be considered as illustrative. The scope of the invention should be determined by reasonable interpretation of the appended claims, and all changes within the equivalent scope of the invention are included in the scope of the invention.

As mentioned above, preferred embodiments of the present invention are disclosed for purposes of illustration, and those skilled in the art can improve and change various other embodiments within the spirit and technical scope of the present invention disclosed in the appended claims below. , Replacement or addition would be possible.

Claims

In the method for decoding the chrominance block of the image,

Determining whether an adaptive transform kernel is applicable to the chrominance block;

Acquiring transform information of the color difference block when the adaptive transform kernel is applicable to the color difference block; And

And inversely transforming the chrominance block using the obtained transform information.
The method of claim 1,

And wherein the adaptive transform kernel determines a combination of a first transform applied in a horizontal direction of the chrominance block and a second transform applied in a vertical direction according to the prediction mode of the chrominance block.
The method of claim 1,

In acquiring the conversion information,

The transformation information includes first transformation information indicating whether the adaptive transformation kernel is applied to a luminance block associated with the chrominance block.

And when the adaptive transform kernel is applied to the luminance block, further comprising second transform information indicating a combination of a transform applied in the horizontal direction and a transform applied in the vertical direction to the luminance block.
The method of claim 3,

If the first transform information indicates that the adaptive transform kernel has been applied to the luminance block,

Inversely transforming the chrominance block,

An inverse transform of the chrominance block is performed using a combination of transforms indicated by the second transform information.
The method of claim 1,

In determining whether the adaptive transform kernel is applicable,

When the prediction mode of the chrominance block is an intra prediction mode,

The applicability of the adaptive transform kernel is determined based on the size of the chrominance block or whether the intra prediction mode of the chrominance block is a specific intra prediction mode.
The method of claim 5,

If the size of the chrominance block is greater than 4x4 or less than 64x64, the adaptive transform kernel may be applied to the chrominance block.
The method of claim 5,

The specific intra prediction mode of the chrominance block may include a first mode for predicting the chrominance pixel value of the chrominance block using the reconstructed luminance pixel value, or a second using the prediction direction of the luminance block at a position corresponding to the chrominance block. How is the mode.
The method of claim 3,

In determining whether the adaptive transform kernel is applicable,

When the prediction mode of the chrominance block is an inter prediction mode,

Based on the magnitude of the chrominance block, the correlation between the prediction block of the chrominance block and the prediction block of the luminance block, or the correlation between the prediction block of the chrominance block and the reconstruction block of the luminance block. How is it applicable?
The method of claim 3,

In acquiring the conversion information,

When the color difference block and the luminance block have the same shape, the conversion information is obtained from a luminance block existing at a position corresponding to the color difference block.

If the shape of the color difference block and the luminance block is different,

Selecting one of a plurality of luminance blocks as the luminance block; And

Obtaining the conversion information from the selected one luminance block.
The method of claim 9,

Selecting one of the plurality of luminance blocks as the luminance block;

A luminance block existing at a position corresponding to a specific position of the chrominance block is selected, or a luminance block having the largest area of the area overlapping the chrominance block is selected, or a size among luminance blocks where the color difference block and the area overlap Where the largest luminance block is selected.
The method of claim 1,

In acquiring the conversion information,

The conversion information is transmitted from an encoder,

The transformation information includes first transformation information indicating whether the adaptive transformation kernel is applied to the chrominance block.

And when the adaptive transform kernel is applied to the chrominance block, the transform information further includes second transform information indicating a combination of a transform applied in a horizontal direction and a transform applied in a vertical direction to the chrominance block.
The method of claim 1,

In acquiring the conversion information,

If the prediction mode of the chrominance block is a specific intra prediction mode, transformation information of the luminance block associated with the chrominance block is obtained as transformation information of the chrominance block,

Otherwise, the conversion information is transmitted from an encoder.
An apparatus for decoding a chrominance block of an image,

The apparatus includes an inverse transform unit for inversely transforming inverse quantized transform coefficients of the chrominance block, wherein the inverse transform unit includes:

An adaptive transform kernel application determiner configured to determine whether an adaptive transform kernel is applicable to the color difference block;

A transformation information obtaining unit obtaining transformation information of the chrominance block when the adaptive transformation kernel is applicable to the chrominance block;

And a color difference block inverse transform unit which inversely transforms the color difference block by using the obtained conversion information.