WO2018097115A1

WO2018097115A1 - Encoding device, decoding device, encoding method, and decoding method

Info

Publication number: WO2018097115A1
Application number: PCT/JP2017/041749
Authority: WO
Inventors: 安倍　清史; 西　孝啓; 遠間　正真; 橋本　隆
Original assignee: パナソニックインテレクチュアルプロパティコーポレーションオブアメリカ
Priority date: 2016-11-22
Filing date: 2017-11-21
Publication date: 2018-05-31
Also published as: TW201826796A

Abstract

An encoding device (100) extracts, from a plurality of candidate motion vectors, one or more predicted motion vector candidates for a block to be encoded, derives a motion vector for the block to be encoded by referring to a reference picture included in a moving image, encodes the difference between a predicted motion vector from among the extracted one or more predicted motion vector candidates and the derived motion vector for the block to be encoded, and uses the derived motion vector for the block to be encoded to perform motion compensation for the block to be encoded. In the extraction of the one or more predicted motion vector candidates, the device extracts all of the one or more predicted motion vector candidates on the basis of evaluation results for each of the plurality of candidate motion vectors obtained using a reconstructed image of an encoded region in the moving image without using the image region of the block to be encoded.

Description

Encoding device, decoding device, encoding method, and decoding method

The present disclosure relates to an encoding device that encodes a moving image including a plurality of pictures.

Conventionally, as a standard for encoding moving images, H.264 265 exists. H. H.265 is also called HEVC (High Efficiency Video Coding).

However, while further improvement of the encoding efficiency is desired, there is a problem that the processing load increases due to the improvement of the encoding efficiency.

Therefore, the present disclosure provides an encoding device and the like that may be able to improve the encoding efficiency while suppressing an increase in processing load.

An encoding apparatus according to an aspect of the present disclosure is an encoding apparatus that encodes a moving image, and includes a processing circuit and a memory connected to the processing circuit, and the processing circuit uses the memory. A plurality of candidate motion vectors based on respective motion vectors of a plurality of encoded blocks corresponding to the encoding target block in the moving image, and from the plurality of candidate motion vectors, the encoding target block At least one prediction motion vector candidate is extracted, a motion vector of the coding target block is derived with reference to a reference picture included in the moving image, and prediction of the extracted at least one prediction motion vector candidate is performed The difference between the motion vector and the derived motion vector of the encoding target block is encoded, and the derived encoding target block of the encoding target block is encoded. Motion vector is used to perform motion compensation on the encoding target block, and in the extraction of the at least one prediction motion vector candidate, all of the at least one prediction motion vector candidate is converted into an image area of the encoding target block. Is extracted based on the respective evaluation results of the plurality of candidate motion vectors using the reconstructed image of the encoded region in the moving image.

Note that these comprehensive or specific aspects may be realized by a system, apparatus, method, integrated circuit, computer program, or non-transitory recording medium such as a computer-readable CD-ROM. The present invention may be realized by any combination of an apparatus, a method, an integrated circuit, a computer program, and a recording medium.

The encoding apparatus and the like according to one aspect of the present disclosure can improve encoding efficiency while suppressing an increase in processing load.

FIG. 1 is a block diagram showing a functional configuration of the encoding apparatus according to Embodiment 1. FIG. 2 is a diagram illustrating an example of block division in the first embodiment. FIG. 3 is a table showing conversion basis functions corresponding to each conversion type. FIG. 4A is a diagram illustrating an example of the shape of a filter used in ALF. FIG. 4B is a diagram illustrating another example of the shape of a filter used in ALF. FIG. 4C is a diagram illustrating another example of the shape of a filter used in ALF. FIG. 5 is a diagram illustrating 67 intra prediction modes in intra prediction. FIG. 6 is a diagram for explaining pattern matching (bilateral matching) between two blocks along the motion trajectory. FIG. 7 is a diagram for explaining pattern matching (template matching) between a template in the current picture and a block in the reference picture. FIG. 8 is a diagram for explaining a model assuming constant velocity linear motion. FIG. 9 is a diagram for explaining the derivation of motion vectors in units of sub-blocks based on the motion vectors of a plurality of adjacent blocks. FIG. 10 is a block diagram showing a functional configuration of the decoding apparatus according to the first embodiment. FIG. 11 is a flowchart illustrating motion compensation by another encoding device that is the basis of the present disclosure. FIG. 12 is a flowchart illustrating motion compensation by another decoding device that is the basis of the present disclosure. FIG. 13 is a diagram for explaining an example of an evaluation value calculation method. FIG. 14 is a diagram for explaining another example of the evaluation value calculation method. FIG. 15 is a flowchart illustrating an example of motion compensation by the encoding device according to the second embodiment. FIG. 16 is a flowchart illustrating an example of motion compensation by the decoding apparatus according to the second embodiment. FIG. 17 is a flowchart illustrating another example of motion compensation by the encoding apparatus according to the second embodiment. FIG. 18 is a flowchart showing another example of motion compensation by the decoding apparatus in the second embodiment. FIG. 19 is a diagram for explaining a method of extracting N predicted motion vector candidates from a plurality of candidate motion vectors in the second embodiment. FIG. 20 is a flowchart illustrating a method of selecting a motion vector predictor by the encoding device and the decoding device according to Embodiment 3. FIG. 21 is a flowchart illustrating an example of motion compensation by the encoding device according to the fourth embodiment. FIG. 22 is a flowchart illustrating an example of motion compensation by the decoding apparatus according to the fourth embodiment. FIG. 23 is a diagram for explaining a method of extracting motion vector predictor candidates according to the fourth embodiment. FIG. 24 is a diagram illustrating an example of a common candidate list in the fourth embodiment. FIG. 25 is a block diagram illustrating an implementation example of the encoding device according to each embodiment. FIG. 26 is a block diagram illustrating an implementation example of the decoding apparatus according to each embodiment. FIG. 27 is an overall configuration diagram of a content supply system that implements a content distribution service. FIG. 28 is a diagram illustrating an example of a coding structure at the time of scalable coding. FIG. 29 is a diagram illustrating an example of a coding structure at the time of scalable coding. FIG. 30 shows an example of a web page display screen. FIG. 31 is a diagram illustrating an example of a web page display screen. FIG. 32 is a diagram illustrating an example of a smartphone. FIG. 33 is a block diagram illustrating a configuration example of a smartphone.

Hereinafter, embodiments will be specifically described with reference to the drawings.

It should be noted that each of the embodiments described below shows a comprehensive or specific example. Numerical values, shapes, materials, components, arrangement positions and connection forms of components, steps, order of steps, and the like shown in the following embodiments are merely examples, and are not intended to limit the scope of the claims. In addition, among the constituent elements in the following embodiments, constituent elements that are not described in the independent claims indicating the highest concept are described as optional constituent elements.

(Embodiment 1)
First, an outline of the first embodiment will be described as an example of an encoding device and a decoding device to which the processing and / or configuration described in each aspect of the present disclosure to be described later can be applied. However, the first embodiment is merely an example of an encoding device and a decoding device to which the processing and / or configuration described in each aspect of the present disclosure can be applied, and the processing and / or processing described in each aspect of the present disclosure. The configuration can also be implemented in an encoding device and a decoding device different from those in the first embodiment.

When applying the processing and / or configuration described in each aspect of the present disclosure to Embodiment 1, for example, one of the following may be performed.

(1) The encoding apparatus or decoding apparatus according to the first embodiment corresponds to the constituent elements described in each aspect of the present disclosure among a plurality of constituent elements constituting the encoding apparatus or decoding apparatus. Replacing the constituent elements with constituent elements described in each aspect of the present disclosure (2) A plurality of constituent elements constituting the encoding apparatus or decoding apparatus with respect to the encoding apparatus or decoding apparatus of the first embodiment The constituent elements corresponding to the constituent elements described in each aspect of the present disclosure are added to the present disclosure after arbitrary changes such as addition, replacement, and deletion of functions or processes to be performed on some constituent elements among the constituent elements. (3) Addition of processes and / or a plurality of processes included in the method to the method performed by the encoding apparatus or decoding apparatus of the first embodiment home Replace any processing corresponding to the processing described in each aspect of the present disclosure with the processing described in each aspect of the present disclosure after performing arbitrary changes such as replacement and deletion of a part of the processing (4) Implementation The constituent elements described in each aspect of the present disclosure, the constituent elements described in the respective aspects of the present disclosure are part of the plurality of constituent elements constituting the encoding device or the decoding device of the first embodiment (5) Encoding device according to Embodiment 1 that is implemented in combination with a component that includes a part of the functions provided, or a component that performs a part of processing performed by a component described in each aspect of the present disclosure Alternatively, a component provided with a part of the functions provided by some of the components included in the decoding device, or a plurality of components included in the encoding device or the decoding device according to the first embodiment. Some of A component that performs a part of processing performed by a component is a component that is described in each aspect of the present disclosure, a component that includes a part of a function included in the component described in each aspect of the present disclosure, (6) A method performed by the encoding device or the decoding device according to Embodiment 1 is performed in combination with a component that performs a part of processing performed by the component described in each aspect of the disclosure. Of the plurality of processes included in the method, the process corresponding to the process described in each aspect of the present disclosure is replaced with the process described in each aspect of the present disclosure. (7) The encoding apparatus according to the first embodiment or A part of the plurality of processes included in the method performed by the decoding device is performed in combination with the processes described in each aspect of the present disclosure

Note that the processes and / or configurations described in each aspect of the present disclosure are not limited to the above examples. For example, the present invention may be implemented in an apparatus used for a different purpose from the moving picture / picture encoding apparatus or moving picture / picture decoding apparatus disclosed in the first embodiment, and the processing and / or described in each aspect. The configuration may be implemented alone. Moreover, you may implement combining the process and / or structure which were demonstrated in the different aspect.

[Outline of encoding device]
First, the outline | summary of the encoding apparatus which concerns on Embodiment 1 is demonstrated. FIG. 1 is a block diagram showing a functional configuration of encoding apparatus 100 according to Embodiment 1. The encoding device 100 is a moving image / image encoding device that encodes moving images / images in units of blocks.

As shown in FIG. 1, an encoding apparatus 100 is an apparatus that encodes an image in units of blocks, and includes a dividing unit 102, a subtracting unit 104, a transforming unit 106, a quantizing unit 108, and entropy encoding. Unit 110, inverse quantization unit 112, inverse transform unit 114, addition unit 116, block memory 118, loop filter unit 120, frame memory 122, intra prediction unit 124, inter prediction unit 126, A prediction control unit 128.

The encoding device 100 is realized by, for example, a general-purpose processor and a memory. In this case, when the software program stored in the memory is executed by the processor, the processor performs the division unit 102, the subtraction unit 104, the conversion unit 106, the quantization unit 108, the entropy encoding unit 110, and the inverse quantization unit 112. , An inverse transform unit 114, an addition unit 116, a loop filter unit 120, an intra prediction unit 124, an inter prediction unit 126, and a prediction control unit 128. The encoding apparatus 100 includes a dividing unit 102, a subtracting unit 104, a transforming unit 106, a quantizing unit 108, an entropy coding unit 110, an inverse quantizing unit 112, an inverse transforming unit 114, an adding unit 116, and a loop filter unit 120. The intra prediction unit 124, the inter prediction unit 126, and the prediction control unit 128 may be implemented as one or more dedicated electronic circuits.

Hereinafter, each component included in the encoding device 100 will be described.

[Division part]
The dividing unit 102 divides each picture included in the input moving image into a plurality of blocks, and outputs each block to the subtracting unit 104. For example, the dividing unit 102 first divides a picture into blocks of a fixed size (for example, 128 × 128). This fixed size block may be referred to as a coding tree unit (CTU). Then, the dividing unit 102 divides each of the fixed size blocks into blocks of a variable size (for example, 64 × 64 or less) based on recursive quadtree and / or binary tree block division. . This variable size block may be referred to as a coding unit (CU), a prediction unit (PU) or a transform unit (TU). In the present embodiment, CU, PU, and TU do not need to be distinguished, and some or all blocks in a picture may be processing units of CU, PU, and TU.

FIG. 2 is a diagram showing an example of block division in the first embodiment. In FIG. 2, a solid line represents a block boundary by quadtree block division, and a broken line represents a block boundary by binary tree block division.

Here, the block 10 is a 128 × 128 pixel square block (128 × 128 block). The 128 × 128 block 10 is first divided into four square 64 × 64 blocks (quadtree block division).

The upper left 64 × 64 block is further divided vertically into two rectangular 32 × 64 blocks, and the left 32 × 64 block is further divided vertically into two rectangular 16 × 64 blocks (binary tree block division). As a result, the upper left 64 × 64 block is divided into two 16 × 64 blocks 11 and 12 and a 32 × 64 block 13.

The upper right 64 × 64 block is horizontally divided into two rectangular 64 × 32 blocks 14 and 15 (binary tree block division).

The lower left 64x64 block is divided into four square 32x32 blocks (quadrant block division). Of the four 32 × 32 blocks, the upper left block and the lower right block are further divided. The upper left 32 × 32 block is vertically divided into two rectangular 16 × 32 blocks, and the right 16 × 32 block is further divided horizontally into two 16 × 16 blocks (binary tree block division). The lower right 32 × 32 block is horizontally divided into two 32 × 16 blocks (binary tree block division). As a result, the lower left 64 × 64 block is divided into a 16 × 32 block 16, two 16 × 16

blocks

17 and 18, two 32 × 32

blocks

19 and 20, and two 32 × 16

blocks

21 and 22.

The lower right 64x64 block 23 is not divided.

As described above, in FIG. 2, the block 10 is divided into 13 variable-size blocks 11 to 23 based on the recursive quadtree and binary tree block division. Such division may be called QTBT (quad-tree plus binary tree) division.

In FIG. 2, one block is divided into four or two blocks (quadrature tree or binary tree block division), but the division is not limited to this. For example, one block may be divided into three blocks (triple tree block division). Such a division including a tri-tree block division may be called an MBT (multi type tree) division.

[Subtraction section]
The subtraction unit 104 subtracts the prediction signal (prediction sample) from the original signal (original sample) in units of blocks divided by the division unit 102. That is, the subtraction unit 104 calculates a prediction error (also referred to as a residual) of a coding target block (hereinafter referred to as a current block). Then, the subtraction unit 104 outputs the calculated prediction error to the conversion unit 106.

The original signal is an input signal of the encoding device 100, and is a signal (for example, a luminance (luma) signal and two color difference (chroma) signals) representing an image of each picture constituting the moving image. Hereinafter, a signal representing an image may be referred to as a sample.

[Conversion section]
The transform unit 106 transforms the prediction error in the spatial domain into a transform factor in the frequency domain, and outputs the transform coefficient to the quantization unit 108. Specifically, the transform unit 106 performs, for example, a predetermined discrete cosine transform (DCT) or discrete sine transform (DST) on a prediction error in the spatial domain.

Note that the conversion unit 106 adaptively selects a conversion type from a plurality of conversion types, and converts a prediction error into a conversion coefficient using a conversion basis function corresponding to the selected conversion type. May be. Such a conversion may be referred to as EMT (explicit multiple core transform) or AMT (adaptive multiple transform).

The plurality of conversion types include, for example, DCT-II, DCT-V, DCT-VIII, DST-I and DST-VII. FIG. 3 is a table showing conversion basis functions corresponding to each conversion type. In FIG. 3, N indicates the number of input pixels. Selection of a conversion type from among these multiple conversion types may depend on, for example, the type of prediction (intra prediction and inter prediction), or may depend on an intra prediction mode.

Information indicating whether or not to apply such EMT or AMT (for example, called an AMT flag) and information indicating the selected conversion type are signaled at the CU level. Note that the signalization of these pieces of information need not be limited to the CU level, but may be other levels (for example, a sequence level, a picture level, a slice level, a tile level, or a CTU level).

Further, the conversion unit 106 may reconvert the conversion coefficient (conversion result). Such reconversion is sometimes referred to as AST (adaptive secondary transform) or NSST (non-separable secondary transform). For example, the conversion unit 106 performs re-conversion for each sub-block (for example, 4 × 4 sub-block) included in the block of the conversion coefficient corresponding to the intra prediction error. Information indicating whether or not NSST is applied and information related to the transformation matrix used for NSST are signaled at the CU level. Note that the signalization of these pieces of information need not be limited to the CU level, but may be other levels (for example, a sequence level, a picture level, a slice level, a tile level, or a CTU level).

Here, the separable conversion is a method of performing the conversion a plurality of times by separating the number of dimensions of the input for each direction, and the non-separable conversion is two or more when the input is multidimensional. In this method, the dimensions are collectively regarded as one dimension, and conversion is performed collectively.

For example, as an example of non-separable conversion, if an input is a 4 × 4 block, it is regarded as one array having 16 elements, and 16 × 16 conversion is performed on the array. The thing which performs the conversion process with a matrix is mentioned.

Similarly, a 4 × 4 input block is regarded as a single array having 16 elements, and then the Givens rotation is performed multiple times on the array (Hypercube Givens Transform) is also a non-separable. It is an example of conversion.

[Quantization unit]
The quantization unit 108 quantizes the transform coefficient output from the transform unit 106. Specifically, the quantization unit 108 scans the transform coefficients of the current block in a predetermined scanning order, and quantizes the transform coefficients based on the quantization parameter (QP) corresponding to the scanned transform coefficients. Then, the quantization unit 108 outputs the quantized transform coefficient (hereinafter referred to as a quantization coefficient) of the current block to the entropy encoding unit 110 and the inverse quantization unit 112.

The predetermined order is an order for quantization / inverse quantization of transform coefficients. For example, the predetermined scanning order is defined in ascending order of frequency (order from low frequency to high frequency) or descending order (order from high frequency to low frequency).

The quantization parameter is a parameter that defines a quantization step (quantization width). For example, if the value of the quantization parameter increases, the quantization step also increases. That is, if the value of the quantization parameter increases, the quantization error increases.

[Entropy encoding unit]
The entropy encoding unit 110 generates an encoded signal (encoded bit stream) by performing variable length encoding on the quantization coefficient that is input from the quantization unit 108. Specifically, the entropy encoding unit 110 binarizes the quantization coefficient, for example, and arithmetically encodes the binary signal.

[Inverse quantization unit]
The inverse quantization unit 112 inversely quantizes the quantization coefficient that is an input from the quantization unit 108. Specifically, the inverse quantization unit 112 inversely quantizes the quantization coefficient of the current block in a predetermined scanning order. Then, the inverse quantization unit 112 outputs the inverse-quantized transform coefficient of the current block to the inverse transform unit 114.

[Inverse conversion part]
The inverse transform unit 114 restores the prediction error by inverse transforming the transform coefficient that is an input from the inverse quantization unit 112. Specifically, the inverse transform unit 114 restores the prediction error of the current block by performing an inverse transform corresponding to the transform by the transform unit 106 on the transform coefficient. Then, the inverse transformation unit 114 outputs the restored prediction error to the addition unit 116.

Note that the restored prediction error does not match the prediction error calculated by the subtraction unit 104 because information is lost due to quantization. That is, the restored prediction error includes a quantization error.

[Addition part]
The adder 116 reconstructs the current block by adding the prediction error input from the inverse transform unit 114 and the prediction sample input from the prediction control unit 128. Then, the adding unit 116 outputs the reconfigured block to the block memory 118 and the loop filter unit 120. The reconstructed block is sometimes referred to as a local decoding block.

[Block memory]
The block memory 118 is a storage unit for storing blocks in an encoding target picture (hereinafter referred to as current picture) that are referred to in intra prediction. Specifically, the block memory 118 stores the reconstructed block output from the adding unit 116.

[Loop filter section]
The loop filter unit 120 applies a loop filter to the block reconstructed by the adding unit 116 and outputs the filtered reconstructed block to the frame memory 122. The loop filter is a filter (in-loop filter) used in the encoding loop, and includes, for example, a deblocking filter (DF), a sample adaptive offset (SAO), an adaptive loop filter (ALF), and the like.

In ALF, a least square error filter is applied to remove coding distortion. For example, for each 2 × 2 sub-block in the current block, a plurality of multiples based on the direction of the local gradient and the activity are provided. One filter selected from the filters is applied.

Specifically, first, sub-blocks (for example, 2 × 2 sub-blocks) are classified into a plurality of classes (for example, 15 or 25 classes). Sub-block classification is performed based on gradient direction and activity. For example, the classification value C (for example, C = 5D + A) is calculated using the gradient direction value D (for example, 0 to 2 or 0 to 4) and the gradient activity value A (for example, 0 to 4). Then, based on the classification value C, the sub-blocks are classified into a plurality of classes (for example, 15 or 25 classes).

The direction value D of the gradient is derived, for example, by comparing gradients in a plurality of directions (for example, horizontal, vertical, and two diagonal directions). The gradient activation value A is derived, for example, by adding gradients in a plurality of directions and quantizing the addition result.

-Based on the result of such classification, a filter for a sub-block is determined from among a plurality of filters.

As the shape of the filter used in ALF, for example, a circularly symmetric shape is used. 4A to 4C are diagrams showing a plurality of examples of filter shapes used in ALF. 4A shows a 5 × 5 diamond shape filter, FIG. 4B shows a 7 × 7 diamond shape filter, and FIG. 4C shows a 9 × 9 diamond shape filter. Information indicating the shape of the filter is signalized at the picture level. It should be noted that the signalization of the information indicating the filter shape need not be limited to the picture level, but may be another level (for example, a sequence level, a slice level, a tile level, a CTU level, or a CU level).

ON / OFF of ALF is determined at the picture level or the CU level, for example. For example, for luminance, it is determined whether to apply ALF at the CU level, and for color difference, it is determined whether to apply ALF at the picture level. Information indicating ALF on / off is signaled at the picture level or the CU level. Signaling of information indicating ALF on / off need not be limited to the picture level or the CU level, and may be performed at other levels (for example, a sequence level, a slice level, a tile level, or a CTU level). Good.

A coefficient set of a plurality of selectable filters (for example, up to 15 or 25 filters) is signalized at the picture level. The signalization of the coefficient set need not be limited to the picture level, but may be another level (for example, sequence level, slice level, tile level, CTU level, CU level, or sub-block level).

[Frame memory]
The frame memory 122 is a storage unit for storing a reference picture used for inter prediction, and is sometimes called a frame buffer. Specifically, the frame memory 122 stores the reconstructed block filtered by the loop filter unit 120.

[Intra prediction section]
The intra prediction unit 124 generates a prediction signal (intra prediction signal) by referring to the block in the current picture stored in the block memory 118 and performing intra prediction (also referred to as intra-screen prediction) of the current block. Specifically, the intra prediction unit 124 generates an intra prediction signal by performing intra prediction with reference to a sample (for example, luminance value and color difference value) of a block adjacent to the current block, and performs prediction control on the intra prediction signal. To the unit 128.

For example, the intra prediction unit 124 performs intra prediction using one of a plurality of predefined intra prediction modes. The plurality of intra prediction modes include one or more non-directional prediction modes and a plurality of directional prediction modes.

One or more non-directional prediction modes are for example H.264. It includes Planar prediction mode and DC prediction mode defined by H.265 / HEVC (High-Efficiency Video Coding) standard (Non-patent Document 1).

The multiple directionality prediction modes are for example H.264. It includes 33-direction prediction modes defined in the H.265 / HEVC standard. In addition to the 33 directions, the plurality of directionality prediction modes may further include 32 direction prediction modes (a total of 65 directionality prediction modes). FIG. 5 is a diagram illustrating 67 intra prediction modes (two non-directional prediction modes and 65 directional prediction modes) in intra prediction. The solid line arrows The 33 directions defined in the H.265 / HEVC standard are represented, and the dashed arrow represents the added 32 directions.

Note that the luminance block may be referred to in the intra prediction of the color difference block. That is, the color difference component of the current block may be predicted based on the luminance component of the current block. Such intra prediction is sometimes called CCLM (cross-component linear model) prediction. The intra prediction mode (for example, called CCLM mode) of the color difference block which refers to such a luminance block may be added as one of the intra prediction modes of the color difference block.

The intra prediction unit 124 may correct the pixel value after intra prediction based on the gradient of the reference pixel in the horizontal / vertical direction. Intra prediction with such correction may be called PDPC (position dependent intra prediction combination). Information indicating whether or not PDPC is applied (for example, referred to as a PDPC flag) is signaled, for example, at the CU level. The signalization of this information need not be limited to the CU level, but may be another level (for example, a sequence level, a picture level, a slice level, a tile level, or a CTU level).

[Inter prediction section]
The inter prediction unit 126 refers to a reference picture stored in the frame memory 122 and is different from the current picture, and performs inter prediction (also referred to as inter-screen prediction) of the current block, thereby generating a prediction signal (inter prediction signal). Prediction signal). Inter prediction is performed in units of a current block or a sub-block (for example, 4 × 4 block) in the current block. For example, the inter prediction unit 126 performs motion estimation in the reference picture for the current block or sub-block. Then, the inter prediction unit 126 generates an inter prediction signal of the current block or sub-block by performing motion compensation using motion information (for example, a motion vector) obtained by motion search. Then, the inter prediction unit 126 outputs the generated inter prediction signal to the prediction control unit 128.

The motion information used for motion compensation is signaled. A motion vector predictor may be used for signalizing the motion vector. That is, the difference between the motion vector and the predicted motion vector may be signaled.

Note that an inter prediction signal may be generated using not only the motion information of the current block obtained by motion search but also the motion information of adjacent blocks. Specifically, the inter prediction signal is generated in units of sub-blocks in the current block by weighted addition of the prediction signal based on the motion information obtained by motion search and the prediction signal based on the motion information of adjacent blocks. May be. Such inter prediction (motion compensation) is sometimes called OBMC (overlapped block motion compensation).

In such an OBMC mode, information indicating the size of a sub-block for OBMC (for example, called OBMC block size) is signaled at the sequence level. Also, information indicating whether or not to apply the OBMC mode (for example, referred to as an OBMC flag) is signaled at the CU level. Note that the level of signalization of these information does not need to be limited to the sequence level and the CU level, and may be other levels (for example, a picture level, a slice level, a tile level, a CTU level, or a sub-block level). Good.

Note that the motion information may be derived on the decoding device side without being converted into a signal. For example, H.M. A merge mode defined in the H.265 / HEVC standard may be used. Further, for example, the motion information may be derived by performing motion search on the decoding device side. In this case, motion search is performed without using the pixel value of the current block.

Here, a mode in which motion search is performed on the decoding device side will be described. The mode in which motion search is performed on the decoding device side is sometimes called a PMMVD (patterned motion vector derivation) mode or an FRUC (frame rate up-conversion) mode.

First, by referring to the motion vector of an encoded block spatially or temporally adjacent to the current block, a list of a plurality of candidates each having a predicted motion vector (may be common with the merge list) is generated Is done. Then, the evaluation value of each candidate included in the candidate list is calculated, and one candidate is selected based on the evaluation value.

Then, a motion vector for the current block is derived based on the selected candidate motion vector. Specifically, for example, the selected candidate motion vector is directly derived as a motion vector for the current block. Further, for example, the motion vector for the current block may be derived by performing pattern matching in the peripheral region at the position in the reference picture corresponding to the selected candidate motion vector.

Note that the evaluation value is calculated by pattern matching between an area in the reference picture corresponding to the motion vector and a predetermined area.

As the pattern matching, the first pattern matching or the second pattern matching is used. The first pattern matching and the second pattern matching may be referred to as bilateral matching and template matching, respectively.

In the first pattern matching, pattern matching is performed between two blocks in two different reference pictures that follow the motion trajectory of the current block. Therefore, in the first pattern matching, a region in another reference picture along the motion trajectory of the current block is used as the predetermined region for calculating the candidate evaluation value described above.

FIG. 6 is a diagram for explaining pattern matching (bilateral matching) between two blocks along a motion trajectory. As shown in FIG. 6, in the first pattern matching, two blocks along the motion trajectory of the current block (Cur block) and two blocks in two different reference pictures (Ref0, Ref1) are used. By searching for the best matching pair, two motion vectors (MV0, MV1) are derived.

Under the assumption of continuous motion trajectory, the motion vectors (MV0, MV1) pointing to the two reference blocks are temporal distances between the current picture (Cur Pic) and the two reference pictures (Ref0, Ref1). It is proportional to (TD0, TD1). For example, when the current picture is temporally located between two reference pictures and the temporal distances from the current picture to the two reference pictures are equal, the first pattern matching uses a mirror-symmetric bi-directional motion vector Is derived.

In the second pattern matching, pattern matching is performed between a template in the current picture (a block adjacent to the current block in the current picture (for example, an upper and / or left adjacent block)) and a block in the reference picture. Therefore, in the second pattern matching, a block adjacent to the current block in the current picture is used as the predetermined region for calculating the candidate evaluation value described above.

FIG. 7 is a diagram for explaining pattern matching (template matching) between a template in the current picture and a block in the reference picture. As shown in FIG. 7, in the second pattern matching, the current block is searched by searching the reference picture (Ref0) for the block that most closely matches the block adjacent to the current block (Cur block) in the current picture (Cur Pic). Of motion vectors are derived.

Information indicating whether or not to apply such FRUC mode (for example, called FRUC flag) is signaled at the CU level. In addition, when the FRUC mode is applied (for example, when the FRUC flag is true), information indicating the pattern matching method (first pattern matching or second pattern matching) (for example, called the FRUC mode flag) is signaled at the CU level. It becomes. Note that the signalization of these pieces of information need not be limited to the CU level, but may be other levels (for example, sequence level, picture level, slice level, tile level, CTU level, or sub-block level). .

Note that motion information may be derived on the decoding device side by a method different from motion search. For example, the motion vector correction amount may be calculated using a peripheral pixel value for each pixel based on a model assuming constant velocity linear motion.

Here, a mode for deriving a motion vector based on a model assuming constant velocity linear motion will be described. This mode may be referred to as a BIO (bi-directional optical flow) mode.

FIG. 8 is a diagram for explaining a model assuming constant velocity linear motion. In FIG. 8, (v _x , v _y ) indicates a velocity vector, and τ ₀ and τ ₁ are the time between the current picture (Cur Pic) and two reference pictures (Ref ₀ , Ref ₁ ), respectively. The distance. (MVx ₀ , MVy ₀ ) indicates a motion vector corresponding to the reference picture Ref ₀ , and (MVx ₁ , MVy ₁ ) indicates a motion vector corresponding to the reference picture Ref ₁ .

At this time, under the assumption of constant velocity linear motion of the velocity vector (v _x , v _y ), (MVx ₀ , MVy ₀ ) and (MVx ₁ , MVy ₁ ) are (v _x τ ₀ , v _y τ), respectively. ₀ ) and (−v _x τ ₁ , −v _y τ ₁ ), and the following optical flow equation (1) holds.

Here, I ^(k) represents the luminance value of the reference image k (k = 0, 1) after motion compensation. This optical flow equation consists of (i) the product of the time derivative of the luminance value, (ii) the horizontal component of the horizontal velocity and the spatial gradient of the reference image, and (iii) the vertical velocity and the spatial gradient of the reference image. Indicates that the sum of the products of the vertical components of is equal to zero. Based on a combination of this optical flow equation and Hermite interpolation, a block-based motion vector obtained from a merge list or the like is corrected in pixel units.

Note that the motion vector may be derived on the decoding device side by a method different from the derivation of the motion vector based on the model assuming constant velocity linear motion. For example, a motion vector may be derived for each subblock based on the motion vectors of a plurality of adjacent blocks.

Here, a mode for deriving a motion vector for each sub-block based on the motion vectors of a plurality of adjacent blocks will be described. This mode may be referred to as an affine motion compensation prediction mode.

FIG. 9 is a diagram for explaining the derivation of motion vectors in units of sub-blocks based on the motion vectors of a plurality of adjacent blocks. In FIG. 9, the current block includes 16 4 × 4 sub-blocks. Here, the motion vector v ₀ of the upper left corner control point of the current block is derived based on the motion vector of the adjacent block, and the motion vector v ₁ of the upper right corner control point of the current block is derived based on the motion vector of the adjacent sub block. Is done. Then, using the two motion vectors v ₀ and v ₁ , the motion vector (v _x , v _y ) of each sub-block in the current block is derived by the following equation (2).

Here, x and y indicate the horizontal position and vertical position of the sub-block, respectively, and w indicates a predetermined weight coefficient.

Such an affine motion compensation prediction mode may include several modes in which the motion vector derivation methods of the upper left and upper right corner control points are different. Information indicating such an affine motion compensation prediction mode (for example, called an affine flag) is signaled at the CU level. Note that the information indicating the affine motion compensation prediction mode need not be limited to the CU level, but other levels (for example, sequence level, picture level, slice level, tile level, CTU level, or sub-block level). ).

[Prediction control unit]
The prediction control unit 128 selects either the intra prediction signal or the inter prediction signal, and outputs the selected signal to the subtraction unit 104 and the addition unit 116 as a prediction signal.

[Outline of Decoding Device]
Next, an outline of a decoding apparatus capable of decoding the encoded signal (encoded bit stream) output from the encoding apparatus 100 will be described. FIG. 10 is a block diagram showing a functional configuration of decoding apparatus 200 according to Embodiment 1. The decoding device 200 is a moving image / image decoding device that decodes moving images / images in units of blocks.

As illustrated in FIG. 10, the decoding device 200 includes an entropy decoding unit 202, an inverse quantization unit 204, an inverse transformation unit 206, an addition unit 208, a block memory 210, a loop filter unit 212, and a frame memory 214. And an intra prediction unit 216, an inter prediction unit 218, and a prediction control unit 220.

The decoding device 200 is realized by, for example, a general-purpose processor and a memory. In this case, when the software program stored in the memory is executed by the processor, the processor executes the entropy decoding unit 202, the inverse quantization unit 204, the inverse transformation unit 206, the addition unit 208, the loop filter unit 212, and the intra prediction unit. 216, the inter prediction unit 218, and the prediction control unit 220. Also, the decoding apparatus 200 is dedicated to the entropy decoding unit 202, the inverse quantization unit 204, the inverse transformation unit 206, the addition unit 208, the loop filter unit 212, the intra prediction unit 216, the inter prediction unit 218, and the prediction control unit 220. It may be realized as one or more electronic circuits.

Hereinafter, each component included in the decoding device 200 will be described.

[Entropy decoding unit]
The entropy decoding unit 202 performs entropy decoding on the encoded bit stream. Specifically, the entropy decoding unit 202 performs arithmetic decoding from a coded bit stream to a binary signal, for example. Then, the entropy decoding unit 202 debinarizes the binary signal. As a result, the entropy decoding unit 202 outputs the quantized coefficient to the inverse quantization unit 204 in units of blocks.

[Inverse quantization unit]
The inverse quantization unit 204 inversely quantizes the quantization coefficient of a decoding target block (hereinafter referred to as a current block) that is an input from the entropy decoding unit 202. Specifically, the inverse quantization unit 204 inversely quantizes each quantization coefficient of the current block based on the quantization parameter corresponding to the quantization coefficient. Then, the inverse quantization unit 204 outputs the quantization coefficient (that is, the transform coefficient) obtained by inverse quantization of the current block to the inverse transform unit 206.

[Inverse conversion part]
The inverse transform unit 206 restores the prediction error by inverse transforming the transform coefficient that is an input from the inverse quantization unit 204.

For example, when the information read from the encoded bit stream indicates that EMT or AMT is applied (for example, the AMT flag is true), the inverse conversion unit 206 determines the current block based on the information indicating the read conversion type. Inversely transform the conversion coefficient of.

Also, for example, when the information read from the encoded bitstream indicates that NSST is applied, the inverse transform unit 206 applies inverse retransformation to the transform coefficient.

[Addition part]
The adder 208 reconstructs the current block by adding the prediction error input from the inverse converter 206 and the prediction sample input from the prediction controller 220. Then, the adding unit 208 outputs the reconfigured block to the block memory 210 and the loop filter unit 212.

[Block memory]
The block memory 210 is a storage unit for storing a block that is referred to in intra prediction and that is within a decoding target picture (hereinafter referred to as a current picture). Specifically, the block memory 210 stores the reconstructed block output from the adding unit 208.

[Loop filter section]
The loop filter unit 212 applies a loop filter to the block reconstructed by the adding unit 208, and outputs the filtered reconstructed block to the frame memory 214, the display device, and the like.

If the ALF on / off information read from the encoded bitstream indicates ALF on, one filter is selected from the plurality of filters based on the local gradient direction and activity, The selected filter is applied to the reconstruction block.

[Frame memory]
The frame memory 214 is a storage unit for storing a reference picture used for inter prediction, and is sometimes called a frame buffer. Specifically, the frame memory 214 stores the reconstructed block filtered by the loop filter unit 212.

[Intra prediction section]
The intra prediction unit 216 performs intra prediction with reference to the block in the current picture stored in the block memory 210 based on the intra prediction mode read from the encoded bitstream, so that a prediction signal (intra prediction Signal). Specifically, the intra prediction unit 216 generates an intra prediction signal by performing intra prediction with reference to a sample (for example, luminance value and color difference value) of a block adjacent to the current block, and performs prediction control on the intra prediction signal. Output to the unit 220.

In addition, when the intra prediction mode that refers to the luminance block is selected in the intra prediction of the color difference block, the intra prediction unit 216 may predict the color difference component of the current block based on the luminance component of the current block. .

In addition, when the information read from the encoded bitstream indicates application of PDPC, the intra prediction unit 216 corrects the pixel value after intra prediction based on the gradient of the reference pixel in the horizontal / vertical direction.

[Inter prediction section]
The inter prediction unit 218 refers to the reference picture stored in the frame memory 214 and predicts the current block. Prediction is performed in units of a current block or a sub-block (for example, 4 × 4 block) in the current block. For example, the inter prediction unit 218 generates an inter prediction signal of the current block or sub-block by performing motion compensation using motion information (for example, a motion vector) read from the encoded bitstream, and generates the inter prediction signal. The result is output to the prediction control unit 220.

When the information read from the encoded bitstream indicates that the OBMC mode is to be applied, the inter prediction unit 218 includes not only the motion information of the current block obtained by motion search but also the motion information of adjacent blocks. To generate an inter prediction signal.

Also, when the information read from the encoded bitstream indicates that the FRUC mode is applied, the inter prediction unit 218 follows the pattern matching method (bilateral matching or template matching) read from the encoded stream. Motion information is derived by performing motion search. Then, the inter prediction unit 218 performs motion compensation using the derived motion information.

In addition, when the BIO mode is applied, the inter prediction unit 218 derives a motion vector based on a model assuming constant velocity linear motion. Also, when the information read from the encoded bitstream indicates that the affine motion compensated prediction mode is applied, the inter prediction unit 218 determines the motion vector in units of subblocks based on the motion vectors of a plurality of adjacent blocks. Is derived.

[Prediction control unit]
The prediction control unit 220 selects either the intra prediction signal or the inter prediction signal, and outputs the selected signal to the adding unit 208 as a prediction signal.

(Embodiment 2)
The encoding device and decoding device in the present embodiment have the same configuration and function as in Embodiment 1, but are characterized by processing operations of

inter prediction units

126 and 218 and the like.

[Knowledge that became the basis of this disclosure]
FIG. 11 is a flowchart illustrating motion compensation by another encoding device that is the basis of the present disclosure. In addition, in each figure after FIG. 11, a motion vector is shown as MV.

The encoding device performs motion compensation for each prediction block corresponding to the prediction unit described above. At this time, the encoding device first determines a plurality of candidate motion vectors for the prediction block based on information such as motion vectors of a plurality of encoded blocks around the prediction block temporally or spatially. Obtain (step S101).

Next, the encoding apparatus determines each of N (N is an integer of 2 or more) candidate motion vectors from among the plurality of candidate motion vectors acquired in step S101 as predicted motion vector candidates. Extraction is performed according to the priority order (step S102). The priority order is predetermined for each of the N candidate motion vectors.

Next, the encoding apparatus selects one predicted motion vector candidate from the N predicted motion vector candidates as the predicted motion vector of the predicted block. At this time, the encoding apparatus encodes prediction motion vector selection information for identifying the selected prediction motion vector into a stream (step S103). The stream is the above-described encoded signal or encoded bit stream.

Next, the encoding device refers to the encoded reference picture and derives a motion vector of the prediction block (step S104). At this time, the encoding apparatus further encodes a difference value between the derived motion vector and the predicted motion vector into a stream as difference motion vector information. An encoded reference picture is a picture composed of a plurality of blocks reconstructed after encoding.

Finally, the encoding apparatus performs motion compensation on the prediction block using the derived motion vector and the encoded reference picture, thereby generating a prediction image of the prediction block (step S105). The predicted image is the above-described inter prediction signal.

FIG. 12 is a flowchart showing motion compensation by another decoding device as a basis of the present disclosure.

The decoding device performs motion compensation for each prediction block for each prediction block. At this time, the decoding apparatus first acquires a plurality of candidate motion vectors for the prediction block based on information such as motion vectors of a plurality of decoded blocks around the prediction block temporally or spatially. (Step S111).

Next, the decoding apparatus sets a predetermined priority as a predicted motion vector candidate for each of N (N is an integer of 2 or more) candidate motion vectors from among the plurality of candidate motion vectors acquired in step S111. Extraction is performed according to the rank (step S112). The priority order is predetermined for each of the N candidate motion vectors.

Next, the decoding apparatus decodes the predicted motion vector selection information from the input stream, and uses the decoded predicted motion vector selection information to select one predicted motion vector from the N predicted motion vector candidates. A candidate is selected as a prediction motion vector of a prediction block (step S113).

Next, the decoding apparatus decodes the difference motion vector information from the input stream, and adds the difference value, which is the decoded difference motion vector information, to the selected prediction motion vector, thereby obtaining the prediction block. A motion vector is derived (step S114).

Finally, the decoding apparatus generates a predicted image of the prediction block by performing motion compensation on the prediction block using the derived motion vector and the decoded reference picture (step S115).

Here, in the example shown in FIGS. 11 and 12, a predetermined priority order is used to extract N predicted motion vector candidates. However, each of the plurality of candidate motion vectors may be evaluated in order to obtain a prediction motion vector having a smaller difference from the motion vector of the prediction block. That is, the respective evaluation values of the plurality of candidate motion vectors acquired in step S101 or step 111 are calculated, and N predicted motion vector candidates are selected from the plurality of candidate motion vectors based on the calculated evaluation values. It may be extracted.

FIG. 13 is a diagram for explaining an example of an evaluation value calculation method.

As a method for calculating the evaluation value, for example, there is a template matching method. In this template matching method, a reconstructed image of a coded region or a decoded region in a moving image is used for calculating an evaluation value. In FIG. 13, the encoded region and the decoded region are collectively referred to as the processed region, and the encoding target picture and the decoding target picture are collectively referred to as the processing target picture. In addition, a prediction block to be encoded and a prediction block to be decoded are collectively referred to as a processing target prediction block.

Specifically, the encoding device includes a reconstructed image of an encoded region around the processing target prediction block in the encoding target picture and a block specified by the candidate motion vector in the encoded reference picture. A difference value from a reconstructed image of a certain encoded area is calculated. For example, the difference value is calculated as a sum of absolute differences of pixel values.

Similar to the encoding device, the decoding device also includes a reconstructed image of a decoded region in the vicinity of the processing target prediction block in the decoding target picture and a decoded image in the vicinity of the block specified by the candidate motion vector in the decoded reference picture. A difference value from the reconstructed image of the area is calculated. For example, the difference value is calculated as a sum of absolute differences of pixel values.

Note that a block designated by a candidate motion vector in a reference picture is hereinafter referred to as a designated block. This designated block is at a position indicated by the candidate motion vector with reference to the spatial position of the processing target prediction block. Further, the relative position of the processed area with respect to the designated block in the reference picture is equal to the relative position of the processed area with respect to the processing target prediction block in the processing target picture. In addition, the processed region around the processing target prediction block or the designated block may be a region adjacent to the left of the block and a region adjacent to the top of the block, or may be only a region adjacent to the left. , Only the region adjacent to the upper side may be used. For example, if there are an adjacent area on the left and an adjacent area on the upper side, those areas are used for calculation of the evaluation value, and if any area does not exist, only the existing area is used for calculation of the evaluation value. used.

The encoding device and the decoding device calculate an evaluation value using the obtained difference value. For example, a higher evaluation value is calculated as the difference value is smaller. Note that the encoding device and the decoding device may calculate the evaluation value using information other than the difference value.

Note that the evaluation value calculation method by the template matching method shown in the example of FIG. 13 is an example, and is not limited to this. For example, the position of the region used for the evaluation or the method for determining whether or not the region can be used is not limited to the example in FIG.

FIG. 14 is a diagram for explaining another example of an evaluation value calculation method.

As a method for calculating the evaluation value, for example, there is a bilateral matching method. Even in this bilateral matching method, a reconstructed image of a coded region or a decoded region in a moving image is used to calculate an evaluation value. In FIG. 14, the encoded region and the decoded region are collectively referred to as a processed region, and the encoding target picture and the decoding target picture are collectively referred to as a processing target picture. In addition, a prediction block to be encoded and a prediction block to be decoded are collectively referred to as a processing target prediction block.

Specifically, the encoding device reconstructs an image of a block specified by a candidate motion vector in the encoded reference picture 1 and a reconstructed image of a block specified by a symmetric motion vector in the encoded reference picture 2 The difference value is calculated. Both the block specified by the candidate motion vector and the block specified by the symmetric motion vector are encoded regions. For example, the difference value is calculated as a sum of absolute differences of pixel values.

Similarly to the encoding device, the decoding device also includes a reconstructed image of a block specified by a candidate motion vector in the decoded reference picture 1 and a reconstructed image of a block specified by a symmetric motion vector in the decoded reference picture 2. The difference value is calculated. Both the block specified by the candidate motion vector and the block specified by the symmetric motion vector are decoded regions. For example, the difference value is calculated as a sum of absolute differences of pixel values.

Note that the symmetric motion vector is a motion vector generated by scaling the candidate motion vector according to the display time interval described above. In addition, the block specified by each of the candidate motion vector and the symmetric motion vector is at a position pointed to based on the spatial position of the processing target prediction block.

Note that the evaluation value calculation method by the bilateral matching method shown in FIG. 14 is an example, and the present invention is not limited to this. For example, the method of specifying the position of the processed region used for evaluation is not limited to the example shown in FIG.

It may be possible to improve the prediction accuracy of the prediction block by extracting N predicted motion vector candidates from a plurality of candidate motion vectors based on such evaluation values. Note that the template matching method and the bilateral matching method are methods used in the above-described FRUC mode. Therefore, a method for extracting a predicted motion vector candidate based on such an evaluation value is also referred to as an extraction method based on an evaluation result by FRUC.

Here, in order to extract N predicted motion vector candidates from the plurality of candidate motion vectors for the prediction block, a first extraction method based on the evaluation result by FRUC and a first priority based on a predetermined priority order are used. Two extraction methods can be used.

For example, the encoding device and the decoding device extract one prediction motion vector candidate using the first extraction method in order to extract two prediction motion vector candidates from a plurality of candidate motion vectors, The remaining one motion vector predictor candidate is extracted using the above extraction method. In such a case, it is assumed that a separate candidate list is used for each of the first extraction method and the second extraction method. These candidate lists are lists indicating a plurality of candidate motion vectors.

Therefore, in the case described above, although there is a possibility that the prediction accuracy can be improved, a plurality of different candidate lists must be created for one prediction block, which increases the processing load. The problem of end up occurs.

Therefore, encoding apparatus 100 and decoding apparatus 200 in the present embodiment use an extraction method based on the evaluation result by FRUC, and perform motion compensation for the prediction block using one candidate list for the prediction block. Do.

Specifically, the encoding apparatus 100 according to the present embodiment extracts at least one prediction motion vector candidate of the encoding target block from a plurality of candidate motion vectors. At this time, the encoding apparatus 100 uses a plurality of candidates using the reconstructed image of the encoded region in the moving image without using the image region of the encoding target block as all of the at least one prediction motion vector candidate. Extraction is performed based on the evaluation results of each motion vector.

Also, the decoding apparatus 200 in the present embodiment extracts at least one predicted motion vector candidate of the decoding target block from a plurality of candidate motion vectors. At this time, the decoding apparatus 200 converts all of the at least one predicted motion vector candidate into a plurality of candidate motion vectors using the reconstructed image of the decoded region in the moving image without using the image region of the decoding target block. Extract based on each evaluation result.

That is, the encoding apparatus 100 and the decoding apparatus 200 in the present embodiment extract all predicted motion vector candidates by an extraction method based on the evaluation result by FRUC. In other words, all predicted motion vector candidates are extracted without using an extraction method based on a predetermined priority order. All the predicted motion vector candidates may be one predicted motion vector candidate or a plurality of predicted motion vector candidates. Therefore, since an extraction method based on a predetermined priority order is not used, a dedicated candidate list for the extraction method is not necessary, and motion compensation using one candidate list is performed for the prediction block. .

[Extract one motion vector candidate using only FRUC]
FIG. 15 is a flowchart illustrating an example of motion compensation by the encoding apparatus 100 according to the present embodiment. When the encoding device 100 illustrated in FIG. 1 encodes a moving image including a plurality of pictures, the inter prediction unit 126 and the like of the encoding device 100 execute the processing illustrated in FIG.

Specifically, the inter prediction unit 126 performs motion compensation for each prediction block corresponding to the above-described prediction unit, with respect to the encoding target block that is the prediction block. At this time, the inter prediction unit 126 first determines a plurality of candidate motion vectors for the prediction block based on information such as motion vectors of a plurality of encoded blocks around the prediction block temporally or spatially. Is acquired (step S201). For example, information such as a motion vector of an encoded block may be a motion vector used for motion compensation of the encoded block, and not only the motion vector but also a picture and code including the encoded block. The display time interval between the pictures to be converted may be included. For example, the plurality of candidate motion vectors are obtained by scaling each of the motion vectors of the plurality of encoded blocks according to the display time interval. Also, the plurality of encoded blocks around the prediction block are different from the encoding target picture, for example, from a plurality of encoded blocks adjacent to the lower left, upper left, and upper right of the prediction block to be encoded. It may be all or a part of a plurality of encoded blocks included in a picture.

Next, the inter prediction unit 126 calculates the evaluation values of the plurality of candidate motion vectors acquired in step S201 using the reconstructed image of the encoded region. That is, the inter prediction unit 126 calculates those evaluation values based on FRUC, that is, the template matching method or the bilateral matching method. Then, the inter prediction unit 126 selects one candidate motion vector having the highest evaluation value from among the plurality of candidate motion vectors as a prediction motion vector of the prediction block (step S202). That is, the inter prediction unit 126 extracts all of the at least one predicted motion vector candidate described above by selecting only one candidate motion vector having the best evaluation result from the plurality of candidate motion vectors.

Note that the inter prediction unit 126 may correct the predicted motion vector by finely moving the selected predicted motion vector in the peripheral region so that the FRUC evaluation value becomes higher. That is, the inter prediction unit 126 may correct the predicted motion vector by finely searching for a region where the FRUC evaluation value is higher.

Next, the inter prediction unit 126 refers to the encoded reference picture and derives a motion vector of the prediction block (step S203). At this time, the inter prediction unit 126 further calculates a difference value between the derived motion vector and the predicted motion vector. The entropy encoding unit 110 encodes the difference value as a difference motion vector information into a stream. That is, the entropy encoding unit 110 encodes the difference between the predicted motion vector that is the selected candidate motion vector and the motion vector of the derived encoding target block.

Finally, the inter prediction unit 126 generates a prediction image of the prediction block by performing motion compensation on the prediction block using the derived motion vector and the encoded reference picture (step S204). .

Note that the inter prediction unit 126 similarly derives a motion vector in units of subblocks obtained by dividing a prediction block instead of motion compensation in units of prediction blocks as described above, and performs motion compensation in units of subblocks. May be performed.

FIG. 16 is a flowchart showing an example of motion compensation by the decoding apparatus 200 according to the present embodiment. When the decoding device 200 illustrated in FIG. 10 decodes a moving image including a plurality of encoded pictures, the inter prediction unit 218 and the like of the decoding device 200 execute the processing illustrated in FIG.

Specifically, the inter prediction unit 218 performs motion compensation on the decoding target block that is a prediction block for each prediction block corresponding to the above-described prediction unit. At this time, the inter prediction unit 218 first selects a plurality of candidate motion vectors for the prediction block based on information such as motion vectors of a plurality of decoded blocks around the prediction block temporally or spatially. Obtain (step S211). For example, the information such as the motion vector of the decoded block may be a motion vector used for motion compensation of the decoded block, and not only the motion vector but also a picture including the decoded block and a decoding target picture. Display time intervals between the two. For example, the plurality of candidate motion vectors are obtained by scaling each of the motion vectors of the plurality of decoded blocks according to the display time interval. In addition, a plurality of decoded blocks around the prediction block are included in, for example, a plurality of decoded blocks adjacent to the lower left, upper left, and upper right of the prediction block to be decoded, and a picture different from the decoding target picture. It may be all or some of the decoded blocks of the plurality of decoded blocks.

Next, the inter prediction unit 218 calculates each evaluation value of the plurality of candidate motion vectors acquired in step S211 using the reconstructed image of the decoded area. That is, the inter prediction unit 218 calculates the evaluation values based on FRUC, that is, the template matching method or the bilateral matching method. Then, the inter prediction unit 218 selects one candidate motion vector having the highest evaluation value from among the plurality of candidate motion vectors as a prediction motion vector of the prediction block (step S212). That is, the inter prediction unit 218 extracts all of the above-described at least one predicted motion vector candidate by selecting only one candidate motion vector having the best evaluation result from the plurality of candidate motion vectors.

Note that the inter prediction unit 218 may correct the predicted motion vector by finely moving the selected predicted motion vector in the peripheral region so that the FRUC evaluation value becomes higher. That is, the inter prediction unit 218 may correct the predicted motion vector by finely searching for a region where the FRUC evaluation value is higher.

Next, the inter prediction unit 218 derives a motion vector of the prediction block using the difference motion vector information decoded by the entropy decoding unit 202 from the stream input to the decoding device 200 (step S213). Specifically, the inter prediction unit 218 derives the motion vector of the prediction block by adding the difference value, which is the decoded difference motion vector information, and the selected prediction motion vector. That is, the entropy decoding unit 202 decodes difference motion vector information that is difference information indicating a difference between two motion vectors. Then, the inter prediction unit 218 derives the motion vector of the prediction block that is the decoding target block by adding the prediction motion vector that is the selected candidate motion vector to the difference indicated by the decoded difference information. .

Finally, the inter prediction unit 218 generates a prediction image of the prediction block by performing motion compensation on the prediction block using the derived motion vector and the decoded reference picture (step S214).

Note that the inter prediction unit 218 similarly derives a motion vector in units of subblocks obtained by dividing a prediction block instead of motion compensation in units of prediction blocks as described above, and performs motion compensation in units of subblocks. May be performed.

In the example shown in FIG. 15 and FIG. 16, one predicted motion vector candidate is extracted, but a plurality of predicted motion vector candidates may be extracted.

[Extract multiple motion vector candidates using only FRUC]
FIG. 17 is a flowchart showing another example of motion compensation by the encoding apparatus 100 according to the present embodiment. When the encoding device 100 illustrated in FIG. 1 encodes a moving image including a plurality of pictures, the inter prediction unit 126 and the like of the encoding device 100 perform the processing illustrated in FIG.

Specifically, the inter prediction unit 126 performs motion compensation for each prediction block corresponding to the above-described prediction unit, with respect to the encoding target block that is the prediction block. At this time, the inter prediction unit 126 first determines a plurality of candidate motion vectors for the prediction block based on information such as motion vectors of a plurality of encoded blocks around the prediction block temporally or spatially. Is acquired (step S201).

Next, the inter prediction unit 126 calculates the evaluation values of the plurality of candidate motion vectors acquired in step S201 using the reconstructed image of the encoded region. That is, the inter prediction unit 126 calculates those evaluation values based on FRUC, that is, the template matching method or the bilateral matching method. Then, based on the evaluation values of the plurality of candidate motion vectors, the inter prediction unit 126 calculates each of N (N is an integer of 2 or more) candidate motion vectors from among the plurality of candidate motion vectors. (Step S202a). That is, the inter prediction unit 126 extracts N candidate motion vectors as all of the at least one predicted motion vector candidate from the plurality of candidate motion vectors based on the evaluation result. More specifically, the inter prediction unit 126 extracts the top N candidate motion vectors from the plurality of candidate motion vectors in the order of good evaluation results as all of the above-described at least one prediction motion vector candidate. In other words, the inter prediction unit 126 extracts the top N candidate motion vectors from the plurality of candidate motion vectors in the descending order of evaluation value as predicted motion vector candidates.

Note that the inter prediction unit 126 finely moves the selected predicted motion vector candidates in the peripheral region so that the FRUC evaluation value becomes higher for each of the extracted N predicted motion vector candidates. The predicted motion vector candidate may be corrected. That is, the inter prediction unit 126 may correct these motion vector predictor candidates by finely searching for a region where the FRUC evaluation value is higher.

Then, the inter prediction unit 126 selects a prediction motion vector of the prediction block from the extracted N prediction motion vector candidates (step S202b). At this time, the inter prediction unit 126 outputs prediction motion vector selection information for identifying the selected prediction motion vector. The entropy encoding unit 110 encodes the predicted motion vector selection information into a stream.

In the selection of the prediction motion vector, the inter prediction unit 126 may use an original image of a prediction block that is a coding target block. For example, the inter prediction unit 126 calculates, for each of N predicted motion vector candidates, the difference between the image of the block specified by the predicted motion vector candidate and the original image of the predicted block. Then, the inter prediction unit 126 selects a motion vector predictor candidate having the smallest difference as a motion vector predictor of the prediction block. Alternatively, the inter prediction unit 126 may derive a motion vector of the prediction block by performing a motion search using the original image of the prediction block. Then, for each of the N predicted motion vector candidates, the inter prediction unit 126 performs an operation between a block image specified by the predicted motion vector candidate and a block image specified by the derived predicted block motion vector. Calculate the difference. Then, the inter prediction unit 126 selects a motion vector predictor candidate having the smallest difference as a motion vector predictor of the prediction block.

Next, the inter prediction unit 126 refers to the encoded reference picture and derives a motion vector of the prediction block (step S203). At this time, the inter prediction unit 126 further calculates a difference value between the derived motion vector and the predicted motion vector. The entropy encoding unit 110 encodes the difference value as a difference motion vector information into a stream.

FIG. 18 is a flowchart showing another example of motion compensation by the decoding apparatus 200 according to this embodiment. When the decoding device 200 illustrated in FIG. 10 decodes a moving image including a plurality of encoded pictures, the inter prediction unit 218 and the like of the decoding device 200 perform the processing illustrated in FIG.

Specifically, the inter prediction unit 218 performs motion compensation on the decoding target block that is a prediction block for each prediction block corresponding to the above-described prediction unit. At this time, the inter prediction unit 218 first selects a plurality of candidate motion vectors for the prediction block based on information such as motion vectors of a plurality of decoded blocks around the prediction block temporally or spatially. Obtain (step S211).

Next, the inter prediction unit 218 calculates each evaluation value of the plurality of candidate motion vectors acquired in step S211 using the reconstructed image of the decoded area. That is, the inter prediction unit 218 calculates the evaluation values based on FRUC, that is, the template matching method or the bilateral matching method. Then, based on the evaluation values of the plurality of candidate motion vectors, the inter prediction unit 218 calculates each of N (N is an integer of 2 or more) candidate motion vectors from among the plurality of candidate motion vectors. (Step S212a). That is, the inter prediction unit 218 extracts N candidate motion vectors as all of the at least one predicted motion vector candidate from the plurality of candidate motion vectors based on the above evaluation result. More specifically, the inter prediction unit 218 extracts the top N candidate motion vectors from the plurality of candidate motion vectors in descending order of evaluation results as all of the above-described at least one prediction motion vector candidate. In other words, the inter prediction unit 218 extracts the top N candidate motion vectors as predicted motion vector candidates from the plurality of candidate motion vectors in descending order of evaluation value.

Note that the inter prediction unit 218 finely moves the selected predicted motion vector candidate in the peripheral region so that the FRUC evaluation value becomes higher for each of the extracted N predicted motion vector candidates. The predicted motion vector candidate may be corrected. That is, the inter prediction unit 218 may correct these motion vector predictor candidates by finely searching for a region where the FRUC evaluation value is higher.

Next, the inter prediction unit 218 uses the prediction motion vector selection information decoded by the entropy decoding unit 202 from the stream input to the decoding apparatus 200, and extracts one prediction motion from the N predicted motion vector candidates extracted. A vector candidate is selected as a prediction motion vector of a prediction block (step S212b). That is, the entropy decoding unit 202 decodes prediction motion vector selection information that is selection information for identifying a prediction motion vector. Then, the inter prediction unit 218 selects a predicted motion vector candidate identified by the decoded predicted motion vector selection information from the extracted N predicted motion vector candidates as the predicted motion vector.

Next, the inter prediction unit 218 derives a motion vector of the prediction block using the difference motion vector information decoded by the entropy decoding unit 202 from the stream input to the decoding device 200 (step S213). Specifically, the inter prediction unit 218 derives the motion vector of the prediction block by adding the difference value, which is the decoded difference motion vector information, and the selected prediction motion vector. That is, the entropy decoding unit 202 decodes difference motion vector information that is difference information indicating a difference between two motion vectors. Then, the inter prediction unit 218 derives the motion vector of the prediction block that is the decoding target block by adding the selected prediction motion vector to the difference indicated by the decoded difference information.

FIG. 19 is a diagram for explaining a method of extracting N predicted motion vector candidates from a plurality of candidate motion vectors.

17 and FIG. 18, the

inter prediction units

126 and 218 respectively extract the top N candidate motion vectors as predicted motion vector candidates from the plurality of candidate motion vectors in descending order of evaluation value. Specifically, when N = 2, as illustrated in FIG. 19A, the

inter prediction units

126 and 218 perform the top two candidate motions in descending order of evaluation value from all candidate motion vectors. The vectors are extracted as motion

vector predictor candidates

1 and 2, respectively.

However, the

inter prediction units

126 and 218 classify the plurality of candidate motion vectors into N groups, and extract one candidate motion vector having the best evaluation result in each group from each of the N groups. All of the above-described at least one predicted motion vector candidate may be extracted. Specifically, when N = 2, the

inter prediction units

126 and 218 classify all candidate motion vectors into two groups as shown in FIG. The first group is a group to which candidate motion vectors obtained based on, for example, the motion vector of a block in the encoding target picture belong. The second group is a group to which candidate motion vectors obtained based on, for example, a motion vector of a block in a picture different from the picture to be encoded belong.

Then, the

inter prediction units

126 and 218 extract one candidate motion vector having the highest evaluation value as the predicted motion vector candidate 1 from the first group. Furthermore, the

inter prediction units

126 and 218 extract one candidate motion vector having the highest evaluation value in the second group as the predicted motion vector candidate 2 from the second group.

Alternatively, the

inter prediction units

126 and 218 may classify a plurality of candidate motion vectors into M groups (M is an integer larger than N). Then, the

inter prediction units

126 and 218 select, from each of the M groups, one candidate motion vector having the best evaluation result in that group as a representative candidate motion vector. Next, the

inter prediction units

126 and 218 select the top N representative candidate motion vectors from the selected M representative candidate motion vectors in the order of good evaluation results, and all the at least one predicted motion vector candidate described above. May be extracted as

Specifically, when M = 3, the

inter prediction units

126 and 218 classify all candidate motion vectors into three groups as shown in (c) of FIG. The first group is a group to which candidate motion vectors obtained based on, for example, the motion vector of the block on the left side of the encoding target block in the encoding target picture belong. The second group is a group to which candidate motion vectors obtained based on, for example, the motion vector of the block above the encoding target block in the encoding target picture belong. The third group is a group to which candidate motion vectors obtained based on, for example, a motion vector of a block in a picture different from the encoding target picture belong.

Then, the

inter prediction units

126 and 218 select one candidate motion vector having the highest evaluation value in the first group as the representative candidate motion vector 1 from the first group. Furthermore, the

inter prediction units

126 and 218 select, from the second group, one candidate motion vector having the highest evaluation value in the second group as the representative candidate motion vector 2. Furthermore, the

inter prediction units

126 and 218 select one candidate motion vector having the highest evaluation value in the third group as the representative candidate motion vector 3 from the third group.

Next, the

inter prediction units

126 and 218 respectively extract the top two representative candidate motion vectors as predicted motion vector candidates in descending order of evaluation value from the extracted three representative candidate motion vectors.

In the example shown in FIG. 19, two prediction motion vector candidates are extracted, but the number is not limited to two, and three or more prediction motion vector candidates may be extracted.

[Effects of Second Embodiment, etc.]
The encoding device in the present embodiment is an encoding device that encodes a moving image, and includes a processing circuit and a memory connected to the processing circuit, and the processing circuit uses the memory, A plurality of candidate motion vectors are acquired based on respective motion vectors of a plurality of encoded blocks corresponding to the encoding target block in the moving image, and at least one of the encoding target blocks is obtained from the plurality of candidate motion vectors. One prediction motion vector candidate is extracted, a motion vector of the coding target block is derived with reference to a reference picture included in the moving image, and a prediction motion vector of the extracted at least one prediction motion vector candidate is extracted And a difference between the derived motion vector of the encoding target block and a motion of the derived encoding target block. Motion compensation is performed on the encoding target block using a vector, and in the extraction of the at least one prediction motion vector candidate, all of the at least one prediction motion vector candidate is converted into an image area of the encoding target block. Extraction is performed based on the respective evaluation results of the plurality of candidate motion vectors using the reconstructed image of the encoded region in the moving image without using it. The memory may be the frame memory 122 or another memory, and the processing circuit may include, for example, the inter prediction unit 126 and the entropy encoding unit 110.

Thereby, all of at least one prediction motion vector candidate is the evaluation result of each of the plurality of candidate motion vectors using the reconstructed image of the encoded region in the moving image without using the image region of the encoding target block. That is, it is extracted based on the evaluation result by FRUC. Therefore, the prediction accuracy of the prediction block that is the encoding target block can be increased, and the encoding efficiency can be improved. Furthermore, in this embodiment, a motion vector predictor candidate is not extracted based on a predetermined priority order. Therefore, if a candidate list for extraction based on the evaluation result by FRUC is generated, all motion vector predictor candidates can be extracted, and it is not necessary to generate a candidate list for extraction based on priority. Therefore, it is possible to improve the encoding efficiency while suppressing an increase in processing load.

In addition, in the extraction of the at least one predicted motion vector candidate, the processing circuit selects only the one candidate motion vector having the best evaluation result from the plurality of candidate motion vectors, whereby the at least one predicted motion vector is selected. Extracting all motion vector candidates, and encoding the difference, the difference between the predicted motion vector that is the selected candidate motion vector and the derived motion vector of the encoding target block may be encoded. Good.

Thereby, for example, as shown in FIG. 15, one predicted motion vector candidate is extracted, and the predicted motion vector candidate is selected as the predicted motion vector. On the other hand, when a plurality of predicted motion vector candidates are extracted and one predicted motion vector is selected from the plurality of predicted motion vector candidates, information for identifying the selected predicted motion vector is encoded. Must be included in the stream. However, in the example illustrated in FIG. 15, one prediction motion vector candidate is extracted and the prediction motion vector candidate is selected as the prediction motion vector, and thus it is not necessary to encode such information. Therefore, the code amount can be reduced.

In the extraction of the at least one predicted motion vector candidate, the processing circuit selects N (N is an integer of 2 or more) candidate motion vectors based on the evaluation result from the plurality of candidate motion vectors. Extracting as all of at least one predicted motion vector candidate, and the processing circuit further selects the predicted motion vector from the extracted N predicted motion vector candidates and identifies the selected predicted motion vector In the encoding of the difference, the difference between the selected predicted motion vector and the derived motion vector of the encoding target block may be encoded.

As a result, for example, as shown in FIG. 17, a plurality of motion vector predictor candidates are extracted, and prediction motion vector candidates with higher prediction accuracy are predicted from among them by using an image of the prediction block that is the encoding target block. It can be selected as a motion vector. Therefore, the encoding efficiency can be improved. In addition, since the selection information for identifying the prediction motion vector selected in this way is encoded, the decoding device decodes the selection information, thereby the prediction selected as the prediction motion vector in the encoding device. The motion vector candidate can be appropriately identified. Therefore, the encoded moving image can be appropriately decoded by the decoding device.

In addition, in the extraction of the at least one predicted motion vector candidate, the processing circuit obtains the top N candidate motion vectors from the plurality of candidate motion vectors in descending order of the evaluation result, and the at least one predicted motion vector. You may extract as all of the candidates. For example, each evaluation result of the plurality of candidate motion vectors has a small difference between the reconstructed image of the first encoded region specified by the candidate motion vector and the second encoded reconstructed image. It is a good evaluation result.

Thereby, for example, as shown in FIG. 19A, N predicted motion vector candidates with high prediction accuracy can be preferentially selected from a plurality of candidate motion vectors.

In addition, in the extraction of the at least one predicted motion vector candidate, the processing circuit classifies the plurality of candidate motion vectors into N groups, and the evaluation result in the group is determined from each of the N groups. All of the at least one predicted motion vector candidate may be extracted by extracting the best candidate motion vector. For example, as shown in FIG. 19B, the plurality of candidate motion vectors are classified into N groups having different properties. Since one predicted motion vector candidate with the best evaluation result is extracted from each of the N groups, N predicted motion vector candidates having different properties and high prediction accuracy can be extracted. . As a result, the range of selection of the prediction motion vector can be expanded, and the possibility that a prediction motion vector with higher prediction accuracy is selected can be increased.

In the extraction of the at least one predicted motion vector candidate, the processing circuit classifies the plurality of candidate motion vectors into M groups (M is an integer greater than N), and from each of the M groups. , One candidate motion vector having the best evaluation result in the group is selected as a representative candidate motion vector, and the top N representative candidates in order of the evaluation result from the selected M representative candidate motion vectors. Motion vectors may be extracted as all of the at least one predicted motion vector candidate.

As a result, for example, as shown in FIG. 19C, even when a plurality of candidate motion vectors are classified into more groups than the number of predicted motion vector candidates to be extracted (that is, N). It is possible to extract N predicted motion vector candidates having different properties and high prediction accuracy.

The decoding device according to the present embodiment is a decoding device that decodes an encoded moving image, and includes a processing circuit and a memory connected to the processing circuit, and the processing circuit includes the memory. To obtain a plurality of candidate motion vectors based on respective motion vectors of a plurality of decoded blocks corresponding to a decoding target block in the moving image, and from the plurality of candidate motion vectors, at least one of the decoding target blocks One prediction motion vector candidate is extracted, difference information indicating a difference between two motion vectors is decoded, and a prediction of the extracted at least one prediction motion vector candidate is obtained as a difference indicated by the decoded difference information. A motion vector of the decoding target block is derived by adding a motion vector, and the derived decoding target block is derived. Motion compensation is performed on the decoding target block using a motion vector of a block, and in the extraction of the at least one prediction motion vector candidate, all of the at least one prediction motion vector candidate Extraction is performed based on the respective evaluation results of the plurality of candidate motion vectors using the reconstructed image of the decoded region in the moving image without using the region. The memory may be the frame memory 214 or another memory, and the processing circuit may include, for example, the inter prediction unit 218 and the entropy decoding unit 202.

Thereby, all of the at least one predicted motion vector candidate are evaluated results of each of the plurality of candidate motion vectors using the reconstructed image of the decoded region in the moving image without using the image region of the decoding target block, that is, Extracted based on the evaluation result by FRUC. Therefore, it is possible to improve the prediction efficiency of the prediction block that is the decoding target block and improve the encoding efficiency. Furthermore, in this embodiment, a motion vector predictor candidate is not extracted based on a predetermined priority order. Therefore, if a candidate list for extraction based on the evaluation result by FRUC is generated, all motion vector predictor candidates can be extracted, and it is not necessary to generate a candidate list for extraction based on priority. Therefore, it is possible to improve the encoding efficiency while suppressing an increase in processing load.

In addition, in the extraction of the at least one predicted motion vector candidate, the processing circuit selects only the one candidate motion vector having the best evaluation result from the plurality of candidate motion vectors, whereby the at least one predicted motion vector is selected. By extracting all motion vector candidates and deriving the motion vector of the decoding target block, the predicted motion vector that is the selected candidate motion vector is added to the difference indicated by the decoded difference information. The motion vector of the decoding target block may be derived.

Thereby, for example, as shown in FIG. 16, one predicted motion vector candidate is extracted, and the predicted motion vector candidate is selected as the predicted motion vector. On the other hand, when a plurality of motion vector predictor candidates are extracted, information for identifying the motion vector predictor selected by the encoding device from the motion vector predictor candidates needs to be decoded from the stream. However, in the example illustrated in FIG. 16, one prediction motion vector candidate is extracted and the prediction motion vector candidate is selected as the prediction motion vector, and thus it is not necessary to decode such information. Therefore, the code amount can be reduced.

In the extraction of the at least one predicted motion vector candidate, the processing circuit selects N (N is an integer of 2 or more) candidate motion vectors based on the evaluation result from the plurality of candidate motion vectors. Extracting as all of at least one predicted motion vector candidate, and the processing circuit further decodes selection information for identifying the predicted motion vector, decoded from the extracted N predicted motion vector candidates The predicted motion vector candidate identified by the selection information is selected as the predicted motion vector, and in the derivation of the motion vector of the decoding target block, the selected predicted motion is added to the difference indicated by the decoded difference information. A motion vector of the decoding target block may be derived by adding the vectors.

Thereby, for example, as shown in FIG. 18, a plurality of motion vector predictor candidates are extracted, and from them, a motion vector predictor candidate with higher prediction accuracy can be selected as a motion vector predictor based on selection information. Therefore, the encoding efficiency can be improved.

In addition, in the extraction of the at least one predicted motion vector candidate, the processing circuit obtains the top N candidate motion vectors from the plurality of candidate motion vectors in descending order of the evaluation result, and the at least one predicted motion vector. You may extract as all of the candidates. For example, the evaluation result of each of the plurality of candidate motion vectors is better as the difference between the reconstructed image of the first decoded area specified by the candidate motion vector and the second decoded reconstructed image is smaller. It is an evaluation result.

In the extraction of the at least one predicted motion vector candidate, the processing circuit classifies the plurality of candidate motion vectors into M groups (M is an integer greater than N), and each of the M groups. Then, one candidate motion vector having the best evaluation result in the group is selected as a representative candidate motion vector, and the top N representatives in order of good evaluation result from the selected M representative candidate motion vectors. Candidate motion vectors may be extracted as all of the at least one predicted motion vector candidate.

These comprehensive or specific aspects may be realized by a non-transitory recording medium such as a system, apparatus, method, integrated circuit, computer program, or computer-readable CD-ROM. The present invention may be realized by any combination of a method, an integrated circuit, a computer program, and a recording medium.

(Embodiment 3)
[FRUC / priority switching]
The encoding apparatus and decoding apparatus in the present embodiment have the same configuration as that of Embodiment 1, but are characterized by the processing operations of

inter prediction units

126 and 218. That is, in the present embodiment as well as in the second embodiment, a problem in the above [knowledge that became the basis of the present disclosure], that is, a plurality of different candidate lists for one prediction block must be created. It solves the problem that the processing load increases.

Such an encoding apparatus 100 according to the present embodiment encodes mode information for identifying an extraction method in the above-described extraction of at least one prediction motion vector candidate. Then, the encoding device 100 selects the extraction method identified by the mode information for the block to be encoded from the first extraction method and the second extraction method, and, according to the selected extraction method, At least one predicted motion vector candidate is extracted. Here, the first extraction method is an extraction method based on the evaluation results of a plurality of candidate motion vectors using the reconstructed image of the encoded region in the moving image without using the image region of the encoding target block. It is. In addition, the second extraction method is an extraction method based on a predetermined priority order for a plurality of candidate motion vectors.

Also, the decoding apparatus 200 according to the present embodiment decodes mode information for identifying an extraction method in the extraction of at least one predicted motion vector candidate. Then, the decoding apparatus 200 selects an extraction method identified for the block to be decoded by the decoded mode information from the first extraction method and the second extraction method, and according to the selected extraction method, At least one predicted motion vector candidate is extracted.

That is, encoding apparatus 100 and decoding apparatus 200 according to the present embodiment, for each prediction block, extract at least one prediction motion vector candidate, an extraction method based on an evaluation result by FRUC, and a predetermined priority order. Switch to the extraction method based on.

This makes it possible to suppress an increase in processing load without creating a plurality of different candidate lists for the prediction block.

FIG. 20 is a flowchart showing a method of selecting a motion vector predictor by the encoding device 100 and the decoding device 200 according to this embodiment.

The

inter prediction units

126 and 218 determine whether the mode information indicates 0 or 1 (step S301). The mode information is information for identifying an extraction method of at least one prediction motion vector candidate. Specifically, when mode information = 0, the mode information indicates a first extraction method, that is, an extraction method based on an evaluation result by FRUC. When mode information = 1, the mode information indicates a second extraction method, that is, an extraction method according to a predetermined priority order.

Here, when it is determined that the mode information indicates 0, the

inter prediction units

126 and 218 extract at least one motion vector predictor candidate based on the evaluation result by FRUC, as in the second embodiment (Step S1). S302). Specifically, the

inter prediction units

126 and 218 perform evaluation using encoded or decoded reconstructed images for each of a plurality of candidate motion vectors. Then, the

inter prediction units

126 and 218 extract at least one prediction motion vector candidate from the plurality of candidate motion vectors based on the evaluation result.

On the other hand, when the

inter prediction units

126 and 218 determine that the mode information indicates 1 in step S301, N predictions (N is an integer equal to or greater than 2) are predicted as in the examples illustrated in FIGS. Motion vector candidates are extracted (step S303). Specifically, the

inter prediction units

126 and 218 extract N predicted motion vector candidates from a plurality of candidate motion vectors according to a predetermined priority order.

When at least one prediction motion vector candidate is extracted in step S302, the

inter prediction units

126 and 218 determine whether or not the number of the extracted prediction motion vector candidates is plural (step S304). When the

inter prediction units

126 and 218 determine that the number of extracted motion vector predictor candidates is one (No in step S304), the inter

motion prediction units

126 and 218 determine the extracted motion vector predictor candidates as encoding target blocks. It selects as a prediction motion vector of a certain prediction block (step S305).

On the other hand, when the

inter prediction units

126 and 218 determine that the number of extracted motion vector predictor candidates is plural (Yes in step S304), the

inter prediction units

126 and 218 are identified by the motion vector predictor selection information from the plurality of motion vector predictor candidates. The predicted motion vector candidate is selected as the predicted motion vector of the predicted block (step S306).

In addition, when N predicted motion vector candidates are extracted in step S303, the

inter prediction units

126 and 218 select a predicted motion vector candidate indicated by the predicted motion vector selection information from the N predicted motion vector candidates. It selects as a prediction motion vector of a prediction block (step S306).

Note that the above-described mode information is encoded into a stream by the entropy encoding unit 110 of the encoding device 100, and is decoded from the stream by the entropy decoding unit 202 of the decoding device 200.

Such mode information is encoded in the header area of any one of the sequence layer, the picture layer, and the slice layer. That is, the entropy encoding unit 110 encodes mode information for identifying an extraction method for each block included in the layer in the header area of the layer. In addition, the entropy decoding unit 202 decodes mode information for identifying an extraction method for each block included in the layer from the header area of the layer.

Thereby, the extraction method based on FRUC and the extraction method according to a predetermined priority order can be switched for each sequence, picture or slice. Further, the mode information may be encoded in the stream in units of prediction blocks. That is, the entropy encoding unit 110 encodes mode information for identifying an extraction method for the block for each block included in the moving image. Further, the entropy decoding unit 202 decodes mode information for identifying an extraction method for each block included in the moving image. Thereby, those extraction methods can be switched for every prediction block.

In addition, the above-mentioned numerical value (0 or 1) shown by mode information is an example, Comprising: Numerical values other than these numerical values may be sufficient. The mode information may indicate an identifier other than a numerical value. That is, the mode information may indicate any identifier as long as it is an identifier that can distinguish the extraction method based on the FRUC evaluation result and the extraction method according to a predetermined priority order.

Also, the above-described prediction motion vector selection information is encoded into a stream in units of prediction blocks by the entropy encoding unit 110 of the encoding device 100, and is decoded from the stream by the entropy decoding unit 202 of the decoding device 200.

[Effects of Embodiment 3 and the like]
The encoding device in the present embodiment is an encoding device that encodes a moving image, and includes a processing circuit and a memory connected to the processing circuit, and the processing circuit uses the memory, A plurality of candidate motion vectors are acquired based on respective motion vectors of a plurality of encoded blocks corresponding to the encoding target block in the moving image, and at least one of the encoding target blocks is obtained from the plurality of candidate motion vectors. One prediction motion vector candidate is extracted, a motion vector of the coding target block is derived with reference to a reference picture included in the moving image, and a prediction motion vector of the extracted at least one prediction motion vector candidate is extracted And a difference between the derived motion vector of the encoding target block and a motion of the derived encoding target block. The vector is used to perform motion compensation on the encoding target block, and in the extraction of the at least one predicted motion vector candidate, mode information for identifying the extraction method is encoded, and the first extraction method and the second The extraction method identified by the mode information is selected for the encoding target block, and the at least one prediction motion vector candidate is extracted according to the selected extraction method, and the first Is an extraction method based on the evaluation results of each of the plurality of candidate motion vectors using the reconstructed image of the encoded region in the moving image without using the image region of the encoding target block. The second extraction method is an extraction method based on a predetermined priority order for the plurality of candidate motion vectors. The memory may be the frame memory 122 or another memory, and the processing circuit may include, for example, the inter prediction unit 126 and the entropy encoding unit 110.

Thereby, for example, the first extraction method based on the evaluation result by FRUC or the second extraction method based on a predetermined priority order is applied to the prediction block which is the encoding target block according to the mode information. Is done. That is, the extraction method is switched. Therefore, the prediction accuracy of the prediction block that is the encoding target block can be increased, and the encoding efficiency can be improved. Furthermore, in this embodiment, since any one of the first extraction method and the second extraction method is applied to the prediction block, the candidate for the first extraction method for the prediction block is used. There is no need to separately generate a list and a candidate list for the second extraction method. Therefore, it is possible to improve the encoding efficiency while suppressing an increase in processing load.

In the encoding of the mode information, an extraction method for each block included in the layer is identified in a header area of any one of a sequence layer, a picture layer, and a slice layer in the moving image stream. Mode information may be encoded.

This makes it possible to switch the extraction method in sequence, picture or slice units. Moreover, the code amount of mode information can be suppressed compared with the case where the extraction method is switched in units of blocks such as prediction blocks.

In the encoding of the mode information, mode information for identifying an extraction method for the block may be encoded for each block included in the moving image.

This makes it possible to switch the extraction method in units of blocks such as prediction blocks. In addition, the possibility of improving the prediction accuracy of a block can be increased as compared with the case of switching the extraction method in units of sequences, pictures, or slices.

The decoding device according to the present embodiment is a decoding device that decodes an encoded moving image, and includes a processing circuit and a memory connected to the processing circuit, and the processing circuit includes the memory. To obtain a plurality of candidate motion vectors based on respective motion vectors of a plurality of decoded blocks corresponding to a decoding target block in the moving image, and from the plurality of candidate motion vectors, at least one of the decoding target blocks One prediction motion vector candidate is extracted, difference information indicating a difference between two motion vectors is decoded, and a prediction of the extracted at least one prediction motion vector candidate is obtained as a difference indicated by the decoded difference information. A motion vector of the decoding target block is derived by adding a motion vector, and the derived decoding target block is derived. A first motion compensation method for performing the motion compensation on the decoding target block by using a motion vector of a signal, and extracting the at least one predicted motion vector candidate by decoding mode information for identifying an extraction method; And an extraction method identified for the decoding target block by the decoded mode information is selected from the second extraction method, and the at least one motion vector predictor candidate is extracted according to the selected extraction method In the first extraction method, extraction is performed based on each evaluation result of the plurality of candidate motion vectors using the reconstructed image of the decoded region in the moving image without using the image region of the decoding target block. The second extraction method is an extraction method based on a predetermined priority order for the plurality of candidate motion vectors.The memory may be the frame memory 214 or another memory, and the processing circuit may include, for example, the inter prediction unit 218 and the entropy decoding unit 202.

Accordingly, for example, the first extraction method based on the evaluation result by FRUC or the second extraction method based on a predetermined priority order is applied to the prediction block which is the decoding target block according to the mode information. The That is, the extraction method is switched. Therefore, it is possible to improve the prediction efficiency of the prediction block that is the decoding target block and improve the encoding efficiency. Furthermore, in this embodiment, since any one of the first extraction method and the second extraction method is applied to the prediction block, the candidate for the first extraction method for the prediction block is used. There is no need to separately generate a list and a candidate list for the second extraction method. Therefore, it is possible to improve the encoding efficiency while suppressing an increase in processing load.

In the decoding of the mode information, in order to identify the extraction method for each block included in the layer from the header area of any one of the sequence layer, the picture layer, and the slice layer in the moving image stream The mode information may be decoded.

Further, in the decoding of the mode information, mode information for identifying an extraction method for the block may be decoded for each block included in the moving image.

(Embodiment 4)
[Extract motion vector candidates with FRUC / priority from common candidate list]
The encoding apparatus and decoding apparatus in the present embodiment have the same configuration as that of Embodiment 1, but are characterized by the processing operations of

inter prediction units

126 and 218. That is, in the present embodiment, as in the second and third embodiments, a plurality of candidate lists different from each other are created for the problem in the above-mentioned [Knowledge on which this disclosure is based], that is, one prediction block. This solves the problem that the processing load increases.

The encoding apparatus 100 according to the present embodiment extracts N predicted motion vector candidates (N is an integer of 2 or more) for the encoding target block from a plurality of candidate motion vectors. At this time, the encoding apparatus 100 generates a candidate list that is a list indicating a plurality of candidate motion vectors and is common to the first extraction method and the second extraction method. Then, encoding apparatus 100 extracts M (M is an integer less than or equal to 1 and less than N) predicted motion vector candidates from a plurality of candidate motion vectors indicated in the common candidate list according to the first extraction method. To do. Furthermore, encoding apparatus 100 extracts L (L = N−M) predicted motion vector candidates from a plurality of candidate motion vectors shown in the common candidate list according to the second extraction method. Here, the first extraction method is an extraction method based on the evaluation results of a plurality of candidate motion vectors using the reconstructed image of the encoded region in the moving image without using the image region of the encoding target block. Specifically, the extraction method is based on the evaluation result by FRUC. In addition, the second extraction method is an extraction method based on a predetermined priority order for a plurality of candidate motion vectors.

Also, decoding apparatus 200 according to the present embodiment extracts N (N is an integer of 2 or more) prediction motion vector candidates for the decoding target block from a plurality of candidate motion vectors. At this time, the decoding apparatus 200 generates a candidate list that is a list indicating a plurality of candidate motion vectors and is common to the first extraction method and the second extraction method. Then, decoding apparatus 200 extracts M (M is an integer of 1 or more and less than N) predicted motion vector candidates from a plurality of candidate motion vectors shown in the common candidate list according to the first extraction method. . Furthermore, the decoding apparatus 200 extracts L (L = N−M) predicted motion vector candidates from a plurality of candidate motion vectors shown in the common candidate list according to the second extraction method.

FIG. 21 is a flowchart showing an example of motion compensation by the encoding apparatus 100 according to the present embodiment. When the encoding device 100 illustrated in FIG. 1 encodes a moving image including a plurality of pictures, the inter prediction unit 126 and the like of the encoding device 100 execute the processing illustrated in FIG.

Specifically, the inter prediction unit 126 performs motion compensation for each prediction block corresponding to the above-described prediction unit, with respect to the encoding target block that is the prediction block. At this time, the inter prediction unit 126 first determines a plurality of candidate motion vectors for the prediction block based on information such as motion vectors of a plurality of encoded blocks around the prediction block temporally or spatially. Is acquired (step S201). At this time, the inter prediction unit 126 is a candidate list indicating a plurality of candidate motion vectors acquired in step S201, and includes an extraction method based on an evaluation result by FRUC and an extraction method based on a predetermined priority order. Generate a common candidate list.

Next, the inter prediction unit 126 calculates the evaluation values of the plurality of candidate motion vectors acquired in step S201 using the reconstructed image of the encoded region. That is, the inter prediction unit 126 calculates those evaluation values based on FRUC, that is, the template matching method or the bilateral matching method. Then, based on the evaluation values of the plurality of candidate motion vectors, the inter prediction unit 126 calculates each of the M candidate motion vectors from among the plurality of candidate motion vectors shown in the common candidate list. Extracted as candidate 1 (step S202aa). That is, the inter prediction unit 126 extracts the top M candidate motion vectors as the predicted motion vector candidates from the plurality of candidate motion vectors in descending order of evaluation value. In addition, the inter prediction unit 126 finely moves the selected motion vector predictor candidate 1 in the peripheral region so that the FRUC evaluation value becomes higher for each of the extracted M motion vector predictor candidates. Thus, the predicted motion vector candidate 1 may be corrected. That is, the inter prediction unit 126 may correct the predicted motion vector candidates 1 by finely searching for a region where the FRUC evaluation value is higher.

Further, the inter prediction unit 126 extracts each of L candidate motion vectors as a predicted motion vector candidate 2 from the plurality of candidate motion vectors shown in the common candidate list according to a predetermined priority order. (Step S202ab).

Then, the inter prediction unit 126 selects one predicted motion vector candidate from the extracted M predicted motion vector candidates 1 and L predicted motion vector candidates 2 as a predicted motion vector of the predicted block. (Step S202b). At this time, the inter prediction unit 126 outputs prediction motion vector selection information for identifying the selected prediction motion vector. The entropy encoding unit 110 encodes the predicted motion vector selection information into a stream.

FIG. 22 is a flowchart showing an example of motion compensation by the decoding apparatus 200 according to this embodiment. When the decoding device 200 illustrated in FIG. 10 decodes a moving image including a plurality of encoded pictures, the inter prediction unit 218 and the like of the decoding device 200 perform the processing illustrated in FIG.

At this time, the inter prediction unit 218 is a candidate list indicating a plurality of candidate motion vectors acquired in step S211, and includes an extraction method based on an evaluation result by FRUC and an extraction method based on a predetermined priority order. Generate a common candidate list.

Next, the inter prediction unit 218 calculates each evaluation value of the plurality of candidate motion vectors acquired in step S211 using the reconstructed image of the decoded area. That is, the inter prediction unit 218 calculates the evaluation values based on FRUC, that is, the template matching method or the bilateral matching method. Then, based on the evaluation values of the plurality of candidate motion vectors, the inter prediction unit 218 predicts each of the M candidate motion vectors from among the plurality of candidate motion vectors indicated in the common candidate list. Extracted as candidate 1 (step S212aa). That is, the inter prediction unit 218 extracts the top M candidate motion vectors as the predicted motion vector candidates from the plurality of candidate motion vectors in descending order of evaluation value. Note that the inter prediction unit 218 finely moves the selected predicted motion vector candidate 1 in the peripheral region so that the FRUC evaluation value becomes higher for each of the extracted M predicted motion vector candidates. Thus, the predicted motion vector candidate 1 may be corrected. That is, the inter prediction unit 218 may correct the motion vector predictor candidates 1 by finely searching for a region where the FRUC evaluation value is higher.

Further, the inter prediction unit 218 extracts each of the L candidate motion vectors as a predicted motion vector candidate 2 from the plurality of candidate motion vectors shown in the common candidate list according to a predetermined priority order. (Step S212ab).

Next, the inter prediction unit 218 uses the motion vector predictor selection information to select one motion vector predictor candidate from the extracted M motion vector predictor candidates 1 and L motion vector predictor candidates 2 as a prediction block. It selects as a prediction motion vector (step S212b). That is, the entropy decoding unit 202 decodes prediction motion vector selection information for identifying a prediction motion vector of a prediction block that is a decoding target block. Then, the inter prediction unit 218 selects a predicted motion vector candidate identified by the decoded predicted motion vector selection information from the extracted N predicted

motion vector candidates

1 and 2 as a predicted motion vector of the predicted block. .

Here, in the present embodiment, in the extraction according to the second extraction method by the

inter prediction units

126 and 218, that is, extraction based on a predetermined priority order, the extraction result of the first extraction method, that is, You may utilize the extraction result based on the evaluation result by FRUC. That is, the

inter prediction units

126 and 218, at least one candidate motion other than the M predicted motion vector candidates extracted by the first extraction method, from among a plurality of candidate motion vectors shown in the common candidate list. L predicted motion vector candidates are extracted from the vector according to the priority order using the evaluation result in the first extraction method.

FIG. 23 is a diagram for explaining a method of extracting a motion vector predictor candidate according to the present embodiment.

For example, the

inter prediction units

126 and 218 classify a plurality of candidate motion vectors shown in the common candidate list into K groups (K is an integer of 2 or more). Then, in the extraction of M motion vector predictor candidates 1, the

inter prediction units

126 and 218 select the top M candidates in descending order of the evaluation result from a plurality of candidate motion vectors indicated in the common candidate list. A motion vector is extracted as M predicted motion vector candidates 1. Further, in the extraction of L predicted motion vector candidates 2, the

inter prediction units

126 and 218 exclude at least one of the common candidate lists excluding the group to which each of the M predicted motion vector candidates 1 belongs. L predicted motion vector candidates 2 are extracted from one or more candidate motion vectors belonging to any of the groups according to priority.

Specifically, as illustrated in (a) of FIG. 23, when K = 3, M = 1, and L = 1, the

inter prediction units

126 and 218 first include a plurality of candidates indicated in the common candidate list. Candidate motion vectors are classified into three groups G1 to C3. The group G1 is a group to which candidate motion vectors obtained based on, for example, the motion vector of the block on the left side of the encoding target block in the encoding target picture belong. The group G2 is a group to which a candidate motion vector obtained based on, for example, a motion vector of a block above the encoding target block in the encoding target picture belongs. The group G3 is a group to which a candidate motion vector obtained based on, for example, a motion vector of a block in a picture different from the encoding target picture belongs.

Next, the

inter prediction units

126 and 218 extract the candidate motion vector having the highest evaluation value as the predicted motion vector candidate 1 from the plurality of candidate motion vectors shown in the common candidate list. Next, the

inter prediction units

126 and 218 prioritize one or more candidate motion vectors belonging to any of the groups G2 and G3, except for the group G1 to which the prediction motion vector candidate 1 belongs, from the common candidate list. One candidate motion vector is extracted as a predicted motion vector candidate 2 according to the rank.

Alternatively, the

inter prediction units

126 and 218 classify a plurality of candidate motion vectors shown in the common candidate list into K groups. Then, in the extraction of M motion vector predictor candidates 1, the

inter prediction units

126 and 218 determine the top M candidate motion vectors in descending order of evaluation results from a plurality of candidate motion vectors indicated in the common candidate list. Are extracted as M predicted motion vector candidates 1. Further, the

inter prediction units

126 and 218, from a plurality of candidate motion vectors belonging to any one of at least one group excluding the group to which each of the M motion vector predictor candidates 1 belongs, from the common candidate list. The candidate motion vector having the best evaluation result is identified as the next predicted motion vector candidate. Then, in the extraction of L predicted motion vector candidates 2, the

inter prediction units

126 and 218 include one or more members belonging to the same group as the group to which the identified next-point predicted motion vector candidate belongs in the common candidate list. L predicted motion vectors 2 are extracted from the candidate motion vectors according to the priority order.

Specifically, as illustrated in (b) of FIG. 23, when K = 3, M = 1, and L = 1, the

inter prediction units

126 and 218 first, as in the above example, A plurality of candidate motion vectors shown in the candidate list are classified into three groups.

Next, the

inter prediction units

126 and 218 extract the candidate motion vector having the highest evaluation value as the predicted motion vector candidate 1 from the plurality of candidate motion vectors shown in the common candidate list. Further, the

inter prediction units

126 and 218 have an evaluation value of a plurality of candidate motion vectors belonging to any of the groups G2 and G3, excluding the group G1 to which the prediction motion vector candidate 1 belongs, from the common candidate list. The highest candidate motion vector 4 is specified as the next predicted motion vector candidate. Then, the

inter prediction units

126 and 218 determine, from one or more candidate motion vectors belonging to the same group as the group G2 to which the candidate motion vector 4 belongs, of the identified next-point predicted motion vector candidates in the common candidate list. One candidate motion vector is extracted as a predicted motion vector candidate 2 in accordance with the priority order.

FIG. 24 is a diagram showing an example of a common candidate list.

For example, the

inter prediction units

126 and 218 may use the common candidate shown in (b) of FIG. 24 for the encoding target block or the decoding target block (hereinafter simply referred to as a processing target block) shown in (a) of FIG. Generate a list. This common candidate list includes an L0 list and an L1 list.

Specifically, the

inter prediction units

126 and 218 include candidate motion vectors based on the motion vectors of

adjacent blocks

1, 2, and 5 adjacent to the processing target block in the common candidate list. The adjacent block 1 is a block adjacent to the lower left of the processing target block, the adjacent block 2 is a block adjacent to the upper right of the processing target block, and the adjacent block 5 is a block adjacent to the upper left of the processing target block. .

For example, the adjacent block 1 is encoded or decoded by the motion vectors mvL01 and mvL11. The adjacent block 2 is encoded or decoded by the motion vectors mvL02 and mvL12. The adjacent block 5 is encoded or decoded by the motion vector mvL05. In such a case, as illustrated in FIG. 24B, the

inter prediction units

126 and 218 include candidate motion vectors based on these motion vectors as a spatial candidate motion vector in a common candidate list. Note that the

inter prediction units

126 and 218 may include candidate motion vectors based on motion vectors of other adjacent blocks in the common candidate list as spatial candidate motion vectors. For example, the other adjacent block is the adjacent block 3 on the right side of the adjacent block 2 or the adjacent block 4 on the lower side of the adjacent block 1. Also, the

inter prediction units

126 and 218 may scale the motion vectors of adjacent blocks based on the display time interval, and include the scaled motion vectors as candidate motion vectors in the candidate list.

Furthermore, the

inter prediction units

126 and 218 may include the temporal candidate motion vector and the combined bi-prediction candidate motion vector (mvL0b, mvL1b) in the candidate list. The temporal candidate motion vectors include, for example, a Col candidate motion vector (mvL0t, mvL1t) and a Universal candidate motion vector (mvL0u, mvL1u). The Col candidate motion vector (mvL0t, mvL1t) is a candidate motion vector based on a motion vector of a block in a picture different from the picture including the processing target block, for example, a block at the same position as the processing target block. Note that the Col candidate motion vector (mvL0t, mvL1t) may be a candidate motion vector obtained by scaling a motion vector of a block in a picture different from the picture including the processing target block at a display time interval. Further, the Col candidate motion vector may be a candidate motion vector based on a motion vector of a block at a position different from the processing target block. Furthermore, a plurality of different Col candidate motion vectors may be included in the candidate list. The Universal candidate motion vector (mvL0u, mvL1u) is a block in a picture different from the picture including the processing target block, and the motion vector of the block at a position that takes into account the amount of movement with the passage of time with respect to the position of the processing target block Is a candidate motion vector based on. The combined bi-prediction candidate motion vector (mvL0b, mvL1b) is a candidate motion vector generated by combining the motion vectors of the L0 list and the L1 list of the candidate list.

In the present embodiment, for example, the candidate list shown in FIG. 24B is commonly used for the extraction method based on the evaluation result by FRUC and the extraction method based on a predetermined priority.

[Effects of Embodiment 4]
The encoding device in the present embodiment is an encoding device that encodes a moving image, and includes a processing circuit and a memory connected to the processing circuit, and the processing circuit uses the memory, A plurality of candidate motion vectors are acquired based on the motion vectors of a plurality of encoded blocks corresponding to the encoding target block in the moving image, and N for the encoding target block are obtained from the plurality of candidate motion vectors. Selection information for extracting prediction motion vector candidates (N is an integer of 2 or more), selecting a prediction motion vector from the extracted N prediction motion vector candidates, and identifying the selected prediction motion vector And a motion vector of the encoding target block is derived with reference to a reference picture included in the moving image, and the derived encoding target block is derived. A difference between the motion vector of the received image and the selected predicted motion vector is encoded, motion compensation is performed on the current block using the derived motion vector of the current block, and the N In the extraction of predicted motion vector candidates, a list showing the plurality of candidate motion vectors, which is a common candidate list for the first extraction method and the second extraction method, is generated and shown in the common candidate list. In accordance with the first extraction method, M (M is an integer of 1 or more and less than N) prediction motion vector candidates are extracted from the plurality of candidate motion vectors, and the plurality of candidates shown in the common candidate list L (L = N−M) predicted motion vector candidates are extracted from motion vectors according to the second extraction method, and the first extraction method includes the encoding target block. Is an extraction method based on the respective evaluation results of the plurality of candidate motion vectors using the reconstructed image of the encoded region in the moving image without using the image region of the moving image, This is an extraction method based on a predetermined priority order for a plurality of candidate motion vectors. The memory may be the frame memory 122 or another memory, and the processing circuit may include, for example, the inter prediction unit 126 and the entropy encoding unit 110.

Thus, M predicted motion vector candidates are extracted according to the first extraction method, that is, based on the evaluation result by FRUC. Therefore, the prediction accuracy of the prediction block that is the encoding target block can be increased, and the encoding efficiency can be improved. In this embodiment, a candidate list common to the first extraction method and the second extraction method is generated. That is, even when M predicted motion vector candidates are extracted according to the first extraction method, L predicted motion vector candidates are extracted according to the second extraction method based on a predetermined priority. Even in this case, a common candidate list is referenced. As a result, it is not necessary to individually generate a candidate list for the first extraction method and a candidate list for the second extraction method for the prediction block. Therefore, it is possible to improve the encoding efficiency while suppressing an increase in processing load.

In addition, in the extraction according to the second extraction method, the processing circuit includes the M number of motion vectors extracted by the first extraction method among the plurality of candidate motion vectors indicated in the common candidate list. The L predicted motion vector candidates may be extracted from the remaining at least one candidate motion vector excluding the predicted motion vector candidates according to the priority order using the evaluation result in the first extraction method.

For example, as shown in FIG. 21, in the extraction according to the second extraction method (for example, step S202ab), the extraction result according to the first extraction method (for example, step S202aa) is referred to. Therefore, it can be suppressed that the same candidate motion vector is extracted as a predicted motion vector candidate in the first extraction method and the second extraction method.

In the extraction of the N predicted motion vector candidates, the processing circuit classifies the plurality of candidate motion vectors shown in the common candidate list into K groups (K is an integer of 2 or more), In the extraction according to the first extraction method, the top M candidate motion vectors in the descending order of the evaluation result are selected from the plurality of candidate motion vectors shown in the common candidate list, and the M predicted motions. In the extraction according to the second extraction method extracted as a vector candidate, any one of at least one group excluding the group to which each of the M motion vector predictor candidates belongs is included in the common candidate list. The L predicted motion vector candidates may be extracted from one or more candidate motion vectors to which the motion vector belongs, according to the priority order. For example, each evaluation result of the plurality of candidate motion vectors has a small difference between the reconstructed image of the first encoded region specified by the candidate motion vector and the second encoded reconstructed image. It is a good evaluation result.

For example, as shown in FIG. 23A, a plurality of candidate motion vectors are classified into K groups having different properties. Then, one (M = 1) prediction motion vector candidate having the best evaluation result is extracted from the entire K groups, and the priority order is determined in advance from groups other than the group to which the prediction motion vector candidate belongs. Therefore, another (L = 1) prediction motion vector candidate is extracted. Therefore, two (N = 2) prediction motion vector candidates having different properties and high prediction accuracy can be extracted. As a result, the range of selection of the prediction motion vector can be expanded, and the possibility that a prediction motion vector with higher prediction accuracy is selected can be increased.

In the extraction of the N predicted motion vector candidates, the processing circuit classifies the plurality of candidate motion vectors shown in the common candidate list into K groups (K is an integer of 2 or more), In the extraction according to the first extraction method, the top M candidate motion vectors in the descending order of the evaluation result are selected from the plurality of candidate motion vectors shown in the common candidate list, and the M predicted motions. The candidate is extracted as a vector candidate, and further, the evaluation is performed from a plurality of candidate motion vectors belonging to any one of at least one group excluding a group to which each of the M predicted motion vector candidates belongs from the common candidate list. The candidate motion vector with the best result is identified as the next predicted motion vector candidate, and in the extraction according to the second extraction method, the common candidate The L predicted motion vector candidates may be extracted in accordance with the priority from one or more candidate motion vectors belonging to the same group as the group to which the identified next-point predicted motion vector candidate belongs. .

For example, as shown in FIG. 23B, the plurality of candidate motion vectors are classified into K groups having different properties. Then, one (M = 1) predicted motion vector candidate having the best evaluation result is extracted from the entire K groups, and the next-point predicted motion vector candidate is obtained from a group other than the group to which the predicted motion vector candidate belongs. Identified. Furthermore, another (L = 1) predicted motion vector candidate is extracted from the same group as the group to which the next-point predicted motion vector candidate belongs, according to the priority order. Therefore, two (N = 2) prediction motion vector candidates having different properties and high prediction accuracy can be extracted. As a result, the range of selection of the prediction motion vector can be expanded, and the possibility that a prediction motion vector with higher prediction accuracy is selected can be increased.

The decoding device according to the present embodiment is a decoding device that decodes an encoded moving image, and includes a processing circuit and a memory connected to the processing circuit, and the processing circuit includes the memory. To obtain a plurality of candidate motion vectors based on respective motion vectors of a plurality of decoded blocks corresponding to a decoding target block in the moving image, and from the plurality of candidate motion vectors, N pieces of motion vectors for the decoding target block are obtained. (N is an integer of 2 or more) predicted motion vector candidates are extracted, selection information for identifying a predicted motion vector of the decoding target block is decoded, and decoding is performed from the extracted N predicted motion vector candidates. A predicted motion vector candidate identified by the selected information is selected as the predicted motion vector, and a difference between the two motion vectors is indicated. The difference information is decoded, and the motion vector of the decoding target block is derived by adding the selected prediction motion vector to the difference indicated by the decoded difference information, and the derived decoding target block The motion vector is subjected to motion compensation using the motion vector, and the extraction of the N predicted motion vector candidates is a list indicating the plurality of candidate motion vectors, the first extraction method and the second A candidate list common to the extraction methods is generated, and M (M is an integer of 1 or more and less than N) according to the first extraction method from the plurality of candidate motion vectors indicated in the common candidate list. Predicted motion vector candidates are extracted, and the L motion vector candidates are extracted from the plurality of candidate motion vectors indicated in the common candidate list according to the second extraction method. L = N−M) as predicted motion vector candidates, and the first extraction method uses the reconstructed image of the decoded region in the moving image without using the image region of the decoding target block. This is an extraction method based on the evaluation results of each of the plurality of candidate motion vectors, and the second extraction method is an extraction method based on a predetermined priority order for the plurality of candidate motion vectors. The memory may be the frame memory 214 or another memory, and the processing circuit may include, for example, the inter prediction unit 218 and the entropy decoding unit 202.

Thus, M predicted motion vector candidates are extracted according to the first extraction method, that is, based on the evaluation result by FRUC. Therefore, it is possible to improve the prediction efficiency of the prediction block that is the decoding target block and improve the encoding efficiency. In this embodiment, a candidate list common to the first extraction method and the second extraction method is generated. That is, even when M predicted motion vector candidates are extracted according to the first extraction method, L predicted motion vector candidates are extracted according to the second extraction method based on a predetermined priority. Even in this case, a common candidate list is referenced. As a result, it is not necessary to individually generate a candidate list for the first extraction method and a candidate list for the second extraction method for the prediction block. Therefore, it is possible to improve the encoding efficiency while suppressing an increase in processing load.

For example, as shown in FIG. 22, in the extraction according to the second extraction method (for example, step S212ab), the extraction result according to the first extraction method (for example, step S212aa) is referred to. Therefore, it can be suppressed that the same candidate motion vector is extracted as a predicted motion vector candidate in the first extraction method and the second extraction method.

In the extraction of the N predicted motion vector candidates, the processing circuit classifies the plurality of candidate motion vectors shown in the common candidate list into K groups (K is an integer of 2 or more), In the extraction according to the first extraction method, the top M candidate motion vectors in the descending order of the evaluation result are selected from the plurality of candidate motion vectors shown in the common candidate list, and the M predicted motions. In the extraction according to the second extraction method extracted as a vector candidate, any one of at least one group excluding the group to which each of the M motion vector predictor candidates belongs is included in the common candidate list. The L predicted motion vector candidates may be extracted from one or more candidate motion vectors to which the motion vector belongs, according to the priority order. For example, the evaluation result of each of the plurality of candidate motion vectors is better as the difference between the reconstructed image of the first decoded area specified by the candidate motion vector and the second decoded reconstructed image is smaller. It is an evaluation result.

[Example of implementation]
FIG. 25 is a block diagram illustrating an implementation example of the encoding device 100 according to each of the above embodiments. The encoding device 100 includes a processing circuit 160 and a memory 162. For example, a plurality of components of the encoding device 100 shown in FIG. 1 are implemented by the processing circuit 160 and the memory 162 shown in FIG.

The processing circuit 160 is a circuit that performs information processing, and is a circuit that can access the memory 162. For example, the processing circuit 160 is a dedicated or general-purpose electronic circuit that encodes a moving image. The processing circuit 160 may be a processor such as a CPU. Further, the processing circuit 160 may be an aggregate of a plurality of electronic circuits. Further, for example, the processing circuit 160 may serve as a plurality of components other than the components for storing information among the plurality of components of the encoding device 100 illustrated in FIG. 1.

The memory 162 is a general purpose or dedicated memory in which information for the processing circuit 160 to encode a moving image is stored. The memory 162 may be an electronic circuit and may be connected to the processing circuit 160. Further, the memory 162 may be included in the processing circuit 160. The memory 162 may be an aggregate of a plurality of electronic circuits. Further, the memory 162 may be a magnetic disk or an optical disk, or may be expressed as a storage or a recording medium. The memory 162 may be a non-volatile memory or a volatile memory.

For example, in the memory 162, a moving image to be encoded may be stored, or a bit string corresponding to the encoded moving image may be stored. The memory 162 may store a program for the processing circuit 160 to encode a moving image.

Also, for example, the memory 162 may serve as a component for storing information among a plurality of components of the encoding device 100 shown in FIG. Specifically, the memory 162 may serve as the block memory 118 and the frame memory 122 shown in FIG. More specifically, the memory 162 may store processed sub-blocks, processed blocks, processed pictures, and the like.

Note that in the encoding device 100, not all of the plurality of components shown in FIG. 1 or the like may be mounted, or all of the plurality of processes described above may not be performed. Some of the plurality of components shown in FIG. 1 and the like may be included in another device, and some of the plurality of processes described above may be executed by another device. In the encoding apparatus 100, a part of the plurality of components shown in FIG. 1 and the like are mounted, and a part of the plurality of processes described above is performed, so that a moving image can be generated with a small code amount. Can be handled appropriately.

FIG. 26 is a block diagram illustrating an implementation example of the decoding device 200 according to each of the above embodiments. The decoding device 200 includes a processing circuit 260 and a memory 262. For example, a plurality of components of the decoding device 200 illustrated in FIG. 10 are implemented by the processing circuit 260 and the memory 262 illustrated in FIG.

The processing circuit 260 is a circuit that performs information processing, and is a circuit that can access the memory 262. For example, the processing circuit 260 is a general-purpose or dedicated electronic circuit that decodes a moving image. The processing circuit 260 may be a processor such as a CPU. Further, the processing circuit 260 may be an aggregate of a plurality of electronic circuits. Further, for example, the processing circuit 260 may serve as a plurality of constituent elements excluding a constituent element for storing information among a plurality of constituent elements of the decoding device 200 illustrated in FIG. 10.

The memory 262 is a general purpose or dedicated memory in which information for the processing circuit 260 to decode a moving image is stored. The memory 262 may be an electronic circuit and may be connected to the processing circuit 260. Further, the memory 262 may be included in the processing circuit 260. The memory 262 may be an aggregate of a plurality of electronic circuits. The memory 262 may be a magnetic disk or an optical disk, or may be expressed as a storage or a recording medium. Further, the memory 262 may be a nonvolatile memory or a volatile memory.

For example, the memory 262 may store a bit sequence corresponding to the encoded moving image, or may store a moving image corresponding to the decoded bit sequence. The memory 262 may store a program for the processing circuit 260 to decode a moving image.

For example, the memory 262 may serve as a component for storing information among a plurality of components of the decoding device 200 illustrated in FIG. Specifically, the memory 262 may serve as the block memory 210 and the frame memory 214 shown in FIG. More specifically, the memory 262 may store processed sub-blocks, processed blocks, processed pictures, and the like.

Note that in the decoding device 200, not all of the plurality of components shown in FIG. 10 and the like may be implemented, or all of the plurality of processes described above may not be performed. Some of the plurality of components shown in FIG. 10 and the like may be included in another device, and some of the plurality of processes described above may be executed by another device. Then, in the decoding device 200, a part of the plurality of components shown in FIG. 10 and the like are mounted, and a part of the plurality of processes described above is performed, so that a moving image is appropriately generated with a small code amount. Can be processed.

[Supplement]
The encoding device 100 and the decoding device 200 in each of the above embodiments may be used as an image encoding device and an image decoding device, or may be used as a moving image encoding device and a moving image decoding device, respectively. . Alternatively, the encoding device 100 and the decoding device 200 can each be used as an inter prediction device. That is, the encoding apparatus 100 and the decoding apparatus 200 may support only the inter prediction unit 126 and the inter prediction unit 218, respectively.

In each of the above embodiments, the prediction block is encoded or decoded as the encoding target block or the decoding target block. However, the encoding target block or the decoding target block is not limited to the prediction block, and may be a sub-block. It may be another block.

Further, in each of the above embodiments, each component may be configured by dedicated hardware or may be realized by executing a software program suitable for each component. Each component may be realized by a program execution unit such as a CPU or a processor reading and executing a software program recorded on a recording medium such as a hard disk or a semiconductor memory.

Specifically, each of the encoding device 100 and the decoding device 200 includes a processing circuit (Processing Circuit) and a storage device (Storage) electrically connected to the processing circuit and accessible from the processing circuit. You may have.

The processing circuit includes at least one of dedicated hardware and a program execution unit, and executes processing using a storage device. Further, when the processing circuit includes a program execution unit, the storage device stores a software program executed by the program execution unit.

Here, the software that realizes the encoding device 100 or the decoding device 200 according to each of the above embodiments is the following program.

That is, this program causes the computer to execute processing according to the flowchart shown in any of FIGS. 15 to 18 and FIGS.

Each component may be a circuit as described above. These circuits may constitute one circuit as a whole, or may be separate circuits. Each component may be realized by a general-purpose processor or a dedicated processor.

Also, another component may execute the process executed by a specific component. In addition, the order in which the processes are executed may be changed, or a plurality of processes may be executed in parallel. Further, the encoding / decoding device may include the encoding device 100 and the decoding device 200.

The first and second ordinal numbers used in the description may be replaced as appropriate. In addition, an ordinal number may be newly given to a component or the like, or may be removed.

As mentioned above, although the aspect of the encoding apparatus 100 and the decoding apparatus 200 was demonstrated based on each embodiment, the aspect of the encoding apparatus 100 and the decoding apparatus 200 is not limited to these embodiment. As long as it does not deviate from the gist of the present disclosure, various modifications conceived by those skilled in the art in the embodiments and forms constructed by combining components in different embodiments are also possible for the encoding apparatus 100 and the decoding apparatus 200. It may be included within the scope of the embodiments.

(Embodiment 5)
In each of the above embodiments, each of the functional blocks can usually be realized by an MPU, a memory, and the like. Further, the processing by each functional block is usually realized by a program execution unit such as a processor reading and executing software (program) recorded on a recording medium such as a ROM. The software may be distributed by downloading or the like, or may be distributed by being recorded on a recording medium such as a semiconductor memory. Naturally, each functional block can be realized by hardware (dedicated circuit).

Further, the processing described in each embodiment may be realized by centralized processing using a single device (system), or may be realized by distributed processing using a plurality of devices. Good. The number of processors that execute the program may be one or more. That is, centralized processing may be performed, or distributed processing may be performed.

The present invention is not limited to the above embodiments, and various modifications are possible, and these are also included in the scope of the present invention.

Furthermore, application examples of the moving picture coding method (picture coding method) or the moving picture decoding method (picture decoding method) shown in the above embodiments and a system using the same will be described. The system includes an image encoding device using an image encoding method, an image decoding device using an image decoding method, and an image encoding / decoding device including both. Other configurations in the system can be appropriately changed according to circumstances.

[Example of use]
FIG. 27 is a diagram illustrating an overall configuration of a content supply system ex100 that implements a content distribution service. The communication service providing area is divided into desired sizes, and base stations ex106, ex107, ex108, ex109, and ex110, which are fixed wireless stations, are installed in each cell.

In the content supply system ex100, devices such as a computer ex111, a game machine ex112, a camera ex113, a home appliance ex114, and a smartphone ex115 via the Internet ex101, the Internet service provider ex102 or the communication network ex104, and the base stations ex106 to ex110. Is connected. The content supply system ex100 may be connected by combining any of the above elements. Each device may be directly or indirectly connected to each other via a telephone network or a short-range wireless communication without using the base stations ex106 to ex110 which are fixed wireless stations. The streaming server ex103 is connected to each device such as a computer ex111, a game machine ex112, a camera ex113, a home appliance ex114, and a smartphone ex115 via the Internet ex101. The streaming server ex103 is connected to a terminal in a hot spot in the airplane ex117 via the satellite ex116.

Note that a wireless access point or a hot spot may be used instead of the base stations ex106 to ex110. Further, the streaming server ex103 may be directly connected to the communication network ex104 without going through the Internet ex101 or the Internet service provider ex102, or may be directly connected to the airplane ex117 without going through the satellite ex116.

The camera ex113 is a device that can shoot still images and moving images such as a digital camera. The smartphone ex115 is a smartphone, a mobile phone, a PHS (Personal Handyphone System), or the like that corresponds to a mobile communication system generally called 2G, 3G, 3.9G, 4G, and 5G in the future.

The home appliance ex118 is a device included in a refrigerator or a household fuel cell cogeneration system.

In the content supply system ex100, a terminal having a photographing function is connected to the streaming server ex103 through the base station ex106 or the like, thereby enabling live distribution or the like. In live distribution, the terminal (computer ex111, game machine ex112, camera ex113, home appliance ex114, smartphone ex115, terminal in airplane ex117, etc.) is used for the still image or video content captured by the user using the terminal. The encoding process described in each embodiment is performed, and the video data obtained by the encoding and the sound data obtained by encoding the sound corresponding to the video are multiplexed, and the obtained data is transmitted to the streaming server ex103. That is, each terminal functions as an image encoding device according to an aspect of the present invention.

On the other hand, the streaming server ex103 streams the content data transmitted to the requested client. The client is a computer or the like in the computer ex111, the game machine ex112, the camera ex113, the home appliance ex114, the smart phone ex115, or the airplane ex117 that can decode the encoded data. Each device that has received the distributed data decrypts and reproduces the received data. That is, each device functions as an image decoding device according to an aspect of the present invention.

[Distributed processing]
The streaming server ex103 may be a plurality of servers or a plurality of computers, and may process, record, and distribute data in a distributed manner. For example, the streaming server ex103 may be realized by a CDN (Contents Delivery Network), and content distribution may be realized by a network connecting a large number of edge servers and edge servers distributed all over the world. In CDN, edge servers that are physically close to each other are dynamically allocated according to clients. Then, the content can be cached and distributed to the edge server, thereby reducing the delay. Also, if some error occurs or the communication status changes due to an increase in traffic, etc., the processing is distributed among multiple edge servers, the distribution subject is switched to another edge server, or the part of the network where the failure has occurred Since detouring can be continued, high-speed and stable distribution can be realized.

In addition to the distributed processing of the distribution itself, the captured data may be encoded at each terminal, may be performed on the server side, or may be shared with each other. As an example, in general, in an encoding process, a processing loop is performed twice. In the first loop, the complexity of the image or the code amount in units of frames or scenes is detected. In the second loop, processing for maintaining the image quality and improving the coding efficiency is performed. For example, the terminal performs the first encoding process, and the server receiving the content performs the second encoding process, thereby improving the quality and efficiency of the content while reducing the processing load on each terminal. it can. In this case, if there is a request to receive and decode in almost real time, the encoded data of the first time performed by the terminal can be received and reproduced by another terminal, enabling more flexible real-time distribution. Become.

As another example, the camera ex113 or the like extracts a feature amount from an image, compresses data relating to the feature amount as metadata, and transmits the metadata to the server. The server performs compression according to the meaning of the image, for example, by determining the importance of the object from the feature amount and switching the quantization accuracy. The feature data is particularly effective for improving the accuracy and efficiency of motion vector prediction at the time of re-compression on the server. Also, simple coding such as VLC (variable length coding) may be performed at the terminal, and coding with a large processing load such as CABAC (context adaptive binary arithmetic coding) may be performed at the server.

As yet another example, in a stadium, a shopping mall, a factory, or the like, there may be a plurality of video data in which almost the same scene is captured by a plurality of terminals. In this case, for example, a GOP (Group of Picture) unit, a picture unit, or a tile obtained by dividing a picture using a plurality of terminals that have performed shooting and other terminals and servers that have not performed shooting as necessary. Distributed processing is performed by assigning encoding processing in units or the like. Thereby, delay can be reduced and real-time property can be realized.

In addition, since the plurality of video data are almost the same scene, the server may manage and / or instruct the video data captured by each terminal to refer to each other. Alternatively, the encoded data from each terminal may be received by the server and the reference relationship may be changed among a plurality of data, or the picture itself may be corrected or replaced to be encoded again. This makes it possible to generate a stream with improved quality and efficiency of each piece of data.

Also, the server may distribute the video data after performing transcoding to change the encoding method of the video data. For example, the server may convert the MPEG encoding system to the VP encoding. H.264 in H.264. It may be converted into H.265.

Thus, the encoding process can be performed by a terminal or one or more servers. Therefore, in the following, description such as “server” or “terminal” is used as the subject performing processing, but part or all of processing performed by the server may be performed by the terminal, or processing performed by the terminal may be performed. Some or all may be performed at the server. The same applies to the decoding process.

[3D, multi-angle]
In recent years, different scenes photographed by terminals such as a plurality of cameras ex113 and / or smartphones ex115 that are substantially synchronized with each other, or images or videos obtained by photographing the same scene from different angles have been increasingly used. Yes. The video captured by each terminal is integrated based on the relative positional relationship between the terminals acquired separately or the region where the feature points included in the video match.

The server not only encodes a two-dimensional moving image, but also encodes a still image automatically based on a scene analysis of the moving image or at a time specified by the user and transmits it to the receiving terminal. Also good. In addition, when the server can acquire the relative positional relationship between the photographing terminals, the server obtains the three-dimensional shape of the scene based on not only the two-dimensional moving image but also the video obtained by photographing the same scene from different angles. Can be generated. The server may separately encode the three-dimensional data generated by the point cloud or the like, and the video to be transmitted to the receiving terminal based on the result of recognizing or tracking the person or the object using the three-dimensional data. Alternatively, the images may be selected or reconstructed from videos captured by a plurality of terminals.

In this way, the user can arbitrarily select each video corresponding to each photographing terminal and enjoy a scene, or can display a video of an arbitrary viewpoint from three-dimensional data reconstructed using a plurality of images or videos. You can also enjoy the clipped content. Furthermore, as with video, sound is collected from a plurality of different angles, and the server may multiplex and transmit sound from a specific angle or space according to the video.

Also, in recent years, content that associates the real world with the virtual world, such as Virtual Reality (VR) and Augmented Reality (AR), has become widespread. In the case of a VR image, the server may create viewpoint images for the right eye and the left eye, respectively, and perform encoding that allows reference between each viewpoint video by Multi-View Coding (MVC) or the like. You may encode as another stream, without referring. At the time of decoding another stream, it is preferable to reproduce in synchronization with each other so that a virtual three-dimensional space is reproduced according to the viewpoint of the user.

In the case of an AR image, the server superimposes virtual object information in the virtual space on the camera information in the real space based on the three-dimensional position or the movement of the user's viewpoint. The decoding device may acquire or hold virtual object information and three-dimensional data, generate a two-dimensional image according to the movement of the user's viewpoint, and create superimposition data by connecting them smoothly. Alternatively, the decoding device transmits the movement of the user's viewpoint to the server in addition to the request for the virtual object information, and the server creates superimposition data according to the movement of the viewpoint received from the three-dimensional data held in the server, The superimposed data may be encoded and distributed to the decoding device. Note that the superimposed data has an α value indicating transparency in addition to RGB, and the server sets the α value of a portion other than the object created from the three-dimensional data to 0 or the like, and the portion is transparent. May be encoded. Alternatively, the server may generate data in which a RGB value of a predetermined value is set as the background, such as a chroma key, and the portion other than the object is set to the background color.

Similarly, the decryption processing of the distributed data may be performed at each terminal as a client, may be performed on the server side, or may be performed in a shared manner. As an example, a terminal may once send a reception request to the server, receive content corresponding to the request at another terminal, perform a decoding process, and transmit a decoded signal to a device having a display. Regardless of the performance of the communicable terminal itself, it is possible to reproduce data with good image quality by distributing processing and selecting appropriate content. As another example, a part of a region such as a tile in which a picture is divided may be decoded and displayed on a viewer's personal terminal while receiving large-size image data on a TV or the like. Accordingly, it is possible to confirm at hand the area in which the person is responsible or the area to be confirmed in more detail while sharing the whole image.

In the future, in the situation where multiple short-distance, medium-distance, or long-distance wireless communications can be used regardless of whether indoors or outdoors, using a distribution system standard such as MPEG-DASH, It is expected that content is received seamlessly while switching appropriate data. Accordingly, the user can switch in real time while freely selecting a decoding device or a display device such as a display installed indoors or outdoors as well as his / her own terminal. Also, decoding can be performed while switching between a terminal to be decoded and a terminal to be displayed based on its own position information. This makes it possible to move while displaying map information on the wall surface of a neighboring building or a part of the ground in which a displayable device is embedded while moving to the destination. Also, access to encoded data on the network, such as when the encoded data is cached in a server that can be accessed from the receiving terminal in a short time, or copied to the edge server in the content delivery service. It is also possible to switch the bit rate of received data based on ease.

[Scalable coding]
The content switching will be described using a scalable stream that is compression-encoded by applying the moving image encoding method shown in each of the above embodiments shown in FIG. The server may have a plurality of streams of the same content and different quality as individual streams, but the temporal / spatial scalable implementation realized by dividing into layers as shown in the figure. The configuration may be such that the content is switched by utilizing the characteristics of the stream. In other words, the decoding side decides which layer to decode according to internal factors such as performance and external factors such as the state of communication bandwidth, so that the decoding side can combine low-resolution content and high-resolution content. You can switch freely and decrypt. For example, when the user wants to continue watching the video that was viewed on the smartphone ex115 while moving on a device such as an Internet TV after returning home, the device only has to decode the same stream to a different layer, so the load on the server side Can be reduced.

Further, as described above, the enhancement layer includes meta information based on image statistical information, etc., in addition to the configuration in which the picture is encoded for each layer and the enhancement layer exists above the base layer. The decoding side may generate content with high image quality by super-resolution of the base layer picture based on the meta information. Super-resolution may be either improvement of the SN ratio at the same resolution or enlargement of the resolution. The meta information includes information for specifying a linear or non-linear filter coefficient used for super-resolution processing, or information for specifying a parameter value in filter processing, machine learning, or least square calculation used for super-resolution processing. .

Alternatively, the picture may be divided into tiles or the like according to the meaning of the object in the image, and the decoding side may select only a part of the region by selecting the tile to be decoded. Also, by storing the object attributes (person, car, ball, etc.) and the position in the video (coordinate position in the same image, etc.) as meta information, the decoding side can determine the position of the desired object based on the meta information. Can be identified and the tile containing the object can be determined. For example, as shown in FIG. 29, the meta information is stored using a data storage structure different from the pixel data such as the SEI message in HEVC. This meta information indicates, for example, the position, size, or color of the main object.

Also, meta information may be stored in units composed of a plurality of pictures, such as streams, sequences, or random access units. Thereby, the decoding side can acquire the time when the specific person appears in the video, etc., and can match the picture in which the object exists and the position of the object in the picture by combining with the information in units of pictures.

[Web page optimization]
FIG. 30 shows an example of a web page display screen on the computer ex111 or the like. FIG. 31 is a diagram illustrating an example of a display screen of a web page on the smartphone ex115 or the like. As shown in FIGS. 30 and 31, the web page may include a plurality of link images that are links to image content, and the appearance differs depending on the browsing device. When a plurality of link images are visible on the screen, the display device until the user explicitly selects the link image, or until the link image approaches the center of the screen or the entire link image enters the screen. The (decoding device) displays a still image or an I picture included in each content as a link image, displays a video like a gif animation with a plurality of still images or I pictures, or receives only a base layer to receive a video. Are decoded and displayed.

When the link image is selected by the user, the display device decodes the base layer with the highest priority. If there is information indicating that the HTML constituting the web page is scalable content, the display device may decode up to the enhancement layer. Also, in order to ensure real-time properties, the display device only decodes forward reference pictures (I picture, P picture, forward reference only B picture) before being selected or when the communication band is very strict. In addition, the delay between the decoding time of the first picture and the display time (delay from the start of content decoding to the start of display) can be reduced by displaying. Further, the display device may intentionally ignore the reference relationship of pictures and roughly decode all B pictures and P pictures with forward reference, and perform normal decoding as the number of received pictures increases over time.

[Automatic driving]
In addition, when transmitting and receiving still image or video data such as two-dimensional or three-dimensional map information for automatic driving or driving support of a car, the receiving terminal adds meta data to image data belonging to one or more layers. Weather or construction information may also be received and decoded in association with each other. The meta information may belong to a layer or may be simply multiplexed with image data.

In this case, since the car, drone, airplane, or the like including the receiving terminal moves, the receiving terminal transmits the position information of the receiving terminal at the time of the reception request, thereby seamless reception and decoding while switching the base stations ex106 to ex110. Can be realized. In addition, the receiving terminal can dynamically switch how much meta-information is received or how much map information is updated according to the user's selection, the user's situation, or the communication band state. become.

As described above, in the content supply system ex100, the encoded information transmitted by the user can be received, decoded and reproduced in real time by the client.

[Distribution of personal contents]
Further, the content supply system ex100 can perform not only high-quality and long-time content by a video distributor but also unicast or multicast distribution of low-quality and short-time content by an individual. Moreover, such personal contents are expected to increase in the future. In order to make personal content superior, the server may perform the encoding process after performing the editing process. This can be realized, for example, with the following configuration.

After shooting, the server performs recognition processing such as shooting error, scene search, semantic analysis, and object detection from the original image or encoded data. Then, the server manually or automatically corrects out-of-focus or camera shake based on the recognition result, or selects a less important scene such as a scene whose brightness is lower than that of other pictures or is out of focus. Edit such as deleting, emphasizing the edge of an object, and changing the hue. The server encodes the edited data based on the editing result. It is also known that if the shooting time is too long, the audience rating will decrease, and the server will move not only in the less important scenes as described above, but also in motion according to the shooting time. A scene with few images may be automatically clipped based on the image processing result. Alternatively, the server may generate and encode a digest based on the result of the semantic analysis of the scene.

In some cases, personal content may include infringements such as copyrights, author's personality rights, or portrait rights, which are inconvenient for individuals, such as exceeding the intended scope of sharing. In some cases. Therefore, for example, the server may change and encode the face of the person in the periphery of the screen or the inside of the house into an unfocused image. In addition, the server recognizes whether or not a face of a person different from the person registered in advance is shown in the encoding target image, and if so, performs processing such as applying a mosaic to the face part. May be. Alternatively, as a pre-processing or post-processing of encoding, the user designates a person or background area that the user wants to process an image from the viewpoint of copyright, etc., and the server replaces the designated area with another video or blurs the focus. It is also possible to perform such processing. If it is a person, the face image can be replaced while tracking the person in the moving image.

In addition, since viewing of personal content with a small amount of data is strongly demanded for real-time performance, the decoding device first receives the base layer with the highest priority and performs decoding and reproduction, depending on the bandwidth. The decoding device may receive the enhancement layer during this time, and may play back high-quality video including the enhancement layer when played back twice or more, such as when playback is looped. A stream that is scalable in this way can provide an experience in which the stream becomes smarter and the image is improved gradually, although it is a rough moving picture when it is not selected or at the beginning of viewing. In addition to scalable coding, the same experience can be provided even if the coarse stream played back the first time and the second stream coded with reference to the first video are configured as one stream. .

[Other usage examples]
In addition, these encoding or decoding processes are generally processed in the LSI ex500 included in each terminal. The LSI ex500 may be configured as a single chip or a plurality of chips. Note that moving image encoding or decoding software is incorporated into some recording medium (CD-ROM, flexible disk, hard disk, etc.) that can be read by the computer ex111 and the like, and encoding or decoding processing is performed using the software. Also good. Furthermore, when the smartphone ex115 has a camera, moving image data acquired by the camera may be transmitted. The moving image data at this time is data encoded by the LSI ex500 included in the smartphone ex115.

Note that the LSI ex500 may be configured to download and activate application software. In this case, the terminal first determines whether the terminal is compatible with the content encoding method or has a specific service execution capability. If the terminal does not support the content encoding method or does not have the capability to execute a specific service, the terminal downloads a codec or application software, and then acquires and reproduces the content.

Further, not only the content supply system ex100 via the Internet ex101, but also a digital broadcasting system, at least the moving image encoding device (image encoding device) or the moving image decoding device (image decoding device) of the above embodiments. Any of these can be incorporated. The difference is that the unicasting of the content supply system ex100 is suitable for multicasting because it uses a satellite or the like to transmit and receive multiplexed data in which video and sound are multiplexed on broadcasting radio waves. However, the same application is possible for the encoding process and the decoding process.

[Hardware configuration]
FIG. 32 is a diagram illustrating the smartphone ex115. FIG. 33 is a diagram illustrating a configuration example of the smartphone ex115. The smartphone ex115 receives the antenna ex450 for transmitting / receiving radio waves to / from the base station ex110, the camera unit ex465 capable of taking video and still images, the video captured by the camera unit ex465, and the antenna ex450. A display unit ex458 for displaying data obtained by decoding the video or the like. The smartphone ex115 further includes an operation unit ex466 that is a touch panel or the like, a voice output unit ex457 that is a speaker or the like for outputting voice or sound, a voice input unit ex456 that is a microphone or the like for inputting voice, and photographing. Memory unit ex467 that can store encoded video or still image, recorded audio, received video or still image, encoded data such as mail, or decoded data, and a user, and network A slot part ex464, which is an interface part with the SIMex 468 for authenticating access to various data. An external memory may be used instead of the memory unit ex467.

Also, a main control unit ex460 that comprehensively controls the display unit ex458, the operation unit ex466, and the like, a power supply circuit unit ex461, an operation input control unit ex462, a video signal processing unit ex455, a camera interface unit ex463, a display control unit ex459, a modulation / Demodulation unit ex452, multiplexing / demultiplexing unit ex453, audio signal processing unit ex454, slot unit ex464, and memory unit ex467 are connected via bus ex470.

When the power key is turned on by a user operation, the power supply circuit unit ex461 starts up the smartphone ex115 in an operable state by supplying power from the battery pack to each unit.

The smartphone ex115 performs processing such as calling and data communication based on the control of the main control unit ex460 having a CPU, a ROM, a RAM, and the like. During a call, the voice signal picked up by the voice input unit ex456 is converted into a digital voice signal by the voice signal processing unit ex454, spread spectrum processed by the modulation / demodulation unit ex452, and digital / analog converted by the transmission / reception unit ex451. After performing the processing and the frequency conversion processing, the data is transmitted via the antenna ex450. Further, the received data is amplified and subjected to frequency conversion processing and analog-digital conversion processing, spectrum despreading processing is performed by the modulation / demodulation unit ex452, and converted to analog audio signal by the audio signal processing unit ex454, and then this is output to the audio output unit ex457. Output from. In the data communication mode, text, still image, or video data is sent to the main control unit ex460 via the operation input control unit ex462 by the operation of the operation unit ex466 of the main body unit, and transmission / reception processing is performed similarly. When transmitting video, still image, or video and audio in the data communication mode, the video signal processing unit ex455 uses the video signal stored in the memory unit ex467 or the video signal input from the camera unit ex465 as described above. The video data is compressed and encoded by the moving image encoding method shown in the form, and the encoded video data is sent to the multiplexing / demultiplexing unit ex453. The audio signal processing unit ex454 encodes the audio signal picked up by the audio input unit ex456 while the camera unit ex465 captures a video or a still image, and sends the encoded audio data to the multiplexing / separating unit ex453. To do. The multiplexing / demultiplexing unit ex453 multiplexes the encoded video data and the encoded audio data by a predetermined method, and the modulation / demodulation unit (modulation / demodulation circuit unit) ex452 and the modulation / demodulation unit ex451 perform modulation processing and conversion. The data is processed and transmitted via the antenna ex450.

In order to decode the multiplexed data received via the antenna ex450 when the video attached to the e-mail or chat, or the video linked to the web page or the like is received, the multiplexing / demultiplexing unit ex453 performs multiplexing By separating the data, the multiplexed data is divided into a bit stream of video data and a bit stream of audio data, and the encoded video data is supplied to the video signal processing unit ex455 via the synchronization bus ex470. The converted audio data is supplied to the audio signal processing unit ex454. The video signal processing unit ex455 decodes the video signal by the video decoding method corresponding to the video encoding method shown in each of the above embodiments, and is linked from the display unit ex458 via the display control unit ex459. A video or still image included in the moving image file is displayed. The audio signal processing unit ex454 decodes the audio signal, and the audio is output from the audio output unit ex457. Since real-time streaming is widespread, depending on the user's situation, there may be occasions where audio playback is not socially appropriate. Therefore, it is desirable that the initial value is a configuration in which only the video data is reproduced without reproducing the audio signal. Audio may be synchronized and played back only when the user performs an operation such as clicking on video data.

In addition, although the smartphone ex115 has been described here as an example, in addition to a transmission / reception terminal having both an encoder and a decoder as a terminal, a transmission terminal having only an encoder and a reception having only a decoder There are three possible mounting formats: terminals. Furthermore, in the digital broadcasting system, it has been described as receiving or transmitting multiplexed data in which audio data or the like is multiplexed with video data. However, multiplexed data includes character data related to video in addition to audio data. Multiplexing may be performed, and video data itself may be received or transmitted instead of multiplexed data.

In addition, although it has been described that the main control unit ex460 including the CPU controls the encoding or decoding process, the terminal often includes a GPU. Therefore, a configuration may be adopted in which a wide area is processed in a lump by utilizing the performance of the GPU by using a memory shared by the CPU and the GPU or a memory whose addresses are managed so as to be used in common. As a result, the encoding time can be shortened, real-time performance can be ensured, and low delay can be realized. In particular, it is efficient to perform motion search, deblocking filter, SAO (Sample Adaptive Offset), and transformation / quantization processing in batches in units of pictures or the like instead of the CPU.

The present disclosure can be used for, for example, a television receiver, a digital video recorder, a car navigation, a mobile phone, a digital camera, a digital video camera, a video conference system, or an electronic mirror.

DESCRIPTION OF SYMBOLS 100 Coding apparatus 102 Division | segmentation part 104 Subtraction part 106 Conversion part 108 Quantization part 110 Entropy encoding part 112,204 Inverse quantization part 114,206 Inverse conversion part 116,208 Adder 118,210 Block memory 120,212

Loop filter Units

122 and 214

Frame memories

124 and 216

Intra prediction units

126 and 218

Inter prediction units

128 and 220

Prediction control units

160 and 260

Processing circuits

162 and 262 Memory 200 Decoding device 202 Entropy decoding unit

Claims

An encoding device for encoding a moving image,
A processing circuit; and a memory connected to the processing circuit;
The processing circuit uses the memory,
Obtaining a plurality of candidate motion vectors based on respective motion vectors of a plurality of encoded blocks corresponding to the encoding target block in the moving image;
Extracting at least one prediction motion vector candidate of the encoding target block from the plurality of candidate motion vectors;
Deriving a motion vector of the encoding target block with reference to a reference picture included in the moving image,
Encoding the difference between the predicted motion vector of the extracted at least one predicted motion vector candidate and the derived motion vector of the encoding target block;
Performing motion compensation on the encoding target block using the derived motion vector of the encoding target block;
In extracting the at least one predicted motion vector candidate,
Evaluation of each of the plurality of candidate motion vectors using all of the at least one predicted motion vector candidate using the reconstructed image of the encoded region in the moving image without using the image region of the encoding target block An encoding device that extracts based on the result.
The processing circuit includes:
In extracting the at least one predicted motion vector candidate,
Extracting all of the at least one predicted motion vector candidate by selecting only one candidate motion vector having the best evaluation result from the plurality of candidate motion vectors;
In the encoding of the difference,
Encoding the difference between the predicted motion vector that is the selected candidate motion vector and the derived motion vector of the encoding target block;
The encoding device according to claim 1.
The processing circuit includes:
In extracting the at least one predicted motion vector candidate,
N candidate motion vectors (N is an integer of 2 or more) are extracted as all of the at least one predicted motion vector candidate from the plurality of candidate motion vectors based on the evaluation result;
The processing circuit further includes:
Selecting the predicted motion vector from the extracted N predicted motion vector candidates;
Encoding selection information for identifying the selected predicted motion vector;
In the encoding of the difference,
Encoding a difference between the selected predicted motion vector and the derived motion vector of the encoding target block;
The encoding device according to claim 1.
The processing circuit includes:
In extracting the at least one predicted motion vector candidate,
The encoding apparatus according to claim 3, wherein the top N candidate motion vectors are extracted as all of the at least one predicted motion vector candidate from the plurality of candidate motion vectors in order of good evaluation results.
The processing circuit includes:
In extracting the at least one predicted motion vector candidate,
Classifying the plurality of candidate motion vectors into N groups;
The encoding according to claim 3, wherein, from each of the N groups, one candidate motion vector having the best evaluation result in the group is extracted to extract all of the at least one predicted motion vector candidate. apparatus.
The processing circuit includes:
In extracting the at least one predicted motion vector candidate,
Classifying the plurality of candidate motion vectors into M groups (M is an integer greater than N);
From each of the M groups, one candidate motion vector having the best evaluation result in the group is selected as a representative candidate motion vector,
4. The code according to claim 3, wherein, from the selected M representative candidate motion vectors, the top N representative candidate motion vectors are extracted as all of the at least one predicted motion vector candidate in order of good evaluation results. Device.
The evaluation result of each of the plurality of candidate motion vectors is better as the difference between the reconstructed image of the first encoded region specified by the candidate motion vector and the second encoded reconstructed image is smaller. The encoding apparatus according to any one of claims 1 to 6, which is an evaluation result.
A decoding device for decoding an encoded moving image,
A processing circuit; and a memory connected to the processing circuit;
The processing circuit uses the memory,
Obtaining a plurality of candidate motion vectors based on respective motion vectors of a plurality of decoded blocks corresponding to a decoding target block in the moving image;
Extracting at least one predicted motion vector candidate of the decoding target block from the plurality of candidate motion vectors;
Decoding difference information indicating the difference between two motion vectors;
Deriving a motion vector of the decoding target block by adding a prediction motion vector of the extracted at least one prediction motion vector candidate to the difference indicated by the decoded difference information;
Performing motion compensation on the decoding target block using the derived motion vector of the decoding target block;
In extracting the at least one predicted motion vector candidate,
All of the at least one predicted motion vector candidate is used as an evaluation result of each of the plurality of candidate motion vectors using a reconstructed image of a decoded region in the moving image without using an image region of the decoding target block. Decoding device that extracts based on.
The processing circuit includes:
In extracting the at least one predicted motion vector candidate,
Extracting all of the at least one predicted motion vector candidate by selecting only one candidate motion vector having the best evaluation result from the plurality of candidate motion vectors;
In deriving the motion vector of the block to be decoded,
The decoding device according to claim 8, wherein the motion vector of the decoding target block is derived by adding the predicted motion vector, which is the selected candidate motion vector, to the difference indicated by the decoded difference information.
The processing circuit includes:
In extracting the at least one predicted motion vector candidate,
N candidate motion vectors (N is an integer of 2 or more) are extracted as all of the at least one predicted motion vector candidate from the plurality of candidate motion vectors based on the evaluation result;
The processing circuit further includes:
Decoding selection information for identifying the predicted motion vector;
Selecting a predicted motion vector candidate identified by the decoded selection information as the predicted motion vector from the extracted N predicted motion vector candidates;
In deriving the motion vector of the block to be decoded,
The decoding device according to claim 8, wherein a motion vector of the decoding target block is derived by adding the selected prediction motion vector to a difference indicated by the decoded difference information.
The processing circuit includes:
In extracting the at least one predicted motion vector candidate,
The decoding apparatus according to claim 10, wherein the top N candidate motion vectors are extracted as all of the at least one predicted motion vector candidate from the plurality of candidate motion vectors in order of good evaluation results.
The processing circuit includes:
In extracting the at least one predicted motion vector candidate,
Classifying the plurality of candidate motion vectors into N groups;
The decoding device according to claim 10, wherein, from each of the N groups, one candidate motion vector having the best evaluation result in the group is extracted to extract all of the at least one predicted motion vector candidate. .
The processing circuit includes:
In extracting the at least one predicted motion vector candidate,
Classifying the plurality of candidate motion vectors into M groups (M is an integer greater than N);
From each of the M groups, one candidate motion vector having the best evaluation result in the group is selected as a representative candidate motion vector,
The decoding according to claim 10, wherein, from the selected M representative candidate motion vectors, the top N representative candidate motion vectors are extracted as all of the at least one predicted motion vector candidate in the order of good evaluation results. apparatus.
The evaluation result of each of the plurality of candidate motion vectors is such that the smaller the difference between the reconstructed image of the first decoded area specified by the candidate motion vector and the second decoded reconstructed image, the better the evaluation result. The decoding device according to any one of claims 8 to 13.
An encoding method for encoding a moving image, comprising:
Obtaining a plurality of candidate motion vectors based on respective motion vectors of a plurality of encoded blocks corresponding to the encoding target block in the moving image;
Extracting at least one prediction motion vector candidate of the encoding target block from the plurality of candidate motion vectors;
Deriving a motion vector of the encoding target block with reference to a reference picture included in the moving image,
Encoding the difference between the predicted motion vector of the extracted at least one predicted motion vector candidate and the derived motion vector of the encoding target block;
Performing motion compensation on the encoding target block using the derived motion vector of the encoding target block;
In extracting the at least one predicted motion vector candidate,
Evaluation of each of the plurality of candidate motion vectors using all of the at least one predicted motion vector candidate using the reconstructed image of the encoded region in the moving image without using the image region of the encoding target block An encoding method that extracts based on the result.
A decoding method for decoding an encoded moving image, comprising:
Obtaining a plurality of candidate motion vectors based on respective motion vectors of a plurality of decoded blocks corresponding to a decoding target block in the moving image;
Extracting at least one predicted motion vector candidate of the decoding target block from the plurality of candidate motion vectors;
Decoding difference information indicating the difference between two motion vectors;
Deriving a motion vector of the decoding target block by adding a prediction motion vector of the extracted at least one prediction motion vector candidate to the difference indicated by the decoded difference information;
Performing motion compensation on the decoding target block using the derived motion vector of the decoding target block;
In extracting the at least one predicted motion vector candidate,
All of the at least one predicted motion vector candidate is used as an evaluation result of each of the plurality of candidate motion vectors using a reconstructed image of a decoded region in the moving image without using an image region of the decoding target block. Decoding method based on extraction.