JP2021093635A - Coding device, decoding device and program - Google Patents

Abstract

To enable a conversion basio which is more suitable to a feature of a prediction residual to be applied, so as to improve coding efficiency, even when a flag is not transmitted in order to switch the conversion basio.SOLUTION: A coding device that divides an image into blocks and codes the blocks, comprises: an intra-prediction part that predicts by intra-prediction a block to be coded, using a reference pixel row selected from a plurality of reference pixel rows including a reference pixel row adjacent to the block to be coded and a reference pixel row not adjacent to the block to be coded; and a conversion part that outputs a conversion coefficient by performing conversion processing for a prediction residual representing a difference between a prediction block outputted by the intra-prediction part and the block to be coded. The conversion part has a conversion basio determining part that determines a conversion basio that is used in the conversion processing, on the basis of the reference pixel row selected out of the plurality of reference pixel rows.SELECTED DRAWING: Figure 6

Description

The present invention relates to a coding device, a decoding device, and a program.

Research on video coding methods is being conducted to compress the amount of data during transmission or storage of still images and moving images. In recent years, ultra-high resolution images such as 8K-SHV have become widespread, and AVC / H.A. 264 and HEVC / H. Coding methods such as 265 are known.

In such a coding technique, intra-prediction using spatial correlation within a frame is used. In intra-prediction, a prediction block is generated by predicting each pixel in the coding target block by referring to the decoded reference pixels adjacent to the coding target block. The coding device performs conversion processing on the predicted residual representing the difference (error) between the coded block and the predicted block to generate a conversion coefficient, and quantizes and entropy-codes the conversion coefficient to output the coded data. To do.

In the draft standard of VVC (Versatile Video Coding), which is a next-generation video coding method jointly standardized by MPEG (Moving Picture Experts Group) and ITU (International Telecommunication Union), one can be selected from a plurality of reference pixels. MRL (Multi Reference Line) intra-prediction is adopted, which is used for intra-prediction by selecting (see, for example, Non-Patent Document 1). The reference pixel string of the plurality of options includes a reference pixel string adjacent to the coded target block and a reference pixel string not adjacent to the coded target block.

Further, in the draft standard of VVC, MTS (Multiple transformation) selection that selectively applies a total of three conversion bases of DCT (Discrete Cosine Transform) 2, DST (Discrete Sine Transform) 7, and DCT8 for each block to be encoded. Is adopted, and it is possible to perform a conversion more suitable for the characteristics of the predicted residual. Regarding the switching of the conversion basis, a function of transmitting and switching the flag and a function of switching according to the block size of the coded block without transmitting the flag can be used.

JVET-P2001 "Versatile Video Coding (Draft 7)"

As described above, when a flag is transmitted to switch the conversion basis, the code amount for the flag increases and the coding efficiency decreases, and the number of trials for determining an appropriate conversion basis on the coding apparatus side is performed. Will increase, leading to an increase in the amount of processing.

On the other hand, there is also a function to switch the conversion base according to the block size of the block to be coded without transmitting the flag, but it is difficult to apply the conversion base suitable for the characteristics of the predicted residual, and the coding efficiency is improved. Realization is difficult.

Therefore, the present invention provides a coding device and a decoding device that improve coding efficiency by making it possible to apply a conversion basis more suitable for the characteristics of the predicted residual even when the flag is not transmitted to switch the conversion basis. The purpose is to provide equipment and programs.

The coding device according to the first aspect is a device that divides an image into blocks and encodes the image. The coding apparatus uses a reference pixel string selected from a plurality of reference pixel strings including a reference pixel string adjacent to the coded target block and a reference pixel string not adjacent to the coded target block. A conversion that outputs a conversion coefficient by performing conversion processing on an intra prediction unit that predicts a coded target block by intra prediction and a prediction residual that represents a difference between the prediction block output by the intra prediction unit and the coded target block. It has a part. The conversion unit has a conversion base determination unit that determines a conversion base used in the conversion process based on the reference pixel sequence selected from the plurality of reference pixel sequences.

The decoding device according to the second aspect includes an entropy decoding unit that decodes the encoded data and outputs a quantization conversion coefficient corresponding to the decoding target block, a reference pixel string adjacent to the decoding target block, and the decoding target. An intra prediction unit that predicts the decoding target block by intra prediction and an entropy decoding unit output using a reference pixel string selected from a plurality of reference pixel strings including a reference pixel string that is not adjacent to the block. An inverse quantization unit that performs an inverse quantization process on the quantization conversion coefficient and outputs a conversion coefficient, and an inverse conversion process that performs an inverse conversion process on the conversion coefficient output by the inverse quantization unit and outputs a predicted residual. It has a part. The inverse transformation unit has a transformation base determination unit that determines a conversion basis to be used in the inverse transformation process based on the reference pixel sequence selected from the plurality of reference pixel sequences.

The program according to the third aspect causes the computer to function as a coding device according to the first aspect or a decoding device according to the second aspect.

According to the present invention, a coding device for improving coding efficiency by making it possible to apply a conversion basis more suitable for the characteristics of the predicted residual even when the flag is not transmitted to switch the conversion basis. Equipment and programs can be provided.

It is a figure which shows the structure of the coding apparatus which concerns on embodiment. It is a figure which shows the total 3 conversion bases of DCT2 (DCT-II), DCT8 (DCT-VIII), and DST7 (DST-VII). It is a figure which shows the conversion basis information output by the conversion basis determination part which concerns on embodiment. It is a figure which shows the candidate of the intra prediction mode which concerns on embodiment. It is a figure which shows the MRL intra prediction which concerns on embodiment. It is a figure which shows the operation flow of the coding apparatus which concerns on embodiment. It is a figure which shows the structure of the decoding apparatus which concerns on embodiment. It is a figure which shows the operation flow of the decoding apparatus which concerns on embodiment.

The coding device and the decoding device according to the embodiment will be described with reference to the drawings. The coding device and the decoding device according to the embodiment encode and decode a moving image represented by MPEG, respectively. In the description of the drawings below, the same or similar parts are designated by the same or similar reference numerals.

<Coding device>
First, the configuration of the coding apparatus according to the present embodiment will be described. FIG. 1 is a diagram showing a configuration of a coding device 1 according to the present embodiment.

As shown in FIG. 1, the coding apparatus 1 includes a block dividing unit 100, a prediction residual generation unit 110, a conversion / quantization unit 120, an entropy coding unit 130, a scan control unit 131, and an inverse quantum. It has a quantization / inverse conversion unit 140, a synthesis unit 150, a memory 160, and a prediction unit 170.

The block division unit 100 divides the original image, which is an input image for each frame (or picture) constituting the moving image, into a plurality of image blocks, and outputs the image block obtained by the division to the prediction residual generation unit 110. The size of the image block is, for example, 32 × 32 pixels, 16 × 16 pixels, 8 × 8 pixels, 4 × 4 pixels, or the like. The shape of the image block is not limited to a square and may be a rectangle (non-square). The image block is a unit for coding by the coding device 1 (that is, a block to be coded) and a unit for decoding by the decoding device (that is, a block to be decoded). Such an image block is sometimes called a CU (Coding Unit).

For example, the block division unit 100 outputs a luminance block and a luminance block by performing block partitioning on the luminance signal and the color difference signal constituting the image. The division may be independently controllable for the luminance signal and the color difference signal. In the following, when the luminance block and the color difference block are not particularly distinguished, they are simply referred to as encoding target blocks.

The prediction residual generation unit 110 calculates a prediction residual representing a difference (error) between the coding target block output by the block division unit 100 and the prediction block obtained by predicting the coding target block by the prediction unit 170. To do. Specifically, the prediction residual generation unit 110 calculates the prediction residual by subtracting each pixel value of the prediction block from each pixel value of the block, and converts the calculated prediction residual into the conversion / quantization unit 120. Output.

The conversion / quantization unit 120 performs conversion processing and quantization processing on a block-by-block basis. The conversion / quantization unit 120 includes a conversion unit 121 and a quantization unit 122.

The conversion unit 121 performs conversion processing on the prediction residual output by the prediction residual generation unit 110 to calculate a conversion coefficient, and outputs the calculated conversion coefficient to the quantization unit 122. The conversion process refers to a process of converting the predicted residual from the spatial domain to the frequency domain, and includes, for example, the discrete cosine transform (DCT), the discrete sine transform (DST), and the carunen-reve transform (KLT). However, the conversion process may include a conversion skip that adjusts by scaling or the like without converting the signal in the pixel region into the frequency domain. The conversion unit 121 outputs information regarding the conversion process applied to the coded target block to the entropy coding unit 130.

The conversion unit 121 uses an MTS that switches a plurality of conversion bases for each coded block in the coding of the luminance signal. The conversion unit 121 has a conversion base determination unit 121a. The conversion basis determination unit 121a determines the conversion basis (conversion type) from DCT-2, DST-7, and DCT-8 in each of the horizontal direction and the vertical direction. That is, the conversion unit 121 switches and applies a plurality of types of conversion processing. The conversion unit 121 outputs the conversion base information indicating the determined conversion base to the entropy coding unit 130 and the inverse conversion unit 142.

FIG. 2 is a diagram showing a total of three conversion bases (Bases function) of DCT2 (DCT-II), DCT8 (DCT-VIII), and DST7 (DST-VII). In FIG. 2, N represents the size of the coded block. DCT8 includes a basis in which the impulse response is monotonically reduced. Specifically, DCT8 is a _{conversion in which the impulse response T 0} (j) of the lowest frequency filter is monotonically reduced (however, j = 0, ..., N-1). DCT8 has a large value at one end of the converted basis waveform and is open. DST7 includes a basis in which the impulse response is monotonically increasing. Specifically, DST7 is a _{conversion in which the impulse response T 0} (j) of the lowest frequency filter is monotonically increased (however, j = 0, ..., N-1). In DST7, one end of the converted basis waveform is closed.

In the present embodiment, three types of conversion processes to be applied to the predicted residuals, DCT2, DCT8, and DST7, will be described as an example, but the conversion including the basis having the characteristics of monotonic increase and monotonous decrease as described above will be described. It is not limited to these three transformation bases as long as it is selectively switched and applied. For example, other DCT / DST such as DCT1 and DCT5 may be used, or a transformation such as a discrete wavelet transform may be used.

FIG. 3 is a diagram showing conversion basis information output by the conversion basis determination unit 121a according to the present embodiment.

As shown in FIG. 3, the conversion basis determination unit 121a outputs a total of three conversion flags of MTS_CU_flag, MTS_Hor_flag, and MTS_Ver_flag as conversion basis information.

When DCT2 is applied to both the horizontal conversion process and the vertical conversion process, the conversion basis determination unit 121a sets MTS_CU_flag = 0. On the other hand, when DCT8 or DST7 is applied to at least one of the horizontal conversion process and the vertical conversion process, the conversion basis determination unit 121a determines MTS_Hor_flag and MTS_Ver_flag according to the combination of conversion bases applied to the horizontal conversion process and the vertical conversion process. To set.

These three conversion flags are subjected to entropy coding processing by the entropy coding unit 130 described later and output as a stream. However, when MTS_CU_flag = 0, the entropy encoding unit 130 does not have to stream output MTS_Hor_flag and MTS_Ver_flag. Further, as will be described in detail later, the entropy coding unit 130 does not have to stream output the three conversion flags when MRL is applied to the coded block.

The quantization unit 122 quantizes the conversion coefficient output by the conversion unit 121 using the quantization parameter (Qp) and the quantization matrix, and the quantization conversion coefficient, which is the quantized conversion coefficient, is converted to the entropy coding unit 130 and the inverse. It is output to the quantization / inverse conversion unit 140.

The entropy coding unit 130 performs entropy coding on the quantization conversion coefficient output by the quantization unit 122, performs data compression to generate coded data (bit stream), and encodes the coded data. Output to the outside of 1. For entropy coding, Huffman coding, CABAC (Context-based Adaptive Binary Arithmetic Coding), or the like can be used.

Further, the entropy coding unit 130 inputs information related to the conversion process from the conversion unit 121, and transmits the information related to the conversion process to the decoding side in the coded data. The information regarding the conversion process may include the above-mentioned conversion basis information.

Further, the entropy encoding unit 130 inputs information on the prediction process from the prediction unit 170, and transmits the input information on the prediction process to the decoding side in the encoded data. The information regarding the prediction process may include reference pixel string information described later.

The inverse quantization / inverse transformation unit 140 performs the inverse quantization process and the inverse transformation process on a block-by-block basis. The inverse quantization / inverse conversion unit 140 includes an inverse quantization unit 141 and an inverse conversion unit 142.

The inverse quantization unit 141 performs an inverse quantization process corresponding to the quantization process performed by the quantization unit 122. Specifically, the inverse quantization unit 141 restores the conversion coefficient by inversely quantizing the quantization conversion coefficient output by the quantization unit 122 using the quantization parameter (Qp) and the quantization matrix. The restored conversion coefficient is output to the inverse conversion unit 142.

The inverse conversion unit 142 performs an inverse transformation process corresponding to the conversion process performed by the conversion unit 121 based on the conversion base information output by the conversion unit 121. For example, when the conversion unit 121 performs the discrete cosine transform, the inverse transform unit 142 performs the inverse discrete cosine transform. The inverse transformation unit 142 performs an inverse transformation process on the conversion coefficient output by the inverse quantization unit 141 to restore the predicted residual, and outputs the restored predicted residual, which is the restored predicted residual, to the synthesis unit 150. ..

The synthesizing unit 150 synthesizes the restoration prediction residual output by the inverse conversion unit 142 with the prediction block output by the prediction unit 170 on a pixel-by-pixel basis. The synthesizing unit 150 adds each pixel value of the restoration prediction residual and each pixel value of the prediction block to decode (reconstruct) the coded block, and outputs the decoded block to the memory 160. The decrypted block is sometimes called a reconstructed block.

The memory 160 stores the decoded blocks output by the compositing unit 150, and stores the decoded blocks as a decoded image in frame units. The memory 160 outputs the stored decoded block or decoded image to the prediction unit 170. A loop filter may be interposed between the compositing unit 150 and the memory 160.

The prediction unit 170 performs prediction processing in block units. The prediction unit 170 includes an inter-prediction unit 171, an intra-prediction unit 172, and a switching unit 173.

The inter-prediction unit 171 performs inter-prediction using the correlation between frames. Specifically, the inter-prediction unit 171 uses the decoded image stored in the memory 160 as a reference image, calculates a motion vector by a method such as block matching, predicts the coded block, and predicts the inter-prediction block. Is generated, and the generated inter-prediction block is output to the switching unit 173. Here, the inter-prediction unit 171 selects the optimum inter-prediction method from inter-prediction using a plurality of reference images (typically bi-prediction) and inter-prediction using one reference image (one-way prediction). Select and perform inter-prediction using the selected inter-prediction method. The inter-prediction unit 171 outputs information (motion vector, etc.) related to the inter-prediction to the entropy coding unit 130.

The intra prediction unit 172 performs intra prediction using the spatial correlation in the frame. Specifically, the intra prediction unit 172 generates an intra prediction block by referring to the decoded pixels around the coded block among the decoded images stored in the memory 160, and the generated intra prediction block. Is output to the switching unit 173. The intra prediction unit 172 selects an intra prediction mode to be applied to the coded target block from a plurality of intra prediction modes, and predicts the target block using the selected intra prediction mode.

FIG. 4 is a diagram showing candidates for the intra prediction mode according to the present embodiment. As shown in FIG. 4, there are 67 candidates for the intra prediction mode for the luminance block from 0 to 66. The mode "0" of the intra prediction mode is Planar prediction, the mode "1" of the intra prediction mode is DC prediction, and the modes "2" to "66" of the intra prediction mode are directional prediction. In the directionality prediction, the direction of the arrow indicates the reference direction, the starting point of the arrow indicates the position of the pixel to be predicted, and the end point of the arrow indicates the position of the reference pixel used to predict the pixel to be predicted.

In the present embodiment, the intra prediction unit 172 has a function of performing MRL intra prediction. FIG. 5 is a diagram showing an MRL intra-prediction according to the present embodiment. Each circle shown in FIG. 5 represents a pixel.

As shown in FIG. 5, in the MRL intra-prediction, in the intra-prediction, the reference pixel string of any one of the plurality of reference pixel strings is used for the prediction to generate a prediction block, and the syntax indicating the reference pixel sequence is generated. This is a method of transmitting to the decoding side.

In FIG. 5, a plurality of reference pixel strings are an adjacent reference pixel string (0th column) adjacent to the coded block and a first reference pixel string (1st column) located outside the adjacent reference pixel string. , An example including a second reference pixel string (second column) located outside the first reference pixel string is shown.

However, for the reference pixel string located outside the adjacent reference pixel string, the reference pixel string farther than the second column may be used. For example, the adjacent reference pixel string in the 0th column is defined as the adjacent reference pixel string, and the reference pixel string in the 1st column located outside the reference pixel string in the 0th column is defined as the 1st reference pixel string in the 1st column. The third reference pixel string located outside the reference pixel string may be used as the second reference pixel string.

In the conventional intra-prediction, each pixel in the coding target block is predicted by using the pixel adjacent to the coding target block as a reference pixel, and the prediction accuracy for the prediction target pixel adjacent to the reference pixel is high. On the other hand, in the conventional intra-prediction, the prediction efficiency of the prediction target pixel at a distance from the reference pixel gradually decreases, and the prediction accuracy of the pixel in the coding target block is not necessarily high depending on the pattern on average. There are drawbacks. Focusing on this phenomenon, in MRL intra-prediction, in order to improve the average prediction accuracy of the coded block, a pixel sequence separated by several lines is used as a reference pixel instead of the pixel adjacent to the coded block. It is possible to do.

For example, the intra-prediction unit 172 applies the MRL intra-prediction when applying the MPM (Most Problem Modes) to the intra-prediction. Since the prediction modes used in the coded block and its surrounding blocks are highly correlated, in MPM, the prediction modes used in the blocks to the left and above the coded block and the directions close to those prediction modes. Prioritize prediction modes and allocate a relatively small amount of data relative to other prediction modes. In addition, MRL intra-prediction is applied to prediction modes other than Planar prediction and DC prediction. That is, the Planar prediction and the DC prediction refer only to the adjacent reference pixel sequence.

The intra prediction unit 172 includes a reference pixel sequence selection unit 172a. When the MRL intra prediction is applied, the reference pixel string selection unit 172a of the adjacent reference pixel string (0th column), the 1st reference pixel string (1st column), and the 2nd reference pixel string (2nd column). From the list, one reference pixel sequence used for intra prediction is selected. As described above, the second reference pixel string may be the third reference pixel string. The reference pixel string selection unit 172a outputs reference pixel string information (reference pixel sequence syntax) indicating the selected reference pixel sequence to the conversion unit 121 and the entropy coding unit 130. Further, the intra prediction unit 172 outputs mode information (prediction mode syntax) indicating the intra prediction mode to the entropy coding unit 130.

The switching unit 173 switches between the inter prediction block output by the inter prediction unit 171 and the intra prediction block output by the intra prediction unit 172, and outputs one of the prediction blocks to the prediction residual generation unit 110 and the synthesis unit 150. ..

As described above, the coding device 1 according to the present embodiment is selected from a plurality of reference pixel strings including a reference pixel string adjacent to the coding target block and a reference pixel string not adjacent to the coding target block. Using the reference pixel sequence, conversion processing is performed on the prediction residual that represents the difference between the intra prediction unit 172 that predicts the coded target block by intra prediction and the prediction block output by the intra prediction unit 172 and the coded target block. It has a conversion unit 121 that outputs a conversion coefficient.

In the present embodiment, the conversion basis determination unit 121a of the conversion unit 121 is used in the conversion process based on the reference pixel sequence selected from the plurality of reference pixel sequences when the intra prediction unit 172 performs the MRL intra prediction. Determine the transformation basis. Specifically, the conversion base determination unit 121a specifies the reference pixel string from the reference pixel string information output by the reference pixel string selection unit 172a, and determines the conversion base associated with the specified reference pixel string.

Here, the conversion base determination unit 121a is based on a reference pixel string selected from a plurality of reference pixel strings, and is used in the horizontal conversion process in the horizontal direction and the vertical direction conversion used in the conversion process in the vertical direction. Determine the basis.

In this way, the conversion base determination unit 121a of the conversion unit 121 determines the conversion base based on the reference pixel string selected in the MRL intra-prediction. That is, by associating the reference pixel sequence with the conversion base, it is not necessary to transmit the flag to the decoding side in order to switch the conversion base. Specifically, the entropy encoding unit 130 transmits the reference pixel string information indicating the selected reference pixel string to the decoding side without transmitting the conversion base information indicating the conversion basis. Therefore, it is possible to reduce the amount of coding for the flag for switching the conversion basis, improve the coding efficiency, and reduce the amount of processing (that is, the number of trials) for determining an appropriate conversion basis.

Table 1 shows an example of the correspondence between the reference pixel sequence and the conversion basis.

In the example shown in Table 1, when the reference pixel string selected by the MRL intra-prediction is the reference pixel string of the 0th column (see FIG. 5), the conversion basis determination unit 121a uses the DST-7 as the horizontal conversion basis and Determined as the vertical transformation basis. When intra-prediction is performed using the reference pixel string of the 0th column, it is considered that the prediction accuracy of the block boundary region adjacent to the reference pixel string of the 0th column is high and the prediction residual of the block boundary region is close to zero. DST7 has one end of the transformed basis waveform closed and is suitable for the predicted residuals having such characteristics. However, the reference pixel sequence in Table 1 is an example, and for the reference pixel string located outside the adjacent reference pixel string, a reference pixel string (for example, the third column) farther than the second column may be used. Absent.

When the reference pixel string selected in the MRL intra-prediction is the reference pixel string of the first column (see FIG. 5), the conversion basis determination unit 121a determines DCT-2 as the horizontal conversion basis and the vertical conversion basis. .. When the intra prediction is performed using the reference pixel sequence of the first row, the DCT-2 is used because the intra prediction is performed so as to improve the average prediction accuracy of the block.

When the reference pixel sequence selected in the MRL intra-prediction is the second reference pixel sequence (see FIG. 5), the conversion basis determination unit 121a determines DST-7 as the horizontal conversion basis and the vertical conversion basis. .. When intra-prediction is performed using the reference pixel sequence of the second column, DST-7 is used because the characteristics of the prediction residual are undefined. However, DCT-2 may be used instead of DST-7.

Next, the operation flow of the coding device 1 according to the present embodiment will be described. FIG. 6 is a diagram showing an operation flow of the coding device 1 according to the present embodiment. Here, when applying the MRL intra-prediction, only the operation related to the determination of the transformation basis is shown.

As shown in FIG. 6, in step S11, the reference pixel string selection unit 172a of the intra prediction unit 172 includes the adjacent reference pixel string (0th column), the 1st reference pixel string (1st column), and the 2nd reference pixel. From the columns (second column), one reference pixel string used for intra-prediction of the coded block is selected. The reference pixel string selection unit 172a outputs reference pixel string information (reference pixel sequence syntax) indicating the selected reference pixel sequence to the conversion unit 121 and the entropy coding unit 130.

In step S12, the intra prediction unit 172 performs intra prediction of the coded target block with reference to the reference pixel string selected in step S11, and outputs the prediction block. Here, the intra prediction unit 172 may perform intra prediction using the intra prediction mode in the MPM.

In step S13, the prediction residual generation unit 110 generates a prediction residual representing the difference between the prediction block output by the intra prediction unit 172 and the coded target block, and outputs the prediction residual to the conversion unit 121.

On the other hand, in step S14, the conversion base determination unit 121a of the conversion unit 121 identifies the reference pixel string from the reference pixel string information output by the reference pixel string selection unit 172a, and the conversion base associated with the specified reference pixel string. To determine. For example, the conversion basis determination unit 121a determines the conversion basis from the number of the reference pixel string according to the correspondence shown in Table 1.

In step S15, the conversion unit 121 performs conversion processing on the predicted residuals using the conversion basis determined by the conversion basis determination unit 121a, and outputs a conversion coefficient. The quantization unit 122 performs a quantization process on the conversion coefficient output by the conversion unit 121, and outputs the quantization conversion coefficient.

In step S16, the entropy coding unit 130 encodes the quantization conversion coefficient output by the quantization unit 122 and outputs the coded data. Here, the entropy encoding unit 130 transmits the reference pixel string information output by the intra prediction unit 172 (reference pixel string selection unit 172a) to the decoding side without transmitting the conversion base information output by the conversion unit 121. ..

<Decoding device>
Next, the configuration of the decoding device according to the present embodiment will be mainly described as being different from the coding device 1. FIG. 7 is a diagram showing a configuration of a decoding device 2 according to the present embodiment.

As shown in FIG. 7, the decoding device 2 includes an entropy decoding unit 200, an inverse quantization / inverse conversion unit 210, a synthesis unit 220, a memory 230, and a prediction unit 240.

The entropy decoding unit 200 decodes the encoded data and outputs the quantization conversion coefficient corresponding to the decoding target block to the inverse quantization / inverse conversion unit 210.

Further, the entropy decoding unit 200 acquires information related to the conversion process and outputs the information related to the conversion process to the inverse quantization / inverse conversion unit 210 (inverse conversion unit 212). The information regarding the conversion process may include the above-mentioned conversion basis information. However, when the MRL intra-prediction is applied, the entropy decoding unit 200 does not acquire the conversion basis information.

Further, the entropy decoding unit 200 acquires information on the prediction process and outputs the information on the prediction process to the prediction unit 240. The information regarding the prediction process may include the reference pixel string information described above. The entropy decoding unit 200 outputs the reference pixel sequence information to the inverse quantization / inverse transformation unit 210 (inverse transformation unit 212).

The inverse quantization / inverse transformation unit 210 performs the inverse quantization process and the inverse transformation process on a block-by-block basis. The inverse quantization / inverse conversion unit 210 includes an inverse quantization unit 211 and an inverse conversion unit 212.

The inverse quantization unit 211 performs an inverse quantization process corresponding to the quantization process performed by the quantization unit 122 of the coding apparatus 1. The inverse quantization unit 211 restores the conversion coefficient of the decoding target block by inversely quantizing the quantization conversion coefficient output by the entropy decoding unit 200 using the quantization parameter (Qp) and the quantization matrix. The restored conversion coefficient is output to the inverse conversion unit 212.

The inverse conversion unit 212 performs an inverse conversion process corresponding to the conversion process performed by the conversion unit 121 of the coding apparatus 1. The inverse conversion unit 212 performs inverse transformation processing on the conversion coefficient output by the inverse quantization unit 211 to restore the predicted residual, and outputs the restored predicted residual (restored predicted residual) to the synthesis unit 220. ..

The inverse transformation unit 212 has a conversion base determination unit 212a. When the MRL intra-prediction is not applied, the conversion basis determination unit 212a determines the conversion basis to be used in the inverse quantization process based on the conversion basis information output by the entropy decoding unit 200 (see FIG. 3).

On the other hand, when the MRL intra-prediction is applied, the conversion basis determination unit 212a determines the conversion basis to be used in the inverse quantization process based on the reference pixel string information output by the entropy decoding unit 200 (see Table 1). .. Specifically, the conversion base determination unit 121a identifies the reference pixel string from the reference pixel string information, and determines the conversion basis associated with the specified reference pixel string. Here, the conversion basis determination unit 121a determines the horizontal transformation basis used in the horizontal inverse transformation process and the vertical transformation basis used in the vertical inverse transformation process based on the selected reference pixel sequence.

The synthesis unit 220 decodes (reconstructs) the original block by synthesizing the prediction residual output by the inverse conversion unit 212 and the prediction block output by the prediction unit 240 on a pixel-by-pixel basis, and decodes the decoded block. Output to memory 230.

The memory 230 stores the decoded blocks output by the compositing unit 220, and stores the decoded blocks as a decoded image in frame units. The memory 230 outputs the decoded block or the decoded image to the prediction unit 240. Further, the memory 230 outputs the decoded image in frame units to the outside of the decoding device 2. A loop filter may be interposed between the compositing unit 220 and the memory 230.

The prediction unit 240 makes a prediction in block units. The prediction unit 240 has an inter prediction unit 241, an intra prediction unit 242, and a switching unit 243.

The inter-prediction unit 241 performs inter-prediction using the correlation between frames. Specifically, the inter-prediction unit 241 encodes the decoded image stored in the memory 230 as a reference image based on the information related to the inter-prediction (for example, motion vector information) output by the entropy decoding unit 200. The target block is predicted, an inter-prediction block is generated, and the generated inter-prediction block is output to the switching unit 243.

The intra prediction unit 242 performs intra prediction using the spatial correlation in the frame. Specifically, the intra prediction unit 242 uses the intra prediction mode corresponding to the information related to the intra prediction output by the entropy decoding unit 200 (for example, the intra prediction mode information), and the intra prediction unit 242 uses the intra prediction mode of the decoded image stored in the memory 230. An intra prediction block is generated with reference to the decoded pixels around the coded block, and the generated intra prediction block is output to the switching unit 243.

In the present embodiment, the intra prediction unit 242 has a function of performing MRL intra prediction. The intra prediction unit 242 generates a prediction block by using any one of the plurality of reference pixel strings for prediction based on the reference pixel string information output by the entropy decoding unit 200.

The switching unit 243 switches between the inter prediction block output by the inter prediction unit 241 and the intra prediction block output by the intra prediction unit 242, and outputs one of the prediction blocks to the synthesis unit 220.

As described above, the decoding device 2 according to the present embodiment is a reference pixel string selected from a plurality of reference pixel strings including a reference pixel string adjacent to the decoding target block and a reference pixel string not adjacent to the decoding target block. Intra prediction unit 242 that predicts the block to be decoded by intra prediction, and inverse quantization unit 211 that performs dequantization processing on the quantization conversion coefficient output by the entropy decoding unit 200 and outputs the conversion coefficient. It has an inverse conversion unit 212 that performs an inverse conversion process on the conversion coefficient output by the inverse quantization unit 211 and outputs a predicted residual. The inverse transformation unit 212 has a transformation base determination unit 212a that determines a conversion basis to be used in the inverse transformation process based on a reference pixel sequence selected from a plurality of reference pixel sequences.

Further, in the present embodiment, the entropy decoding unit 200 does not acquire the conversion base information indicating the conversion base, but refers to the reference pixel string information indicating the reference pixel string selected on the coding side from the plurality of reference pixel strings. To get. The conversion base determination unit 212a determines the conversion base based on the reference pixel string indicated by the acquired reference pixel string information.

Therefore, according to the decoding device 2 according to the present embodiment, even when the flag is not transmitted in order to switch the conversion base, the conversion base more suitable for the characteristics of the predicted residual can be applied for coding. Efficiency can be improved.

Next, the operation flow of the decoding device 2 according to the present embodiment will be described. FIG. 8 is a diagram showing an operation flow of the decoding device 2 according to the present embodiment. Here, when applying the MRL intra-prediction, only the operation related to the determination of the transformation basis is shown.

As shown in FIG. 8, in step S21, the entropy decoding unit 200 decodes the encoded data and outputs the quantization conversion coefficient corresponding to the decoding target block to the inverse quantization unit 211. The entropy decoding unit 200 acquires the reference pixel string information indicating the reference pixel string selected on the coding side without acquiring the conversion base information indicating the conversion base, and the acquired reference pixel string information is used as the inverse conversion unit 212. And output to the intra prediction unit 242.

In step S22, the intra prediction unit 242 refers to the reference pixel string indicated by the reference pixel string information, predicts the decoding target block from the MRL intra prediction, and outputs the prediction block.

On the other hand, in step S23, the conversion base determination unit 212a of the inverse conversion unit 212 identifies the reference pixel string from the reference pixel string information output by the entropy decoding unit 200, and determines the conversion base associated with the specified reference pixel string. decide. For example, the conversion basis determination unit 121a determines the conversion basis from the number of the reference pixel string according to the correspondence shown in Table 1.

In step S24, the inverse quantization unit 211 performs an inverse quantization process on the quantization conversion coefficient output by the entropy decoding unit 200 and outputs the conversion coefficient. The inverse transformation unit 212 performs an inverse transformation process on the conversion coefficient output by the inverse quantization unit 211 using the conversion basis determined in step S23, and outputs the predicted residual to the synthesis unit 220.

In step S25, the synthesizing unit 220 decodes (reconstructs) the original block by synthesizing the prediction residual output by the inverse conversion unit 212 and the prediction block output by the prediction unit 240 on a pixel-by-pixel basis. The decoded block is output to the memory 230.

<Other Embodiments>
In the above-described embodiment, when the MRL intra-prediction is applied, an example of determining the conversion basis based only on the selected reference pixel sequence has been described. However, the transformation basis may be determined based on other criteria, not just the selected reference pixel sequence. For example, the conversion basis may be determined based on the selected reference pixel sequence and the block size.

That is, in the coding apparatus 1 according to the other embodiment, the conversion basis determination unit 121a converts the reference pixel string selected from the plurality of reference pixel strings and the block size of the coded block. Determine the transformation basis used in the process. In the decoding device 2 according to another embodiment, the conversion basis determination unit 212a is used in the inverse conversion process based on the reference pixel string selected from the plurality of reference pixel strings and the block size of the block to be decoded. Determine the transformation basis.

For example, when the block size is equal to or smaller than a predetermined size, the conversion basis determination units 121a and 212a adaptively determine the conversion basis from the reference pixel sequence according to the correspondence as shown in Table 1. On the other hand, when the block size exceeds a predetermined size, the conversion basis determination units 121a and 212a determine a predetermined conversion basis (for example, DCT-2) regardless of the association as shown in Table 1.

A program that causes a computer to execute each process performed by the coding device 1 may be provided. A program that causes a computer to execute each process performed by the decoding device 2 may be provided. The program may be recorded on a computer-readable medium. Computer-readable media allow you to install programs on your computer. Here, the computer-readable medium on which the program is recorded may be a non-transient recording medium. The non-transient recording medium is not particularly limited, but may be, for example, a recording medium such as a CD-ROM or a DVD-ROM.

A circuit that executes each process performed by the coding device 1 may be integrated, and the coding device 1 may be configured by a semiconductor integrated circuit (chipset, SoC). A circuit that executes each process performed by the decoding device 2 may be integrated, and the decoding device 2 may be configured by a semiconductor integrated circuit (chipset, SoC).

Although the embodiments have been described in detail with reference to the drawings, the specific configuration is not limited to the above, and various design changes and the like can be made within a range that does not deviate from the gist.

1: Coding device 2: Decoding device 100: Block dividing unit 110: Predicted residual generation unit 120: Conversion / quantization unit 121: Conversion unit 121a: Conversion basis determination unit 122: Quantization unit 130: Entropy coding unit 131 : Scan control unit 140: Inverse quantization / inverse conversion unit 141: Inverse quantization unit 142: Inverse conversion unit 150: Synthesis unit 160: Memory 170: Prediction unit 171: Inter prediction unit 172: Intra prediction unit 172a: Reference pixel string Selection unit 173: Switching unit 200: Entropy decoding unit 210: Inverse quantization / inverse conversion unit 211: Inverse quantization unit 212: Inverse conversion unit 212a: Conversion base determination unit 220: Synthesis unit 230: Memory 240: Prediction unit 241: Inter prediction unit 242: Intra prediction unit 243: Switching unit

Claims (10)

A coding device that divides an image into blocks and encodes it.
Intra-prediction of the coding target block using a reference pixel string selected from a plurality of reference pixel strings including a reference pixel string adjacent to the coding target block and a reference pixel string not adjacent to the coding target block. Intra-prediction part that predicts by
It is provided with a conversion unit that performs conversion processing on a prediction residual representing a difference between the prediction block output by the intra prediction unit and the coded target block and outputs a conversion coefficient.
The conversion unit
A coding apparatus having a conversion base determination unit that determines a conversion base used in the conversion process based on the reference pixel sequence selected from the plurality of reference pixel sequences. A quantization unit that performs quantization processing on the conversion coefficient output by the conversion unit and outputs a quantization conversion coefficient.
Further provided with an entropy coding unit that encodes the quantization conversion coefficient and outputs the coded data.
The entropy encoding unit is
The first aspect of the present invention is that the reference pixel string information indicating the reference pixel string selected from the plurality of reference pixel strings is transmitted to the decoding side without transmitting the conversion base information indicating the conversion base. The encoding device according to. The transformation basis determination unit
Based on the reference pixel sequence selected from the plurality of reference pixel sequences, the horizontal conversion base used in the conversion process in the horizontal direction and the vertical conversion base used in the conversion process in the vertical direction are determined. The coding apparatus according to claim 1 or 2. The transformation basis determination unit
Claims 1 to 1, wherein the conversion base used in the conversion process is determined based on the reference pixel sequence selected from the plurality of reference pixel sequences and the block size of the coded block. The coding apparatus according to any one of 3. An entropy decoding unit that decodes the coded data and outputs the quantization conversion coefficient corresponding to the block to be decoded,
Using a reference pixel string selected from a plurality of reference pixel strings including a reference pixel string adjacent to the decoding target block and a reference pixel string not adjacent to the decoding target block, the decoding target block is subjected to intra-prediction. Intra-prediction part to predict and
An inverse quantization unit that performs an inverse quantization process on the quantization conversion coefficient output by the entropy decoding unit and outputs a conversion coefficient.
It is provided with an inverse transformation unit that performs an inverse transformation process on the conversion coefficient output by the inverse quantization unit and outputs a predicted residual.
The inverse conversion unit
A decoding device comprising a conversion base determination unit that determines a conversion base used in the inverse transformation process based on the reference pixel sequence selected from the plurality of reference pixel sequences. The entropy decoding unit
Without acquiring the conversion base information indicating the conversion base, the reference pixel string information indicating the reference pixel string selected on the coding side from the plurality of reference pixel strings is acquired.
The transformation basis determination unit
The decoding device according to claim 5, wherein the conversion basis is determined based on the reference pixel sequence indicated by the acquired reference pixel sequence information. The transformation basis determination unit
Based on the reference pixel sequence selected from the plurality of reference pixel sequences, the horizontal transformation base used in the horizontal inverse transformation process and the vertical transformation basis used in the vertical inverse transformation process are determined. The decoding device according to claim 5 or 6, wherein the decoding device is characterized by the above. The transformation basis determination unit
5. A aspect of claim 5, wherein the conversion base used in the inverse transformation process is determined based on the reference pixel sequence selected from the plurality of reference pixel sequences and the block size of the decoding target block. 7. The decoding device according to any one of 7. A program characterized in that a computer functions as the coding device according to any one of claims 1 to 4. A program characterized in that a computer functions as the decoding device according to any one of claims 5 to 8.

Images