KR20190013380A - Method and Apparatus for Expressing Position of Non-zero Coefficients - Google Patents

Abstract

The present disclosure relates to efficiently expressing locations of non-zero coefficients in a coefficient block. According to an aspect of the present disclosure, in encoding a coefficient block which can be divided into a plurality of sub-blocks, with respect to the coefficient block, a block determined to be a valid block is recursively divided into small blocks having uniform sizes until a single significant coefficient is reached, and whether each of generated blocks is a valid block is signaled. According to another aspect of the present disclosure, in encoding the coefficient block which can be divided into the plurality of sub-blocks, provided is a method for efficiently expressing locations of non-zero coefficients in the coefficient block on the basis of location of the last valid sub-block.

Description

Field of the Invention [0001] The present invention relates to a method and an apparatus for expressing the position of non-zero coefficients,

Field of the Invention [0002] The present invention relates to image encoding or decoding, and more particularly, to expressing the position of non-zero coefficients.

The contents described in this section merely provide background information on the present embodiment and do not constitute the prior art.

Video compression techniques include spatial prediction and / or temporal prediction to reduce or eliminate the inherent redundancy in video sequences. In block-based video coding, a video frame or slice may be partitioned into blocks. Each block can be further partitioned. Blocks within an intra-coded (I) frame or slice are encoded using spatial prediction for reference samples in neighboring blocks in the same frame or slice. Blocks in an inter-coded (P or B) frame or slice may use spatial prediction for reference samples in neighboring blocks in the same frame or slice or temporal prediction for reference samples in different reference frames. Spatial or temporal prediction results in a prediction block for the block to be coded. The residual data represents the pixel differences between the original block to be encoded and the prediction block.

The inter-coded block is encoded according to the motion vector indicating the block of reference samples forming the prediction block and the residual data indicating the difference between the coded block and the prediction block. The intra-coded block is encoded according to the intra-coding mode and the residual data. For further compression, the residual data may be transformed from the pixel domain to the transform domain to generate residual transform coefficients, which may then be quantized. The quantized transform coefficients arranged in a two-dimensional array may be scanned in a specific order to generate a one-dimensional vector of transform coefficients for entropy coding.

The main object of the present invention is to efficiently express the position of non-zero coefficients in a coefficient block having an array of coefficients that can be divided into a plurality of (sub) blocks.

According to an aspect of the present invention, there is provided an image encoding method comprising encoding a distribution of significant coefficients that are non-zero coefficients in a coefficient block. The image encoding method comprising the steps of: encoding a significant flag indicating whether the current coefficient block is a valid block having at least one non-zero coefficient; Wherein, when the valid flag indicates that the current coefficient block is a valid block, for the current coefficient block, the sub-block determined as the valid block is divided into n (n is 2 or more Determining whether each of the n subblocks is a valid block, while dividing the subblocks recursively into subblocks of equal size, Skipping the coding of the valid flag associated with the n subblocks if the size of the n subblocks is greater than or equal to a predetermined threshold value; And encoding the valid flag associated with the n subblocks if the size of the n subblocks is less than a predetermined threshold value.

According to another aspect of the present invention there is provided an image decoding method comprising determining a distribution of significant coefficients that are non-zero coefficients in a coefficient block. The image decoding method comprising: parsing, from a bitstream, a significant flag indicating whether the current coefficient block is a valid block having at least one non-zero coefficient; Wherein, when the valid flag indicates that the current coefficient block is a valid block, for the current coefficient block, the sub-block determined as the valid block is divided into n (n is 2 or more Determining whether each of the n subblocks is a valid block, while dividing the subblocks recursively into subblocks of equal size of the number of subblocks. If the size of the n subblocks is greater than or equal to a predetermined threshold value, the step of determining whether each of the n subblocks is an effective block may include the step of skipping the parsing of the validity flag associated with the n subblocks, ; And parsing valid flags associated with the n subblocks if the size of the n subblocks is less than a preset threshold value.

According to yet another aspect of the present invention, there is provided an image decoding apparatus comprising determining a distribution of significant coefficients that are non-zero coefficients in a coefficient block, the apparatus comprising: a memory; And one or more processors, wherein the one or more processors are configured to perform a method comprising the steps of: The method includes parsing, from the bitstream, a significant flag indicating whether the current coefficient block is a valid block having at least one non-zero coefficient; If the valid flag indicates that the current coefficient block is a valid block, for the current coefficient block, a sub-block determined as a valid block is divided into n equal-sized sub- The method comprising the steps of: determining whether each of the n subblocks is a valid block, while dividing the blocks recursively into blocks. If the size of the n subblocks is greater than or equal to a predetermined threshold value, the step of determining whether each of the n subblocks is an effective block may include the step of skipping the parsing of the validity flag associated with the n subblocks, ; And parsing valid flags associated with the n subblocks if the size of the n subblocks is less than a predetermined threshold.

According to another aspect of the present invention, there is provided a method of encoding a coefficient block having an array of coefficients divisible by coefficients of a plurality of subblocks. Wherein the last valid subblock is the last subblock in a subblock scan order having at least one non-zero coefficient; Coding information indicating a location of the determined last valid subblock; Wherein each sub-block except the left-most sub-block of the coefficient block among the sub-blocks preceding the last valid sub-block in the sub-block scan order is an effective sub-block having at least one non- Encoding a syntax element for a sub-block; And encoding coefficients of the last valid subblock, coefficients of the left-ordered subblock, and coefficients of subblocks indicating that the syntax element is an effective subblock having at least one non-zero coefficient.

According to another aspect of the present invention, there is provided a method of decoding a coefficient block having an array of coefficients divisible by coefficients of a plurality of subblocks. Wherein the decoding method of the coefficient block comprises decoding information indicating the position of the last valid subblock, wherein the last valid subblock is the last subblock in the subblock scan order having at least one non-zero coefficient; Wherein each sub-block except the left-most sub-block of the coefficient block among the sub-blocks preceding the last valid sub-block in the sub-block scan order is an effective sub-block having at least one non- Decoding the syntax element for the sub-block; And decoding coefficients of the last valid sub-block, coefficients of the left-ordered sub-block, and coefficients of sub-blocks indicating that the syntax element is an effective sub-block having at least one non-zero coefficient.

According to yet another aspect of the present invention, there is provided an apparatus for decoding a coefficient block having an array of coefficients divisible into coefficients of a plurality of subblocks. The apparatus comprising: a memory; And one or more processors, wherein the one or more processors are configured to perform a method comprising the steps of: The method includes decoding information indicating a position of a last valid subblock, wherein the last valid subblock is the last subblock in the subblock scan order having at least one non-zero coefficient; Wherein each sub-block except the left-most sub-block of the coefficient block among the sub-blocks preceding the last valid sub-block in the sub-block scan order is an effective sub-block having at least one non- Decoding the syntax element for the sub-block; And decoding coefficients of the last valid sub-block, coefficients of the left-ordered sub-block, and coefficients of sub-blocks indicating that the syntax element is an effective sub-block having at least one non-zero coefficient.

1 is an exemplary block diagram of an image encoding apparatus in which the techniques of the present disclosure may be implemented.
2 is an exemplary view of neighboring blocks of the current block.
3 is an exemplary block diagram of an image decoding apparatus capable of implementing the techniques of the present disclosure;
4 is a diagram illustrating exemplary scan schemes used for encoding quantized coefficients of a square coefficient block.
5 is a diagram illustrating a scan sequence of subblocks and coefficients in more detail for a diagonal scan scheme.
6 is a diagram showing an example of a 32x32 coefficient block.
FIG. 7 is a diagram illustrating sub-blocks preceding the last valid sub-block in the scan order in the 32 × 32 coefficient block of FIG.
FIG. 8 is a diagram showing sub-blocks trailing the last valid sub-block in the scan order in the 32x32 coefficient block of FIG. 6. FIG.
9 is a flowchart illustrating a method of encoding a coefficient block according to an embodiment of the present invention.
10 is a flowchart illustrating a method of encoding a coefficient block by an image decoding apparatus according to an embodiment of the present invention.
Fig. 11 is a diagram showing blocks generated by dividing a valid block hierarchically until a single effective coefficient is reached, for the 32x32 block illustrated in Fig. 6. Fig.
FIG. 12 is a tree formed by scanning blocks generated in the coefficient block of FIG. 11 in a diagonal fashion.
FIG. 13 is a diagram illustrating a tree of a simple structure composed of three layers.
14 is a diagram illustrating a 64x64 coefficient block.
15 is a flowchart showing a process of a large coefficient block in the image decoding apparatus.

Hereinafter, some embodiments of the present invention will be described in detail with reference to exemplary drawings. It should be noted that, in adding the identification codes to the constituent elements of the respective drawings, the same constituent elements are denoted by the same reference symbols as possible even if they are displayed on different drawings. In the following description of the present invention, a detailed description of known functions and configurations incorporated herein will be omitted when it may make the subject matter of the present invention rather unclear.

The techniques of this disclosure generally relate to efficiently representing the location of non-zero coefficients (i.e., significant coefficients) for a coefficient block that is an array of quantized coefficients that is the result of transformations and quantization.

1 is an exemplary block diagram of an image encoding apparatus in which the techniques of the present disclosure may be implemented.

The image encoding apparatus includes a block division unit 110, a prediction unit 120, a subtracter 130, a transform unit 140, a quantization unit 145, an encoding unit 150, an inverse quantization unit 160, 165, an adder 170, a filter unit 180, and a memory 190. Each component of the image encoding apparatus may be embodied as a hardware chip, or may be embodied as software, and the microprocessor may be implemented to execute the function of the software corresponding to each component.

The block dividing unit 110 divides each picture constituting an image into a plurality of CTUs (Coding Tree Units), and then recursively divides the CTUs using a tree structure. A leaf node in a tree structure becomes a coding unit (CU) which is a basic unit of coding. In the tree structure, a quad tree (QuadTree, QT) in which an upper node (or a parent node) is divided into four sub nodes (or child nodes) of the same size, or a QT structure and an upper node are divided into two lower nodes A QuadTree plus BinaryTree (QTBT) structure combining a binary tree (BT) structure can be used. That is, QTBT can be used to partition the CTU into multiple CUs.

In a QTBT (QuadTree plus BinaryTree) structure, a CTU can be first divided into a QT structure. The quadtree partitioning can be repeated until the size of the splitting block reaches the minimum block size (MinQTSize) of the leaf node allowed in QT. If the leaf node of the quadtree is not greater than the maximum block size (MaxBTSize) of the root node allowed in BT, it can be further partitioned into a BT structure. In BT, there may be a plurality of split types. For example, in some examples, there may be two types: a symmetric horizontal splitting and a vertically splitting type (i.e., symmetric vertical splitting) that divides a block of the node into two blocks of equal size horizontally. In addition, there may be additional types for dividing the block of the node into two asymmetric blocks. In the asymmetric form, the block of the node may be divided into two rectangular blocks having a size ratio of 1: 3, or the block of the corresponding node may be divided into diagonal directions.

The division information generated by the block division unit 110 by dividing the CTU by the QTBT structure is encoded by the encoding unit 150 and transmitted to the image decoding apparatus.

Hereinafter, a block corresponding to a CU to be encoded or decoded (i.e., a leaf node of QTBT) is referred to as a 'current block'.

The prediction unit 120 generates a prediction block by predicting the current block. The prediction unit 120 includes an intra prediction unit 122 and an inter prediction unit 124.

In general, the current blocks in a picture may each be predictively coded. Prediction of the current block may be performed using an intra prediction technique (using data from a picture containing the current block) or an inter prediction technique (using data from a picture previously coded for a picture containing a current block) And can be generally performed. Inter prediction includes both unidirectional prediction and bidirectional prediction.

For each inter-predicted block, a set of motion information may be available. A set of motion information may include motion information for forward and backward prediction directions. Here, the forward and backward prediction directions are two prediction directions of the bi-directional prediction mode, and the terms " forward " and " backward " do not necessarily have geometric meanings. Instead, they generally correspond to whether the reference picture is to be displayed before the current picture (" backward ") or later (" forward "). In some examples, the "forward" and "backward" prediction directions may correspond to reference picture list 0 (RefPicList0) and reference picture list 1 (RefPicList1) of the current picture.

For each prediction direction, the motion information includes a reference index and a motion vector. The reference index can be used to identify the reference picture in the current reference picture list (RefPicList0 or RefPicList1). The motion vector has horizontal (x) and vertical (y) components. In general, the horizontal component represents the horizontal displacement in the reference picture relative to the position of the current block in the current picture, which is needed to locate the x-coordinate of the reference block. The vertical component represents the vertical displacement in the reference picture relative to the position of the current block, which is needed to locate the y coordinate of the reference block.

The inter-prediction unit 124 searches for a block most similar to the current block in the reference picture coded and decoded earlier than the current picture, and generates a prediction block for the current block using the searched block. Then, a motion vector corresponding to the displacement between the current block in the current picture and the prediction block in the reference picture is generated. In general, motion estimation is performed on a luma component, and motion vectors calculated based on luma components are used for both luma components and chroma components. The motion information including the information on the reference picture used for predicting the current block and the information on the motion vector is encoded by the encoding unit 150 and transmitted to the video decoding apparatus.

Various methods can be used to minimize the amount of bits required to encode motion information.

For example, if the reference picture and the motion vector of the current block are the same as the reference picture and the motion vector of the neighboring block, the motion information of the current block can be transmitted to the decoding apparatus by encoding information capable of identifying the neighboring block. This method is referred to as a 'merge mode'.

In the merge mode, the inter-prediction unit 124 selects a predetermined number of merge candidate blocks (hereinafter, merge candidates) from neighboring blocks of the current block.

As shown in FIG. 2, the neighboring blocks for deriving the merge candidate include a left block L, an upper block A, an upper right block AR and a lower left block BL adjacent to the current block in the current picture ), And upper left block AL may be used in whole or in part. In addition, a block located in a reference picture (which may be the same as or different from a reference picture used for predicting the current block) rather than the current picture in which the current block is located may be used as the merge candidate. For example, a co-located block co-located with the current block in the reference picture or blocks adjacent to the same co-located block may be further used as merge candidates.

The inter-prediction unit 124 constructs a merged list including a predetermined number of merge candidates using these neighboring blocks. The merge candidate to be used as the motion information of the current block among the merge candidates included in the merge list is selected and merge index information for identifying the selected candidate is generated. The generated merge index information is encoded by the encoding unit 150 and transmitted to the decoding apparatus.

Another method of coding motion information is to encode a differential motion vector.

In this method, the inter-prediction unit 124 derives the predicted motion vector candidates for the motion vector of the current block using neighboring blocks of the current block. The neighboring blocks used for deriving the predicted motion vector candidates include a left block L, an upper block A, an upper right block AR, a lower left block (FIG. 5) adjacent to the current block in the current picture shown in FIG. BL), and upper left block AL may be used in whole or in part. In addition, a block located in a reference picture (which may be the same as or different from a reference picture used for predicting the current block), which is not the current picture in which the current block is located, is used as a neighboring block used to derive the predicted motion vector candidates It is possible. For example, a co-located block in the reference picture or a block adjacent to the co-located block in the reference picture may be used.

The inter-prediction unit 124 derives the predicted motion vector candidates using the motion vectors of the neighboring blocks, and determines a predicted motion vector for the motion vector of the current block using the predicted motion vector candidates. Then, a differential motion vector is calculated by subtracting the predicted motion vector from the motion vector of the current block.

The predictive motion vector can be obtained by applying a predefined function (e.g., median value, mean value calculation, etc.) to the predicted motion vector candidates. In this case, the image decoding apparatus also knows a predefined function. Also, since the neighboring blocks used for deriving the predicted motion vector candidates are already encoded and decoded, the motion vector of the neighboring blocks is already known by the video decoding apparatus. Therefore, the image encoding apparatus does not need to encode information for identifying a predicted motion vector candidate. Therefore, in this case, the information on the differential motion vector and the information on the reference picture used for predicting the current block are coded.

Meanwhile, the predicted motion vector may be determined by selecting one of the predicted motion vector candidates. In this case, information for identifying the predictive motion vector candidates selected is further coded together with the information about the differential motion vector and the reference pictures used for predicting the current block.

The intra prediction unit 122 predicts pixels in the current block using pixels (reference pixels) located in the vicinity of the current block in the current picture including the current block. There are a plurality of intra prediction modes according to the prediction direction, and the neighboring pixels to be used and the calculation formula are defined differently according to each prediction mode. In particular, the intra prediction unit 122 may determine an intra prediction mode to be used for coding the current block. In some examples, the intra-prediction unit 122 may encode the current block using a plurality of intra-prediction modes and may select an appropriate intra-prediction mode to use from the tested modes. For example, the intra-prediction unit 122 may calculate rate-distortion values using a rate-distortion analysis for various tested intra-prediction modes, and may employ rate- The intra prediction mode may be selected.

The intra prediction unit 122 selects one intra prediction mode from a plurality of intra prediction modes, and predicts the current block using neighboring pixels (reference pixels) determined by the selected intra prediction mode and an equation. The information on the selected intra prediction mode is encoded by the encoding unit 150 and transmitted to the video decoding apparatus.

The subtractor 130 subtracts the prediction block generated by the intra prediction unit 122 or the inter prediction unit 124 from the current block to generate a residual block.

The transform unit 140 transforms the residual signal in the residual block having pixel values in the spatial domain into transform coefficients in the frequency domain. The transform unit 140 may transform the residual signals in the residual block by using the size of the current block as a transform unit or divide the residual block into a plurality of smaller subblocks and transform residual signals into subblock- Conversion. There are various ways of dividing the residual block into smaller sub-blocks. For example, it may be divided into sub blocks of the same size that have been set, or a QT (quadtree) type partition using a residual block as a root node.

The quantization unit 145 quantizes the transform coefficients output from the transform unit 140, and outputs the quantized transform coefficients to the encoding unit 150.

The encoding unit 150 encodes the quantized transform coefficients using a coding scheme such as CABAC to generate a bitstream. This encoding is typically performed on the quantized transform coefficients using one of a plurality of available scan patterns.

One aspect of the teachings of the present disclosure generally relates to efficiently representing the location of non-zero coefficients (i.e., significant coefficients) for a coefficient block that is an array of quantized coefficients that is the result of transformations and quantization. As such, certain techniques of the present disclosure may be performed by the encoding unit 150. [ That is, for example, the encoding unit 150 may perform the techniques of the present disclosure described with reference to FIGS. 6 to 15 below. In other instances, one or more other units of the encoding apparatus may additionally be involved in performing the techniques of the present disclosure.

The encoding unit 150 encodes information such as a CTU size, a MinQTSize, a MaxBTSize, a MaxBTDepth, a MinBTSize, a QT division flag, a BT division flag, and a division type associated with the block division so that the video decoding apparatus So that the block can be divided.

The encoding unit 150 encodes information on a prediction type indicating whether the current block is coded by intraprediction or inter prediction, and encodes the intra prediction information or the inter prediction information according to the prediction type.

When the current block is intra-predicted, a syntax element for the intra-prediction mode is encoded as intra-prediction information. When the current block is inter-predicted, the encoding unit 150 encodes a syntax element for inter-prediction information. The syntax elements for the inter prediction information include the following.

(1) mode information indicating whether motion information of the current block is coded in the merge mode or in a mode in which the differential motion vector is coded

(2) Syntax element for motion information

When the motion information is coded by the merge mode, the encoding unit 150 converts the merge index information indicating which candidate of the merge candidates is selected as a candidate for extracting the motion information of the current block as a syntax element for the motion information .

On the other hand, when motion information is coded by a mode for coding a differential motion vector, information on a differential motion vector and information on a reference picture are encoded into syntax elements for motion information. If the predictive motion vector is determined in a manner of selecting one of the plurality of predictive motion vector candidates, the syntax element for the motion information further includes predictive motion vector identification information for identifying the selected candidate .

The inverse quantization unit 160 dequantizes the quantized transform coefficients output from the quantization unit 145 to generate transform coefficients. The inverse transform unit 165 transforms the transform coefficients output from the inverse quantization unit 160 from the frequency domain to the spatial domain and restores the residual block.

The adder 170 adds the reconstructed residual block and the prediction block generated by the predictor 120 to reconstruct the current block. The pixels in the reconstructed current block are used as reference pixels when intra prediction of the next-order block is performed.

The filter unit 180 deblock-filters the boundaries between the restored blocks and stores them in the memory 190 to remove blocking artifacts caused by coding / decoding on a block-by-block basis. When all the blocks in one picture are reconstructed, the reconstructed picture is used as a reference picture for inter prediction of a block in a picture to be coded later.

Hereinafter, a video decoding apparatus will be described.

3 is an exemplary block diagram of an image decoding apparatus capable of implementing the techniques of the present disclosure;

The image decoding apparatus includes a decoding unit 310, an inverse quantization unit 320, an inverse transform unit 330, a prediction unit 340, an adder 350, a filter unit 360, and a memory 370. 2, each component may be implemented as a hardware chip, or may be implemented as software, and a microprocessor may be implemented to execute functions of software corresponding to each component.

The decoding unit 310 decodes the bitstream received from the image encoding apparatus to extract information related to the block division to determine a current block to be decoded and outputs prediction information and residual signal information necessary for restoring the current block .

The decoding unit 310 extracts information on a CTU size from an SPS (Sequence Parameter Set) or a PPS (Picture Parameter Set) to determine a size of the CTU, and divides the picture into CTUs of a predetermined size. Then, the CTU is determined as the top layer of the tree structure, that is, the root node, and the CTU is divided using the tree structure by extracting the partition information for the CTU. For example, when the CTU is divided using the QTBT structure, the first flag (QT_split_flag) related to the division of the QT is first extracted and each node is divided into four nodes of the lower layer. For the node corresponding to the leaf node of the QT, the second flag (BT_split_flag) related to the BT division and the division type information are extracted and the corresponding leaf node is divided into the BT structure.

On the other hand, when the current block to be decoded is determined through division of the tree structure, the decoding unit 310 extracts information on a prediction type indicating whether the current block is intra-predicted or inter-predicted.

When the prediction type information indicates intra prediction, the decoding unit 310 extracts a syntax element for the intra prediction information (intra prediction mode) of the current block.

When the prediction type information indicates inter prediction, the decoding unit 310 extracts a syntax element for the inter prediction information. First, mode information indicating whether a motion information of a current block is coded by a mode among a plurality of coding modes is extracted. Here, the plurality of coding modes include a merge mode including a skip mode and a differential motion vector coding mode. When the mode information indicates the merge mode, the decoding unit 310 extracts merge index information indicating whether to derive the motion vector of the current block from any of the merge candidates as a syntax element for the motion information. On the other hand, when the mode information indicates the differential motion vector coding mode, the decoding unit 310 extracts information on the differential motion vector and information on the reference picture referred to by the motion vector of the current block as a syntax element for the motion vector do. On the other hand, when the video encoding apparatus uses any one of the plurality of predicted motion vector candidates as the predicted motion vector of the current block, the predictive motion vector identification information is included in the bitstream. Therefore, in this case, not only the information on the difference motion vector and the information on the reference picture but also the prediction motion vector identification information is extracted as a syntax element for the motion vector.

Meanwhile, the decoding unit 310 extracts information on the quantized transform coefficients of the current block as information on the residual signal. Another aspect of the teachings of the present disclosure generally relates to efficiently decoding the location of non-zero coefficients (i.e., significant coefficients) for a coefficient block that is an array of quantized coefficients that is the result of transform and quantization. Accordingly, certain techniques of the present disclosure may be performed by the decryption unit 310. [ That is, for example, the decryption unit 310 may perform the techniques of the present disclosure described below with reference to FIGS. In other instances, one or more other units of the encoding apparatus may additionally be involved in performing the techniques of the present disclosure.

The inverse quantization unit 320 dequantizes the quantized transform coefficients and the inverse transform unit 330 generates the residual block for the current block by inversely transforming the dequantized transform coefficients from the frequency domain to the spatial domain to recover the residual signals.

The prediction unit 340 includes an intra prediction unit 342 and an inter prediction unit 344. The intra prediction unit 342 is activated when the intra prediction is the prediction type of the current block, and the inter prediction unit 344 is activated when the intra prediction is the prediction type of the current block.

The intra prediction unit 342 determines an intra prediction mode of the current block among the plurality of intra prediction modes from the syntax element for the intra prediction mode extracted from the decoding unit 310, To predict the current block.

The inter-prediction unit 344 determines the motion information of the current block using the syntax element for the intra-prediction mode extracted from the decoding unit 310, and predicts the current block using the determined motion information.

First, the inter prediction unit 344 confirms the mode information in the inter prediction, which is extracted from the decoding unit 310. When the mode information indicates the merge mode, the inter-prediction unit 344 constructs a merge list including a predetermined number of merge candidates using the neighboring blocks of the current block. The way in which the inter prediction unit 344 constructs the merge list is the same as that of the inter prediction unit 124 of the image encoding apparatus. Then, one merge candidate is selected from merge candidates in the merge list using merge index information transmitted from the decoding unit 310. [ Then, the motion information of the selected merge candidate, that is, the motion vector of the merge candidate and the reference picture are set as the motion vector of the current block and the reference picture.

If the mode information indicates the differential motion vector coding mode, the inter-prediction unit 344 derives the predicted motion vector candidates using the motion vectors of the neighboring blocks of the current block, The predicted motion vector for the motion vector of the current frame is determined. The way in which the inter-prediction unit 344 derives the predicted motion vector candidates is the same as that of the inter-prediction unit 124 of the image encoding apparatus. If the video coding apparatus uses any one of the plurality of candidate prediction motion vector candidates as the prediction motion vector of the current block, the syntax element for the motion information includes the prediction motion vector identification information. Accordingly, in this case, the inter-prediction unit 344 can select a candidate indicated by the predicted motion vector identification information among the predicted motion vector candidates as a predicted motion vector. However, when the image encoding apparatus determines a predicted motion vector using a function predefined for a plurality of predicted motion vector candidates, the inter-prediction unit may determine a predicted motion vector by applying the same function as that of the image encoding apparatus. When the predicted motion vector of the current block is determined, the inter-prediction unit 344 adds the predicted motion vector and the differential motion vector received from the decoding unit 310 to determine a motion vector of the current block. The reference picture referred to by the motion vector of the current block is determined using the information on the reference picture transmitted from the decoding unit 310.

When the motion vector and the reference picture of the current block are determined in the merge mode or differential motion vector coding mode, the inter prediction unit 342 generates a prediction block of the current block using a block indicated by the motion vector in the reference picture do.

The adder 350 adds the residual block output from the inverse transform unit and the prediction block output from the inter prediction unit or the intra prediction unit to reconstruct the current block. The pixels in the reconstructed current block are utilized as reference pixels for intra prediction of a block to be decoded later.

The filter unit 360 deblock-filters the boundaries between the restored blocks and stores them in the memory 370 in order to remove blocking artifacts caused by decoding on a block-by-block basis. When all blocks in one picture are reconstructed, the reconstructed picture is used as a reference picture for inter prediction of a block in a picture to be decoded later.

As described above, the techniques of the present disclosure are concerned with efficiently encoding and decoding the location of non-zero coefficients (i.e., significant coefficients) for a coefficient block that is an array of quantized coefficients that is the result of transform and quantization. Some of the disclosed techniques can be applied directly to residual blocks that have not undergone transformation.

In the H.265 (HEVC) standard, there are six syntaxes for representing the position of a non-zero coefficient among the quantized coefficients after the quantization process.

1. last_sig_coeff_x_prefix

2. last_sig_coeff_y_prefix

3. last_sig_coeff_x_suffix

4. last_sig_coeff_y_suffix

5. coded_sub_block_flag

6. sig_coeff_flag

The previous four syntaxes relate to the location of the last significant coefficient in the scan order within the coefficient block, where the x and y components for that location are shown separately, Prefix, and suffix. The coded_sub_block_flag is a flag indicating whether each subblock includes one or more non-zero coefficients by dividing the coefficient block into a plurality of subblocks. Here, coded_sub_block_flag is represented by " 0 " if all the coefficients in the corresponding subblock are zero and " 1 " if there is more than one nonzero coefficient. sig _ coeff _flag is that each sub-block coefficients within a non-zero flag indicating whether the recognition agent. sig _ coeff _flag if the zero coefficient is represented by "0", the non-zero if the coefficient is represented by "1".

The coded_sub_block_flag syntax is signaled only for the sub-block preceding the last significant sub-block in which the last significant coefficient exists in the scan order for the coefficient block. If coded_sub_block_flag is "1" in, the sig _ coeff _flag syntax for each of all the coefficients in the sub-block is signaled.

The coding of the coefficients in the coefficient block is typically performed using one of a plurality of available scan schemes. 4 is a diagram illustrating exemplary scan schemes used for encoding quantized coefficients of a square coefficient block. These scan methods include an up-right diagonal method, a horizontal method, and a vertical method. When the current block to be coded is coded using the inter picture prediction scheme, the coefficients of the current block are scanned in an up-right diagonal manner, and when the current block is coded in the intra prediction mode, One of the branch types is selected and the coefficients of the corresponding block are scanned.

The exemplary scan scheme shows the same scan pattern for the subblocks in the coefficient block and the coefficients in each subblock. For example, in the case of the horizontal scanning method, the scanning order of the sub-blocks is also a horizontal method, and the scanning order of the coefficients in each sub-block is also a horizontal method. Figure 5 shows the scan order of subblocks and coefficients in more detail for the diagonal scan scheme. However, the order stored in the actual bit stream is stored in the reverse order of the scan order. In other words, the bit stream is stored in the order of the pixel at position 255 to the pixel at position 0 in FIG. In other words, once the last valid (non-zero) coefficient position is known, the coefficients can be decoded in the opposite direction of the scan order towards the first coefficient in the scan pattern, as will be described further below.

6 is a diagram showing an example of a 32x32 coefficient block. Here, the scanning method of the 32x32 coefficient block uses a diagonal method, the pixels marked with gray indicate non-zero coefficients, and the pixels marked with black indicate the last non-zero coefficient in the scan order. All white coefficients have zero values. Non-Applying the coded_sub_block_flag and sig _ coeff _flag syntax of the above-described three pieces of information indicating the location of the zero coefficients in the block illustrated in Figure 6, the number of bits required to represent these two syntax is as follows. coded_sub_block_flag needs 30 bits for 30 subblocks (total of subblocks shown in FIG. 6) among the total 64 subblocks considering the position of the last significant coefficient. Coded_sub_block_flag is not signaled for the left- ordered subblock including the DC component. Coefficient block of Figure 6 is one or more non-a 7 a which includes a coefficient zero sub-block (i.e., the valid sub-block), and therefore to display the sig _ coeff _flag for seven significant sub-block, substantially 112 bits (= 7 x 16; less than 16 bits may be needed for the last valid sub-block). As a result, the number of bits required to represent the two syntaxes for the coefficient block of FIG. 6 is approximately 142 bits (= 30 + 112).

As described above, in order to reduce the overhead of the flag bits, the last valid coefficient position is used. The problem is that since the size of the transform block currently being discussed can be as large as 64 × 64, 128 × 128, it is necessary to consider that a significant bit number is required for encoding the syntax for expressing the position of the last significant coefficient as the x- will be.

In order to provide similar functions while requiring fewer bits, the present invention proposes a method of signaling information indicating the location of the last significant sub-block within the coefficient block to be coded. Here, the last valid subblock is the last subblock in the subblock scan order having at least one non-zero coefficient. There are four schemes for efficiently expressing the position of the last valid subblock proposed by the present invention.

(1) X axis index and Y axis index in the coefficient block of the last valid subblock

(2) the number of sub-blocks preceding the last valid sub-block in the scan order

(3) the number of subblocks following the last valid subblock in the scan order

(4) determine a favorable scheme of (2) scheme or (3) scheme depending on the position of the last valid sub-block in the coefficient block, and transmit flag information indicating the determined scheme and the number information according to the determined scheme

For example, with respect to the 32x32 coefficient block illustrated in FIG. 6, the positions of the last valid subblock are expressed by the four proposed methods, respectively.

( 3, 4) - Referring to FIG. 7 or 8, the last valid sub-block in the scan order in the coefficient block is divided into X-axis index 3 and Y-axis index 4 .

(2) 31 (when diagonal scanning method is used) - Referring to FIG. 7, the number of subblocks preceding the last valid subblock in the scan order is 31.

(3) 32 (when diagonal scanning method is used) - Referring to FIG. 8, the number of subblocks following the last valid subblock in the scan order is 32.

(4) 0 31 or 1 32 , where underlined " 0 ", " 1 " is a flag value indicating which of the methods (2) and (3) is used.

The method of coding the coefficient block by the image coding apparatus and the overall process of decoding the coefficient block by the image decoding apparatus will be described with reference to FIGS. 9 and 10, respectively, based on the signaling method of the position of the last valid subblock .

9 is a flowchart illustrating a method of encoding a coefficient block according to an embodiment of the present invention.

First, the image encoding apparatus determines the last significant sub-block in the current coefficient block to be encoded (S910). The last valid subblock is the last subblock in the subblock scan order having at least one non-zero coefficient.

The image encoding apparatus encodes information indicating the position of the last valid sub-block determined (S920). In some embodiments, the information indicating the location of the last valid sub-block may be (1) an X-axis index and a Y-axis index value within the coefficient block of the last valid sub-block. In some other embodiments, the information indicating the location of the last valid subblock may be information on the number of subblocks preceding the last valid subblock in the scan order. In yet another embodiment, the information indicating the position of the last valid subblock may be information on the number of subblocks following the last valid subblock in the scan order. In yet another embodiment, the information indicating the position of the last valid subblock includes information on a selected one of the number of subblocks preceding the last valid subblock or the number of succeeding subblocks in the scan order, And may include a flag value indicating whether it is selected.

The image encoding apparatus determines whether each sub-block excluding the left-most sub-block of the coefficient block among the sub-blocks preceding the last valid sub-block in the sub-block scan order is an effective sub-block having at least one non-zero coefficient (Step S930). In step S930, the syntax element for each sub-block is encoded. At this time, the leftmost subblock and the last valid subblock of the coefficient block may be set as valid subblocks having at least one non-zero coefficient.

The image encoding apparatus includes coefficients of the leftmost subblock of the coefficient block, coefficients of the last valid subblock, coefficients of the subblocks indicating that the syntax element is an effective subblock having at least one non-zero coefficient, The coefficients of sub-blocks set as valid sub-blocks are encoded (S940). The process of encoding these coefficients may include encoding a flag indicating whether each coefficient in the corresponding sub-block is non-zero or zero. The image coding apparatus skips encoding of the coefficients of the sub-blocks indicating that the syntax element is not an effective sub-block having at least one non-zero coefficient. Also, the image encoding apparatus skips encoding of the coefficients of the sub-blocks trailing the last valid sub-block in the sub-block scan order.

10 is a flowchart illustrating a method of encoding a coefficient block by an image decoding apparatus according to an embodiment of the present invention.

The video decoding apparatus decodes information indicating the position of the last valid subblock in the current coefficient block to be decoded from the coded bitstream (S1010).

The video decoding apparatus may further comprise a decoding unit for decoding, from the encoded bit stream, at least one non-zero coefficient of each sub-block excluding the left-most sub-block of the coefficient block among the sub-blocks preceding the last valid sub- Blocks for each sub-block indicating whether the sub-block is an effective sub-block are decoded (S1020). At this time, the leftmost subblock and the last valid subblock of the coefficient block may be set as valid subblocks having at least one non-zero coefficient.

The video decoding apparatus is characterized in that, from the encoded bit stream, coefficients of the left-ordered sub-block of the coefficient block, coefficients of the last valid sub-block, and the syntax element are valid sub-blocks having at least one non- And the coefficients of the sub-blocks set as the valid sub-blocks are decoded (S1030). The process of decoding these coefficients may include decoding a flag indicating whether each coefficient in the corresponding sub-block is non-zero or zero. The video decoding apparatus skips the decoding of the coefficients of the sub-blocks indicating that the syntax element is not an effective sub-block having at least one non-zero coefficient. In addition, the image decoding apparatus skips the decoding of the coefficients of the sub-blocks following the last valid sub-block in the sub-block scan order.

In the method described above, the location of the non-zero coefficients after the transform is mostly collected in the low frequency component and is not located in the high frequency, and the position of the last valid coefficient or the last valid subblock is preferentially signaled. In the following, we propose a technique that can efficiently express the position of nonzero coefficients wherever they are located. This technique may be more useful when the transform of a block to be coded is skipped and the residual value itself is sent instead of high frequency and low frequency components (or coefficients).

Hierarchical partitioning of valid blocks

Another technique proposed by the present invention is to use subblocks (effective blocks) including at least one non-zero coefficient for n current subblocks (nb 2 or more natural number) equal size ( Significant squares), while signaling whether it is a valid block for each of the subblocks generated using the significant_flag . The validity flag is expressed, for example, as " 1 " if there is more than one non-zero coefficient (i.e., effective coefficient) in one subblock, and for example, " 0 " If the valid flag value is " 0 ", the corresponding sub-block is no longer divided, and the valid flag value for the lower layer blocks is no longer needed. On the other hand, if the valid flag value is " 1 ", the corresponding block is divided into n subblocks (natural number of 2 or more), and signals validity flags for the subblocks in the lower hierarchy.

If the valid flag value of a given coefficient block is " 0 ", then the block is no longer partitioned, and therefore the valid flag value for blocks in that lower layer is no longer needed. On the other hand, if the effective flag value of a given coefficient block is "1" and the coefficient block is a square-shaped coefficient block, it is divided into four square blocks of the lower hierarchy, and each of the four blocks is expressed as a non- . The effective block is repeatedly divided until a single effective coefficient is reached, so that the size of the generated block includes the size of the coefficient block up to a size of 1 x 1 (i.e., a single coefficient). For example, if the coefficient block is 8x8, the block for which the validity flag is determined includes all the sizes of 8x8, 4x4, 2x2, and 1x1.

 If the effective flag value of a given coefficient block is " 1 " and the coefficient block has a rectangular shape, the number of square blocks in the lower layer may be changed according to the size of the coefficient block. For example, if the coefficient block is 16 × 8, the square block divided into the lower hierarchy is 2 × 8, and if the coefficient block is 4 × 16, the square block divided into the lower hierarchy has 4 × 4 do. That is, when a given coefficient block is a rectangle of M × kM or kM × M size, a given coefficient block can be first divided into n = k squares (M × M). At this time, the order of k squares proceeds in reverse order from left to right or top to bottom according to the shape of the coefficient block. For example, if the coefficient block is a horizontally long rectangle, the order of the lower k square blocks proceeds in reverse order from left to right. If the coefficient block is a vertically long rectangle, the order of the lower k square blocks is It proceeds. The process after dividing the rectangular coefficient block into k squares in the lower layer proceeds in the same way as (4) square subblocks in the lower layer of the square block.

11 is a diagram showing blocks generated by applying the proposed technique to the 32x32 block illustrated in FIG. According to the proposed method, valid flags for the blocks indicated by bold solid lines in FIG. 11 are signaled. The hierarchical structure of each block and the valid flag value of each block are expressed in a tree form as shown in FIG. A node indicated by black means that the valid flag value of the corresponding block is " 1 ", and a node indicated by white indicates that the valid flag value of the corresponding block is " 0 " . Also, in the case of a node indicated by white, the corresponding node becomes a leaf node, and no child node exists. That is, when the valid flag of a given block is " 0 ", the corresponding block becomes a leaf node, and there is no lower layer block. Here, the corresponding block may be referred to as a parent node, and the blocks of the lower layer may be referred to as a child node. If the valid flag of a given block is " 1 ", the corresponding block always has four child nodes.

The tree of FIG. 12 is constructed by scanning the blocks generated in the coefficient block of FIG. 11 in a diagonal manner. In addition, a tree is constructed in consideration of the order of storing the bitstream (i.e., the reverse order of the scan order; upper left, lower left, upper right, and lower right). The diagonal method used in FIG. 12 is an example, and the above-mentioned conventional three types of scan methods can be used according to existing conditions. If a scanning method other than diagonal is used, the tree for the coefficient block of FIG. 11 is configured differently from that of FIG. It is also possible to construct a tree in a z-scan (reverse order) manner and use the bit stream storage order (i.e., reverse order of z-scan order; top left, top right, bottom left, bottom right).

The valid flag for a block of the same size as the coefficient block can be regarded as a coded block flag (cbf). For example, in the case of the coefficient block of FIG. 11, the valid flag for the 32x32 block performs the cbf function. Since the valid flag value of the 32x32 block for the coefficient block illustrated in FIG. 11 is " 1 ", the valid flags for the lower four 16x16 blocks are obtained. At this time, the four 16x16 blocks are scanned in the reverse order of the upper left, lower left, upper right, and lower right blocks according to the diagonal scanning method. The four valid flags become " 0011 ". Only the valid flags of the lower 8x8 blocks for the lower left and upper left 16x16 blocks corresponding to " 1 " The valid flag for the lower 4 8 × 8 blocks of the lower left 16 × 16 block is "0100", and the valid flags for the lower 4 8 × 8 blocks of the upper left 16 × 16 block are "1111". It is only necessary to express the valid flag value of the lower 4 × 4 blocks for the block having the valid flag of "1" among the 8 × 8 blocks. The above operation is repeated until the size of the block is 1 x 1.

The validity flags for the (32 x 32 to 1 x 1) blocks generated as a result of applying the quadrature to the coefficient block of Fig. 11 are as shown in the following table.

Block size Valid flag 32 x 32 blocks One 16 × 16 blocks 0011 8 x 8 blocks 0100
1111 4 × 4 blocks 0100
0110 0001 0001 1001 2 x 2 blocks 0010
0001 0001 0001 1000 0001 0001 1 x 1 block 0100
1000 1000 0010 0001 1000 0001

Hereinafter, an example of searching the tree of FIG. 12 for the coefficient block of FIG. 11 is shown. A depth-first search or a breadth-first search may be used to search the tree of FIG. Here, the depth-first method and the width-first method refer to the order of searching the nodes of the tree. That is, according to the present invention, bits allocated to the node are read while searching for nodes of the tree according to a predetermined method among the depth-first method and the width-first method, thereby forming a bit-stream. For example, referring to the tree structure illustrated in FIG. 13A, (1) if a bit stream is formed in a depth-first manner, bits are read in order of ABCDFGHIE; (2) , The bits are read in order of ABCDEFGHI.

Therefore, a bitstream is constructed by searching the tree of FIG. 12 proposed by the present invention by the width-first search, as follows.

One. One

2. 0011

3. 0100 1111

4. 0100 0110 0001 0001 1001

5. 0010 0001 0001 0001 1000 0001 0001

6. 0100 1000 1000 0010 0001 1000 0001

Conversely, the tree of FIG. 12 is searched by depth-first search to construct a bit stream as follows.

One. One

2. 00

3. 1 01 01 001 0100 0 00 00

4. 1 1 01 0001 1000 1 0001 1000 0

5.   1 0001 0001 0010

6.   1 0001 1 0001 000

7.   1 1 0001 1000 001 0001 0001

Exception rule

On the other hand, if a width-first search method is applied to the tree illustrated in FIG. 13B, a bit stream can be composed of "1", "0001", and "1000". At this time, the valid flag value of the uppermost parent node is " 1 ", which means that at least one of the four child nodes of the block must have a valid flag value of " 1 ". However, if all the valid flag values of the three child nodes of the four child nodes in the scanning order are all " 0 ", the flag value of the last remaining child node must be " 1 ". As a result, the flag value of the fourth child node can be deduced to be " 1 " in both the encoding apparatus and the decoding apparatus, so that it is not necessary to be signaled. When this exception rule is applied to the tree illustrated in (b) of FIG. 13, the bit stream is composed of " 1 ", " 000 ", and " 1000 ".

By applying such an exception rule, a bit stream is constructed by the breadth first search method for the tree illustrated in FIG. 12 as follows. Here, a total of nine " 1 " values that are not signaled are indicated by " - ".

One. One

2. 0011

3. 0100 1111

4. 0100 0110 000 - 000 - 1001

5. 0010 000 - 000 - 000 - 1000 000 - 000 -

6. 0100 1000 1000 0010 000 - 1000 000 -

The exception rule is applied to the tree of FIG. 12 to construct a bitstream by depth-first search. Here, a total of nine " 1 " values that are not signaled are indicated by " - ".

One. One

2. 00

3. 1 01 01 001 0100 0 00 00

4. 1 1 01 000 - 1000 1 000 - 1000 0

5. 1 000 - 000 - 0010

6. 1 000 - 1 000 - 000

7. 1 1 000 - 1000 001 000 - 000 -

The high- Zero

Another technique proposed in this disclosure is to set the coefficient values of a certain portion corresponding to the high frequency component in the coefficient block to " 0 " regardless of the actual value, when the coefficient block (or the quantized transform block) .

For example, when the size of the coefficient block is 64x64, all the coefficient values of the remaining blocks except for the upper left 32x32 block (upper left, lower left, and right lower 32x32 blocks) are forcibly zeroed. FIG. 14 illustrates a 64x64 coefficient block. If the predetermined threshold size is 32x32, non-zero coefficients (indicated by black) located in an area excluding the left upper 32x32 block are all zeroed. Therefore, all of the valid flags for the 32 × 32 left-justified block, the right-justified 32 × 32 block, and the 32 × 32 right-right block are set to "0".

If a given coefficient block is a rectangle of M × kM or kM × M size, a given coefficient block can be divided into k × M × M blocks. If the predetermined threshold size is h × h (h ≤ M) All the coefficient values of the blocks other than the upper hxh block or the leftmost hxh block can be forcibly zeroed. Conversely, if the predetermined threshold size is hxh (M < h < kM), all the coefficient values of the remaining blocks excluding the uppermost Mxh block or the leftmost hxM block can be forcibly zeroed. For example, if the coefficient block is 16 × 4 and the predetermined threshold size is 8 × 8, the non-zero coefficients of the right two 4 × 4 blocks out of the four 4 × 4 blocks are all zeroed.

Valid flag values for the 64x64 coefficient block illustrated in FIG. 14 are expressed as follows through the diagonal scanning method (reverse order) and width-first search. Of the four 32x32 blocks, the preceding three blocks in the scan order are set to " 000 " despite the fact that the non-zero coefficients actually exist, and only the last fourth block becomes the flag value " 1 ". In addition, applying the exception rule results in information that can be inferred without signaling all 4 bits (" 0001 ") for 32x32 blocks. Further, the coefficients of the zeroed blocks are not signaled.

Block size Valid flag 64 × 64 blocks One 32 x 32 blocks 0001 16 × 16 blocks 1111 8 x 8 blocks 0001 0001  1001 1111 4 × 4 blocks 0100 0001 0100 0100
0001 1000 0001 1001 2 x 2 blocks 0010
1000 0010 0100 1000 0001 1 x 1 block 0100 0010 0100 0001
0010 0100 0010 0001 0001

The zeroing of the high-frequency block can be applied again to the four blocks corresponding to the grandchild node if the block size of the grandchild node of the coefficient block is still larger than the threshold value.

This exception rule is particularly useful for increasing the coding efficiency of a large coefficient block in that the larger the coefficient block size, the non-zero coefficients are mostly concentrated on the low frequency components after conversion and hardly exist in the high frequency components. Therefore, the larger the transform size, the more the compression performance can be increased by the proposed zeroing technique. This zeroing technique may be combined with the techniques described with reference to FIGS.

Using the techniques and rules described above, for the 32x32 coefficient block illustrated in FIG. 11, the valid flag values signaled to represent the distribution of non-zero coefficients are: The rules applied are as follows.

- Zeroing of high frequency blocks: Set th value to 64

- Scan method: use diagonal method

- Tree navigation method: use width priority

- Apply exception rule

One. One

2. 0011

3. 0100 1111

4. 0100 0110 000 - 000 - 1001

5. 0010 000 - 000 - 000 - 1000 000 - 000 -

6. 0100 1000 1000 0010 000 - 1000 000 -

Thus, a total of 80 bits is required to represent the position of the non-zero coefficients of the 32x32 coefficient block illustrated in FIG. As described above, if the coefficient block of FIG. 6, which is the same as the coefficient block of FIG. 11, is expressed by the method adopted in the existing HEVC, even if four syntaxes for expressing the position of the last significant coefficient are excluded, the remaining two syntaxes coded_sub_block_flag and _ sig coeff number of bits required for _flag) is approximately 140 bits. If the information for expressing the position of the last significant coefficient is added to 140 bits, more bits are required.

If the threshold value (th) is 16, and the other conditions are identical as above, the 32 × 32 block valid flag for the coefficients illustrated in Figure 4 are as follows.

One. One

2. ----

3. 1111

4. 0110 000 - 000 - 1001

5. 000 - 000 - 000 - 1000 000 - 000 -

6. 1000 1000 0010 000 - 1000 000 -

Here, since the effective flag value for the entire coefficient block (i.e., the 32x32 block) is " 1 &quot;, and the 16x16 block as the lower node is equal to the threshold value ( th = 16) The flag value for the 16th block is set to " 000 &quot;, and the flag value for the fourth block is set to " 1 &quot;. Therefore, the validity flag " 0001 " for four 16x16 blocks need not be signaled.

15 is a flowchart showing a process of a large coefficient block in the image decoding apparatus. It should be noted that although the use of the width-first search method is assumed in FIG. 15, depth-first search may be used. For the sake of simplicity, FIG. 15 does not include the above-mentioned exception rule, and it is assumed that the coefficient block has a square shape.

The image decoding apparatus sets the size (i.e., TU size) of the current coefficient block to be decoded to N. [ The video decoding apparatus parses the valid flag for the current coefficient block and, if the valid flag is not " 0 &quot;, divides the coefficient block into four blocks. If the size (N / 2) of the generated block is equal to or larger than the threshold value th , the image decoding apparatus sets valid flags for the generated four blocks to "0001" without parsing the valid flag. Conversely, when the size (N / 2) of the generated block is less than the threshold value th , the image decoding apparatus parses valid flags for the generated four blocks.

For each of the four generated blocks, if the valid flag value of the corresponding block is not " 0 ", the corresponding block is repeatedly repeated with small blocks of four equal sizes until a single effective coefficient (N = 1) is reached and parses valid flags for the four blocks generated each time the block is divided.

As described above, the technique described with reference to FIGS. 11 to 15 is more useful when a transform of a block to be coded is skipped and a residual value itself is sent instead of high frequency and low frequency components (or coefficients) . Focusing on this point, the image encoding apparatus applies the technique described with reference to Figs. 11 to 15 when the conversion of the block to be coded is skipped, and Figs. 6 to 10 are applied to the quantized transform coefficient block to which the transform is performed And may be configured to apply the techniques described above.

The foregoing description is merely illustrative of the technical idea of the present embodiment, and various modifications and changes may be made to those skilled in the art without departing from the essential characteristics of the embodiments. Therefore, the present embodiments are to be construed as illustrative rather than restrictive, and the scope of the technical idea of the present embodiment is not limited by these embodiments. The scope of protection of the present embodiment should be construed according to the following claims, and all technical ideas within the scope of equivalents thereof should be construed as being included in the scope of the present invention.

Claims (20)

CLAIMS 1. An image encoding method comprising encoding a distribution of non-zero coefficients in a coefficient block,
Encoding a significant flag indicating whether the current coefficient block is a valid block having at least one non-zero coefficient;
Wherein, when the valid flag indicates that the current coefficient block is a valid block, for the current coefficient block, the sub-block determined as the valid block is divided into n (n is 2 or more Determining whether each of the n subblocks is a valid block, while dividing the subblocks recursively into subblocks of equal size,
Skipping the coding of the valid flag associated with the n subblocks if the size of the n subblocks is greater than or equal to a predetermined threshold value; And
If the size of the n subblocks is smaller than a preset threshold value, encoding the validity flag associated with the n subblocks
And an image encoding method. The method according to claim 1,
Wherein the encoding of the validity flag associated with the n subblocks comprises:
And if n-1 sub-blocks preceding in the sub-block scan order among the n sub-blocks are not all valid blocks, valid flags associated with the preceding n-1 sub-blocks are encoded, And skipping the encoding of the valid flag associated with the last sub-block in the sub-block scan order. The method according to claim 1,
The image encoding method includes:
And skipping the encoding of the validity flag associated with the n subblocks,
Encoding the coefficients of the leftmost subblock among the n subblocks; And
And skipping coding of coefficients of sub-blocks of right-wing, left-justified, and right-wing among the n sub-blocks. Determining a distribution of non-zero coefficients in a coefficient block, the method comprising:
Parsing, from the bitstream, a significant flag indicating whether the current coefficient block is a valid block having at least one non-zero coefficient;
Wherein, when the valid flag indicates that the current coefficient block is a valid block, for the current coefficient block, the sub-block determined as the valid block is divided into n (n is 2 or more Determining whether each of the n subblocks is a valid block, while recursively dividing the subblocks into equal sized subblocks,
Wherein determining whether each of the n subblocks is a valid block comprises:
If the size of the n subblocks is greater than or equal to a predetermined threshold, skipping the parsing of the validity flag associated with the n subblocks; And
If the size of the n subblocks is smaller than a preset threshold value, parsing valid flags associated with the n subblocks
And decodes the decoded image. 5. The method of claim 4,
If the current coefficient block is square,
Wherein the n equal size sub-blocks are four squares. 5. The method of claim 4,
If the current coefficient block is a rectangle of M × kM or kM × M size,
The current coefficient block is first divided into k square blocks, and then the sub-blocks determined as valid blocks are divided into four equal-sized square blocks of sub-blocks And the image is divided into a plurality of blocks. 5. The method of claim 4,
Wherein skipping parsing of the validity flag associated with the n subblocks comprises:
Setting an effective flag associated with left sub-blocks of the n sub-blocks to a first value and setting an effective flag associated with sub-blocks of righting, left-justifying, and righting to a second value To be decoded. 8. The method of claim 7,
Further comprising skipping decoding of coefficients of subblocks having validity flags set to the second value. &Lt; Desc / Clms Page number 22 &gt; 5. The method of claim 4,
Wherein parsing valid flags associated with the n subblocks comprises:
Blocks among the n sub-blocks are not all valid blocks, decodes valid flags related to the preceding n-1 sub-blocks, and selects one of the n sub-blocks And skipping the decoding of the valid flag associated with the last sub-block in the block scan order. An image decoding apparatus comprising determining a distribution of non-zero coefficients in a coefficient block,
Memory; And
Comprising one or more processors,
The one or more processors,
Parsing, from the bitstream, a significant flag indicating whether the current coefficient block is a valid block having at least one non-zero coefficient;
If the valid flag indicates that the current coefficient block is a valid block, for the current coefficient block, a sub-block determined as a valid block is divided into n equal-sized sub- Determining whether each of the n subblocks is a valid block, while recursively dividing the blocks into blocks,
Wherein determining whether each of the n subblocks is a valid block comprises:
If the size of the n subblocks is greater than or equal to a predetermined threshold, skipping the parsing of the validity flag associated with the n subblocks; And
If the size of the n subblocks is smaller than a preset threshold value, parsing valid flags associated with the n subblocks
And decodes the decoded image. 11. The method of claim 10,
Wherein skipping parsing of the validity flag associated with the n subblocks comprises:
Setting an effective flag associated with left sub-blocks of the n sub-blocks to a first value and setting an effective flag associated with sub-blocks of righting, left-justifying, and righting to a second value To the image decoding apparatus. 11. The method of claim 10,
Wherein parsing valid flags associated with the n subblocks comprises:
Blocks, if the n-1 sub-blocks preceding the sub-block scan order among the n sub-blocks are not all valid blocks, decodes valid flags related to the preceding n-1 sub-blocks, And skipping the decoding of the valid flag associated with the last sub-block in the sub-block scan order. A method of encoding a coefficient block having an array of coefficients divisible by coefficients of a plurality of subblocks,
Determining a last significant sub-block, wherein the last valid sub-block is the last sub-block in a sub-block scan order having at least one non-zero coefficient;
Coding information indicating a location of the determined last valid subblock;
Wherein each sub-block except the left-most sub-block of the coefficient block among the sub-blocks preceding the last valid sub-block in the sub-block scan order is an effective sub-block having at least one non- Encoding a syntax element for a sub-block; And
Encoding the coefficients of the left-ordered subblock, the coefficients of the last valid subblock, and coefficients of subblocks indicating that the syntax element is an effective subblock having at least one non-zero coefficient
And a coefficient block. 14. The method of claim 13,
The position of the last valid sub-
An index of the X axis and a Y axis index in the coefficient block of the last valid subblock,
The number of sub-blocks trailing the last valid sub-block in the sub-block scan order,
Wherein the number of subblocks preceding the last valid subblock is represented by the number of subblocks preceding the last valid subblock in the subblock scan order. 14. The method of claim 13,
The information indicating the position of the last valid sub-
A second information indicating the number of subblocks preceding the last valid subblock in the subblock scan order or the number of subblocks preceding the last valid subblock in the subblock scan order; And
Third information indicating which of the first information or the second information is signaled;
And a coefficient block coding step of coding the coefficient block. A method of decoding a coefficient block having an array of coefficients divisible by coefficients of a plurality of subblocks,
Block, the last valid sub-block being the last sub-block in the sub-block scan order having at least one non-zero coefficient;
Wherein each sub-block except the left-most sub-block of the coefficient block among the sub-blocks preceding the last valid sub-block in the sub-block scan order is an effective sub-block having at least one non- Decoding the syntax element for the sub-block; And
Decoding coefficients of the left-ordered sub-block, coefficients of the last valid sub-block, and coefficients of sub-blocks indicating that the syntax element is an effective sub-block having at least one non-zero coefficient
And decodes the coefficient block. 17. The method of claim 16,
The position of the last valid sub-
Wherein the coefficient of the last valid sub-block is represented by X-axis coordinates and Y-axis coordinates in the coefficient block of the last valid sub-block. 17. The method of claim 16,
The position of the last valid sub-
A number of subblocks preceding the last valid subblock in the subblock scan order or a number of subblocks preceding the last valid subblock in the subblock scan order, / RTI &gt; 17. The method of claim 16,
The information indicating the position of the last valid sub-
A second information indicating the number of subblocks preceding the last valid subblock in the subblock scan order or the number of subblocks preceding the last valid subblock in the subblock scan order; And
Third information indicating which of the first information or the second information is signaled;
And a coefficient block coding step of coding the coefficient block. An apparatus for decoding a coefficient block having an array of coefficients divisible by coefficients of a plurality of subblocks,
Memory; And
Comprising one or more processors,
The one or more processors,
Block, the last valid sub-block being the last sub-block in the sub-block scan order having at least one non-zero coefficient;
Wherein each sub-block except the left-most sub-block of the coefficient block among the sub-blocks preceding the last valid sub-block in the sub-block scan order is an effective sub-block having at least one non- Decoding the syntax element for the sub-block; And
Decoding coefficients of the left-ordered sub-block, coefficients of the last valid sub-block, and coefficients of sub-blocks indicating that the syntax element is an effective sub-block having at least one non-zero coefficient
To perform the decoding process.

Images