240 and output in the form of a bitstream.

The intra predictor 222 may predict a current block with reference to samples within a current picture. The referenced samples may be located neighboring to the current block, or may also be located away from the current block according to the prediction mode. The prediction modes in the intra prediction may include a plurality of non-directional modes and a plurality of directional modes. The non-directional mode may include, for example, a DC mode or a planar mode. The directional mode may include, for example, 33 directional prediction modes or 65 directional prediction modes according to the fine degree of the prediction direction. However, this is illustrative and the directional prediction modes which are more or less than the above number may be used according to the setting. The intra predictor 222 may also determine the prediction mode applied to the current block using the prediction mode applied to the neighboring block.

The inter predictor 221 may induce a predicted block of the current block based on a reference block (reference sample array) specified by a motion vector on a reference picture. At this time, in order to decrease the amount of motion information transmitted in the inter prediction mode, the motion information may be predicted in units of a block, a sub-block, or a sample based on the correlation of the motion information between the neighboring block and the current block. The motion information may include a motion vector and a reference picture index. The motion information may further include inter prediction direction (L0 prediction, L1 prediction, Bi prediction, or the like) information. In the case of the inter prediction, the neighboring block may include a spatial neighboring block existing within the current picture and a temporal neighboring block existing in the reference picture. The reference picture including the reference block and the reference picture including the temporal neighboring block may also be the same as each other, and may also be different from each other. The temporal neighboring block may be called the name such as a collocated reference block, a collocated CU (colCU), or the like, and the reference picture including the temporal neighboring block may also be called a collocated picture (colPic). For example, the inter predictor 221 may configure a motion information candidate list based on the neighboring blocks, and generate information indicating what candidate is used to derive the motion vector and/or the reference picture index of the current block. The inter prediction may be performed based on various prediction modes, and for example, in the case of a skip mode and a merge mode, the inter predictor 221 may use the motion information of the neighboring block as the motion information of the current block. In the case of the skip mode, the residual signal may not be transmitted unlike the merge mode. A motion vector prediction (MVP) mode may indicate the motion vector of the current block by using the motion vector of the neighboring block as a motion vector predictor, and signaling a motion vector difference.

The predictor 220 may generate a prediction signal based on various prediction methods to be described below. For example, the predictor 220 may apply intra prediction or inter prediction for prediction of one block and may simultaneously apply intra prediction and inter prediction. This may be called combined inter and intra prediction (CIIP). In addition, the predictor may be based on an intra block copy (IBC) prediction mode or based on a palette mode for prediction of a block. The IBC prediction mode or the palette mode may be used for image/video coding of content such as games, for example, screen content coding (SCC). IBC basically performs prediction within the current picture, but may be performed similarly to inter prediction in that a reference block is derived within the current picture. That is, IBC may use at least one of the inter prediction techniques described in the present document. The palette mode may be viewed as an example of intra coding or intra prediction. When the palette mode is applied, a sample value in the picture may be signaled based on information on the palette table and the palette index.

The prediction signal generated by the predictor (including the inter predictor 221 and/or the intra predictor 222) may be used to generate a reconstructed signal or may be used to generate a residual signal.

The transformer 232 may generate transform coefficients by applying a transform technique to the residual signal. For example, the transform technique may include at least one of a discrete cosine transform (DCT), a discrete sine transform (DST), a graph-based transform (GBT), or a conditionally non-linear transform (CNT). Here, GBT refers to transformation obtained from a graph when expressing relationship information between pixels in the graph. CNT refers to transformation obtained based on a prediction signal generated using all previously reconstructed pixels. Also, the transformation process may be applied to a block of pixels having the same size as a square or may be applied to a block of a variable size that is not a square.

The quantizer 233 quantizes the transform coefficients and transmits the same to the entropy encoder 240, and the entropy encoder 240 encodes the quantized signal (information on the quantized transform coefficients) and outputs the encoded signal as a bitstream. Information on the quantized transform coefficients may be referred to as residual information. The quantizer 233 may rearrange the quantized transform coefficients in the block form into a one-dimensional vector form based on a coefficient scan order and may generate information on the transform coefficients based on the quantized transform coefficients in the one-dimensional vector form.

The entropy encoder 240 may perform various encoding methods such as, for example, exponential Golomb, context-adaptive variable length coding (CAVLC), and context-adaptive binary arithmetic coding (CABAC). The entropy encoder 240 may encode information necessary for video/image reconstruction (e.g., values of syntax elements, etc.) other than the quantized transform coefficients together or separately. Encoded information (e.g., encoded video/image information) may be transmitted or stored in units of a network abstraction layer (NAL) unit in the form of a bitstream. The video/image information may further include information on various parameter sets, such as an adaptation parameter set (APS), a picture parameter set (PPS), a sequence parameter set (SPS), or a video parameter set (VPS). Also, the video/image information may further include general constraint information. In the present document, information and/or syntax elements transmitted/signaled from the encoding apparatus to the decoding apparatus may be included in video/image information. The video/image information may be encoded through the encoding procedure described above and included in the bitstream. The bitstream may be transmitted through a network or may be stored in a digital storage medium. Here, the network may include a broadcasting network and/or a communication network, and the digital storage medium may include various storage media such as USB, SD, CD, DVD, Blu-ray, HDD, and SSD. A transmitting unit (not shown) and/or a storing unit (not shown) for transmitting or storing a signal output from the entropy encoder 240 may be configured as internal/external elements of the encoding apparatus 200, or the transmitting unit may be included in the entropy encoder 240.

The quantized transform coefficients output from the quantizer 233 may be used to generate a prediction signal. For example, the residual signal (residual block or residual samples) may be reconstructed by applying dequantization and inverse transform to the quantized transform coefficients through the dequantizer 234 and the inverse transform unit 235. The adder 250 may add the reconstructed residual signal to the prediction signal output from the inter predictor 221 or the intra predictor 222 to generate a reconstructed signal (reconstructed picture, reconstructed block, reconstructed sample array). When there is no residual for the processing target block, such as when the skip mode is applied, the predicted block may be used as a reconstructed block. The adder 250 may be referred to as a restoration unit or a restoration block generator. The generated reconstructed signal may be used for intra prediction of a next processing target block in the current picture, or may be used for inter prediction of the next picture after being filtered as described below.

Meanwhile, luma mapping with chroma scaling (LMCS) may be applied during a picture encoding and/or reconstruction process.

The filter 260 may improve subjective/objective image quality by applying filtering to the reconstructed signal. For example, the filter 260 may generate a modified reconstructed picture by applying various filtering methods to the reconstructed picture, and store the modified reconstructed picture in the memory 270, specifically, in a DPB of the memory 270. The various filtering methods may include, for example, deblocking filtering, a sample adaptive offset, an adaptive loop filter, a bilateral filter, and the like. The filter 260 may generate various kinds of information related to the filtering, and transfer the generated information to the entropy encoder 240 as described later in the description of each filtering method. The information related to the filtering may be encoded by the entropy encoder 240 and output in the form of a bitstream.

The modified reconstructed picture transmitted to the memory 270 may be used as a reference picture in the inter predictor 221. When the inter prediction is applied through the encoding apparatus, prediction mismatch between the encoding apparatus 200 and the decoding apparatus may be avoided and encoding efficiency may be improved.

The DPB of the memory 270 may store the modified reconstructed picture for use as the reference picture in the inter predictor 221. The memory 270 may store motion information of a block from which the motion information in the current picture is derived (or encoded) and/or motion information of blocks in the picture, having already been reconstructed. The stored motion information may be transferred to the inter predictor 221 to be utilized as motion information of the spatial neighboring block or motion information of the temporal neighboring block. The memory 270 may store reconstructed samples of reconstructed blocks in the current picture, and may transfer the reconstructed samples to the intra predictor 222.

FIG. 3 is a diagram for schematically explaining the configuration of a video/image decoding apparatus to which the embodiments of the present document may be applied.

Referring to FIG. 3, the decoding apparatus 300 may include and configured with an entropy decoder 310, a residual processor 320, a predictor 330, an adder 340, a filter 350, and a memory 360. The predictor 330 may include an inter predictor 331 and an intra predictor 332. The residual processor 320 may include a dequantizer 321 and an inverse transformer 322. The entropy decoder 310, the residual processor 320, the predictor 330, the adder 340, and the filter 350, which have been described above, may be configured by one or more hardware components (e.g., decoder chipsets or processors) according to an embodiment. Further, the memory 360 may include a decoded picture buffer (DPB), and may be configured by a digital storage medium. The hardware component may further include the memory 360 as an internal/external component.

When the bitstream including the video/image information is input, the decoding apparatus 300 may reconstruct the image in response to a process in which the video/image information is processed in the encoding apparatus illustrated in FIG. 2. For example, the decoding apparatus 300 may derive the units/blocks based on block split-related information acquired from the bitstream. The decoding apparatus 300 may perform decoding using the processing unit applied to the encoding apparatus. Therefore, the processing unit for the decoding may be, for example, a coding unit, and the coding unit may be split according to the quad-tree structure, the binary-tree structure, and/or the ternary-tree structure from the coding tree unit or the maximum coding unit. One or more transform units may be derived from the coding unit. In addition, the reconstructed image signal decoded and output through the decoding apparatus 300 may be reproduced through a reproducing apparatus.

The decoding apparatus 300 may receive a signal output from the encoding apparatus of FIG. 2 in the form of a bitstream, and the received signal may be decoded through the entropy decoder 310. For example, the entropy decoder 310 may parse the bitstream to derive information (e.g., video/image information) necessary for image reconstruction (or picture reconstruction). The video/image information may further include information on various parameter sets such as an adaptation parameter set (APS), a picture parameter set (PPS), a sequence parameter set (SPS), or a video parameter set (VPS). In addition, the video/image information may further include general constraint information. The decoding apparatus may further decode picture based on the information on the parameter set and/or the general constraint information. Signaled/received information and/or syntax elements described later in the present document may be decoded may decode the decoding procedure and obtained from the bitstream. For example, the entropy decoder 310 decodes the information in the bitstream based on a coding method such as exponential Golomb coding, context-adaptive variable length coding (CAVLC), or context-adaptive arithmetic coding (CABAC), and output syntax elements required for image reconstruction and quantized values of transform coefficients for residual. More specifically, the CABAC entropy decoding method may receive a bin corresponding to each syntax element in the bitstream, determine a context model by using a decoding target syntax element information, decoding information of a decoding target block or information of a symbol/bin decoded in a previous stage, and perform an arithmetic decoding on the bin by predicting a probability of occurrence of a bin according to the determined context model, and generate a symbol corresponding to the value of each syntax element. In this case, the CABAC entropy decoding method may update the context model by using the information of the decoded symbol/bin for a context model of a next symbol/bin after determining the context model. The information related to the prediction among the information decoded by the entropy decoder 310 may be provided to the predictor (inter predictor 332 and intra predictor 331), and residual values on which the entropy decoding has been performed in the entropy decoder 310, that is, the quantized transform coefficients and related parameter information, may be input to the residual processor 320.

The residual processor 320 may derive a residual signal (residual block, residual samples, or residual sample array). Also, information on filtering among the information decoded by the entropy decoder 310 may be provided to the filter 350. Meanwhile, a receiving unit (not shown) for receiving a signal output from the encoding apparatus may be further configured as an internal/external element of the decoding apparatus 300, or the receiving unit may be a component of the entropy decoder 310. Meanwhile, the decoding apparatus according to the present document may be called a video/image/picture decoding apparatus, and the decoding apparatus may be divided into an information decoder (video/image/picture information decoder) and a sample decoder (video/image/picture sample decoder). The information decoder may include the entropy decoder 310, and the sample decoder may include at least one of the dequantizer 321, the inverse transformer 322, the adder 340, the filter 350, the memory 360, an inter predictor 332, and an intra predictor 331.

The dequantizer 321 may dequantize the quantized transform coefficients to output the transform coefficients. The dequantizer 321 may rearrange the quantized transform coefficients in a two-dimensional block form. In this case, the rearrangement may be performed based on a coefficient scan order performed by the encoding apparatus. The dequantizer 321 may perform dequantization for the quantized transform coefficients using a quantization parameter (e.g., quantization step size information), and acquire the transform coefficients.

The inverse transformer 322 inversely transforms the transform coefficients to acquire the residual signal (residual block, residual sample array).

The predictor 330 may perform the prediction of the current block, and generate a predicted block including the prediction samples of the current block. The predictor may determine whether the intra prediction is applied or the inter prediction is applied to the current block based on the information on prediction output from the entropy decoder 310, and determine a specific intra/inter prediction mode.

The predictor 330 may generate a prediction signal based on various prediction methods to be described later. For example, the predictor may apply intra prediction or inter prediction for prediction of one block, and may simultaneously apply intra prediction and inter prediction. This may be called combined inter and intra prediction (CIIP). In addition, the predictor may be based on an intra block copy (IBC) prediction mode or based on a palette mode for prediction of a block. The IBC prediction mode or the palette mode may be used for image/video coding of content such as games, for example, screen content coding (SCC). IBC may basically perform prediction within the current picture, but may be performed similarly to inter prediction in that a reference block is derived within the current picture. That is, IBC may use at least one of the inter prediction techniques described in the present document. The palette mode may be considered as an example of intra coding or intra prediction. When the palette mode is applied, information on the palette table and the palette index may be included in the video/image information and signaled.

The intra predictor 331 may predict the current block by referring to the samples in the current picture. The referred samples may be located in the neighborhood of the current block, or may be located apart from the current block according to the prediction mode. In intra prediction, prediction modes may include a plurality of non-directional modes and a plurality of directional modes. The intra predictor 331 may determine the prediction mode to be applied to the current block by using the prediction mode applied to the neighboring block.

The inter predictor 332 may derive a predicted block for the current block based on a reference block (reference sample array) specified by a motion vector on a reference picture. In this case, in order to reduce the amount of motion information being transmitted in the inter prediction mode, motion information may be predicted in the unit of blocks, subblocks, or samples based on correlation of motion information between the neighboring block and the current block. The motion information may include a motion vector and a reference picture index. The motion information may further include information on inter prediction direction (L0 prediction, L1 prediction, Bi prediction, and the like). In case of inter prediction, the neighboring block may include a spatial neighboring block existing in the current picture and a temporal neighboring block existing in the reference picture. For example, the inter predictor 332 may construct a motion information candidate list based on neighboring blocks, and derive a motion vector of the current block and/or a reference picture index based on the received candidate selection information. Inter prediction may be performed based on various prediction modes, and the information on the prediction may include information indicating a mode of inter prediction for the current block.

The adder 340 may generate a reconstructed signal (reconstructed picture, reconstructed block, or reconstructed sample array) by adding the obtained residual signal to the prediction signal (predicted block or predicted sample array) output from the predictor (including inter predictor 332 and/or intra predictor 331). If there is no residual for the processing target block, such as a case that a skip mode is applied, the predicted block may be used as the reconstructed block.

The adder 340 may be called a reconstructor or a reconstructed block generator. The generated reconstructed signal may be used for the intra prediction of a next block to be processed in the current picture, and as described later, may also be output through filtering or may also be used for the inter prediction of a next picture.

Meanwhile, a luma mapping with chroma scaling (LMCS) may also be applied in the picture decoding process.

The filter 350 may improve subjective/objective image quality by applying filtering to the reconstructed signal. For example, the filter 350 may generate a modified reconstructed picture by applying various filtering methods to the reconstructed picture, and store the modified reconstructed picture in the memory 360, specifically, in a DPB of the memory 360. The various filtering methods may include, for example, deblocking filtering, a sample adaptive offset, an adaptive loop filter, a bilateral filter, and the like.

The (modified) reconstructed picture stored in the DPB of the memory 360 may be used as a reference picture in the inter predictor 332. The memory 360 may store the motion information of the block from which the motion information in the current picture is derived (or decoded) and/or the motion information of the blocks in the picture having already been reconstructed. The stored motion information may be transferred to the inter predictor 332 so as to be utilized as the motion information of the spatial neighboring block or the motion information of the temporal neighboring block. The memory 360 may store reconstructed samples of reconstructed blocks in the current picture, and transfer the reconstructed samples to the intra predictor 331.

In the present document, the embodiments described in the filter 260, the inter predictor 221, and the intra predictor 222 of the encoding apparatus 200 may be applied equally or to correspond to the filter 350, the inter predictor 332, and the intra predictor 331.

The video/image coding method according to the present document may be performed based on the following partitioning structure. Specifically, procedures of prediction, residual processing ((inverse) transform and (de) quantization), syntax element coding, and filtering to be described later may be performed based on CTU and CU (and/or TU and PU) derived based on the partitioning structure. A block partitioning procedure may be performed by the image partitioner 210 of the above-described encoding apparatus, and partitioning-related information may be (encoding) processed by the entropy encoder 240, and may be transferred to the decoding apparatus in the form of a bitstream. The entropy decoder 310 of the decoding apparatus may derive the block partitioning structure of the current picture based on the partitioning-related information obtained from the bitstream, and based on this, may perform a series of procedures (e.g., prediction, residual processing, block/picture reconstruction, in-loop filtering, and the like) for image decoding. The CU size and the TU size may be equal to each other, or a plurality of TUs may be present within a CU region. Meanwhile, the CU size may generally represent a luma component (sample) coding block (CB) size. The TU size may generally represent a luma component (sample) transform block (TB) size. The chroma component (sample) CB or TB size may be derived based on the luma component (sample) CB or TB size in accordance with a component ratio according to a color format (chroma format, e.g., 4:4:4, 4:2:2, 4:2:0 and the like) of a picture/image. The TU size may be derived based on maxTbSize. For example, if the CU size is larger than the maxTbSize, a plurality of TUs (TBs) of the maxTbSize may be derived from the CU, and the transform/inverse transform may be performed in the unit of TU (TB). Further, for example, in case that intra prediction is applied, the intra prediction mode/type may be derived in the unit of CU (or CB), and neighboring reference sample derivation and prediction sample generation procedures may be performed in the unit of TU (or TB). In this case, one or a plurality of TUs (or TBs) may be present in one CU (or CB) region, and in this case, the plurality of TUs (or TBs) may share the same intra prediction mode/type.

Further, in the video/image coding according to the present document, an image processing unit may have a hierarchical structure. One picture may be partitioned into one or more tiles, bricks, slices, and/or tile groups. One slice may include one or more bricks. On brick may include one or more CTU rows within a tile. The slice may include an integer number of bricks of a picture. One tile group may include one or more tiles. One tile may include one or more CTUs. The CTU may be partitioned into one or more CUs. A tile represents a rectangular region of CTUs within a particular tile column and a particular tile row in a picture. A tile group may include an integer number of tiles according to a tile raster scan in the picture. A slice header may carry information/parameters that can be applied to the corresponding slice (blocks in the slice). In case that the encoding/decoding apparatus has a multi-core processor, encoding/decoding processes for the tiles, slices, bricks, and/or tile groups may be processed in parallel. In the present document, the slice or the tile group may be used exchangeably. That is, a tile group header may be called a slice header. Here, the slice may have one of slice types including intra (I) slice, predictive (P) slice, and bi-predictive (B) slice. In predicting blocks in I slice, inter prediction may not be used, and only intra prediction may be used. Of course, even in this case, signaling may be performed by coding the original sample value without prediction. With respect to blocks in P slice, intra prediction or inter prediction may be used, and in case of using the inter prediction, only uni-prediction can be used. Meanwhile, with respect to blocks in B slice, the intra prediction or inter prediction may be used, and in case of using the inter prediction, up to bi-prediction can be maximally used.

The encoder may determine the tile/tile group, brick, slice, and maximum and minimum coding unit sizes in consideration of the coding efficiency or parallel processing, or according to the characteristics (e.g., resolution) of a video image, and information for them or information capable of inducing them may be included in the bitstream.

The decoder may obtain information representing the tile/tile group, brick, and slice of the current picture, and whether the CTU in the tile has been partitioned into a plurality of coding units. By making such information be obtained (transmitted) only under a specific condition, the efficiency can be enhanced.

The slice header (slice header syntax) may include information/parameters that can be commonly applied to the slice. APS (APS syntax) or PPS (PPS syntax) may include information/parameters that can be commonly applied to one or more pictures. The SPS (SPS syntax) may include information/parameters that can be commonly applied to one or more sequences. The VPS (VPS syntax) may include information/parameters that can be commonly applied to multiple layers. The DPS (DPS syntax) may include information/parameters that can be commonly applied to overall video. The DPS may include information/parameters related to concatenation of a coded video sequence (CVS).

In the present document, an upper-level syntax may include at least one of the APS syntax, PPS syntax, SPS syntax, VPS syntax, DPS syntax, and slice header syntax.

Further, for example, information about partitioning and configuration of the tile/tile group/brick/slice may be configured by an encoding end through the upper-level syntax, and may be transferred to the decoding apparatus in the form of a bitstream.

In the present document, at least one of quantization/dequantization and/or transform/inverse transform may be omitted. When the quantization/dequantization is omitted, the quantized transform coefficient may be referred to as a transform coefficient. When the transform/inverse transform is omitted, the transform coefficient may be called a coefficient or a residual coefficient or may still be called the transform coefficient for uniformity of expression.

In the present document, the quantized transform coefficient and the transform coefficient may be referred to as a transform coefficient and a scaled transform coefficient, respectively. In this case, the residual information may include information on transform coefficient(s), and the information on the transform coefficient(s) may be signaled through residual coding syntax. Transform coefficients may be derived based on the residual information (or information on the transform coefficient(s)), and scaled transform coefficients may be derived through inverse transform (scaling) on the transform coefficients. Residual samples may be derived based on inverse transform (transform) of the scaled transform coefficients. This may be applied/expressed in other parts of the present document as well.

As described above, the encoding apparatus may perform various encoding methods such as exponential Golomb, context-adaptive variable length coding (CAVLC), and context-adaptive binary arithmetic coding (CABAC). In addition, the decoding apparatus may decode information in a bitstream based on a coding method such as exponential Golomb coding, CAVLC or CABAC, and output a value of a syntax element required for image reconstruction and quantized values of transform coefficients related to residuals. For example, the coding methods described above may be performed as in the following contents.

FIG. 4 exemplarily shows context-adaptive binary arithmetic coding (CABAC) for encoding a syntax element.

An encoding process of CABAC may include a process of transforming an input signal into a binary value through binarization in case that the input signal is not the binary value, but is a syntax element. If the input signal is already the binary value (i.e., if the value of the input signal is the binary value), the corresponding input signal may be bypassed without being binarized. Here, each binary number 0 or 1 constituting the binary value may be called a bin. For example, if a binary string after binarization is 110, each of 1, 1, and 0 is called one bin. The bin(s) for one syntax element may indicate a value of the syntax element.

The binarized bins of the syntax element may be input to a regular encoding engine or a bypass encoding engine. The regular encoding engine may allocate a context model that reflects a probability value to the corresponding bin, and may code the corresponding bin based on the allocated context model. The regular encoding engine may update the context model for the corresponding bin after performing the coding for the respective bins. The bin being coded as described above may be called a context-coded bin.

Meanwhile, in case that the binarized bins of the syntax element are input to the bypass encoding engine, they may be coded as follows. For example, the bypass encoding engine of the encoding apparatus omits a procedure of estimating probability for the input bin and a procedure of updating a probability model applied to the bin after encoding. In case that the bypass encoding is applied, the encoding apparatus may encode the input bin by applying a uniform probability distribution instead of allocating the context model, and through this, the encoding speed can be improved. The bin being encoded as described above may be called a bypass bin.

Entropy decoding performs the same process as the entropy encoding as described above in reverse order. For example, in case that the syntax element is decoded based on the context model, the decoding apparatus may receive the bin corresponding to the syntax element through a bitstream. Further, the decoding apparatus may determine the context model using the syntax element, decoding information of a decoding target block or a neighboring block, or information of a symbol/bin decoded in a previous step, and may derive the value of the syntax element by performing arithmetic decoding of the bin through prediction of the probability of occurrence of the received bin according to the determined context model. Thereafter, the context model of the bin being next decoded may be updated with the determined context model.

Further, for example, in case that the syntax element is bypass-decoded, the decoding apparatus may receive the bin corresponding to the syntax element through the bitstream, and may decode the input bin by applying the uniform probability distribution. In this case, the procedure of deriving the context model of the syntax element and the procedure of updating the context model applied to the bin after decoding may be omitted.

The residual samples may be derived as quantized transform coefficients through transform and quantization processes. The quantized transform coefficients may be called transform coefficients. In this case, the transform coefficients in the block may be signaled in the form of residual information. The residual information may include a residual coding syntax. That is, the encoding apparatus may configure and encode the residual coding syntax based on the residual information to output the residual coding syntax in the form of a bitstream, and the decoding apparatus may derive the residual (quantized) transform coefficients by decoding the residual coding syntax obtained from the bitstream. As described below, the residual coding syntax may include syntax elements representing whether transform has been applied to the corresponding block, where is the location of the last effective transform coefficient in the block, whether the effective transform coefficient is present in the subblock, and what is the size/sign of the effective transform coefficient.

For example, the (quantized) transform coefficient may be encoded and/or decoded based on the syntax elements, such as last_sig_coeff_x_prefix, last_sig_coeff_y_prefix, last_sig_coeff_x_suffix, last_sig_coeff_y_suffix, coded_sub_block_flag, sig_coeff_flag, par_level_flag, abs_level_gtX_flag, abs_remainder, coeff_sign_flag, and dec_abs_level. This may be called residual (data) coding or (transform) coefficient coding. The syntax elements related to the encoding/decoding of the residual data may be represented as in Table 1 (see FIGS. 15A-15F) or Table 2 (see FIGS. 16-16F).

In Table 1 and Table 2, transform_skip_flag represents whether the transform for the associated block is omitted. The transform_skip_flag may be a syntax element of the transform skip flag. The associated block may be a coding block (CB) or a transform block (TB). With respect to the transform (and quantization) and residual coding procedure, the CB and the TB may be interchangeably used. For example, residual samples may be derived with respect to the CB, and (quantized) transform coefficients may be derived through the transform and quantization for the residual samples. Through the residual coding procedure, information (e.g., syntax elements) efficiently representing the location, size, sign, and the like of the (quantized) transform coefficients may be generated and signaled. The quantized transform coefficients may be simply called transform coefficients. In general, if the CB is not larger than the maximum TB, the size of the CB may be equal to the size of the TB, and in this case, the target block being transformed (and quantized) and residual-coded may be called the CB or the TB. Meanwhile, if the CB is larger than the maximum TB, the target block being transformed (and quantized) and residual-coded may be called the TB. Hereinafter, although it is explained that the syntax elements related to the residual coding are signaled in the unit of a transform block (TB), this is merely exemplary, and the TB may be interchangeably used with the CB as described above.

The syntax for the residual coding according to the transform skip flag may be the same as that in Table 3 or Table 4.

TABLE 3

Descriptor

residual_ts_coding( x0, y0, log2TbWidth, log2TbHeight, cIdx ) {

log2SbSize = ( Min( log2TbWidth, log2TbHeight ) < 2 ? 1 : 2 )

numSbCoeff = 1 << ( log2SbSize << 1 )

lastSubBlock = ( 1 << ( log2TbWidth + log2TbHeight − 2 * log2SbSize ) ) − 1

inferSbCbf = 1

MaxCcbs = 2 * ( 1 << log2TbWidth ) * ( 1<< log2TbHeight )

for( i =0; i <= lastSubBlock; i++ ) {

xS = DiagScanOrder[ log2TbWidth − log2SbSize ][ log2TbHeight − log2SbSize ][ i ][ 0 ]

yS = DiagScanOrder[ log2TbWidth − log2SbSize ][ log2TbHeight − log2SbSize ][ i ][ 1 ]

if( ( i != lastSubBlock | | !inferSbCbf ) {

coded_sub_block_flag[ xS ][ yS ]
ae(v)

}

if( coded_sub_block_flag[ xS ][ yS ] && i < lastSubBlock )

inferSbCbf = 0

/* First scan pass */

inferSbSigCoeffFlag = 1

for( n = 0; n <= numSbCoeff − 1; n++ ) {

xC = ( xS << log2SbSize ) + DiagScanOrder[ log2SbSize ][ log2SbSize ][ n ][ 0 ]

yC = ( yS << log2SbSize ) + DiagScanOrder[ log2SbSize ][ log2SbSize ][ n ][ 1 ]

if( coded_sub_block_flag[ xS ][ yS ] &&

( n != numSbCoeff − 1 | | !inferSbSigCoeffFlag ) ) {

sig_coeff_flag[ xC ][ yC ]
ae(v)

MaxCcbs− −

if( sig_coeff_flag[ xC ][ yC ] )

inferSbSigCoeffFlag = 0

}

CoeffSignLevel[ xC ][ yC ] = 0

if( sig_coeff_flag[ xC ][ yC ] {

coeff_sign_flag[ n ]
ae(v)

MaxCcbs− −

CoeffSignLevel[ xC ][ yC ] = ( coeff_sign_flag[ n ] > 0 ? −1 : 1 )

abs_level_gtx_flag[ n ][ 0 ]
ae(v)

MaxCcbs− −

if( abs_level_gtx_flag[ n ][ 0 ] ) {

par_level_flag[ n ]
ae(v)

MaxCcbs− −

}

}

AbsLevelPassX[ xC ][ yC ] =

sig_coeff_flag[ xC ][ yC ] + par_level_flag[ n ] + abs_level_gtx_flag[ n ][ 0 ]

}

/* Greater than X scan pass (numGtXFlags=5) */

for( n = 0; n <= numSbCoeff − 1; n++ ) {

xC = ( xS << log2SbSize ) + DiagScanOrder[ log2SbSize ][ log2SbSize ][ n ][ 0 ]

yC = ( yS << log2SbSize ) + DiagScanOrder[ log2SbSize ][ log2SbSize ][ n ][ 1 ]

for( j = 1; j < 5; j++ ) {

if( abs_level_gtx_flag[ n ][ j − 1 ] )

abs_level_gtx_flag[ n ]
ae(v)

MaxCcbs− −

AbsLevelPassX[ xC ][ yC ] + = 2 * abs_level_gtx_flag[ n ][ j ]

}

}

/* remainder scan pass */

for( n = 0; n <= numSbCoeff − 1; n++ ) {

xC = ( xS << log2SbSize ) + DiagScanOrder[ log2SbSize ][ log2SbSize ][ n ][ 0 ]

yC = ( yS << log2SbSize ) + DiagScanOrder[ log2SbSize ][ log2SbSize ][ n ][ 1 ]

if( abs_level_gtx_flag[ n ][ 4 ] )

abs_remainder[ n ]
ae(v)

if( intra_bdpcm_flag = = 0 ) {

absRightCoeff = abs( TransCoeffLevel[ x0 ][ y0 ][ cIdx ][ xC − 1 ][ yC ] )

absBelowCoeff = abs( TransCoeffLevel[ x0 ][ y0 ][ cIdx ][ xC ][ yC − 1 ] )

predCoeff = Max( absRightCoeff, absBelowCoeff )

if( AbsLevelPassX[ xC ][ yC ] + abs_remainder[ n ] = = 1 && predCoeff >

0 )

TransCoeffLevel[ x0 ][ y0 ][ cIdx ][ xC ][ yC ] =

( 1 − 2 * coeff_sign_flag[ n ] ) * predCoeff

else if( AbsLevelPassX[ xC ][ yC ] + abs_remainder[ n ] <= predCoeff )

TransCoeffLevel[ x0 ][ y0 ][ cIdx ][ xC ][ yC ] = ( 1 − 2 * coeff_sign_flag[ n ] ) *

( AbsLevelPassX[ xC ][ yC ] + abs_remaninder[ n ] − 1)

else

TransCoeffLevel[ x0 ][ y0 ][ cIdx ][ xC ][ yC ] = ( 1 − 2 * coeff_sign_flag[ n ] ) *

( AbsLevelPassX[ xC ][ yC ] + abs_remainder[ n ] )

} else

TransCoeffLevel[ x0 ][ y0 ][ cIdx ][ xC ][ yC ] = ( 1 − 2 * coeff_sign_flag[ n ] ) *

( AbsLevelPassX[ xC ][ yC ] + abs_remainder[ n ] )

}

}

}

TABLE 4

Descriptor

residual_ts_coding( x0, y0, log2TbWidth, log2TbHeight, cIdx ) {

log2SbW = ( Min( log2TbWidth, log2TbHeight ) < 2 ? 1 : 2 )

log2SbH = log2SbW

if( log2TbWidth + log2TbHeight > 3 )

if( log2TbWidth < 2 ) {

log2SbW = log2TbWidth

log2SbH = 4 − log2SbW

} else if( log2TbHeight < 2 ) {

log2SbH = log2TbHeight

log2SbW = 4 − log2SbH

}

numSbCoeff = 1 << ( log2SbW + log2SbH )

lastSubBlock = ( 1 << ( log2TbWidth + log2TbHeight − ( log2SbW + log2SbH ) ) ) − 1

inferSbCbf = 1

RemCcbs = ( ( 1 << ( log2TbWidth + log2TbHeight ) ) * 7 ) >> 2

for( i =0; i <= lastSubBlock; i++ ) {

xS = DiagScanOrder[ log2TbWidth − log2SbW ][ log2TbHeight − log2SbH ][ i ][ 0 ]

yS = DiagScanOrder[ log2TbWidth − log2SbW ][ log2TbHeight − log2SbH ][ i ][ 1 ]

if( i != lastSubBlock | | !inferSbCbf )

sb_coded_flag[ xS ][ yS ]
ae(v)

if( sb_coded_flag[ xS ][ yS ] && i < lastSubBlock )

inferSbCbf = 0

/* First scan pass */

inferSbSigCoeffFlag = 1

lastScanPosPass1 = −1

for( n = 0; n <= numSbCoeff − 1 && RemCcbs >= 4; n++ ) {

xC = ( xS << log2SbW ) + DiagScanOrder[ log2SbW ][ log2SbH ][ n ][ 0 ]

yC = ( yS << log2SbH ) + DiagScanOrder[ log2SbW ][ log2SbH ][ n ][ 1 ]

lastScanPosPass1 = n

if( sb_coded_flag[ xS ][ yS ] &&

( n != numSbCoeff − 1 | | !inferSbSigCoeffFlag ) ) {

sig_coeff_flag[ xC ][ yC ]
ae(v)

RemCcbs− −

if( sig_coeff_flag[ xC ][ yC ] )

inferSbSigCoeffFlag = 0

}

CoeffSignLevel[ xC ][ yC ] = 0

if( sig_coeff_flag[ xC ][ yC ] ) {

coeff_sign_flag[ n ]
ae(v)

RemCcbs− −

CoeffSignLevel[ xC ][ yC ] = ( coeff_sign_flag[ n ] > 0 ? −1 : 1 )

abs_level_gtx_flag[ n ][ 0 ]
ae(v)

RemCcbs− −

if( abs_level_gtx_flag[ n ][ 0 ] ) {

par_level_flag[ n ]
ae(v)

RemCcbs− −

}

}

AbsLevelPass1[ xC ][ yC ] =

sig_coeff_flag[ xC ][ yC ] + par_level_flag[ n ] + abs_level_gtx_flag[ n ][ 0 ]

}

/* Greater than X scan pass (numGtXFlags=5) */

lastScanPosPass2 = −1

for( n = 0; n <= numSbCoeff − 1 && RemCcbs >= 4; n++ ) {

xC = ( xS << log2SbW ) + DiagScanOther[ log2SbW ][ log2SbH ][ n ][ 0 ]

yC = ( yS << log2SbH ) + DiagScanOrder[ log2SbW ][ log2SbH ][ n ][ 1 ]

AbsLevelPass2[ xC ][ yC ] = AbsLevelPass1[ xC ][ yC ]

for( j = 1; j < 5; j++ ) {

if( abs_level_gtx_flag[ n ][ j − 1 ] ) {

abs_level_gtx_flag[ n ][ j ]
ae(v)

RemCcbs− −

}

AbsLevelPass2[ xC ][ yC ] += 2 * abs_level_gtx_flag[ n ][ j ]

}

lastScanPosPass2 = n

}

/* remainder scan pass */

for( n = 0; n <= numSbCoeff − 1; n++ ) {

xC = ( xS << log2SbW ) + DiagScanOrder[ log2SbW ][ log2SbH ][ n ][ 0 ]

yC = ( yS << log2SbH ) + DiagScanOrder[ log2SbW ][ log2SbH ][ n ][ 1 ]

if( ( n <= lastScanPosPass2 && AbsLevelPass2[ xC ][ yC ] >= 10 ) | |

( n > lastScanPosPass2 && n <= lastScanPosPass1 &&

AbsLevelPass1[ xC ][ yC ] >= 2 ) | |

( n > lastScanPosPass1 && sb_coded_flag[ xS ][ yS ] ) )

abs_remanieder[ n ]
ae(v)

if( n <= lastScanPosPass2 )

AbsLevel[ xC ][ yC ] = AbsLevelPass2[ xC ][ yC ] + 2 * abs_remainder[ n ]

else if(n <= lastScanPosPass1 )

AbsLevel[ xC ][ yC ] = AbsLevelPass1[ xC ][ yC ] + 2 * abs_remainder[ n ]

else { /* bypass */

AbsLevel[ xC ][ yC ] = abs_remainder[ n ]

if( abs_remainder[ n ] )

coeff_sign_flagp[ n ]
ae(v)

}

if( BdpcmFlag[ x0 ][ y0 ][ cIdx ] = = 0 && n <= lastScanPosPass1 ) {

absLeftCoeff = xC > 0 ? AbsLevel[ xC − 1 ][ yC ] ) : 0

absAboveCoeff = yC > 0 ? AbsLevel[ xC ][ yC − 1 ] ) : 0

predCoeff = Max( absLeftCoeff, absAboveCoeff )

if( AbsLevel[ xC ][ yC ] = = 1 && predCoeff > 0 )

AbsLevel[ xC ][ yC ] = predCoeff

else if( AbsLevel[ xC ][ yC ] > 0 && AbsLevel[ xC ][ yC ] <= predCoeff )

AbsLevel[ xC ][ yC ]− −

}

TransCoeffLevel[ x0 ][ y0 ][ cIdx ][ xC ][ yC ] = ( 1 − 2 * coeff_sign_flag[ n ] ) *

AbsLevel[ xC ][ yC ]

}

}

According to the present embodiment, the residual coding may be branched according to the value of the transform skip flag transform_skip_flag. That is, different syntax elements may be used for the residual coding based on the value of the transform skip flag (based on whether to skip the transform). The residual coding being used in case that the transform skip is not applied (i.e., in case that the transform is applied) may be called regular residual coding (RRC), and the residual coding in case that the transform skip is not applied (i.e., in case that the transform is not applied) may be called transform skip residual coding (TSRC). Further, the regular residual coding may also be called general residual coding. Further, the regular residual coding may be called a regular residual coding syntax structure, and the transform skip residual coding may be called a transform skip residual coding syntax structure. Table 1 and Table 2 may represent a residual coding syntax element in case that the value of transform_skip_flag is 0, that is, in case that the transform is applied, and Table 3 and Table 4 may represent a residual coding syntax element in case that the value of transform_skip_flag is 1, that is, in case that the transform is not applied.

Specifically, as an example, the transform skip flag indicating whether the transform of the transform block is skipped may be parsed, and whether the transform skip flag is 1 may be determined. If the value of the transform skip flag is 0, syntax elements last_sig_coeff_x_prefix, last_sig_coeff_y_prefix, last_sig_coeff_x_suffix, last_sig_coeff_y_suffix, sb_coded_flag, sig_coeff_flag, abs_level_gtx_flag, par_level_flag, abs_remainder, dec_abs_level, and/or coeff_sign_flag for residual coefficients of the transform block as illustrated in Table 1 or Table 2 may be parsed, and the residual coefficients may be derived based on the syntax elements. In this case, the syntax elements may be sequentially parsed, and the parsing order may be changed. Further, the abs_level_gtx_flag may represent the abs_level_gt1_flag and/or the abs_level_gt3_flag. For example, the abs_level_gtx_flag[n][0] may be an example of the first transform coefficient level flag abs_level_gt1_flag, and the abs_level_gtx_flag[n][1] may be an example of the second transform coefficient level flag abs_level_gt3_flag.

In an embodiment, the encoding apparatus may encode (x, y) position information of the last non-zero transform coefficient in a transform block based on the syntax elements last_sig_coeff_x_prefix, last_sig_coeff_y_prefix, last_sig_coeff_x_suffix, and last_sig_coeff_y_suffix. More specifically, the last_sig_coeff_x_prefix represents a prefix of a column position of a last significant coefficient in a scanning order within the transform block, the last_sig_coeff_y_prefix represents a prefix of a row position of the last significant coefficient in the scanning order within the transform block, the last_sig_coeff_x_suffix represents a suffix of a column position of the last significant coefficient in the scanning order within the transform block, and the last_sig_coeff_y_suffix represents a suffix of a row position of the last significant coefficient in the scanning order within the transform block. Here, the significant coefficient may represent a non-zero coefficient. In addition, the scanning order may be a right diagonal scanning order. Alternatively, the scanning order may be a horizontal scanning order or a vertical scanning order. The scanning order may be determined based on whether intra/inter prediction is applied to a target block (a CB or a CB including a TB) and/or a specific intra/inter prediction mode.

Thereafter, the encoding apparatus may divide the transform block into 4×4 sub-blocks, and then indicate whether there is a non-zero coefficient in the current sub-block using a 1-bit syntax element coded_sub_block_flag for each 4×4 sub-block.

If a value of coded_sub_block_flag is 0, there is no more information to be transmitted, and thus, the encoding apparatus may terminate the encoding process on the current sub-block. Conversely, if the value of coded_sub_block_flag is 1, the encoding apparatus may continuously perform the encoding process on sig_coeff_flag. Since the sub-block including the last non-zero coefficient does not require encoding for the coded_sub_block_flag and the sub-block including the DC information of the transform block has a high probability of including the non-zero coefficient, coded_sub_block_flag may not be coded and a value thereof may be assumed as 1.

If the value of coded_sub_block_flag is 1 and thus it is determined that a non-zero coefficient exists in the current sub-block, the encoding apparatus may encode sig_coeff_flag having a binary value according to a reverse scanning order. The encoding apparatus may encode the 1-bit syntax element sig_coeff_flag for each transform coefficient according to the scanning order. If the value of the transform coefficient at the current scan position is not 0, the value of sig_coeff_flag may be 1. Here, in the case of a subblock including the last non-zero coefficient, sig_coeff_flag does not need to be encoded for the last non-zero coefficient, so the coding process for the sub-block may be omitted. Level information coding may be performed only when sig_coeff_flag is 1, and four syntax elements may be used in the level information encoding process. More specifically, each sig_coeff_flag[xC][yC] may indicate whether a level (value) of a corresponding transform coefficient at each transform coefficient position (xC, yC) in the current TB is non-zero. In an embodiment, the sig_coeff_flag may correspond to an example of a syntax element of a significant coefficient flag indicating whether a quantized transform coefficient is a non-zero significant coefficient.

The remaining level value after encoding of the sig_coeff_flag may be derived as in the following equation. That is, the syntax element remAbsLevel representing the level value to be encoded may be derived as in the following equation.

remAbsLevel
[
n
]

=

❘
"[LeftBracketingBar]"

coeff
[
n
]

❘
"[RightBracketingBar]"

-
1

[

Equation
⁢

1

]

Here, coeff[n] means a real transform coefficient value.

Further, the abs_level_gtx_flag[n][0] may represent whether the remAbsLevel[n] at a corresponding scanning position n is larger than 1. For example, if the value of the abs_level_gtx_flag[n][0] is 0, an absolute value of the transform coefficient at the corresponding position may be 1. Further, if the value of the abs_level_gtx_flag[n][0] is 1, the remAbsLevel[n] representing the level value to be encoded hereinafter may be updated as in the equation below.

remAbsLevel
[
n
]

=

remAbsLevel
[
n
]

-
1

[

Equation
⁢

2

]

Further, the value of the least significant coefficient (LSB) of the remAbsLevel[n] described in Equation 2 may be encoded through the par_level_flag as in Equation 3 below.

par_level
⁢

_flag
[
n
]

=

remAbsLevel
[
n
]

&

⁢

1

[

Equation
⁢

3

]

Here, par_level_flag[n] may represent a parity of the transform coefficient level (value) at the scanning position n.

After the par_leve_flag[n] is encoded, the transform coefficient level value remAbsLevel[n] to be encoded may be updated as in the following equation.

remAbsLevel
[
n
]

=

remAbsLevel
[
n
]

≫
1

[

Equation
⁢

4

]

abs_level_gtx_flag[n][1] may represent whether the remAbsLevel at the corresponding scanning position n is larger than 3. Only in case that the abs_level_gtx_flag[n][1] is 1, encoding of the abs_remainder[n] can be performed. The relationship between the real transform coefficient value coeff and each syntax element may be as in the following equation.

[

Equation
⁢

5

]

❘
"[LeftBracketingBar]"

coeff
[
n
]

❘
"[RightBracketingBar]"

=

sig_coeff
⁢

_flag
[
n
]

+

abs_level
⁢
_gtX
⁢

_flag
[
n
]

[
0
]

+

par_level
⁢

_flag
[
n
]

+

2
*

(

abs_level
⁢
_gtX
⁢

_flag
[
n
]

[
1
]

+

abs_remainder
[
n
]

)

Further, the following Table 5 represents examples related to the above-described Equation 5.

TABLE 5

|coeff[n]|
sig_coeff_flag[n]
abs_level_gtX_flag[n][0]
par_level_flag[n]
abs_level_gtX_flag[n][1]
abs_remainder[n]

0
0

1
1
0

2
1
1
0
0

3
1
1
1
0

4
1
1
0
1
0

5
1
1
1
1
0

6
1
1
0
1
1

7
1
1
1
1
1

8
1
1
0
1
2

9
1
1
1
1
2

10
1
1
0
1
3

11
1
1
1
1
3

. . .
. . .
. . .
. . .

Here, |coeff[n]| represents the transform coefficient level (value), and may be indicated as AbsLevel for the transform coefficient. Further, the sign of each coefficient may be encoded using coeff_sign_flag that is a one-bit symbol.

Further, as another example, if the value of the transform skip flag is 1, syntax elements sb_coded_flag, sig_coeff_flag, coeff_sign_flag, abs_level_gtx_flag, par_level_flag, and/or abs_remainder for the residual coefficient of the transform block as shown in Table 3 or Table 4 may be parsed, and the residual coefficient may be derived based on the syntax elements. In this case, the syntax elements may be sequentially parsed, or the parsing order may be changed. Further, the abs_level_gtx_flag may represent abs_level_gt1_flag, abs_level_gt3_flag, abs_level_gt5_flag, abs_level_gt7_flag, and/or abs_level_gt9_flag. For example, abs_level_gtx_flag[n][j] may be a flag representing whether an absolute value or level (value) of the transform coefficient at the scanning position n is larger than (j<<1)+1. In some cases, the (j<<1)+1 may be replaced by a predetermined threshold value, such as a first threshold value or a second threshold value.

Meanwhile, the CABAC provides high performance, but has the disadvantage that the throughput performance is not good. This is caused by the regular encoding engine of the CABAC, and since the regular encoding (encoding through the regular encoding engine of the CABAC) uses the probability state and range updated through encoding of the previous bin, it may show high data dependency, and it may take a lot of time to read the probability interval and to determine the current state. The throughput problem of the CABAC may be solved by limiting the number of context-coded bins. For example, as in the above-described Table 5, the sum of bins used to express the sig_coeff_flag[n], abs_level_gtx_flag[n][0], par_level_flag[n], and abs_level_gtx_flag[n][1] may be limited to 1.75 per pixel in the transform block depending on the size of the transform block. In this case, if all of the limited number of context-coded bins are used to encode the context element, the encoding apparatus may perform the bypass coding by binarizing the remaining coefficients through a binarization method to be described later without using the context coding. In other words, if the number of coded context-coded bins becomes TU width*TU height*1.75 in TU, the sig_coeff_flag[n], abs_level_gtx_flag[n][0], par_level_flag[n], and abs_level_gtx_flag[n][1], which are coded as the context-coded bins, may not be coded any more, and the value of |coeff[n]| may be directly coded into dec_abs_level[n] as in the following Table 6.

TABLE 6

|coeff[n]|
dec_abs_level[n]

0
0

1
1

2
2

3
3

4
4

5
5

6
6

7
7

8
8

9
9

10
10

11
11

. . .
. . .

In this case, the sign of each coefficient may be coded using coeff_sign_flag[n] that is a one-bit symbol.

FIG. 5 exemplarily illustrates transform coefficients in a 4×4 block.

The 4×4 block of FIG. 5 represents an example of quantized coefficients. The block illustrated in FIG. 5 may be a 4×4 transform block, or may be a 4×4 subblock of an 8×8, 16×16, 32×32, or 64×64 transform block. The 4×4 block of FIG. 5 may represent a luma block or a chroma block. However, this is merely exemplary, and in the present embodiment, the transform of a large block size (maximum 64×64 size) is possible, and this may be useful for a high-resolution video (e.g., 1080p and 4K sequence). The high-frequency transform coefficients may be zeroed with respect to the transform block of which the size (width or height or both width and height) is 64, and only low-frequency coefficients may be maintained. For example, in case of the M×N transform block having the block width of M and the block height of N, only 32 left columns of the transform coefficients may be maintained when M is 64. Further, only 32 top rows of the transform coefficients may be maintained when N is 64.

In case that the transform skip mode is used for a block having a large size, the whole block may be used without being zeroed even with respect to any values. Since the maximum transform size configurable in SPS is supported, the encoding apparatus may adaptively select the transform size having a length of 16, 32, or 64 at maximum in accordance with necessity of a specific implementation. Specifically, binarization of the coefficient position coding that is not the last 0 may be coded based on the reduced TU size, and the selection of the context model for the coefficient position coding that is not the last 0 may be determined by the original TU size.

In FIG. 5, as an example, the encoding results for the coefficients being scanned reverse diagonally are illustrated. In FIGS. 5, n (0 to 15) designates the scan position of the coefficients according to the reverse diagonal scan. If n is 15, it represents the coefficient at the bottom right corner, which is firstly scanned in the 4×4 block, and if n is 0, it represent the coefficient at the top left corner, which is lastly scanned.

Meanwhile, as described above, if the input signal is not a binary value, but is the syntax element, the encoding apparatus may transform the input signal into a binary value by binarizing the value of the input signal. Further, the decoding apparatus may derive the binarized value (i.e., binarized bin) of the syntax element by decoding the syntax element, and may derive the value of the syntax element by reversely binarizing the binarized value. The binarization process may be performed as a truncated rice (TR) binarization process, a k-th order Exp-Golomb (EGk) binarization process, a limited k-th order Exp-Golomb (limited EGk) or fixed-length (FL) binarization process to be described later. Further, the reverse binarization process may be performed based on the TR binarization process, the EGk binarization process, the limited EGk binarization process, or the FL binarization process, and may represent a process of deriving the value of the syntax element.

For example, the TR binarization process may be performed as follows.

An input of the TR binarization process may be a request for TR binarization and cMax and cRiceParam for the syntax element. Further, an output of the TR binarization process may be the TR binarization for a value symbolVal corresponding to a bin string.

As an example, if a suffix bin string for the syntax element is present, the TR bin string for the syntax element may be a concatenation of a prefix bin string and a suffix bin string, and if the suffix bin string is not present, the TR bin string for the syntax element may be the prefix bin string. For example, the prefix bin string may be derived as follows.

The prefix value of the symbolVal may be derived as in the following equation.

prefixVal
=

symbolVal
≫
cRiceParam

[

Equation
⁢

6

]

Here, prefix Val may represent a prefix value of symbolVal. The prefix of the TR bin string (i.e., prefix bin string) may be derived as follows.

For example, if the prefix Val is smaller than cMax>>cRiceParam, the prefix bin string may be a bit string having a length of prefix Val+1, which is indexed by binIdx. That is, if the prefix Val is smaller than cMax>>cRiceParam, the prefix bin string may be a bit string having the number of bits prefix Val+1 indicated by the binIdx. The bin for the binIdx that is smaller the prefix Val may be 1. Further, the bin for the binIdx equal to the prefix Val may be 0.

For example, a bin string derived as unary binarization for the prefix Val may be as in Table 7 below.

TABLE 7

prefixVal
Bin string

0
0

1
1
0

2
1
1
0

3
1
1
1
0

4
1
1
1
1
0

5
1
1
1
1
1
0

. . .

binIdx
0
1
2
3
4
5

Meanwhile, if the prefix Val is not smaller than cMax>>cRiceParam, the prefix bin string may be a bit string of which the length is the cMax>>cRiceParam and of which all bins are 1.

Further, the suffix bin string of the TR bin string may be present in case that the cMax is larger than the symbolVal and the cRiceParam is larger than 0. For example, the prefix bin string may be derived as follows.

The suffix value of the symbolVal for the syntax element may be derived as in the following equation.

suffixVal
=

symbolVal
-

(

(
prefixVal
)

≪
cRiceParam

)

[

Equation
⁢

7

]

Here, the suffix Val may represent a suffix value of the symbolVal.

The suffix (i.e., suffix bin string) of the TR bin string may be derived based on an FL binarization process for the suffix Val of which the cMax value is 1<<cRiceParam)−1.

For the input parameter cRiceParam=0, the TR binarization is exactly a truncated unary binarization and it is always invoked with a cMax value equal to the largest possible value of the syntax element being decoded.

Meanwhile, the EGk binarization process may be performed as follows.

The input of the EGk binarization process may be a request for the EGk binarization. Further, the output of the EGk binarization process may be the EGk binarization for the value symbolVal corresponding to the bin string.

The bit string of the EGk binarization process for the symbolVal may be derived as follows.

TABLE 8

absV = Abs( symbolVal )

stopLoop = 0

do

if( absV >= ( 1 << k ) ) {

put( 1 )

absV = absV − ( 1 << k )

k++

} else {

put( 0 )

while( k− − )

put( ( absV >> k ) & 1 )

stopLoop = 1

}

while( !stopLoop )

Referring to Table 8, a binary value X may be added at the end of a bin string through each call of put (X). Here, X may be 0 or 1.

Further, a limited EGk binarization process may be performed as follows.

The input of the limited EGk binarization process may be a request for the limited EGk binarization and the rice parameter riceParam. Further, the output of the limited EGk binarization process may be the limited EGk binarization for the value symbolVal related to the corresponding bin string.

The bin string of the limited EGk binarization process for the symbolVal may be derived as follows.

TABLE 9

codeValue = symbolVal >> riceParam

preExtLen = 0

while( ( preExtLen < maxPreExtLen ) && ( codeValue > ( ( 2 << preExtLen ) − 2 ) ) ) {

preExtLen++

put( 1 )

}

if( preExtLen = = maxPreExtLen )

escapeLength = log2TransformRange

else {

escapeLength = preExtLen + riceParam

put( 0 )

}

symbolVal = symbolVal − ( ( ( 1 << preExtLen ) − 1 ) << riceParam )

while( ( escapeLength− − ) > 0 )

put( ( symbolVal >> escapeLength ) & 1 )

Referring to Table 9, a binary value X may be added at the end of the bin string through each call of put (X). Here, X may be 0 or 1.

Variables log 2TransformRange and maxPrefixExtensionLength may be derived as follows.

log
⁢
2
⁢
TransformRange

=
15

[

Equation
⁢

8

]

maxPrefixExtensionLength
=

26
-

log
⁢
2
⁢
TransformRange

Further, the FL binarization process may be performed as follows.

The input of the FL binarization process may be a request for the FL binarization and cMax for the syntax element. Further, the output of the FL binarization process may be the FL binarization for the value symbolVal corresponding to the bin string.

The FL binarization may be configured using the bin string having the number of bits of a fixed length of the symbol value symbolVal. Here, the fixed length may be derived as in the following equation.

fixedLength
=

Ceil
(

Log
⁢
2
⁢

(

c
⁢
Max

+
1

)

)

[

Equation
⁢

9

]

That is, the bin string for the symbol value symbolVal may be derived through the FL binarization, and the bin length (i.e., the number of bits) of the bin string may be the fixed length.

Indexing of bins for the FL binarization may be a method for using a value increasing in order from the most significant bit to the least significant bit. For example, the bin index related to the most significant bit may be binIdx=0.

Meanwhile, a binarization process for a syntax element abs_remainder[n] among residual information may be performed as follows.

The input of the binarization process for the abs_remainder[n] may be a request for binarization of the syntax element abs_remainder[n], color component cIdx, luma position (x0, y0) representing the top left sample of the current luma transform block based on the top left luma sample of a picture, current coefficient scan position (xC, yC), binary logarithm of the transform block width longTbWidth, and binary logarithm of the transform block height log 2TbHeight. The output of the binarization process for the abs_remainder may be binarization of the abs_remainder (i.e., binarized bin string of the abs_remainder). Through the binarization process, available bin strings for the abs_remainder may be derived.

The rice parameter cRiceParam for the abs_remainder[n] may be derived through a rice parameter deriving process being performed through inputting of the color component cIdx and the luma position (x0, y0), the current coefficient scan position (xC, yC), log 2TbWidth that is the binary logarithm of the width of the transform block, and log 2TbHeight that is the binary logarithm of the height of the transform block. The rice parameter deriving process will be described in detail later.

The cMax for the abs_remainder[n] being currently coded may be derived based on the rice parameter cRiceParam. For example, the cMax may be derived as in the following equation.

c
⁢
Max

=

6
≪
cRiceParam

[

Equation
⁢

10

]

Meanwhile, in case that the suffix bin string is present, the binarization for the syntax element abs_remainder[n], that is, the bin string for the abs_remainder[n] may be the concatenation of the prefix bin string and the suffix bin string. In case that the suffix bin string is not present, the bin string for the abs_remainder[n] may be the prefix bin string.

For example, the prefix bin string for the abs_remainder[n] may be derived as follows.

The prefix value prefix Val of the abs_remainder[n] may be derived as in the following equation.

prefixVal
=

Min
⁡
(

c
⁢
Max

,

abs_remainder
[
n
]

)

[

Equation
⁢

11

]

The prefix bin string of the abs_remainder[n] may be derived through the TR binarization process for the prefix Val using the cMax and the cRiceParam as an input.

If the prefix bin string is equal to a bit string of which all bits are 1 and the bit length is 6, the suffix bin string of the abs_remainder[n] may be present, and may be derived as follows.

The suffix value suffix Val of the abs_remainder[n] may be derived as in the following equation.

suffixVal
=

abs_remainder
[
n
]

-

c
⁢
Max

[

Equation
⁢

12

]

The suffix bin string of the abs_remainder[n] may be derived through the limited EGk binarization process for the binarization of the suffix Val that uses cRiceParam+1 and cRiceParam as an input.

The rice parameter for the abs_remainder[n] may be derived by the following process.

The input of the rice parameter deriving process may be a base level baseLevel, a color component index cIdx, a luma position (x0, y0), a current coefficient scan position (xC, yC), a binary logarithm log 2TbWidth of the width of the transform block and a binary logarithm log 2TbHeight of the height of the transform block. The luma position (x0, y0) may represent the top left sample of the current luma transform block based on the top left luma sample of the picture. The output of the rice parameter deriving process may be the rice parameter cRiceParam.

For example, the variable locSumAbs may be derived by a pseudo code disclosed in the following table based on the arrangement AbsLevel [x][y] for the given component index cIdx and the transform block having the top left luma position (x0, y0).

TABLE 10

If transform_skip_flag[ x0 ][ y0 ] is equal to 1, trafoSkip is set equal to 1 and the following

applies:

locSumAbs = 0

if( xC > 0 )

locSumAbs += AbsLevel[ xC − 1 ][ yC ]

if( yC > 0 )

locSumAbs += AbsLevel[ xC ][ yC − 1 ]

locSumAbs = Clip3( 0, 31, locSumAbs )

Otherwise (transform_skip_flag[ x0 ][ y0 ] is equal to 0), trafoSkip is set equal to 0 and the

following applies:

locSumAbs = 0

if( xC < (1 << log2TbWidth) − 1 ) {

locSumAbs += AbsLevel[ xC + 1 ][ yC ]

if( xC < (1 << log2TbWidth) − 2 )

locSumAbs += AbsLevel[ xC + 2 ][ yC ]

if( yC < (1 << log2TbHeight) − 1 )

locSumAbs += AbsLevel[ xC + 1 ][ yC + 1 ]

}

if( yC < (1 << log2TbHeight) − 1 ) {

locSumAbs += AbsLevel[ xC ][ yC + 1 ]

if( yC < (1 << log2TbHeight) − 2 )

locSumAbs += AbsLevel [ xC ][ yC + 2 ]

}

locSumAbs = Clip3( 0, 31, locSumAbs − baseLevel * 5 )

In case that baseLevel is 0 in Table 10, the variable s may be configured as Max (0, QState−1), and the rice parameter cRiceParam and the variable ZeroPos[n] may be derived as in the following Table 11 based on the variables locSumAbs, trafoSkip, and s. If the baseLevel is larger than 0, the rice parameter cRiceParam may be derived as in Table 11 based on the variables locSumAbs and trafoSkip.

TABLE 11

trafoSkip
s
locSumAbs
0
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15

0

cRiceParam
0
0
0
0
0
0
0
1
1
1
1
1
1
1
2
2

1

cRiceParam
0
0
0
0
0
0
0
0
0
0
0
0
1
1
1
1

0
ZeroPos[ n ]
0
0
0
0
0
1
2
2
2
2
2
2
4
4
4
4

1
ZeroPos[ n ]
1
1
1
1
2
3
4
4
4
6
6
6
8
8
8
8

2
ZeroPos[ n ]
1
1
2
2
2
3
4
4
4
6
6
6
8
8
8
8

locSumAbs
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31

0

cRiceParam
2
2
2
2
2
2
2
2
2
2
2
2
3
3
3
3

1

cRiceParam
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2

0
ZeroPos[ n ]
4
4
4
4
4
4
4
8
8
8
8
8
16
16
16
16

1
ZeroPos[ n ]
4
4
12
12
12
12
12
12
12
12
16
16
16
16
16
16

2
ZeroPos[ n ]
8
8
12
12
12
12
12
12
12
16
16
16
16
16
16
16

Meanwhile, the binarization process for the syntax element dec_abs_level among the residual information may be performed as follows.

The input of the binarization process for the dec_abs_level may be a request for the binarization of the syntax element dec_abs_level[n], a color component cIdx, a luma position (x0, y0), a current coefficient scan position (xC, yC), a binary logarithm log 2TbWidth of the width of the transform block, and a binary logarithm log 2TbHeight of the height of the transform block. The luma position (x0, y0) may represent the top left sample of the current luma transform block based on the top left luma sample of the picture.

The output of the binarization process for the dec_abs_level may be binarization of the dec_abs_level (i.e., binarized bin string of the dec_abs_level). Through the binarization process, available bin strings for the dec_abs_level may be derived.

The rice parameter cRiceParam for the dec_abs_level[n] may be derived through the rice parameter deriving process that is performed through inputting of the color component cIdx and the luma position (x0, y0), the current coefficient scan position (xC, yC), the log 2TbWidth that is the binary logarithm of the width of the transform block, and the log 2TbHeight that is the binary logarithm of the height of the transform block. The rice parameter deriving process will be described in detail later.

Further, for example, the cMax for the dec_abs_level[n] may be derived based on the rice parameter cRiceParam. The cMax may be derived as in Equation 10.

Meanwhile, in case that the suffix bin string is present, the binarization for the dec_abs_level[n], that is, the bin string for the dec_abs_level[n], may be the concatenation of the prefix bin string and the suffix bin string. Further, in case that the suffix bin string is not present, the bin string for the dec_abs_level[n] may be the prefix bin string.

For example, the prefix bin string may be derived as follows.

The prefix value prefix Val of the dec_abs_level[n] may be derived as in the following equation.

prefixVal
=

Min
⁡
(

c
⁢
Max

,

dec_abs
⁢

_level
[
n
]

)

[

Equation
⁢

13

]

The prefix bin string of the dec_abs_level[n] may be derived through the TR binarization process for the prefix Val that uses the cMax and the cRiceParam as an input.

If the prefix bin string is equal to the bit string of which all bits are 1 and the bit length is 6, the suffix bin string of the dec_abs_level[n] may be present, and this may be derived as follows.

The suffix value suffix Val of the dec_abs_level[n] may be derived as in the following equation.

suffixVal
=

dec_abs
⁢

_level
[
n
]

-

c
⁢
Max

[

Equation
⁢

14

]

The suffix bin string of the dec_abs_level[n] may be derived through the limited EGk binarization process for the binarization of the suffix Val of which the Exp-Golomb degree K is configured as cRiceParam+1.

The rice parameter for the dec_abs_level[n] may be derived by the pseudo code of Table 10.

Meanwhile, the regular residual coding (RRC) and the transform skip residual coding (TSRC) as described above may have the following difference.

For example, the rice parameter cRiceParam of the syntax element abs_remainder[ ] in the regular residual coding may be derived as described above, but the rice parameter cRiceParam of the syntax element abs_remainder[ ] in the transform skip residual coding may be derived as 1. That is, for example, in case that the transform skip is applied to the current block (e.g., current TB), the rice parameter cRiceParam for the abs_remainder[ ] of the transform skip residual coding for the current block may be derived as 1.

Further, referring to Table 1 to Table 4, although abs_level_gtx_flag[n][0] and/or abs_level_gtx_flag[n][1] may be signaled in the regular residual coding, abs_level_gtx_flag[n][0], abs_level_gtx_flag[n][1], abs_level_gtx_flag[n][2], abs_level_gtx_flag[n][3], and abs_level_gtx_flag[n][4] may be signaled in the transform skip residual coding. Here, the abs_level_gtx_flag[n][0] may be represented as abs_level_gt1_flag or the first coefficient level flag, the abs_level_gtx_flag[n][1] may be represented as abs_level_gt3_flag or the second coefficient level flag, the abs_level_gtx_flag[n][2] may be represented as abs_level_gt5_flag or the third coefficient level flag, the abs_level_gtx_flag[n][3] may be represented as abs_level_gt7_flag or the fourth coefficient level flag, and the abs_level_gtx_flag[n][4] may be represented as abs_level_gt9_flag or the fifth coefficient level flag. Specifically, the first coefficient level flag may be a flag representing whether the coefficient level is larger than the first threshold value (e.g., 1), the second coefficient level flag may be a flag representing whether the coefficient level is larger than the second threshold value (e.g., 3), the third coefficient level flag may be a flag representing whether the coefficient level is larger than the third threshold value (e.g., 5), the fourth coefficient level flag may be a flag representing whether the coefficient level is larger than the fourth threshold value (e.g., 7), and the fifth coefficient level flag may be a flag representing whether the coefficient level is larger than the fifth threshold value (e.g., 9).

As described above, in comparison to the regular residual coding, the transform skip residual coding may further include abs_level_gtx_flag[n][2], abs_level_gtx_flag[n][3], and abs_level_gtx_flag[n][4] in addition to abs_level_gtx_flag[n][0] and abs_level_gtx_flag[n][1].

Further, for example, in the regular residual coding, the syntax element coeff_sign_flag may be bypass-coded, whereas in the transform skip residual coding, the syntax element coeff_sign_flag may be bypass-coded or context-coded.

The following description has been prepared to explain specific examples of the present document. Hereinafter, since the name of a specific device or the name of specific signal/information is exemplarily presented, the technical feature of the present specification is not limited to the specific names used in the following explanation.

Hereinafter, disclosed is a method for efficiently deriving a rice parameter for binarization of a syntax (e.g., dec_abs_level) representing a level value of a transform coefficient in level coding and/or a syntax (e.g., syntax having abs_remainder or a value corresponding to a level value of −x, here, x may have a value in a range of 1 to 10) representing a residual level value.

FIG. 6 illustrates an example of neighboring coefficients for deriving a rice parameter.

According to the present embodiment, in case of determining a rice parameter for a target transform coefficient, availability of neighboring transform coefficients may be used. Through this, a codeword length after binarization in level coding may be reduced, and thus the coding performance may be improved.

As an example, FIG. 6 illustrates 5 neighboring transform coefficients that can be used to determine a rice parameter for a target transform coefficient in case that regular residual coding is applied to a target transform block. In case of deriving the rice parameter, in order to determine availability of corresponding neighboring coefficients for the target transform coefficient, the encoding apparatus and/or the decoding apparatus may confirm whether the corresponding neighboring coefficients are within a boundary of the transform block and/or within an effective area after high-frequency zeroing. If a specific neighboring coefficient gets out of a range of the current transform block or gets out of an effective area after zero-out, the value thereof is not used to derive the rice parameter.

Referring to FIG. 6, the neighboring transform coefficient adjacent to a bottom side of the target transform coefficient is located within the boundary of the current transform block. Accordingly, the neighboring transform coefficient adjacent to the bottom side of the target transform coefficient may be used to determine the rice parameter. Since the remaining neighboring transform coefficients (neighboring transform coefficients adjacent to the right side and the bottom-right side of the target transform coefficient) are located out of the boundary of the current transform block, they are not used to determine the rice parameter. Accordingly, in case of the transform block illustrated in FIG. 6, the number of available neighboring coefficients when deriving the rice parameter for the target transform coefficient is 1.

In case of deriving the rice parameter for the target transform coefficient as described above, the number of available neighboring coefficients may not be always 5, but may be 1, 2, 3, or 4 depending on the locations of the target transform coefficient and/or the neighboring transform coefficient. However, referring to Table 10, in the related art, level values of corresponding neighboring coefficients are accumulated through accumulation of accessible (or available) effective neighboring coefficients for the target transform coefficient, but in case of finally deriving locSumAbs that directly exerts an influence on the determination of the rice parameter, the number of the effective neighboring coefficient(s) is not considered. In other words, in the related art, in case of regular residual coding, the rice parameter is derived using the sum locSumAbs of absolute values (level values) of 5 neighboring transform coefficients based on the coefficient (target transform coefficient) intended to be currently coded (encoded/decoded). Accordingly, according to the related art, there is a problem in that an accurate rice parameter is not derived.

In order to solve such a problem, in the present embodiment, a pseudocode disclosed in the following table may be used when deriving the rice parameter for the target transform coefficient.

TABLE 12

locSumAbs = 0

locNumNb = 0

if( xC < ( 1 << log2TbWidth) − 1 ) {

locNumNb++

locSumAbs += AbsLevel[ xC + 1 ][ yC ]

if( xC < ( 1 << log2TbWidth ) − 2 ) {

locNumNb++

locSumAbs += AbsLevel[ xC + 2 ][ yC ]

}

if( yC < ( 1 << log2TbHeight) − 1 ) {

locNumNb++

locSumAbs += AbsLevel[ xC + 1 ][ yC + 1 ]

}

}

if( yC < ( 1 << log2TbHeight) − 1 ) {

locNumNb++

locSumAbs += AbsLevel[ xC ][ yC + 1 ]

if( yC < ( 1 << log2TbHeight) − 2 ) {

locNumNb++

locSumAbs += AbsLevel [ xC ][ yC + 2 ]

}

}

locSumAbs = Clip3( 0, 31, locSumAbs − baseLevel * locNumNb )

The pseudocode of Table 12 may be used in case that a value of a transform skip flag transform_skip_flag for the corresponding transform block is 0, that is, in case that the corresponding transform block is not the transform skip block.

In Table 12, locNumNb represents the number of available neighboring coefficient(s) for the target transform coefficient. log 2TbWidth and log 2TbHeight represent the horizontal/vertical sizes of the transform block, respectively. In case that high-frequency zeroing is performed for the transform block, log 2TbWidth and log 2TbHeight may represent the horizontal/vertical sizes of an effective area after zero-out. baseLevel may have various integer values, such as 0, 1, 2, or 10, depending on a syntax type being coded prior to level value coding, but the present embodiment is not limited to such values.

The encoding apparatus and the decoding apparatus may derive the rice parameter using the number locSumAbs of available neighboring coefficient(s) for the target transform coefficient derived through Table 12 based on a rice parameter lookup table. In this case, as the rice parameter lookup table, the table of Table 11 may be used. Alternatively, a table different from Table 11 may be used for the coding performance.

According to the present embodiment, through the method as described above, a more accurate rice parameter may be derived, and through this, the length of a codeword being used when the level value is binarized may be reduced to improve the coding performance.

FIG. 7 illustrates another example of neighboring coefficients for deriving a rice parameter.

As an example, FIG. 7 illustrates 2 neighboring transform coefficients that can be used to determine a rice parameter for a target transform coefficient in case that transform skip residual coding is applied to a target transform block. In case of deriving the rice parameter, in order to determine availability of corresponding neighboring coefficients for the target transform coefficient, the encoding apparatus and/or the decoding apparatus may confirm whether the corresponding neighboring coefficients are within a boundary of the transform block. If a specific neighboring coefficient gets out of a range of the current transform block, the value thereof is not used to derive the rice parameter.

In FIG. 7, since both the neighboring transform coefficient adjacent to a left side of the target transform coefficient and the neighboring transform coefficient adjacent to a top side of the target transform coefficient are located within the boundary of the current transform block, they may be used to determine the rice parameter for the target transform coefficient. In case that the neighboring transform coefficient adjacent to the left side of the target transform coefficient and/or the neighboring transform coefficient adjacent to the top side of the target transform coefficient are located out of the current transform block, the corresponding neighboring transform coefficient is not used to derive the rice parameter.

In case of deriving the rice parameter for the target transform coefficient as described above, the number of available neighboring coefficient(s) may not be always 2, but may be 0, 1, or 2 depending on the locations of the target transform coefficient and/or the neighboring transform coefficient. However, referring to Table 10, in the related art, the level values of the corresponding neighboring coefficients are accumulated through accumulation of the accessible (or available) effective neighboring coefficients for the target transform coefficient, but in case of finally deriving the locSumAbs that directly exerts an influence on the determination of the rice parameter, the number of the effective neighboring coefficients is not considered. In other words, in case of the transform skip residual coding in the related art, the rice parameter is derived using the sum locSumAbs of absolute values (level values) of 2 neighboring transform coefficients based on the coefficient (target transform coefficient) intended to be currently coded (encoded/decoded). Accordingly, according to the related art, there is a problem in that an accurate rice parameter is not derived.

In order to solve such a problem, in the present embodiment, a pseudocode disclosed in the following table may be used when deriving the rice parameter for the target transform coefficient.

TABLE 13

locSumAbs = 0

locNumNb = 0

if( xC > 0 ) {

locNumNb++

locSumAbs += AbsLevel[ xC − 1][ yC ]

}

if( yC > 0 ){

locNumNb++

locSumAbs += AbsLevel[ xC ][ yC − 1 ]

}

locSumAbs = Clip3( 0, 31, locSumAbs − baseLevel * locNumNb )

The pseudocode of Table 13 may be used in case that the value of a transform skip flag transform_skip_flag for the corresponding transform block is 1, that is, in case that the corresponding transform block is the transform skip block.

In Table 13, locNumNb represents the number of available neighboring coefficient(s) for the target transform coefficient. log 2TbWidth and log 2TbHeight represent the horizontal/vertical sizes of the transform block, respectively. baseLevel may have various integer values, such as 0, 1, 2, or 10, depending on a syntax type being coded prior to the level value coding, but the present embodiment is not limited to such values.

The encoding apparatus and the decoding apparatus may derive the rice parameter using the number locSumAbs of available neighboring coefficient(s) for the target transform coefficient derived through Table 13 based on a rice parameter lookup table. In this case, as the rice parameter lookup table, the table of Table 11 may be used. Alternatively, a table different from Table 11 may be used for the coding performance.

The rice parameter derived through the above-described process may be used for binarization of a syntax (e.g., dec_abs_level) representing the level value of the transform coefficient in the level coding and/or a syntax (e.g., syntax having abs_remainder or a value corresponding to a level value of −x) representing the residual level value. Here, x may have a value in the range of 1 to 10, but the present embodiment is not limited thereto.

Meanwhile, in deriving the rice parameter for the target transform coefficient, the pseudocode disclosed in Table 12 and the pseudocode disclosed in Table 13 may be separately used, or may be used in combination with each other as in Table 14 below.

TABLE 14

If transform_skip_flag[ x0 ][ y0 ] is equal to 1, trafoSkip is set equal to 1 and the following applies:

locSumAbs = 0

locNumNb = 0

if( xC > 0 ) {

locNumNb++

locSumAbs += AbsLevel[ xC − 1 ][ yC ]

}

if( yC > 0 ){

locNumNb++

locSumAbs += AbsLevel[ xC ][ yC − 1 ]

}

locSumAbs = Clip3( 0, 31, locSumAbs − baseLevel * locNumNb )

Otherwise (transform_skip_flag[ x0 ][ y0 ] is equal to 0), trafoSkip is set equal to 0 and the following applies:

locSumAbs = 0

locNumNb = 0

if( xC < (1 << log2TbWidth) − 1 ) {

locNumNb++

locSumAbs += AbsLevel[ xC + 1 ][ yC ]

if( xC < (1 << log2TbWidth) − 2 ) {

locNumNb++

locSumAbs += AbsLevel[ xC + 2 ][ yC ]

}

if( yC < (1 << log2TbHeight) − 1 ) {

locNumNb++

locSumAbs += AbsLevel[ xC + 1 ][ yC + 1 ]

}

}

if( yC < (1 << log2TbHeight) − 1 ) {

locNumNb++

locSumAbs += AbsLevel[ xC ][ yC + 1 ]

if( yC < ( 1 << log2TbHeight) − 2 ) {

locNumNb++

locSumAbs += AbsLevel [ xC ][ yC + 2 ]

}

}

locSumAbs = Clip3( 0, 31, locSumAbs − baseLevel * locNumNb )

In Table 14, locNumNb represents the number of available neighboring coefficient(s) for the target transform coefficient. locSumAbs represents the sum of level value(s) of available neighboring coefficient(s) for the target transform coefficient. log 2TbWidth and log 2TbHeight represent the horizontal/vertical sizes of the transform block, respectively. In case that high-frequency zeroing is performed for the transform block, log 2TbWidth and log 2TbHeight may represent the horizontal/vertical sizes of an effective area after zero-out. baseLevel may have various integer values, such as 0, 1, 2, or 10, depending on a syntax type being coded prior to the level value coding.

FIGS. 8 and 9 schematically illustrate an entropy encoding method and an example of related components according to an embodiment of the present document.

A rice parameter deriving method disclosed in FIG. 8 may be performed by the encoding apparatus 200 disclosed in FIG. 2 and FIG. 9. Specifically, for example, S800 and S810 of FIG. 8 may be performed by a rice parameter deriver 241 of an entropy encoder 240. S820 of FIG. 8 may be performed by a binarizer 242 of the entropy encoder 240, and S830 of FIG. 8 may be performed by an entropy encoding processor 243 of the entropy encoder 240.

The entropy encoding method disclosed in FIG. 8 may include the above-described embodiments in the present document.

Referring to FIG. 8 and FIG. 9, the entropy encoder performs a residual coding procedure for (quantized) transform coefficients. The encoding apparatus may perform residual coding of the (quantized) transform coefficients in the current block (current CB or current TB) in a scan order. The encoding apparatus may generate and encode various syntax elements for residual information, for example, as indicated in Table 1 to Table 4. As an example, with respect to the (quantized) transform coefficients in the current block, the encoding apparatus may code at least one of sig_coeff_flag[xC][yC], abs_level_gtx_flag[n][0], par_level_flag[n], abs_level_gtx_flag[n][1], abs_remainder[n], and dec_abs_level[n]. In this case, the encoding apparatus may derive the value of the abs_remainder[n] or dec_abs_level[n], and may derive the rice parameter for the abs_remainder[n] or dec_abs_level[n].

As disclosed in embodiments of the present document, the rice parameter for (the transform coefficient of) the current scanning position may be derived based on the number locNumNb of the available neighboring coefficients for the target transform coefficient. Specifically, the rice parameter deriver of the entropy encoder may generate at least one of information representing the level value of the target transform coefficient in the current transform block or information representing the residual level value (S800). Here, the information representing the level value of the transform coefficient may correspond to the dec_abs_level syntax element, and the information representing the residual level value may correspond to the abs_remainder syntax element.

Thereafter, the rice parameter deriver of the entropy encoder may derive the rice parameter for at least one of the information representing the level value of the target transform coefficient or the information representing the residual level value based on the neighboring transform coefficient of the target transform coefficient (S810). As an example, the rice parameter deriver of the entropy encoder may determine that the corresponding neighboring transform coefficient is available based on whether the neighboring transform coefficient of the target transform coefficient is present within the boundary of the transform block and/or within the effective area after the high-frequency zeroing. Specifically, the rice parameter deriver of the entropy encoder may derive the number of available neighboring transform coefficients for the target transform coefficient based on whether the neighboring transform coefficient of the target transform coefficient is present within the boundary of the transform block and/or whether the neighboring transform coefficient of the target transform coefficient is present within the effective area after the high-frequency zeroing, and may derive a sum of level values of the available neighboring transform coefficients. Further, based on at least one of pseudocodes indicated in Table 12 to Table 14, the rice parameter for the target transform coefficient may be derived using a difference value between the sum of the level values of the available neighboring transform coefficients and a product between a base level and the number of the available neighboring transform coefficients. Here, the sum of the level values of the available neighboring transform coefficients may be equal to or larger than 0 and equal to or smaller than 31. Further, if the current transform block is not a transform skip block, that is, if regular residual coding is applied to the current transform block, the number of the available neighboring transform coefficients may have an integer value that is equal to or smaller than 5. If the current transform block is the transform skip block, that is, if transform skip residual coding is applied to the current transform block, the number of the available neighboring transform coefficients may have an integer value that is equal to or smaller than 2.

As an example, the rice parameter deriver of the entropy encoder may use the table of Table 11 in deriving the rice parameter. Specifically, in case that the current transform block is not the transform skip block, the rice parameter deriver of the entropy encoder may derive the value of the rice parameter as 0 if the sum of the level values of the available neighboring transform coefficients is equal to or larger than 0 and equal to or smaller than 6, may derive the value of the rice parameter as 1 if the sum of the level values of the available neighboring transform coefficients is equal to or larger than 7 and equal to or smaller than 13, may derive the value of the rice parameter as 2 if the sum of the level values of the available neighboring transform coefficients is equal to or larger than 14 and equal to or smaller than 27, and may derive the value of the rice parameter as 3 if the sum of the level values of the available neighboring transform coefficients is equal to or larger than 28 and equal to or smaller than 31. Further, in case that the current transform block is the transform skip block, the rice parameter deriver of the entropy encoder may derive the value of the rice parameter as 0 if the sum of the level values of the available neighboring transform coefficients is equal to or larger than 0 and equal to or smaller than 11, may derive the value of the rice parameter as 1 if the sum of the level values of the available neighboring transform coefficients is equal to or larger than 12 and equal to or smaller than 24, and may derive the value of the rice parameter as 2 if the sum of the level values of the available neighboring transform coefficients is equal to or larger than 25 and equal to or smaller than 31.

The binarizer of the entropy encoder may derive a bin string by performing binarization for at least one of the information representing the level value of the target transform coefficient or the information representing the residual level value based on the rice parameter derived by the rice parameter deriver (S820). According to the present embodiment, based on the rice parameter, the length of the bin string for values of the information representing the level value of the target transform coefficient or the information representing the residual level value may be adaptively determined.

It is apparent to those skilled in the art that the above-described rice parameter deriving procedure can be omitted with respect to other syntax elements (sig_coeff_flag[xC][yC], abs_level_gtx_flag[n][0], par_level_flag[n], abs_level_gtx_flag[n][1], and the like) which are FL-based binarized without using the rice parameter. With respect to the sig_coeff_flag[xC][yC], abs_level_gtx_flag[n][0], par_level_flag[n], abs_level_gtx_flag[n][1], and the like, TR-based binarization, EGk-based binarization, limited EGk-based binarization, or fixed-length binarization, which is not the rice parameter-based binarization, may be performed.

The entropy encoding processor of the entropy encoder may perform entropy encoding based on the bin string for at least one of the information representing the level value of the target transform coefficient or the information representing the residual level value (S830). The entropy encoding processor may perform context-based entropy encoding for the bin string based on the entropy coding technique, such as CABAC and the like, and the output thereof may be included in the bitstream. As described above, the bitstream may include various kinds of information for image/video decoding, such as prediction information, in addition to residual information including the information representing the level value of the target transform coefficient and/or the information representing the residual level value. The bitstream may be transferred to the decoding apparatus through a (digital) storage medium or a network.

FIGS. 10 and 11 schematically illustrate a rice parameter deriving method and an example of related components according to an embodiment of the present document.

A rice parameter deriving method disclosed in FIG. 10 may be performed by the decoding apparatus 300 disclosed in FIG. 3 and FIG. 11. Specifically, for example, S1000 and S1010 of FIG. 10 may be performed by the rice parameter deriver 311 of the entropy decoder 310. S1020 of FIG. 10 may be performed by the binarizer 312 of the entropy decoder 310, and S1030 of FIG. 10 may be performed by the entropy decoding processor 313 of the entropy decoder 310.

The entropy decoding method disclosed in FIG. 10 may include the above-described embodiments in the present document.

Referring to FIGS. 10 and 11, the entropy decoder may derive (quantized) transform coefficients by decoding encoded residual information. The decoding apparatus may derive the (quantized) transform coefficients by decoding the encoded residual information for the current block (current CB or current TB). The decoding apparatus may decode various syntax elements for the residual information, for example, as indicated in Table 1 to Table 4, and may derive the (quantized) transform coefficients based on analysis of values of related syntax elements.

As an example, with respect to the (quantized) transform coefficients in the current block, the decoding apparatus may decode at least one of sig_coeff_flag[xC][yC], abs_level_gtx_flag[n][0], par_level_flag[n], abs_level_gtx_flag[n][1], abs_remainder[n], and dec_abs_level[n]. In this case, the decoding apparatus may derive the rice parameter for the abs_remainder[n] and/or dec_abs_level[n].

As disclosed in embodiments of the present document, the rice parameter for (the transform coefficient of) the current scanning position may be derived based on the number locNumNb of the available neighboring coefficients for the target transform coefficient. Specifically, the rice parameter deriver of the entropy decoder may obtain the residual information for the current transform block from the bitstream (S1000). Here, the residual information may include at least one of information representing the level value of the target transform coefficient or information representing the residual level value in the current transform block. The information representing the level value of the transform coefficient may correspond to the dec_abs_level syntax element, and the information representing the residual level value may correspond to the abs_remainder syntax element.

The rice parameter deriver of the entropy decoder may derive the rice parameter for at least one of the information representing the level value of the target transform coefficient or the information representing the residual level value based on the neighboring transform coefficient of the target transform coefficient (S1010). As an example, the rice parameter deriver of the entropy decoder may determine that the corresponding neighboring transform coefficient is available based on whether the neighboring transform coefficient of the target transform coefficient is present within the boundary of the transform block and/or within the effective area after the high-frequency zeroing. Specifically, the rice parameter deriver of the entropy decoder may derive the number of available neighboring transform coefficients for the target transform coefficient based on whether the neighboring transform coefficient of the target transform coefficient is present within the boundary of the transform block and/or whether the neighboring transform coefficient of the target transform coefficient is present within the effective area after the high-frequency zeroing, and may derive a sum of level values of the available neighboring transform coefficients. Further, based on at least one of pseudocodes indicated in Table 12 to Table 14, the rice parameter for the target transform coefficient may be derived using a difference value between the sum of the level values of the available neighboring transform coefficients and a product between a base level and the number of the available neighboring transform coefficients. Here, the sum of the level values of the available neighboring transform coefficients may be equal to or larger than 0 and equal to or smaller than 31. Further, if the current transform block is not a transform skip block, that is, if regular residual coding is applied to the current transform block, the number of the available neighboring transform coefficients may have an integer value that is equal to or smaller than 5. If the current transform block is the transform skip block, that is, if transform skip residual coding is applied to the current transform block, the number of the available neighboring transform coefficients may have an integer value that is equal to or smaller than 2.

As an example, the rice parameter deriver of the entropy decoder may use the table of Table 11 in deriving the rice parameter. Specifically, in case that the current transform block is not the transform skip block, the rice parameter deriver of the entropy decoder may derive the value of the rice parameter as 0 if the sum of the level values of the available neighboring transform coefficients is equal to or larger than 0 and equal to or smaller than 6, may derive the value of the rice parameter as 1 if the sum of the level values of the available neighboring transform coefficients is equal to or larger than 7 and equal to or smaller than 13, may derive the value of the rice parameter as 2 if the sum of the level values of the available neighboring transform coefficients is equal to or larger than 14 and equal to or smaller than 27, and may derive the value of the rice parameter as 3 if the sum of the level values of the available neighboring transform coefficients is equal to or larger than 28 and equal to or smaller than 31. Further, in case that the current transform block is the transform skip block, the rice parameter deriver of the entropy decoder may derive the value of the rice parameter as 0 if the sum of the level values of the available neighboring transform coefficients is equal to or larger than 0 and equal to or smaller than 11, may derive the value of the rice parameter as 1 if the sum of the level values of the available neighboring transform coefficients is equal to or larger than 12 and equal to or smaller than 24, and may derive the value of the rice parameter as 2 if the sum of the level values of the available neighboring transform coefficients is equal to or larger than 25 and equal to or smaller than 31.

The binarizer of the entropy decoder may derive available bin strings for available values of the information representing the level value of the target transform coefficient or the information representing the residual level value by performing binarization for at least one of the information representing the level value of the target transform coefficient or the information representing the residual level value based on the rice parameter derived by the rice parameter deriver (S1020). According to the present embodiment, based on the rice parameter, the length of the bin string for the values of the information representing the level value of the target transform coefficient or the information representing the residual level value may be adaptively determined.

The entropy decoding processor of the entropy decoder may derive the level value of the target transform coefficient based on the bin string for at least one of the information representing the level value of the target transform coefficient or the information representing the residual level value (S1030). For example, the entropy decoding processor of the entropy decoder may compare the derived bin string with the available bin strings while sequentially parsing and decoding the respective bins for the abs_remainder[n] and/or the dec_abs_level[n]. If the derived bin string is the same as one of the available bin strings, the value corresponding to the corresponding bin string is derived as the value of the abs_remainder[n] and/or the dec_abs_level[n]. If not, the above comparison procedure is performed after further parsing and decoding the next bit in the bitstream. Through such a process, the corresponding information can be signaled using a variable length bit without using a start bit or an end bit for specific information (specific syntax element) in the bitstream. Through this, a relatively smaller number of bits can be allocated with respect to a small value, and thus the overall coding efficiency can be enhanced.

The entropy decoding processor of the entropy decoder may perform context-based entropy decoding for the respective bins in the bin string from the bitstream based on the entropy coding technique, such as CABAC or the like.

It is apparent to those skilled in the art that the above-described rice parameter deriving procedure can be omitted with respect to other syntax elements (sig_coeff_flag[xC][yC], abs_level_gtx_flag[n][0], par_level_flag[n], abs_level_gtx_flag[n][1], and the like) which are FL-based binarized without it using the rice parameter. With respect to the sig_coeff_flag[xC][yC], abs_level_gtx_flag[n][0], par_level_flag[n], abs_level_gtx_flag[n][1], and the like, TR-based binarization, EGk-based binarization, limited EGk-based binarization, or fixed-length binarization, which is not the rice parameter-based binarization, may be performed.

As described above, the bitstream may include various kinds of information for image/video decoding, such as prediction information, in addition to the residual information including the information on the abs_remainder[n] and/or the dec_abs_level[n]. The bitstream may be transferred to the decoding apparatus through a (digital) storage medium or a network.