JP6887296B2 - Coding device, decoding device and program - Google Patents

Description

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

Conventionally, research on coding methods has been conducted in order to compress the amount of data when transmitting or storing still images and moving images (videos).

In recent years, in video coding technology, ultra-high resolution video represented by 8K-SHV (Super Hi-Vision) has become widespread, and moving images with a huge amount of data can be broadcast waves or I.
As a method for transmitting on the P network, H.264 / AVC (Advanced Video)
Coding) and H.265 / HEVC (High Efficiency Video)
Coding methods such as Coding) are known.

In the moving image coding method represented by H.265 / HEVC, prediction is performed while switching between two types of predictions, inter-prediction using temporal correlation between frames and intra-prediction using spatial correlation within frames. After generating the residual signal between the original image and the predicted image, a stream is generated by performing orthogonal transformation processing, quantization processing, entropy coding processing, and the like.

In the intra prediction, prediction modes such as Planar prediction, DC prediction, and direction prediction are prepared, and the intra prediction is performed using the adjacent decoded reference pixels according to the prediction mode determined by the encoder. ..

Here, in the intra prediction in H.265 / HEVC, in a CU in which there is no adjacent decoded reference pixel such as a coded target block (hereinafter referred to as “CU: Coding Unit”) located at the upper left in the frame. , It is configured to create a reference pixel to be used when generating a predicted image by a process of filling in a specified value.

Further, in the conventional intra prediction in H.265 / HEVC, the coding process is C in the upper left.
The reference pixel may not have been decoded because it is performed in the coding order (Z scan order or raster scan order) of U → upper right CU → lower left CU → lower right CU. In such a case, the predicted image is generated by using the value obtained by extrapolating the nearest decoded reference pixel to the 0th order (see, for example, Non-Patent Document 1).

Here, in the conventional intra-prediction in H.265 / HEVC, the CU is a plurality of blocks to which a prediction mode is assigned to each (hereinafter, “PU: Prediction Un”).
There is a case where it is divided into (referred to as "it") (hereinafter, referred to as "when N × N division is performed"). In such a case, since the PU coding process is performed in the Z scan order described above, for example, when a prediction mode in a prediction direction other than the upper left is assigned to a certain PU, the reference pixel is not decoded. Therefore, there is a problem that the prediction accuracy is lowered and the coding efficiency is lowered.

Hereinafter, such a problem will be specifically described with reference to FIGS. 12 (a) to 12 (d). In the figures of the present specification, the arrow indicating the prediction direction of the prediction mode is H.265 / HE.
Similar to the description in the VC standard, it is assumed that the pixel targeted for intra-prediction goes to the reference pixel (the same applies hereinafter).

FIG. 12 shows an example of intra prediction when CU # 1 is divided into PU # 0 to PU # 3 in the conventional H.265 / HEVC.

In the example of FIG. 12, the prediction mode in PU # 0 is "34", the prediction mode in PU # 1 is "2", the prediction mode in PU # 2 is "18", and the prediction mode in PU # 3 is "18". The mode is assumed to be "2".

As shown in FIGS. 12 (a) and 12 (c), all the reference pixels of PU # 0 and PU # 2 have been decoded.

On the other hand, as shown in FIG. 12B, when PU # 1 is encoded, the reference pixel located in PU # 2 has not been decoded at this point, so that PU # 1 is predicted as it is. It cannot be a reference pixel for generating an image. Therefore, the conventional conventional H.265 / H
In EVC, the reference pixel P # 0 located at the bottom of the decoded reference pixel located in PU # 0
The value of is specified to be copied to the undecrypted reference pixel located in the same column in PU # 2.

Therefore, when the direction prediction as shown in FIG. 12B is performed, the prediction accuracy is lowered because a part of the generated prediction image is composed of undecoded reference pixels filled with copies.
There is a problem that the coding efficiency is reduced. The same applies when the direction prediction as shown in FIG. 12D is performed.

In order to solve such a problem, in the intra-prediction, in addition to the above-mentioned Z scan order,
There is known a technique for improving prediction accuracy by giving a degree of freedom in the coding order by using a U-type scan order, an X-type scan order, or the like (see, for example, Non-Patent Document 2).

Supervised by Ei Okubo, "Impress Standard Textbook Series H.265 / HEVC Textbook", Impress Japan Co., Ltd., October 21, 2013 Mochizuki et al., "Applicable intra-prediction method based on average value coordinates", IPSJ Research Report, vol, 2012-AVM-77, No.12

However, in the technique described in Non-Patent Document 2 and the like described above, it is necessary to separately transmit a flag indicating what kind of coding order is used for each CU, so that the amount of data to be transmitted increases. There was a problem that it would end up.

Therefore, the present invention has been made to solve the above-mentioned problems, and provides a coding device, a decoding device, and a program capable of preventing a decrease in coding efficiency without increasing the amount of data to be transmitted. The purpose is to do.

The first feature of the present invention is a coding device configured to divide an original image of a frame unit constituting a moving image into coding target blocks and encode them, and the coding target block is When divided into blocks to which a prediction mode is assigned, a coding order control unit configured to determine the coding order of the blocks based on the combination of the prediction modes in each of the blocks. It is a gist to include a decoded image generation unit configured to generate a decoded image based on the coding order and the method of dividing the coded block into the blocks.

A second feature of the present invention is a decoding device configured to divide an image in units of frames constituting a moving image into coding target blocks and decode the coding target block.
When the block is divided into blocks to which a prediction mode is assigned, a decoding order control unit configured to determine the decoding order of the blocks based on the combination of the prediction modes in each of the blocks, and the decoding. It is a gist to include a decoded image generation unit configured to generate a decoded image based on the order and the method of dividing the coded block into the blocks.

A third feature of the present invention is a program for causing a computer to function as the coding device described in the first feature described above.

A fourth feature of the present invention is a program for making a computer function as a decoding device according to the second feature described above.

According to the present invention, it is possible to provide a coding device, a decoding device and a program capable of preventing a decrease in coding efficiency without increasing the amount of data to be transmitted.

FIG. 1 is a diagram showing an example of a method of dividing CU into PUs in the first embodiment. FIG. 2 is a functional block diagram of the coding device 1 according to the first embodiment. FIG. 3 is a diagram showing an example of a method for grouping prediction modes according to the first embodiment. FIG. 4 is a diagram showing an example of a table used for determining the coding order and the decoding order in the first embodiment. FIG. 5 is a diagram showing an example of a coding order and a decoding order in the first embodiment. FIG. 6 is a diagram showing an example of a coding order and a decoding order in the first embodiment. FIG. 7 is a flowchart showing the operation of the coding device 1 according to the first embodiment. FIG. 8 is a functional block diagram of the decoding device 2 according to the first embodiment. FIG. 9 is a flowchart showing the operation of the decoding device 2 according to the first embodiment. FIG. 10 is a diagram showing an example of a table used for determining the coding order and the decoding order in the second embodiment. FIG. 11 is a diagram showing an example of a coding order and a decoding order in the second embodiment. FIG. 12 is a diagram for explaining the prior art.

(First Embodiment)
Hereinafter, the coding device 1 and the decoding device 2 according to the first embodiment of the present invention will be described with reference to FIGS. 1 to 9. Here, the coding device 1 and the decoding device 2 according to the present embodiment are described by H.A.
It is configured to support intra-prediction in moving image coding methods such as 265 / HEVC. The coding device 1 and the decoding device 2 according to the present embodiment are configured to be compatible with any moving image coding method as long as it is a moving image coding method that performs intra-prediction.

The coding device 1 according to the present embodiment is configured to divide the original image of each frame constituting the moving image into coding target blocks and encode the original image. Hereinafter, in the present embodiment, a case where “CU” used in the current H.265 / HEVC is used as an example of the coded block will be described, but the present invention is limited to such a case. However, it can also be applied to cases where a block to be encoded with another name is used.

Further, the coding device 1 according to the present embodiment is configured so that the coded block can be divided into a plurality of blocks. Here, it is assumed that a prediction mode is assigned to each of such blocks. Hereinafter, in the present embodiment, as such a block,
The case using "PU" used in the current H.265 / HEVC will be described as an example, but the present invention is not limited to such a case, and a block having another name is also used. Applicable.

Hereinafter, in the present embodiment, as shown in FIG. 1, a case where CU # 1 is divided into PU # 0 to PU # 3 will be described as an example.

Further, in the present embodiment, in the CU to be encoded in which the adjacent decoded reference pixel does not exist, such as the CU located at the upper left in the frame, a specified value (for example, "512" for a 10-bit moving image Since it is configured to create reference pixels to be used when generating a predicted image by the process of filling in), all the pixels adjacent to the left side of the CU to be encoded can be used as reference pixels. To do.

As shown in FIG. 2, the coding apparatus 1 according to the present embodiment includes a prediction mode determination unit 11, a division determination unit 12, a coding order control unit 13, a recalculation control unit 14, and a decoded image generation unit. It includes 15, an entropy coding unit 16, and a memory 17.

The prediction mode determination unit 11 is configured to determine an appropriate prediction mode to be applied to the CU and PU.

For example, as shown in FIG. 3, in the current H.265 / HEVC, the prediction mode is from "0" to "0".
Taking any value of "34", the prediction mode 0 corresponds to the Planar prediction, and the prediction mode 1 corresponds to the DC prediction.

In the present embodiment, the prediction mode is divided into three regions, the prediction modes 2 to 9 belong to the prediction region A, the prediction modes 0, 1, 10 to 26 belong to the prediction region B, and the prediction modes 27 to 34 belong to the prediction region B. ,
It is assumed that it belongs to the prediction area C.

That is, when there is a reference pixel on the lower left side from the center of the CU and PU (when the prediction direction is the lower left side), the prediction mode belongs to the prediction area A, and the reference pixel is on the upper right side from the center of the CU and PU. In some cases (when the prediction direction is in the upper right), the prediction mode belongs to the prediction area C.
Otherwise, the prediction mode belongs to the prediction area B.

The present invention is also applicable to cases where the number of prediction modes is larger than "35" in the current H.265 / HEVC.

The division determination unit 12 is configured to determine whether or not to divide the CU (CU # 1 in the present embodiment) into a plurality of PUs (PU # 0 to PU # 3 in the present embodiment). .. In the present embodiment, as a method of dividing the CU into a plurality of PUs, a case of four divisions is described as an example, but the number of divisions and the division shape when the CU is divided into a plurality of PUs are described. , Not limited to such cases.

The coding order control unit 13 is determined by the division determination unit 12 to divide the CU (CU # 1 in the present embodiment) into a plurality of PUs (PU # 0 to PU # 3 in the present embodiment). If,
It is configured to determine the coding order of the PUs based on the combination of the prediction modes in each of the PUs determined by the prediction mode determination unit 11.

Here, the coding order control unit 13 may be configured to determine the coding order of the PU so as to use the largest number of decoded reference pixels when generating the predicted image.

For example, the coding order control unit 13 sets CU # 1 to PU # 0 to PU # by the division determination unit 12.
When it is decided to divide into three, it may be configured to determine the coding order of PUs based on the table shown in FIG.

Specifically, the coding order control unit 13 sets CU # 1 to PU # 0 to P by the division determination unit 12.
When it is decided to divide into U # 3, and as in the case of (1) in FIG. 4, PU
The prediction mode in # 0 belongs to the prediction area A, and the prediction mode in PU # 1 belongs to the prediction area A.
When the prediction mode in PU # 2 belongs to the prediction area B and the prediction mode in PU # 3 belongs to the prediction area B (for example, as shown in FIG. 5, the prediction mode in PU # 0 is "2". Yes, the prediction mode in PU # 1 is "2", the prediction mode in PU # 2 is "18", and the prediction mode in PU # 3 is "18"), FIG. 12 (a).
) To PU # 0 (upper left PU in CU # 1) → PU as shown in FIGS. 5 (a) to 5 (d) instead of the conventional Z scan order as shown in FIGS. 12 (d). # 2 (PU in the lower left of CU # 1)
→ PU # 1 (the upper right PU in CU # 1) → PU # 3 (the lower right PU in CU # 1) may be configured to adopt the coding order.

Alternatively, the coding order control unit 13 sets CU # 1 to PU # 0 to PU # by the division determination unit 12.
When it is decided to divide into 3, and as in the case of (2) in FIG. 4, PU # 0
When the prediction mode in PU # 1 belongs to the prediction area C, the prediction mode in PU # 1 belongs to the prediction area A, the prediction mode in PU # 2 belongs to the prediction area B, and the prediction mode in PU # 3 belongs to the prediction area A ( For example, as shown in FIG. 6, the prediction mode in PU # 0 is "3.
4 ”, the prediction mode in PU # 1 is“ 2 ”, the prediction mode in PU # 2 is“ 18 ”, and the prediction mode in PU # 3 is“ 2 ”), FIG. 12 (a). ) ~
PU # 0 (upper left PU in CU # 1) → PU # 2 as shown in FIGS. 6 (a) to 6 (d) instead of the conventional Z scan order as shown in FIG. 12 (d). (PU in the lower left of CU # 1) → P
It may be configured to adopt the coding order of U # 3 (the lower right PU in the CU # 1) → PU # 1 (the upper right PU in the CU # 1).

The combination of the prediction region and the coding order (or decoding order) of each PU is not limited to the combination shown in FIG. 4, and other effects (for example, improvement of coding speed and coding processing) are not limited to those shown in FIG. If it can be expected to be simplified, other combinations may be used.

Here, if the coding order is changed as described above, a PU that needs to be recalculated by changing the content of the reference pixel to be used is generated. Therefore, the recalculation control unit 14 is configured to selectively erase and recalculate a part of the block data of the memory 17 held so that it can be used as a reference image.

Specifically, the recalculation control unit 14 is CU (in the present embodiment, C) by the division determination unit 12.
When it is decided to divide U # 1) into a plurality of PUs (PU # 0 to PU # 3 in this embodiment), and the coding order of the PUs determined by the coding order control unit 13. However, Fig. 12 (
When the order is not the conventional Z scan order as shown in a) to FIG. 12 (d), the PU coding order determined by the coding order control unit 13 and the conventional Z scan order are compared from the beginning. The decoded images already created for all PUs after the PUs in different coding orders are erased, and then the decoding processing is performed in the new coding order using new reference pixels to obtain the decoded images. It is configured to be created and stored in the memory 17.

For example, the recalculation control unit 14 uses the coding order control unit 13 to perform PU # 0 (upper left PU in CU # 1) → PU # 2 (lower left PU in CU # 1) → PU # 1 (CU). P in the upper right of # 1
When it is decided to adopt the coding order (see FIGS. 5 (a) to 5 (d)) of U) → PU # 3 (the lower right PU in CU # 1), the conventional Z scan order. P with a different coding order than
All PU # 2 (lower left PU in CU # 1) after U # 2 (lower left PU in CU # 1),
Delete the decoded images already created for PU # 1 (the upper right PU in CU # 1) and PU # 3 (the lower right PU in CU # 1), and then use a new reference pixel. The coding process is performed in the new coding order, a decoded image is created, and the decoded image is stored in the memory 17.

Similarly, the recalculation control unit 14 uses the coding order control unit 13 to perform PU # 0 (upper left PU in CU # 1) → PU # 2 (lower left PU in CU # 1) → PU # 3 (PU # 1). Lower right P in CU # 1
When it is decided to adopt the coding order (see FIGS. 6 (a) to 6 (d)) of U) → PU # 1 (PU in the upper right of CU # 1), the conventional Z scan order is used. P with different coding order
All PU # 2 (lower left PU in CU # 1) after U # 2 (lower left PU in CU # 1),
Delete the decoded images already created for PU # 1 (the upper right PU in CU # 1) and PU # 3 (the lower right PU in CU # 1), and then use a new reference pixel. The coding process is performed in the new coding order, a decoded image is created, and the decoded image is stored in the memory 17.

The decoded image generation unit 15 has a PU coding order and a CU determined by the coding order control unit 13.
Based on the method of dividing (CU # 1 in this embodiment) into PUs (PU # 0 to PU # 3 in this embodiment), a decoded image for each PU is generated.

Specifically, the decoded image generation unit 15 uses the division determination unit 12 to perform CU # 1 on a plurality of PU #s.
When it is decided to divide into 0 to PU # 3, the decoded image for each PU is sequentially generated according to the coding order of the PU determined by the coding order control unit 13.

As shown in FIG. 2, the decoded image generation unit 15 includes a prediction unit 15a, a residual signal generation unit 15b, a conversion / quantization unit 15c, an inverse quantization / inverse orthogonal transformation unit 15d, and a locally decoded image generation. Part 15
It is equipped with e.

The prediction unit 15a is configured to generate a prediction image using the prediction mode determined by the prediction mode determination unit 11. That is, the prediction unit 15a is configured to determine the position of the reference pixel used when generating the prediction image.

Specifically, the prediction unit 15a uses the division determination unit 12 to perform CU # 1 on a plurality of PUs # 0 to P.
When it is decided to divide into U # 3, and as in the case of (1) in FIG. 4, PU
The prediction mode in # 0 belongs to the prediction area A, and the prediction mode in PU # 1 belongs to the prediction area A.
When the prediction mode in PU # 2 belongs to the prediction area B and the prediction mode in PU # 3 belongs to the prediction area B (for example, the case of FIG. 5), in FIGS. 5 (a) to 5 (d). As shown, PU # 0 (upper left PU in CU # 1) → PU # 2 (lower left PU in CU # 1) → P
It may be configured to generate a predicted image in the coding order of U # 1 (the upper right PU in CU # 1) → PU # 3 (the lower right PU in CU # 1).

Alternatively, the prediction unit 15a sets the CU # 1 by the division determination unit 12 to a plurality of PU # 0 to PU #.
When it is decided to divide into 3, and as in the case of (2) in FIG. 4, PU # 0
When the prediction mode in PU # 1 belongs to the prediction area C, the prediction mode in PU # 1 belongs to the prediction area A, the prediction mode in PU # 2 belongs to the prediction area B, and the prediction mode in PU # 3 belongs to the prediction area A ( For example, as shown in the case of FIG. 6) and FIGS. 6 (a) to 6 (d), PU # 0 (upper left PU in CU # 1) → PU # 2 (lower left PU in CU # 1). ) → PU #
It may be configured to generate a predicted image in the coding order of 3 (lower right PU in CU # 1) → PU # 1 (upper right PU in CU # 1).

Here, the prediction unit 15a of the decoded image generation unit 15 may be configured to generate a prediction image in consideration of the distance between the pixels of PU # 0 to PU # 3 and the decoded reference pixel.

For example, as shown in FIG. 6D, the prediction unit 15a is configured to generate a prediction image of PU # 1 by using the decoded reference pixels in the PU adjacent to the lower side of PU # 1. You may be.

The residual signal generation unit 15b is configured to generate a residual signal based on the difference between the predicted image generated by the prediction unit 15a and the original image.

The conversion / quantization unit 15c performs conversion processing (for example, orthogonal conversion processing) and quantization processing on the residual signal generated by the residual signal generation unit 15b to generate a quantized conversion coefficient. It is configured in.

The inverse quantization / inverse orthogonal transform unit 15d again performs inverse quantization processing and inverse orthogonal transformation processing on the quantized conversion coefficient generated by the transformation / quantization unit 15c to generate a residual signal. It is configured as follows.

The locally decoded image generation unit 15e generates a locally decoded image for each PU by adding the predicted image generated by the prediction unit 15a to the residual signal generated by the inverse quantization / inverse orthogonal transform unit 15d. It is configured in.

The entropy coding unit 16 is configured to perform entropy coding processing on flag information including a prediction mode determined by the prediction mode determination unit 11 and a quantized conversion coefficient to output a stream.

The memory 17 is configured to hold the decoded image for each PU generated by the decoded image generation unit 15 so as to be available as a reference image.

An example of the operation of the coding apparatus 1 according to the present embodiment will be described with reference to FIG. 7.

As shown in FIG. 7, in step S101, whether the coding apparatus 1 divides the CU (CU # 1 in the present embodiment) into a plurality of PUs (PU # 0 to PU # 3 in the present embodiment). Determine whether or not (whether or not to apply the N × N division).

In the case of "Yes", this operation proceeds to step S102, and in the case of "No", this operation is
finish.

In step S102, the coding device 1 determines the prediction mode applied to each of PU # 0 to PU # 3.

In step S103, the coding device 1 uniquely determines the coding order of the PUs based on the combination of prediction modes in each of the PUs.

In step S104, the coding device 1 is the PU determined in step S103.
It is determined whether or not the coding order of is different from the conventional Z scan order.

In the case of "Yes", this operation proceeds to step S105, and in the case of "No", this operation is
finish.

In step S105, the coding device 1 is the PU determined in step S103.
It is determined whether or not the first PU in the coding order of is the same as the first PU in the conventional Z scan order.

In the case of "Yes", this operation proceeds to step S107, and in the case of "No", this operation is
The process proceeds to step S106.

In step S106, the encoding device 1 deletes all the decoded images of PU # 0 to PU # 3 from the memory. After that, this operation proceeds to step S110, and the coding apparatus 1 recalculates using the coding order of the PUs determined in step S103, and all PUs # 0 to 0.
Generate a decoded image of PU # 3.

In step S107, the coding device 1 is the PU determined in step S103.
It is determined whether or not the second PU in the coding order of is the same as the second PU in the conventional Z scan order.

In the case of "Yes", this operation proceeds to step S109, and in the case of "No", this operation is
The process proceeds to step S108.

In step S108, the encoding device 1 deletes the decoded image of the PU other than the head from the memory. After that, this operation proceeds to step S110, and the coding device 1 advances to step S1.
Recalculation is performed using the coding order of the PUs determined in 03, and the decoded images of the second and subsequent PUs are generated.

In step S109, the coding apparatus 1 deletes the decoded image of the third and subsequent PUs from the memory, and in step S110, recalculates using the coding order of the PUs determined in step S103, and performs recalculation for the third and subsequent PUs. Generate a decoded image of.

According to the coding device 1 according to the present embodiment, it is possible to prevent a decrease in coding efficiency without increasing the amount of data to be transmitted.

Further, the decoding device 2 according to the present embodiment is configured to divide an image in frame units constituting a moving image into CUs and decode the image. Further, the decoding device 2 according to the present embodiment is configured so that the CU can be divided into a plurality of PUs, similarly to the coding device 1 according to the present embodiment.

As shown in FIG. 8, the decoding device 2 according to the present embodiment includes an entropy decoding unit 21, a decoding order control unit 22, a decoded image generation unit 23, and a memory 24.

The entropy decoding unit 21 is configured to perform entropy decoding processing on the stream output from the coding device 1 to decode the conversion coefficient, flag information, and the like from the stream output from the coding device 1. ing. Here, the conversion coefficient is the coding device 1
The above-mentioned quantized conversion coefficient obtained by dividing the original image in frame units into CUs and obtaining a coded signal. In addition, the flag information includes accompanying information such as a prediction mode.

The decoding order control unit 22 is configured to determine the decoding order of the PUs based on the prediction mode of each PU.

Specifically, the decoding order control unit 22 is N × N output by the entropy decoding unit 21.
It is configured to determine the decoding order of the PUs in the CU according to the flag indicating whether or not the division has been performed (whether or not the CU is divided into a plurality of PUs) and the direction of the prediction mode.

Here, the decoding order control unit 22 may be configured to determine the decoding order of the PU so as to use the largest number of decoded reference pixels when generating the predicted image.

For example, in the decoding order control unit 22, CU # 1 is PU # 0 to P, similarly to the coding order control unit 13.
When divided into U # 3, it may be configured to determine the decoding order of the PUs based on the same table as the table shown in FIG. 4 held by the coding apparatus 1.

Specifically, the decoding order control unit 22 predicts in PU # 0 when CU # 1 is divided into PU # 0 to PU # 3, and as in the case of (1) of FIG. Mode is prediction area A
When the prediction mode in PU # 1 belongs to the prediction area A, the prediction mode in PU # 2 belongs to the prediction area B, and the prediction mode in PU # 3 belongs to the prediction area B (for example,
As shown in FIG. 5, the prediction mode in PU # 0 is "2", the prediction mode in PU # 1 is "2", the prediction mode in PU # 2 is "18", and the prediction mode in PU # 3 is "18". When the prediction mode is "18"), as shown in FIGS. 5 (a) to 5 (d), instead of the conventional Z scan order as shown in FIGS. 12 (a) to 12 (d). PU # 0 (upper left PU in CU # 1) → PU # 2 (lower left PU in CU # 1) → PU # 1 (upper right PU in CU # 1)
) → PU # 3 (the lower right PU in CU # 1) may be configured to adopt the decoding order.

Alternatively, the decoding order control unit 22 predicts the prediction mode in PU # 0 when CU # 1 is divided into PU # 0 to PU # 3 and as in the case of (2) of FIG. When the prediction mode in PU # 1 belongs to the prediction area A, the prediction mode in PU # 2 belongs to the prediction area B, and the prediction mode in PU # 3 belongs to the prediction area A (for example, FIG. 6).
As shown in, the prediction mode in PU # 0 is "34", the prediction mode in PU # 1 is "2", the prediction mode in PU # 2 is "18", and the prediction mode in PU # 3 Is "2"), instead of the conventional Z scan order as shown in FIGS. 12 (a) to 12 (d), PU # is shown in FIGS. 6 (a) to 6 (d). 0 (upper left PU in CU # 1) → PU # 2 (lower left PU in CU # 1) → PU # 3 (lower right PU in CU # 1) →
It may be configured to adopt the decoding order of PU # 1 (PU in the upper right of CU # 1).

The decoded image generation unit 23 uses the PU decoding order and the CU (CU) determined by the decoding order control unit 22.
In this embodiment, a decoded image for each PU is generated based on the method of dividing CU # 1) into PUs (PU # 0 to PU # 3 in this embodiment).

Specifically, when the CU # 1 is divided into a plurality of PUs # 0 to PU # 3, the decoded image generation unit 23 sequentially follows the PU coding order determined by the decoding order control unit 22. , PU
It is configured to generate a decoded image for each.

More specifically, the decoding image generation unit 23 is an entropy decoding unit according to the decoding order determined by the decoding order control unit 22 when the CU # 1 is divided into a plurality of PU # 0 to PU # 3. It is configured to generate a decoded image for each PU by sequentially performing a predictive image generation process, an inverse quantization process, and an inverse orthogonal transformation process on the quantized conversion coefficient output by 21. ..

As shown in FIG. 4, the decoded image generation unit 23 includes a predictive image generation unit 23a, an inverse quantization / inverse conversion unit 23b, and a local decoding image generation unit 23c.

The prediction image generation unit 23a may be configured to generate a prediction image using the prediction mode output by the entropy decoding unit 21 according to the decoding order determined by the decoding order control unit 22.

Specifically, in the prediction image generation unit 23a, when CU # 1 is divided into a plurality of PUs # 0 to PU # 3, and as in the case of (1) of FIG. 4, PU # 0 When the prediction mode in PU # 1 belongs to the prediction area A, the prediction mode in PU # 1 belongs to the prediction area A, the prediction mode in PU # 2 belongs to the prediction area B, and the prediction mode in PU # 3 belongs to the prediction area B ( For example, as shown in FIGS. 5 (a) to 5 (d)) and 5 (a) to 5 (d).
PU # 0 (upper left PU in CU # 1) → PU # 2 (lower left PU in CU # 1) → PU # 1
It may be configured to generate a predicted image in the decoding order of (PU in the upper right in CU # 1) → PU # 3 (PU in the lower right in CU # 1).

Alternatively, the prediction unit 23a predicts the prediction mode in PU # 0 when CU # 1 is divided into a plurality of PU # 0 to PU # 3, and as in the case of (2) of FIG. When the prediction mode in PU # 1 belongs to the prediction area A, the prediction mode in PU # 2 belongs to the prediction area B, and the prediction mode in PU # 3 belongs to the prediction area A (for example, FIG. 6). Case), as shown in FIGS. 6 (a) to 6 (d), PU # 0 (P in the upper left of CU # 1).
U) → PU # 2 (lower left PU in CU # 1) → PU # 3 (lower right PU in CU # 1) → P
It may be configured to generate a predicted image in the decoding order of U # 1 (PU in the upper right of CU # 1).

Here, the predicted image generation unit 23a of the decoded image generation unit 23 may be configured to generate a predicted image in consideration of the distance between the pixels of PU # 0 to PU # 3 and the decoded reference pixel. ..

For example, as shown in FIG. 6D, the prediction image generation unit 23a generates a prediction image of PU # 1 by using the decoded reference pixels in the PU adjacent to the lower side of PU # 1. It may be configured.

The inverse quantization / inverse transformation unit 23b performs the inverse quantization processing and the inverse transformation processing (for example, the inverse orthogonal transformation processing) on the quantized conversion coefficient output by the entropy decoding unit 21 to obtain a residual. It is configured to generate a signal.

The locally decoded image generation unit 23c is inversely quantized with the prediction image generated by the prediction unit 23a.
It is configured to generate a decoded image for each PU by adding the residual signal generated by the inverse conversion unit 23b.

The memory 24 is configured to hold the decoded image for each PU generated by the decoded image generation unit 23 so as to be available as a reference image for intra-prediction and inter-prediction.

An example of the operation of the decoding device 2 according to the present embodiment will be described with reference to FIG.

As shown in FIG. 9, in step S201, the decoding device 2 acquires the conversion coefficient and the flag information from the stream output from the coding device 1.

In step S202, the decoding device 2 has a plurality of PUs (CU # 1 in the present embodiment) based on the flag information included in the stream output from the coding device 1.
In the present embodiment, it is determined whether or not the division is divided into PU # 0 to PU # 3) (whether or not the N × N division is applied).

In the case of "Yes", this operation proceeds to step S203, and in the case of "No", this operation is
finish.

In step S203, the decoding device 2 acquires each prediction mode of PU # 0 to PU # 3 based on the flag information included in the stream output from the coding device 1.

In step S204, the decoding device 2 uniquely determines the decoding order of the PUs based on the combination of the prediction modes in each of PU # 0 to PU # 3.

In step S205, the decoding device 2 generates the decoded images of PU # 0 to PU # 3 according to the decoding order of PU determined in step S204.

According to the decoding device 2 according to the present embodiment, it is possible to prevent a decrease in coding efficiency without increasing the amount of data to be transmitted.

(Second embodiment)
Hereinafter, with reference to FIGS. 10 and 11, the coding device 1 and the decoding device 2 according to the second embodiment of the present invention will be referred to as the coding device 1 and the decoding device 2 according to the first embodiment described above. We will focus on the differences between the two.

In the coding apparatus 1 according to the present embodiment, whether or not the prediction unit 15a of the decoded image generation unit 15 uses a prediction mode using decoded reference pixels in a plurality of PUs adjacent to the PU as the prediction mode. It is configured to generate a predicted image in consideration of.

Here, examples of the prediction mode using the decoded reference pixels in a plurality of PUs adjacent to the PU include Planar prediction and DC prediction, but the prediction mode is not limited to these and is adjacent to the PU. It may be in any prediction mode using decoded reference pixels in a plurality of PUs.

For example, in the coding device 1 of the present embodiment, when the pixels in the PU in the three directions or all directions adjacent to the PU to be coded are already decoded in the coding order determined by the prediction mode determination unit 11. In addition, when Planar prediction or DC prediction is assigned to such a PU, the prediction unit 15a of the decoding image generation unit 15 will generate a prediction image using all available reference pixels. The coding order control unit 13 may be configured to determine the coding order of the PU based on the table shown in FIG.

Here, the table shown in FIG. 10 is different from the table shown in FIG. 4 in the shaded areas.

The table shown in FIG. 4 is created with priority given to shortening the time required for the coding process (or decoding process), that is, as shown in FIGS. 12 (a) to 12 (d) as much as possible. By approaching the conventional Z scan order, the number of PUs that require recalculation in the coding process (or decoding process) is reduced.

On the other hand, the table shown in FIG. 10 considers the use of reference pixels in PUs that are adjacent to the PU to be encoded (decoding target) in three or all directions when Planar prediction or DC prediction is used. It is created with priority given to improving prediction accuracy.

Therefore, in the portion of the prediction region B in the shaded area in the table shown in FIG. 10, the coding order control unit 13 is shown in FIG. 10 only when the prediction mode is Planar prediction or DC prediction. The coding order of the PUs may be determined based on the table, and in other cases, the coding order of the PUs may be determined based on the table shown in FIG.

In this way, by using the table shown in FIG. 4 and the table shown in FIG. 10 in combination, it is possible to improve the prediction system without increasing the time required for the coding process (or the decoding process). ..

The combination of the prediction region and the coding order (or decoding order) of each PU is not limited to the combination shown in FIG. 10, and other effects (for example, improvement of coding speed and coding processing) are not limited to those shown in FIG. If it can be expected to be simplified, other combinations may be used.

Specifically, the coding order control unit 13 sets CU # 1 to PU # 0 to P by the division determination unit 12.
When it is decided to divide into U # 3, and as in the case of (1) of FIG. 10, P
The prediction mode in U # 0 belongs to the prediction area B, the prediction mode in PU # 1 belongs to the prediction area C, the prediction mode in PU # 2 belongs to the prediction area A, and the prediction mode in PU # 3 belongs to the prediction area A. In the case of belonging and the prediction mode in PU # 0 belonging to the prediction area B is "0" or "1" (Planar prediction or DC prediction), FIG. 11 (a)
)-As shown in FIG. 11 (d), PU # 1 (PU in the upper right of CU # 1) → PU # 2 (CU)
It is configured to adopt the coding order of PU # 3 (lower left PU in CU # 1) → PU # 3 (lower right PU in CU # 1) → PU # 0 (upper left PU in CU # 1). May be good.

On the other hand, the coding order control unit 13 sets CU # 1 to PU # 0 to PU # 3 by the division determination unit 12.
PU # 0 when it is decided to divide into
In the case where the prediction mode in PU # 1 belongs to the prediction area B, the prediction mode in PU # 1 belongs to the prediction area C, the prediction mode in PU # 2 belongs to the prediction area A, and the prediction mode in PU # 3 belongs to the prediction area A. And the prediction mode in PU # 0 belonging to the prediction area B is "0".
If it is not "1" (not Planar prediction or DC prediction), PU # 0 (CU #)
Upper left PU in 1) → PU # 1 (upper right PU in CU # 1) → PU # 2 (lower left PU in CU # 1) → PU # 3 (lower right PU in CU # 1) It may be configured to adopt the coding order.

In this way, when the division determination unit 12 determines that CU # 1 is divided into PU # 0 to PU # 3, and as in the case of FIG. 10 (1), the prediction in PU # 0 When the mode belongs to the prediction area B, the prediction mode in PU # 1 belongs to the prediction area C, the prediction mode in PU # 2 belongs to the prediction area A, and the prediction mode in PU # 3 belongs to the prediction area A, PU # As the predicted images of 1 to PU # 3, the same predicted images as in the case of the first embodiment described above are generated. However, in PU # 0, since the pixels in the PU adjacent to PU # 0 in all directions have already been decoded, the prediction accuracy can be further improved as compared with the case of the first embodiment described above (as compared with the case of the first embodiment described above). See FIG. 11 (d)).

In the decoding device 2 according to the present embodiment, similarly to the coding device 1 according to the present embodiment, the prediction unit 23a of the decoding image generation unit 23 sets the prediction mode as a decoded reference in a plurality of PUs adjacent to the PU. It is configured to generate a prediction image in consideration of whether or not a prediction mode using pixels is used.

(Other embodiments)
As mentioned above, the invention has been described by the embodiments described above, but the statements and drawings that form part of the disclosure in such embodiments should not be understood as limiting the invention. Such disclosure will reveal to those skilled in the art various alternative embodiments, examples and operational techniques.

Further, although not particularly mentioned in the above-described embodiment, a program for causing a computer to execute each process performed by the above-mentioned coding device 1 and decoding device 2 may be provided. The program may also be recorded on a computer readable medium. Computer-readable media can be used to install such programs on computers. Here, the computer-readable medium on which such a 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.

Alternatively, even if a chip composed of a memory for storing a program for realizing at least a part of the functions in the encoding device 1 and the decoding device 2 described above and a processor for executing the program stored in the memory is provided. Good.

1 ... Coding device 11 ... Prediction mode determination unit 12 ... Division determination unit 13 ... Coding order control unit 14 ... Recalculation control unit 15 ... Decoded image generation unit 15a ... Prediction unit 15b ... Residual signal generation unit 15c ... Conversion. Quantization unit 15d ... Inverse quantization / inverse conversion unit 15e ... Locally decoded image generation unit 16 ... Entropy coding unit 17 ... Memory 2 ... Decoding device 21 ... Entropy decoding unit 22 ... Decoding order control unit 23 ... Decoding image generation unit 23a ... Prediction image generation unit 23b ... Inverse quantization / inverse conversion unit 23c ... Locally decoded image generation unit 24 ... Memory

Claims (6)

It is a coding device configured to divide the original image of each frame constituting the moving image into coding target blocks and encode it.
When the coded block is divided into blocks to which a prediction mode is assigned, the coding order of the blocks is determined based on the combination of the prediction modes in each of the blocks. Coding order control unit and
A coding apparatus including a decoded image generation unit configured to generate a decoded image based on the coding order and a method of dividing the coded block into the blocks. The first aspect of the present invention, wherein the coding order control unit is configured to determine the coding order so as to use the largest number of decoded reference pixels when generating a predicted image. Coding device. A decoding device configured to divide an image in units of frames constituting a moving image into blocks to be encoded and decode it.
When the coded block is divided into blocks to which a prediction mode is assigned, decoding is configured to determine the decoding order of the blocks based on the combination of the prediction modes in each of the blocks. Forward control unit and
A decoding device including a decoded image generation unit configured to generate a decoded image based on the decoding order and a method of dividing the coded block into the blocks. The decoding device according to claim 3 , wherein the decoding order control unit is configured to determine the decoding order so as to use the largest number of decoded reference pixels when generating a predicted image. .. A program for causing a computer to function as the coding device according to claim 1 or 2. A program for causing a computer to function as the decoding device according to claim 3 or 4.

Images