KR20210118768A - Method of processing a video and device therefor - Google Patents

Abstract

An image decoding method according to the present disclosure includes determining whether to apply an adaptive loop filter to a current block, and when the adaptive loop filter is applied, whether the adaptive loop filter has linearity or nonlinearity determining whether or not there is, determining a filter coefficient set of the adaptive loop filter, and filtering reconstructed samples in the current block by using the filter coefficient set.

Description

Video signal processing method and apparatus

The present disclosure relates to a video signal processing method and apparatus.

Recently, demand for high-resolution and high-quality images such as HD (High Definition) images and UHD (Ultra High Definition) images is increasing in various application fields. As the image data becomes higher resolution and higher quality, the amount of data relatively increases compared to the existing image data. storage costs will increase. High-efficiency image compression techniques can be used to solve these problems that occur as image data becomes high-resolution and high-quality.

Inter-screen prediction technology that predicts pixel values included in the current picture from pictures before or after the current picture with image compression technology, intra-picture prediction technology that predicts pixel values included in the current picture using pixel information in the current picture, Various techniques exist, such as entropy encoding technology in which a short code is assigned to a value with a high frequency of occurrence and a long code is assigned to a value with a low frequency of occurrence.

Meanwhile, as the demand for high-resolution images increases, the demand for stereoscopic image content as a new image service is also increasing. A video compression technique for effectively providing high-resolution and ultra-high-resolution stereoscopic image content is being discussed.

An object of the present disclosure is to provide a method and an apparatus for applying an in-loop filter in encoding/decoding a video signal.

An object of the present disclosure is to provide a method and an apparatus for applying a nonlinear adaptive loop filter in encoding/decoding a video signal.

An object of the present disclosure is to provide a method of applying various adaptive loop filters in encoding/decoding a video signal, and a method for determining filter coefficients under each adaptive loop filter application method and an apparatus therefor. .

The technical problems to be achieved in the present disclosure are not limited to the technical problems mentioned above, and other technical problems not mentioned will be clearly understood by those of ordinary skill in the art to which the present disclosure belongs from the description below. will be able

A video signal decoding method according to the present disclosure includes determining whether to apply an adaptive loop filter to a current block, and determining whether the adaptive loop filter has linearity or nonlinearity when the adaptive loop filter is applied. determining whether or not there is a filter coefficient set of the adaptive loop filter, and filtering reconstructed samples in the current block by using the filter coefficient set.

A video signal encoding method according to the present disclosure includes determining whether to apply an adaptive loop filter to a current block, and determining whether the adaptive loop filter has linearity or nonlinearity when the adaptive loop filter is applied. determining whether or not there is, determining a filter coefficient set of the adaptive loop filter, and filtering reconstructed samples in the current block by using the filter coefficient set.

In the video signal decoding method according to the present disclosure, when it is determined that the adaptive loop filter has nonlinearity, the filtering is performed by applying a clipping function to a difference between the reconstructed sample and a neighboring reconstructed sample, A clipping value for applying a clipping function may be determined by an index specifying one of a plurality of clipping value candidates.

In the video signal decoding method according to the present disclosure, the index may be determined by a 1-bit flag decoded from a bitstream.

In the video signal decoding method according to the present disclosure, the number of clipping value candidates may be adaptively determined based on at least one of a color component, a color format, and a bit depth.

In the video signal decoding method according to the present disclosure, the plurality of clipping value candidates may be derived from one of a plurality of lookup tables defining a mapping relationship between bit depth and clipping value candidates.

In the video signal decoding method according to the present disclosure, one of the plurality of lookup tables may be selected by index information parsed from a bitstream.

In the video signal decoding method according to the present disclosure, the flag may be decoded only for positions to which a predefined filter coefficient is applied among the reconstructed samples in the current block.

In the video signal decoding method according to the present disclosure, the clipping function may be applied only at a position where the predefined filter coefficient is applied.

The features briefly summarized above with respect to the present disclosure are merely exemplary aspects of the detailed description of the present disclosure that follows, and do not limit the scope of the present disclosure.

According to the present disclosure, by applying the in-loop filter, it is possible to reduce the restoration error.

According to the present disclosure, by providing a method of applying a nonlinear adaptive loop filter, it is possible to reduce a reconstruction error.

According to the present disclosure, while various methods of applying an adaptive loop filter are provided, a method of determining filter coefficients under each method of applying an adaptive loop filter is provided.

Effects obtainable in the present disclosure are not limited to the above-mentioned effects, and other effects not mentioned may be clearly understood by those of ordinary skill in the art to which the present disclosure belongs from the description below. will be.

1 is a block diagram illustrating an image encoding apparatus according to an embodiment of the present disclosure.
2 is a block diagram illustrating an image decoding apparatus according to an embodiment of the present disclosure.
3 is a flowchart of a method of applying an adaptive loop filter according to an embodiment of the present disclosure.
4 shows an example in which the size of a diamond-shaped filter is defined.
5 shows an example in which a rectangular adaptive loop filter is applied.
6 is a diagram illustrating a block and filter coefficient derivation region.
7 is a diagram illustrating an example in which the number of samples used to derive a block gradient is reduced.
8 shows an example of geometric transformation types.
9 shows an example in which a reconstructed image is divided into a plurality of regions.
10 is an exemplary diagram for explaining an application aspect of the filter when the filter deviates from the boundary of the region.
11 shows an example in which indexes are coded for only some of filter coefficients.
12 is a diagram for explaining an example in which coefficient merging is applied under a block-based filter application method.
13 illustrates an example in which encoding of index information is omitted for some classes.
14 is a diagram for explaining an example in which coefficient merging is applied under a method of applying a region-based filter.

Since the present disclosure can make various changes and can have various embodiments, specific embodiments are illustrated in the drawings and described in detail in the detailed description. However, this is not intended to limit the present disclosure to specific embodiments, and should be understood to include all modifications, equivalents and substitutes included in the spirit and scope of the present disclosure. In describing each figure, like reference numerals have been used for like elements.

Terms such as first, second, etc. may be used to describe various elements, but the elements should not be limited by the terms. The above terms are used only for the purpose of distinguishing one component from another. For example, without departing from the scope of the present disclosure, a first component may be referred to as a second component, and similarly, a second component may also be referred to as a first component. and/or includes a combination of a plurality of related listed items or any of a plurality of related listed items.

When a component is referred to as being “connected” or “connected” to another component, it is understood that the other component may be directly connected or connected to the other component, but other components may exist in between. it should be On the other hand, when it is said that a certain element is "directly connected" or "directly connected" to another element, it should be understood that no other element is present in the middle.

The terms used in the present application are used only to describe specific embodiments, and are not intended to limit the present disclosure. The singular expression includes the plural expression unless the context clearly dictates otherwise. In the present application, terms such as “comprise” or “have” are intended to designate that a feature, number, step, operation, component, part, or combination thereof described in the specification exists, but one or more other features It should be understood that this does not preclude the existence or addition of numbers, steps, operations, components, parts, or combinations thereof.

Hereinafter, preferred embodiments of the present disclosure will be described in more detail with reference to the accompanying drawings. Hereinafter, the same reference numerals are used for the same components in the drawings, and repeated descriptions of the same components are omitted.

1 is a block diagram illustrating an image encoding apparatus according to an embodiment of the present disclosure.

Referring to FIG. 1 , the image encoding apparatus 100 includes a picture division unit 110 , prediction units 120 and 125 , a transform unit 130 , a quantization unit 135 , a rearrangement unit 160 , and an entropy encoding unit ( 165 ), an inverse quantization unit 140 , an inverse transform unit 145 , a filter unit 150 , and a memory 155 .

Each of the constituent units shown in FIG. 1 is independently illustrated to represent different characteristic functions in the image encoding apparatus, and does not mean that each constituent unit is composed of separate hardware or one software constituent unit. That is, each component is listed as each component for convenience of description, and at least two components of each component are combined to form one component, or one component can be divided into a plurality of components to perform a function, and each Integrated embodiments and separate embodiments of the components are also included in the scope of the present disclosure without departing from the essence of the present disclosure.

In addition, some components are not essential components to perform an essential function in the present disclosure, but may be optional components for merely improving performance. The present disclosure may be implemented by including only essential components to implement the essence of the present disclosure, except for components used for performance improvement, and a structure including only essential components excluding optional components used for performance improvement Also included in the scope of the present disclosure.

The picture divider 110 may divide the input picture into at least one processing unit. In this case, the processing unit may be a prediction unit (PU), a transform unit (TU), or a coding unit (CU). The picture splitter 110 divides one picture into a combination of a plurality of coding units, prediction units, and transformation units, and combines one coding unit, prediction unit, and transformation unit based on a predetermined criterion (eg, a cost function). can be selected to encode the picture.

For example, one picture may be divided into a plurality of coding units. In order to split a coding unit in a picture, a recursive tree structure such as a quad tree, a ternary tree, or a binary tree may be used. A coding unit divided into other coding units with the coding unit as a root may be divided having as many child nodes as the number of the divided coding units. A coding unit that is no longer split according to certain restrictions becomes a leaf node. For example, when it is assumed that quad tree splitting is applied to one coding unit, one coding unit may be split into up to four different coding units.

Hereinafter, in an embodiment of the present disclosure, a coding unit may be used as a unit for performing encoding or may be used as a meaning for a unit for performing decoding.

A prediction unit may be split in the form of at least one square or rectangle of the same size within one coding unit, and one prediction unit among the split prediction units within one coding unit is a prediction of another. It may be divided to have a shape and/or size different from that of the unit.

In intra prediction, the transformation unit and the prediction unit may be set to be the same. In this case, after dividing the coding unit into a plurality of transform units, intra prediction may be performed for each transform unit. A coding unit may be divided in a horizontal direction or a vertical direction. The number of transformation units generated by dividing the coding unit may be 2 or 4 according to the size of the coding unit.

The prediction units 120 and 125 may include an inter prediction unit 120 performing inter prediction and an intra prediction unit 125 performing intra prediction. Whether to use inter prediction or to perform intra prediction on a coding unit may be determined, and specific information (eg, intra prediction mode, motion vector, reference picture, etc.) according to each prediction method may be determined. In this case, a processing unit in which prediction is performed and a processing unit in which a prediction method and specific content are determined may be different. For example, a prediction method and a prediction mode may be determined in a coding unit, and prediction may be performed in a prediction unit or a transformation unit. A residual value (residual block) between the generated prediction block and the original block may be input to the transform unit 130 . Also, prediction mode information, motion vector information, etc. used for prediction may be encoded by the entropy encoder 165 together with the residual value and transmitted to the decoding apparatus. When a specific encoding mode is used, it is also possible to encode the original block as it is without generating the prediction block through the prediction units 120 and 125 and transmit it to the decoder.

The inter prediction unit 120 may predict a prediction unit based on information on at least one of a picture before or after a picture of the current picture, and in some cases, prediction based on information of a partial region in the current picture that has been encoded Units can also be predicted. The inter prediction unit 120 may include a reference picture interpolator, a motion prediction unit, and a motion compensator.

The reference picture interpolator may receive reference picture information from the memory 155 and generate pixel information of integer pixels or less in the reference picture. In the case of luminance pixels, a DCT-based 8-tap interpolation filter (DCT-based Interpolation Filter) with different filter coefficients may be used to generate pixel information of integer pixels or less in units of 1/4 pixels. In the case of the color difference signal, a DCT-based 4-tap interpolation filter in which filter coefficients are different to generate pixel information of integer pixels or less in units of 1/8 pixels may be used.

The motion prediction unit may perform motion prediction based on the reference picture interpolated by the reference picture interpolator. As a method for calculating the motion vector, various methods such as Full search-based Block Matching Algorithm (FBMA), Three Step Search (TSS), and New Three-Step Search Algorithm (NTS) may be used. The motion vector may have a motion vector value of 1/2 or 1/4 pixel unit based on the interpolated pixel. The motion prediction unit may predict the current prediction unit by using a different motion prediction method. Various methods, such as a skip method, a merge method, an AMVP (Advanced Motion Vector Prediction) method, an intra block copy method, etc., may be used as the motion prediction method.

The intra prediction unit 125 may generate a prediction block based on reference pixel information that is pixel information in the current picture. Reference pixel information may be derived from a selected one of a plurality of reference pixel lines. An Nth reference pixel line among the plurality of reference pixel lines may include left pixels having an x-axis difference of N from an upper-left pixel in the current block and upper pixels having an y-axis difference of N from the upper-left pixel. The number of reference pixel lines that the current block can select may be one, two, three, or four.

When a neighboring block of the current prediction unit is a block on which inter prediction is performed, and thus a reference pixel is a pixel on which inter prediction is performed, a reference pixel included in the block on which inter prediction is performed is a reference pixel of the block on which intra prediction has been performed. information can be used instead. That is, when the reference pixel is not available, the unavailable reference pixel information may be replaced with information of at least one of the available reference pixels.

In intra prediction, the prediction mode may have a directional prediction mode in which reference pixel information is used according to a prediction direction and a non-directional mode in which directional information is not used when prediction is performed. A mode for predicting luminance information and a mode for predicting chrominance information may be different, and intra prediction mode information used for predicting luminance information or predicted luminance signal information may be utilized to predict chrominance information.

When intra prediction is performed, if the size of the prediction unit and the size of the transformation unit are the same, intra prediction for the prediction unit based on the pixel present on the left side, the pixel present on the upper left side, and the pixel present on the upper side of the prediction unit can be performed.

The intra prediction method may generate a prediction block after applying a smoothing filter to a reference pixel according to a prediction mode. Whether to apply the smoothing filter may be determined according to the selected reference pixel line.

In order to perform the intra prediction method, the intra prediction mode of the current prediction unit may be predicted from the intra prediction mode of the prediction unit existing around the current prediction unit. When the prediction mode of the current prediction unit is predicted using the mode information predicted from the neighboring prediction unit, if the intra prediction mode of the current prediction unit and the neighboring prediction unit are the same, the current prediction unit and the neighboring prediction unit are used using predetermined flag information It is possible to transmit information that the prediction modes of . , and if the prediction modes of the current prediction unit and the neighboring prediction units are different from each other, entropy encoding may be performed to encode prediction mode information of the current block.

In addition, a residual block including residual information, which is a difference value between a prediction unit and an original block of the prediction unit, in which prediction is performed based on the prediction unit generated by the prediction units 120 and 125 may be generated. The generated residual block may be input to the transform unit 130 .

The transform unit 130 converts the original block and the residual block including residual information of the prediction unit generated by the prediction units 120 and 125 to DCT (Discrete Cosine Transform), DST (Discrete Sine Transform), KLT and It can be converted using the same conversion method. Whether to apply DCT, DST, or KLT to transform the residual block is based on at least one of the size of the transform unit, the shape of the transform unit, the prediction mode of the prediction unit, or the intra prediction mode information of the prediction unit. can decide

The quantizer 135 may quantize the values transformed by the transform unit 130 into the frequency domain. The quantization coefficient may change according to blocks or the importance of an image. The value calculated by the quantization unit 135 may be provided to the inverse quantization unit 140 and the rearrangement unit 160 .

The rearrangement unit 160 may rearrange the coefficient values on the quantized residual values.

The reordering unit 160 may change the two-dimensional block form coefficient into a one-dimensional vector form through a coefficient scanning method. For example, the rearranging unit 160 may use a Zig-Zag Scan method to scan from DC coefficients to coefficients in a high-frequency region and change them into a one-dimensional vector form. Depending on the size of the transform unit and the intra prediction mode, instead of a zig-zag scan, a vertical scan that scans a two-dimensional block shape coefficient in a column direction, a horizontal scan that scans a two-dimensional block shape coefficient in a row direction, or a two-dimensional block A diagonal scan that scans the shape coefficients in the diagonal direction may be used. That is, it may be determined whether any of the zig-zag scan, vertical scan, horizontal scan, or diagonal scan is to be used according to the size of the transform unit and the intra prediction mode.

The entropy encoding unit 165 may perform entropy encoding based on the values calculated by the reordering unit 160 . For entropy encoding, various encoding methods such as Exponential Golomb, Context-Adaptive Variable Length Coding (CAVLC), and Context-Adaptive Binary Arithmetic Coding (CABAC) may be used.

The entropy encoding unit 165 receives the residual value coefficient information and block type information, prediction mode information, division unit information, prediction unit information and transmission unit information, motion of the coding unit from the reordering unit 160 and the prediction units 120 and 125 . Various information such as vector information, reference frame information, interpolation information of a block, and filtering information may be encoded.

The entropy encoder 165 may entropy-encode the coefficient values of the coding units input from the reordering unit 160 .

The inverse quantizer 140 and the inverse transform unit 145 inversely quantize the values quantized by the quantizer 135 and inversely transform the values transformed by the transform unit 130 . The residual values generated by the inverse quantizer 140 and the inverse transform unit 145 are combined with the prediction units predicted through the motion estimation unit, the motion compensator, and the intra prediction unit included in the prediction units 120 and 125 and restored. You can create a Reconstructed Block.

The filter unit 150 may include at least one of a deblocking filter, an offset correcting unit, and an adaptive loop filter (ALF).

The deblocking filter may remove block distortion caused by the boundary between blocks in the reconstructed picture. In order to determine whether to perform deblocking, it may be determined whether to apply the deblocking filter to the current block based on pixels included in several columns or rows included in the block. When a deblocking filter is applied to a block, a strong filter or a weak filter can be applied according to the required deblocking filtering strength. In addition, in applying the deblocking filter, horizontal filtering and vertical filtering may be concurrently processed when performing vertical filtering and horizontal filtering.

The offset correcting unit may correct an offset from the original image in units of pixels with respect to the image on which the deblocking has been performed. In order to perform offset correction on a specific picture, a method of dividing pixels included in an image into a certain number of regions, determining the region to be offset and applying the offset to the region, or taking edge information of each pixel into consideration can be used to apply

Adaptive loop filtering (ALF) may be performed based on a value obtained by comparing the filtered reconstructed image and the original image. After dividing the pixels included in the image into a predetermined group, one filter to be applied to the corresponding group is determined, and filtering can be performed differentially for each group. As for information on whether to apply ALF, the luminance signal may be transmitted for each coding unit (CU), and the shape and filter coefficients of the ALF filter to be applied may vary according to each block. In addition, the ALF filter of the same type (fixed type) may be applied regardless of the characteristics of the target block.

The memory 155 may store the reconstructed block or picture calculated through the filter unit 150 , and the stored reconstructed block or picture may be provided to the predictors 120 and 125 when inter prediction is performed.

2 is a block diagram illustrating an image decoding apparatus according to an embodiment of the present disclosure.

Referring to FIG. 2 , the image decoding apparatus 200 includes an entropy decoding unit 210, a reordering unit 215, an inverse quantization unit 220, an inverse transform unit 225, prediction units 230 and 235, and a filter unit ( 240) and a memory 245 may be included.

When an image bitstream is input by the image encoding apparatus, the input bitstream may be decoded by a procedure opposite to that of the image encoding apparatus.

The entropy decoding unit 210 may perform entropy decoding in a procedure opposite to that performed by the entropy encoding unit of the image encoding apparatus. For example, various methods such as Exponential Golomb, Context-Adaptive Variable Length Coding (CAVLC), and Context-Adaptive Binary Arithmetic Coding (CABAC) may be applied corresponding to the method performed by the image encoding apparatus.

The entropy decoding unit 210 may decode information related to intra prediction and inter prediction performed by the encoding apparatus.

The reordering unit 215 may perform reordering based on a method of rearranging the entropy-decoded bitstream by the entropy decoding unit 210 by the encoder. Coefficients expressed in the form of a one-dimensional vector may be restored and rearranged as coefficients in the form of a two-dimensional block. The reordering unit 215 may receive information related to coefficient scanning performed by the encoder and perform the reordering by performing a reverse scanning method based on the scanning order performed by the corresponding encoder.

The inverse quantization unit 220 may perform inverse quantization based on the quantization parameter provided by the encoding apparatus and the reordered coefficient values of the blocks.

The inverse transform unit 225 may perform inverse transforms, ie, inverse DCT, inverse DST, and inverse KLT, on the transforms performed by the transform unit, ie, DCT, DST, and KLT, on the quantization result performed by the image encoding apparatus. Inverse transform may be performed based on a transmission unit determined by the image encoding apparatus. In the inverse transform unit 225 of the image decoding apparatus, a transformation technique (eg, DCT, DST, KLT) may be selectively performed according to a plurality of pieces of information such as a prediction method, a size, a shape of a current block, a prediction mode, and an intra prediction direction. can

The prediction units 230 and 235 may generate a prediction block based on the prediction block generation related information provided from the entropy decoding unit 210 and previously decoded block or picture information provided from the memory 245 .

As described above, when intra prediction is performed in the same manner as in the operation in the image encoding apparatus, when the size of the prediction unit and the size of the transformation unit are the same, the pixel present on the left side of the prediction unit, the pixel present on the upper left side, and the upper Intra prediction is performed on the prediction unit based on the existing pixel, but when the size of the prediction unit and the size of the transformation unit are different when performing intra prediction, intra prediction is performed using the reference pixel based on the transformation unit can do. Also, intra prediction using NxN splitting may be used only for the smallest coding unit.

The prediction units 230 and 235 may include a prediction unit determiner, an inter prediction unit, and an intra prediction unit. The prediction unit determining unit receives various information such as prediction unit information input from the entropy decoder 210, prediction mode information of the intra prediction method, and motion prediction related information of the inter prediction method, and divides the prediction unit from the current coding unit, and predicts It may be determined whether the unit performs inter prediction or intra prediction. The inter prediction unit 230 uses information required for inter prediction of the current prediction unit provided from the image encoding apparatus based on information included in at least one of a picture before or after the current picture including the current prediction unit. Inter prediction may be performed on the prediction unit. Alternatively, inter prediction may be performed based on information of a pre-restored partial region in the current picture including the current prediction unit.

In order to perform inter prediction, a motion prediction method of a prediction unit included in a corresponding coding unit based on a coding unit is selected from among skip mode, merge mode, AMVP mode, and intra block copy mode. You can decide which way to go.

The intra prediction unit 235 may generate a prediction block based on pixel information in the current picture. When the prediction unit is a prediction unit on which intra prediction is performed, intra prediction may be performed based on intra prediction mode information of the prediction unit provided by the image encoding apparatus. The intra prediction unit 235 may include an Adaptive Intra Smoothing (AIS) filter, a reference pixel interpolator, and a DC filter. The AIS filter is a part that performs filtering on the reference pixel of the current block, and may be applied by determining whether to apply the filter according to the prediction mode of the current prediction unit. AIS filtering may be performed on the reference pixel of the current block by using the prediction mode and AIS filter information of the prediction unit provided by the image encoding apparatus. When the prediction mode of the current block is a mode in which AIS filtering is not performed, the AIS filter may not be applied.

When the prediction mode of the prediction unit is a prediction unit in which intra prediction is performed based on a pixel value obtained by interpolating the reference pixel, the reference pixel interpolator may interpolate the reference pixel to generate a reference pixel of a pixel unit having an integer value or less. When the prediction mode of the current prediction unit is a prediction mode that generates a prediction block without interpolating the reference pixel, the reference pixel may not be interpolated. The DC filter may generate the prediction block through filtering when the prediction mode of the current block is the DC mode.

The reconstructed block or picture may be provided to the filter unit 240 . The filter unit 240 may include a deblocking filter, an offset correction unit, and an ALF.

Information on whether a deblocking filter is applied to a corresponding block or picture and information on whether a strong filter or a weak filter is applied when the deblocking filter is applied may be provided from the image encoding apparatus. The deblocking filter of the image decoding apparatus may receive deblocking filter-related information provided from the image encoding apparatus, and the image decoding apparatus may perform deblocking filtering on the corresponding block.

The offset correction unit may perform offset correction on the reconstructed image based on the type of offset correction applied to the image during encoding, information on the offset value, and the like.

ALF may be applied to a coding unit based on information on whether ALF is applied, ALF coefficient information, etc. provided from the encoding apparatus. Such ALF information may be provided by being included in a specific parameter set.

The memory 245 may store the reconstructed picture or block to be used as a reference picture or reference block, and may also provide the reconstructed picture to an output unit.

As described above, hereinafter, in the embodiments of the present disclosure, a coding unit is used as a term for a coding unit for convenience of description, but may also be a unit for performing decoding as well as coding.

In addition, the current block indicates a block to be encoded/decoded, and according to the encoding/decoding step, a coding tree block (or coding tree unit), a coding block (or a coding unit), a transform block (or a transform unit), and a prediction block (or a prediction unit) or a block to which an in-loop filter is applied. In this specification, a 'unit' may indicate a basic unit for performing a specific encoding/decoding process, and a 'block' may indicate a pixel array of a predetermined size. Unless otherwise specified, 'block' and 'unit' may be used interchangeably. For example, in the embodiments to be described below, it may be understood that the coding block (coding block) and the coding unit (coding unit) are mutually equivalent.

Under the lossy coding scheme, a restoration error occurs in the restoration area. In order to attenuate the above reconstruction error, an in-loop filter may be applied to the reconstruction block. Specifically, at least one of a deblocking filter, a sample adaptive offset filter (SAO) filter, and an adaptive loop filter may be applied to the reconstruction block.

In-loop filters may be applied according to a predefined order. As an example, the filters may be applied in the order of a deblocking filter, a sample adaptive offset filter, and an adaptive loop filter.

In this case, whether to apply each filter may be independently determined. For example, even if the deblocking filter or the sample adaptive offset filter is not applied, the adaptive loop filter may be applied to the reconstruction block.

Alternatively, it may be set to have dependencies between filters. As an example, the adaptive loop filter may be set to be applied only when the deblocking filter or the sample adaptive offset filter is applied to the reconstruction block.

An application aspect of the adaptive loop filter among the listed in-loop filters will be described in detail.

3 is a flowchart of a method of applying an adaptive loop filter according to an embodiment of the present disclosure.

The adaptive loop filter refers to a method of applying a polygonal type filter to a reconstruction block. Hereinafter, a block to be determined whether or not to apply the adaptive loop filter is referred to as a current block.

Referring to FIG. 3 , first, it may be determined whether or not to apply the adaptive loop filter to the current block ( S301 ). A flag indicating whether the adaptive loop filter is applied may be encoded and signaled. As an example, alf_ctb_flag indicates whether the adaptive loop filter is applied to the current block.

Whether to apply the adaptive loop filter may be determined in units of a coding tree, a coding unit, or a transform unit.

Alternatively, information indicating the size of a block (hereinafter, a reference block) for determining whether an adaptive loop filter is applied may be signaled through a bitstream. For example, the information may be signaled at a slice, picture, or sequence level. As an example, the information may indicate a difference between the size of the coding tree unit and the size of the reference block. For example, the difference information may indicate a difference between a value obtained by taking Log_2 as the size of a coding tree unit and a value obtained by taking Log_2 as the size of a reference block. For example, when the coding tree unit is 128x128 and the reference block is 32x32, the difference information _{may be set to 2 (Log 2} (132/32)).

Alternatively, the size of the mood block may be predefined in the encoder and the decoder. For example, the reference block may have a fixed size of 64x64.

Whether the adaptive loop filter is applied may be independently determined for each color component. Here, the color component represents a luminance (Luma) component or a color difference (Cb, Cr) component. For example, the alf_ctb_flag may be encoded and signaled for each color component.

Alternatively, whether to independently determine whether to apply the adaptive loop filter for each color component may be determined according to a color format. Here, the color format indicates a composition ratio of a luminance component and a chrominance component (eg, 4:4:4, 4:2:2, 4:2:0). For example, when the color format is 4:4:4, it is possible to independently determine whether to apply the adaptive loop filter to each of the luminance component and the color component. On the other hand, when the color format is 4:2:2 or 4:2:0, the result of determining whether to apply the adaptive loop filter to the luminance component may be directly applied to the chrominance component.

When it is decided to apply the adaptive loop filter, a filter to be applied to the current block may be determined (S302). Determining the filter refers to determining at least one of a filter type, filter coefficients, or filter size.

The filter shape may be a polygonal shape. For example, the adaptive loop filter may have a square shape, a non-square shape, a diamond shape, a trapezoid shape, a rhombus shape, or a cross shape.

The form of the adaptive loop filter may be predefined in the encoder and the decoder. Alternatively, information for specifying the type of the adaptive loop filter may be signaled through the bitstream. For example, index information indicating one of a plurality of filter types may be signaled through a bitstream.

Alternatively, the filter type may be determined based on at least one of a color component, a color format, or a bit depth.

For example, the diamond filter may be fixedly applied to both the luminance component and the chrominance component. Alternatively, when the color format is 4:4:4 or 4:2:0, diamond-shaped filters are applied to both the luminance component and the chrominance component, and when the color format is 4:2:2, diamond is applied to the luminance component. A type filter may be applied, and a non-diamond type filter (eg, a square filter) may be applied to the color difference component.

In the encoder, filter coefficients that minimize the reconstruction error can be derived. In addition, information on filter coefficients indicating optimal efficiency may be encoded and signaled to a decoder.

The size of the filter may be defined by at least one of the number of taps in the horizontal direction and the number of taps in the vertical direction of the filter.

4 shows an example in which the size of a diamond-shaped filter is defined.

The size of the diamond-shaped filter may be defined as a maximum length (maximum number of taps) in a horizontal direction and a maximum length (maximum number of taps) in a vertical direction. For example, a diamond filter having a size of 7x7 indicates that the maximum length in the horizontal direction and the maximum length in the vertical direction are 7, respectively.

In FIG. 4 , it is illustrated that the maximum length in the horizontal direction and the maximum length in the vertical direction of the diamond-shaped filter are the same. Unlike the illustrated example, a diamond-shaped filter having a different maximum length in a horizontal direction and a maximum length in a vertical direction may be used.

A filter size may be set differently for each color component. For example, a diamond-shaped filter having a size of 7x7 may be used for the luminance component and a diamond-shaped filter having a size of 5x5 may be used for the chrominance component.

Alternatively, the size of the filter applied to the color difference component may be determined in consideration of the color format. For example, when the color format is 4:2:0, the diamond filter of 5x5 size shown in FIG. 4B may be applied to the color difference component. On the other hand, when the color format is 4:4:4, a filter having the same size as the luminance component, for example, a 7x7 diamond filter shown in FIG. 4A may be applied to the chrominance component.

Even when the color format is 4:2:2, a filter having the same size as the luminance component may be applied to the chrominance component. Alternatively, when the color format is 4:2:2, a filter having at least one of a maximum length in a horizontal direction and a maximum length in a vertical direction shorter than a filter applied to the luminance component may be applied to the color difference component.

As in the described example, whether to apply the same size filter to the luminance component and the chrominance component may be determined according to the color format.

When the filter is determined, the reconstructed sample may be filtered using the determined filter (S303). Specifically, the reconstructed sample may be filtered using reconstructed samples around the reconstructed sample. In this case, the position and the number of neighboring reconstructed samples may be determined by the filter type.

5 shows an example in which a rectangular adaptive loop filter is applied.

In the example of FIG. 5 , I(x, y) represents a value of a sample before the adaptive loop filter is applied, and O(x, y) represents a value of a sample after the loop filter is applied. Equation 1 below represents the application aspect of the adaptive loop filter shown in FIG. 5 as an equation.

When whether or not the adaptive loop filter is applied in units of coding tree units, in Equation 1, (x, y) represents the coordinates of the samples in the coding tree unit, and (i, j) represents the coordinates according to the filter shape. indicates. For example, assuming that the central position of the filter is (0, 0), values of i and j may be determined. For example, as in the example shown in FIG. 5 , when a rectangular filter having a size of 3x3 is applied, i and j may be set to integers between -1 and 1, respectively. Depending on the type of filter, ranges of values of i and j may be different.

)은 1로 설정될 수 있다. (

)는 중앙 위치의 필터 계수(w(0, 0))를 나타낸다. O(x, y)는 필터 처리된 샘플 값을 나타내고, I(x, y)는 입력 값(즉, 필터 적용 전 샘플 값)을 나타낸다. S는 적응적 루프 필터의 적용 영역을 나타낸다. In Equation 1, the sum of filter coefficients (eg,

) may be set to 1. (

) represents the filter coefficient (w(0, 0)) of the central position. O(x, y) represents the filtered sample value, and I(x, y) represents the input value (ie, the sample value before applying the filter). S represents the application area of the adaptive loop filter.

In FIG. 4 , filter coefficients w(i,j) are expressed as a one-dimensional variable f[n]. For example, in FIG. 4(a), f[12] marked at the filter center position _{represents w (0, 0)} , and in FIG. 4(b), f[6] marked at the filter center position is w _{(0, 0)} is represented.

As in the example shown in FIG. 4 , one filter coefficient may be applied to a plurality of positions. As an example, filter coefficients in the filter may be distributed in a symmetrical form. It is possible to derive the filter coefficients only for some, but not all, positions in the filter. Using the symmetry of the filter, the filter coefficients of the residual position can be derived.

When applying the adaptive loop filter, either a block-based filter application method or a region-based filter application method may be applied.

Which one of the above two filter application methods is to be used may be predefined in the encoder and/or the decoder. Alternatively, information specifying one of the above two filter application methods may be encoded and signaled. For example, the information may be encoded through a higher header such as a slice, a picture, or a sequence.

Alternatively, the information may be encoded for each block of a predetermined size (eg, for each coding tree unit, for each coding block, for each transform block, or for each reference block). In this case, a different filter application method may be applied to each block of a predetermined size.

Alternatively, the filter application method may be adaptively determined according to whether another in-loop filter is applied. As an example, the filter application method may be determined based on at least one of whether a deblocking filter is applied or whether a sample adaptive offset filter is applied.

When the block-based filter application method is applied, filter coefficients may be determined in units of blocks of a predetermined size. As an example, filter coefficients may be determined in units of blocks having a predefined size (eg, 4x4) in the coding tree unit. In this case, the filter coefficients of each block may be determined based on a filter coefficient set corresponding to any one of the plurality of class candidates. When the block-based filter application method is applied, a class for identifying a filter coefficient set may be obtained for each block of a predetermined size.

On the other hand, when the region-based filter application method is applied, filter coefficients may be determined for each region. Here, the region may represent a coding tree unit. In this case, one set of filter coefficients may be applied to the coding tree unit.

Alternatively, the region may be defined as a tile, a slice, a sub picture, or the like, or may be defined as a processing unit different from the above-described processing units. In this case, one filter coefficient set may be applied to a region, and the filter coefficient set may be commonly applied to a plurality of blocks included in one region.

Each filter application method will be described in detail.

When the block-based filter application method is applied, the reconstructed image may be divided into a plurality of blocks.

In this case, information indicating the size of the block may be signaled through the bitstream. The information may include at least one of a width or a height of a block. When the block is a square, only one of the block width information and the block height information may be encoded and signaled. On the other hand, when the block is non-square, width maintenance of the block and height information of the block may be encoded and signaled, respectively. The information may be encoded and signaled at a slice, picture, or sequence level.

The block type may be predefined in the encoder and the decoder. Alternatively, information indicating the shape of a block may be encoded and signaled. For example, a flag indicating whether the shape of the block is a square or a non-square may be encoded and signaled.

Alternatively, at least one of a size or a shape of a block may be determined based on at least one of whether a deblocking filter is applied or whether a sample adaptive offset filter is applied.

Alternatively, at least one of the block shape and size may be predefined in the encoder and the decoder. For example, in order to derive the filter coefficients, the reconstructed image may be set to be divided into 4x4 blocks.

After dividing the reconstructed image into a plurality of blocks, a filter coefficient derivation region may be set around each block. The filter coefficient derivation region may be composed of a block and N lines around the block. Here, a line represents a row or column.

For example, when the size of the block is 4x4, the filter coefficient derivation region may be defined with the size of 8x8. Here, the filter coefficient derivation region may include a block and two lines around each boundary of the block.

Information on the size of the filter coefficient derivation region may be signaled through a bitstream. For example, the information may indicate the number (N) of lines surrounding the block included in the filter coefficient derivation region.

Alternatively, based on at least one of a color component, a color format, or a block size, the number N of lines around the block may be adaptively determined.

6 is a diagram illustrating a block and filter coefficient derivation region.

In FIG. 6 , the filter coefficient derivation region is illustrated as including a 4×4 block and two lines adjacent to each of the boundaries of the block. In the embodiment to be described later, it is assumed that the size of the block is 4x4 and the size of the filter coefficient derivation region is 8x8, as in the illustrated example.

When the filter coefficient derivation region is determined, a gradient of a block within the filter coefficient derivation region may be calculated. Specifically, the block inclination in at least one of a horizontal direction, a vertical direction, an upper right diagonal direction, or an upper left diagonal direction may be calculated.

Equation 2 below shows an example of calculating the gradient in the filter coefficient derivation region.

In Equation 2, g _v denotes a vertical block gradient, g _h denotes a horizontal block gradient, g _d1 denotes an upper right diagonal block gradient, and g _d2 denotes an upper left diagonal block gradient.

As in the example shown in Equation 2, the block gradient may be derived by summing the gradients of each of the samples in the filter coefficient derivation region. Here, the sample gradient may be derived based on the reconstructed sample and adjacent samples of the reconstructed sample. Here, the positions of adjacent samples may be determined by the desired gradient direction. As an example, the sample gradient with respect to the vertical direction is a reconstructed sample (R(k, l) in Equation 2), a lower neighbor reconstructed sample (R(k, l-1) in Equation 2), and an upper neighbor reconstructed sample ( It can be derived based on R(k, l+1)) in Equation 2). The sample gradient with respect to the horizontal direction is a reconstructed sample (R(k, l) in Equation 2), a left-neighbor reconstructed sample (R(k-1, l) in Equation 2), and a right-neighbor reconstructed sample (Equation 2). It can be derived based on R(k+1, l)) of

In Equation 2, (i, j) represents the position of the upper left sample in the block, and R(k, l) represents the restored sample at the (k, l) position. |x| represents the absolute value of x.

Thereafter, a class of the block may be determined based on the block gradient. One of the plurality of class candidates may be selected as the class of the block.

The number of class candidates may be predefined in the encoder and decoder. Alternatively, information indicating the number of class candidates may be encoded and signaled.

The class of the block may be determined based on at least one of a maximum value, a minimum value, a median value, or an average value thereof among block gradients in a plurality of directions. As an example, it is possible to determine whether a block slope is large or small with respect to a horizontal block slope and a vertical direction block slope, and large or small with respect to an upper right diagonal block slope and an upper left diagonal block slope may be determined. Thereafter, the class of the block may be determined with reference to the above determination result. Hereinafter, for convenience of description, a horizontal direction and a vertical direction will be referred to as a first group direction, and an upper right diagonal direction and an upper left diagonal direction will be referred to as a second group direction.

Equation 3 below shows an example of determining the minimum and maximum values for the inclinations in the first group direction.

g ^max _h,v represents a larger value among g _h and g _v ^{, and g min} _h,v represents a smaller value among g _h and g _v.

In addition, by comparing the horizontal block gradient and the vertical block gradient, a variable DirHV indicating a tendency toward the first group direction may be derived. For example, when the vertical block gradient g _v is greater than the horizontal block gradient g _h , the variable DirHV may be set to 1. On the other hand, when the horizontal block gradient g _h is greater than the vertical block gradient g _v , the variable DirHV may be set to 3 .

Equation 4 below shows an example of determining the minimum and maximum values for the inclinations in the second group direction.

g ^max _d0,d1 represents a larger value among g _d0 and g _d1 ^{, and g min} _d0,d1 represents a smaller value among g _d0 and g _{d1 .}

In addition, by comparing the horizontal block gradient and the vertical block gradient, a variable DirD indicating a tendency toward the second group direction may be derived. For example, when the upper right diagonal block gradient gd0 is _{greater than the horizontal block gradient g h} , the variable DirHV may be set to 1. On the other hand, when the horizontal block gradient g _h is greater than the vertical block gradient g _v , the variable DirHV may be set to 3 .

Using the minimum (eg g ^min _h,v , g ^min _d0,d1 ) and maximum (eg g ^min _h,v , g ^min _d0,d1 ) derived using the respective block gradients, the class of the block Variables can be derived to determine

Equation 5 shows an example of deriving variables for determining the class of the block.

As illustrated in Equation 5 above, a value obtained by multiplying the maximum value among the gradients in the first group direction and the minimum value among the gradients in the second group direction, and the minimum value among the gradients in the first group direction and the maximum value among the gradients in the second group direction By comparing the multiplied values, the variables hvd1, hvd0, dir1 and dir2 can be derived. In this case, the variables hvd0 and hvd1 are variables set equal to one of the inclinations of the block, and the variables dir1 and dir2 are variables set equal to the tendency of the block.

Using the variables derived through Equation 5, it is possible to derive a variable dirS indicating the direction of a tendency that appears strongly throughout the block. As an example, the variable dirS may be derived based on Equation 6 below.

When the variable dirS is derived, the index ClassIdx indicating the class of the block can be derived using this. As an example, the variable ClassIdx may be derived based on Equation 7 below.

In Equation 7, Q[g _h +g _v ] represents a value derived by quantizing the sum of the horizontal gradient g _h and the vertical gradient g _{v .} As an example, Q[g _h +g _v ] may be output as a value between 0 and 4.

The variable ClassIdx may indicate one of a plurality of class candidates. For example, when the class of the block indicates one of 25 class candidates, the variable ClassIdx may be set to a value between 0 and 24.

In order to simplify the determination of the class of the block, the range (eg, the number) of samples used when the gradient of the block is derived may be adjusted. As an example, the number of samples used in deriving the block gradient may be reduced by reducing the number N of lines around the sub-sampling or block boundary region.

7 is a diagram illustrating an example in which the number of samples used to derive a block gradient is reduced.

In FIG. 7 , samples marked with markers indicate samples obtained through subsampling. Only samples marked with a marker can be used to derive the block slope. That is, the block gradient may be derived by adding the gradients of samples marked with the marker.

For example, as in the example shown in (a) of FIG. 7 , the block gradient using only samples in which both the x and y coordinates are even and samples in which both the x and y coordinates are odd numbers in the filter coefficient derivation region. can induce

Alternatively, as in the example shown in FIG. 7B , the block gradient may be derived using only samples included inside the block, not the filter coefficient derivation region.

Alternatively, as in the example shown in (c) of FIG. 7 , the block gradient can be derived using only samples in which both the x and y coordinates are even and samples in which both the x and y coordinates are odd numbers. have.

Alternatively, as in the example shown in (d) of FIG. 7 , the block gradient may be derived using only samples at a central position within the block.

When there are a plurality of simplification methods, information specifying one of the plurality of simplification methods may be encoded and signaled. In this case, the information may be encoded through an upper header such as a sequence, a picture, or a slice.

Alternatively, information indicating a sampling rate for subsampling may be encoded and signaled.

Alternatively, the sampling rate may be adaptively determined according to a color component, a color format, or a block size.

In order to simplify the derivation of the filter coefficients, the class of the block may be derived using only the gradients in the first group direction or the class of the block may be derived using only the gradients in the second group direction.

At this time, based on at least one of the size and shape of the block, whether a deblocking filter is applied, or whether a sample adaptive offset filter is applied, whether to use only gradients in the first group direction or only gradients in the second group direction It can be decided whether or not to use

Alternatively, information specifying at least one of the first group direction and the second group direction may be encoded and signaled. The information may indicate a first group direction, a second group direction, or one of the first and second group directions. Only the gradients in the group direction selected by the above information can be used to derive the class of the block.

As another example, a range of samples used to derive a block gradient or directions used to derive a block class may be determined for each predefined region in the reconstructed image. Here, the predefined regions may indicate regions in which mutually parallel processing is possible. For example, the predefined region may indicate a tile, a slice, or a sub-picture.

For example, for each predefined region, information specifying directions used to derive a block class may be encoded and signaled. Accordingly, directions used to derive the block class may be different for each predefined region. For example, in the first tile in the picture, the block class is derived using gradients in the first group direction and the second group direction, and in the second tile, the block class is derived using only gradients in the second group direction. have.

When the class of the block is determined, a filter coefficient set corresponding to the determined class may be called, and a geometric transformation type for the called filter coefficient set may be selected. Specifically, filter coefficients for the current block may be derived by selecting one of M geometric transformation types.

Information for determining the number M of geometric transformation types may be encoded and signaled. The information may be signaled through an upper header such as a slice, a picture, or a sequence. Alternatively, the number of geometric transformation types usable in the encoder and the decoder may be predefined.

The geometric transformation types may include at least one of a non-transformation, a vertical transformation, a horizontal transformation, a diagonal transformation, or a rotation. When a geometric transformation type is selected, filter coefficients included in the filter coefficient set may be allocated according to an arrangement order of filter coefficients in the selected geometric transformation type.

8 shows an example of geometric transformation types.

In the example shown in FIG. 8 , when the vertical transformation type is selected, the positions of filter coefficients under the non-transform type may be inverted around a central axis in the horizontal direction (eg, f9-f9 of the non-transform type).

When the diagonal transformation type is selected, the positions of filter coefficients under the non-transform type may be inverted about a central axis in the diagonal direction (eg, f7-f7 or f5-f5 of the non-transform type).

When the rotation type is selected, the positions of the filter coefficients under the non-transform type may be rotated by a predetermined angle (eg, 90 degrees) in a clockwise or counterclockwise direction.

Although not shown, when the horizontal transform type is selected, the positions of filter coefficients under the non-transform type may be inverted around a central axis in the vertical direction (eg, f0-f0 of the non-transform type).

For convenience of description, it is assumed that the number of available geometric transformation types is four shown in the example of FIG. 8 in the embodiments described below.

By using the variables dir1 and dir2 derived based on the block gradients, the geometric transformation type of the current block may be determined. Table 1 shows an example in which the geometric transformation type is determined by the variables dir1 and dir2.

Type dir1 dir2 Diagonal 0 One One 0 Vertical 0 3 3 0 Rotation One 2 2 One No geometric transformation 2 3 3 2

By using the variables dir1 and dir2 derived based on the block gradients, the geometric transformation type of the current block may be determined. Table 1 shows an example in which the geometric transformation type is determined by the variables dir1 and dir2. When only gradients in the first group direction or gradients in the second group direction are used, a lookup table defined differently from Table 1 This can be used.

Alternatively, a plurality of lookup tables may be predefined in the encoder and the decoder. The encoder may encode and signal information specifying one of a plurality of lookup tables. Each lookup table contains block gradients (eg, _{at least one of g h} , g _v , g _d1 or g _d2 ) or a variable derived from the block gradients (eg, at least one of dir1 or dir2 ) and a geometric transformation type. It may contain correspondences between

Or, at least among the size, shape, and block gradients of the current block, the maximum value, the minimum value, the median value, or their average value, whether the deblocking filter is applied to the current block, or whether the sample adaptive offset filter is applied to the current block Considering one, one of the plurality of lookup tables may be selected.

Filter coefficient sets may be predefined in the encoder and decoder.

Alternatively, configuration information of each of the essential coefficient sets may be signaled through a bitstream. Here, the configuration information may include at least one of the number of filter coefficients included in the filter coefficient set (ie, the number of filter taps) or the value of the filter coefficients.

In this case, filter coefficient sets may be independently defined in units of predefined regions. Here, the predefined regions may indicate regions in which parallel processing is possible. For example, the predefined region may be a tile, a slice, or a sub-picture.

As an example, for each of the tiles in the picture, filter coefficient sets may be individually defined. For example, when a picture includes 4 tiles and 25 classes are defined for each tile, configuration information for a total of 100 filter coefficient sets may be transmitted through an upper header. In this case, a block belonging to a specific tile may use one of the filter coefficient sets defined for the corresponding tile.

Next, an area-based filter application method will be described.

When the region-based filter application method is applied, the reconstructed image may be divided into a plurality of regions, and a filter may be determined for each region. Specifically, for each of the plurality of regions, at least one of a filter shape, a filter coefficient, or a filter size may be determined.

Information for determining a filter may be signaled through a bitstream. Specifically, information on at least one of a filter shape, filter coefficients, and filter size may be encoded and signaled.

When the region-based filter application method is applied, segmentation information of a reconstructed image may be encoded and signaled through a bitstream. The information may indicate at least one of the number of regions included in the reconstructed image, whether the regions are divided into uniform sizes, or sizes of the regions.

Alternatively, at least one of the number of regions, whether regions are divided into uniform sizes, or sizes of regions may be predefined in the encoder and decoder.

Alternatively, at least one of the number of regions, whether the regions are divided into uniform sizes, or sizes of the regions may be adaptively determined based on at least one of a picture resolution, a color component, a color format, or an HDR image.

Alternatively, each of the arbitrary processing units in the picture may be set as one area. Here, the processing unit may be a tile, a slice, or a sub-picture. When each of the arbitrary processing units in the picture is set as one region, encoding of partition information for the region may be omitted.

Alternatively, information indicating whether each of the processing units is set as one region may be encoded and signaled. For example, the information may be a 1-bit flag. When the flag is 1 (True), it indicates that each of the processing units is set to one area. In this case, encoding of partition information for a region may be omitted. The flag being 0 (False) indicates that each of the processing units is not necessarily set to one area. In this case, segmentation information for the region may be additionally encoded.

The boundary of the region may be set so that one processing unit is not divided into a plurality of units. As an example, a set of at least one coding tree unit may be defined as one region.

9 shows an example in which a reconstructed image is divided into a plurality of regions.

When regions are divided into uniform sizes, at least one of information indicating the number of rows constituting the regions, information indicating the number of columns constituting regions, or information indicating the size (eg, at least one of width or height) of regions may be encoded and signaled.

On the other hand, when the regions are not divided into uniform sizes, at least information indicating the number of columns constituting the regions, information indicating the number of rows constituting the regions, information indicating the width of each column, or information indicating the height of each row One may be encoded and signaled.

As an example, in FIG. 9 , it is exemplified that the reconstructed image is divided into regions of non-uniform size, and the reconstructed image is exemplified as divided into three columns and three rows. Accordingly, the value of information (eg, flag) indicating whether the reconstructed image is divided into regions of uniform size may be set to false and encoded. Also, the number of columns included in the reconstructed image may be set to represent 3 and encoded, and the number of rows included in the reconstructed image may be set to represent 3 and encoded.

In addition, information on the width of each column and the height of each row may be encoded and signaled. Here, the column width indicates the number of predetermined processing unit columns included in the corresponding column, and the column height indicates the number of predetermined processing unit rows included in the corresponding column. When the processing unit is a coding tree unit, in the example shown in FIG. 9 , the width of the first column and the height of the first row may be set to 4 and 3, respectively.

Also, encoding of width information may be omitted for the last column in the reconstructed image, and encoding of height information may be omitted for the last row in the reconstructed image.

For convenience of hardware implementation and memory saving, restrictions may be applied to applying a filter at the boundary between regions. As an example, when the filter is out of the boundary of the region, filtering may be performed except for a sample at a position outside the boundary of the region and/or a sample at a position symmetrical thereto.

10 is an exemplary diagram for explaining an application aspect of the filter when the filter deviates from the boundary of the region.

In the example shown in FIG. 10 , the sample to be applied with the filter is located at the center position of the filter (eg, f[12]). In this case, when the neighboring sample for filtering the filter application target sample is out of the boundary of the region, the neighboring sample may be set to be unavailable.

For example, in the example shown in (a) of FIG. 10 , samples at locations (eg, f[0] to f[8]) outside the upper boundary of the region are set to be unavailable for filtering the sample to be applied. can

In addition, the above unavailable samples and samples at symmetric positions may also be set as unavailable.

Accordingly, in the example shown in (a) of FIG. 10 , the filter application target sample is filtered using only samples belonging to the same row as the filter application target sample (eg, f[9] to f[11]). can be

Alternatively, filtering may be performed by setting a value of an unavailable sample to a default value (eg, 0) or converting a filter coefficient for an unavailable sample to 0.

In the example shown in (b) of FIG. 10 , samples at positions (eg, f[0] to f[8]) outside the lower boundary of the region may be set to be unavailable for filtering the sample to be applied to the filter.

In addition, the above unavailable samples and samples at symmetric positions may also be set as unavailable.

Accordingly, in the example shown in (b) of FIG. 10 , the filter application target sample uses only samples belonging to the same row as the filter application target sample (eg, f[9] to f[11]), can be filtered.

Alternatively, filtering may be performed by setting a value of an unavailable sample to a default value (eg, 0) or converting a filter coefficient for an unavailable sample to 0.

At the upper and lower boundaries of the region, samples at a symmetric position about the horizontal axis of the filter may be determined, and at the left and right boundaries of the region, samples at a symmetric position about the vertical axis of the filter may be determined.

At the boundary between regions, information indicating whether the above-described constraint is applied may be encoded and signaled. For example, a flag indicating whether the above-described constraint is applied may be encoded and signaled.

In this case, the flag may be individually encoded for each region.

Alternatively, the flag may be encoded at a higher level than zero. For example, the flag may be signaled through an upper header such as a slice, a picture, or a sequence. In this case, with respect to regions referring to the same higher header, whether the above-described restriction is applied may be determined in common.

Alternatively, a variable indicating whether the above-described constraint is applied or not may be defined, and whether the above-described constraint is applied to each region may be determined by the variable. For example, the variable ResFlag may indicate whether a constraint is applied. The value of the variable ResFlag may be determined based on the above-described flag value. Alternatively, the value of the variable ResFlag may be adaptively determined based on at least one of a size of a region, the number of regions, a position of the region, a color component, or a color format. Alternatively, the value of the variable ResFlag may be predefined in the encoder and the decoder.

When the above-mentioned constraint does not apply (eg, when the variable ResFlag is 0), samples outside the boundary of the region may also be used for filtering.

In addition, information indicating whether to set the unavailable sample and the sample present at a symmetric position as unavailable may be encoded and signaled. The information may be signaled for each area. As an example, a flag indicating whether to set an unavailable sample and a sample having a symmetric position to be unavailable may be signaled.

Alternatively, whether to set the unavailable sample and the sample present in a symmetric position to be unavailable may be predefined in the encoder and the decoder.

Alternatively, it may be determined whether or not to set the samples present in the symmetric position as unavailable by considering at least one of the filter type, the filter size, or the number of unavailable samples (eg, the number of samples outside the region boundary).

Instead of excluding unusable samples during filtering, padding may be performed on an unavailable sample position, and then filtering may be performed using the padded sample. Padding can be performed by copying pixels that are symmetric with the unavailable samples or to which the same filter coefficients are applied to the unavailable samples. Alternatively, padding may be performed by copying available pixels adjacent to unavailable samples.

In the above example, it has been described that the picture is divided into a plurality of regions when the region-based filter application method is applied. As another example, an area-based filter application method may be applied to each processing unit in which parallel processing is performed. Here, the processing unit may represent a slice, a tile, or a subpicture. That is, region division may be performed independently for each processing unit.

As another example, at least one of a size or a shape of a filter may be adaptively determined according to a location of a sample within a block or a location of a block.

When the adaptive loop filter is applied, the reconstructed sample may be filtered using the reconstructed sample and neighboring reconstructed samples. As an example, Equations 8 and 9 below are examples illustrating a process of obtaining a filtered reconstructed sample.

In Equation 8, f[n] means a filter coefficient. I represents the input reconstruction sample. h _x indicates that the horizontal direction (ie, x-axis) coordinate is x, and v _y indicates that the vertical direction (ie, y-axis) coordinate is y. That is, I[h _x , v _y ] represents a restored sample at the (x, y) position.

In Equation 9, O(x, y) represents a filtered reconstructed sample.

As in the examples shown in Equations 8 and 9, by shifting the temporary value tmp obtained by summing values derived by multiplying each reconstructed sample by a filter coefficient based on the shifting parameter shift, the filtered reconstructed sample O (x, y) can be obtained.

Here, the shifting parameter may be predefined in the encoder and the decoder. As an example, the shifting parameter shift may be fixed to 7.

Alternatively, information for determining a shifting parameter may be signaled through a bitstream. The information may be encoded through an upper header such as a slice, a tile, or a sub picture.

Alternatively, the shifting parameter may be adaptively determined based on at least one of a size of a current block, a filter type, a filter size, a bit depth, a color component, or a color format.

The embodiments described above relate to the application of a linear adaptive loop filter. Contrary to the description, non-linearity may be applied when the adaptive loop filter is applied. When a nonlinear adaptive loop filter is applied, a filtered sample may be output by applying a Clip function. For example, when a nonlinear adaptive loop filter is applied, Equation 1 may be changed to Equation 10 below.

The clip function may be defined as in Equation 11 below.

In addition, a detailed process to which the nonlinear adaptive loop filter is applied can be expressed as Equation 12 below.

In Equation 12, c[n] represents the clipping value k(i, j) in one dimension.

When the temporary value tmp is derived through Equation 12, a filtered reconstructed sample may be obtained through Equation 9.

Information indicating whether the nonlinear adaptive loop filter is applied may be encoded and signaled. For example, a 1-bit flag indicating whether a nonlinear adaptive loop filter is applied may be encoded and signaled.

Information indicating whether a nonlinear adaptive loop filter is applied to each color component may be encoded.

When a nonlinear adaptive loop filter is applied, information for determining a clipping value corresponding to a filter coefficient may be encoded and signaled. As an example, information indicating a clipping value c[n] corresponding to the filter coefficient f[n] may be encoded.

In this case, after setting a plurality of clipping values as candidates, an index specifying a clipping value corresponding to the filter coefficient among the plurality of clipping value candidates may be encoded and signaled for each filter coefficient.

Table 2 shows an example in which different clipping indices are assigned to each clipping value candidate.

BitDepth clipIdx 0 One 2 3 8 2 ⁸ 2 ⁵ 2 ³ 2 ¹ 9 2 ⁹ 2 ⁶ 2 ⁴ 2 ² 10 2 ¹⁰ 2 ⁷ 2 ⁵ 2 ³ 11 2 ¹¹ 2 ⁸ 2 ⁶ 2 ⁴ 12 2 ¹² 2 ⁹ 2 ⁷ 2 ⁵ 13 2 ¹³ 2 ¹⁰ 2 ⁸ 2 ⁶ 14 2 ¹⁴ 2 ¹¹ 2 ⁹ 2 ⁷ 15 2 ¹⁵ 2 ¹² 2 ¹⁰ 2 ⁸ 16 2 ¹⁶ 2 ¹³ 2 ¹¹ 2 ⁹

In Table 2, Bitdepth indicates a bit depth of an image, and clipIdx indicates a clipping index assigned to a clipping value candidate. The configuration of clipping value candidates may be different depending on the bit depth. Here, for example, according to the bit depth, at least one of the number of candidates or the values of the candidates may be different. In Table 2, it is exemplified that the values of the clipping value candidates are set differently according to the bit depth.

Alternatively, a configuration of clipping value candidates may be determined based on at least one of a picture resolution, a color component, or a color format. For example, according to a picture resolution, a color component, or a color format, at least one of the number of candidates or the value of the candidates may be different.

To reduce the bit length of the index identifying the clipping value, a smaller number of candidates than illustrated in Table 2 may be used. For example, only two clipping value candidates may be defined for each bit depth. When two clipping value candidates are defined, a clipping value may be determined using a 1-bit flag.

Tables 3 to 5 illustrate simplified lookup tables.

BitDepth clipIdx 0 One 8 2 ⁸ 2 ⁵ 9 2 ⁹ 2 ⁶ 10 2 ¹⁰ 2 ⁷ 11 2 ¹¹ 2 ⁸ 12 2 ¹² 2 ⁹ 13 2 ¹³ 2 ¹⁰ 14 2 ¹⁴ 2 ¹¹ 15 2 ¹⁵ 2 ¹² 16 2 ¹⁶ 2 ¹³

BitDepth clipIdx 0 One 8 2 ⁸ 2 ³ 9 2 ⁹ 2 ⁴ 10 2 ¹⁰ 2 ⁵ 11 2 ¹¹ 2 ⁶ 12 2 ¹² 2 ⁷ 13 2 ¹³ 2 ⁸ 14 2 ¹⁴ 2 ⁹ 15 2 ¹⁵ 2 ¹⁰ 16 2 ¹⁶ 2 ¹¹

BitDepth clipIdx 0 One 8 2 ⁸ 2 ¹ 9 2 ⁹ 2 ² 10 2 ¹⁰ 2 ³ 11 2 ¹¹ 2 ⁴ 12 2 ¹² 2 ⁵ 13 2 ¹³ 2 ⁶ 14 2 ¹⁴ 2 ⁷ 15 2 ¹⁵ 2 ⁸ 16 2 ¹⁶ 2 ⁹

When a plurality of lookup tables are pre-stored, information identifying one of the plurality of lookup tables may be encoded and signaled. As an example, an index specifying one of the lookup tables shown in Tables 2 to 5 may be encoded and signaled. Alternatively, the lookup table may be adaptively selected based on at least one of a bit depth, a color component, or a color format. As another example, in order to reduce memory occupancy, a clipping value may be determined using a formula instead of using a lookup table. Equation 13 shows an example in which a clipping value is determined by a bit depth and a clipping index.

In Equation 13, the variable clipIdx may be determined by an index signaled through a bitstream.

As another example, the number of times of application of the clipping function may be further reduced than in Equation (12). As an example, Equation 14 shows another example in which the temporary value tmp is derived.

In Equation 12, it is exemplified that the clipping function is applied twice to one filter coefficient, but in Equation 14, it is exemplified that the clipping function is applied once to one filter coefficient.

Even if the number of times of application of the clipping function is decreased, the clipping value may not be changed. For example, in the case of deriving the temporary value tmp by Equation 12 and the deriving the temporary value tmp by Equation 14, the values of c[n] may be the same.

Alternatively, when the number of times of application of the clipping function decreases, the clipping value may be changed. For example, the clipping value c[n] in the case of deriving the temporary value tmp by Equation 14 may be twice the clipping value c[n] in the case of deriving the temporary value tmp by Equation 12.

In the above-described example, it is exemplified that an index for determining a clipping value is encoded for each filter coefficient.

As another example, indexes may be coded only for filter coefficients at preset positions.

11 shows an example in which indexes are coded for only some of filter coefficients.

In (a) of FIG. 11 , it is exemplified that indexes are encoded only for f[0], f[1], f[3], f[4], f[8], and f[9].

In (b) of FIG. 11 , it is exemplified that indexes are encoded only for f[2], f[5], f[7], and f[10].

In (c) of FIG. 11 , it is exemplified that indexes are encoded only for f[6] and f[11].

For filter coefficients whose index is not encoded, a fixed clipping value may be used or a clipping value of an adjacent position filter coefficient may be used. Here, the adjacent position refers to at least one of the upper end, the lower end, the left side, or the right side. Depending on the geometric transformation type, the adjacent position may be adaptively determined.

Alternatively, the clipping function may be applied only at the position where the index is encoded. That is, the clipping function may not be applied at a position where the index is not coded. As an example, in the example shown in (c) of FIG. 11 , the clipping function is applied at the position to which f[6] and f[11] are applied, while the position to which f[6] and f[11] are applied. The clipping function may not be applied to the remaining positions excepted.

Information related to the adaptive loop filter may be encoded and signaled through an upper header. Here, the upper header indicates a parameter set commonly referenced by a plurality of blocks (eg, a coding tree unit, a coding unit, or a transform unit) for which application of the adaptive loop filter is determined. In particular, an upper header in which information related to the adaptive loop filter is encoded may be defined as a filter header.

At least one of information about a filter coefficient set or configuration information of each filter coefficient set may be signaled through the filter header. For example, the filter header may include at least one of information on a filter coefficient set corresponding to each of a plurality of classes for a block-based filter application method, or information on a filter coefficient set for each region for an area-based filter application method may contain one.

For example, if 25 classes are used under the block-based filter application method and the reconstructed picture is set to be divided into 20 regions under the area-based filter application method, a filter coefficient set corresponding to each of the 25 classes through the filter header or At least one of the filter coefficient sets corresponding to each of the 20 regions may be encoded and signaled.

Information indicating whether the filter header includes information on filter coefficient sets used in the block-based filter application method or filter coefficient sets used in the region-based filter application method may be encoded.

A plurality of filter headers may be encoded and signaled. In this case, a different identifier (ID, Identifier) may be assigned to each filter head.

Information indicating the number of filter headers for each image may be encoded and signaled. For example, the number information may be signaled through a higher header such as a picture or a sequence. The number information may represent a value obtained by subtracting 1 from the total number of filter headers.

For each color component, a different filter header may be selected. As an example, for a luminance component (eg, y), a filter header with an identifier (ID) of 0 is used, while for a first chrominance component (eg, u or cb), a filter header with an identifier of 1 is used , for the second chrominance component (eg, v or cr), a filter header having an identifier of 2 may be used.

Instead of using a different filter header for each color difference component, one filter header may be commonly applied to the color difference components. For example, a filter header having an identifier of 0 may be used for the luminance component, while a filter header having an identifier of 1 may be used for the first chrominance component and the second chrominance component.

Alternatively, one filter header may be commonly applied to all color components.

Information indicating whether the filter header is individually determined for each color component or for each color difference component may be encoded. Based on the information, it may be determined whether an index indicating one of a plurality of filter headers is signaled for each color component or for each color difference component.

When one filter header is commonly applied to a plurality of color components, the filter header may include filter coefficient set information for each of the plurality of color components. In this case, information indicating which color component the filter coefficient set in the filter header relates to may be further encoded and signaled.

For example, when one filter header is commonly applied to all color components, the filter coefficient set in the filter header relates to the luminance component (Y), the first chrominance component (eg, Cb), or the second Information indicating whether or not it relates to two color difference components (eg, Cr) may be additionally encoded and signaled.

A filter header storing filter coefficient sets for the block-based filter application method and a filter header storing filter coefficient sets for the region-based filter application method may be independently encoded and signaled.

Alternatively, one filter header may include filter coefficient sets for the block-based filter application method and filter coefficient sets for the region-based filter application method in one filter header. In this case, information indicating whether the filter coefficient set in the filter header is for the block-based filter application method or the region-based filter application method may be additionally encoded and signaled.

As another example, the filter coefficient application method may be fixedly set for each color component. As an example, the block-based filter application method may be fixedly applied to the luma component, while the region-based filter application method may be fixedly applied to the chrominance component. In this case, a filter header may be selected with reference to a filter application method of each color component.

Coefficient merging between filter coefficient sets may be performed. Through coefficient merging, information on a filter coefficient set may be encoded only for a part of a plurality of classes or a plurality of regions. A class in which the filter coefficient set is not coded or a region in which the filter coefficient set is not coded may be set to use a filter coefficient set of another class or another region.

12 is a diagram for explaining an example in which coefficient merging is applied under a block-based filter application method.

Fig. 12 (a) is an embodiment in which coefficient merging is not applied, and Fig. 12 (b) is an embodiment in which coefficient merging is applied.

When coefficient merging is not applied, a filter coefficient set for each of the classes may be encoded and signaled. For example, when there are 25 classes, a filter coefficient set may be encoded for each of the 25 classes, as in the example shown in FIG. 12A .

When coefficient merging is not applied, filter coefficient sets as many as the number of classes may be encoded and signaled.

When coefficient merging is applied, a number of filter coefficient sets less than the number of classes may be encoded and signaled. In this case, one filter coefficient set may be allocated to a plurality of classes. As an example, as in the example shown in FIG. 12B , filter coefficient set 1 may be allocated to classes 1 to 3 .

When coefficient merging is applied, since the number of filter coefficient sets to be encoded is reduced, from the viewpoint of Rate-Distortion Optimization (RDO), encoding efficiency may be improved. That is, the signaling overhead for the filter coefficient set can be reduced.

When a filter coefficient set is defined for each of the classes, information indicating a mapping relationship between the class and the filter coefficient set is unnecessary. That is, since the number of classes and filter coefficient sets are transmitted (ie, 1:1 mapping between classes and filter coefficient sets), when the class of the block is determined to be N, it may be determined that the filter coefficient set N is used.

On the other hand, when the number of filter coefficient sets is smaller than the number of classes, information indicating a mapping relationship between classes and filter coefficient sets is additionally required. For example, in the example shown in (b) of FIG. 12 , since filter coefficient set 1 rather than filter coefficient set 2 is assigned to class 2, mapping information indicating that filter coefficient set 1 is assigned to class 2 is additionally provided is required The mapping information may be an index for identifying a filter coefficient set having a mapping relationship with a class.

When coefficient merging is applied, information indicating the number of filter coefficient sets may be further encoded and signaled. For example, in the example shown in (b) of FIG. 12 , a total of five filter coefficient sets are encoded and signaled. Accordingly, number information indicating that the number of filter coefficient sets is five may be encoded and signaled.

As an example, number information derived by differentiating a predetermined constant value from the number of filter coefficient sets may be encoded. Here, the constant value may be a natural number such as 1 or 2.

Alternatively, number information derived based on the difference between the number of classes and filter coefficient sets may be encoded.

For each class, index information for identifying a filter coefficient set that is a mapping relationship may be encoded and signaled. For example, in the example shown in (b) of FIG. 12 , since there are a total of five filter coefficient sets, index information indicating a value of 0 to 4 for each class may be encoded and signaled. .

In this case, the index information may indicate an index of a filter coefficient set assigned to a class.

Alternatively, index information of each class may be encoded based on differential coding (DPCM). For example, for class N, between the index of the filter coefficient set assigned to class N and the index of the filter coefficient set assigned to the previous class (eg, class N-1) or the next class (eg, class N+1) The index difference indicating the difference may be encoded as index information.

Alternatively, a flag indicating whether the filter coefficient set assigned to the class is the same as the previous class or the next class may be encoded and signaled. When the flag is 1 (True), encoding of index information for a corresponding class may be omitted. On the other hand, when the flag is 0 (False), index information may be encoded for a corresponding class.

It may be set so that the filter coefficient set index assigned to the class is not allowed to have a value smaller than the index of the filter coefficient set assigned to the previous class. In this case, for some classes, encoding of index information may be omitted.

13 illustrates an example in which encoding of index information is omitted for some classes.

13A shows a filter coefficient set assigned to each class. As in the illustrated example, the index of the filter coefficient set allocated to class N may be forced to have a value equal to or greater than the index of the filter coefficient set allocated to the previous class (ie, class N-1). In this case, the filter coefficient set having the smallest index is always allocated to the first class (class 0), and the filter coefficient set having the largest index is always allocated to the last class (class 24).

Accordingly, with respect to the first class and the last class, encoding of index information may be omitted.

13 (b) shows an encoding aspect of index information. As in the illustrated example, encoding of index information may be omitted for the first class (class 0) and the last class (class 24). Only for the remaining classes except for the first class and the last class, index information may be encoded and signaled.

When the signaling of the index information is omitted, the decoder may determine the filter coefficient set assigned to the corresponding class according to whether the class is the first or the last. For example, when a class in which signaling of index information is omitted is the first class, it may be inferred that the first filter coefficient set is selected. On the other hand, when the class in which the signaling of index information is omitted is the last class, it may be considered that the last filter coefficient set is selected.

The above-described embodiment related to merging filter coefficients may be applied not only to the block-based filter application method but also to the region-based filter application method.

14 is a diagram for explaining an example in which coefficient merging is applied under a method of applying a region-based filter.

Fig. 14 (a) is an embodiment in which coefficient merging is not applied, and Fig. 14 (b) is an embodiment in which coefficient merging is applied.

When coefficient merging is not applied, a filter coefficient set for each of the regions may be encoded and signaled. For example, when there are 25 regions, a filter coefficient set may be encoded for each of the 25 regions, as in the example shown in FIG. 14A .

When coefficient merging is not applied, filter coefficient sets as many as the number of regions may be encoded and signaled.

When coefficient merging is applied, a number of filter coefficient sets less than the number of regions may be encoded and signaled. In this case, one filter coefficient set may be allocated to a plurality of regions. As an example, as in the example shown in (b) of FIG. 14 , the filter coefficient set 1 may be allocated to regions 1 to 3 .

When the coefficient merging method is applied, information for specifying a filter coefficient set allocated to each region may be encoded and signaled. In this case, since the method of encoding the index information of each region is the same as the above-described block-based filter application method, a detailed description thereof will be omitted.

Information on filter coefficients constituting the filter coefficient set may be encoded and signaled. Referring to the example of FIG. 4 , values of 13 filter coefficients (f[0] to f[12]) are coded and signaled for a luminance component, and 7 filter coefficients (f[0] to f[0]) are signaled for a chrominance component. ] to f[6]) may be encoded and signaled.

In this case, encoding may be omitted for the filter coefficients at the filter center position. For example, in the example of FIG. 4 , encoding of f[12], which is the central position of the filter for the luminance component, may be omitted, and encoding of f[6], which is the central position of the filter for the chrominance component, may be omitted. .

The filter coefficient at the center position may be derived by differentiating the sum of the remaining filter coefficients from a predetermined constant value. As an example, the filter coefficient f[12] for the luminance component and the filter coefficient f[6] for the chrominance component may be derived as shown in Equation 15 below.

The information on the filter coefficients may include at least one of absolute value information of the filter coefficients and sign information of the filter coefficients.

As another example, the filter may be applied by quantizing the filter coefficient f[n] to the power of 2. In this case, instead of the filter coefficient absolute value information, an index of the filter coefficient may be encoded and signaled. The index may represent one of the values expressed as an exponential power of 2. For example, assuming that the absolute value of the filter coefficient has a value between 0 and 64 and that the filter coefficient is quantized to a value expressed by the power of 2, the absolute value of the filter coefficient is one of the values included in the following set. determined by one

{0, 1, 2, 4, 8, 16, 32, 64}

Accordingly, an index specifying one of the values included in the above set may be encoded. For example, index 0 may indicate absolute value 0, and index 7 may indicate absolute value 64.

Instead of quantizing the values of the filter coefficients to the power of 2, each of the quantized values may be set to have an equivalence or an equal ratio. For example, the filter coefficients may be quantized to be one of the arithmetic sequences such as {0, 9, 18, 27, 36, 45, 54, 64}.

When encoding an index indicating an absolute value of a filter coefficient, a truncated unary binarization method may be applied.

As another example, when encoding an index indicating an absolute value of a filter coefficient, a fixed length method may be applied. For example, if an index indicating one of a total of eight values is encoded, an index having a value between 000 and 111 may be encoded and signaled.

When the absolute value of the filter coefficient is not 0, the signal may be signaled by additionally encoding filter coefficient sign information. The sign information may be a flag indicating whether the filter coefficients are positive or negative.

The range of the absolute value of the filter coefficient may be predefined in the encoder and the decoder.

As another example, information indicating a range of absolute values of filter coefficients may be signaled through a bitstream. The information may be signaled through an upper header such as a slice, picture, sequence, or filter header. The information may indicate at least one of a minimum value or a maximum value of the range. For example, when the maximum value of the range ^{is expressed as 2 N} , information for specifying the N may be encoded and signaled.

As another example, the range may be adaptively determined based on at least one of a color component, a color format, a filter size, a filter type, or a filter application method.

When a plurality of filter headers exist, identification information for identifying one of the filter headers may be signaled for each processing unit of a predefined level. For example, identification information indicating one of a plurality of filter headers may be encoded and signaled for each slice, subpicture, or tile. Blocks belonging to a processing unit of a predefined level may use the filter header determined at the processing unit level.

As another example, after a filter header is determined for each processing unit of a predetermined level, information indicating whether to use the filter header determined at the processing unit level as it is for each sub processing unit belonging to the processing unit is encoded and signaled. can The information may be a 1-bit flag. For example, when the flag is 1 (True), the filter header determined at the processing unit level may be used as it is. On the other hand, when the flag is 0 (False), identification information for identifying a filter header to be used in a sub-processing unit may be additionally encoded. The identification information may specify one of the remaining filter headers excluding the filter header determined at the processing unit level.

As an example, it is assumed that the dependency relationship between processing units follows.

Slice ⊆ Tile ⊆ Sub Picture ⊆ Picture

In this case, the processing unit for which the filter header is determined may be any one of a tile, a subpicture, or a picture, and the subprocessing unit may be any one of a slice, a tile, or a subpicture.

As another example, a processing unit in which identification information of a filter header is encoded may be adaptively determined according to a method of determining a slice. Here, the slice determination method may be one of a raster scan slice determination method or a square slice determination method.

The raster scan slice determination method refers to a method of determining successive tiles as one slice when the raster scan order is followed. The rectangular shape slice determination method refers to a determination method that allows only slices having a rectangular shape. When the rectangular slice determination method is applied, a tile located at a vertex of the rectangle may belong to the same row and/or the same column as a tile located at another vertex.

For example, when the raster scan slice determining method is applied, identification information of a filter header may be encoded for each slice. On the other hand, when the rectangular slice determination method is applied, identification information of the filter header may be encoded for each tile.

Alternatively, in contrast to the above, when the raster scan slice determining method is applied, identification information of the filter header is encoded for each tile, and when the rectangular slice determining method is applied, the identification information of the filter header can be encoded for each slice.

As another example, the filter header may be independently encoded for each of the parallel processing target regions. For example, when parallel processing for each tile is performed, a filter header may be encoded for each of the tiles. As an example, when a picture consists of two tiles, three filter headers may be encoded for a first tile and two filter headers may be encoded for a second tile. Accordingly, a total of five filter headers may be encoded for the current picture.

In this case, one filter header may be redundantly coded for a plurality of tiles.

At least one of the number of filter headers for each tile or identifier information of a filter header allocated to each of the tiles may be further encoded and signaled.

As another example, information for identifying a filter header for each block may be encoded and signaled. For example, when it is determined that the adaptive loop filter is applied to the current coding tree unit, information for identifying a filter header applied to the current coding tree unit may be further encoded and signaled.

In the above-described example, it has been exemplified that information indicating a method of applying the adaptive loop filter is encoded and signaled. As another example, a method of applying the adaptive loop filter may be determined according to the selected filter header. For example, when a filter header including only filter coefficient sets for the block-based filter application method is selected, the block-based filter application method may be applied to the current block. On the other hand, when a filter header including only filter coefficient sets for the region-based filter application method is selected, the region-based filter application method may be applied to the current block.

A fixed filter application method for each color component may be applied. For example, a block-based filter application method may be always applied to a luminance component, and a region-based filter application method may be always applied to a chrominance component. In this case, with respect to the luminance component block, information for identifying one of the filter headers for the block-based filter application method may be additionally encoded and signaled. Also, with respect to the chrominance component region, information for identifying a current one of filter headers for the region-based filter application method may be additionally encoded and signaled.

While the region-based filter application method is applied, when a picture includes only one region, a plurality of filter coefficient set candidates may be configured for the single region. In this case, in order to determine the filter coefficient set for the single region, index information for identifying one of the plurality of filter coefficient set candidates may be encoded and signaled. In this case, the index information may be encoded and signaled in units of blocks (eg, coding tree units). That is, even when the region-based filter application method is applied, different filter coefficient sets may be used for each block.

A filter header can be determined independently for each color component.

Alternatively, color components may be set to use the same filter header. In this case, the filter header may include both information on filter coefficient sets for a luminance component and information on filter coefficient sets for a chrominance component. Also, for the luminance component, information identifying one of the plurality of filter headers may be encoded, and for the color difference component, the encoding of the information may be omitted.

The names of syntaxes used in the above-described embodiments are merely named for convenience of description.

Applying the decoding process or the embodiments described based on the encoding process to the encoding process or the decoding process is included in the scope of the present disclosure. It is also within the scope of the present disclosure to change the embodiments described in a certain order in an order different from that described.

Although the above-described embodiment has been described based on a series of steps or a flowchart, this does not limit the time-series order of the invention, and may be performed simultaneously or in a different order, if necessary. In addition, each of the components (eg, unit, module, etc.) constituting the block diagram in the above-described embodiment may be implemented as a hardware device or software, or a plurality of components may be combined to form one hardware device or software. may be implemented. The above-described embodiment may be implemented in the form of program instructions that can be executed through various computer components and recorded in a computer-readable recording medium. The computer-readable recording medium may include program instructions, data files, data structures, etc. alone or in combination. Examples of the computer-readable recording medium include a hard disk, a magnetic medium such as a floppy disk and a magnetic tape, an optical recording medium such as a CD-ROM, a DVD, and a magneto-optical medium such as a floppy disk. media), and hardware devices specially configured to store and execute program instructions, such as ROM, RAM, flash memory, and the like. The hardware device may be configured to operate as one or more software modules to perform processing according to the present disclosure, and vice versa.

Claims (15)

determining whether to apply an adaptive loop filter to the current block;
when the adaptive loop filter is applied, determining whether the adaptive loop filter has linearity or nonlinearity;
determining a filter coefficient set of the adaptive loop filter; and
and filtering the reconstructed samples in the current block by using the filter coefficient set. According to claim 1,
When it is determined that the adaptive loop filter has nonlinearity,
The filtering is performed by applying a clipping function to the difference between the reconstructed sample and the surrounding reconstructed sample,
The clipping value for applying the clipping function is determined by an index specifying one of a plurality of clipping value candidates. 3. The method of claim 2,
The index is determined by a 1-bit flag decoded from the bitstream, the video decoding method. 3. The method of claim 2,
The number of clipping value candidates is adaptively determined based on at least one of a color component, a color format, and a bit depth. 3. The method of claim 2,
An image decoding method, characterized in that the plurality of clipping value candidates are derived from one of a plurality of lookup tables defining a mapping relationship between bit depth and clipping value candidates. 6. The method of claim 5,
An image decoding method, characterized in that one of the plurality of lookup tables is selected according to index information parsed from a bitstream. 4. The method of claim 3,
The image decoding method, characterized in that the flag is decoded only for a position to which a predefined filter coefficient is applied among the reconstructed samples in the current block. 8. The method of claim 7,
An image decoding method, characterized in that the clipping function is applied only at a position where the predefined filter coefficient is applied. determining whether to apply an adaptive loop filter to the current block;
when the adaptive loop filter is applied, determining whether the adaptive loop filter has linearity or nonlinearity;
determining a filter coefficient set of the adaptive loop filter; and
and filtering the reconstructed samples in the current block by using the filter coefficient set. 10. The method of claim 9,
When it is determined that the adaptive loop filter has nonlinearity,
The filtering is performed by applying a clipping function to the difference between the reconstructed sample and the surrounding reconstructed sample,
The clipping value for applying the clipping function is determined by an index specifying one of a plurality of clipping value candidates. 11. The method of claim 10,
A video encoding method, characterized in that a 1-bit flag indicating the index is encoded in a bitstream. 11. The method of claim 10,
The number of clipping value candidates is adaptively determined based on at least one of a color component, a color format, and a bit depth. 11. The method of claim 10,
The video encoding method, characterized in that the plurality of clipping value candidates are derived from one of a plurality of lookup tables defining a mapping relationship between bit depth and clipping value candidates. 14. The method of claim 13,
An index specifying one of the plurality of lookup tables is encoded in a bitstream. determining whether to apply an adaptive loop filter to the current block;
when the adaptive loop filter is applied, determining whether the adaptive loop filter has linearity or nonlinearity;
determining a filter coefficient set of the adaptive loop filter; and
and filtering the reconstructed sample in the current block by using the filter coefficient set.

Images