JP7026287B1 - Image decoder, image decoding method and program - Google Patents

Abstract

PROBLEM TO BE SOLVED: To ensure that the value of a quantization parameter is always "0" or more. An image decoding device 200 according to the present invention corrects the value of a decoded quantization parameter of a target block depending on whether or not adaptive color conversion is applied to the target block, and then the corrected quantum. It is provided with a quantization parameter derivation unit 220A configured to use the larger value of the quantization parameter value and the predetermined value of "0" or more as the quantization parameter of the target block. [Selection diagram] Fig. 3

Description

The present invention relates to an image decoding device, an image decoding method and a program.

Non-Patent Document 1 discloses a method for decoding a quantization parameter used in the case of dequantization.

Versatile Video Coding (Draft 7), JVET-N2001

However, the inventor of the present application has found in Non-Patent Document 1 that the value of the decoded quantization parameter may become a negative value.

Therefore, the present invention has been made in view of the above-mentioned problems, and provides an image decoding device, an image decoding method, and a program capable of ensuring that the value of the quantization parameter is always "0" or more. The purpose is.

The first feature of the present invention is an image decoding device, in which the value of the decoded quantization parameter of the target block is corrected depending on whether or not the adaptive color conversion is applied to the target block, and then after the correction. The gist is to include a quantization parameter derivation unit configured so that the larger of the value of the quantization parameter of the above and the predetermined value of "0" or more is the quantization parameter of the target block. And.

The second feature of the present invention is an image decoding method, in which the value of the decoded quantization parameter of the target block is corrected depending on whether or not the adaptive color conversion is applied to the target block, and then after the correction. The gist is to have a step of using the larger value of the value of the quantization parameter of the above and the predetermined value of "0" or more as the quantization parameter of the target block.

The third feature of the present invention is a program used in an image decoding device, in which the value of the decoded quantization parameter of the target block is corrected by the computer depending on whether or not the adaptive color conversion is applied to the target block. After that, the gist is to execute the step of using the larger value of the corrected quantization parameter value and the predetermined value of "0" or more as the quantization parameter of the target block.

INDUSTRIAL APPLICABILITY According to the present invention, it is possible to provide an image decoding device, an image decoding method and a program capable of ensuring that the value of the quantization parameter is always "0" or more.

It is a figure which shows an example of the structure of the image processing system 10 which concerns on one Embodiment. It is a figure which shows an example of the functional block of the image decoding apparatus 200 which concerns on one Embodiment. It is a figure which shows an example of the functional block of the inverse transformation / inverse quantization unit 220 of the image decoding apparatus 200 which concerns on one Embodiment. It is a flowchart which shows an example of the processing flow of the quantization parameter derivation unit 220A of the inverse conversion / dequantization unit 220 of the image decoding apparatus 200 which concerns on one Embodiment. It is a figure which shows an example of the functional block of the intra prediction unit 240 of the image decoding apparatus 200 which concerns on one Embodiment. It is a flowchart which shows an example of the decoding process flow of the intra prediction mode of a color difference signal by the intra prediction mode decoding unit 242B of the intra prediction mode 240 of the image decoding apparatus 200 which concerns on one Embodiment. It is a flowchart which shows an example of the processing flow in the case of applying BDPCM to the luminance CU or the color difference CU by the prediction signal generation unit 242C of the intra prediction unit 240 of the image decoding apparatus 200 which concerns on one Embodiment. It is a figure which shows an example of the functional block of the in-loop filter processing unit 250 of the image decoding apparatus 200 which concerns on one Embodiment. It is a figure which shows an example of the functional block of the deblocking filter part 250A of the in-loop filter processing part 250 of the image decoding apparatus 200 which concerns on one Embodiment. It is a figure for demonstrating an example of the boundary strength determination method by the boundary strength determination unit 253A / 253B of the deblocking filter unit 250A of the in-loop filter processing unit 250 of the image decoding apparatus 200 which concerns on one Embodiment. It is a figure for demonstrating the weighting coefficient of TPM and GEO. It is a flowchart which shows an example of the processing flow of the filter processing unit 255A / 255B of the in-loop filter processing unit 250 of the image decoding apparatus 200 which concerns on one Embodiment. It is a figure which shows an example of the functional block of the adaptive loop filter unit 250C of the in-loop filter processing unit 250 of the image decoding apparatus 200 which concerns on one Embodiment. It is a figure which shows an example of the filtering processing in the Cb CC-ALF part 257A of the adaptive loop filter part 250C of the in-loop filter processing part 250 of the image decoding apparatus 200 which concerns on one Embodiment, and the processing at the boundary part. It is a figure which shows an example of the filtering processing in the Cb CC-ALF part 257A of the adaptive loop filter part 250C of the in-loop filter processing part 250 of the image decoding apparatus 200 which concerns on one Embodiment, and the processing at the boundary part. It is a figure which shows an example of the functional block of the image coding apparatus 100 which concerns on one Embodiment. It is a figure which shows an example of the functional block of the intra prediction unit 112 of the image coding apparatus 100 which concerns on one Embodiment.

Hereinafter, embodiments of the present invention will be described with reference to the drawings. The components in the following embodiments can be replaced with existing components as appropriate, and various variations including combinations with other existing components are possible. Therefore, the description of the following embodiments does not limit the content of the invention described in the claims.

(First Embodiment)
Hereinafter, the image processing system 10 according to the first embodiment of the present invention will be described with reference to FIGS. 1 to 17. FIG. 1 is a diagram showing an image processing system 10 according to an embodiment according to the present embodiment.

As shown in FIG. 1, the image processing system 10 includes an image coding device 100 and an image decoding device 200.

The image coding device 100 is configured to generate encoded data by encoding an input image signal. The image decoding device 200 is configured to generate an output image signal by decoding the coded data.

Here, such coded data may be transmitted from the image coding device 100 to the image decoding device 200 via a transmission path. Further, the coded data may be stored in the storage medium and then provided from the image coding device 100 to the image decoding device 200.

(Image Decoding Device 200)
Hereinafter, the image decoding apparatus 200 according to the present embodiment will be described with reference to FIG. 2. FIG. 2 is a diagram showing an example of a functional block of the image decoding apparatus 200 according to the present embodiment.

As shown in FIG. 2, the image decoding device 200 includes a decoding unit 210, an inverse conversion / inverse quantization unit 220, an adder 230, an inter-prediction unit 241 and an intra-prediction unit 242, and an in-loop filter processing unit. It has 250 and a frame buffer 260.

The decoding unit 210 is configured to decode the coded data generated by the image coding device 100 and decode the coefficient level value.

Here, for example, decoding is entropy-encoded data decoding to which codes of different lengths are assigned based on the probability of occurrence of a coefficient level value.

The decoding unit 210 may be configured to acquire control data by decoding the coded data.

Here, the control data may include size data such as a coding block (CU: Coding Unit) size, a prediction block (PU: Precision Unit) size, and a conversion block (TU: Transfer Unit) size.

The inverse conversion / inverse quantization unit 220 is configured to perform an inverse conversion process of the coefficient level value output from the decoding unit 210. Here, the inverse transformation / inverse quantization unit 220 may be configured to perform inverse quantization of the coefficient level value prior to the inverse transformation process.

Further, the inverse conversion / inverse quantization unit 220 decodes the quantization parameter (QP: Quantization Parameter), which is a parameter defining the quantization step size used in the inverse quantization, for each block as described later. It may be configured in.

Further, the inverse conversion / inverse quantization unit 220 may be configured not to perform the inverse conversion process in the block to which the conversion skip mode (Transform Skip) described later is applied.

The adder 230 adds a prediction signal to the prediction residual signal output from the inverse conversion / inverse quantization unit 220 to generate a pre-filter processing decoding signal, and the pre-filter processing decoding signal is used by the intra prediction unit 242 and the in-loop. It is configured to output to the filter processing unit 250.

Here, the pre-filtered decoding signal constitutes a reference block used by the intra prediction unit 242.

The inter-prediction unit 241 is configured to generate a prediction signal by inter-prediction (inter-frame prediction).

Specifically, the inter-prediction unit 241 is configured to generate a prediction signal for each prediction block based on the motion vector decoded from the coded data and the reference signal included in the reference frame. The inter-prediction unit 241 is configured to output a prediction signal to the adder 230.

Here, the inter-prediction unit 241 may include a plurality of modes as a specific method of inter-prediction. For example, the inter-prediction unit 241 may include a Triangle Partition Mode (TPM) or a Geometric Partitioning (GEO) mode, which will be described later.

The intra prediction unit 242 is configured to generate a prediction signal by intra prediction (in-frame prediction).

Specifically, the intra prediction unit 242 is configured to specify a reference block included in the target frame and generate a prediction signal for each prediction block based on the specified reference block. The intra prediction unit 242 is configured to output a prediction signal to the adder 230.

The in-loop filter processing unit 250 is configured to perform filter processing on the pre-filter processing decoding signal output from the adder 230 and output the post-filter processing decoding signal to the frame buffer 260.

Here, the in-loop filter processing may be composed of a plurality of filtering processes. For example, the filtering process includes a deblocking filter process that reduces distortion that occurs at the boundary portion of a block (encoded block, a predicted block, a conversion block, or a subblock that divides them), or a filter transmitted from the image coding apparatus 100. This is an adaptive loop filter process that switches filters based on factors such as coefficients, filter selection information, and local properties of the image pattern.

The frame buffer 260 is configured to store reference frames used by the inter-prediction unit 241.

Here, the filtered decoded signal constitutes a reference frame used by the inter-prediction unit 241.

(Inverse conversion / inverse quantization unit 220)
Hereinafter, the inverse transformation / inverse quantization unit 220 according to the present embodiment will be described with reference to FIG. FIG. 3 is a diagram showing an example of a functional block of the inverse transformation / inverse quantization 220 of the image decoding apparatus 200 according to the present embodiment.

As shown in FIG. 3, the inverse conversion / inverse quantization 220 includes a quantization parameter derivation unit 220A, an inverse quantization unit 220B, and an inverse conversion unit 220C.

The quantization parameter derivation unit 220A is configured to receive control data as an input, decode the quantization parameter for each block, and output it to the inverse quantization unit 220B.

Specifically, the quantization parameter derivation unit 220A corrects the value of the decoded quantization parameter of the target block depending on whether or not the adaptive color conversion is applied to the target block, and then determines the corrected quantization parameter. It is configured so that the larger value of the value and the predetermined value of "0" or more is used as the quantization parameter of the target block.

For example, the quantization parameter derivation unit 220A corrects the value of the decoded quantization parameter of the target block depending on whether or not the adaptive color conversion is applied to the target block, and then corrects the value of the quantization parameter and ". The larger value of "0" may be configured to be the quantization parameter of the target block.

Alternatively, the quantization parameter derivation unit 220A determines whether or not the adaptive color conversion is applied to the target block by applying the decoded quantization parameter value of the target block to the target block when the conversion skip mode is not applied to the target block. After the correction, the value of the corrected quantization parameter and the larger value of "0" may be configured to be the quantization parameter of the target block.

Alternatively, the quantization parameter derivation unit 220A determines the value of the quantization parameter of the target block when the conversion skip mode is applied to the target block, regardless of whether or not the adaptive color conversion is applied to the target block. It may be configured to be equal to or larger than the minimum value of the quantization parameter when the conversion skip mode is applied.

Alternatively, the quantization parameter derivation unit 220A corrects the decoded quantization parameter value depending on whether or not the adaptive color conversion is applied to the target block when the conversion skip mode is applied to the target block. , The larger of the corrected quantization parameter value and the minimum value of the quantization parameter when the conversion skip mode is applied may be configured to be the quantization parameter of the target block.

Alternatively, the quantization parameter derivation unit 220A determines the value of the quantization parameter of the target block when the conversion skip mode is applied to the target block, regardless of whether or not the adaptive color conversion is applied to the target block. It may be configured to be "4" or more.

Alternatively, the quantization parameter derivation unit 220A determines whether or not the adaptive color conversion is applied to the target block by applying the value of the decoded quantization parameter of the target block to the target block when the conversion skip mode is applied to the target block. After the correction, the value of the corrected quantization parameter and the larger value of "4" may be configured to be the quantization parameter of the target block.

An example of the specific processing content of the quantization parameter derivation unit 220A will be described later.

The dequantization unit 220B is configured to input the quantization parameter and the coefficient level value output from the decoding unit 210, perform dequantization processing, and output the coefficient level value after dequantization. .. As for the specific processing of the dequantization unit 220B, a known method can be used, and thus the description thereof will be omitted.

The inverse conversion unit 220C is configured to receive the coefficient level value after the inverse quantization processing output by the inverse quantization unit 220B as an input, perform the inverse conversion processing, and output the residual signal. As for the specific processing of the inverse conversion unit 220C, a known method can be used, and thus the description thereof will be omitted.

When the conversion skip mode (Transform Skip) is applied to the block, the inverse conversion unit 220C is not processed and the output of the inverse quantization unit 220B is used as the output of the inverse conversion / inverse quantization unit 220. May be good. Whether or not the Transform Skip is applied to the block can be determined, for example, by checking the value of the transform # skip # flag in Non-Patent Document 1.

FIG. 4 is a flowchart showing an example of the processing flow of the quantization parameter derivation unit 220A. Hereinafter, an example of the processing of the quantization parameter derivation unit 220A according to the present embodiment will be described with reference to FIG.

As shown in FIG. 4, in step S41, the quantization parameter derivation unit 220A derives the value of the variable QpY. QpY can be calculated by the following formula, for example, as in Non-Patent Document 1.

QpY = ((qP _{Y # PRED} + CuQpDeltaVal + 64 + 2 × QpBdOffset)% (64 + QpBdOffset))-QpBdOffset ... (Equation 1)
Here, qP _{Y # PRED} is a predicted value calculated from the surrounding blocks. Further, CuQpDeltaVal is a value indicating the difference between the QpY of the adjacent block and the QpY of the target block, and is obtained by decoding the value transmitted from the image coding apparatus 100 by the decoding unit 210. QpBdOffset is a positive value set according to the internal bit depth of the pixel value.

As a specific calculation method for these variables, for example, the method described in Non-Patent Document 1 can be used.

The portion of "+ 64 + 2 x QpBdOffset" in the above (Equation 1) is added to ensure that the portion before the operator% is "0" or more, and QpY calculated by (Equation 1). The minimum value of is "-QpBdOffset".

In step S42, the quantization parameter derivation unit 220A calculates Qp'Y, which is the quantization parameter value of the luminance signal. For example, the quantization parameter derivation unit 220A can calculate Qp'Y by the following (Equation 2) as in Non-Patent Document 1.

Qp'Y = QpY + QpBdOffset ... (Equation 2)
Here, since the minimum value of QpY is "-QpBdOffset" as described above, the minimum value of Qp'Y is "0".

In step S43, the quantization parameter derivation unit 220A calculates Qp'Cb, Qp'Cr, and Cp'CbCr, which are the quantization parameter values of the color difference signal.

Actually, the quantization parameters Qp'Cb, Qp'Cr, and Cp'CbCr are defined for the case of using the functions of Cb signal, Cr signal, and Joint Chromium Coding, respectively, but the calculation method is basic. Since they are the same, only the calculation method of the parameter Qp'Cb will be described below, and the other two explanations will be omitted.

For example, the quantization parameter derivation unit 220A can calculate Qp'Cb by the following (Equation 3) as in Non-Patent Document 1.

Qp'Cb = Clip3 (-QpBdOffset, 63, qPcb + pps # cb # qp # offset + slice # cb # qp # offset + CuQpOffsetCb) + QpBdOffset ... (Equation 3)
Here, qPcb is a parameter calculated using the above-mentioned QpY. pps # cb # qp #offset is a value specified by PPS (Picture Parameter Set). slice # cb # qp # offset is a value specified by slice Header. CuQpOffsetCb is obtained by decoding the value transmitted from the image coding device 100 by the decoding unit 210.

As a specific calculation method for these parameters, for example, the method described in Non-Patent Document 1 can be used.

"Clip3 (min, max, val)" in the above-mentioned (Equation 3) returns the value of "min" when the value of the variable val is smaller than the variable min, and returns the value of "min" when the value of the variable val is larger than the variable max. It is a function that returns the value of "max", and otherwise returns the value of "val" as it is. Therefore, the minimum value of the first term on the right side of (Equation 3) is "-QpBdOffset", and "QpBdOffset" is added to the value of the first term on the right side of (Equation 3) in the second term of the right side of (Equation 3). Therefore, the minimum value of Qp'Cb is "0".

The method for deriving Qp'Y, Qp'Cb, Qp'Cr, and Cp'CbCr described above is merely an example. As long as the derivation method is such that the minimum value of each value is "0", it may be derived by a method other than the above.

In step S44, the quantization parameter derivation unit 220A describes the following with respect to the luminance signal quantization parameter Qp'Y calculated as described above and the color difference signal quantization parameters Qp'Cb, Qp'Cr, and Cp'CbCr. Correct the street.

First, consider the variable qP in which Qp'Y, Qp'Cb, Qp'Cr, and Cp'CbCr are substituted.

In the quantization parameter derivation unit 220A, when the conversion skip (Transform Skip) is not applied to the color signal (Y, Cb, Cr or CbCr) of the target block (for example, when the value of transform # skip # flag is "0"). ), And as shown in the following (Equation 4) and (Equation 5), make corrections.

qP = qP- (cu # act # enabled # flag? Offset: 0)
… (Equation 4)
qP = Max (0, qP) ... (Equation 5)
Here, cup # act # enabled # flag is a flag indicating whether or not adaptive color conversion (Adaptive Color Transfer: ACT) is applied to the target block, and offset is a positive integer. Further, Max (a, b) is a function that returns the same value as the larger value of the two variables a and b.

Further, when the Transform Skip is applied to the target block (for example, when the value of the transform #skip # flag is "1"), the quantization parameter derivation unit 220A corrects as shown in the following (Equation 6). do.

qP = Max (QpPrimeTsMin, qP- (cu # act # enabled # flag? Offset: 0)) ... (Equation 6)
Here, QpPrimeTsMin is a variable that defines the minimum value of the quantization parameter in the block to which the Transform Skip is applied.

Further, when the Quantization Parameter Derivation Unit 220A applies the Transform Skip to the target block (for example, when the value of the transform #skip #flag is "1"), the following (Equation 7) and (Equation 8) It may be corrected as shown in.

qP = Max (QpPrimeTsMin, qP)-(cu # act # enabled # flag? Offset: 0)) ... (Equation 7)
qP = Max (4, qP) ... (Equation 8)
Further, the quantization parameter derivation unit 220A may switch the values of the variables offset of the above-mentioned (Equation 4), (Equation 6) and (Equation 7) for each color signal.

For example, the quantization parameter derivation unit 220A is in the case of a Y signal and a Cb signal (corresponding to a Cg signal when ACT is applied) (for example, when the value of cIdx in Non-Patent Document 1 is “0” or “1”). Is "offset = 5", and in the case of Cr signal (corresponding to Co signal when ACT is applied) (for example, when the value of cIdx in Non-Patent Document 1 is "2"), "offset = 3" may be set. can.

By correcting the value of the quantization parameter as described above, it is possible to guarantee that the value of the quantization parameter is "0" or more for all the blocks.
Further, when the Transform Skip mode is applied to the target block, for example, by making a correction as in the above (Equation 6), it is possible to guarantee that the value of the quantization parameter is "QpPrimeTsMin" or more. ..

With such a configuration, the minimum value of the quantization parameter in the block to which the Transform Skip is applied can be unified regardless of the suitability of the adaptive color conversion.

When applying the Transform Skip to the target block, for example, by making corrections as described in (Equation 7) and (Equation 8) above, the value of the quantization parameter is guaranteed to be "4" or more. can do. When the Transform Skip mode is applied, when "QP = 4", the quantization step width becomes "1" and lossless decoding is possible. Therefore, such a configuration makes it unnecessary to use the quantization parameter. It is possible to prevent the value from becoming small.

Further, the processing of the quantization parameter correction described in step S45 can be performed as shown in the following (Equation 9) and (Equation 10) in step S42 and step S43, respectively.

Qp'Y = Max (MinVal, QpY + QpBdOffset- (cu # act # enabled # flag? Offset: 0)) ... (Equation 6)
Qp'Cb = Clip3 (-QpBdOffset + MinVal, 63, qPcb + pps # cb # qp # offset + slice # cb # qp # offset + CuQpOffsetCb) + QpBdOffset ... (Equation 10)
Here, MinVal is "0" when the Transform Skip is not applied to the target block, and is "QpPrimeTsMin" when the Transform Skip is applied to the target block.

With the above configuration, it is possible to ensure that the value of the quantization parameter is "0" or more regardless of the suitability of the adaptive color conversion. For example, in Non-Patent Document 1, the remainder calculation is performed using the value of the quantization parameter, and the divisor is required to be “0” or more, so that the stability of the calculation can be ensured.
Further, with the above configuration, it is possible to guarantee that the quantization parameter value is "QpPrimeTsMin" or more in the block to which the Transform Skip is applied, regardless of the suitability of the adaptive color conversion.

When Transform Skip is applied, the quantization step width becomes "1" when "QP = 4", and lossless decoding becomes possible. Therefore, if the above processing is performed after defining "QpPrimeTsMin = 4," it is possible to prevent the quantization parameter from becoming unnecessarily small in the Transform Skip application block.

As described above, the quantization parameter value corresponding to each color signal can be calculated for each block.

(Intra Prediction Unit 242)
Hereinafter, the intra prediction unit 242 according to the present embodiment will be described with reference to FIG. FIG. 5 is a diagram showing an example of a functional block of the intra prediction unit 242 of the image decoding device 200 according to the present embodiment.

As shown in FIG. 5, the intra prediction unit 242 includes an intra prediction mode decoding unit 242B and a prediction signal generation unit 242C.

The intra prediction unit 242 is an example of a prediction unit configured to generate a prediction signal by intra prediction (in-frame prediction).

The intra prediction mode decoding unit 242B is configured to decode information necessary for performing intra prediction for each block.

FIG. 6 is a flowchart showing an example of the flow of decoding processing of the intra prediction mode of the color difference signal by the intra prediction mode decoding unit 242B. Hereinafter, an example of the flow of the decoding process of the intra-prediction mode of the color difference signal by the intra-prediction mode decoding unit 242B will be described with reference to FIG.

In step S601, the intra prediction mode decoding unit 242B determines whether or not the predetermined condition 1 is satisfied.

The predetermined condition 1 may include a condition that the width of the target CU is equal to or less than the maximum size when the Transform Skip is applied (for example, MaxTsSize of Non-Patent Document 1).

Further, the predetermined condition 1 may include a condition that the height of the target CU is equal to or less than the maximum size (for example, MaxTsSize of Non-Patent Document 1) when the Transform Skip is applied.

Here, the intra prediction mode decoding unit 242B may set the width and height of the CU as the actual CU size (width / height) of the color difference signal according to the sampling ratio of the color difference signal. For example, in the case of YUV4: 2: 0 sampling, the CU size (width / height) of the color difference signal is halved as compared with the case of the luminance signal.

Further, under the predetermined condition 1, BDPCM (Block DPCM) is enabled in SPS (Sequence Parameter Set) (for example, the value of sps # bdpcm # chroma # enabled # flag is larger than "0". ) May be included.

If all of the predetermined conditions 1 are satisfied, this operation proceeds to step S602, and if any one of the predetermined conditions 1 is not satisfied, this operation proceeds to step S603.

In step S602, the intra prediction mode decoding unit 242B decodes intra # bdpcm # chroma # flag. When intra # bdpcm # chroma # flag is "1", it means that BDPCM is applied in the target color difference CU, and when such intra # bdpcm # chroma # flag is "0", it means that BDPCM is not applied. ..

In step S603, the intra prediction mode decoding unit 242B determines whether or not the value of intra # bdpcm # chroma # flag is larger than “0”.

Here, if it is determined that the value of intra # bdpcm # chroma # flag is larger than "0", this operation proceeds to step S608, and if not, this operation proceeds to step S604. If the above-mentioned predetermined condition 1 is not satisfied and the process of step S602 is skipped, the intra prediction mode decoding unit 242B considers the value of intra # bdpcm # chroma # flag to be “0”.

In step S608, the intra prediction mode decoding unit 242B decodes intra # bdpcm # chroma # dir # flag. intra # bdpcm # chroma # dir # flag indicates whether the prediction is performed from the vertical direction or the horizontal direction when BDPCM is applied to the target color difference CU. After the processing of step S608 is completed, this operation moves to step S610 and ends the processing.

In step S604, the intra prediction mode decoding unit 242B determines whether or not the predetermined condition 2 is satisfied.

The predetermined condition 2 may include a condition that CCLM (color difference linear prediction) can be used in the target color difference CU.

Here, if it is determined that the predetermined condition 2 is satisfied, the operation proceeds to step S605, and if not, the operation proceeds to step S606.

In step S605, the intra prediction mode decoding unit 242B decodes the clm # mode # flag. When cclm # mode # flag is "1", it means that CCLM is applied in the target color difference CU, and when cclm # mode # flag is "0", it means that CCLM is not applied.

In step S606, the intra prediction mode decoding unit 242B determines whether or not the value of cllm # mode # flag is larger than “0”.

Here, if it is determined that the value of cclm # mode # flag is larger than "0", this operation proceeds to step S609, and if not, this operation proceeds to step S607. If the predetermined condition 2 is not satisfied and the process of step S605 is skipped, the intra prediction mode decoding unit 242B considers the value of cclm # mode # flag to be “0”.

In step S609, the intra prediction mode decoding unit 242B decodes cclm # mode # idx. After the process of step S609 is completed, this operation proceeds to step S610 and ends the process.

In step S607, the intra-prediction mode decoding unit 242B decodes the normal intra-prediction mode (intra # chroma # pred # mode). After the process of step S607 is completed, this operation proceeds to step S610 and ends the process.

As described above, when the BDPCM is applied to the color difference block, the information regarding the prediction direction of the BDPCM is decoded, and when the BDPCM is not applied to the color difference block, the information regarding the CCLM or the normal intra prediction mode is decoded. By configuring the intra-prediction mode decoding unit 242B as described above, it is possible to ensure that the BDPCM, CCLM, and normal intra-prediction mode can be appropriately selected for each block.

Further, as described above, by configuring the intra prediction mode decoding unit 242B so as to use the size of the color difference CU for determining the applicability of the color difference BDPCM, the color difference sampling method can be 4: 2: 0, 4: 2: 2. Even if it changes to 2, 4: 4: 4, it is possible to make a determination according to the size of the actual color difference CU.

The prediction signal generation unit 242C shown in FIG. 5 is configured to generate a prediction signal based on the prediction mode of the processing target block decoded by the intra prediction mode decoding unit 242B. Here, as a method for generating a prediction signal, for example, a known method described in Non-Patent Document 1 can be used.

FIG. 7 is a flowchart showing an example of a processing flow when BDPCM is applied to the luminance CU or the color difference CU by the prediction signal generation unit 242C. Hereinafter, an example of such a processing flow will be described with reference to FIG. 7.

In step S71, the prediction signal generation unit 242C determines whether or not the predetermined condition 3 is satisfied.

Here, the predetermined condition 3 may include a condition that filtering to the reference pixel used for the prediction of BDPCM is enabled.

For example, by providing a flag for controlling the enable / disable of reference pixel filtering at the time of BDPCM in the headers of SPS (Sequence Parameter Set), PPS (Picture Parameter Set), Picture Header, Slice Header, etc., the filtering can be enabled / disabled. You can switch.

Further, the predetermined condition 3 may include a condition that the coding tool for screen contents such as Palette Mode is invalid in the slice to which the target CU belongs.

Whether the encoding tool for screen content such as Palette Mode is valid or invalid, for example, the value of the flag that controls the valid / invalidation of Palette Mode contained in the headers such as SPS, PPS, PictureHeader, and SpecteHeader. It can be determined by checking.

Here, if it is determined that the predetermined condition 3 is satisfied, the operation proceeds to step S72, and if not, the operation proceeds to step S73.

In step S72, the prediction signal generation unit 242C performs a filtering process for smoothing the reference pixel used for prediction.

For example, in the case of a color difference signal, the prediction signal generation unit 242C can determine the reference pixel position used for prediction based on the value of intra # bdpcm # chroma # dir # flag. In the case of the luminance signal, the prediction signal generation unit 242C can determine the reference pixel position used for prediction based on the same information.

The filtering process can be executed, for example, by convolving a linear filter having a coefficient weight of {1/4, 2/4, 1/4}. After the completion of the filtering process, this operation proceeds to step S73.

In step S73, the prediction signal generation unit 242C generates a prediction value using the reference pixel smoothed or the reference pixel not smoothed in step S72.

As the details of the predicted value generation process, a known method can be used, so the details will be omitted. After the prediction value generation process is completed, this operation proceeds to step S74 and ends the process.

As described above, the intra prediction unit 242 includes the prediction signal generation unit 242C, and the prediction signal generation unit 242C performs smoothing processing on the reference pixels used for prediction value generation prior to the prediction value generation processing by BDPCM. Even if noise such as imaging noise is superimposed on the reference pixel, the effect of reducing such noise can be expected, the prediction accuracy is improved, and the coding efficiency is improved. can.

Further, as described above, the intra prediction unit 242 includes the prediction signal generation unit 242C, and the prediction signal generation unit 242C is included in the header according to the value of the flag for controlling the enable / disable of the reference pixel filtering at the time of BDPCM. By configuring the reference pixel used for predictive value generation to control whether or not smoothing is performed prior to the predicted value generation process, the image taken by the camera considered to be effective for smoothing is performed. By performing smoothing on the screen content and not smoothing the screen content for which smoothing is considered unnecessary, it is possible to prevent an unintended decrease in the coding efficiency.

Further, as described above, the intra prediction unit 242 includes the prediction signal generation unit 242C, and the prediction signal generation unit 242C precedes the prediction value generation processing by BDPCM depending on whether the coding tool for screen content is valid or invalid. By configuring the reference pixel used to generate the predicted value to control whether or not to perform smoothing processing, the image taken by a camera for which smoothing is effective is smoothed and smoothed. By not smoothing the screen content that does not require the above, it is possible to prevent an unintended decrease in the coding efficiency.

(In-loop filter processing unit 250)
Hereinafter, the in-loop filter processing unit 250 according to the present embodiment will be described with reference to FIG. FIG. 8 is a diagram showing an example of a functional block of the in-loop filter processing unit 250 of the image decoding device 200 according to the present embodiment.

As shown in FIG. 8, the in-loop filter processing unit 250 includes a deblocking filter unit 250A, a SAO unit 250B, and an adaptive loop filter unit 250C.

The SAO unit 250B is configured to perform sample adaptive offset (SAO: Simple Adaptive Offset) processing. Since a known process can be used for the SAO unit 250B, the description of the specific process content will be omitted.

The configuration shown in FIG. 8 is merely an example. The processing order of each loop filter may be changed. Further, a configuration in which a part of the loop filter processing shown in FIG. 8 is omitted may be used. Further, a loop filter process (not shown in FIG. 8) may be added.

Hereinafter, an example of the processing contents of the deblocking filter unit 250A and the adaptive loop filter processing unit 250C will be described.

As shown in FIG. 9, the deblocking filter unit 250A includes a block boundary detection unit 251 (251A / 251B), a boundary strength determination unit 253 (253A / 253B), a filter determination unit 254 (254A / 254B), and a filter. It has a processing unit 255 (255A / 255B).

Here, the configuration with "A" at the end is the configuration related to the deblocking filter processing for the vertical block boundary, and the configuration with "B" at the end is the configuration for the horizontal block boundary. This is a configuration related to deblocking filter processing.

Hereinafter, a case where the deblocking filter processing for the horizontal block boundary is performed after the deblocking filter processing for the vertical block boundary is performed will be illustrated.

As described above, the deblocking filtering process may be applied to the coded block, the predicted block, or the transformed block. Further, the deblocking filter processing may be applied to the subblocks obtained by dividing each of the above blocks. That is, the target block and the adjacent block may be a coded block, a prediction block, a conversion block, or a subblock obtained by dividing them.

When applying a deblocking filter to a subblock, the block described below can be read as a subblock as appropriate.

Since the deblocking filter processing for the vertical block boundary and the deblocking filter processing for the horizontal block boundary are the same processing, the deblocking filter processing for the vertical block boundary will be described below.

The block boundary detection unit 251A is configured to detect the boundary (block boundary) of two adjacent blocks based on the control data indicating the block size.

Here, the blocks are a coding block (CU), a prediction block (PU), and a conversion block (TU). Since a known method can be applied as a specific detection method, detailed description thereof will be omitted.

The boundary strength determination unit 253A is configured to determine the boundary strength of the block boundary between the target block and the adjacent block.

Further, the boundary strength determination unit 253A may be configured to determine the boundary strength of the block boundary based on the control data indicating whether or not the target block and the adjacent block are intra prediction blocks.

For example, as shown in FIG. 10, the boundary strength determination unit 253A determines that the boundary strength of the block boundary is "0" when at least one of the target block and the adjacent block is a lossless block. It may be configured as follows.

Here, the lossless block may be defined as a block to which the Transform Skip is applied and the value of the quantization parameter is equal to or less than a predetermined threshold value (first threshold value).

Alternatively, the lossless block may be defined as a block to which the Transform Skip is applied and the value of the quantization parameter is equal to the first threshold value.

For example, the first threshold value may be the minimum value of the quantization parameter in the conversion skip mode (QpPrimeTsMin described above). Further, for example, the first threshold value may be set to "4".

Alternatively, the lossless block may be defined as a block in which the value of the quantization parameter is equal to or less than the first threshold value regardless of whether or not the Transform Skip is applied. For example, the first threshold value may be the minimum value of the quantization parameter in the conversion skip mode (QpPrimeTsMin described above). For example, such a threshold value may be set to "4".

For example, as shown in FIG. 10, the boundary strength determination unit 253A applies when the above-mentioned block boundary is a sub-block boundary, the affine motion compensation prediction is applied to the target CU, and PROF is valid in the Picture Header. It may be configured to determine that the boundary strength of the block boundary is "0".

For example, as shown in FIG. 10, in the boundary strength determination unit 253A, when at least one block of the target block and the adjacent block is an intra prediction block (that is, at least one of the blocks on both sides of the block boundary is intra. In the case of a predicted block), it may be configured to determine that the boundary strength of the block boundary is "2".

Further, the boundary strength determination unit 253A controls whether or not the target block and the adjacent block include a non-zero orthogonal transformation coefficient and whether or not the block boundary is the boundary of the transformation block (TU boundary). It may be configured to determine the boundary strength of such block boundaries based on the data.

For example, as shown in FIG. 10, in the boundary strength determination unit 253A, when at least one of the target block and the adjacent block contains a non-zero orthogonal transformation coefficient and the block boundary is the boundary of the transformation block. (That is, when a non-zero conversion coefficient exists in at least one of the blocks on both sides of the block boundary and it is the boundary of the TU), it is configured to determine that the boundary strength of the block boundary is "1". It may have been done.

Further, the boundary strength determination unit 253A is based on the control data indicating whether or not the absolute value of the difference between the motion vectors of the target block and the adjacent block is equal to or more than the threshold value (for example, 1/2 pixel), and is based on the control data of the block boundary. It may be configured to determine the boundary strength.

For example, as shown in FIG. 10, the boundary strength determination unit 253A has a case where the absolute value of the difference between the motion vectors of the target block and the adjacent block is equal to or more than a threshold value (for example, 1/2 pixel) (that is, both sides of the block boundary). It may be configured to determine that the boundary strength of the block boundary is "1" when the absolute value of the difference between the motion vectors of the blocks is equal to or more than the threshold value (for example, 1/2 pixel). ..

Further, the boundary strength determination unit 253A is configured to determine the boundary strength of the block boundary based on the control data indicating whether or not the reference blocks referred to in the prediction of the motion vectors of the target block and the adjacent block are different. It may have been done.

For example, as shown in FIG. 10, when the boundary strength determination unit 253A has different reference blocks referred to in the prediction of the motion vector of the target block and the adjacent block (that is, the reference images are different in the blocks on both sides of the block boundary). ) May be configured to determine that the boundary strength of the block boundary is "1".

The boundary strength determination unit 253A may be configured to determine the boundary strength of the block boundary based on the control data indicating whether or not the numbers of motion vectors of the target block and the adjacent block are different.

For example, as shown in FIG. 10, the boundary strength determination unit 253A is applied when the number of motion vectors of the target block and the adjacent block is different (that is, the number of motion vectors is different between the blocks on both sides of the block boundary). It may be configured to determine that the boundary strength of the block boundary is "1".

The boundary strength determination unit 253A may be configured to determine the boundary strength of the block boundary based on the type of the inter-prediction mode of the target block and the adjacent block.

For example, as shown in FIG. 10, the boundary strength determination unit 253A determines such a block boundary when the above-mentioned block boundary is a CU boundary and TPM is used in at least one of the target block and the adjacent block. It may be configured to determine that the boundary strength of is "1".

Alternatively, in the boundary strength determination unit 253A, when the above-mentioned block boundary is a CU boundary and TPM is used in at least one of the target block and the adjacent block, and the value of the weighting coefficient described later satisfies a predetermined condition. In addition, it may be configured to determine that the boundary strength of the block boundary is "1".

For example, as shown in FIG. 10, the boundary strength determination unit 253A determines such a block boundary when the above-mentioned block boundary is a CU boundary and GEO is used in at least one of the target block and the adjacent block. It may be configured to determine that the boundary strength of is "1".

Alternatively, in the boundary strength determination unit 253A, when the above-mentioned boundary is a CU boundary, GEO is used in at least one of the target block and the adjacent block, and the value of the weighting coefficient described later satisfies a predetermined condition. , The boundary strength of the block boundary may be determined to be "1".

For example, as shown in FIG. 10, the boundary strength determination unit 253A may be configured to determine that the boundary strength of the block boundary is "0" when none of the above conditions are satisfied. ..

It should be noted that the larger the value of the boundary strength, the higher the possibility that the block distortion generated at the block boundary is large.

In the above-mentioned boundary strength determination method, the luminance signal and the color difference signal may be determined by a common method or may be determined by using some different conditions.

Hereinafter, the weighting coefficients of TPM and GEO will be described with reference to FIGS. 11 (a) to 11 (d). TPM and GEO are modes in which a rectangular block (including a square) is divided into two regions by a straight line, and motion compensation prediction is performed using different motion vectors in each region.

In TPM, the division boundary is limited to the diagonal line of the rectangle, and GEO can perform division with a higher degree of freedom. As for the details of TPM and GEO, the specifications described in Non-Patent Document 1 and Non-Patent Document 2 can be followed, and therefore detailed description thereof will be omitted.

In TPM and GEO, in the pixels near the above-mentioned division boundary, the predicted value by the two vectors corresponding to each division area is weighted and synthesized, and the weight is defined to be changed in a gradation shape. Is synthesized smoothly.
11 (a) to 11 (d) are diagrams showing an example of weighting factors at the time of synthesis. In FIGS. 11 (a) to 11 (d), the weight coefficient is set to 0/8 to 8/8, and the numerator of the weight of the predicted value (P1) by the first motion vector is shown. The weight of the predicted value (P2) by the second motion vector can be calculated by 1- (weight of P1).
The predetermined condition 4 of the weighting coefficient at the time of determining the boundary strength described above can be defined as follows, for example, when the boundary strength is determined every four pixels.

In the predetermined condition 4, the value obtained by adding the weight coefficient values of the pixels at the four corners of the 4 × 4 pixel block adjacent to the target block boundary (4 pixels) is equal to or greater than the second threshold value and equal to or less than the third threshold value (or the second threshold value). It may be defined as larger and smaller than the third threshold value).

Alternatively, the predetermined condition 4 defines that at least one of the weight coefficient values of each pixel of the target boundary (4 pixels) is other than 0 (= 0/8) or 1 (= 8/8). Is also good.

The filter determination unit 254A is configured to determine the type of filter processing (for example, deblocking filter processing) applied to the block boundary.

For example, the filter determination unit 254A determines whether or not to apply the filter processing to the block boundary based on the boundary strength of the block boundary, the quantization parameter included in the target block and the adjacent block, and the weak filter processing. It may be configured to determine which of the strong filtering applications to apply.

The filter determination unit 254A may be configured to determine that the filter processing is not applied when the boundary strength of the block boundary is “0”.

The filter processing unit 255A is configured to perform processing on the pre-deblocking image based on the determination of the filter determination unit 254A. The processing for the image before deblocking is no filter processing, weak filter processing, strong filter processing, or the like.

That is, the deblocking filter unit 250A is based on whether or not at least one of two adjacent blocks satisfies the first condition that the value of the quantization parameter is equal to or less than a predetermined threshold value. , It is configured to control the deblocking filter applied to the boundary between the two blocks.

Alternatively, the deblocking filter unit 250A is based on whether or not at least one of two adjacent blocks satisfies the first condition that the value of the quantization parameter is a predetermined threshold value. It is configured to control the deblocking filter applied to the boundary between the two blocks.

Alternatively, when at least one of the two adjacent blocks satisfies the first condition, the deblocking filter unit 250A sets the boundary strength used for controlling the deblocking filter applied to the boundary to "0". It is configured to be set to.

Alternatively, when the deblocking filter unit 250A satisfies both the first condition and the second condition that the conversion skip mode is applied, at least one of the two adjacent blocks is satisfied. The boundary strength used to control the deblocking filter applied to the boundary is set to "0".

As described above, by setting the boundary strength of the boundary of the lossless block to "0", the deblocking filter is not applied to all the pixels in the target block, and losslessness can be ensured in block units.

FIG. 12 is a flowchart showing an example of the processing flow of the filter processing unit 255A. Hereinafter, an example of the processing flow of the filter processing unit 255A will be described with reference to FIG.

In step S121, the filter processing unit 255A calculates the pixel value after applying the filter according to the filter processing determined by the filter determination unit 254A.

In step S122, the filter processing unit 255A determines whether or not the block P satisfies the predetermined condition 5.

Hereinafter, of the two blocks adjacent to the target block boundary, one is referred to as a block P and the other is referred to as a block Q.

For example, the predetermined condition 5 may include a condition that the Palette Mode is applied to the block P.

Alternatively, for example, the predetermined condition 5 may include a condition that the block P is a lossless block. Here, as the definition of the lossless block, for example, the same definition as the definition described in the boundary strength determination unit 253A can be used.

If it is determined that the predetermined condition 5 is satisfied, the operation proceeds to step S126, and if not, the operation proceeds to step S123.

In step S123, the filter processing unit 255A reflects the pixel value after applying the filter generated in step S121 to the image after the filter processing.

On the other hand, in step S126, the filter processing unit 255A processes the pixel value of the block P so as not to reflect the pixel value after applying the filter generated in step S121.

Specifically, for example, it can be realized by setting the value of the parameter n, which means "applying the pixel value after applying the filter to the pixel n pixels away from the boundary on the block P side", to "0".

Alternatively, for example, it can be realized by replacing the pixel value after applying the filter with the pixel value before applying the filter. If the pixels in the block P do not reflect the pixel values after applying the filter as a result, a method other than the illustrated method may be used.

Next, in steps S124, S125, and S127, the block Q is subjected to the same processing as in steps S122, S123, and S126 described above, and then this operation proceeds to step S128 and ends the processing.

That is, the deblocking filter unit 250A is configured so that the deblocking filter is not applied to the pixels inside the block that satisfy the first condition that the value of the quantization parameter is equal to or less than a predetermined threshold value.

Alternatively, the deblocking filter unit 250A is configured so that the deblocking filter is not applied to the pixels inside the block that satisfy the first condition that the value of the quantization parameter is a predetermined threshold value.

Alternatively, the deblocking filter unit 250A applies the deblocking filter to the boundary based on whether or not the block satisfies both the first condition and the second condition that the conversion skip mode is applied. Is configured to control.

Alternatively, the deblocking filter unit 250A is configured so that the deblocking filter is not applied to the pixels inside the block that satisfy both the first condition and the second condition.

The threshold value of the first condition may be "4". Alternatively, the threshold value of the first condition may be the minimum value of the quantization parameter in the conversion skip mode, that is, QpPrimeTsMin.

As described above, by not reflecting the pixel value after applying the filter in the lossless block, losslessness in block units can be guaranteed.

Further, as described above, at the block boundary to which TPM or GEO is applied, the boundary strength may be controlled according to the value of the weighting coefficient.

Further, as described above, at the block boundary to which TPM or GEO is applied, the value obtained by adding the weight coefficient values of the pixels at the four corners of the 4 × 4 pixel block adjacent to the target block boundary (4 pixels) is the first value. It may be configured to control the boundary strength depending on whether or not the condition of 2 threshold values or more and 3rd threshold value or less (or larger than the 2nd threshold value and smaller than the 3rd threshold value) is satisfied.

Further, as described above, at the block boundary to which TPM or GEO is applied, at least one of the weight coefficient values of each pixel of the target boundary (4 pixels) is 0 (= 0/8) or 1 (=). It may be configured to control the boundary strength depending on whether or not the condition other than 8/8) is satisfied.

With the above configuration, even if the motion vector (MV) value or number of motion vectors (MV) between adjacent blocks does not meet the desired conditions, the weight coefficient value can be set at the block boundary to which TPM or GEO is applied. Deblocking filters can be applied appropriately for boundaries that are different and are likely to actually cause block distortion.

Further, as described above, when the above-mentioned block boundary is a sub-block boundary, the Affin motion compensation prediction is applied to the CU to which the adjacent sub-block belongs, and PROF is valid in the Picture Header, the boundary of the block boundary is concerned. It may be configured to determine that the intensity is "0".

With the above configuration, it is possible to prevent the deblocking filter from being unnecessarily applied to the value after the PROF process for refining the predicted value on a pixel-by-pixel basis, and it is expected that the subjective image quality will be improved.

As shown in FIG. 13, the adaptive loop filter unit 250C has a luminance ALF unit 256, a Cb CC-ALF unit 257A, a Cr CC-ALF unit 257B, a Cb ALF unit 258A, and a Cr ALF unit 258B.

As the processing of the luminance ALF unit 256, the Cb ALF unit 258A, and the Cr ALF unit 258B, for example, the processing described in Non-Patent Document 1 can be applied, and therefore detailed description thereof will be omitted.

Hereinafter, an example of the processing contents of the Cb CC-ALF unit 257A and the Cr CC-ALF unit 257B will be described. The Cb CC-ALF unit 257A and the Cr CC-ALF unit 257B differ only in that the target is a Cb signal or a Cr signal, and the processing contents are the same. Only the operation of the part 257A will be described, and the description of the Cr CC-ALF part 257B will be omitted.

As shown in FIG. 13, the Cb CC-ALF unit 257A takes a luminance signal before filtering processing by the luminance ALF unit 256 as an input, and adds the output value to the output of the Cb ALF unit 258A to obtain a Cb signal. It is configured to generate a filtered signal of.

14 and 15 are diagrams showing an example of the filtering process in the Cb CC-ALF section 257A and the process at the boundary portion.

Here, for example, as shown in FIGS. 14A to 14E, a two-dimensional filter in which eight filter coefficients C0 to C7 are arranged as shown in FIG. 14 is applied to the luminance signal and added to the Cb signal. Consider the case of calculating the value.

The pixel positions [y] and [x] of the luminance signal corresponding to the coefficient C2 are the pixel positions corresponding to the pixels of the Cb signal to which the values calculated by the Cb CC-ALF unit 257A are added.

The value to be added to the Cb signal can be calculated, for example, by the formula shown in FIG. Here, P [y] [x] means the pixel value of the luminance signal existing at the pixel positions [y] [x].

14 (a) to 14 (e) show the relationship between the processing boundaries such as virtual boundaries and the arrangement of filter coefficients. Here, the boundary is a boundary that prohibits reference to pixels located outside the boundary, such as the boundary of parallel processing units such as slices and tiles defined in the standard, and the virtual boundary provided to realize pipeline processing. It is a line.

Of the two regions divided by the boundary, the one where the pixel corresponding to the filter coefficient C2 exists is considered to be inside the boundary, and the one not present is considered to be outside the boundary. If the reference pixel position exists outside the boundary, the pixel value at that position cannot actually be referred to, so it is necessary to perform padding processing. Specifically, as shown in the table shown in FIG. 15, the padding process can be realized by changing the actual reference pixel position according to the pattern of the boundary and the arrangement of the filter coefficients.

FIG. 14 (e) shows an example when padding is not required. The values of offset1 to offset3 when padding is not required are as shown in the table shown in FIG.

On the other hand, for example, in the arrangement shown in FIG. 14C, since the pixels corresponding to the filter coefficients C4 to C7 are outside the boundary, the corresponding values of offset2 and offset3 can be set to "0". Filtering can be realized.

Further, in the arrangement of FIG. 14C, the pixel corresponding to the filter coefficient C0 can be referred to because it is located inside the boundary. When the same method as the padding method (mirror padding) performed in the part 258B is used, the filter coefficient C0 side also performs padding, so that offset1 is also set to "0".

Mirror padding is a method of padding a line that appears outside the boundary when the filter coefficient is folded and inverted around the horizontal line in which the filter coefficient C2 exists in FIG. 14 (c).

In the case of FIG. 14B, since the line corresponding to offset1 is outside the boundary, offset1 is set to "0", and offset2 and offset3 are also set to "0" according to the mirror padding method.

In the case of FIG. 14D, since the line corresponding to offset3 is outside the boundary, offset1 and offset2 are the same as in the case of FIG. 14E, and only offset3 is changed to “1”.

Also in the case of FIG. 14A, when the filter coefficient arrangement is folded around the line where the filter coefficient C2 exists, the line corresponding to the filter coefficient C7 appears outside the boundary. Therefore, when performing mirror padding, the offset 3 is set to 1. Correct to and perform padding processing.

On the other hand, in the case of FIG. 14A, padding is not always necessary because there is actually no filter coefficient referencing the outside of the boundary. Therefore, in the case of FIG. 14A, the offset value can be set in the same manner as in FIG. 14E. In other words, if the padding process in the case of FIG. 14 (a) is not defined and the padding process is not defined, the setting in FIG. 14 (e) may be used.

As described above, in the CC-ALF unit that takes the luminance signal as an input and adds the output to the luminance signal, the luminance signal pixel corresponding to the luminance signal pixel to which the output value is added (luminance signal pixel position corresponding to the above coefficient C2). When is one line below the boundary ((b) above) or one line above the boundary ((c) above), the padding process is performed by setting all the offsets to be added to the reference pixel position to "0". By configuring, for example, in Non-Patent Document 1, it is possible to perform the padding process by the same method as the padding method (mirror padding) performed in the luminance ALF section 256, Cb ALF section 258A, and Cr ALF section 258B. Therefore, the control logic can be shared.

Further, as described above, in the CC-ALF unit that takes the luminance signal as an input and adds the output to the color difference signal, when all the reference pixel positions are inside the boundary, the padding process is not performed and at least the reference pixel positions are at least. The padding process may be performed only when one is outside the boundary.

That is, by not defining the padding process in the case of FIG. 14A, unnecessary padding process can be eliminated and the coding efficiency can be improved.

(Image Coding Device 100)
Hereinafter, the image coding apparatus 100 according to the present embodiment will be described with reference to FIG. FIG. 16 is a diagram showing an example of a functional block of the image coding apparatus 100 according to the present embodiment.

As shown in FIG. 16, the image coding apparatus 100 includes an inter-prediction unit 111, an intra-prediction unit 112, a subtractor 121, an adder 122, a conversion / quantization unit 131, and an inverse conversion / dequantization. It has a unit 132, a coding unit 140, an in-loop filter processing unit 150, and a frame buffer 160.

The inter-prediction unit 111 is configured to generate a prediction signal by inter-prediction (inter-frame prediction), similarly to the inter-prediction unit 241.

Specifically, the inter-prediction unit 111 identifies and identifies the reference block included in the reference frame by comparing the frame to be encoded (hereinafter referred to as the target frame) with the reference frame stored in the frame buffer 160. It is configured to determine the motion vector for the reference block.

Further, the inter-prediction unit 111 is configured to generate a prediction signal included in the prediction block for each prediction block based on the reference block and the motion vector. The inter-prediction unit 111 is configured to output a prediction signal to the subtractor 121 and the adder 122. Here, the reference frame is a frame different from the target frame.

Like the intra prediction unit 242, the intra prediction unit 112 is configured to generate a prediction signal by intra prediction (in-frame prediction).

Specifically, the intra prediction unit 112 is configured to specify a reference block included in the target frame and generate a prediction signal for each prediction block based on the specified reference block. Further, the intra prediction unit 112 is configured to output a prediction signal to the subtractor 121 and the adder 122.

Here, the reference block is a block referred to for the block to be predicted (hereinafter referred to as the target block). For example, the reference block is a block adjacent to the target block.

The subtractor 121 is configured to subtract the prediction signal from the input image signal and output the prediction residual signal to the conversion / quantization unit 131. Here, the subtractor 121 is configured to generate a prediction residual signal, which is the difference between the prediction signal generated by the intra prediction or the inter prediction and the input image signal.

The adder 122 adds a prediction signal to the prediction residual signal output from the inverse conversion / inverse quantization unit 132 to generate a pre-filter processing decoding signal, and the pre-filter processing decoding signal is used by the intra prediction unit 112 and the input. It is configured to output to the loop filter processing unit 150.

Here, the pre-filtered decoding signal constitutes a reference block used by the intra prediction unit 112.

The conversion / quantization unit 131 is configured to perform conversion processing of the predicted residual signal and acquire a coefficient level value. Further, the conversion / quantization unit 131 may be configured to quantize the coefficient level value.

Here, the conversion process is a process of converting the predicted residual signal into a frequency component signal. In such a conversion process, a base pattern (transformation matrix) corresponding to a discrete cosine transform (DCT) may be used, or a base pattern (transformation matrix) corresponding to a discrete sine transform (DST). May be used.

The inverse conversion / inverse quantization unit 132 is configured to perform an inverse conversion process of the coefficient level value output from the conversion / quantization unit 131, similarly to the inverse conversion / inverse quantization unit 220. Here, the inverse transformation / inverse quantization unit 132 may be configured to perform inverse quantization of the coefficient level value prior to the inverse transformation process.

Here, the inverse conversion process and the inverse quantization are performed in the reverse procedure of the conversion process and the quantization performed by the conversion / quantization unit 131.

The coding unit 140 is configured to encode the coefficient level value output from the conversion / quantization unit 131 and output the coded data.

Here, for example, coding is entropy coding in which codes of different lengths are assigned based on the probability of occurrence of a coefficient level value.

Further, the coding unit 140 is configured to encode the control data used in the decoding process in addition to the coefficient level value.

As described above, the control data may include size data such as a coded block size, a predicted block size, and a conversion block size.

Similar to the in-loop filter processing unit 250, the in-loop filter processing unit 150 performs filter processing on the pre-filter processing decoded signal output from the adder 122, and outputs the post-filter processing decoded signal to the frame buffer 160. It is configured to do.

The frame buffer 160 is configured to store reference frames used by the inter-prediction unit 111.

Here, the filtered decoded signal constitutes a reference frame used by the inter-prediction unit 111.

(Intra Prediction Unit 112)
Hereinafter, the intra prediction unit 112 according to the present embodiment will be described with reference to FIG. 17. FIG. 17 is a diagram showing an example of a functional block of the intra prediction unit 112 of the image coding device 112 according to the present embodiment.

As shown in FIG. 17, the intra prediction unit 112 includes an intra prediction mode determination unit 112A, an intra prediction mode coding unit 112B, and a prediction signal generation unit 112C.

The intra prediction unit 112 is an example of a prediction unit configured to generate a prediction signal by intra prediction (in-frame prediction).

The intra prediction mode determination unit 112A is configured to determine information to be encoded by the subsequent intra prediction mode coding unit 112B, such as the intra prediction mode of the block and whether or not BDPCM is applied. As the determination method, a known method can be used, and details thereof will be omitted.

The intra prediction mode coding unit 112B is configured to encode the prediction mode of the target block. The processing content is the same as that of the intra prediction mode decoding unit 242B. Specifically, for example, in the processing flowchart of the intra prediction mode decoding unit 242B shown in FIG. 6, the “decoding” part is changed to “encoding”.

The prediction signal generation unit 112C is configured to generate a prediction signal of the block according to the prediction mode encoded by the intra prediction mode 112B. The method of generating the prediction signal is the same as that of the prediction signal generation unit 242C.

Further, the above-mentioned image coding device 100 and image decoding device 200 may be realized by a program that causes a computer to execute each function (each step).

In each of the above embodiments, the present invention has been described by using the application to the image coding device 100 and the image decoding device 200 as an example, but the present invention is not limited to such an example, and the image code is not limited to the above. It can be similarly applied to an image coding / decoding system having each function of the computer 100 and the image decoding device 200.

According to the present invention, it is possible to provide an image decoding device 200, an image decoding method, and a program having a function of ensuring that a specific block in a frame is lossless.

10 ... Image processing system 100 ... Image coding device 111, 241 ... Inter prediction unit 112, 242 ... Intra prediction unit 112A ... Intra prediction mode determination unit 112B ... Intra prediction mode coding unit 112C, 242C ... Prediction signal generation unit 121 ... Subtractors 122, 230 ... Adder 131 ... Conversion / quantization unit 132, 220 ... Inverse conversion / inverse quantization unit 140 ... Coding unit 150, 250 ... In-loop filter processing unit 160, 260 ... Frame buffer 200 ... Image decoding Device 210 ... Decoding unit 220A ... Quantization parameter derivation unit 220B ... Inverse quantization unit 220C ... Inverse conversion unit 242B ... Intra prediction mode decoding unit 250A ... Deblocking filter unit 251, 251A, 251B ... Block boundary detection unit 253, 253A, 253B ... Boundary strength determination unit 254, 254A, 254B ... Filter determination unit 255, 255A, 255B ... Filter processing unit 250B ... SAO unit 250C ... Adaptive loop filter unit 256 ... Brightness ALF unit 257A ... Cb CC-ALF unit 257B ... Cr CC -ALF part 258A ... Cb-ALF part 258B ... Cr-ALF part

Claims (3)

It is an image decoder
When the conversion skip mode is applied to the target block, the value of the quantization parameter of the target block is the quantum when the conversion skip mode is applied, regardless of whether or not the adaptive color conversion is applied to the target block. An image decoding apparatus including a quantization parameter derivation unit configured to have a minimum value of a quantization parameter or more and a minimum value of a quantization parameter when the conversion skip mode is applied to be "4" or more. .. When the conversion skip mode is applied to the target block, the value of the quantization parameter of the target block is the quantum when the conversion skip mode is applied, regardless of whether or not the adaptive color conversion is applied to the target block. An image decoding method comprising a step of setting the minimum value of the quantization parameter or more and setting the minimum value of the quantization parameter when the conversion skip mode is applied to "4" or more. A program used in an image decoder, which can be used on a computer.
When the conversion skip mode is applied to the target block, the value of the quantization parameter of the target block is the quantum when the conversion skip mode is applied, regardless of whether or not the adaptive color conversion is applied to the target block. A program characterized in that a step of setting the minimum value of the quantization parameter to "4" or more and setting the minimum value of the quantization parameter when the conversion skip mode is applied to "4" or more is executed.

Images