KR20230067502A - Video decoding method, video decoding apparatus, video encoding method, and video encoding apparatus - Google Patents

Abstract

A plurality of steps for obtaining information on a first motion vector of the current block from a bitstream, determining the first motion vector based on the information on the first motion vector, and determining a second motion vector of the current block. Determines a candidate list including candidate motion vectors of <RTI ID=0.0>and assigns one of the plurality of candidate motion vectors to the second motion based on the distance between the plurality of candidate motion vectors and the first motion vector. vector, and using the first motion vector and the second motion vector to determine the motion vector of the current block, a video decoding method and apparatus are proposed.

Description

Video decoding method, video decoding apparatus, video encoding method, and video encoding apparatus

The present disclosure relates to a video decoding method and apparatus, and a video encoding method and apparatus, and more specifically, in bi-directional prediction, a first motion vector among two pieces of motion vector information (first motion vector information and second motion vector information). Video decoding in which related information is transmitted only for information, and second motion vector information is derived using a first motion vector obtained using the transmitted first motion vector information and a motion vector candidate list for the second motion vector; and It relates to an encoding method and apparatus.

In codecs such as H.264 Advanced Video Coding (AVC), High Efficiency Video Coding (HEVC), and Versatile Video Coding (VVC), an image is divided into blocks and inter prediction or intra prediction is performed. Each block is predicted-encoded and predicted-decoded through.

Intra prediction is a method of compressing an image by removing spatial redundancy in images, and inter prediction is a method of compressing an image by removing temporal redundancy between images.

In the encoding process, a prediction block is generated through intra prediction or inter prediction, a residual block is generated by subtracting the prediction block from a current block, and residual samples of the residual block are transformed and quantized.

In the decoding process, residual samples of the residual block are generated by inverse quantization and inverse transformation of quantized transform coefficients of the residual block, and a prediction block generated through intra prediction or inter prediction is combined with the residual block to reconstruct the current block. The restored current block may be output after being processed according to one or more filtering algorithms.

In codecs such as H.264 AVC, HEVC, and VVC, in the case of bi-directional prediction, motion information for both directions is transmitted and motion estimation is performed twice to find two reference blocks.

Since the resolution of video transmitted in real time tends to increase to 4K, 8K UHD or higher, a bi-directional prediction method advantageous to real-time transmission may be required in terms of complexity.

A video decoding method according to an embodiment includes obtaining information on a first motion vector of a current block from a bitstream; determining the first motion vector based on information on the first motion vector; determining a candidate list including a plurality of candidate predictive motion vectors for determining a second motion vector of the current block; determining one of the plurality of candidate motion vectors as the second motion vector based on a distance between the plurality of candidate motion vectors and the first motion vector; and determining a motion vector of the current block using the first motion vector and the second motion vector.

A video decoding apparatus according to an embodiment includes a memory for storing one or more instructions; and at least one processor operating according to the one or more instructions. At least one processor may obtain information about a first motion vector of a current block from a bitstream. At least one processor may determine the first motion vector based on information on the first motion vector. At least one processor may determine a candidate list including a plurality of candidate prediction motion vectors for determining the second motion vector of the current block. At least one processor may determine one of the plurality of candidate motion vectors as the second motion vector based on a distance between the plurality of candidate motion vectors and the first motion vector. At least one processor may determine the motion vector of the current block using the first motion vector and the second motion vector.

A video encoding method according to an embodiment includes generating information about a first motion vector of a current block; determining the first motion vector based on information on the first motion vector; determining a candidate list including a plurality of candidate predictive motion vectors for determining a second motion vector of the current block; determining one of the plurality of candidate motion vectors as the second motion vector based on a distance between the plurality of candidate motion vectors and the first motion vector; and determining a motion vector of the current block by using the first motion vector and the second motion vector, and encoding information on the first motion vector.

A video encoding apparatus based on AI according to an embodiment includes a memory for storing one or more instructions; and at least one processor operating according to the one or more instructions. At least one processor may generate information about the first motion vector of the current block. At least one processor may determine the first motion vector based on information on the first motion vector. At least one processor may determine a candidate list including a plurality of candidate prediction motion vectors for determining the second motion vector of the current block. At least one processor may determine one of the plurality of candidate motion vectors as the second motion vector based on a distance between the plurality of candidate motion vectors and the first motion vector. At least one processor may determine a motion vector of the current block using the first motion vector and the second motion vector, and encode information on the first motion vector.

1 is a diagram illustrating a process of encoding and decoding an image.
2 is a diagram illustrating blocks divided from an image according to a tree structure.
3A is a diagram for explaining an example of a reference block in bi-directional prediction.
3B is a diagram for explaining an example of a reference block in bidirectional prediction.
3C is a diagram for explaining an example of a reference block in bidirectional prediction.
3D is a diagram for explaining an example of a reference block in bi-directional prediction.
3E is a diagram for explaining a method of obtaining a current block by using a reference block in bi-directional prediction.
4A is a diagram for explaining a method of obtaining a first reference block of a current block through motion estimation according to an embodiment of the present disclosure.
4B is a diagram for explaining a method of deriving a second reference block using a first reference block obtained through motion estimation and candidate blocks of a current block according to an embodiment of the present disclosure.
5A is a diagram for explaining a method of deriving a second reference block using a first reference block obtained through motion estimation and candidate blocks of a current block according to an embodiment of the present disclosure.
5B is a diagram for explaining a method of deriving a second reference block using a first reference block obtained through motion estimation and candidate blocks of a current block according to an embodiment of the present disclosure.
6 is a flowchart of a video decoding method according to an embodiment of the present disclosure.
7 is a diagram showing the configuration of a video decoding apparatus according to an embodiment of the present disclosure.
8 is a flowchart of a video encoding method according to an embodiment of the present disclosure.
9 is a diagram showing the configuration of a video encoding apparatus according to an embodiment of the present disclosure.

In this disclosure, the expression “at least one of a, b, or c” means “a”, “b”, “c”, “a and b”, “a and c”, “b and c”, “a, b” and c”, or variations thereof.

Since the present disclosure may have various changes and various embodiments, specific embodiments are illustrated in the drawings, and will be described in detail through detailed description. However, this is not intended to limit the embodiments of the present disclosure, and it should be understood that the present disclosure includes all modifications, equivalents, and substitutes included in the spirit and scope of the various embodiments.

In describing the embodiments, if it is determined that a detailed description of a related known technology may unnecessarily obscure the subject matter of the present disclosure, the detailed description will be omitted. In addition, numbers (eg, 1st, 2nd, etc.) used in the description process of the specification are only identifiers for distinguishing one component from another.

In addition, in the present disclosure, when one component is referred to as "connected" or "connected" to another component, the one component may be directly connected or directly connected to the other component, but in particular Unless otherwise described, it should be understood that they may be connected or connected via another component in the middle.

In addition, in the present disclosure, components expressed as '~ unit (unit)', 'module', etc., are two or more components combined into one component, or one component is divided into two or more for each function. may be differentiated into In addition, each of the components to be described below may additionally perform some or all of the functions of other components in addition to its own main function, and some of the main functions of each component may be different from other components. Of course, it may be performed exclusively by a component.

Also, in the present disclosure, an 'image or picture' may mean a still image (or frame), a motion picture composed of a plurality of continuous still images, or a video.

Also, in the present disclosure, 'current block' means a block that is currently a processing target. The current block may be a slice, tile, maximum coding unit, coding unit, prediction unit, or transformation unit divided from the current image.

In addition, in the present disclosure, 'sample' refers to data to be processed as data allocated to a sampling position within data such as an image or a block. For example, a sample may include a pixel in a 2D image.

Prior to describing an image decoding method, an image decoding apparatus, an image encoding method, and an image encoding apparatus according to an exemplary embodiment, video encoding and decoding processes will be described with reference to FIGS. 1 and 2 .

1 is a diagram illustrating a process of encoding and decoding an image.

The encoding device 110 transmits the bitstream generated through encoding of the image to the decoding device 150, and the decoding device 150 receives and decodes the bitstream to restore the image.

Specifically, in the encoding device 110, the prediction encoder 115 outputs a prediction block through inter prediction and intra prediction, and the transform and quantization unit 120 outputs a residual sample of a residual block between the prediction block and the current block. are transformed and quantized to output quantized transform coefficients. The entropy encoding unit 125 encodes the quantized transform coefficient and outputs it as a bitstream.

The quantized transform coefficient is reconstructed into a residual block including residual samples in the spatial domain through the inverse quantization and inverse transform unit 130 . The reconstructed block obtained by adding the prediction block and the residual block is output as a filtered block after passing through the deblocking filtering unit 135 and the loop filtering unit 140. The reconstructed image including the filtered block may be used as a reference image for the next input image in the predictive encoder 115 .

The bitstream received by the decoding device 150 is reconstructed into a residual block including residual samples in the spatial domain through an entropy decoding unit 155 and an inverse quantization and inverse transform unit 160 . A reconstructed block is generated by combining the prediction block output from the prediction decoding unit 175 and the residual block, and the reconstructed block is output as a filtered block through the deblocking filtering unit 165 and the loop filtering unit 170. The reconstructed image including the filtered block may be used as a reference image for the next image in the prediction decoding unit 175.

The loop filtering unit 140 of the encoding device 110 performs loop filtering using filter information input according to user input or system settings. The filter information used by the loop filtering unit 140 is transmitted to the decoding apparatus 150 through the entropy encoding unit 125 . The loop filtering unit 170 of the decoding device 150 may perform loop filtering based on filter information input from the entropy decoding unit 155 .

In the process of encoding and decoding an image, an image is hierarchically segmented, and encoding and decoding are performed on blocks divided from the image. A block segmented from an image will be described with reference to FIG. 2 .

FIG. 2 is a diagram showing blocks divided from the image 200 according to a tree structure.

One image 200 may be divided into one or more slices or one or more tiles. One slice may include a plurality of tiles.

One slice or one tile may be a sequence of one or more Maximum Coding Units (Maximum CUs).

One maximum coding unit may be divided into one or more coding units. A coding unit may be a reference block for determining a prediction mode. In other words, it may be determined whether the intra prediction mode or the inter prediction mode is applied to each coding unit. In the present disclosure, a maximum coding unit may be referred to as a maximum coding block, and a coding unit may be referred to as a coding block.

The size of the coding unit may be equal to or smaller than the maximum coding unit. Since a maximum coding unit is a coding unit having a maximum size, it may be referred to as a coding unit.

One or more prediction units for intra prediction or inter prediction may be determined from the coding unit. The size of the prediction unit may be equal to or smaller than the coding unit.

Also, one or more transform units for transform and quantization may be determined from the coding unit. The size of the conversion unit may be equal to or smaller than the coding unit. A transform unit is a reference block for transform and quantization, and residual samples of a coding unit may be transformed and quantized for each transform unit within the coding unit.

In the present disclosure, a current block may be a slice, a tile, a maximum coding unit, a coding unit, a prediction unit, or a transformation unit divided from the image 200 . Also, a sub-block of the current block is a block divided from the current block, and for example, if the current block is a maximum coding unit, the sub-block may be a coding unit, a prediction unit, or a transformation unit. Also, an upper block of the current block is a block that includes the current block as a part. For example, if the current block is a maximum coding unit, the upper block may be a picture sequence, picture, slice, or tile.

Hereinafter, a video decoding method, a video decoding apparatus, a video encoding method, and a video encoding apparatus according to an exemplary embodiment will be described with reference to FIGS. 3A to 9 .

3A is a diagram for explaining an example of a reference block in bi-directional prediction.

Referring to FIG. 3A, according to bi-directional prediction, two reference blocks, that is, a first reference block 315 and a second reference block 325 are obtained in a first reference frame 310, and the first reference block 315 ) and the second reference block 325, the current block 305 in the current frame 300 may be predicted.

3B is a diagram for explaining an example of a reference block in bidirectional prediction.

Referring to FIG. 3B , according to bidirectional prediction, a first reference block 315 in a first reference frame 320 of a past view of a current frame 300 and a second reference frame of a future view of the current frame 300 ( 330), the current block 305 in the current frame 300 may be predicted using the first reference block 315 and the second reference block 325.

3C is a diagram for explaining an example of a reference block in bidirectional prediction.

Referring to FIG. 3C , according to bidirectional prediction, a first reference block 315 in a first reference frame 340 of a past view of the current frame 300 and a second reference frame of a past view of the current frame 300 ( The second reference block 325 in 350 is obtained, and the current block 305 in the current frame 300 may be predicted using the first reference block 315 and the second reference block 325 .

3D is a diagram for explaining an example of a reference block in bi-directional prediction.

Referring to FIG. 3D , according to bidirectional prediction, a first reference block 315 in a first reference frame 360 of a future view of a current frame 300 and a second reference frame of a future view of the current frame 300 ( The second reference block 325 in 370 is obtained and the current block 305 in the current frame 300 may be predicted using the first reference block 315 and the second reference block 325 .

3E is a diagram for explaining a method of obtaining a current block by using a reference block in bi-directional prediction.

Referring to FIG. 3E , a reference image 335 is acquired using the first reference block 315 and the second reference block 325 acquired according to FIGS. 3A to 3D , and the reference image 335 is the current block. (305).

3A to 3E, in order to obtain the reference image 335 on the decoding side using the first reference block 315 and the second reference block 325, the first reference block 315 and the second reference block 315 are used. Information for block 325 should be transmitted. The information on the reference block may include information on the reference frame including the reference block (eg, reference index) and positional information (eg, motion vector) of the reference block relative to the position of the current block. there is. For example, the motion vector may express the displacement of each x-axis and y-axis in units of specific fractional pixels or specific integer pixels based on the position of the upper left pixel of the current block. Also, in motion vector transmission, a motion vector difference value, which is a difference between neighboring motion vector information and current motion vector information, may be transmitted using motion vector information of neighboring blocks. In addition, a motion vector prediction index for information used when selecting and using information of several neighboring blocks may be additionally transmitted. That is, in order to acquire one reference block, reference frame information, a motion vector prediction index, and a motion vector difference value must be transmitted. In the case of bi-directional prediction, two reference blocks are used, so reference frame information, motion vector prediction index, Two motion vector difference values are required. Accordingly, since motion estimation is performed for each reference block to acquire two reference blocks, two motion estimation is required, such as a system requiring real-time low-delay or low latency video transmission in terms of complexity. It may not be suitable for real-time transmission.

For real-time transmission, a method in which the motion vector difference value of the second reference block is not transmitted may be considered. In this method, motion estimation is performed by transmitting information on the first reference block only for the first reference block, and motion estimation is not performed on the second reference block. Therefore, although there may be some loss in terms of compression efficiency, it is advantageous for real-time transmission applications in terms of complexity.

Specifically, for a block that performs bi-directional prediction, a motion vector difference value for a first reference block is transmitted, and a motion vector difference value for a second reference block is not transmitted, and motion vector prediction for a second reference block is performed. A motion vector of the second reference block may be determined using the candidate. In this case, if the number of motion vector prediction candidates selectable for the second reference block is plural, since a flag or index must be signaled to select one of the plurality of motion vector prediction candidates, the encoding side additionally predicts the motion vector. A decision process must be performed to select one of the candidates. In general, for this determination process, for each of the selectable motion vector prediction candidates, a reference block at a corresponding location is fetched from a reference picture referred to by each of the motion vector prediction candidates and obtained through the motion vector of the first reference block. A predicted block of the current block is generated by averaging with the first reference block, and then a cost is calculated using a cost function such as sum of absolute differences (SAD), and then a block with a small cost is selected. In this case, since data must be fetched from reference pictures existing in the external memory as much as the number of motion vector prediction candidates, the required external memory bandwidth increases, which is not desirable for real-time transmission.

As a method for solving this problem, it may be configured to unconditionally transmit a value of a specific index position among motion vector prediction candidates for the selectable second reference block. Specifically, the first of motion vector prediction candidates?? It can be set to always have an index. Accordingly, on the encoding side, the flag of the motion vector prediction candidate of the second reference block may be always set to 0 and transmitted. In this case, there is no need to fetch several reference blocks to determine motion vector prediction candidates, which has an advantage in terms of complexity, but may degrade performance in terms of compression efficiency.

To solve this problem, according to an embodiment of the present disclosure, one of the motion vector prediction candidates of the second reference block is determined by considering the relationship between the motion vector of the first reference block and the motion vector prediction candidates of the second reference block. 2 can be selected as a reference block. In this case, there is no need to fetch several reference blocks from the external memory, so it has an advantage in terms of complexity, and in terms of compression efficiency, performance is improved compared to simply using a specific index. That is, it is efficient in terms of complexity while minimizing loss in terms of image quality.

4A is a diagram for explaining a method of obtaining a first reference block of a current block through motion estimation according to an embodiment of the present disclosure.

Referring to FIG. 4A , in order to find the first reference block 315 in the first reference frame 310 of the current block 305 of the current frame 300, a motion search area is used by using information on the first reference block. A motion search in 400 is performed, and motion compensation is performed to obtain the first reference block 315 .

4B is a diagram for explaining a method of deriving a second reference block using a first reference block obtained through motion estimation and candidate blocks of a current block according to an embodiment of the present disclosure.

Referring to FIG. 4B , a first reference block 315 is first acquired according to motion compensation as in FIG. 4A . When included in the first reference frame 310, which is the same frame as the first reference block 315 and the second reference block, the first reference frame 310 from neighboring blocks of the current block 305 of the current frame 300 Candidate positions 405 and 410 of the second reference block in ) are obtained. The distance between each of the first candidate block 405 and the second candidate block 410 of the second reference block and the first reference block 315 is identified to determine the first candidate block 405 that is the most distant candidate block. It can be determined as this second reference block.

The distance between each of the first candidate block 405 and the second candidate block 410 of the second reference block and the first reference block 315 is a Euclidean distance measurement method or a Manhattan distance measurement method. can be identified using

Specifically, the motion vector of the first reference block is (x0, y0), the motion vector of the second reference block is (x1, y1), the motion vector prediction candidates of the second reference block MV1 are two, 2 If the first candidate (MVP_1_1) of the reference block is (x11, y11) and the second candidate (MVP_1_2) of the second reference block is (x12, y12), the Manhattan distance between each candidate and the first reference block is measured By comparing the values, the second reference block may be determined according to a pseudo code such as Equation 1 below.

[Equation 1]

When selecting a second reference block within the same frame, when a block close to the first reference block is selected, an image similar to using only the first reference block is created, but when a block far from the first reference block is selected, a new image is created. Since a reference image of the form can be generated, coding efficiency can be improved by using the reference image in various ways.

That is, the decoding side receives motion information (motion vector difference value, motion vector prediction candidate index) for the first reference block, determines the motion vector of the first reference block using the transmitted motion information, and determines the motion vector of the current block. A candidate list including candidate blocks for the second reference block is determined using the block, and a block farthest from the location of the first reference block among the candidate blocks for the second reference block is determined as the second reference block. and use it Accordingly, only the motion information for the first reference block is transmitted, and the motion vector for the second reference block is determined using the transmitted motion information for the first reference block without transmission of the motion information for the second reference block. , compression efficiency can be improved by reducing the size of transmission information and selecting an efficient reference block.

Specifically, the decoding side obtains information on the motion vector of the first reference block. The information on the motion vector of the first reference block includes an index indicating one of motion vector prediction candidates for the first reference block and a motion vector difference value. The decoding side obtains the motion vector of the first reference block by using information on the motion vector of the first reference block, and determines a candidate list including motion vector prediction candidates of the second reference block. The decoding side identifies a distance between a motion vector of the first reference block and motion vectors of motion vector prediction candidates of the second reference block. If the motion vector of the first reference block is (x0, y0) and the motion vector of the motion vector prediction candidate is (x1, y1), |x0 - x1| + |y0 - y1| Based on the measured distance, a motion vector of a motion vector prediction candidate having the furthest distance from the motion vector of the first reference block is determined as the motion vector of the second reference block.

According to an embodiment, among the distances between the motion vector of the first reference block and motion vectors of the motion vector prediction candidates of the second reference block, the motion vector prediction candidate of the candidate block having the shortest distance is the motion vector of the second reference block. may be determined by

In addition, instead of the first reference block 315, the distance between the current block 305 and the first candidate block 405 and the second candidate block 410 of the second reference block is identified, and the candidate block is the furthest away from each other. A motion vector prediction candidate of may be determined as the motion vector of the second reference block or a motion vector prediction candidate of the candidate block having the closest distance may be determined as the motion vector of the second reference block.

In addition, based on the distance between the first reference block 315 and the current block 305, the motion vector prediction candidate of the candidate block furthest from the first reference block 315 is determined as the motion vector of the second reference block. or a motion vector prediction candidate of a candidate block having the closest distance may be determined as the motion vector of the second reference block. Specifically, when the distance between the first reference block 315 and the current block 305 is equal to or less than a predetermined size K, the motion vector prediction candidate of the candidate block furthest from the first reference block 315 is the second reference block. is determined as a motion vector of , and when the distance between the first reference block 315 and the current block 305 is greater than K, the motion vector prediction candidate of the candidate block closest to the first reference block 315 is the second It may be determined as a motion vector of a reference block. In addition, when the distance between the first reference block 315 and the current block 305 is smaller than a predetermined size K, the motion vector prediction candidate of the candidate block furthest from the first reference block 315 is the second reference block. is determined as a motion vector of , and when the distance between the first reference block 315 and the current block 305 is greater than K, the motion vector prediction candidate of the candidate block closest to the first reference block 315 is the second reference block. It may be determined as a motion vector of a block. Even in this case, since there is no need to load several reference blocks from an external memory, it has an advantage in terms of complexity and performance can be improved in terms of compression efficiency compared to simply using a specific index.

According to an embodiment, there may be two or more motion vector candidates of the second reference block.

According to an embodiment, when there are two motion vector candidates for the motion vector of the first reference block, there may be two or more motion vector candidates for the motion vector of the second reference block.

According to an embodiment, when a motion vector difference value with respect to a second reference block is transmitted in a specific picture or specific block, an index or a flag may be signaled for motion vector candidates of the second reference block, or a first signal may be signaled without signaling. The motion vector of the second reference block may be determined based on the distance between the motion vector of the first reference block and the motion vectors of the motion vector prediction candidates of the second reference block.

5A is a diagram for explaining a method of deriving a second reference block using a first reference block obtained through motion estimation and candidate blocks of a current block according to an embodiment of the present disclosure.

Referring to FIG. 5A , a first reference block 315 is first acquired according to motion compensation. When the first reference block 315 is included in the first reference frame 320 and the second reference block is included in the second reference frame 330 at a different viewpoint from the first reference frame 320, the current frame 300 Candidate positions 505 and 510 of the second reference block in the second reference frame 330 are obtained from neighboring blocks of the current block 305 of ). A distance between each of the first candidate block 505 and the second candidate block 510 of the second reference block and the first reference block 315 is identified to determine the distance between the first candidate block 505 and the second candidate block 510. ), a candidate block having the closest distance may be determined as the second reference block.

In bi-prediction, when two reference blocks are located in different reference frames, the probability of pointing to the same object is high when the distance between the two reference blocks is short. Based on the distance between each of the candidate blocks of the first reference block and the second reference block, a candidate block having a short distance is determined as the second reference block, so that coding efficiency may be improved.

Specifically, the decoding side obtains information on the motion vector of the first reference block. Information on the motion vector of the first reference block includes the direction of the first reference frame to which the first reference block belongs, that is, a direction index indicating whether it is a past view or a future view based on the current frame, a reference index indicating the first reference frame, and a first reference frame. 1 includes an index indicating one of motion vector prediction candidates for a reference block and a motion vector difference value. The decoding side obtains a motion vector of the first reference block by using information on the motion vector of the first reference block. The decoding side determines a candidate list including motion vector prediction candidates of the second reference block having a direction different from that of the motion vector of the first reference block among motion vector prediction candidates of the second reference block by using the direction index. The decoding side obtains the picture order count (POC) of the reference frame using the reference index. POC means a value numbered according to the display order of pictures. The decoding side identifies a scaled distance between a motion vector of the first reference block and motion vectors of motion vector prediction candidates having a direction different from that of the first reference block among motion vector prediction candidates of the second reference block by using the POC. The scaled distance is measured according to Equation 2 when the motion vector and POC of the first reference block are (x0, y0, POC0) and the motion vector and POC of the motion vector prediction candidate are (x1, y1, POC1).

[Equation 2]

A motion vector of a motion vector prediction candidate having the closest distance from the motion vector of the first reference block based on the measured distance (dist), that is, the scaled distance, is determined as the motion vector of the second reference block.

According to an embodiment, a motion vector of a motion vector prediction candidate having the furthest distance from the motion vector of the first reference block based on the scaled distance may be determined as the motion vector of the second reference block. If there is one motion vector having a POC distant from POC0, the corresponding motion vector may be used as it is.

According to an embodiment, if there are a plurality of motion vectors having POCs distant from POC0, the motion vector having the farthest distance from the motion vector of the first reference block among motion vector candidates having POCs farther from POC0 can be used

According to an embodiment, if there is one motion vector having a POC close to POC0, the corresponding motion vector may be used as it is.

According to an embodiment, if there are a plurality of motion vectors having POCs that are close to POC0, the motion vector having the closest distance to the motion vector of the first reference block among motion vector candidates having POCs that are close to POC0 is the motion vector. can be used

5B is a diagram for explaining a method of deriving a second reference block using a first reference block obtained through motion estimation and candidate blocks of a current block according to an embodiment of the present disclosure.

Referring to FIG. 5B , a first reference block 315 is obtained first according to motion compensation. When the first reference block 315 is included in the first reference frame 340 and the second reference block is included in the second reference frame 350 at the same time as the first reference frame 340, the current frame 300 Candidate positions 515 and 520 of the second reference block in the second reference frame 350 are obtained from neighboring blocks of the current block 305 of ). A distance between each of the first candidate block 515 and the second candidate block 520 of the second reference block and the first reference block 315 is identified to determine the distance between the first candidate block 515 and the second candidate block 520. ), a candidate block having the closest distance may be determined as the second reference block.

Specifically, the decoding side obtains information on the motion vector of the first reference block. The information on the motion vector of the first reference block includes a reference index indicating a first reference frame to which the first reference block belongs, an index indicating one of motion vector prediction candidates for the first reference block, and a motion vector difference value. The decoding side obtains the motion vector of the first reference block by using information on the motion vector of the first reference block, and determines a candidate list including motion vector prediction candidates of the second reference block. The decoding side obtains the picture order count (POC) of the reference frame using the reference index. The decoding side identifies a scaled distance between a motion vector of the first reference block and motion vectors of motion vector prediction candidates of the second reference block by using the POC. The scaled distance is measured according to Equation 3 when the motion vector and POC of the first reference block are (x0, y0, POC0) and the motion vector and POC of the motion vector prediction candidate are (x1, y1, POC1).

[Equation 3]

A motion vector of a motion vector prediction candidate having the closest distance to the motion vector of the first reference block based on the measured distance (dist), that is, the scaled distance, is determined as the motion vector of the second reference block.

According to an embodiment, a motion vector of a motion vector prediction candidate having the furthest distance from the motion vector of the first reference block based on the scaled distance may be determined as the motion vector of the second reference block.

According to an embodiment, if there is one motion vector having a far POC based on POC0, the corresponding motion vector may be used as it is.

According to an embodiment, if there are a plurality of motion vectors having POCs distant from POC0, the motion vector having the farthest distance from the motion vector of the first reference block among motion vector candidates having POCs farther from POC0 can be used

According to an embodiment, if there is one motion vector having a POC close to POC0, the corresponding motion vector may be used as it is.

According to an embodiment, if there are a plurality of motion vectors having POCs that are close to POC0, the motion vector having the closest distance to the motion vector of the first reference block among motion vector candidates having POCs that are close to POC0 is the motion vector. can be used

According to an embodiment, when there are a plurality of reference pictures or reference frames, a distance between the current picture and the first reference picture to which the first reference block belongs based on the motion vector of the first reference block and the motion vector of the second reference block A candidate block of the second reference block may be obtained by identifying a distance between motion vector candidates of the first reference block and the second reference block by scaling based on the distance between the second reference pictures to which the candidate belongs.

Specifically, the distance may be identified according to the cost function f(N) of Equation 4 below.

[Equation 4]

POC _c is the POC of the current picture belonging to the current block, and if the motion vector of the first reference block is MvL0, MvL0[0] is the x-component of the motion vector of the first reference block, and MvL0[1] is the first reference block is the y component of the motion vector of , POC _L0 is the POC of the first reference picture to which the first reference block belongs, and the motion vector candidate of the second reference block is MvL1[N], MvL1[N][0] is 2 is the x component of the motion vector candidate of the reference block, MvL1[N][1] is the y component of the motion vector candidate of the second reference block, and POC _L1 is the second reference picture to which the motion vector candidate of the second reference block belongs. is a POC of , and N is an index indicating a specific motion vector prediction candidate, which is 1 or more and M or less.

A final candidate N _F having the greatest cost is determined according to Equation 5 below using the cost function f(N) of Equation 4 above, and the final candidate N _F may be determined as a motion vector of the second reference block.

[Equation 5]

According to an embodiment, a final candidate having the smallest cost is determined using the cost function f(N) of Equation 4, unlike Equation 5, and the final candidate may be determined as a motion vector of the second reference block.

According to an embodiment, when there is only one reference picture, the term for comparing the POC may be removed and expressed as in Equation 6 below.

[Equation 6]

According to an embodiment, candidates may be selected in the order in which the second reference picture indicated by the motion vector candidate of the second reference block has the farthest distance from the first reference picture indicated by the motion vector of the first reference block. In addition, when there is only one motion vector candidate included in the second reference picture having the farthest distance, the corresponding candidate may be selected as the motion vector of the second reference block.

According to an embodiment, candidates may be selected in the order in which the second reference picture indicated by the motion vector candidate of the second reference block has the shortest distance from the first reference picture indicated by the motion vector of the first reference block. In addition, when there is only one motion vector candidate included in the second reference picture having the shortest distance, the candidate may be selected as the motion vector of the second reference block.

According to an embodiment, a second reference picture indicated by the motion vector candidate of the second reference block is a motion vector candidate included in the second reference picture that is the furthest away from the first reference picture indicated by the motion vector of the first reference block. If there are a plurality, the distance may be identified according to Equation 7.

[Equation 7]

POC _c is the POC of the current picture belonging to the current block, and if the motion vector of the first reference block is MvL0, MvL0[0] is the x-component of the motion vector of the first reference block, and MvL0[1] is the first reference block is the y component of the motion vector of , POC _L0 is the POC of the first reference picture to which the first reference block belongs, and the motion vector candidate of the second reference block is MvL1[N], MvL1[N][0] is 2 is an x-component of a motion vector candidate belonging to a second reference picture farthest from the first reference picture among reference pictures to which the reference block belongs, and MvL1[N][1] is among the reference pictures to which the second reference block belongs. A y-component of a motion vector candidate belonging to a second reference picture farthest from the first reference picture, and POC _L1 is the second reference picture furthest from the first reference picture among reference pictures to which the second reference block belongs. POC, and N is an index meaning a specific motion vector prediction candidate, which is 2 or more and M or less.

A final candidate N _F having the greatest cost is determined according to Equation 8 below using the cost function f(N) of Equation 7, and the final candidate N _F may be determined as a motion vector of the second reference block.

[Equation 8]

According to an embodiment, in Equation 7, MvL1[N][0] is an x-component of a motion vector candidate belonging to a second reference picture having the closest distance from the first reference picture among reference pictures to which the second reference block belongs. , MvL1[N][1] is a y-component of a motion vector candidate belonging to a second reference picture having the closest distance from the first reference picture among reference pictures to which the second reference block belongs, and POC _L1 is the second reference block It may be the POC of a second reference picture that is closest to the first reference picture among the reference pictures to which it belongs. Accordingly, unlike Equation 8, a final candidate having the lowest cost is determined using the cost function f(N) of Equation 7, and the final candidate may be determined as a motion vector of the second reference block.

According to an embodiment, as shown in Equation 9 below, the distance may be identified by scaling the motion vector of the first reference block. In this case, there is an advantage in that scaling is performed only once for the motion vector of the first reference block.

[Equation 9]

According to an embodiment, the Euclidean distance measurement method as shown in Equation 10 below may be used instead of the Manhattan distance measurement method of Equation 1 above.

[Equation 10]

Here, if the motion vector of the first reference block is MvL0, MvL0[0] is the x component of the motion vector of the first reference block, MvL0[1] is the y component of the motion vector of the first reference block, and the second reference block If the motion vector candidate of the block is MvL1[N], MvL1[N][0] is the x-component of the motion vector candidate of the second reference block, and MvL1[N][1] is the motion vector candidate of the second reference block. is the y component of

6 is a flowchart of a video decoding method according to an embodiment of the present disclosure.

In S610, the video decoding apparatus 700 may obtain information on a first motion vector of the current block from the bitstream.

According to an embodiment, information on the first motion vector of the current block may be determined through SATD (Sum of Transform Difference) or RDO (Rate Distortion Optimization) calculation at the encoding side and transmitted through the bitstream.

According to an embodiment, between a first reference frame including the first reference block and a second reference frame including motion vector candidates of the second reference block according to information included in the motion vector information of the first reference block. relationship can be judged.

According to an embodiment, when the information on the motion vector of the first reference block includes an index indicating one of motion vector prediction candidates for the first reference block and a motion vector difference value, the first reference block including the first reference block The reference frame and the second reference frame including motion vector candidates of the second reference block are the same.

According to an embodiment, if the information on the motion vector of the first reference block includes a reference index indicating the first reference frame, an index indicating one of motion vector prediction candidates for the first reference block, and a motion vector difference value, The first reference frame including the first reference block and the second reference frame including motion vector candidates of the second reference block are different, and the viewpoints of the first reference frame and the second reference frame are the current frame including the current block. It is the same point in time based on .

According to an embodiment, the information on the motion vector of the first reference block includes the direction of the first reference frame to which the first reference block belongs, that is, a direction index indicating whether it is a past view or a future view based on the current frame, and the first reference frame. If a reference index indicating , an index indicating one of the motion vector prediction candidates for the first reference block, and a motion vector difference value are included, the motion vector candidates of the first reference frame including the first reference block and the second reference block are Included second reference frames are different, and viewpoints of the first reference frame and the second reference frame are different viewpoints based on the current frame including the current block.

In S630, the video decoding apparatus 700 may determine a first motion vector based on information on the first motion vector.

In S650, the video decoding apparatus 700 may determine a candidate list including a plurality of candidate predictive motion vectors for determining a second motion vector of the current block.

In operation S670, the video decoding apparatus 700 may determine one of a plurality of candidate motion vectors as a second motion vector based on a distance between the plurality of candidate motion vectors and the first motion vector.

According to an embodiment, when the first reference frame of the first motion vector is the same as the second reference frame of the plurality of candidate predictive motion vectors, the second motion vector is the first among the plurality of candidate predictive motion vectors. It may be a candidate prediction motion vector that has the furthest distance from the 2 motion vectors.

According to an embodiment, when the first reference frame of the first motion vector is the same as the second reference frame of the plurality of candidate motion vectors, an x component of the plurality of candidate motion vectors and a y of the first motion vector The distance can be identified using a component.

According to an embodiment, when a first reference frame of the first motion vector is different from a second reference frame of the plurality of candidate predictive motion vectors, the second motion vector is the first one of the plurality of candidate predictive motion vectors. It may be a candidate prediction motion vector having the closest distance to the 2 motion vectors.

According to an embodiment, a first reference frame of the first motion vector is different from a second reference frame of the plurality of candidate predicted motion vectors, and a direction of the first reference frame and a direction of the second reference frame are the same. , the distance may be identified by scaling the plurality of candidate predictive motion vectors based on the Picture Order Count (POC) of the first reference frame, the POC of the plurality of candidate predictive motion vectors, and the POC of the current block. there is.

According to an embodiment, a first reference frame of the first motion vector is different from a second reference frame of the plurality of candidate predicted motion vectors, and a direction of the first reference frame and a direction of the second reference frame are different. , second candidate predicted motion vectors having a direction different from that of the first motion vector among the plurality of candidate motion vectors are determined, the POC of the first reference frame, the POC of the second candidate motion vectors, and the distance may be identified by scaling the second candidate predictive motion vectors based on the POC of the current block.

According to an embodiment, when the reference frames of the plurality of candidate motion vectors are a plurality of second reference frames, the POC of the first reference frame, the POC of the second candidate prediction motion vectors, and the POC of the current block The distance may be identified by scaling the second candidate predictive motion vectors based on .

According to an embodiment, when the reference frames of the plurality of candidate predicted motion vectors are a plurality of second reference frames, a distance closest to the first reference frame of the first motion vector among the plurality of second reference frames A third reference frame may be determined, and the second motion vector may be determined based on a distance between a plurality of third candidate predicted motion vectors included in the third reference frame and the first motion vector.

According to an embodiment, when the reference frames of the plurality of candidate predicted motion vectors are a plurality of second reference frames, a distance closest to the first reference frame of the first motion vector among the plurality of second reference frames When a third reference frame is determined and the third candidate motion vector included in the third reference frame is one, the third candidate motion vector motion vector may be determined as the second motion vector.

In S690, the video decoding apparatus 700 may determine a motion vector of the current block by using the first motion vector and the second motion vector.

7 is a diagram showing the configuration of a video decoding apparatus according to an embodiment of the present disclosure.

Referring to FIG. 7 , a video decoding apparatus 700 may include a receiving unit 710 and a decoding unit 720.

The receiver 710 and the decoder 720 may be implemented with one or more processors. The receiver 710 and the decoder 720 may operate according to instructions stored in memory.

Although FIG. 7 shows the receiver 710 and the decoder 720 individually, the receiver 710 and the decoder 720 may be implemented by a single processor. In this case, the receiver 710 and the decoder 720 may be implemented as a dedicated processor, or a combination of software and a general-purpose processor such as an application processor (AP), central processing unit (CPU), or graphic processing unit (GPU). can be implemented through In addition, a dedicated processor may include a memory for implementing an embodiment of the present disclosure or a memory processing unit for using an external memory.

The receiving unit 710 and the decoding unit 720 may be composed of a plurality of processors. In this case, it may be implemented with a combination of dedicated processors, or it may be implemented with a combination of software and a plurality of general-purpose processors such as APs, CPUs, or GPUs. Also, the processor may include a dedicated artificial intelligence processor. As another example, an AI-only processor may be configured as a separate chip from the processor.

The receiving unit 710 obtains information on a first motion vector of a current block from a bitstream.

According to an embodiment, information on the first motion vector of the current block may be determined through SATD (Sum of Transform Difference) or RDO (Rate Distortion Optimization) calculation at the encoding side and transmitted through the bitstream.

The decoder 720 determines a first motion vector based on information on the first motion vector.

The decoder 720 determines a candidate list including a plurality of candidate prediction motion vectors for determining the second motion vector of the current block.

The decoder 720 determines one of the plurality of candidate motion vectors as the second motion vector based on the distance between the plurality of candidate motion vectors and the first motion vector.

The decoder 720 determines the motion vector of the current block using the first motion vector and the second motion vector.

8 is a flowchart of a video encoding method according to an embodiment of the present disclosure.

In S810, the video encoding apparatus 900 may generate information on a first motion vector of the current block.

According to an embodiment, according to a relationship between a first reference frame including the first reference block and a second reference frame including motion vector candidates of the second reference block, motion vectors of the first reference block are included in the information. information can be determined.

According to an embodiment, if the first reference frame including the first reference block and the second reference frame including motion vector candidates of the second reference block are the same, information on the motion vector of the first reference block is It includes an index indicating one of the motion vector prediction candidates for the reference block and a motion vector difference value.

According to an embodiment, the first reference frame including the first reference block and the second reference frame including motion vector candidates of the second reference block are different, and views of the first reference frame and the second reference frame are currently If it is the same view based on the current frame including the block, information on the motion vector of the first reference block includes a reference index indicating the first reference frame, an index indicating one of motion vector prediction candidates for the first reference block, and motion Contains vector difference values.

According to an embodiment, the first reference frame including the first reference block and the second reference frame including motion vector candidates of the second reference block are different, and views of the first reference frame and the second reference frame are currently If it is a different view based on the current frame including the block, the information on the motion vector of the first reference block indicates the direction of the first reference frame to which the first reference block belongs, that is, whether it is a past view or a future view based on the current frame. It includes a direction index, a reference index indicating the first reference frame, an index indicating one of motion vector prediction candidates for the first reference block, and a motion vector difference value.

In S830, the video encoding apparatus 900 may determine a first motion vector based on information on the first motion vector.

In S850, the video encoding apparatus 900 may determine a candidate list including a plurality of candidate prediction motion vectors for determining a second motion vector of the current block.

In operation S870, the video encoding apparatus 900 may determine a first one of a plurality of candidate motion vectors as a second motion vector based on a distance between the plurality of candidate motion vectors and the first motion vector.

According to an embodiment, when the first reference frame of the first motion vector is the same as the second reference frame of the plurality of candidate predictive motion vectors, the second motion vector is the first among the plurality of candidate predictive motion vectors. It may be a candidate prediction motion vector that has the furthest distance from the 2 motion vectors.

According to an embodiment, when the first reference frame of the first motion vector is the same as the second reference frame of the plurality of candidate motion vectors, an x component of the plurality of candidate motion vectors and a y of the first motion vector The distance can be identified using a component.

According to an embodiment, when a first reference frame of the first motion vector is different from a second reference frame of the plurality of candidate predictive motion vectors, the second motion vector is the first one of the plurality of candidate predictive motion vectors. It may be a candidate prediction motion vector having the closest distance to the 2 motion vectors.

According to an embodiment, a first reference frame of the first motion vector and a second reference frame of the plurality of candidate predicted motion vectors are different, and directions of the first reference frame and directions of the second reference frame are the same. , the distance may be identified by scaling the plurality of candidate predictive motion vectors based on the Picture Order Count (POC) of the first reference frame, the POC of the plurality of candidate predictive motion vectors, and the POC of the current block. there is.

According to an embodiment, a first reference frame of the first motion vector is different from a second reference frame of the plurality of candidate predicted motion vectors, and a direction of the first reference frame and a direction of the second reference frame are different. , second candidate predicted motion vectors having a direction different from that of the first motion vector among the plurality of candidate motion vectors are determined, the POC of the first reference frame, the POC of the second candidate motion vectors, and the distance may be identified by scaling the second candidate predictive motion vectors based on the POC of the current block.

According to an embodiment, when the reference frames of the plurality of candidate motion vectors are a plurality of second reference frames, the POC of the first reference frame, the POC of the second candidate prediction motion vectors, and the POC of the current block The distance may be identified by scaling the second candidate predictive motion vectors based on .

According to an embodiment, when the reference frames of the plurality of candidate predicted motion vectors are a plurality of second reference frames, a distance closest to the first reference frame of the first motion vector among the plurality of second reference frames A third reference frame may be determined, and the second motion vector may be determined based on a distance between a plurality of third candidate predicted motion vectors included in the third reference frame and the first motion vector.

According to an embodiment, when the reference frames of the plurality of candidate predicted motion vectors are a plurality of second reference frames, a distance closest to the first reference frame of the first motion vector among the plurality of second reference frames When a third reference frame is determined and the third candidate motion vector included in the third reference frame is one, the third candidate motion vector motion vector may be determined as the second motion vector.

In S890, the video encoding apparatus 900 may encode information about the first motion vector by determining the motion vector of the current block using the first motion vector and the second motion vector.

According to an embodiment, information on the first motion vector of the current block may be determined and signaled through calculation of Sum of Transform Difference (SATD) or Rate Distortion Optimization (RDO) at the encoding side.

9 is a diagram showing the configuration of a video encoding apparatus according to an embodiment of the present disclosure.

Referring to FIG. 9 , a video encoding apparatus 900 may include a generator 910 and an encoder 920 .

The generator 910 and the encoder 920 may be implemented with one or more processors. The generator 910 and the encoder 920 may operate according to instructions stored in memory.

Although FIG. 9 shows the generator 910 and the encoder 920 separately, the generator 910 and the encoder 920 may be implemented by a single processor. In this case, the generator 910 and the encoder 920 may be implemented as a dedicated processor, or a combination of software and a general-purpose processor such as an application processor (AP), central processing unit (CPU), or graphic processing unit (GPU). It can also be implemented through In addition, a dedicated processor may include a memory for implementing an embodiment of the present disclosure or a memory processing unit for using an external memory.

The generating unit 910 and the encoding unit 920 may be composed of a plurality of processors. In this case, it may be implemented with a combination of dedicated processors, or it may be implemented with a combination of software and a plurality of general-purpose processors such as APs, CPUs, or GPUs. Also, the processor may include a dedicated artificial intelligence processor. As another example, an AI-only processor may be configured as a separate chip from the processor.

The generation unit 910 generates information about a first motion vector of a current block from a bitstream.

The encoder 920 determines a first motion vector based on information on the first motion vector.

The encoder 920 determines a candidate list including a plurality of candidate prediction motion vectors for determining the second motion vector of the current block.

The encoder 920 determines one of the plurality of candidate motion vectors as the second motion vector based on the distance between the plurality of candidate motion vectors and the first motion vector.

The encoder 920 encodes information on the first motion vector by determining a motion vector of the current block using the first motion vector and the second motion vector.

According to an embodiment, information on the first motion vector of the current block may be determined and signaled through calculation of Sum of Transform Difference (SATD) or Rate Distortion Optimization (RDO) at the encoding side.

A video decoding method according to an embodiment includes obtaining information on a first motion vector of a current block from a bitstream; determining the first motion vector based on information on the first motion vector; determining a candidate list including a plurality of candidate predictive motion vectors for determining a second motion vector of the current block; determining one of the plurality of candidate motion vectors as the second motion vector based on a distance between the plurality of candidate motion vectors and the first motion vector; and determining a motion vector of the current block using the first motion vector and the second motion vector.

A video decoding method according to an embodiment of the present disclosure performs motion compensation by determining a second motion vector based on a distance between a first motion vector and a plurality of candidate predicted motion vectors without signaling information on the second motion vector. Since this is performed only once, there is no need to load multiple reference blocks from external memory, which can have an advantage in terms of complexity.

According to an embodiment, when the first reference frame of the first motion vector is the same as the second reference frame of the plurality of candidate predictive motion vectors, the second motion vector is the first among the plurality of candidate predictive motion vectors. It may be a candidate prediction motion vector that has the furthest distance from the 2 motion vectors.

According to an embodiment, when the first reference frame of the first motion vector is the same as the second reference frame of the plurality of candidate motion vectors, an x component of the plurality of candidate motion vectors and the first motion vector The distance can be identified using the y component.

According to an embodiment, when a first reference frame of the first motion vector is different from a second reference frame of the plurality of candidate predictive motion vectors, the second motion vector is the first one of the plurality of candidate predictive motion vectors. It may be a candidate prediction motion vector having the closest distance to the 2 motion vectors.

According to an embodiment, a first reference frame of the first motion vector is different from a second reference frame of the plurality of candidate predicted motion vectors, and a direction of the first reference frame and a direction of the second reference frame are the same. , the distance may be identified by scaling the plurality of candidate predictive motion vectors based on the Picture Order Count (POC) of the first reference frame, the POC of the plurality of candidate predictive motion vectors, and the POC of the current block. there is.

According to an embodiment, a first reference frame of the first motion vector is different from a second reference frame of the plurality of candidate predicted motion vectors, and a direction of the first reference frame and a direction of the second reference frame are different. , second candidate predicted motion vectors having a direction different from that of the first motion vector among the plurality of candidate motion vectors are determined, the POC of the first reference frame, the POC of the second candidate motion vectors, and the distance may be identified by scaling the second candidate predictive motion vectors based on the POC of the current block.

According to an embodiment, when the reference frames of the plurality of candidate motion vectors are a plurality of second reference frames, the POC of the first reference frame, the POC of the second candidate prediction motion vectors, and the POC of the current block The distance may be identified by scaling the second candidate predictive motion vectors based on .

According to an embodiment, when the reference frames of the plurality of candidate predicted motion vectors are a plurality of second reference frames, a distance closest to the first reference frame of the first motion vector among the plurality of second reference frames A third reference frame may be determined, and the second motion vector may be determined based on a distance between a plurality of third candidate predicted motion vectors included in the third reference frame and the first motion vector.

According to an embodiment, when the reference frames of the plurality of candidate predicted motion vectors are a plurality of second reference frames, a distance closest to the first reference frame of the first motion vector among the plurality of second reference frames When a third reference frame is determined and the third candidate motion vector included in the third reference frame is one, the third candidate motion vector motion vector may be determined as the second motion vector.

A video decoding method according to an embodiment of the present disclosure performs coding for a current block by determining a second motion vector according to a relationship between a first reference frame and a second reference frame without signaling information on the second motion vector. Efficiency can be improved.

A video decoding apparatus according to an embodiment includes a memory for storing one or more instructions; and at least one processor operating according to the one or more instructions, wherein the at least one processor obtains information on a first motion vector of a current block from a bitstream, and provides information on the first motion vector. to determine the first motion vector, determine a candidate list including a plurality of candidate predictive motion vectors for determining a second motion vector of the current block, and determine the plurality of candidate predictive motion vectors and the first motion vector. Based on the distance between the motion vectors, one of the plurality of candidate predicted motion vectors is determined as the second motion vector, and the motion vector of the current block is determined using the first motion vector and the second motion vector. can decide

A video decoding apparatus according to an embodiment of the present disclosure performs motion compensation by determining a second motion vector based on a distance between a first motion vector and a plurality of candidate predicted motion vectors without signaling information on the second motion vector. Since this is performed only once, there is no need to load multiple reference blocks from external memory, which can have an advantage in terms of complexity.

According to an embodiment, when the first reference frame of the first motion vector is the same as the second reference frame of the plurality of candidate predictive motion vectors, the second motion vector is the first among the plurality of candidate predictive motion vectors. It may be a candidate prediction motion vector that has the furthest distance from the 2 motion vectors.

According to an embodiment, when the first reference frame of the first motion vector is the same as the second reference frame of the plurality of candidate motion vectors to be predicted, an x component of the plurality of candidate motion vectors to be predicted and the first motion vector The distance can be identified using the y component.

According to an embodiment, when a first reference frame of the first motion vector is different from a second reference frame of the plurality of candidate predictive motion vectors, the second motion vector is the first one of the plurality of candidate predictive motion vectors. It may be a candidate prediction motion vector having the closest distance to the 2 motion vectors.

According to an embodiment, a first reference frame of the first motion vector is different from a second reference frame of the plurality of candidate predicted motion vectors, and a direction of the first reference frame and a direction of the second reference frame are the same. , the distance may be identified by scaling the plurality of candidate predictive motion vectors based on the Picture Order Count (POC) of the first reference frame, the POC of the plurality of candidate predictive motion vectors, and the POC of the current block. there is.

According to an embodiment, a first reference frame of the first motion vector is different from a second reference frame of the plurality of candidate predicted motion vectors, and a direction of the first reference frame and a direction of the second reference frame are different. , second candidate predicted motion vectors having a direction different from that of the first motion vector among the plurality of candidate motion vectors are determined, the POC of the first reference frame, the POC of the second candidate motion vectors, and the distance may be identified by scaling the second candidate predictive motion vectors based on the POC of the current block.

According to an embodiment, when the reference frames of the plurality of candidate motion vectors are a plurality of second reference frames, the POC of the first reference frame, the POC of the second candidate prediction motion vectors, and the POC of the current block The distance may be identified by scaling the second candidate predictive motion vectors based on .

According to an embodiment, when the reference frames of the plurality of candidate predicted motion vectors are a plurality of second reference frames, a distance closest to the first reference frame of the first motion vector among the plurality of second reference frames A third reference frame may be determined, and the second motion vector may be determined based on a distance between a plurality of third candidate predicted motion vectors included in the third reference frame and the first motion vector.

According to an embodiment, when the reference frames of the plurality of candidate predicted motion vectors are a plurality of second reference frames, a distance closest to the first reference frame of the first motion vector among the plurality of second reference frames When a third reference frame is determined and the third candidate motion vector included in the third reference frame is one, the third candidate motion vector motion vector may be determined as the second motion vector.

A video decoding apparatus according to an embodiment of the present disclosure performs coding for a current block by determining a second motion vector according to a relationship between a first reference frame and a second reference frame without signaling information on the second motion vector. Efficiency can be improved.

A video encoding method according to an embodiment includes generating information about a first motion vector of a current block; determining the first motion vector based on information on the first motion vector; determining a candidate list including a plurality of candidate predictive motion vectors for determining a second motion vector of the current block; determining one of the plurality of candidate motion vectors as the second motion vector based on a distance between the plurality of candidate motion vectors and the first motion vector; and determining a motion vector of the current block by using the first motion vector and the second motion vector, and encoding information on the first motion vector.

A video encoding method according to an embodiment of the present disclosure performs decoding by determining a second motion vector based on distances between a first motion vector and a plurality of candidate predicted motion vectors without generating information on the second motion vector. Compression efficiency can be improved by reducing the size of information transmitted to the side.

According to an embodiment, when the first reference frame of the first motion vector is the same as the second reference frame of the plurality of candidate predictive motion vectors, the second motion vector is the first among the plurality of candidate predictive motion vectors. It may be a candidate prediction motion vector that has the furthest distance from the 2 motion vectors.

According to an embodiment, when the first reference frame of the first motion vector is the same as the second reference frame of the plurality of candidate motion vectors, an x component of the plurality of candidate motion vectors and the first motion vector The distance can be identified using the y component.

According to an embodiment, when a first reference frame of the first motion vector is different from a second reference frame of the plurality of candidate predictive motion vectors, the second motion vector is the first one of the plurality of candidate predictive motion vectors. It may be a candidate prediction motion vector having the closest distance to the 2 motion vectors.

According to an embodiment, a first reference frame of the first motion vector is different from a second reference frame of the plurality of candidate predicted motion vectors, and a direction of the first reference frame and a direction of the second reference frame are the same. , the distance may be identified by scaling the plurality of candidate predictive motion vectors based on the Picture Order Count (POC) of the first reference frame, the POC of the plurality of candidate predictive motion vectors, and the POC of the current block. there is.

According to an embodiment, a first reference frame of the first motion vector is different from a second reference frame of the plurality of candidate predicted motion vectors, and a direction of the first reference frame and a direction of the second reference frame are different. , second candidate predicted motion vectors having a direction different from that of the first motion vector among the plurality of candidate motion vectors are determined, the POC of the first reference frame, the POC of the second candidate motion vectors, and the distance may be identified by scaling the second candidate predictive motion vectors based on the POC of the current block.

According to an embodiment, when the reference frames of the plurality of candidate motion vectors are a plurality of second reference frames, the POC of the first reference frame, the POC of the second candidate prediction motion vectors, and the POC of the current block The distance may be identified by scaling the second candidate predictive motion vectors based on .

According to an embodiment, when the reference frames of the plurality of candidate predicted motion vectors are a plurality of second reference frames, a distance closest to the first reference frame of the first motion vector among the plurality of second reference frames A third reference frame may be determined, and the second motion vector may be determined based on a distance between a plurality of third candidate predicted motion vectors included in the third reference frame and the first motion vector.

According to an embodiment, when the reference frames of the plurality of candidate predicted motion vectors are a plurality of second reference frames, a distance closest to the first reference frame of the first motion vector among the plurality of second reference frames When a third reference frame is determined and the third candidate motion vector included in the third reference frame is one, the third candidate motion vector motion vector may be determined as the second motion vector.

In a video encoding method according to an embodiment of the present disclosure, coding of a current block is performed by determining a second motion vector according to a relationship between a first reference frame and a second reference frame without generating information on the second motion vector. Efficiency can be improved.

A video encoding apparatus according to an embodiment includes a memory for storing one or more instructions; and at least one processor that operates according to the one or more instructions, wherein the at least one processor generates information about a first motion vector of a current block, and based on the information about the first motion vector, the at least one processor A first motion vector is determined, a candidate list including a plurality of candidate predictive motion vectors for determining a second motion vector of the current block is determined, and a distance between the plurality of candidate predictive motion vectors and the first motion vector is determined. Based on the distance of , one of the plurality of candidate predicted motion vectors is determined as the second motion vector, and a motion vector of the current block is determined using the first motion vector and the second motion vector to obtain a second motion vector. 1 Information on a motion vector can be encoded.

A video encoding apparatus according to an embodiment of the present disclosure performs decoding by determining a second motion vector based on a distance between a first motion vector and a plurality of candidate predicted motion vectors without generating information on the second motion vector. Compression efficiency can be improved by reducing the size of information transmitted to the side.

According to an embodiment, when the first reference frame of the first motion vector is the same as the second reference frame of the plurality of candidate predictive motion vectors, the second motion vector is the first among the plurality of candidate predictive motion vectors. It may be a candidate prediction motion vector that has the furthest distance from the 2 motion vectors.

According to an embodiment, when the first reference frame of the first motion vector is the same as the second reference frame of the plurality of candidate motion vectors, an x component of the plurality of candidate motion vectors and the first motion vector The distance can be identified using the y component.

According to an embodiment, when a first reference frame of the first motion vector is different from a second reference frame of the plurality of candidate predictive motion vectors, the second motion vector is the first one of the plurality of candidate predictive motion vectors. It may be a candidate prediction motion vector having the closest distance to the 2 motion vectors.

According to an embodiment, a first reference frame of the first motion vector is different from a second reference frame of the plurality of candidate predicted motion vectors, and a direction of the first reference frame and a direction of the second reference frame are the same. , the distance may be identified by scaling the plurality of candidate predictive motion vectors based on the Picture Order Count (POC) of the first reference frame, the POC of the plurality of candidate predictive motion vectors, and the POC of the current block. there is.

According to an embodiment, a first reference frame of the first motion vector is different from a second reference frame of the plurality of candidate predicted motion vectors, and a direction of the first reference frame and a direction of the second reference frame are different. , second candidate predicted motion vectors having a direction different from that of the first motion vector among the plurality of candidate motion vectors are determined, the POC of the first reference frame, the POC of the second candidate motion vectors, and the distance may be identified by scaling the second candidate predictive motion vectors based on the POC of the current block.

According to an embodiment, when the reference frames of the plurality of candidate motion vectors are a plurality of second reference frames, the POC of the first reference frame, the POC of the second candidate prediction motion vectors, and the POC of the current block The distance may be identified by scaling the second candidate predictive motion vectors based on .

According to an embodiment, when the reference frames of the plurality of candidate predicted motion vectors are a plurality of second reference frames, a distance closest to the first reference frame of the first motion vector among the plurality of second reference frames A third reference frame may be determined, and the second motion vector may be determined based on a distance between a plurality of third candidate predicted motion vectors included in the third reference frame and the first motion vector.

According to an embodiment, when the reference frames of the plurality of candidate predicted motion vectors are a plurality of second reference frames, a distance closest to the first reference frame of the first motion vector among the plurality of second reference frames When a third reference frame is determined and the third candidate motion vector included in the third reference frame is one, the third candidate motion vector motion vector may be determined as the second motion vector.

A video encoding apparatus according to an embodiment of the present disclosure performs coding for a current block by determining a second motion vector according to a relationship between a first reference frame and a second reference frame without generating information on the second motion vector. Efficiency can be improved.

The device-readable storage medium may be provided in the form of a non-transitory storage medium. Here, 'non-temporary storage medium' only means that it is a tangible device and does not contain signals (e.g., electromagnetic waves), and this term refers to the case where data is stored semi-permanently in the storage medium and temporary It does not discriminate if it is saved as . For example, a 'non-temporary storage medium' may include a buffer in which data is temporarily stored.

According to one embodiment, the method according to various embodiments disclosed in this document may be provided by being included in a computer program product. Computer program products may be traded between sellers and buyers as commodities. A computer program product is distributed in the form of a device-readable storage medium (eg compact disc read only memory (CD-ROM)), or through an application store or between two user devices (eg smartphones). It can be distributed (e.g., downloaded or uploaded) directly or online. In the case of online distribution, at least a part of a computer program product (eg, a downloadable app) is stored on a device-readable storage medium such as a memory of a manufacturer's server, an application store server, or a relay server. It can be temporarily stored or created temporarily.

Claims (15)

obtaining information on a first motion vector of a current block from a bitstream (S610);
determining the first motion vector based on information on the first motion vector (S630);
determining a candidate list including a plurality of candidate predictive motion vectors for determining a second motion vector of the current block (S650);
determining one of the plurality of candidate motion vectors as the second motion vector based on the distance between the plurality of candidate motion vectors and the first motion vector (S670); and
and determining a motion vector of the current block by using the first motion vector and the second motion vector (S690). According to claim 1,
When the first reference frame of the first motion vector is the same as the second reference frame of the plurality of candidate predicted motion vectors,
The second motion vector is a candidate prediction motion vector having the furthest distance from the second motion vector among the plurality of candidate motion vectors. According to claim 1 or 2,
When the first reference frame of the first motion vector is the same as the second reference frame of the plurality of candidate predicted motion vectors,
The video decoding method of claim 1 , wherein the distance is identified using an x component of the plurality of candidate motion vectors and a y component of the first motion vector. According to any one of claims 1 to 3,
When a first reference frame of the first motion vector is different from a second reference frame of the plurality of candidate predicted motion vectors,
The second motion vector is a candidate prediction motion vector having the closest distance to the second motion vector among the plurality of candidate motion vectors. According to any one of claims 1 to 4,
When a first reference frame of the first motion vector is different from a second reference frame of the plurality of candidate predicted motion vectors, and directions of the first reference frame and directions of the second reference frame are the same,
video decoding, wherein the distance is identified by scaling the plurality of candidate predictive motion vectors based on a Picture Order Count (POC) of the first reference frame, the POC of the plurality of candidate predictive motion vectors, and the POC of the current block. method. According to any one of claims 1 to 5,
When a first reference frame of the first motion vector is different from a second reference frame of the plurality of candidate predicted motion vectors, and a direction of the first reference frame and a direction of the second reference frame are different,
Among the plurality of candidate motion vectors, second candidate prediction motion vectors having a direction different from that of the first motion vector are determined;
and the distance is identified by scaling the second candidate predictive motion vectors based on the POC of the first reference frame, the POC of the second candidate predictor motion vectors, and the POC of the current block. According to any one of claims 1 to 6,
When the reference frames of the plurality of candidate predictive motion vectors are a plurality of second reference frames,
and the distance is identified by scaling the second candidate predictive motion vectors based on the POC of the first reference frame, the POC of the second candidate predictor motion vectors, and the POC of the current block. According to any one of claims 1 to 7,
When the reference frames of the plurality of candidate predicted motion vectors are a plurality of second reference frames,
Among the plurality of second reference frames, a third reference frame having the closest distance from the first reference frame of the first motion vector is determined;
Wherein the second motion vector is determined based on a distance between a plurality of third candidate predicted motion vectors included in the third reference frame and the first motion vector. According to any one of claims 1 to 8,
When the reference frames of the plurality of candidate predicted motion vectors are a plurality of second reference frames,
Among the plurality of second reference frames, a third reference frame having the closest distance from the first reference frame of the first motion vector is determined;
When the third candidate predicted motion vector included in the third reference frame is one,
The video decoding method of claim 1 , wherein the third candidate predicted motion vector is determined as the second motion vector. a memory that stores one or more instructions; and
at least one processor operating in accordance with the one or more instructions;
The at least one processor,
obtaining information on a first motion vector of a current block from a bitstream;
determining the first motion vector based on information on the first motion vector;
determining a candidate list including a plurality of candidate predictive motion vectors for determining a second motion vector of the current block;
determining one of the plurality of candidate motion vectors as the second motion vector based on a distance between the plurality of candidate motion vectors and the first motion vector;
and determining a motion vector of the current block by using the first motion vector and the second motion vector. According to claim 10,
When the first reference frame of the first motion vector is the same as the second reference frame of the plurality of candidate predicted motion vectors,
The second motion vector is a candidate prediction motion vector having the furthest distance from the second motion vector among the plurality of candidate motion vectors. According to claim 10 or 11,
When a first reference frame of the first motion vector is different from a second reference frame of the plurality of candidate predicted motion vectors,
The second motion vector is a candidate predictive motion vector having the closest distance to the second motion vector among the plurality of candidate predictive motion vectors. Generating information on a first motion vector of a current block (S810);
determining the first motion vector based on information on the first motion vector (S830);
determining a candidate list including a plurality of candidate predictive motion vectors for determining a second motion vector of the current block (S850);
determining one of the plurality of candidate motion vectors as the second motion vector based on the distance between the plurality of candidate motion vectors and the first motion vector (S870); and
and encoding information on the first motion vector by determining a motion vector of the current block using the first motion vector and the second motion vector (S890). According to claim 13,
When the first reference frame of the first motion vector is the same as the second reference frame of the plurality of candidate predicted motion vectors,
The video encoding method of claim 1 , wherein the second motion vector is a candidate prediction motion vector having the furthest distance from the second motion vector among the plurality of candidate motion vectors. The method of claim 13 or 14,
When a first reference frame of the first motion vector is different from a second reference frame of the plurality of candidate predicted motion vectors,
The second motion vector is a candidate predictive motion vector having the closest distance to the second motion vector among the plurality of candidate predictive motion vectors.

Images