Decoding and encoding methods

Abstract

To provide an intra-frame prediction mode mapping method for improving decoding efficiency.SOLUTION: A decoding method decodes a syntax element of information on a candidate intra prediction mode when one of the candidate intra prediction modes for the current block is the same as the intra prediction mode, and decodes syntax elements of the remaining intra prediction modes when they are not the same. A prediction block is generated on the basis of the intra prediction mode of the current block, and a reconstructed block is generated on the basis of the prediction block. The remaining intra prediction mode information has a shorter codeword length of an index than the candidate intra prediction mode. The syntax element of the remaining intra prediction mode information is a value coded using a fixed 5 bits.SELECTED DRAWING: None

Description

[Technical Field]

[0001] The present invention relates to an intra-frame prediction method and an image decoding device, and more particularly to an encoding / decoding method and an image decoding device. The present invention relates to a method and apparatus for [Background technology]

[0002] In recent years, HD (High Definition) images and UHD (Ultra High Definition) images have become increasingly There are various demands for high-resolution, high-quality images such as (High Definition) images. The higher the resolution and quality of image data, the more difficult it is to use existing image data. Since the amount of data increases relatively compared to existing wired or wireless broadband lines, When transmitting image data using a computer and storing it using an existing storage medium, the transmission costs and Storage costs increase as image data becomes higher in resolution and quality. To solve such problems, highly efficient image compression techniques can be utilized.

[0003] As an image compression technique, it is possible to compress the current image (picture) from the previous or subsequent images. Inter-frame prediction technology predicts pixel values ​​contained in an image, and uses pixel information from the current image to Intra-screen prediction technology predicts pixel values ​​contained in an image, and assigns short codes to values ​​that appear frequently. Various techniques are used, such as entropy coding, which assigns longer codes to values ​​that appear less frequently. There are techniques for effectively compressing image data using such image compression techniques, and for transmitting or can be stored. Summary of the Invention [Problem to be solved by the invention]

[0004] A first object of the present invention is to provide an intra-frame prediction mode mapping method for improving image coding efficiency. The purpose is to provide a method.

[0005] A second object of the present invention is to provide an intra-frame prediction mode mapping method for improving image coding efficiency. The object of the present invention is to provide an apparatus for carrying out the method. [Means for solving the problem]

[0006] In order to achieve the first object of the present invention, an intra prediction method according to one aspect of the present invention includes: One candidate intra prediction mode among multiple candidate intra prediction modes for the current block and the intra prediction mode of the current block are the same, and One candidate intra prediction mode among multiple candidate intra prediction modes for the current block and the intra prediction mode of the current block are not the same, Decoding a syntax element containing information about the intra prediction mode of the current block to derive the and the intra prediction mode of the current block is derived by using the intra prediction mode of the current block. Decoding a syntax element containing information about the mode is equivalent to decoding a syntax element containing intra prediction mode information. The table is based on the intra-frame prediction mode and the index of the intra-frame prediction mode. In the table, the intra prediction mode is If the mode is anar mode, the index is 0. If the mode is DC mode, the index is In the case of directional intra prediction mode, the index is determined by the direction of the intra prediction mode. The current block can be mapped to boxes 2 to 34. The syntax elements containing the information are values ​​coded using a fixed 5-bit The information of the block is the remaining intra prediction modes excluding the multiple candidate intra prediction modes for the current block. The prediction mode information may be information indicating one of the intra-frame prediction modes. Multiple candidate intra prediction modes for a block are derived based on the neighboring blocks of the current block. Three different prediction modes derived based on the selected intra-frame prediction mode and the additional intra-frame prediction mode. If the flag information is 1, the flag information and one of the candidate intra prediction modes of the current block based on the Indicates that the intra prediction mode is the same as that of the current block, and the flag information is 0. Based on the flag information, a plurality of candidate intra-frame prediction modes of the current block and the image of the current block are selected. It can be indicated that the intra prediction mode of the current block is not the same as the intra prediction mode. Decoding a syntax element containing information about the intra prediction mode of the current block to derive the In the table, three intra-frame prediction modes are selected as candidate intra-frame prediction modes for the current block. The remaining 32 intra-picture prediction mode information is derived by excluding the 32 intra-picture prediction modes. Syntax element containing information about the intra prediction mode of the current block and the intra prediction mode of the current block. The remaining 32 intra-frame prediction mode information is mapped to syntax elements. The intra prediction mode of the current block is set to the selected intra prediction mode. It can include.

[0007] In order to achieve the second object of the present invention, an image decoding apparatus according to one aspect of the present invention comprises: One candidate intra prediction mode among multiple candidate intra prediction modes for the current block and flag information indicating whether the intra-frame prediction mode of the current block is the same as that of the current block. One of multiple candidate intra-frame prediction modes for the block and the current block If the intra prediction mode of the current block is not the same as the intra prediction mode of the previous block, The entropy decoder decodes and constructs syntax elements containing information about the current block. One of the candidate intra prediction modes and the current block If the intra prediction mode is not the same as the current block's a prediction unit for deriving an intra-frame prediction mode, and the table includes an intra-frame prediction mode and an intra-frame prediction This is a table that maps the image data to the index information of the mode. If the prediction mode is planar mode, the index is 0. If the intra-frame prediction mode is DC mode, the index is 0. If it is a directional intra prediction mode, the index is 1. If it is a directional intra prediction mode, the index is 1. The syntax elements can be mapped to indices 2 to 34 by The value is coded using 5 bits, and the 5 bits of information are the One of the remaining intra prediction mode information excluding the multiple candidate intra prediction modes It may be information indicating an intra prediction mode. The code is an intra-frame prediction mode derived based on the neighboring blocks of the current block and an additional frame. There may be three different intra prediction modes derived based on the intra prediction mode. If the flag information is 1, the current block candidate screen is displayed based on the flag information. One candidate intra prediction mode among the intra prediction modes and the intra prediction mode of the current block If the flag information is 0, the current block is The multiple candidate intra prediction modes of the block and the intra prediction mode of the current block are not the same. The prediction unit can select a plurality of candidate images for the current block in the table. As the intra prediction mode, the remaining 32 intra prediction modes are The remaining 32 intra prediction mode information and the intra prediction mode of the current block are calculated. The remaining 32 pieces of intra-frame prediction mode information are mapped. One intra prediction mode mapped to the syntax element is used as the intra prediction mode of the current block. This can be realized by setting [Effects of the Invention]

[0008] As described above, according to the intra prediction method and the image decoding device according to the embodiment of the present invention, By encoding and decoding the intra-frame prediction mode information using a short number of bits, the image coding This can improve the encoding efficiency. [Brief explanation of the drawings]

[0009] [Figure 1] 1 is a block diagram illustrating an encoding device according to an embodiment of the present invention. [Figure 2] FIG. 2 is a block diagram illustrating a decoder according to an embodiment of the present invention. [Figure 3] 10 is a flowchart illustrating a method for decoding an intra-frame prediction mode of a current predictor according to an embodiment of the present invention. [Figure 4] FIG. 1 is a conceptual diagram illustrating a method for encoding an intra-frame prediction mode according to an embodiment of the present invention. [Figure 5] FIG. 1 is a conceptual diagram illustrating a method for decoding an intra-frame prediction mode according to an embodiment of the present invention. [Figure 6] FIG. 10 is a conceptual diagram illustrating a case where a code number mapping table according to an embodiment of the present invention is not used. [Figure 7]10 illustrates non-directional intra prediction modes and directional intra prediction modes when using 35 intra prediction modes according to an embodiment of the present invention. [Figure 8] 1 illustrates non-directional intra prediction modes and directional intra prediction modes when using 19 intra prediction modes according to an embodiment of the present invention. [Figure 9] 1 illustrates non-directional intra prediction modes and directional intra prediction modes when using 11 intra prediction modes according to an embodiment of the present invention. [Figure 10] 1 is a conceptual diagram illustrating a code number mapping and a code word mapping method excluding an MPM according to an embodiment of the present invention. [Figure 11] 1 is a conceptual diagram illustrating a part of an image encoding device and an image decoding device that perform encoding and decoding in an intra-frame prediction mode according to an embodiment of the present invention. [Figure 12] 1 is a conceptual diagram illustrating a part of an image encoding device and an image decoding device that perform encoding and decoding in an intra-frame prediction mode according to an embodiment of the present invention. DETAILED DESCRIPTION OF THE INVENTION

[0010] The components disclosed in the embodiments and drawings of the present invention are different from each other in characteristics of the image encoding device. Each component is shown as an independent unit to demonstrate its characteristic function. This does not mean that the software is made as a single piece of hardware or software configuration. , each component is arranged so as to be included individually in each component for the sake of convenience of explanation. At least two of the components are combined to form one component. Alternatively, one component may be divided into multiple components to perform the same function. The embodiments in which each component is integrated and separated do not depart from the essence of the present invention. As long as the above is true, it is within the scope of the present invention.

[0011] In addition, some of the components disclosed in the present invention may perform essential functions in the present invention. It is not a required component, but may simply be an optional component to improve performance. The invention is not limited to the components used merely to improve performance, but may be used to achieve the essence of the invention. It can be implemented by including only the essential components, and is a choice used solely for performance enhancement. A structure including only essential components, excluding essential components, is also included within the scope of the present invention.

[0012] FIG. 1 is a block diagram showing an encoding device according to an embodiment of the present invention.

[0013] As shown in FIG. 1, the encoding device includes a division unit 100, a prediction unit 110, an intra-frame prediction unit 103, and a , an inter-picture prediction unit 106, a transform unit 115, a quantization unit 120, a reordering unit 125, and an entropy The encoder 130, the inverse quantizer 135, the inverse transformer 140, the filter 145, and the memory 1 It can have 50.

[0014] The encoding device is realized by an image encoding method described in the following embodiments of the present invention. However, some of the components may be operated in a different way to reduce the complexity of the encoder and For example, intra-frame prediction is not performed in the prediction section. When performing this, all intra-frame prediction modes are used for real-time encoding. Instead of using a method for selecting the optimal intra-picture coding method, a limited number of intra-picture prediction methods are used. The prediction modes are used, and one of them is selected as the final intra prediction mode. As yet another example, a method for selecting between intra-frame prediction and inter-frame prediction may be used. When performing the above, it is also possible to restrict the type of prediction block to be used.

[0015] The unit of blocks processed by the encoding device is a coding unit for encoding and a prediction unit for prediction. The coding unit is called a CU (Coding Unit). The prediction unit is PU (Prediction Unit), and the transformation unit is TU (Transformation Unit). This can be expressed in terms of the sform unit.

[0016] The division unit 100 divides one image (picture) into a plurality of coding blocks and prediction blocks. , and transform block combinations, and then, based on a predetermined criterion (e.g., a cost function), Select one combination of coding block, prediction block, and transformation block from For example, to divide an image into coding units, Inductive trees such as QuadTree Structures In the following, in the embodiment of the present invention, the meaning of the coding block is defined as follows: Not only in the sense of a block that encodes, but also in the sense of a block that decodes It can be used.

[0017] A prediction block can be a unit for performing intra-frame prediction or inter-frame prediction. The blocks may be square blocks such as 2N x 2N or N x N. Using DIP (Short Distance Intra Prediction) The block for which inter prediction is performed may have a rectangular block shape. Prediction blocks of square shape such as 2Nx2N, NxN or square shape are the same shape 2N×N, N×2N, or asymmetric AMP (Asymetr There is a prediction block division method using ic motion partitioning. The conversion method in the conversion unit 115 may vary depending on the form of the prediction block.

[0018] The prediction unit 110 includes an intra prediction unit 103 that performs intra prediction and an inter prediction unit 104 that performs inter prediction. and a unit 106. The prediction block may be subjected to inter-frame prediction or It may be determined whether to perform intra-frame prediction. The processing unit in which prediction is performed and the prediction method may be determined. For example, when intra-frame prediction is performed, the unit of the prediction mode may differ from the unit of the processing block. The prediction code is determined based on the predicted block, and the prediction process is performed based on the transform block. The residual between the generated predicted block and the original block can be calculated. The difference value (residual block) can be input to the transform unit 115. The prediction mode information, motion vector information, etc. used are entropy coded together with the residual values. The signal can be encoded by the unit 130 and transmitted to the decoder.

[0019] When using PCM (Pulse Coded Modulation) encoding mode In this case, the original block is directly coded and decoded without performing prediction via the prediction unit 110. It is also possible to transmit the signal to a decoding unit.

[0020] In the intra prediction unit 103, the prediction unit 103 predicts the blocks existing around the current block (the block to be predicted). An intra-predicted block can be generated based on the reference pixels. In order to calculate the optimal intra prediction mode for the current block, the intra prediction is performed on the current block. The blocks are generated using multiple intra-frame prediction modes, and then One predicted block can be selected and used as the predicted block for the current block. In the intra-frame prediction method, the intra-frame prediction mode uses reference pixel information depending on the prediction direction. and a non-directional mode that does not use directional information when making predictions. It is possible to have a mode for predicting luminance information and a mode for predicting chrominance information. In order to predict color difference information, intra prediction is used to predict luminance information. Mode information or predicted luminance signal information can be utilized.

[0021] The current block for which intra prediction is performed using one intra prediction mode is the current block. The current block is determined based on the intra prediction mode information used when the surrounding blocks perform intra prediction. The intra prediction mode of the current block is predicted, and the intra prediction mode information of the current block is coded. In other words, the intra-frame prediction mode of the current block can be set to the Prediction can be performed from the intra-frame prediction mode of the predicted block to be predicted. The following method is used to predict the intra prediction mode of the current block using the mode information. Such a method can be used.

[0022] 1) The intra prediction mode of the current block is the same as the intra prediction mode of the surrounding blocks. If so, a predetermined flag information is coded to determine the intra prediction mode of the current block and the surrounding blocks. In this case, information that the intra prediction mode is the same as that of the other frame can be transmitted.

[0023] 2) If the intra prediction mode of the current block and the intra prediction mode of the surrounding blocks are different, If different, the intra prediction mode of the current block is used to encode the prediction mode information of the current block. The measurement mode information can be entropy coded.

[0024] In 1) and 2), the intra prediction mode of the current block is used to encode the The intra prediction modes of the surrounding blocks are defined as candidate intra prediction modes. Used.

[0025] When performing the above methods 1) and 2), if the intra-frame prediction mode of the surrounding blocks cannot be used, (e.g., there are no surrounding blocks or the surrounding blocks perform inter prediction) In this case, the intra prediction mode of the current block is set to a specific predefined intra prediction mode. It can be set as a candidate intra-frame prediction mode value for prediction.

[0026] The intra-screen prediction unit 103 predicts the pixel information of the current block from the reference pixels around the current block. The intra-predicted block can be generated based on the surrounding information of the current block. The edge blocks are blocks that have undergone inter-frame prediction, and the reference pixels are blocks that have undergone inter-frame prediction. In this case, the pixel in question can be used to The pixel of the block where the intra prediction was performed is used as a reference pixel. The current block can be intra-predicted, i.e., the reference pixels are not available. In this case, other pixels can be used in place of the unused reference pixels.

[0027] A prediction block can contain multiple transform blocks. , when the size of the prediction block and the size of the transformation block are the same, Based on the pixel on the side, the pixel on the top left, and the pixel on the top Intra prediction can be performed on the predicted block. However, when intra prediction is performed, The size of the prediction block is different from the size of the transformation block, and multiple If a transformation block is included, the reference pixel determined based on the transformation block is used. Intra-frame prediction can be performed using this method.

[0028] Also, one coding block may be divided into multiple prediction blocks, but the coding block Only in the smallest coding block corresponding to the smallest block size, one coding block is Intra prediction is performed using an NxN partition where the block is divided into four square prediction blocks. It is possible.

[0029] Intra-frame prediction method uses MDIS (Mode Decoder Information Set) for reference pixels depending on the intra-frame prediction mode. After applying the Endent Intra Smoothing filter, the predicted block The type of MDIS filter applied to the reference pixel can be varied. An additional filter applied to the predicted block within a picture after intra prediction has been performed. The MDIS filter, which is a filter for the reference pixel and the image generated after the prediction, It can be used to reduce the discrepancy between the predicted block and the MD When performing IS filtering, the reference pixel and the image pixel are selected depending on the direction of the intra-frame prediction mode. Various filtering is performed on some columns included in the predicted block within a plane. It is possible to do so.

[0030] According to an embodiment of the present invention, when performing intra prediction according to the size of the current block, The number of available intra prediction modes may vary. The number of intra-frame prediction modes that can be used depends on the size of the current target block. Therefore, when intra prediction is performed on the current block, The size of the block is determined, and the image that can be used by performing intra-image prediction is An intra prediction mode may be determined.

[0031] The inter prediction unit 106 predicts at least one of the images before and after the current image. Prediction can be performed by referencing the block information contained in one image. 106 may include a reference image interpolation unit, a motion prediction unit, and a motion compensation unit.

[0032] The reference image interpolation unit receives reference image information from the memory 150 and interpolates integer pixels in the reference image. For luminance pixels, it is possible to generate integer pixel information in 1 / 4 pixel units. To generate the following pixel information, a DCT-based 8-tab interpolation filter with different filter coefficients is used. A DCT-based Interpolation Filter can be used. In the case of color difference signals, in order to generate pixel information of less than an integer pixel in 1 / 8 pixel units, DCT-based 4-tab interpolation filter with different filter coefficients A polarization filter may be used.

[0033] The inter-picture prediction unit 106 performs motion prediction based on the reference picture interpolated by the reference picture interpolation unit. As a method for calculating motion vectors, FBMA (Full Motion Transformation Algorithm) search-based Block Matching Algorithm), T SS(Three Step Search), NTS(New Three-Step Various methods can be used, such as the Motion Vector Search Algorithm. The motion vector values ​​may be in 1 / 2 or 1 / 4 pixel units based on the interpolated pixels. The inter prediction unit 106 performs one of various inter prediction methods. The current block can be predicted by applying the inter-frame prediction method. For example, the Skip method, the Merge method, and the Advanced Multimedia Privacy Policy (AMVP) Various methods, such as the Implemented Motion Vector Prediction (EMP) method can be used.

[0034] The predicted block (intra-predicted block or inter-predicted block) generated by the prediction unit 110 The residual is the difference between the predicted block and the original block. A residual block containing the residual information may be generated.

[0035] The residual block may be input to a transform unit 115. The transform unit 115 transforms the original block The residual block containing the residual value information between the predicted block and the (Discrete Cosine Transform) or DST(Discrete It can be transformed using a transformation method such as the sine transform. Whether to apply DCT or DST to transform the residual block depends on the residual Intra-frame prediction mode information of the predicted block used to generate the block and the predicted block That is, the transform unit can determine the size of the predicted block. The conversion method can be applied differently depending on the size and prediction method of .

[0036] The quantization unit 120 can quantize the value converted into the frequency domain by the conversion unit 115. The quantization coefficient can vary depending on the block or the importance of the image. The values ​​calculated by the quantization unit 120 are provided to the inverse quantization unit 135 and the reordering unit 125. can.

[0037] The reordering unit 125 may perform reordering of coefficient values ​​with respect to the quantized residual values. The reordering unit 125 performs coefficient scanning. The modulus allows us to convert two-dimensional block-type coefficients into one-dimensional vector form. For example, the reordering unit 125 uses a zig-zag scan method. and scans from DC coefficients to high frequency coefficients, and converts them into one-dimensional vector form. The size of the transformation block and the intra-frame prediction mode can be changed. As an alternative to the row-scan method, a vertical scan method is used, which scans two-dimensional block-type coefficients in the column direction. Scanning method: horizontal scanning method is used, which scans two-dimensional block-type coefficients in the row direction. That is, depending on the size of the transform block and the intra-frame prediction mode, zigzag One of the scanning methods is used: vertical scanning, horizontal scanning, or vertical scanning. The decision can be made as to whether or not the information will be used.

[0038] The entropy coding unit 130 performs entropy coding based on the value calculated by the reordering unit 125. Entropy coding can be performed using, for example, Exponential Golomb (Exp onential Golomb), CAVLC(Context-Adaptive Variable Length Coding), CABAC(Context-Ad Various coding schemes such as Adaptive Binary Arithmetic Coding Encryption methods can be used.

[0039] The entropy coding unit 130 receives the coded block from the reordering unit 125 and the prediction unit 110. Residual value coefficient information and block type information, prediction mode information, division unit information, prediction block block information and transmission unit information, motion vector information, reference frame information, block interpolation information, It receives various information, such as filtering information, and generates entropy based on a predetermined encoding method. In addition, the entropy coding unit 130 can perform the reordering unit 125. The coefficient values ​​of the coding unit input from can be entropy coded.

[0040] The entropy coding unit 130 performs HEB (High Efficiency Block Coder) coding using CABAC. Concurrency Binarization method or CABAC bypass coding The HTB (High Throughput) algorithm utilizes the CAVLC coefficient binarization method. Entropy coding can be performed using the Binarization method. do.

[0041] The entropy coding unit 130 performs binarization on the intra-frame prediction mode information to obtain the current The entropy coding unit 13 can encode the intra-frame prediction mode information of the block. 0 may include a codeword mapping unit for performing such a binarization operation. Depending on the size of the prediction block used for intra-picture prediction, binarization can be performed in various ways. The code word mapping unit uses the code word mapping table for the binary conversion operation. or by pre-storing a codeword mapping table. In yet another embodiment, the entropy coding unit 130 may A code number mapping unit performs code word mapping and a code word mapping unit performs code word mapping. The intra prediction mode information of the current block can be expressed using the coding part. The code number mapping unit and the code word mapping unit are configured with a code number mapping table and a code A word mapping table can be generated or stored.

[0042] The inverse quantization unit 135 and the inverse transform unit 140 inversely quantize the values ​​quantized by the quantization unit 120. The inverse quantization unit 135 and the inverse transformation unit 140 perform inverse transformation on the values ​​transformed by the transformation unit 115. The residual value (Residual) generated in the prediction unit 110 is passed through a motion estimation unit, a motion A reconstructed block is obtained by combining the predicted block predicted by the compensation unit and the intra prediction unit. (Reconstructed Block) can be generated.

[0043] The filter unit 145 includes a deblocking filter, an offset correction unit, and an ALF (Adaptive Lattice Filter). The filter may include at least one of the following:

[0044] The deblocking filter removes blockages caused by inter-block boundaries in the reconstructed image. To determine whether to perform deblocking, the block distortion can be removed. Based on the pixels contained in some columns or rows contained in the block, It is determined whether to apply a deblocking filter. When applying a filter, the stronger the filter ( Apply a Strong Filter or a Weak Filter. A deblocking filter can also be applied to remove vertical filtering and watermarks. When flat filtering is performed, horizontal filtering and vertical filtering are performed in parallel. Row processing can be performed.

[0045] The offset correction section corrects the original pixel-by-pixel ratio for the deblocked image. It is possible to correct the offset between the image and the target image. To do this, the pixels in the image are divided into a certain number of regions, and the regions to be offset are determined as follows: Determine how to apply an offset to the region, or consider the edge information of each pixel. A method of applying an offset by

[0046] ALF (Adaptive Loop Filter) is a filtered restoration Filtering can be performed based on the value of comparing the image with the original image. The pixels are divided into at least one group and then applied to the group. You can define one filter and filter each group individually. Information regarding whether F is applied is provided for each coding unit (CU). The size and coefficients of the ALF applied by each block may vary. The ALF can have a variety of forms, and the filters it contains can be The number of coefficients can also be changed. filter coefficient information, ALF On / Off information, filter type information) are stored in the bitstream. The parameter may be transmitted in the form of a parameter set.

[0047] The memory 150 stores the reconstructed blocks or images calculated through the filter unit 145. When performing inter-frame prediction, the stored reconstruction block or image can be 110.

[0048] FIG. 2 is a block diagram illustrating a decoder according to an embodiment of the present invention.

[0049] As shown in FIG. 2, the decoder includes an entropy decoding unit 210, a reordering unit 215, an inverse The image processing system includes a decoder 220, an inverse converter 225, a predictor 230, a filter 235, and a memory 240. It may also be used.

[0050] When a bitstream is input from the encoder, the input bitstream is coded The data may be decoded in the reverse order of the encoder.

[0051] The entropy decoding unit 210 is an entropy coder in the entropy coding unit of the encoder. Entropy decoding can be performed by reversing the procedure used to entropy decryption. Among the information decoded by the decoder 210, the information for generating a prediction block is input to the predictor 220. 30, and the residual value obtained by entropy decoding in the entropy decoding unit is sent to the reordering unit. 215.

[0052] The entropy decoding unit 210 also uses CABAC in the same way as the entropy coding unit. Among the HTB methods that utilize the HEB or CAVLC coefficient coding methods, at least Another method can be used to perform the inverse transformation.

[0053] The entropy decoding unit 210 performs the entropy decoding on the intra-frame prediction and inter-frame prediction performed by the encoder. The entropy decoding unit can decode the information related to the intra-frame prediction mode number. A codeword map to generate a codeword received as A codeword mapping section may be included to include a mapping table. The code mapping table can be pre-stored or adaptively generated. When using a code number mapping table, the code number mapping table is used to perform code number mapping. A node number mapping unit may further be provided.

[0054] The reordering unit 215 reorders the bits entropy decoded by the entropy decoding unit 210. The reordering can be done based on the way the stream was reordered in the encoding unit. Coefficients expressed in block form are restored to coefficients in two-dimensional block form and rearranged. The reordering section provides information related to the coefficient scanning performed in the coding section. and scanning the data in reverse order based on the scanning order performed by the encoding unit. Thus, realignment can be performed.

[0055] The inverse quantization unit 220 uses the quantization parameters provided by the encoder and the realigned blocks. Inverse quantization can be performed based on the coefficient values ​​of

[0056] The inverse transform unit 225 performs DCT and The inverse DCT and inverse DST can be performed on the DST. In the transform part of the encoder, the DCT and DST are performed based on the predicted transmission unit. It is selectively performed depending on multiple pieces of information such as the measurement method, the size of the current block, and the prediction direction. The inverse transform unit 225 of the decoder may perform the following on the basis of the transform information performed by the transform unit of the encoder: The inverse transformation can be performed.

[0057] When performing a transformation, it is possible to perform the transformation based on a coding block that is not a transformation block. Cut.

[0058] The prediction unit 230 receives the information related to the generation of the predicted block from the entropy decoding unit 210. based on the information and previously decoded block or image information provided by memory 240. A prediction block can be generated using the

[0059] As mentioned above, when intra prediction is performed as in the encoder, the predicted block If the size of the predicted block is the same as the size of the transform block, the pixels to the left of the predicted block The predicted block is based on the pixel in the upper left corner, the pixel in the upper left corner, and the pixel in the upper left corner. When performing intra-frame prediction, the transform block is not included in the predicted block. In this case, intra prediction can be performed using reference pixels based on the transform block. Also, as mentioned above, NxN division is used only for the minimum size coding block. Intra prediction may be used.

[0060] The prediction unit 230 may include a prediction unit determination unit, an inter-picture prediction unit, and an intra-picture prediction unit. The prediction unit discrimination unit determines the prediction unit information input from the entropy decoding unit, the intra-frame prediction It receives various information such as prediction mode information for the measurement method, motion prediction related information for the inter-frame prediction method, etc. Then, the prediction block is classified into the current coding block, and whether the prediction block is an inter-picture prediction block or not is determined. In either case, it is possible to determine whether or not to perform intra-frame prediction. The current prediction block is generated by using the information necessary for inter-frame prediction of the current prediction block provided by At least one image before or after the current image contains Based on the obtained information, inter prediction can be performed on the current prediction block.

[0061] In order to perform inter-frame prediction, the coding block is used as a reference and the The motion prediction method for the predicted block is Skip Mode, Merge Mode, Merge Mode, AMVP Mode, or You can determine whether it is possible or not.

[0062] Intra prediction can generate a predicted block based on pixel information in the current image. If the predicted block is an intra-predicted block, the Intra prediction can be performed based on the intra prediction mode information of the selected prediction block. The intra-picture prediction unit may include an MDIS filter, a reference pixel interpolation unit, and a DC filter. The MDIS filter is a part that filters the reference pixels of the current block. , and determines whether to apply a filter depending on the intra-frame prediction mode of the current prediction block, and applies the filter. The prediction mode and MDIS filter of the prediction block provided by the encoder can be used. MDIS filtering may be performed on the reference pixels of the current block using this information. If the prediction mode of the current block is a mode that does not perform MDIS filtering, MDI The S filter may not be applied. Also, like the encoder, it generates the prediction block in the same way. After this, further filtering may be performed together with the reference pixels.

[0063] The reference pixel interpolation unit determines whether the prediction mode of the prediction block is based on the pixel value obtained by interpolating the reference pixel. In the case of a predicted block that performs intra-frame prediction, the reference pixels are interpolated to obtain a pixel unit of an integer value or less. Reference pixels can be generated. The prediction mode of the current prediction block does not interpolate reference pixels. If the prediction mode is a prediction mode that generates a prediction block in D, the reference pixels do not need to be interpolated. The C filter is used to filter out the current block if the prediction mode is DC mode. The predicted block can be generated by:

[0064] The reconstructed block or image may be provided to a filter unit 235. 235 may include a deblocking filter, an offset correction unit, and an ALF.

[0065] Whether the encoder has applied a deblocking filter to the block or image. If a deblocking filter is applied, whether a strong filter is applied or not can receive information about whether a weak filter was applied. The deblocking filter uses the information provided by the encoder to provide the deblocking filter. The decoder can then perform deblocking filtering on the block. As with the encoder, first, vertical deblocking filtering and horizontal deblocking filtering are performed. In the overlapping area, vertical deblocking and water filtering are performed. At least one of horizontal deblocking and vertical deblocking may be performed. Blocking filtering or horizontal deblocking filtering is performed in conjunction with vertical deblocking. Even if blocking filtering and horizontal deblocking filtering are performed in the overlapping area, Through this deblocking filtering process, Parallel processing of the data is possible.

[0066] The offset correction unit determines the type and offset of the offset correction applied to the image during encoding. Offset correction can be performed on the restored image based on bit value information, etc.

[0067] ALF (adaptive loop filter) is a filter that , and filtering can be performed based on the value of the restored image compared to the original image. Encoding is performed based on the ALF applicability information and ALF coefficient information provided by the encoder. ALF can be applied to units. Such ALF information is available for a particular parameter set. The information may be provided in a package.

[0068] The memory 240 stores the reconstructed image or block and stores the reference image or block. The image can be made available as a link and the restored image can be provided to the output. This can be done.

[0069] As mentioned above, in the following embodiments of the present invention, for convenience of explanation, the coding unit Although the term "Coding Unit" is used as the coding block, The block may be a block that performs not only encoding but also decoding. The intra prediction methods described in FIGS. 3 to 12 are the same as those described in FIGS. 1 and 2. Such an encoder and decoder can be implemented according to the functions of the present invention. falls within the scope of rights.

[0070] The intra prediction mode of the current predicted block is predicted from the intra prediction mode of the surrounding blocks. Such a prediction method for intra-frame prediction mode is called MPM (Most Probable Multiplier). The intra prediction mode of the blocks on the left and above the current block is called the The first MPM and the second MPM are determined based on the mode or frequently used intra-frame prediction mode. and the intra prediction mode of the current block is the first MPM and the second MPM. If the prediction mode is the same as at least one of the prediction modes, the intra prediction mode of the current block is determined to be the same as the intra prediction mode of the current block. "prev_i" is information indicating that it is the same as the intra-frame prediction mode of MPM. ntra_pred_flag" and the first MP via "mpm_idx" information M, which of the second MPMs is an intra-frame prediction mode and the intra-frame prediction mode of the current prediction block? If the current block's intra prediction mode is the same as the If the intra prediction mode of the current block is not the same as that of the MPM, The information to be used is coded in the "rem_intra_luma_pred_mode" information. It is possible.

[0071] FIG. 3 illustrates a method for decoding an intra prediction mode of a current block according to an embodiment of the present invention. FIG.

[0072] As shown in FIG. 3, in order to decode the intra prediction mode information of the current block, The "prev_intra_pred_flag" information can be decoded (step Top S300).

[0073] The decoded "prev_intra_pred_flag" information is either 1 or 0. It is determined whether or not the data can be read (step S310).

[0074] In the embodiment of the present invention, when the "prev_intra_pred_flag" information is 1, If so, it is determined that the MPM and the intra prediction mode of the current block are the same, and If the "_intra_pred_flag" information is 0, the MPM and the current block screen It is determined that the intra prediction mode information is different, but "prev_intra_pred_flag Other binarization methods for the information or other methods of representing the information may also be used.

[0075] If the current block is within the screen via "prev_intra_pred_flag" If it is determined that the prediction mode and the intra-frame prediction mode of the MPM are the same, dx" and the intra prediction mode of the current block is the first MPM or the second MPM. It is possible to obtain information on whether the intra-frame prediction mode is the same as the (Step S320). If "prev_intra_pred_flag" is currently If it is determined that the intra prediction mode of the block and the intra prediction mode of the MPM are not the same, In this case, the intra prediction mode of the current block is the remaining mode information "rem_intra _luma_pred_mode" information is decoded to obtain the intra prediction mode information for the current block. The information can be obtained (step S330).

[0076] To encode and decode "rem_intra_luma_pred_mode" information For this purpose, the codeword information mapped to the intra prediction mode can be used. Table 1 below shows the exponential arithmetic, which is one of the methods for binarizing intra-frame prediction mode information. This shows the encoding method.

[0077] TIFF0007822430000001.tif69169

[0078] As shown in Table 1, the code number (codeNum ) can be mapped with a shorter codeword. If you map a short codeword onto the information you want to share, you can convert the same information into a shorter bitstream. This increases the coding and decoding efficiency.

[0079] Table 2 below lists the order of intra-frame prediction modes according to an embodiment of the present invention. Table 2 can also be written as Table 3.

[0080] TIFF0007822430000002.tif63169

[0081] TIFF0007822430000003.tif45169

[0082] Table 4 below shows a mapping between intra prediction modes and codewords according to an embodiment of the present invention. The code words in Table 4 are arbitrarily set. Without departing from the essence of the present invention, other codewords may be used to select the current intra prediction mode. It is possible to map information related to

[0083] TIFF0007822430000004.tif54169

[0084] As shown in Table 4, the mapping between the intra prediction mode and the codeword according to the embodiment of the present invention is In our coding method, we map short codewords to frequently occurring intra prediction modes, The more frequently an intra-frame prediction mode occurs, the shorter the codewords can be. Cut.

[0085] Hereinafter, based on the order generated in the embodiment of the present invention, the shortest or earliest mapping is performed. The codeword to be coded is defined as the first codeword, and then the second codeword is defined as the , the third codeword, the fourth codeword, the nth codeword, etc. That is, the length of the nth codeword is shorter than the length of the n+1th codeword. The length of the nth codeword can be equal to or less than the length of the n+1th codeword. (where n is an integer)

[0086] When encoding using intra prediction, the non-directional intra prediction mode is used. The planar mode and DC mode, which are the modes used, are frequently used. The intra-picture prediction mode is used relatively infrequently. Short codewords are mapped to directional intra prediction modes, and non-directional intra prediction modes are mapped to A codeword longer than the codeword mapped to the prediction mode is used for the directional intra prediction mode. By mapping the image to a code, the efficiency of encoding and decoding the image can be improved. This can be done.

[0087] Tables 5 and 6 show different binary encoding methods according to the intra-frame prediction mode in the embodiment of the present invention. 1 is a table showing how to do this.

[0088] TIFF0007822430000005.tif50169

[0089] TIFF0007822430000006.tif60169

[0090] Table 5 shows the cases where 19 modes are used as intra-frame prediction modes. Table 6 shows the intra prediction modes when using 35 modes as intra prediction modes. This shows the binary method for representing it.

[0091] As shown in Tables 5 and 6, the binary code for representing the intra-frame prediction mode of the current block is There are two encoding methods: unary code and fixed length. As shown in Table 5, when the intra prediction mode is 0 or 1, In this case, the prefix using the unary code is fixed to 0, Fixed length to distinguish between 0 and 1, which are intra prediction modes. In addition, when the intra-frame prediction mode is 2 to 5, The prefix using ary code is fixed to 10, and intra-frame prediction modes 2 to 5 are distinguished. In order to achieve this, the fixed length can be expressed in 2 bits. In the formula, the codeword and intra-frame prediction are calculated using a unary code and fixed length. Table 6 also shows how to map intra-frame prediction mode information in this way. The information can be expressed in binary.

[0092] In the cases of Tables 5 and 6, as in Table 2, the smaller the intra-frame prediction mode number, the lower the binarization rate. Therefore, the method of generating short code words can be used. According to the embodiment, the intra prediction modes with small numbers are frequently generated. By placing the code, the information can be expressed with fewer bits, improving coding efficiency. can be increased.

[0093] FIG. 4 is a conceptual diagram illustrating a method for encoding an intra-frame prediction mode according to an embodiment of the present invention. be.

[0094] As shown in FIG. 4, according to an embodiment of the present invention, when encoding an intra prediction mode, , using the codeNum mapping table It can be used.

[0095] Unlike the embodiment of Table 5 or Table 6 described above, the code number mapping table (co deNum mapping table), any intra prediction mode If determined, the determined intra prediction mode, which is the t-th in the code number mapping table, is By swapping the (t-1)th intra-frame prediction mode with the (t-1)th intra-frame prediction mode, By increasing the rank of the new intra-frame prediction mode in the code number mapping table, If the intra prediction mode is the intra prediction mode with a high occurrence frequency, a smaller number of codes As a result, a short code word is mapped to that code number. That is, the code number ranking for the t-th intra prediction mode is raised. The rearranged code number mapping table is used to generate the intra prediction code for the next prediction block. As shown in Figure 4, the intra-frame prediction mode is determined as 2", and the code number "2" mapped to the second intra-frame prediction mode is The second code is read from the code word mapping table. The codeword ("1000") mapped to the number is used as the result of the second intra-frame prediction mode. The code number mapping table is displayed on the second screen. The first intra-frame prediction mode is swapped with the first intra-frame prediction mode located immediately above it to produce a second intra-frame prediction mode. The code number for the th intra-frame prediction mode is "1", and for the 1st intra-frame prediction mode is "2". The code number for the next step will be revised to "2". The code number mapping table is used when encoding the next predicted block in intra prediction mode. In still another embodiment, the occurrence frequency of the t-th and (t-1)-th intra prediction modes is Based on the frequency, it is determined whether or not to swap the t-th and (t-1)-th intra-frame prediction modes. That is, the occurrence frequency of the t-th intra prediction mode is determined as the (t-1)-th intra prediction mode. If the occurrence frequency of the t-th and (t-1)-th intra prediction modes is greater than the occurrence frequency of the t-th and (t-1)-th intra prediction modes, After swapping, the rearranged code number mapping table is used to Conversely, the tth intra prediction mode can be used to perform intra prediction for If the occurrence frequency of the (t-1)th intra-frame prediction mode is smaller than the occurrence frequency of the (t-1)th intra-frame prediction mode, The current code number mapping is used without swapping between the two intra-frame prediction modes. The table can be used when performing intra prediction for the next predicted block. In order to prevent the number of occurrences for all intra-frame prediction modes from increasing infinitely, Therefore, it is possible to periodically reduce the occurrence frequency for all intra prediction modes by the same percentage. In addition to the swapping process for the intra-frame prediction mode based on the frequency of the present embodiment, Other steps (e.g., determining code words) may be performed in the same manner as in the previous embodiment (directly without using frequency counts). It may be the same as swapping.

[0096] For convenience of explanation, in FIG. 4, the intra-frame prediction mode number and code number mapping table are shown. In this example, the intra prediction mode number and the code number are assumed to be the same. An initial code number mapping table for mapping the code number is determined in advance, and the intra-frame prediction is performed. It is possible to map code numbers which are numbers different from mode numbers.

[0097] As shown in FIG. 4, when an arbitrary intra-frame prediction mode is determined, the intra-frame prediction mode is The code number may be mapped to a code number value in a code number mapping table. The code number values ​​are mapped to code word values ​​via a code word mapping table. It may be pinged.

[0098] FIG. 5 is a conceptual diagram illustrating a method for decoding an intra-frame prediction mode according to an embodiment of the present invention. be.

[0099] As shown in Figure 5, the decoder uses the same initial codeword mapping table as the encoder. and an initial code number mapping table. A code word is read, and a code word is generated from the code word mapping table based on the code word. and obtains a mapped code number value from the The measurement mode is obtained from the code number mapping table, and finally the current block screen is In the decoding process, the intra prediction mode can be determined in the same way as in the encoder. When the prediction mode is determined, the code number mapping table is used to determine the intra-frame prediction mode. This allows for swapping.

[0100] When using the above method, the initial code number is used to prevent unnecessary swapping. The value of a mapping table is important because such a table can be used to create a slice or can be reinitialized in arbitrary units such as frames, and the features of the current slice or frame Swapping is performed after initialization to generate a code number mapping table that reflects the Therefore, the intra prediction mode and the code number value must be mapped. When setting the code number mapping table to be used, the frequently occurring intra-frame prediction mode value It is important to map smaller code number values ​​to the According to the above, the more frequently the intra-frame prediction mode number occurs, the smaller the code number. As a result, the length of the codeword for the intra prediction mode is By shortening it, coding efficiency can be expected.

[0101] FIG. 6 shows different uses of the code number mapping table according to the embodiment of the present invention. This is a conceptual diagram showing the case where

[0102] As shown in Figure 6, the binary coding of the intra prediction mode is performed without using MPM. Hexadecimal encoding can be done without generating separate flag information for the MPM. Based on the intra-frame prediction mode information of the blocks on the left and above the specific prediction block, A method for rearranging the code number mapping table can be used. For example, If the intra prediction modes of the blocks on the left and top are 2 and 3, 2 and 3 are used as the The remaining intra-frame prediction modes are placed at the top of the mode number mapping table. A method for constructing a code number mapping table can be used. The intra-frame prediction mode of the block existing above is set to the upper Then, the remaining blocks are placed in the left and upper side, excluding the intra-frame prediction modes of the blocks. The method of arranging the intra-picture prediction modes on the code number mapping table is used to A codeword corresponding to the intra prediction mode can be generated.

[0103] In the above-mentioned method of rearranging the code number mapping table, the picture of a specific predicted block Intra prediction mode is intra prediction of a block located to the left and / or above a specific prediction block. It is assumed that there is a high possibility that the measurement mode is identical to at least one of the modes. Therefore, the intra-frame prediction modes that occur relatively frequently are included in the code number mapping. By placing it above the mapping table, the above-mentioned code number mapping table can be rearranged. Here, the intra prediction mode that occurs frequently can be minimized in a specific prediction block. The prediction mode may be the intra prediction mode of the block on the left and / or top of the block. If the intra prediction mode that occurs frequently is located on the left side and If the intra-frame prediction mode of the block located above or below the frame is used, a relative rearrangement occurs. This reduces the probability of unnecessary rearrangement, and therefore prevents unnecessary rearrangement. Intra prediction of a block whose prediction mode is to the left and / or above a specific prediction block If not in the mode, the screen of the blocks to the left and / or above the specific predicted block The intra prediction mode is positioned at the top of the code number mapping table, and the remaining intra prediction modes are When rearranging the code number mapping table, the shorter code word is In order to allocate the most frequently occurring intra prediction modes among the remaining intra prediction modes, As a result, the code numbers that occur more frequently should be mapped to the code numbers that occur less frequently. Higher intra prediction modes are present to the left and / or above a particular predicted block. Regardless of whether it is the same as the intra-frame prediction mode, a smaller code number can be assigned. This results in advantages in terms of compression performance and / or complexity.

[0104] In yet another embodiment, MPM is used in binary coding of intra prediction modes. Binary encoding can be done, but without generating separate flag information for the MPM. Based on the intra-frame prediction mode information of the first MPM and the second MPM of a specific prediction block, The code word for the MPM is then rearranged using the method of rearranging the code number mapping table. For example, the first MPM and the second MPM intra prediction can be assigned. If the mode is 2 and 3, place 2 and 3 at the top of the code number mapping table. , how to construct the code number mapping table by postponing the remaining intra-frame prediction modes. Other processes can be used to the left and above a particular prediction block. Before the intra prediction mode of the block to be coded is positioned at the top of the code number mapping table, It may be the same as the above embodiment.

[0105] According to an embodiment of the present invention, the intra prediction mode as described above is A different number of intra prediction modes can be used depending on the size.

[0106] Table 5 below shows the number of intra-frame prediction modes that can be used depending on the size of the prediction block. This is a table.

[0107] TIFF0007822430000007.tif55169

[0108] As shown in Table 7, the size of the prediction block is either 4x4 or 64x64. In this case, the number of intra prediction modes may be 11, 18, or 19. Intra prediction mode when block size is 16x16, 32x32, 64x64 The number of available items is 35.

[0109] The 35 intra-frame prediction modes are listed in Table 8 below, with the intra-frame prediction mode number and the corresponding It can have a name.

[0110] TIFF0007822430000008.tif108169

[0111] FIG. 7 shows a non-directional prediction method when using 35 intra-frame prediction modes according to an embodiment of the present invention. 10 shows an intra-frame prediction mode and a directional intra-frame prediction mode.

[0112] In the case where there are 35 intra-frame prediction modes for performing intra-frame prediction on a prediction block, In this case, Planar, DC, Ver+x (x is an integer between -8 and 8), Hor+x (x may be an integer between -7 and 8.

[0113] FIG. 8 shows a non-directional prediction scheme when using 19 intra-frame prediction modes according to an embodiment of the present invention. 10 shows an intra-frame prediction mode and a directional intra-frame prediction mode.

[0114] In the case where there are 19 intra prediction modes for performing intra prediction on a prediction block, In this case, planar, DC, Ver+2x (x is an integer between -4 and 4), Hor+2x (x is an integer between -3 and 4). That is, 19 intra-frame prediction modes Unlike the case of using 35 intra-frame prediction modes in FIG. 7, To use a method for selecting only intra-frame prediction modes for angular motions that are multiples of 2 from can be done.

[0115] When using 18 intra-frame prediction modes, all the modes except for the 18th intra-frame prediction mode are used. It is possible to perform intra-frame prediction using intra-frame prediction modes from 1 to 17. .

[0116] FIG. 9 shows a non-directional prediction scheme when using 11 intra-frame prediction modes according to an embodiment of the present invention. 10 shows an intra-frame prediction mode and a directional intra-frame prediction mode.

[0117] In the case of 11 intra-frame prediction modes, planar, DC, Ver+4x (x is Hor+4x (where x is an integer between -1 and 2) can be used. That is, the 11 intra-frame prediction modes are used in the 35 intra-frame prediction modes in FIG. Unlike the case of the 3D image, the intra-frame prediction mode is for angular motions that are multiples of 4 from the vertical and horizontal directions. The 11 intra-frame prediction modes are shown in Figure 9. In order to generate nine modes from mode 4 to mode 10, VER (mode 2) The space based on the 9 modes (mode 1) and HOR (mode 3) is divided into 4 equal parts. Adding lanar (mode 0) and DC (mode 1), there are a total of 11 modes. It is possible.

[0118] As shown in Figures 8 and 9, 11 or 19 intra-frame prediction modes are used for 64x64 blocks. It can be used on a 4x4 or 4x4 block.

[0119] If the size of the prediction block is 64x64, the block is divided into smaller blocks (for example, This means that there is no need to use a different block size (e.g., 32x32, 16x16, etc.). Therefore, it can be determined that there is no significant change in the pixel values ​​of the black lines. When all 35 global modes are evaluated for a block, most of the in-screen predictions are The predicted values ​​for the observed modes were similar, and the magnitude of performance for each mode with similar directionality was Therefore, considering the complexity, it is not possible to evaluate all 35 modes. In terms of complexity, it is more effective to perform intra-frame prediction based only on a partial intra-frame prediction mode than to perform intra-frame prediction based on a partial intra-frame prediction mode. Therefore, in the present invention, as shown in FIGS. 8 and 9, a 64×64 prediction block is used. 11 intra-frame prediction modes, 18 intra-frame prediction modes, 19 intra-frame prediction modes Any one of the modes can be used for intra-frame prediction.

[0120] Even in the case of a 4x4 prediction block, the size of the prediction block is small, so 35 The predicted values ​​for each mode created as an intra-frame prediction mode may be similar, and all 35 Rather than using all intra-frame prediction modes, we use prediction blocks of 64x64 size. Intra prediction may be performed using one of 18 or 19 intra prediction modes. .

[0121] As mentioned above, 11, 19 or 35 intra prediction modes are used, and the MPM mode If the number of codes is three, the code number mapping for remaining modes other than MPM is By generating a coding table, the coding efficiency can be improved.

[0122] Hereinafter, the MPM used in the embodiment of the present invention is the intra prediction mode of the current block. This is used as a general concept of candidate intra-frame prediction modes for predicting the frame value. A similar concept to MPRM (Most Probable Remaining Model) ode) may be used and is within the scope of the present invention, but for the sake of convenience, we will use MPM. Disclosures will only be made about:

[0123] FIG. 10 shows the remaining (remaining) intra prediction excluding the MPM according to the embodiment of the present invention. FIG. 10 is a conceptual diagram illustrating a method for mapping code numbers to modes.

[0124] As shown in FIG. 10, when the MPM is an intra-frame prediction mode number of 2, 4, or 5, the code In the number mapping table, the code numbers for the remaining intra-frame prediction modes excluding 2, 4, and 5 are used. In other words, when encoding in remaining mode, the MPM The corresponding intra-frame prediction mode is not selected, so it is not applicable to the remaining modes except for MPM. Code number values ​​can only be assigned by codeword mapping. A link table can be used to map the link to a codeword. The codeword is mapped only to the intra prediction mode corresponding to the main mode. This prevents unnecessary codewords from being wasted and improves coding efficiency. do.

[0125] Similarly, when performing a decoding operation, a code number is generated based on the input code word. Except for the intra-frame prediction mode that corresponds to MPM, the mapped code number mapping The intra-frame prediction mode information can be decoded using the decoding table.

[0126] FIG. 10 shows a case where 11 intra-frame prediction modes are used. Intra prediction mode is used to perform intra prediction for the predicted block, and three MPMs are used. When the remaining intra prediction modes are used, the codewords are mapped to the eight remaining intra prediction modes. If the remaining intra prediction mode is expressed as a fixed length, the codeword length is 3 bits. The remaining eight intra prediction modes can be mapped using the codewords.

[0127] Table 9 below shows the remaining intra-frame prediction mode according to an embodiment of the present invention in fixed length. This is what was done.

[0128] TIFF0007822430000009.tif76169

[0129] As shown in Table 9, the code numbers are mapped to code words using fixed-length coding. obtain.

[0130] In this method, the intra-frame prediction modes are 11, 19, and 35, and the MPM If there are 3, the remaining modes will be 8, 16, and 32. The number of remaining modes is an exponent power of 2, and the remaining modes are fixed by each exponent bit. The table below shows the fixed length of the exponent bits depending on the number of remaining bits. If so, this indicates the length of the codeword to indicate the remaining mode.

[0131] TIFF0007822430000010.tif37169

[0132] As shown in Table 9, when the number of intra-frame prediction modes is 11 and the number of MPMs is 3, In this case, the codeword can use 3 bits to indicate the remaining mode. If the number of intra-frame prediction modes is 19 and the number of MPMs is 3, The codeword for indicating the mode can use 4 bits. If the number of is 35 and the number of MPMs is 3, A codeword can use 5 bits, i.e., according to an embodiment of the present invention , the number of intra prediction modes of the current block corresponds to the remaining intra mode. The remaining modes may be generated by exponential powers of two of the number, i.e., each exponential bit In the above embodiment, it can be determined that the M The number of PMs is assumed to be 3. If the number of MPMs changes, the image quality of the predicted block The number of intra prediction modes used for intra prediction may vary.

[0133] FIG. 11 shows an image coder for encoding and decoding in an intra-frame prediction mode according to an embodiment of the present invention. FIG. 1 is a conceptual diagram showing a part of an image encoding device and an image decoding device.

[0134] Referring to the upper part of FIG. 11, the image is coded by a codeword mapping unit 1100 of the image coding device. It is possible to generate a codeword by receiving an intra-picture prediction mode number. To perform the code word mapping, the code word mapping unit 1100 needs to The codeword mapping table may also be stored. The codeword mapping unit may perform a binarization operation to generate the entropy. It may be included as part of the p-coding unit. Among the modes, non-directional modes such as DC mode or planar mode, which occur frequently, A small number of intra prediction mode numbers can be assigned to the intra prediction modes. The smaller the code number, the shorter the codeword that is mapped. A mapping table can be generated.

[0135] As shown in the lower part of FIG. 11, the codeword mapping unit 1110 of the image decoding device The codeword is received and the input codeword is converted into a codeword included in the codeword mapping unit. It can be generated as an intra-frame prediction mode number based on the keyword mapping table. Cut.

[0136] FIG. 12 shows an image coder for encoding and decoding in an intra-frame prediction mode according to an embodiment of the present invention. FIG. 1 is a conceptual diagram showing a part of an image encoding device and an image decoding device.

[0137] As shown in the upper part of FIG. 12, a part of the image encoding device includes a code number mapping unit 120. 0 and a codeword mapping unit 1220. The code number mapping unit 1200 receives the intra-frame prediction mode number and outputs the code number. The codeword mapping unit 1220 receives the code number and maps the codeword The code number mapping and code word according to the above embodiment can be output. A mode mapping may be performed, and the more frequently the intra-frame prediction mode number occurs, the smaller the number. As mentioned above, the code number mapping is not possible except for the MPM. When using the mapping method, the code number mapping section is Performing code number mapping for the remaining intra-frame prediction modes excluding the intra-frame prediction mode can be done.

[0138] As shown in the lower part of FIG. 12, the image decoding apparatus includes a codeword mapping unit 1240 and The codeword mapping unit 1260 may be provided. 240 can calculate a code number based on the input code word, The number mapping unit 1260 receives the code number and calculates the intra-frame prediction mode number. The code word mapping unit 1240 has a A codeword mapping table for the code number mapping unit 126 may be included. 0 is a code for calculating the intra-frame prediction mode number based on the received code number. A codeword mapping table and a code number mapping table may be included. The code mapping portion can be pre-stored or adaptively generated.

[0139] Although the above description has been given with reference to the embodiments, those skilled in the art will recognize that the following patents Various modifications of the present invention may be made without departing from the spirit and scope of the present invention as defined in the appended claims. It will be understood that this may be adjusted and varied.

Claims

1. A decoding method for an intra prediction mode, comprising: Decoding flag information indicating whether one of the candidate intra prediction modes for the current block is the same as the intra prediction mode of the current block; If one of the candidate intra prediction modes for the current block is identical to the intra prediction mode of the current block, decoding a syntax element of information regarding the candidate intra prediction mode for the current block; If the candidate intra prediction mode for the current block and the intra prediction mode of the current block are not identical, decoding syntax elements of remaining intra prediction modes of the current block, the syntax elements including information about the intra prediction mode of the current block; generating a predicted block of the current block based on the intra prediction mode of the current block; generating a reconstruction block of the current block based on the predicted block; The intra prediction mode of the current block is mapped to index information of the intra prediction mode based on a table; When the intra-frame prediction mode is a planar mode, the intra-frame prediction mode is mapped to index 0; when the intra-frame prediction mode is a DC mode, the intra-frame prediction mode is mapped to index 1; when the intra-frame prediction mode is a directional intra-frame prediction mode, the intra-frame prediction mode is mapped to indexes 2 to 34 according to the directionality of the intra-frame prediction mode; a codeword length of the index of the candidate intra prediction mode for the current block is shorter than a codeword length of the index of the remaining intra prediction mode information; the candidate intra-prediction mode for the current block is a mode for predicting luminance information of the current block, the chrominance information of the current block is predicted based on intra-frame prediction mode information corresponding to the luminance information of the current block; A decoding method in which the syntax elements of the remaining intra prediction mode information are values ​​coded using a fixed 5 bits.

2. An intra-frame prediction mode encoding method, determining and encoding flag information indicating whether one of the candidate intra prediction modes for the current block is the same as the intra prediction mode of the current block; If one of the candidate intra prediction modes for the current block is the same as the intra prediction mode of the current block, encoding a syntax element of information about the candidate intra prediction mode for the current block; If the candidate intra prediction mode for the current block is not identical to the intra prediction mode of the current block, encoding syntax elements of remaining intra prediction modes of the current block, the syntax elements including information about the intra prediction mode of the current block; generating a predicted block of the current block based on the intra prediction mode of the current block; generating a reconstruction block of the current block based on the predicted block; The intra prediction mode of the current block is mapped to index information of the intra prediction mode based on a table; When the intra-frame prediction mode is a planar mode, the intra-frame prediction mode is mapped to index 0; when the intra-frame prediction mode is a DC mode, the intra-frame prediction mode is mapped to index 1; when the intra-frame prediction mode is a directional intra-frame prediction mode, the intra-frame prediction mode is mapped to indexes 2 to 34 according to the directionality of the intra-frame prediction mode; a codeword length of the index of the candidate intra prediction mode for the current block is shorter than a codeword length of the index of the remaining intra prediction mode information; the candidate intra-prediction mode for the current block is a mode for predicting luminance information of the current block, the chrominance information of the current block is predicted based on intra-frame prediction mode information corresponding to the luminance information of the current block; The encoding method, wherein the syntax elements of the remaining intra prediction mode information are values ​​encoded using a fixed 5 bits.

Images