JP2018121318A - Encoder, decoder and program - Google Patents

Abstract

PROBLEM TO BE SOLVED: To reduce increase of entropy in intra-prediction, even when using a reference pixel on the underside or the right side.SOLUTION: An encoder 1 includes an intra-prediction section 14a configured to generate a prediction image by using an intra-prediction mode, a residual signal generation section 14b configured to generate a residual signal by the difference of the prediction image and the original image, and an orthogonal transformation section 14c configured to perform orthogonal transformation processing of the residual signal, after inverting the base of at least one of the horizontal direction and the vertical direction, when the intra-prediction section 14a generates a prediction image by using a reference pixel located on at least one of the right side and the underside.SELECTED DRAWING: Figure 3

Description

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

H.265 / HEVC (High Efficiency Video Coding)
In a moving image (video) encoding method represented by the above, prediction is performed while switching between two types of prediction, inter prediction using temporal correlation between frames and intra prediction using spatial correlation within a frame. After the difference signal is generated, a stream obtained by performing orthogonal transform processing, loop filter processing, and entropy encoding processing is output.

In the intra prediction in HEVC, a total of 35 of the Planer prediction, the DC prediction, and the direction prediction.
Different types of modes are prepared and configured to perform intra prediction using adjacent decoded reference pixels according to the mode determined by the encoder.

Here, in intra prediction, a specified value (10) for a CU in which there is no adjacent decoded reference pixel, such as a coding target block (hereinafter referred to as “CU: Coding Unit”) located at the upper left in the frame. In the case of a bit moving image, “512”) is embedded to create a reference pixel used when generating a predicted image.

Further, in the conventional HEVC, since the encoding process is performed in the order of raster scan from the upper left, the reference pixel may not be decoded. In such a case, a predicted image is generated using a value obtained by extrapolating the nearest decoded reference pixel by zero order.

In particular, in the intra prediction in the conventional HEVC, the reference pixels located at the lower left and upper right of the CU have been decoded by the encoding process in the raster scan order shown in FIG. There are many cases (see FIG. 8B), and in such a case, performing direction prediction from a direction in which reference pixels that have not been decoded exist predicts that the prediction accuracy decreases and the coding efficiency decreases. There was a problem.

In order to solve such a problem, in intra prediction, a raster scan order (for example, Z-type) is used as an encoding process order for a plurality of transform blocks (hereinafter referred to as “TU: Transform Unit”) existing in a CU. In addition, there is known a technique for improving the prediction accuracy by providing a degree of freedom in the encoding order such as U type and X type (see Non-Patent Document 1).
.

In the example of FIGS. 8A and 8B, the direction prediction is performed in the direction from the lower left to the upper right (the direction opposite to the direction indicated by the dashed arrow in FIGS. 8A and 8B). The pixel on the broken line arrow is predicted using the lower left reference pixel. In the drawings of the present specification, the arrow indicating the direction (prediction direction) of the intra prediction mode is assumed to be directed from the pixel subject to intra prediction to the reference pixel, as described in the HEVC standard (the same applies hereinafter). .

Mochizuki et al., “Applied Intra Prediction Method Based on Average Coordinates”, Information Processing Society of Japan, Vol. 2012-AVM-77, No. 12

In the conventional HEVC, as shown in FIG. 9, intra prediction is prediction using the upper or left decoded reference pixels that are spatially adjacent, and the accuracy of the predicted image near the decoded reference pixels is high, Using the tendency that the accuracy of the predicted image far from the decoded reference pixel tends to be low, the discrete sine transform (hereinafter referred to as “DST”) from the left side and the upper side where the decoded reference pixel is located to the horizontal and vertical directions. Or orthogonal transform such as discrete cosine transform (hereinafter referred to as “DCT”) is applied to reduce the entropy of the residual signal.

In particular, the shape of the impulse response of the DST has an asymmetric shape in which one end is closed and the other end is widened as shown in FIG. 10, so that the generated residual signal is shown in FIG. By applying DST in accordance with the signal strength, it is possible to effectively reduce entropy.

As described above, in the technique described in Non-Patent Document 1, there are cases where lower or right reference pixels are used in intra prediction.

In such a case, in the residual signal, the signal strength on the lower side and the right side close to the position of the reference pixel decreases, and the signal strength on the upper side and the left side far from the position of the reference pixel tends to increase.
When orthogonal transformation is applied as usual, entropy may increase, which causes a problem of reducing encoding efficiency.

Therefore, the present invention has been made to solve the above-described problem, and is an encoding that can reduce an increase in entropy even when a lower or right reference pixel is used in intra prediction. An object is to provide a device, a decoding device, and a program.

A first feature of the present invention is an encoding device configured to divide and encode an original image in frame units constituting a moving image into encoding target blocks, and uses an intra prediction mode. An intra prediction unit configured to generate a prediction image, a residual signal generation unit configured to generate a residual signal based on a difference between the prediction image and the original image, and the intra prediction unit When the predicted image is generated using reference pixels located on at least one of the right side and the lower side, the base is orthogonal to the residual signal by inverting at least one base in the horizontal direction and the vertical direction. The gist of the present invention is to include an orthogonal transform unit configured to perform transform processing.

A second feature of the present invention is a decoding apparatus configured to divide a frame-unit original image constituting a moving image into decoding target blocks and decode the predicted image using an intra prediction mode. An intra prediction unit configured to generate and the intra prediction unit,
When the predicted image is generated using reference pixels located on at least one of the right side and the lower side, an inverse orthogonal transform process is performed on the transform coefficient after inverting at least one base in the horizontal direction and the vertical direction. And an inverse transform unit configured to generate a residual signal.

A third feature of the present invention is an encoding device configured to divide and encode an original image in frame units constituting a moving image into encoding target blocks, and uses an intra prediction mode. An intra prediction unit configured to generate a prediction image, a residual signal generation unit configured to generate a residual signal based on a difference between the prediction image and the original image, and the intra prediction unit When the predicted image is generated using reference pixels located on at least one of the right side and the lower side, the residual signal is inverted in at least one of the horizontal direction and the vertical direction and then subjected to orthogonal transform processing. The gist of the present invention is to include a configured orthogonal transform unit.

A fourth feature of the present invention is a decoding apparatus configured to divide a frame-unit original image constituting a moving image into decoding target blocks and decode the prediction image using an intra prediction mode. An intra prediction unit configured to generate and the intra prediction unit,
When the predicted image is generated using reference pixels located on at least one of the right side and the lower side, a signal obtained by performing an inverse orthogonal transform process on the transform coefficient is inverted to at least one of the horizontal direction and the vertical direction. And an inverse transform unit configured to generate a residual signal.

The fifth feature of the present invention is summarized as being a program for causing a computer to function as the encoding device described in the first or third feature.

The gist of the sixth feature of the present invention is that it is a program for causing a computer to function as the decoding device described in the second or fourth feature.

ADVANTAGE OF THE INVENTION According to this invention, even if it is a case where a lower or right side reference pixel is used in intra prediction, the encoding apparatus, decoding apparatus, and program which can reduce an entropy increase can be provided.

FIG. 1 is a functional block diagram of an encoding apparatus 1 according to the first embodiment. FIG. 2 is a diagram illustrating an example of intra prediction when TU partitioning is performed in the first embodiment. FIG. 3 is a flowchart showing the operation of the encoding apparatus 1 according to the first embodiment. FIG. 4 is a functional block diagram of the decoding device 3 according to the first embodiment. FIG. 5 is a flowchart showing the operation of the decoding device 3 according to the first embodiment. FIG. 6 is a flowchart showing the operation of the encoding apparatus 1 according to the second embodiment. FIG. 7 is a flowchart showing the operation of the decoding device 3 according to the second embodiment. FIG. 8 is a diagram for explaining a conventional HEVC. FIG. 9 is a diagram for explaining a conventional HEVC. FIG. 10 is a diagram for explaining a conventional HEVC. FIG. 11 is a diagram for explaining the conventional HEVC.

(First embodiment)
Hereinafter, the encoding device 1 and the decoding device 3 according to the first embodiment of the present invention will be described with reference to FIGS. Here, the encoding device 1 and the decoding device 3 according to this embodiment are HE
It is configured to support intra prediction in a moving picture coding scheme such as VC. Note that the encoding device 1 and the decoding device 3 according to the present embodiment are configured to be compatible with any video encoding scheme as long as the video encoding scheme performs intra prediction.

The encoding apparatus 1 according to the present embodiment is configured to divide and encode a frame-unit original image constituting a moving image into CUs. In addition, the encoding apparatus 1 according to the present embodiment includes a CU
May be divided into a plurality of TUs. Hereinafter, in the present embodiment, a case where a CU is divided into a plurality of TUs will be described as an example. However, the present invention is also applicable to a case where a CU is not divided into a plurality of TUs.

In the present embodiment, for a CU to be encoded that does not have an adjacent decoded reference pixel, such as the CU located at the upper left in the frame, a specified value (if a 10-bit moving image is “
512 ”) is created so as to generate reference pixels to be used when generating a predicted image, so that all pixels adjacent to the left side of the CU to be encoded can be used as reference pixels. To do.

As illustrated in FIG. 1, the encoding device 1 according to the present embodiment includes an intra prediction mode determination unit 11.
A TU partition determination unit 12, a coding order control unit 13, a sequential local decoded image generation unit 14, a memory 15, and an entropy coding unit 16.

The intra prediction mode determination unit 11 is configured to determine an optimal intra prediction mode to be applied to the CU.

The TU partition determination unit 12 is configured to determine whether to divide a CU into a plurality of TUs. In this embodiment, as a method of dividing a CU into a plurality of TUs, a case of four divisions is described as an example. However, regarding the number of divisions and division shapes when a CU is divided into a plurality of TUs. However, it is not limited to such a case.

The encoding order control unit 13 is configured to determine the encoding order of the TUs in the CU based on the intra prediction mode (for example, the direction of the intra prediction mode).

Specifically, when the TU division determination unit 12 determines to divide a CU into a plurality of TUs, the encoding order control unit 13 as illustrated in FIGS. 2 (a) to 2 (d). When the direction of the intra prediction mode determined by the intra prediction mode determination unit 11 is the direction from the lower left to the upper right (that is, when the direction prediction is performed from the lower left to the upper right), the CU
The conventional raster scan order (Z type as shown in FIG. 8 (a))
TU # A3 (lower left TU in CU # A) → TU # A4 (lower right TU in CU # A)
) → TU # A1 (upper left TU in CU # A) → TU # A2 (upper right TU in CU # A) or TU # A3 (lower left TU in CU # A) → TU # A1 (CU # A
Among the coding orders of TU # A4 (TU # A4 (bottom right TU) → TU # A2 (CU #A top right TU)), a predetermined coding order is adopted. It may be configured to.

In addition, the encoding order control unit 13 determines that the TU partition determination unit 12 determines to divide a CU into a plurality of TUs, and the intra prediction mode direction determined by the intra prediction mode determination unit 11 is When the direction is from the upper right to the lower left (that is, when direction prediction is performed from the upper right to the lower left), instead of the conventional raster scan order (Z type as shown in FIG. 8A), TU # A2 (upper right TU in CU # A) → TU # A4 (CU # A
TU # A1 (upper left TU in CU # A) → TU # A3 (lower left TU in CU # A) or TU # A2 (in CU # A) TU at the top right of) → TU #
A1 (upper left TU in CU # A) → TU # A4 (lower right TU in CU # A) → TU # A3
Of the encoding orders (lower left TU in CU # A), a predetermined encoding order may be adopted.

The sequential local decoded image generation unit 14 is configured to generate a local decoded image (decoded image for each TU) based on the encoding order determined by the encoding order control unit 13 and the method of dividing the CU into TUs. ing.

Specifically, the sequential local decoded image generation unit 14 follows the encoding order determined by the encoding order control unit 13 when the TU division determination unit 12 determines to divide a CU into a plurality of TUs. The local decoded image is generated sequentially.

As shown in FIG. 1, the sequential local decoded image generation unit 14 includes an intra prediction unit 14a, a residual signal generation unit 14b, an orthogonal transform / quantization unit 14c, an inverse quantization unit / inverse orthogonal transform unit 14d, ,
A local decoded image generation unit 14e.

The intra prediction unit 14a is configured to generate a prediction image using the intra prediction mode determined by the intra prediction mode determination unit 11. That is, the intra prediction unit 14a is configured to determine the position of the reference pixel used when generating a predicted image.

Specifically, the intra prediction unit 14a uses the TU partition determination unit 12 to convert a CU into a plurality of TUs.
2 (a) to 2 (d), when the intra prediction mode direction (prediction direction) is the direction from the lower left to the upper right, TU # A3 (CU
TU # A4 (lower right TU in CU # A) → TU # A1 (CU # A)
The upper left TU) → TU # A2 (upper right TU in CU # A), or T
U # A3 (lower left TU in CU # A) → TU # A1 (upper left TU in CU # A) → TU #
Of the coding orders of A4 (lower right TU in CU # A) → TU # A2 (upper right TU in CU # A), a prediction image is generated in a predetermined coding order. It may be.

Here, as shown in FIG. 2C and FIG. 2D, the intra prediction unit 14a TU # A1 (upper left TU in CU # A) in which the adjacent lower reference pixel is decoded And TU # A2
(Upper right TU in CU # A) may be configured to generate a prediction image using decoded reference pixels adjacent to the left side and the lower side.

Also, in the encoding device 1 according to the present embodiment, the intra prediction unit 14a is a case where the TU partition determination unit 12 determines to divide a CU into a plurality of TUs, and the direction of the intra prediction mode (prediction Direction) is a direction from the upper right to the lower left, TU # A2 (CU #
TU # A4 (lower right TU in CU # A) → TU # A1 (upper left TU in CU # A) → TU # A3 (lower left TU in CU # A) Coding order or TU
# A2 (upper right TU in CU # A) → TU # A1 (upper left TU in CU # A) → TU # A
4 (lower right TU in CU # A) → TU # A3 (lower left TU in CU # A) is configured to generate a predicted image in a predetermined coding order. It may be.

Here, the intra prediction unit 14a performs TU # A in which the adjacent right reference pixel is decoded.
For 1 (upper left TU in CU # A) and TU # A3 (lower left TU in CU # A),
You may be comprised so that a prediction image may be produced | generated using the decoded reference pixel adjacent to the upper side and the right side.

Alternatively, the intra prediction unit 14a calculates TU # A3 (lower left TU in CU # A) → TU # A.
4 (lower right TU in CU # A) → TU # A1 (upper left TU in CU # A) → TU # A2 (
The encoding order of TU # A in the upper right TU), or TU # A3 (lower left T in CU # A)
U) → TU # A1 (upper left TU in CU # A) → TU # A4 (lower right TU in CU # A)
→ When the encoding order of TU # A2 (upper right TU in CU # A) is used, the upper adjacent reference pixel is the decoded TU (located in the uppermost stage of the divided TU group) TU
In the example of FIG. 2, for TU # A1 and TU # A2), decoded reference pixels adjacent to the left side, upper side, and lower side of the TU are used instead of the intra prediction direction common in CU # A. You may be comprised so that prediction previously prescribed | regulated, such as linear interpolation, may be performed.

That is, the intra prediction unit 14a may be configured to generate a predicted image using decoded reference pixels adjacent in three directions such as the lower side, the left side, and the upper side.

Alternatively, the intra prediction unit 14a may be configured to generate a predicted image using decoded reference pixels adjacent in three directions such as the right side, the left side, and the upper side.

The residual signal generation unit 14b is configured to generate a residual signal based on the difference between the predicted image generated by the intra prediction unit 14a and the original image.

The orthogonal transform / quantization unit 14c is configured to perform orthogonal transform processing and quantization processing on the residual signal generated by the residual signal generation unit 14b to generate a quantized transform coefficient. .

Here, the orthogonal transform / quantization unit 14c determines whether to invert the base used for the orthogonal transform process based on the position of the reference pixel used when generating the predicted image determined by the intra prediction unit 14a. Is configured to do.

For example, when the intra prediction unit 14a generates a predicted image using reference pixels located on at least one of the right side and the lower side (that is, reference pixels adjacent to at least one of the right side and the lower side), The quantizing unit 14c is configured to perform orthogonal transform processing on the residual signal generated by the residual signal generating unit 14b after inverting at least one of the bases in the vertical direction and the horizontal direction. .

On the other hand, when the intra prediction unit 14a does not generate a predicted image using reference pixels located on either the right side or the lower side, the orthogonal transform / quantization unit 14c includes the residual signal generation unit 14
An orthogonal transform process is performed on the residual signal generated by b without inverting the base.

For example, when the intra prediction unit 14a generates a predicted image using reference pixels located on the left side and the lower side, the orthogonal transform / quantization unit 14c performs the orthogonal transform process after inverting the base in the vertical direction. You may be comprised so that it may give.

In addition, when the intra prediction unit 14a generates a prediction image using reference pixels located on the right side and the upper side, the orthogonal transform / quantization unit 14c performs the orthogonal transform process after inverting the horizontal base. It may be configured as follows.

Further, when the intra prediction unit 14a generates a prediction image using reference pixels located on the right side and the lower side, the orthogonal transform / quantization unit 14c performs orthogonality after inverting the bases in the vertical direction and the horizontal direction. You may be comprised so that a conversion process may be performed.

Note that the intra prediction unit 14a generates a predicted image using reference pixels located on at least one of the right side and the lower side, and the applied orthogonal transform process is an asymmetric orthogonal transform process (for example, DST or the like). ), The orthogonal transform / quantization unit 14c outputs the residual signal generation unit 1
The residual signal generated by 4b may be configured to invert at least one base in the vertical direction and the horizontal direction used for orthogonal transform processing.

That is, even when the intra prediction unit 14a generates a predicted image using reference pixels located on at least one of the right side and the lower side, the applied orthogonal transform process is a symmetric orthogonal transform process (for example, DCT Etc.), the orthogonal transform / quantization unit 14c may be configured not to invert the basis used for the orthogonal transform process with respect to the residual signal generated by the residual signal generation unit 14b. .

In addition, when the intra prediction unit 14a generates a prediction image using reference pixels adjacent in three directions such as the lower side, the left side, and the upper side, the orthogonal transform / quantization unit 14c includes the residual signal generation unit 14
The base used for the orthogonal transform process may not be inverted with respect to the residual signal generated by b.

According to such a configuration, in the residual signal, there is a high possibility that the signal intensity is low on both the upper side and the lower side close to the reference pixel. Therefore, the encoding apparatus 1 is not performed by performing the above inversion processing. The amount of processing can be reduced.

Further, when the intra prediction unit 14a generates a predicted image using reference pixels adjacent in three directions such as the right side, the left side, and the upper side, the orthogonal transform / quantization unit 14c includes the residual signal generation unit 14
The base used for the orthogonal transform process may not be inverted with respect to the residual signal generated by b.

According to such a configuration, in the residual signal, there is a high possibility that the signal strength is low on both the right side and the left side close to the reference pixel. Therefore, by performing the above inversion processing, The amount of processing can be reduced.

The inverse quantization unit / inverse orthogonal transform unit 14d performs inverse quantization processing and inverse orthogonal transform processing again on the quantized transform coefficient generated by the orthogonal transform / quantization unit 14c, and generates a residual signal. Configured to generate.

Here, when the orthogonal transform / quantization unit 14c inverts the base used in the orthogonal transform process, the inverse quantization unit / inverse orthogonal transform unit 14d performs the inverse orthogonal transform process after inverting the base. It is configured to apply.

The local decoded image generation unit 14e generates a local decoded image by adding the prediction image generated by the intra prediction unit 14a to the residual signal generated by the inverse quantization unit / inverse orthogonal transformation unit 14d. It is configured.

The memory 15 is configured to hold the local decoded image generated by the sequential local decoded image generation unit 14 so that it can be used as a reference image.

The entropy encoding unit 16 is configured to perform entropy encoding processing on flag information including the intra prediction mode determined by the intra prediction mode determination unit 11 and the quantized transform coefficient, and output the stream. Yes.

FIG. 3 shows a flowchart for explaining an example of the operation of the encoding apparatus 1 according to the present embodiment.

As illustrated in FIG. 3, in step S101, the encoding device 1 determines an optimal intra prediction mode to be applied to the CU.

In step S102, the encoding apparatus 1 determines whether to divide a CU into a plurality of TUs. When it is determined in step S102 that the CU is divided into a plurality of TUs, the operation proceeds to step S103. On the other hand, when it is determined in step S102 that the CU is not divided into a plurality of TUs, the operation proceeds to step S108.

When it is determined in step S103 that the direction of the intra prediction mode is the direction from the lower left to the upper right or the direction from the upper right to the lower left, the operation is performed in step S10.
Proceed to 5. On the other hand, when it is determined in step S103 that the direction of the intra prediction mode is other than the direction from the lower left to the upper right and the direction from the upper right to the lower left, the operation proceeds to step S104.

In step S104, the encoding apparatus 1 uses the conventional HEVC as the encoding order described above.
The raster scan order (Z-type as shown in FIG. 8A) used in FIG.

In step S108, the encoding apparatus 1 performs the TU on the TU to be encoded.
Predetermined prediction is performed using decoded reference pixels adjacent to the left side and the upper side.

When it is determined that the direction of the intra prediction mode is the direction from the lower left to the upper right (step S105), in step S106, the encoding device 1 sets TU # A3 (CU # A) as the above-described encoding order. TU # A4 (bottom right TU in CU # A) →
TU # A1 (upper left TU in CU # A) → TU # A2 (upper right TU in CU # A) or TU # A3 (lower left TU in CU # A) → TU # A1 (upper left TU in CU # A) → TU # A4 (lower right TU in CU # A) → TU # A2 (upper right TU in CU # A) in the encoding order defined in advance Encoding order is adopted.

On the other hand, when it is determined that the direction of the intra prediction mode is not the direction from the lower left to the upper right (step S105), in step S111, the encoding device 1 uses TU # A2 (CU #) as the encoding order described above. TU in the upper right in A) → TU # A4 (T in the lower right in CU # A)
U) → TU # A1 (upper left TU in CU # A) → TU # A3 (lower left TU in CU # A)
Or TU # A2 (upper right TU in CU # A) → TU # A1 (CU #
Among the coding orders of the upper left TU in A) → TU # A4 (lower right TU in CU # A) → TU # A3 (lower left TU in CU # A), a predetermined coding order is used. adopt.

In step S107, the encoding device 1 determines whether or not the reference pixel adjacent to the upper side of the TU to be encoded has been decoded. In step S107, if the decoding has been completed, the operation proceeds to step S109. If the decoding has not been completed, the operation proceeds to step S110.

In step S109, the encoding apparatus 1 performs the TU on the TU to be encoded.
Predetermined prediction is performed using decoded reference pixels adjacent to the left side, upper side, and lower side.

In step S110, the encoding device 1 performs the TU on the TU to be encoded.
Predetermined prediction is performed using the decoded reference pixels adjacent to the left side and the lower side.

In step S112, the encoding device 1 determines whether or not the reference pixel adjacent to the left side of the TU to be encoded has been decoded. In step S112, when the decoding has been completed, the operation proceeds to step S113. When the decoding has not been completed, the operation proceeds to step S114.

In step S113, the encoding apparatus 1 performs the TU on the TU to be encoded.
Predetermined prediction is performed using decoded reference pixels adjacent to the left side, the upper side, and the right side.

In step S114, the encoding apparatus 1 performs the TU on the TU to be encoded.
Predetermined prediction is performed using the decoded reference pixels adjacent to the right side and the upper side.

In step S115, the encoding apparatus 1 performs orthogonal transform processing on the residual signal after inverting the base, and performs subsequent processing.

In step S116, the encoding apparatus 1 performs orthogonal transform processing on the residual signal without inverting the base, and performs subsequent processing.

According to the encoding device 1 according to the present embodiment, when the prediction image is generated using the reference pixels located on at least one of the right side and the lower side, the base is inverted with respect to the residual signal. Therefore, an increase in entropy can be reduced.

Also, the decoding device 3 according to the present embodiment converts the original image in units of frames constituting the moving image into the CU.
It is comprised so that it may divide | segment into and decode. Also, the decoding device 3 according to the present embodiment is configured to be able to divide a CU into a plurality of TUs, similarly to the encoding device 1 according to the present embodiment.

As illustrated in FIG. 4, the decoding device 3 according to the present embodiment includes an entropy decoding unit 31, a decoding order control unit 32, a sequential local decoded image generation unit 33, and a memory 34.

The entropy decoding unit 31 is configured to decode transform coefficients, flag information, and the like from the stream output from the encoding device 1. Here, the transform coefficient is a quantized transform coefficient obtained by the encoding apparatus 1 as a signal encoded by dividing an original image in frame units into CUs. The flag information includes accompanying information such as a prediction mode.

The decoding order control unit 32 is configured to determine the decoding order of the TUs in the CU based on the intra prediction mode.

Specifically, the decoding order control unit 32 includes a flag indicating whether the TU division output by the entropy decoding unit 31 has been performed (whether the CU is divided into a plurality of TUs) and the intra prediction mode. The decoding order of TUs in the CU is determined according to the direction.

For example, the decoding order control unit 32, like the coding order control unit 13, is a case where the CU is divided into a plurality of TUs and the direction of the intra prediction mode is a direction from the lower left to the upper right. TU # A3 (lower left TU in CU # A) → TU # A4 (lower right TU in CU # A)
→ TU # A1 (upper left TU in CU # A) → decoding order TU # A2 (upper right TU in CU # A) or TU # A3 (lower left TU in CU # A) → TU # Of the decoding order of A1 (upper left TU in CU # A) → TU # A4 (lower right TU in CU # A) → TU # A2 (upper right TU in CU # A) The decoding process may be performed in order.

Similarly to the encoding order control unit 13, the decoding order control unit 32 is a case where the CU is divided into a plurality of TUs and the direction of the intra prediction mode is a direction from the upper right to the lower left. TU # A2 (upper right TU in CU # A) → TU # A4 (lower right TU in CU # A) →
Decoding order of TU # A1 (upper left TU in CU # A) → TU # A3 (lower left TU in CU # A), or TU # A2 (upper right TU in CU # A) → TU # A1 Of the decoding order of (upper left TU in CU # A) → TU # A4 (lower right TU in CU # A) → TU # A3 (lower left TU in CU # A) The decoding process may be performed in order.

The sequential local decoded image generation unit 33 includes the decoding order and the CU determined by the decoding order control unit 32.
The local decoded image (decoded image for each TU) is generated based on the method of dividing the TU into TUs.

Specifically, when the CU is divided into a plurality of TUs, the sequential local decoded image generation unit 33 performs the quantization output by the entropy decoding unit 31 according to the decoding order determined by the decoding order control unit 32. A local decoded image is generated by sequentially performing intra prediction, inverse quantization processing, and inverse orthogonal transform processing on the transformed coefficients.

As shown in FIG. 4, the sequential local decoded image generation unit 33 includes an intra prediction unit 33a, an inverse quantization / inverse conversion unit 33b, and a decoded image generation unit 33c.

The intra prediction unit 33a may be configured to generate a prediction image using the intra prediction mode output by the entropy decoding unit 31 according to the decoding order determined by the decoding order control unit 32.

Specifically, when the CU is divided into a plurality of TUs and the direction of the intra prediction mode is the direction from the lower left to the upper right, TU # A3 (C
TU # A4 (bottom right TU in CU # A) → TU # A1 (CU #)
Decoding order of upper left TU in A) → TU # A2 (upper right TU in CU # A), or T
U # A3 (lower left TU in CU # A) → TU # A1 (upper left TU in CU # A) → TU #
Among the decoding orders of A4 (lower right TU in CU # A) → TU # A2 (upper right TU in CU # A), a prediction image is generated in a predetermined decoding order. Also good.

Here, as shown in FIGS. 2C and 2D, the intra prediction unit 33a TU # A1 (upper left TU in CU # A) in which the adjacent lower reference pixel is decoded And TU # A2
(Upper right TU in CU # A) may be configured to generate a prediction image using decoded reference pixels adjacent to the left side and the lower side.

Further, in the decoding device 3 according to the present embodiment, the intra prediction unit 33a includes a TU having a plurality of TUs.
And the direction of the intra prediction mode (prediction direction) is the direction from the upper right to the lower left, TU # A2 (upper right TU in CU # A) → TU # A4 (CU #)
Decoding order of TU # A1 (upper left TU in CU # A) → TU # A3 (lower left TU in CU # A) → TU # A2 (in CU # A) TU at the top right of) → TU #
A1 (upper left TU in CU # A) → TU # A4 (lower right TU in CU # A) → TU # A3
Of the decoding order (the lower left TU in CU # A), the prediction image may be generated in a predetermined decoding order.

Here, the intra prediction unit 33a performs TU # A in which the adjacent right reference pixel is decoded.
For 1 (upper left TU in CU # A) and TU # A3 (lower left TU in CU # A),
The prediction image may be generated using the decoded reference pixels adjacent to the upper side and the right side.

Alternatively, the intra prediction unit 33a determines TU # A3 (lower left TU in CU # A) → TU # A.
4 (lower right TU in CU # A) → TU # A1 (upper left TU in CU # A) → TU # A2 (
Decoding order of TU # A upper right TU), or TU # A3 (lower left TU in CU # A)
) → TU # A1 (upper left TU in CU # A) → TU # A4 (lower right TU in CU # A) →
When the decoding order of TU # A2 (upper right TU in CU # A) is used, the TU in which the reference pixel adjacent on the upper side has been decoded (TU located in the uppermost stage among the divided TU groups, In the example of FIG. 2, for TU # A1 and TU # A2), linearity using decoded reference pixels adjacent to the left side, upper side, and lower side of the TU, not the common intra prediction direction in CU # A. You may be comprised so that prediction previously prescribed | regulated, such as interpolation, may be performed.

The inverse quantization / inverse transform unit 33b performs an inverse quantization process and an inverse transform process (for example, an inverse orthogonal transform process) on the quantized transform coefficient output from the entropy decoding unit 31, thereby obtaining a residual. It is configured to generate a signal.

For example, when the intra prediction unit 33a generates a prediction image using reference pixels located on at least one of the right side and the lower side (that is, reference pixels adjacent to at least one of the right side and the lower side), inverse quantization is performed. The inverse transform unit 33b is configured to generate a residual signal by performing inverse orthogonal transform processing on the transform coefficient after inverting at least one base in the vertical direction and the horizontal direction. Yes.

On the other hand, if the intra prediction unit 33a has not generated a predicted image using reference pixels located on either the right side or the lower side, the inverse quantization / inverse transform unit 33b does not invert the base. The residual signal is generated by performing an inverse orthogonal transform process on the transform coefficient described above.

For example, when the intra prediction unit 33a generates a prediction image using the reference pixels located on the left side and the lower side, the inverse quantization / inverse conversion unit 33b inverts the vertical base and The residual signal may be generated by performing an inverse orthogonal transform process on the transform coefficient.

In addition, when the intra prediction unit 33a generates a prediction image using reference pixels located on the right side and the upper side, the inverse quantization / inverse conversion unit 33b inverts the horizontal base and The residual signal may be generated by performing inverse orthogonal transform processing on the transform coefficient.

Further, when the intra prediction unit 33a generates a prediction image using the reference pixels located on the right side and the lower side, the inverse quantization / inverse conversion unit 33b is obtained by inverting the bases in the vertical direction and the horizontal direction. Thus, the residual signal may be generated by performing an inverse orthogonal transform process on the transform coefficient described above.

Note that when the intra prediction unit 33a generates a prediction image using reference pixels located on at least one of the right side and the lower side, and the applied orthogonal transform process is an asymmetric orthogonal transform process (for example, DST or the like). ), The inverse quantization / inverse transform unit 33b is configured to generate a residual signal by performing an inverse orthogonal transform process on the transform coefficient after inverting the base. May be.

That is, even when the intra prediction unit 33a generates a prediction image using reference pixels located on at least one of the right side and the lower side, the applied orthogonal transformation process is a symmetric orthogonal transformation process (for example, DCT Etc.), the inverse quantization / inverse transform unit 33b is configured to generate a residual signal by performing the inverse orthogonal transform process on the transform coefficient without inverting the base. May be.

In addition, when the intra prediction unit 33a generates a prediction image using reference pixels adjacent in three directions such as the lower side, the left side, and the upper side, the inverse quantization / inverse conversion unit 33b You may comprise so that the base used for an inverse orthogonal transformation process may not be reversed.

In addition, when the intra prediction unit 33a generates a predicted image using reference pixels adjacent in three directions such as the right side, the left side, and the upper side, the inverse quantization / inverse transform unit 33b performs an inverse operation on the transform coefficient described above. The base used for the orthogonal transformation process may be configured not to be inverted.

The decoded image generation unit 33c is configured to generate a local decoded image by adding the prediction image generated by the intra prediction unit 33a and the residual signal generated by the inverse quantization / inverse conversion unit 33b. .

The memory 34 is configured to hold the local decoded image generated by the sequential local decoded image generation unit 33 so as to be usable as a reference image for intra prediction and inter prediction.

FIG. 5 shows a flowchart for explaining an example of the operation for determining the decoding order described above by the decoding device 3 according to the present embodiment.

As illustrated in FIG. 5, in step S <b> 201, the decoding device 3 acquires an intra prediction mode from the stream output from the encoding device 1.

In step S202, the decoding device 3 determines whether the CU is divided into a plurality of TUs based on the flag information included in the stream output from the encoding device 1. If it is determined in step S202 that the CU is divided into a plurality of TUs, the operation proceeds to step S203. On the other hand, in step S202, the CU
If it is determined that is not divided into a plurality of TUs, the operation proceeds to step S205.

In step S205, the decoding apparatus 3 performs a predetermined prediction on the decoding target TU using the decoded reference pixels adjacent to the left side and the upper side of the TU.

In step S203, the decoding device 3 determines whether or not the direction of the intra prediction mode is a direction from the lower left to the upper right or a direction from the upper right to the lower left. When it is determined in step S203 that the direction of the intra prediction mode is the direction from the lower left to the upper right or the direction from the upper right to the lower left, the operation proceeds to step S206.

On the other hand, when it is determined in step S203 that the direction of the intra prediction mode is other than the direction from the lower left to the upper right and the direction from the upper right to the lower left, the operation proceeds to step S204.

In step S204, the decoding apparatus 3 employs the conventional raster scan order (Z type as shown in FIG. 8A) used in HEVC as the above-described decoding order.

When it is determined that the direction of the intra prediction mode is the direction from the lower left to the upper right (step S206), in step S207, the decoding device 3 determines that the decoding order is as described above.
TU # A3 (lower left TU in CU # A) → TU # A4 (lower right TU in CU # A) → TU
Decoding order of # A1 (upper left TU in CU # A) → TU # A2 (upper right TU in CU # A) or TU # A3 (lower left TU in CU # A) → TU # A1 ( TU # A, upper left TU) → TU # A4 (lower right TU in CU # A) → TU # A2 (upper right TU in CU # A)
The decoding order defined in advance is adopted among the decoding orders.

On the other hand, when it is determined that the direction of the intra prediction mode is not the direction from the lower left to the upper right (step S206), in step S211, the decoding device 3 uses TU # A2 (in CU # A as the decoding order). TU in the upper right of) → TU # A4 (lower right TU in CU # A)
→ TU # A1 (upper left TU in CU # A) → TU # A3 (lower left TU in CU # A) or TU # A2 (upper right TU in CU # A) → TU # Of the decoding order A1 (upper left TU in CU # A) → TU # A4 (lower right TU in CU # A) → TU # A3 (lower left TU in CU # A) Adopt order.

In step S208, the decoding device 3 determines whether or not the reference pixel adjacent to the upper side of the decoding target TU has been decoded. If it has been decoded in step S208, the operation proceeds to step S209. If it has not been decoded, the operation is performed in step S209.
Proceed to 210.

In step S209, the decoding device 3 performs a predetermined prediction on the decoding target TU using decoded reference pixels adjacent to the left side, the upper side, and the lower side of the TU.

In step S210, the decoding device 3 performs a predetermined prediction on the decoding target TU using the decoded reference pixels adjacent to the left side and the lower side of the TU.

In step S212, the decoding device 3 determines whether the reference pixel adjacent to the left side of the decoding target TU has been decoded. In step S212, when the decoding has been completed, the operation proceeds to step S213. When the decoding has not been completed, the operation proceeds to step S213.
Proceed to 214.

In step S213, the decoding device 3 performs a predetermined prediction on the decoding target TU using the decoded reference pixels adjacent to the left side, the upper side, and the right side of the TU.

In step S214, the decoding device 3 performs a predetermined prediction on the decoding target TU using the decoded reference pixels adjacent to the right side and the upper side of the TU.

In step S215, the decoding device 3 inverts at least one of the bases in the vertical direction and the horizontal direction, and then performs an inverse orthogonal transform process on the transform coefficient described above to generate a residual signal, and then Process.

In step S216, the decoding device 3 generates a residual signal by performing an inverse orthogonal transform process on the above-described transform coefficient without inverting the base, and performs the subsequent processing.

According to the decoding device 3 according to the present embodiment, when a predicted image is generated using reference pixels located on at least one of the right side and the lower side, the basis is inverted with respect to the transform coefficient and then inverted. Since it is configured to perform orthogonal transform processing, an increase in entropy can be reduced.

(Second Embodiment)
Hereinafter, with reference to FIGS. 6 and 7, the encoding device 1 and the decoding device 3 according to the second embodiment of the present invention are described below. The difference will be described below.

In the encoding device 1 according to the present embodiment, when the intra prediction unit 14a generates a predicted image using reference pixels located on at least one of the right side and the lower side, the orthogonal transform / quantization unit 1
4c is configured to invert the residual signal generated by the residual signal generation unit 14b in at least one of the horizontal direction and the vertical direction and then perform orthogonal transformation processing.

For example, when the intra prediction unit 14a generates a predicted image using the reference pixels located on the left side and the lower side, the orthogonal transform / quantization unit 14c uses the residual signal generated by the residual signal generation unit 14b. May be configured to perform orthogonal transformation processing after inverting in the vertical direction.

In addition, when the intra prediction unit 14a generates a prediction image using reference pixels located on the right side and the upper side, the orthogonal transform / quantization unit 14c uses the residual signal generated by the residual signal generation unit 14b. You may comprise so that an orthogonal transformation process may be performed after inverting in a horizontal direction.

Further, when the intra prediction unit 14a generates a prediction image using the reference pixels located on the right side and the lower side, the orthogonal transform / quantization unit 14c generates the residual signal generated by the residual signal generation unit 14b. It is also possible to perform orthogonal transformation processing after inverting in the vertical and horizontal directions.

Note that the intra prediction unit 14a generates a predicted image using reference pixels located on at least one of the right side and the lower side, and the applied orthogonal transform process is an asymmetric orthogonal transform process (for example, DST or the like). ), The orthogonal transform / quantization unit 14c outputs the residual signal generation unit 1
The residual signal generated by 4b may be inverted in at least one of the horizontal direction and the vertical direction and then subjected to orthogonal transform processing.

That is, even when the intra prediction unit 14a generates a predicted image using reference pixels located on at least one of the right side and the lower side, the applied orthogonal transform process is a symmetric orthogonal transform process (for example, DCT Or the like, the orthogonal transform / quantization unit 14c may be configured to perform orthogonal transform processing without inverting the residual signal generated by the residual signal generation unit 14b.

In addition, when the intra prediction unit 14a generates a prediction image using reference pixels adjacent in three directions such as the lower side, the left side, and the upper side, the orthogonal transform / quantization unit 14c includes the residual signal generation unit 14
You may comprise so that an orthogonal transformation process may be performed, without inverting the residual signal produced | generated by b.

Further, when the intra prediction unit 14a generates a predicted image using reference pixels adjacent in three directions such as the right side, the left side, and the upper side, the orthogonal transform / quantization unit 14c includes the residual signal generation unit 14
You may comprise so that an orthogonal transformation process may be performed, without inverting the residual signal produced | generated by b.

FIG. 6 shows a flowchart for explaining an example of the operation of the encoding apparatus 1 according to the present embodiment.

As shown in FIG. 6, the operations in steps S301 to S314 are the same as those in step S10 shown in FIG.
1 to S114.

In step 315, the encoding apparatus 1 performs orthogonal transform processing without inverting the above-described residual signal, and performs subsequent processing.

In step 316, the encoding apparatus 1 inverts the above-described residual signal in at least one of the horizontal direction and the vertical direction, performs orthogonal transform processing, and then performs subsequent processing.

According to the encoding device 1 according to the present embodiment, when a predicted image is generated using reference pixels located on at least one of the right side and the lower side, the residual signal is set to at least one of the horizontal direction and the vertical direction. Since it is configured to perform the orthogonal transformation process after being inverted, an increase in entropy can be reduced.

In the decoding device 3 according to the present embodiment, when the intra prediction unit 33a generates a prediction image using reference pixels located on at least one of the right side and the lower side, the inverse quantization / inverse conversion unit 33 is used.
b is configured to invert a signal obtained by performing inverse orthogonal transform processing on the transform coefficient described above in at least one of the horizontal direction and the vertical direction.

For example, when the intra prediction unit 33a generates a predicted image using reference pixels located on the left side and the lower side, the inverse quantization / inverse transform unit 33b performs an inverse orthogonal transform process on the transform coefficient described above. The signal obtained in (1) may be inverted in the vertical direction.

In addition, when the intra prediction unit 33a generates a prediction image using reference pixels located on the right side and the upper side, the inverse quantization / inverse transform unit 33b performs inverse orthogonal transform processing on the transform coefficient described above. The obtained signal may be configured to be inverted in the horizontal direction.

Further, when the intra prediction unit 33a generates a predicted image using the reference pixels located on the right side and the lower side, the inverse quantization / inverse transform unit 33b performs an inverse orthogonal transform process on the transform coefficient described above. The signal obtained in (1) may be inverted in the vertical and horizontal directions.

Note that when the intra prediction unit 33a generates a prediction image using reference pixels located on at least one of the right side and the lower side, and the applied orthogonal transform process is an asymmetric orthogonal transform process (for example, DST or the like). ), The inverse quantization / inverse transform unit 33b may be configured to invert a signal obtained by performing an inverse orthogonal transform process on the transform coefficient described above.

That is, even when the intra prediction unit 33a generates a prediction image using reference pixels located on at least one of the right side and the lower side, the applied orthogonal transformation process is a symmetric orthogonal transformation process (for example, DCT Etc.), the inverse quantization / inverse transform unit 33b may be configured not to invert a signal obtained by performing an inverse orthogonal transform process on the transform coefficient described above.

In addition, when the intra prediction unit 33a generates a prediction image using reference pixels adjacent in three directions such as the lower side, the left side, and the upper side, the inverse quantization / inverse transform unit 33b performs inverse orthogonal transform to the transform coefficient described above. You may be comprised so that the signal obtained by processing may not be reversed.

In addition, when the intra prediction unit 33a generates a prediction image using reference pixels adjacent in three directions such as the right side, the left side, and the upper side, the inverse quantization / inverse transform unit 33b performs an inverse orthogonal transform process on the transform coefficient described above. The signal obtained by applying may be configured not to be inverted.

FIG. 7 shows a flowchart for explaining an example of the operation of the decoding device 3 according to the present embodiment.

As shown in FIG. 7, the operations in steps S401 to S414 are the same as those in step S20 shown in FIG.
1 to S214.

In step 415, the decoding device 3 performs the subsequent processing without inverting the signal obtained by performing the inverse orthogonal transform process on the transform coefficient described above.

In step 416, the decoding device 3 inverts the signal obtained by performing the inverse orthogonal transform process on the above-described transform coefficient in at least one of the horizontal direction and the vertical direction, and then performs the subsequent process.

(Other embodiments)
As described above, the present invention has been described by using the above-described embodiment. However, it should not be understood that the description and drawings constituting a part of the disclosure in the embodiment limit the present invention. From this disclosure, various alternative embodiments, examples and operational techniques will be apparent to those skilled in the art.

Further, although not particularly mentioned in the above-described embodiment, a program for causing a computer to execute each process performed by the above-described encoding device 1 and decoding device 3 may be provided. Such a program may be recorded on a computer-readable medium. If a computer readable medium is used, such a program can be installed in the computer. Here, the computer-readable medium on which such a program is recorded may be a non-transitory recording medium. The non-transitory recording medium is not particularly limited, but may be a recording medium such as a CD-ROM or a DVD-ROM.

Alternatively, a chip configured by a memory that stores a program for realizing at least a part of the functions in the encoding device 1 and the decoding device 3 and a processor that executes the program stored in the memory may be provided. Good.

DESCRIPTION OF SYMBOLS 1 ... Coding apparatus 11 ... Intra prediction mode determination part 12 ... TU division | segmentation determination part 13 ... Coding order control part 14 ... Sequential local decoded image generation part 14a ... Intra prediction part 14b ... Residual signal generation part 14c ... Orthogonal transformation * Quantization unit 14d ... inverse quantization unit / inverse orthogonal transformation unit 14e ... local decoded image generation unit 15 ... memory 16 ... entropy encoding unit 3 ... decoding device 31 ... entropy decoding unit 32 ... decoding order control unit 33 ... sequential local decoding Image generation unit 33a ... Intra prediction unit 33b ... Inverse quantization / inverse conversion unit 33c ... Decoded image generation unit 34 ... Memory

Claims (10)

An encoding device configured to divide and encode an original image of a frame unit constituting a moving image into encoding target blocks,
An intra prediction unit configured to generate a prediction image using the intra prediction mode;
A residual signal generator configured to generate a residual signal based on a difference between the predicted image and the original image;
When the intra prediction unit generates the predicted image using reference pixels located on at least one of the right side and the lower side, the base of at least one of the horizontal direction and the vertical direction is inverted with respect to the residual signal. And an orthogonal transform unit configured to perform an orthogonal transform process. When the intra prediction unit generates the predicted image using reference pixels located on at least one of the right side and the lower side, and the orthogonal transform process to be applied is an asymmetric orthogonal transform process, The orthogonal transform unit is configured to perform orthogonal transform processing on the residual signal after inverting at least one base in a horizontal direction and a vertical direction. Encoding device. A decoding device configured to divide and decode an original image of a frame unit constituting a moving image into encoding target blocks,
An intra prediction unit configured to generate a prediction image using the intra prediction mode;
When the intra prediction unit generates the prediction image using reference pixels located on at least one of the right side and the lower side, the transform coefficient is obtained after inverting at least one base in the horizontal direction and the vertical direction. And an inverse transform unit configured to generate a residual signal by performing an inverse orthogonal transform process. When the intra prediction unit generates the predicted image using reference pixels located on at least one of the right side and the lower side, and the orthogonal transform process to be applied is an asymmetric orthogonal transform process, The inverse transform unit is configured to generate a residual signal by performing an inverse orthogonal transform process on the transform coefficient after inverting at least one base in the horizontal direction and the vertical direction. The decoding device according to claim 3. An encoding device configured to divide and encode an original image of a frame unit constituting a moving image into encoding target blocks,
An intra prediction unit configured to generate a prediction image using the intra prediction mode;
A residual signal generator configured to generate a residual signal based on a difference between the predicted image and the original image;
When the intra prediction unit generates the predicted image using reference pixels located on at least one of the right side and the lower side, the residual signal is inverted in at least one of the horizontal direction and the vertical direction and then orthogonally transformed. An encoding device comprising: an orthogonal transform unit configured to perform processing. When the intra prediction unit generates the predicted image using reference pixels located on at least one of the right side and the lower side, and the orthogonal transform process to be applied is an asymmetric orthogonal transform process, 6. The encoding apparatus according to claim 5, wherein the orthogonal transform unit is configured to perform an orthogonal transform process after inverting the residual signal in at least one of a horizontal direction and a vertical direction. A decoding device configured to divide and decode an original image of a frame unit constituting a moving image into encoding target blocks,
An intra prediction unit configured to generate a prediction image using the intra prediction mode;
When the intra prediction unit generates the predicted image using reference pixels located on at least one of the right side and the lower side, a signal obtained by performing an inverse orthogonal transform process on the transform coefficient is obtained in the horizontal direction and the vertical direction. And a reverse transform unit configured to generate a residual signal by inverting at least one of the decoding device. When the intra prediction unit generates the predicted image using reference pixels located on at least one of the right side and the lower side, and the orthogonal transform process to be applied is an asymmetric orthogonal transform process, The inverse transform unit is configured to generate a residual signal by inverting a signal obtained by performing an inverse orthogonal transform process on the transform coefficient in at least one of a horizontal direction and a vertical direction. The decoding device according to claim 7. The program for functioning a computer as an encoding apparatus of Claim 1, 2, 5 or 6. The program for functioning a computer as a decoding apparatus of Claim 3, 4, 7, or 8.