KR20130124920A - Method and apparatus for image encoding/decoding - Google Patents

Abstract

The present invention discloses an image coding/decoding method and a device. An image coding device (100) comprises a movement prediction part (111), a movement compensation part (112), an intra prediction part (120), a switch (115), a subtractor (125), a conversion part (130), quantization part (140), an entropy coding part. [Reference numerals] (AA) Start;(S1200) Intra && is the present block 4�4 size?;(S1210) Brightness?;(S1220) Mode = an intra prediction mode for a brightness signal;(S1240) Horizontal direction = DST, Vertical direction = DST;(S1250) Horizontal direction = DCT, Vertical direction = DCT

Description

METHOD AND APPARATUS FOR IMAGE ENCODING / DECODING [0002]

The present invention relates to encoding / decoding of an image, and more particularly, to a transform / scan method for a residual signal.

Recently, a broadcast service having a high definition (HD) resolution (1280x1024 or 1920x1080) has been expanding worldwide as well as in Korea. As a result, many users are getting used to high resolution and high quality video, and many institutions are accelerating the development of next generation video equipment. In addition, as interest in UHD (Ultra High Definition) with resolution more than four times that of HDTV with HDTV increased, video standardization organizations became aware of the necessity of compression technology for higher resolution, high image quality. In addition, there is a need for a new standard that can achieve the same picture quality through the compression efficiency higher than H.264 / AVC used in HDTV, cell phone, and Blu-ray player,

Currently, Moving Picture Experts Group (MPEG) and Video Coding Experts Group (VCEG) jointly standardize High Efficiency Video Coding (HEVC), the next generation video codec. The video including UHD video is twice as high as H.264 / AVC And aims at encoding with compression efficiency. In addition to HD and UHD images, 3D broadcasting and mobile communication networks can provide high-quality images at lower frequencies.

The present invention provides an image encoding / decoding method and apparatus capable of improving encoding / decoding efficiency.

The present invention provides a method and apparatus for converting a residual signal capable of improving encoding / decoding efficiency.

The present invention provides a method and apparatus for scanning a residual signal capable of improving encoding / decoding efficiency.

According to one aspect of the present invention, an image decoding method is provided. The method includes classifying into a predetermined group based on the directionality of the intra prediction mode and determining a frequency conversion scheme for each of the classified predetermined groups.

By performing grouping according to the intra prediction mode of the residual signal and inducing different frequency conversion schemes and / or scan directions for each group, encoding efficiency of the residual signal may be increased.

1 is a block diagram illustrating a configuration of an image encoding apparatus according to an embodiment of the present invention.
2 is a block diagram illustrating a configuration of an image decoding apparatus according to an embodiment of the present invention.
3 is a diagram schematically illustrating a segmentation structure of an image when an image is encoded.
4 is a diagram illustrating a form of a prediction unit PU that a coding unit CU may include.
FIG. 5 is a diagram illustrating a form of a transform unit (TU) that a coding unit CU may include.
6 is a diagram illustrating an example of an intra prediction mode.
7 is a diagram for describing a method of predicting and encoding a color difference signal.
8 is a diagram illustrating an example of an up-right scan method for transform coefficients.
9 is a flowchart illustrating an embodiment of a method of determining a scan direction according to an intra side mode.
10 is a flowchart illustrating another embodiment of a method of determining a scan direction according to an intra prediction mode.
11 is a flowchart illustrating an example of a method of selecting a frequency conversion method for a residual signal (residual image).
12 is a flowchart illustrating a process of determining a transformation scheme according to an intra prediction mode according to an embodiment of the present invention.
13 is a flowchart illustrating a process of determining a transformation scheme according to an intra prediction mode according to another embodiment of the present invention.
14 is a diagram illustrating an example of a resolution difference between a luminance block and a chrominance block for explaining a method of determining an application range of a frequency conversion method according to an embodiment of the present invention.
FIG. 15 is a diagram illustrating another example of a resolution difference between a luminance block and a chrominance block for explaining a method of determining an application range of a frequency conversion method according to an embodiment of the present invention.
16 is a flowchart illustrating a process of determining a scan method according to an intra prediction mode according to an embodiment of the present invention.

Hereinafter, embodiments of the present invention will be described in detail with reference to the drawings. In describing the embodiments of the present specification, when it is determined that a detailed description of a related well-known configuration or function may obscure the gist of the present specification, the description may be omitted.

When an element is referred to herein as being connected or connected to another element, it may mean directly connected or connected to the other element, It may mean something. In addition, the content described as including a specific configuration in this specification does not exclude a configuration other than the configuration, it means that additional configuration may be included in the scope of the technical idea of the present invention or the present invention.

Terms such as first and second may be used to describe various configurations, but the configurations are not limited by the terms. The terms are used to distinguish one configuration from another. For example, without departing from the scope of the present invention, the first configuration may be referred to as the second configuration, and similarly, the second configuration may also be referred to as the first configuration.

In addition, the components shown in the embodiments of the present invention are independently shown to represent different characteristic functions, and do not mean that each component is made of separate hardware or one software component unit. In other words, each component is listed as a component for convenience of description, and at least two of the components may form one component, or one component may be divided into a plurality of components to perform a function. The integrated and separated embodiments of each component are also included in the scope of the present invention without departing from the spirit of the present invention.

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

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

1, the image encoding apparatus 100 includes a motion prediction unit 111, a motion compensation unit 112, an intra prediction unit 120, a switch 115, a subtractor 125, a transform unit 130, A quantization unit 140, an entropy encoding unit 150, an inverse quantization unit 160, an inverse transformation unit 170, an adder 175, a filter unit 180, and a reference image buffer 190.

The image encoding apparatus 100 may encode an input image in an intra mode or an inter mode and output a bit stream. In the intra mode, the switch 115 is switched to the intra mode, and in the inter mode, the switch 115 can be switched to the inter mode. Intra prediction is intra prediction, and inter prediction is inter prediction. The image encoding apparatus 100 may generate a prediction block for an input block of an input image, and then may code a residual between the input block and the prediction block. At this time, the input image may mean an original picture.

In the intra mode, the intra predictor 120 may generate a prediction block by performing spatial prediction using pixel values of blocks that are already encoded / decoded around the current block.

In the inter mode, the motion predicting unit 111 can obtain a motion vector by searching an area of the reference picture stored in the reference picture buffer 190 that is best matched with the input block. The motion compensation unit 112 may generate a prediction block by performing motion compensation using a motion vector. Here, the motion vector is a two-dimensional vector used for inter prediction, and can represent an offset between the current image to be encoded / decoded and the reference image.

The subtractor 125 may generate a residual block by a difference between the input block and the generated prediction block.

The transforming unit 130 may perform a transform on the residual block to output a transform coefficient. The quantization unit 140 may quantize the input transform coefficients according to a quantization parameter (or a quantization parameter) to output a quantized coefficient.

The entropy encoding unit 150 may output a bit stream by performing entropy encoding based on the values calculated by the quantization unit 140 or the encoding parameter values calculated in the encoding process. When entropy encoding is applied, a small number of bits are allocated to a symbol having a high probability of occurrence, and a large number of bits are allocated to a symbol having a low probability of occurrence, thereby expressing symbols, The size of the column can be reduced. Therefore, the compression performance of the image encoding can be enhanced through the entropy encoding. The entropy encoding unit 150 may use an encoding method such as Exponential-Golomb, Context-Adaptive Variable Length Coding (CAVLC), and Context-Adaptive Binary Arithmetic Coding (CABAC) for entropy encoding.

Since the image encoding apparatus 100 according to the embodiment of FIG. 1 performs inter-prediction encoding, that is, inter-view prediction encoding, the currently encoded image needs to be decoded and stored for use as a reference image. Accordingly, the quantized coefficients are inversely quantized in the inverse quantization unit 160 and inversely transformed in the inverse transformation unit 170. The inverse quantized and inverse transformed coefficients are added to the prediction block through the adder 175 and a reconstruction block is generated.

The restoration block passes through the filter unit 180 and the filter unit 180 applies at least one of a deblocking filter, a sample adaptive offset (SAO), and an adaptive loop filter (ALF) can do. The filter unit 180 may be referred to as an adaptive in-loop filter. The deblocking filter can remove block distortion occurring at the boundary between the blocks. The SAO may add a proper offset value to the pixel value to compensate for coding errors. ALF can perform filtering based on the comparison between the reconstructed image and the original image. The reconstruction block having passed through the filter unit 180 can be stored in the reference picture buffer 190.

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

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

The video decoding apparatus 200 receives the bit stream output from the encoder and decodes the video stream into the intra mode or the inter mode, and outputs the reconstructed video, that is, the reconstructed video. In the intra mode, the switch is switched to the intra mode, and in the inter mode, the switch can be switched to the inter mode.

The image decoding apparatus 200 may obtain a reconstructed residual block from the input bitstream, generate a prediction block, and add the reconstructed residual block and the prediction block to generate a reconstructed block, i.e., a reconstructed block .

The entropy decoding unit 210 may entropy-decode the input bitstream according to a probability distribution to generate symbols including a symbol of a quantized coefficient type.

When the entropy decoding method is applied, a small number of bits are assigned to a symbol having a high probability of occurrence, and a large number of bits are assigned to a symbol having a low probability of occurrence, so that the size of a bit string for each symbol is Can be reduced.

The quantized coefficients are inversely quantized in the inverse quantization unit 220 and inversely transformed in the inverse transformation unit 230. The reconstructed residual block can be generated as a result of inverse quantization / inverse transformation of the quantized coefficients.

In the intra mode, the intraprediction unit 240 may generate a prediction block by performing spatial prediction using the pixel value of the already decoded block around the current block. In the inter mode, the motion compensation unit 250 may generate a prediction block by performing motion compensation using a motion vector and a reference image stored in the reference picture buffer 270. [

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

3 is a diagram schematically illustrating a segmentation structure of an image when an image is encoded.

In High Efficiency Video Coding (HEVC), coding is performed by a coding unit (CU) in order to efficiently divide an image.

Referring to FIG. 3, in HEVC, an image 300 is sequentially divided into units of a largest coding unit (LCU) (hereinafter referred to as an LCU), and then a division structure is determined by an LCU unit. The partition structure refers to a distribution of coding units (hereinafter, referred to as CUs) for efficiently encoding an image in the LCU 310, which is reduced to one CU by half of its horizontal and vertical sizes. It may be determined according to whether to split into CUs. The partitioned CU may be recursively divided into four CUs whose horizontal size and vertical size are reduced by half with respect to the CU partitioned in the same manner.

At this time, the division of the CU can be recursively divided to a predetermined depth. The depth information is information indicating the size of the CU, and is stored for each CU. For example, the depth of the LCU may be zero, and the depth of the SCU may be a predefined maximum depth. Here, the LCU is a coding unit having a maximum coding unit size as described above, and the Smallest Coding Unit (SCU) is a coding unit having a minimum coding unit size.

Each time the division from the LCU 310 to half of the horizontal and vertical sizes increases the depth of the CU by one. For each depth, the size of the CU that does not perform the division is 2Nx2N, and the size of the CU that performs the division is divided into 4 NxN size CUs. The size of N decreases by half every time the depth increases by one.

Referring to FIG. 3, an LCU having a minimum depth of 0 may be 64x64 pixels, and an SCU having a maximum depth of 3 may be 8x8 pixels. At this time, the CU (LCU) of 64x64 pixels can be represented by depth 0, the CU of 32x32 pixels by depth 1, the CU of 16x16 pixels by depth 2, and the CU (SCU) of 8x8 pixel by depth 3. [

In addition, information on whether to divide a specific CU can be expressed through division information of 1 bit for each CU. This division information can be included in all the CUs except for the SCU. For example, when the CU is not divided, 0 can be stored in the division information, and when the CU is divided, 1 can be stored in the division information.

The CU divided from the LCU includes a prediction unit (PU) or a prediction unit (PB), which is a basic unit for prediction, and a transform unit (TU or Transform Block; TB) .

Fig. 4 is a diagram showing a form of a prediction unit (PU) that a coding unit (CU) can include.

A CU that is no longer subdivided among the CUs segmented from the LCU is divided into one or more prediction units, and this action itself is also referred to as a partition (or partition). The prediction unit (hereinafter referred to as PU) is a basic unit for performing prediction, and is encoded in any one of a skip mode, an inter mode, and an intra mode, and is divided into various forms according to each mode .

Referring to FIG. 4, in the skip mode, the 2N × 2N mode 410 having the same size as the CU may be supported without a partition in the CU.

In the inter mode, eight partitioned forms within a CU, such as 2Nx2N mode 410, 2NxN mode 415, Nx2N mode 420, NxN mode 425, 2NxnU mode 430, 2NxnD mode 435 nLx2N mode 440 and nRx2N mode 445 may be supported.

In the intra mode, the 2Nx2N mode 410 and the NxN mode 425 can be supported in the CU.

FIG. 5 is a diagram illustrating a form of a transform unit (TU) that a coding unit CU may include.

A conversion unit (hereinafter referred to as a TU) is a basic unit used for a space conversion and a quantization process in a CU. The TU may have a square or rectangular shape. A CU that is not further divided among the CUs segmented from the LCU may be divided into one or more TUs. In this case, the partition structure of the TU may have a quad-tree shape as shown in FIG. 5.

On the other hand, the HEVC performs intraprediction (intra prediction) coding as in H.264 / AVC, and predictive coding is performed using neighboring blocks located around the current block according to the intra prediction mode of the current block. H.264 / AVC has 9 directional prediction modes, while HEVC performs 36 types of prediction modes including 33 directional prediction modes and 3 non-directional prediction modes.

6 is a diagram illustrating an example of an intra prediction mode. Different intra mode prediction modes may be assigned different mode numbers.

Referring to FIG. 6, there are a total of 36 prediction modes, and 33 directional modes and three types according to a direction and / or a prediction method in which reference pixels used to predict pixel values of a current block are located. It may include a non-directional mode.

Three non-directional modes are Planar (Intra_Planar) mode, DC (Intra_DC) mode, and LM mode (Intra_FromLuma) for predicting the color difference signal from the reconstructed luminance signal.

The encoding for the 36 intra prediction modes as shown in FIG. 6 may be applied to each of the luminance signal and the chrominance signal. At this time, in the case of the luminance signal, the LM mode is excluded. The encoding of the intra prediction mode for the color difference signal can be performed by the following three methods as shown in Table 1 below.

Table 1 shows an example of a method of encoding an intra prediction mode for a color difference signal.

The encoding method of the intra-prediction mode for three color-difference signals will be described with reference to Table 1. The first method uses a derived mode (DM) in which an intraprediction mode of a luminance signal is directly applied to an intraprediction mode of a color difference signal. The second method is a method using an encoding mode (EM (Explicit Mode) applying an actual intra prediction mode. The EM mode, which is an intra prediction mode of the color difference signal to be encoded, has a plane mode (Planar), an average mode (DC), a horizontal mode (Hor), a vertical mode (Ver), and the eighth position in the vertical direction (Ver + 8 or 34). Mode). The third method is a method using an LM mode for predicting a color difference signal from a restored luminance signal. The most efficient encoding method of the three modes described above can be selected.

7 is a diagram for describing a method of predicting and encoding a color difference signal.

As a method of predicting and encoding a color difference signal, a method (Intra_FromLuma) for predicting a color difference signal from a reconstructed luminance signal sample may be used. This method may use a linear correlation between the color difference signal and the luminance signal, and may be represented by Equation 1.

Here, Pred _c [x, y] is a predicted value of the chrominance signal 710, and Rec _L '[x, y] is a value calculated by using a luminance signal for matching the 4: 2: 0 sampling ratio of the chrominance signal. to be. At this time, Rec _L '[x, y] may be calculated through Equation 2.

In Equation 1, α is a weight between the down-sampled luminance signal and the color difference signal, and β represents a compensation value.

The prediction image for the color difference signal obtained through Equations 1 and 2 may have a difference value from the original image, and the difference value is entropy encoded through frequency domain transformation and quantization. In this case, the frequency domain transform may include an integer transform, a discrete cosine transform (DCT), an integer discrete sine transform (DST), or an intra prediction mode dependent DCT / DST.

As described above, when the difference value between the original image and the predicted image is entropy coded after undergoing frequency domain transformation and quantization, the efficiency of entropy coding is rearranged into a 1-dimensional form to increase the efficiency of entropy coding. In order to reorder the quantization coefficients, a zigzag scan method is used in a conventional video coding method such as H.264 / AVC, but an upper-right scan is basically used in HEVC. In addition, the frequency domain transform in HEVC uses integer transform, DCT, DST, or DCT / DST dependent on intra prediction mode.

8 is a diagram illustrating an example of an up-right scan method for transform coefficients.

When encoding quantization coefficients for a block having a random size, a block having a random size is divided into 4 × 4 subblocks to perform encoding.

Referring to FIG. 8, a 16x16 sized block may be divided into 16 4x4 sized subblocks and encoded. At the time of decoding, whether or not a transform coefficient exists in each subblock may be confirmed through flag information parsed from a bitstream. For example, the flag may be significant_coeff_group_flag (sigGrpFlag). If the significant_coeff_group_flag value is 1, it means that there is at least one quantized transform coefficient in the 4x4 subblock. On the contrary, if the significant_coeff_group_flag value is 0, it may mean that there is no quantized transform coefficient in the 4x4 subblock.

The scan direction for the 4x4 subblock and the scan direction for the significant_coeff_group_flag shown in FIG. 8 basically used the upper right scan direction.

Scanning methods for the quantization coefficients used in HEVC include upper-right, horizontal, and vertical scans. In the inter prediction, an up-right scan method is basically used, and in the intra prediction, an up-right, horizontal and vertical scan can be selectively used.

The scan direction in the intra prediction may be selected differently according to the intra prediction mode, which may be applied to both the luminance signal and the chrominance signal. Table 2 below shows an example of a scan direction according to an intra prediction mode.

In Table 2, IntraPredModeValue means intra prediction mode. In this case, the luminance signal corresponds to the IntraPredMode value, and the color difference signal corresponds to the IntraPredModeC value. log2TrafoSize means that the size of the current transform block is expressed using the log. For example, an IntraPredModeValue value of 1 means an average (DC; Intra_DC) mode, and a value of log2TrafoSize-2 of 1 means an 8x8 block.

In addition, in Table 2, scan directions according to the intra prediction mode (IntraPredModeValue) and the current transform block size (log2TrafoSize-2) are indicated by numbers 0, 1, and 2. For example, an upper right scan direction (Up-right) may be represented by 0, a horizontal scan direction (Horizontal) is 1, and a vertical scan direction (Vertical) is represented by 2.

9 is a flowchart illustrating an embodiment of a method of determining a scan direction according to an intra side mode. The method of FIG. 9 may be performed by the encoding apparatus of FIG. 1 or the decoding apparatus of FIG. 2 described above. In the embodiment of FIG. 9, the method of FIG. 9 is described as being performed by the encoding apparatus for the sake of convenience of explanation, but it can also be applied to the decoding apparatus as well.

Referring to FIG. 9, the encoding apparatus determines whether the intra prediction mode is present (S900).

If not in the intra prediction mode, the encoding apparatus determines to use the upper right direction scan as a scan for the quantization coefficients (S960).

In the intra prediction mode, the encoding apparatus determines whether the intra prediction mode IntraPredModeC for the color difference signal is the LM mode (S910).

If the intra prediction mode (IntraPredModeC) for the chrominance signal is the LM mode, the encoding apparatus determines to use the upper right direction scan as a scan for the quantization coefficients (S960).

If the intra prediction mode (IntraPredModeC) for the color difference signal is not the LM mode, the encoding apparatus determines whether the value of the intra prediction mode (IntraPredMode (C)) is 6 or more and 14 or less (S920). In this case, the intra prediction mode IntraPredMode (C) may be a luminance signal or a chrominance signal according to a signal component.

If the value of the intra prediction mode IntraPredMode (C) is 6 or more and 14 or less, the encoding apparatus determines to use the vertical scan as a scan for the quantization coefficients (S940).

If the value of the intra prediction mode IntraPredMode (C) is 6 or more and 14 or less, the encoding apparatus determines whether the value of the intra prediction mode IntraPredMode (C) is 22 or more and 30 or less (S930).

If the value of the intra prediction mode IntraPredMode (C) is greater than or equal to 22 and less than or equal to 30, the encoding apparatus determines to use a horizontal scan as a scan for the quantization coefficients (S950). Otherwise, the encoding apparatus determines to use the upper right direction scan as a scan for the quantization coefficients (S960).

10 is a flowchart illustrating another embodiment of a method of determining a scan direction according to an intra prediction mode. The method of FIG. 10 may be performed by the encoding apparatus of FIG. 1 or the decoding apparatus of FIG. 2 described above. In the embodiment of FIG. 10, for convenience of description, the method of FIG. 10 is described as being performed by the encoding apparatus, but the same may be applied to the decoding apparatus.

In FIG. 10, an index is an indicator indicating the size of a transform block, and an index value may be allocated as follows according to the size of each transform block.

If the size of the transform block is 64x64, the index value is 1; if the size of the transform block is 32x32, the index value is 2; if the size of the transform block is 16x16, the index value is 3; if the size of the transform block is 8x8, the index value is 4, if the size of the transform block is 4x4, the index value may be 5, and if the size of the transform block is 2x2, the index value may be 6.

According to the method of determining the scan direction according to the intra prediction mode of FIG. 10, a luminance signal is applied to an 8x8 size transform block and a 4x4 size transform block, and a color difference signal allows a 2x2 size transform block in the current HEVC. As you can see, it is applied only to 4x4 size conversion block. In HEVC, when the scan according to the intra prediction mode is not applied, all of the upper right scans are applied.

Meanwhile, the scan direction obtained from Table 2 for encoding the quantization coefficients in intra prediction may be used as the scan direction for the 4x4 subblock and the scan direction for the significant_coeff_group_flag as shown in FIG. 8.

Referring to FIG. 10 in more detail, the encoding apparatus determines whether it is an intra prediction mode (S1000).

If not in the intra prediction mode, the encoding apparatus determines to use the upper right direction scan as a scan for the quantization coefficients (S1070).

In the intra prediction mode, the encoding apparatus converts the width of the current transform block into an index value (S1005). Here, the index is an indicator indicating the size of the transform block as described above, and an index value may be assigned according to the size of each transform block.

In the case of the luminance signal in step S1010, the encoding apparatus obtains an intra prediction mode (IntraPredMode) for the luminance signal (S1015).

If the index value of the current transform block is greater than 3 and less than 6 in step S1020, that is, if the size of the current transform block is 8x8 or 4x4, the encoding apparatus has an intra prediction mode (IntraPredMode) value of the luminance signal of 6 or more. It is determined whether it is 14 or less (S1025).

As a result of the determination in step S1025, if the value of the intra prediction mode (IntraPredMode) for the luminance signal is 6 or more and 14 or less, the encoding apparatus determines to use the vertical scan as a scan for the quantization coefficients (S1060). Otherwise, the encoding apparatus determines whether an intra prediction mode (IntraPredMode) value of the luminance signal is 22 or more and 30 or less (S1030).

As a result of the determination in step S1030, if the value of the intra prediction mode (IntraPredMode) for the luminance signal is 22 or more and 30 or less, the encoding apparatus determines to use the horizontal scan as a scan for the quantization coefficients (S1065). Otherwise, the encoding apparatus determines to use the upper right direction scan as a scan for the quantization coefficients (S1070).

When the index value of the current transform block is greater than 3 and less than 6 in step S1020, that is, when the size of the current transform block is 64x64, 32x32, 16x16, or 2x2 except for 8x8 or 4x4, the encoding apparatus performs The scan decides to use the upper right direction scan (S1070).

If it is not the luminance signal in step S1010, that is, the color difference signal, the encoding apparatus determines whether the DM mode (S1035). Here, the DM mode is a mode in which the intra prediction mode of the luminance signal is applied as it is to the intra prediction mode of the chrominance signal.

In the DM mode, the encoding apparatus obtains an intra prediction mode (IntraPredMode) for the luminance signal (S1040), and sets an intra prediction mode (IntraPredMode) for the obtained luminance signal in the intra prediction mode (IntraPredModeC) for the chrominance signal. (S1045).

If not in the DM mode, the encoding apparatus obtains an intra prediction mode (IntraPredModeC) for the color difference signal (S1050).

The encoding apparatus determines whether the index value of the current transform block is larger than 4 and smaller than 7 (S1055). That is, it is determined whether the size of the current transform block is 4x4 or 2x2.

According to the determination result of step S1055, the encoding apparatus may perform steps S1025 and S1030, and may determine a scan method for the quantization coefficients according to the result.

Meanwhile, as described above, the difference value (or residual signal) between the original image and the predicted image is entropy coded after undergoing frequency domain transformation and quantization. In this case, to improve the coding efficiency due to the frequency domain transform, an integer transform, an integer discrete cosine transform (DCT), an integer discrete sine transform (DST), or an intra prediction mode dependent DCT / DST may be selectively and adapted according to the block size. It is applied to the enemy.

11 is a flowchart illustrating an example of a method of selecting a frequency conversion method for a residual signal (residual image). The method of FIG. 11 may be performed by the encoding apparatus of FIG. 1 or the decoding apparatus of FIG. 2 described above. In the embodiment of FIG. 11, for convenience of description, the method of FIG. 11 is described as being performed by the encoding apparatus, but the same may be applied to the decoding apparatus.

Referring to FIG. 11, in operation S1100, if the current block is encoded in the intra mode and is not a block of the luma signal, the encoding apparatus applies an integer transform or a DCT as a frequency transform method for the residual signal of the color difference signal of the current block. (S1190).

If the current block is encoded in the intra mode in step S1100 and is a block of the luminance signal, the encoding apparatus obtains an intra prediction mode (IntraPredMode) for the luminance signal of the current block (S1110).

The encoding apparatus checks whether the current block is a block of 4x4 size (iWidth == 4) (S1120).

If the current block is not a block of 4x4 size (iWidth == 4), the encoding apparatus applies integer transform or DCT as a frequency transform method for the residual signal of the luminance signal of the current block (S1190).

Otherwise, if the current block is a block of 4x4 size (iWidth == 4), the encoding apparatus checks the intra prediction mode of the current block.

If the value of the intra prediction mode of the current block is 2 or more and 10 or less (S1130), the encoding apparatus applies DST in the horizontal direction and DCT in the vertical direction in the frequency conversion method for the luminance signal of the current block (S1160). ).

Otherwise, if the value of the intra prediction mode of the current block is 0, 11 or more and 25 or less (S1140), the encoding apparatus applies the DST in both the horizontal and vertical directions as a frequency conversion method for the luminance signal of the current block ( S1170).

Otherwise, if the value of the intra prediction mode of the current block is 26 or more and 34 or less (S1150), the encoding apparatus applies DCT in the horizontal direction and DST in the vertical direction as a frequency conversion method for the luminance signal of the current block. (S1180).

Otherwise, the encoding apparatus applies DCT in both the horizontal and vertical directions as a frequency conversion method for the residual signal of the luminance signal of the current block (S1190).

In FIG. 11 described above, iWidth is an indicator indicating the size of a transform block, and an iWidth value according to the size of each transform block can be allocated as follows.

For example, iWidth is 64 if the transform block is 64x64, iWidth is 32 if the transform block is 32x32, iWidth is 16 if the transform block is 16x16, iWidth is 8 if the transform block is 8x8, and 8 is the transform block. If the size of 4x4 iWidth may be 4, if the size of the transform block 2x2 iWidth may be 2.

Meanwhile, contents related to the above-described content of FIG. 11 are posted in a transformation process for scaled transform coefficients in a high efficiency video coding (HEVC) working document. Is as follows.

Where the input is:

Width of the current transform block; nW

Height of the current transform block; nH

An array of scaled transform coefficients with element dij; (nWxnH) array d

An index for the luminance signal and the chrominance signal of the current block; cIdx

If cIdx is 0, it means a luminance signal. If cIdx is 1 or cIdx is 2, it means a color difference signal. In addition, cIdx of 1 means Cb in the color difference signal, and cIdx of 2 means Cr in the color difference signal.

Where the output is

An array of residual signals obtained by inverse transforming the scaled transform coefficients; (nWxnH) array r

If the prediction mode (PredMode) for the current block is intra prediction mode (Intra), Log2 (nW * nH) is equal to 4 and cIdx value is 0, then the variables horizTrType and vertTrType depend on the intra prediction direction mode of the luminance signal. Obtained through Table 3. If not, the variables horizTrType and vertTrType are set to zero.

An inverse transform process is performed on scaled transform coefficients with the variables horizTrType and vertTrType. First, the size of the current block (nW, nH), the scaled transform coefficient array (nWxnH array d), and the variable horizTrType are input to perform a one-dimensional inverse transformation in the horizontal direction to output the array (nWxnH array e).

Next, the array (nWxnH array e) is input and the array (nWxnH array g) is derived as shown in Equation 3 below.

Next, one-dimensional inverse transform is performed in the vertical direction by receiving the size of the current block (nW, nH), the array (nWxnH array g), and the variable vertTrType.

Next, according to cIdx, an array (nWxnH) array r for residual signals is set as in Equation 4 below.

Here, shift is shift = 20-BitDepth _Y when cIdx is 0, otherwise shift = 20-BitDepth _C. BitDepth means the number of bits (eg, 8 bits) of the sample for the current image.

In HEVC, intra prediction mode information of a luminance signal is used to select a frequency conversion scheme for a residual signal. As described above, a total of 36 intra prediction modes are classified into an intra prediction mode having 33 directionalities and an intra prediction mode having three directionalities.

In this case, one of the following four methods may be selected as the frequency conversion method according to the intra prediction mode.

1. Perform DCT conversion in both horizontal and vertical directions

2. Perform DST conversion in both horizontal and vertical directions

3. Perform DCT conversion in the horizontal direction and DST conversion in the vertical direction.

4. Perform DST conversion in the horizontal direction and DCT conversion in the vertical direction.

As described above, if a total of four frequency conversion schemes exist, the encoder and the decoder should all have hardware modules for the four frequency conversion schemes. This has the disadvantage of increasing the price of the hardware. Therefore, it is desirable to reduce the type of frequency conversion scheme in order to lower the production cost of hardware while maintaining the coding efficiency. In addition, it is desirable to find an appropriate balance between increasing coding efficiency and increasing complexity by using various frequency transforms.

For example, in the above four examples of HEVC transform schemes, only 1 and 2 are allowed, and 3 and 4 are not allowed, so that the complexity may not be greatly increased while maintaining an increase in encoding efficiency due to various transforms.

In general, when various frequency conversions are adaptively performed on the residual signal generated through intra prediction or inter prediction, encoding efficiency may be increased. However, there is an overhead of transmitting signaling information on which frequency transform is used to the decoder.

Hereinafter, the present invention provides a method and apparatus for maximizing the coding efficiency while increasing the complexity by reducing the overhead of signaling information indicating the type of frequency transform of the residual signal while adaptively selecting various frequency transforms. to provide. Hereinafter, in this specification, the case of intra prediction of HEVC will be described as an example.

The intra prediction mode of HEVC includes a prediction mode having 33 directions and a prediction mode having three directions. The prediction modes having 33 directionality are arranged by dividing into equal angles (the angle between the prediction directions is constant). For example, intra prediction mode 5 has only a slight angle difference from intra prediction mode 6. Therefore, the prediction block predicted and generated based on the intra prediction mode 5 may have little or no difference from the prediction block predicted and generated based on the intra prediction mode 6. In other words, the residual signal generated based on intra prediction mode 5 and the residual signal generated based on intra prediction mode 6 may be almost similar or identical.

In general, when various frequency conversions are adaptively performed on the residual signal, encoding efficiency may be increased. However, there is an overhead of transmitting signaling information on which frequency conversion is used.

Suppose that the residual signal generated based on intra prediction mode 5 and the residual signal generated based on intra prediction mode 6 are very similar. In this case, if DCT transform is performed in mode 5 and DST transform is performed in mode 6, the encoder can select a transform method with better coding efficiency for similar residual signals only by selecting the intra prediction direction. It is not necessary to transmit signaling information about whether this is used.

Hereinafter, a specific method of increasing coding efficiency for a residual signal using these characteristics, for example, for intra prediction of HEVC will be described. According to an embodiment of the present invention, prediction modes having similar directions of intra prediction modes are grouped with respect to the residual signal, and different frequency conversion schemes are used according to prediction mode values based on the group derived by the grouping. Such an embodiment of determining a frequency conversion method for the residual signal according to the present invention may be applied to intra prediction and / or inter prediction. It may also be applied only to the luminance signal or only to the chrominance signal or to both the luminance signal and the chrominance signal. It may also be restricted to only apply to one or more specific block size (s) or to only apply to or below (or more) a particular block size.

For example, in order to group the modes of similar intra prediction modes with respect to the residual signal, the prediction modes may be grouped according to whether the value of the intra prediction mode is odd or even, and the value of the intra prediction mode divided by 4 The prediction modes may be grouped according to the rest.

When grouping intra prediction modes, grouping may be applied to all intra prediction modes, or grouping may be applied to some intra prediction modes. For example, when grouping the intra prediction mode, the grouping may be applied only to the planar mode (hereinafter referred to as planar mode) and the average mode (hereinafter referred to as DC mode), and conversely, the directional prediction mode except for the planar mode and DC mode. You can also apply grouping only to.

Different frequency conversion schemes may be used for groups derived by grouping. The following is an example of the frequency conversion method.

1. Perform DCT conversion in both horizontal and vertical directions

2. Perform DST conversion in both horizontal and vertical directions

3. Perform DCT conversion in the horizontal direction and DST conversion in the vertical direction.

4. Perform DST conversion in the horizontal direction and DCT conversion in the vertical direction.

5. Perform no change in both horizontal and vertical directions

6. Perform DCT conversion in the horizontal direction and no conversion in the vertical direction.

7. Perform no conversion in the horizontal direction and DCT conversion in the vertical direction.

Here, no conversion means not performing frequency conversion (or frequency inverse conversion).

12 is a flowchart illustrating a process of determining a transformation scheme according to an intra prediction mode according to an embodiment of the present invention. 12 may be performed by the encoding apparatus of FIG. 1 or the decoding apparatus of FIG. 2 described above. In the embodiment of FIG. 12, for convenience of description, the method of FIG. 12 is described as being performed by the encoding apparatus. However, the same may be applied to the decoding apparatus.

Referring to FIG. 12, if the current block is encoded in an intra prediction mode and the current block is a block of 4x4 size (trafoSize (or iWidth) == 4) (S1200), the encoding apparatus performs a frequency conversion scheme on the current block. DST or DCT can be applied. Otherwise, the encoding apparatus may apply the DCT as the frequency transform scheme of the current block (S1250).

If the current block is a luminance block (S1210), the encoding apparatus obtains an intra prediction mode (Mode) for the luminance signal (S1220). If not (S1210), that is, if the current block is a chrominance block, the encoding apparatus may apply DCT as a frequency conversion scheme of the current block (S1250).

If the intra prediction mode (Mode) of the current block is even (eg, Mode% 2 == 0) (S1230), the encoding apparatus may apply the DST in both the horizontal direction and the vertical direction by using the frequency conversion method of the current block. There is (S1240).

Otherwise, if the intra prediction mode of the current block is odd (e.g., Mode% 2 == 1) (S1230), the encoding apparatus may apply the DCT in both the horizontal direction and the vertical direction using the frequency conversion scheme of the current block. It may be (S1250).

In the above-described embodiment of FIG. 12, the intra prediction mode is divided into odd groups and even groups according to the mode value, and different frequency conversion schemes are applied to each group (even and odd groups).

In this case, the planar mode and the DC mode among the intra prediction modes may be considered as one group and thus different frequency conversion methods may be determined. For example, the planar mode may apply DST, and the DC mode may apply DCT.

In addition, in the LM mode of the color difference signal, DST may be applied or DCT may be applied. In addition, in the LM mode, the above-described embodiment of FIG. 12 may be applied as a frequency conversion method for the residual signal of the color difference signal using the intra prediction mode of the luminance signal.

Meanwhile, in the above-described method of FIG. 12, a transformation process for scaled transform coefficients may be modified as follows based on HEVC.

Where the input is:

Width of the current transform block; nW

Height of the current transform block; nH

An array of scaled transform coefficients with element d _ij ; (nWxnH) array d

An index for the luminance signal and the chrominance signal of the current block; cIdx

If cIdx is 0, it means a luminance signal. If cIdx is 1 or cIdx is 2, it means a color difference signal. In addition, cIdx of 1 means Cb in the color difference signal, and cIdx of 2 means Cr in the color difference signal.

Where the output is:

An array of residual signals obtained by inverse transforming the scaled transform coefficients; (nWxnH) array r

If the Prediction Mode (PredMode) for the current block is Intra Prediction Mode (Intra), Log2 (nW * nH) is equal to 4 and cIdx is 0, then the following process is performed, otherwise the variables horizTrType and vertTrType Is set to zero.

If the intra prediction mode of the luminance signal is even, the variables horizTrType and vertTrType are set to 1 (ie, apply DST to horizontal and vertical). Otherwise, if the intra prediction mode of the luminance signal is odd, the variables horizTrType and vertTrType are set to 0 (i.e., apply DCT to horizontal and vertical).

An inverse transform process is performed on scaled transform coefficients with the variables horizTrType and vertTrType. First, the size of the current block (nW, nH), the scaled transform coefficient array (nWxnH array d), and the variable horizTrType are input to perform a one-dimensional inverse transformation in the horizontal direction to output the array (nWxnH array e).

Next, an array (nWxnH array e) is input and an array (nWxnH array g) is derived as shown in Equation 5 below.

Next, one-dimensional inverse transform is performed in the vertical direction by receiving the size of the current block (nW, nH), the array (nWxnH array g), and the variable vertTrType.

Next, the array (nWxnH) array r for the residual signal according to cIdx is set as in Equation 6 below.

Here, shift is shift = 20-BitDepth _Y when cIdx is 0, otherwise shift = 20-BitDepth _C. BitDepth means the number of bits (eg, 8 bits) of the sample for the current image.

On the other hand, the intra prediction mode for selecting the frequency conversion scheme can be brought in a variety of ways as shown in the following example.

1. The intra prediction mode of the luminance signal of the current block may be applied to both the luminance signal and the residual signal (or residual image) for the chrominance signal to be used to derive the frequency conversion scheme.

2. The intra prediction mode for the chrominance signal of the current block may be applied to both the luminance signal and the residual signal (or residual image) for the chrominance signal to be used to derive the frequency conversion scheme.

3. DCT or DST can be used only for the intra prediction mode for the luminance signal of the current block. The intra prediction mode for the color difference signal of the current block may be used to derive a frequency conversion method of the residual signal (or residual image) for the color difference signal.

In addition to the above-described method, the frequency conversion scheme may be derived in various ways.

In this case, the frequency conversion scheme according to the intra prediction mode may be used in various ways. For example, the DC mode may apply DCT, and other modes except the DC mode may apply DST in the horizontal and vertical directions.

13 is a flowchart illustrating a process of determining a transformation scheme according to an intra prediction mode according to another embodiment of the present invention. The method of FIG. 13 may be performed by the encoding apparatus of FIG. 1 or the decoding apparatus of FIG. 2 described above. In the embodiment of FIG. 13, for convenience of description, the method of FIG. 13 is described as being performed by an encoding apparatus, but the same may be applied to the decoding apparatus.

Referring to FIG. 13, when the current block is encoded in the intra prediction mode and the current block is a block of 4x4 size (trafoSize (or iWidth) == 4) (S1300), the encoding apparatus performs a frequency conversion scheme on the current block. DST or DCT can be applied. If not (S1300), the encoding apparatus may apply DCT as the frequency conversion method of the current block (S1390).

If the current block is a luminance block (S1310), the encoding apparatus obtains an intra prediction mode for the luminance signal (S1320). If not (S1310), that is, if the current block is a chrominance block, the encoding apparatus may apply DCT as a frequency conversion scheme of the current block (S1390).

If the remainder obtained by dividing the intra prediction mode (Mode) of the current block by 4 is 0 (eg, Mode% 4 == 0) (S1330), the encoding apparatus performs both horizontal and vertical directions in the frequency conversion scheme of the current block. It is possible to apply the DST to (S1360).

Otherwise, if the remainder obtained by dividing the intra prediction mode (Mode) of the current block by 4 is 1 (for example, Mode% 4 == 1) (S1340), the encoding apparatus performs DST in the horizontal direction by using the frequency conversion method of the current block. DCT may be applied in the vertical direction (S1370). Alternatively, DCT may be applied in the horizontal direction and DST may be applied in the vertical direction.

Otherwise, if the remainder of dividing the intra prediction mode (Mode) of the current block by 4 is 2 (eg, Mode% 4 == 2) (S1350), the encoding apparatus performs DCT in the horizontal direction by using the frequency conversion method of the current block. DST may be applied in the vertical direction (S1380). Alternatively, DST may be applied in the horizontal direction and DCT may be applied in the vertical direction.

Otherwise, if the intra prediction mode (Mode) of the current block divided by 4 is 3 (eg, Mode% 4 == 3) (S1350), the encoding apparatus performs horizontal and vertical directions in the frequency conversion scheme of the current block. DCT can be applied to the control (S1390).

In the above-described embodiment of FIG. 13, the intra prediction mode is divided into four groups according to the mode value (for example, four groups according to the remainder obtained by dividing the mode value of the intra prediction mode by 4), and the frequency conversion is different for each group. The method was applied.

The number of groups of intra prediction modes according to the embodiments of FIGS. 12 and 13 described above is just one example, and the present invention is not limited thereto. In addition to the above-described method, different prediction methods may be applied to each group by grouping the intra prediction modes using other methods.

Meanwhile, in the aforementioned method of FIG. 13, a transformation process for scaled transform coefficients may be modified as follows based on HEVC.

Where the input is:

Width of the current transform block; nW

Height of the current transform block; nH

An array of scaled transform coefficients with element d _ij ; (nWxnH) array d

An index for the luminance signal and the chrominance signal of the current block; cIdx

If cIdx is 0, it means a luminance signal. If cIdx is 1 or cIdx is 2, it means a color difference signal. In addition, cIdx of 1 means Cb in the color difference signal, and cIdx of 2 means Cr in the color difference signal.

Where the output is:

An array of residual signals obtained by inverse transforming the scaled transform coefficients; (nWxnH) array r

If the Prediction Mode (PredMode) for the current block is Intra Prediction Mode (Intra), Log2 (nW * nH) is equal to 4 and cIdx is 0, then the following process is performed, otherwise the variables horizTrType and vertTrType Is set to zero.

If the intra prediction mode of the luminance signal divided by 4 is 0, the variables horizTrType and vertTrType are set to 1 (i.e., apply DST to horizontal and vertical). Otherwise, if the intra prediction mode of the luminance signal divided by 4 is 1, the variable horizTrType is set to 1 and vertTrType is set to 0 (ie, DST is applied horizontally and DCT is applied vertically). Otherwise, if the intra prediction mode of the luminance signal divided by 4 is 2, the variable horizTrType is set to 0 and vertTrType is set to 1 (ie, DCT is applied horizontally and DST is applied vertically). Otherwise, if the intra prediction mode of the luminance signal divided by 4 is 3 (if not 0, 1, 2), the variables horizTrType and vertTrType are set to 0 (ie, apply DCT to horizontal and vertical).

An inverse transform process is performed on scaled transform coefficients with the variables horizTrType and vertTrType. First of all, the size of the current block (nW, nH), the scaled transform coefficient array (nWxnH array d), and the variable horizTrType are input to perform a one-dimensional inverse transformation in the horizontal direction to output the array (nWxnH array e).

Next, an array (nWxnH array e) is input and an array (nWxnH array g) is derived as shown in Equation 7 below.

Next, one-dimensional inverse transformation is performed in the vertical direction by receiving the size of the current block (nW, nH), the array (nWxnH array g), and the variable vertTrType.

Next, the array (nWxnH) array r for the residual signal according to cIdx is set as in Equation 8.

Here, shift is shift = 20-BitDepth _Y when cIdx is 0, otherwise shift = 20-BitDepth _C. BitDepth means the number of bits (eg, 8 bits) of the sample for the current image.

In the above-described embodiments, a hierarchical grouping in two units may be used as a method for more effectively grouping. For example, when the hierarchical grouping method is described using the embodiment of FIG. 13 described above, grouping is performed first when the Mode% 2 value is 0 and 1, and then (Mode >> 1) for each group. The conversion method can be determined by grouping according to whether the value of% 2 is 0 or 1.

In addition, the above-described methods of FIGS. 9 to 13 may be variously applied to the luminance and color difference signals of the current block, and may be applied as follows, for example.

If the current block is a luminance block (or a chrominance block and the intra prediction mode of the chrominance block is an LM mode), an intra prediction mode (IntraPredMode) for the luminance signal is obtained. Here, the intra prediction mode for the luminance signal of the current block may be used to derive a frequency conversion scheme of the residual signal (or residual image) for the luminance signal and the chrominance signal. If not, it is determined whether the color difference block is in DM mode. If the current block is a color difference block and a DM block, DCT may be applied as a frequency conversion method of the current block. Otherwise, if the current block is a chrominance block and not a DM block, the intra prediction mode for the chrominance signal may be used to derive the frequency conversion scheme of the residual signal (or residual image) for the chrominance signal.

Meanwhile, the above-described embodiments may vary the application range of the frequency conversion scheme according to the block size or the CU depth or the TU depth. The variable (for example, the size or depth information of the block) for determining the coverage can be set such that the encoder and the decoder use a predetermined value or use a predetermined value according to the profile or the level, If the value is written in the bitstream, the decoder may use this value from the bitstream.

Table 4 shows an example of a method of varying the application range of the frequency conversion scheme according to the CU depth. For example, Method A may apply a specific transformation method only to a depth above a given depth, Method B may apply a specific transformation method only below a given depth, and Method C may apply a specific transformation method only to a given depth. In Table 4 below, when the method is applied to the corresponding depth of the CU (or TU), it is marked as O, and when the method is not applied to the corresponding depth of the CU (or TU), it is marked as X.

Referring to Table 4, when the CU (or TU) depth is 2, all of the method A, method B, method C can be applied to the embodiments of the present invention.

If the embodiments of the present invention are not applied to all depths of a CU (or TU), they may be indicated using any indicator (eg, a flag), and one greater than the maximum value of the CU depth is applied. It may be expressed by signaling with a CU depth value indicating a range.

In addition, the method of varying the application range of the frequency conversion scheme according to the above-described block size or block depth may be applied even when the resolution between the luminance signal and the color difference signal is different. Hereinafter, a method of determining an application range of the frequency conversion method when the resolution between the luminance signal and the color difference signal is different will be described with reference to FIGS. 14 and 15.

14 is a diagram illustrating an example of a resolution difference between a luminance block and a chrominance block for explaining a method of determining an application range of a frequency conversion method according to an embodiment of the present invention.

Referring to FIG. 14, when it is assumed that a chrominance image is 1/4 size of a luminance image (for example, the luminance image is 416x240 size and the chrominance image is 208x120 size), the 8x8 luminance block 1410 is 4x4 size. Corresponds to the color difference block 1420.

In this case, the 8x8 sized luminance block 1410 may have four 4x4 sized luminance blocks, and each 4x4 sized luminance block may have an intra prediction mode. On the other hand, the 4x4 sized color difference block 1420 may not be divided into a 2x2 sized color difference block. The 4x4 sized color difference block 1420 may have one intra prediction mode.

In this case, when the 4x4 sized color difference block 1420 is encoded in the LM mode or the 4x4 sized color difference block 1420 is in the DM mode (the mode that uses the intra prediction mode of the luminance signal as the intra prediction mode of the color difference signal). When encoded, the intra prediction mode of the 8x8 luminance block 1410 for inducing a frequency conversion scheme for the residual signal (or residual image) of the 4x4 sized chrominance block 1420 is an intra of four 4x4 luminance blocks. One of the prediction modes may be used.

In order to selectively apply a frequency conversion scheme to the residual signal of the chrominance signal, a method of deriving an intra prediction mode may be variously applied as follows.

1. In one embodiment, the intra prediction mode of the block located at the upper left of the luminance signal block may be used.

2. In another embodiment, an intra prediction mode of a block located at the upper right end, the lower left end, or the lower right end of the luminance signal block may be used.

3. In another embodiment, an average value or a median value of four luminance signal blocks may be used.

4. In another embodiment, an average value or an intermediate value using intra prediction modes of the four luminance signal blocks of the current block and the color difference signal blocks of the neighboring blocks of the current block may be used.

In addition to the above-described method, the intra prediction mode for the color difference signal may be derived in various ways.

FIG. 15 is a diagram illustrating another example of a resolution difference between a luminance block and a chrominance block for explaining a method of determining an application range of a frequency conversion method according to an embodiment of the present invention.

Referring to FIG. 15, a luminance block 1510 having a size of 16 × 16 may have one intra prediction mode. On the other hand, the 8x8 sized chrominance block 1520 may be divided into four 4x4 sized chrominance blocks and may have an intra prediction mode for each of the 4x4 sized chrominance blocks.

At this time, when the 8x8 color difference block 1520 is encoded in the LM mode or the 8x8 size color difference block 1520 is in the DM mode (a mode in which the intra prediction mode of the luminance signal is used as the intra prediction mode of the color difference signal). When encoded, the intra prediction mode of the luminance block 1510 of size 16x16 may be used to derive a frequency conversion scheme for the residual signal (or residual image) of the color difference block 1520 of size 8x8. Alternatively, an intra prediction mode may be derived from a neighboring block (a luminance block or a chrominance block) of the current block to derive a frequency conversion method for the residual signal of the 8x8 color difference block 1520.

The aforementioned application range of the frequency conversion scheme may be differently applied to the chrominance block according to the size of the luminance block, may be applied differently to the luminance signal and the chrominance signal, or may be differently applied according to the horizontal direction and the vertical direction.

Table 5 shows an example of a method of determining the application range of the frequency conversion method according to the block size, the color difference signal and the luminance signal, the vertical direction and the horizontal direction.

Referring to Method 4 of Table 5, when the size of the luminance block is 8 (8x8, 8x4, 2x8, etc.), and the size of the color difference block is 4 (4x4, 4x2, 2x4), The method of determining the conversion scheme according to the intra prediction mode may be applied to the luminance signal, the color difference signal, the horizontal signal, and the vertical signal.

Looking at the method wave 2 of the method of Table 5, when the size of the luminance block is 16 (16x16, 8x16, 4x16, etc.), and the size of the color difference block is 4 (4x4, 4x2, 2x4), The method of determining the conversion scheme according to the intra prediction mode according to the embodiment may be applied to the luminance signal, the color difference signal, and the horizontal signal, but not to the vertical signal.

The above-described embodiments of the present invention are also applicable for determining the transform coefficient scan method. Hereinafter, a specific method of applying the present invention to an example of a transform coefficient scan method according to the intra prediction mode of HEVC will be described. According to an embodiment of the present invention, a prediction mode having similar characteristics of residual signals is grouped, and different transform coefficient scan orders are used according to prediction mode values based on the group derived by the grouping. Such a method according to an embodiment of the present invention may be applied to intra prediction and / or inter prediction. Further, the present invention may be applied only to the luminance signal, only to the chrominance signal, or both to the luminance signal and the chrominance signal. It may also be restricted to only apply to one or more specific block size (s) or to only apply to or below (or more) a particular block size.

For example, in order to group the modes of similar intra prediction modes with respect to the residual signal, the prediction modes may be grouped according to the remainder obtained by dividing the value of the intra prediction mode by 3, and the planar mode and the DC mode among the intra prediction modes. As a group, different scan schemes may be determined.

In this case, different scan methods may be used for each group, and the following scan methods may be used.

1. Scan the top right direction (DIAG; Up-right = 0)

2. Scan horizontal (HOR; Horizontal = 1)

3. Vertical (VER; Vertical = 2) scan

In addition, there may be a zigzag scan method.

16 is a flowchart illustrating a process of determining a scan method according to an intra prediction mode according to an embodiment of the present invention. The method of FIG. 16 may be performed by the encoding apparatus of FIG. 1 or the decoding apparatus of FIG. 2 described above. In the embodiment of FIG. 16, for convenience of description, the method of FIG. 16 is described as being performed by the encoding apparatus, but the same may be applied to the decoding apparatus.

Referring to FIG. 16, the derivation of the scan direction with respect to the luminance and color difference signals is determined as an intra prediction mode of the luminance and color difference signals.

If the current block is not encoded in the intra prediction mode (S1600), the encoding apparatus scans the current block in the upper right direction (S1655).

If the current block is encoded in the intra prediction mode (intra) and the current block is the luminance block (S1605), the encoding apparatus obtains an intra prediction mode for the luminance signal of the current block (S1610).

Otherwise, if the current block is encoded in the intra prediction mode (intra), the current block is not the luminance block (the current block is the chrominance block), and the current block is the DM mode (S1615), the encoding apparatus determines the luminance signal of the current block. An intra prediction mode for is obtained (S1620).

Otherwise, if the current block is encoded in the intra prediction mode, the current block is not the luminance block (the current block is the chrominance block), and the current block is not in the DM mode, the encoding apparatus may be used for the chrominance signal of the current block. An intra prediction mode is obtained (S1625).

As described above, the following steps may be performed based on the intra prediction mode of the current block acquired according to the luminance signal or the color difference signal.

If the size of the current block is not 4x4 or 8x8 (S1630), the encoding apparatus scans the current block in the upper right direction (S1655).

If the size of the current block is 4x4 or 8x8 and the remainder of the intra prediction mode divided by 3 of the current block is 1 (for example, Mode% 3 == 1) (S1635), the encoding apparatus scans the current block. It performs in the vertical direction (S1645). Alternatively, it may be performed in the horizontal direction or the upper right direction.

Otherwise, if the size of the current block is 4x4 or 8x8, and the remainder obtained by dividing the intra prediction mode (Mode) of the current block by 3 is 2 (eg, Mode% 3 == 2) (S1640), the encoding apparatus determines the current block. Scan in the horizontal direction (S1650). Alternatively, it may be performed in the vertical direction or the upper right direction.

Otherwise, if the size of the current block is 4x4 or 8x8, and the remainder obtained by dividing the intra prediction mode of the current block by 3 is 0 (eg, Mode% 3 == 0) (S1640), the encoding apparatus determines that the current block Scan in the upper right direction (S1655). Alternatively, it may be performed in the horizontal direction or the vertical direction.

In the above-described embodiment of FIG. 16, the intra prediction mode is divided into three groups according to the mode value (for example, three groups according to the remainder obtained by dividing the mode value of the intra prediction mode by 3) and each scan method is different for each group. Was applied.

In this case, the planar mode and the DC mode among the intra prediction modes may be considered as one group, and thus different scan methods may be determined. For example, the planar mode may apply one of the upper right direction scan, the horizontal direction scan, and the vertical direction scan, and the DC mode may apply one of the upper right direction scan, the horizontal direction scan, and the vertical direction scan. Here, the scan mode of the planar mode and the DC mode may be different from each other, or may be the same.

In the LM mode of the color difference signal, one of the upper right direction scan, the horizontal direction scan, and the vertical direction scan may be applied. In addition, in the case of the LM mode, the method of FIG. 16 may be applied as a scan method for the residual signal of the color difference signal using the intra prediction mode of the luminance signal.

Meanwhile, in the aforementioned method of FIG. 16, the TU syntax may be modified as shown in Table 6 below based on HEVC.

Referring to Table 6, Transform_unit means a bitstream order for one TU (Transform) block. log2TrafoSize means a result of right shift operation of the sum of the input log2TrafoWidth and log2TrafoHeight, and means the TU block size for the luminance signal. log2TrafoSizeC means the TU block size for the color difference signal. PredMode means a prediction encoding mode for the current block, Intra for intra coding, and Inter for inter coding. scanIdx represents the scan direction information on the luminance signal of the current TU block, and includes the upper right direction (DIAG; Up-right = 0), the horizontal direction (HOR; Horizontal = 1), and the vertical direction (VER; Vertical = 2). . scanIdxC represents the scan direction information on the color difference signal of the current TU block, and includes the upper right direction (DIAG; Up-right = 0), the horizontal direction (HOR; Horizontal = 1), and the vertical direction (VER; Vertical = 2). . IntraPredMode means intra prediction mode information on the luminance signal, and IntraPredModeC means intra prediction mode information on the chrominance signal.

The method of FIG. 16 described above may modify TU semantics based on HEVC as follows.

The transform coefficient levels are parsed into the transCoeffLevel array. In this case, when PredMode is equal to Intra, different scan directions are applied according to the intra prediction mode. This scan direction can be obtained from the remainder of the intra prediction mode divided by three.

In this case, the scanning direction of the color difference signal may be used as it is.

In the above-described embodiments, the methods are described on the basis of a flowchart as a series of steps or blocks, but the present invention is not limited to the order of the steps, and some steps may occur in different orders or simultaneously . It will also be understood by those skilled in the art that the steps depicted in the flowchart illustrations are not exclusive, that other steps may be included, or that one or more steps in the flowchart may be deleted without affecting the scope of the present invention. You will understand.

The foregoing description is merely illustrative of the technical idea of the present invention, and various changes and modifications may be made by those skilled in the art without departing from the essential characteristics of the present invention. Therefore, the embodiments disclosed in the present invention are intended to illustrate rather than limit the scope of the present invention, and the scope of the technical idea of the present invention is not limited by these embodiments. The scope of protection of the present invention should be construed according to the claims, and all technical ideas within the scope of the claims should be construed as being included in the scope of the present invention.

Claims (1)

Classifying into a predetermined group based on the directionality of the intra prediction mode; And
And determining a frequency conversion scheme for each of the classified predetermined groups.

Images