WO2020175965A1 - 인트라 예측 기반 비디오 신호 처리 방법 및 장치 - Google Patents
인트라 예측 기반 비디오 신호 처리 방법 및 장치 Download PDFInfo
- Publication number
- WO2020175965A1 WO2020175965A1 PCT/KR2020/002920 KR2020002920W WO2020175965A1 WO 2020175965 A1 WO2020175965 A1 WO 2020175965A1 KR 2020002920 W KR2020002920 W KR 2020002920W WO 2020175965 A1 WO2020175965 A1 WO 2020175965A1
- Authority
- WO
- WIPO (PCT)
- Prior art keywords
- block
- transform
- current block
- intra prediction
- prediction
- Prior art date
Links
Classifications
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04N—PICTORIAL COMMUNICATION, e.g. TELEVISION
- H04N19/00—Methods or arrangements for coding, decoding, compressing or decompressing digital video signals
- H04N19/10—Methods or arrangements for coding, decoding, compressing or decompressing digital video signals using adaptive coding
- H04N19/102—Methods or arrangements for coding, decoding, compressing or decompressing digital video signals using adaptive coding characterised by the element, parameter or selection affected or controlled by the adaptive coding
- H04N19/103—Selection of coding mode or of prediction mode
- H04N19/105—Selection of the reference unit for prediction within a chosen coding or prediction mode, e.g. adaptive choice of position and number of pixels used for prediction
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04N—PICTORIAL COMMUNICATION, e.g. TELEVISION
- H04N19/00—Methods or arrangements for coding, decoding, compressing or decompressing digital video signals
- H04N19/10—Methods or arrangements for coding, decoding, compressing or decompressing digital video signals using adaptive coding
- H04N19/134—Methods or arrangements for coding, decoding, compressing or decompressing digital video signals using adaptive coding characterised by the element, parameter or criterion affecting or controlling the adaptive coding
- H04N19/157—Assigned coding mode, i.e. the coding mode being predefined or preselected to be further used for selection of another element or parameter
- H04N19/159—Prediction type, e.g. intra-frame, inter-frame or bidirectional frame prediction
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04N—PICTORIAL COMMUNICATION, e.g. TELEVISION
- H04N19/00—Methods or arrangements for coding, decoding, compressing or decompressing digital video signals
- H04N19/10—Methods or arrangements for coding, decoding, compressing or decompressing digital video signals using adaptive coding
- H04N19/102—Methods or arrangements for coding, decoding, compressing or decompressing digital video signals using adaptive coding characterised by the element, parameter or selection affected or controlled by the adaptive coding
- H04N19/117—Filters, e.g. for pre-processing or post-processing
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04N—PICTORIAL COMMUNICATION, e.g. TELEVISION
- H04N19/00—Methods or arrangements for coding, decoding, compressing or decompressing digital video signals
- H04N19/10—Methods or arrangements for coding, decoding, compressing or decompressing digital video signals using adaptive coding
- H04N19/102—Methods or arrangements for coding, decoding, compressing or decompressing digital video signals using adaptive coding characterised by the element, parameter or selection affected or controlled by the adaptive coding
- H04N19/119—Adaptive subdivision aspects, e.g. subdivision of a picture into rectangular or non-rectangular coding blocks
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04N—PICTORIAL COMMUNICATION, e.g. TELEVISION
- H04N19/00—Methods or arrangements for coding, decoding, compressing or decompressing digital video signals
- H04N19/10—Methods or arrangements for coding, decoding, compressing or decompressing digital video signals using adaptive coding
- H04N19/169—Methods or arrangements for coding, decoding, compressing or decompressing digital video signals using adaptive coding characterised by the coding unit, i.e. the structural portion or semantic portion of the video signal being the object or the subject of the adaptive coding
- H04N19/17—Methods or arrangements for coding, decoding, compressing or decompressing digital video signals using adaptive coding characterised by the coding unit, i.e. the structural portion or semantic portion of the video signal being the object or the subject of the adaptive coding the unit being an image region, e.g. an object
- H04N19/176—Methods or arrangements for coding, decoding, compressing or decompressing digital video signals using adaptive coding characterised by the coding unit, i.e. the structural portion or semantic portion of the video signal being the object or the subject of the adaptive coding the unit being an image region, e.g. an object the region being a block, e.g. a macroblock
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04N—PICTORIAL COMMUNICATION, e.g. TELEVISION
- H04N19/00—Methods or arrangements for coding, decoding, compressing or decompressing digital video signals
- H04N19/50—Methods or arrangements for coding, decoding, compressing or decompressing digital video signals using predictive coding
- H04N19/593—Methods or arrangements for coding, decoding, compressing or decompressing digital video signals using predictive coding involving spatial prediction techniques
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04N—PICTORIAL COMMUNICATION, e.g. TELEVISION
- H04N19/00—Methods or arrangements for coding, decoding, compressing or decompressing digital video signals
- H04N19/60—Methods or arrangements for coding, decoding, compressing or decompressing digital video signals using transform coding
- H04N19/61—Methods or arrangements for coding, decoding, compressing or decompressing digital video signals using transform coding in combination with predictive coding
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04N—PICTORIAL COMMUNICATION, e.g. TELEVISION
- H04N19/00—Methods or arrangements for coding, decoding, compressing or decompressing digital video signals
- H04N19/70—Methods or arrangements for coding, decoding, compressing or decompressing digital video signals characterised by syntax aspects related to video coding, e.g. related to compression standards
Definitions
- the present invention relates to a video signal processing method and apparatus, and more particularly, to a video signal processing method and apparatus for encoding or decoding a video signal based on intra prediction.
- Compression encoding refers to a series of signal processing technologies for transmitting digitized information through a communication line or storing it in a form suitable for a storage medium.
- Objects such as audio, video, and text are present in the object of compression encoding.
- the technique of performing compression encoding for an image is called video image compression.
- Compression encoding for video signals is performed by removing excess information in consideration of spatial correlation, temporal correlation, and probability correlation.
- more and more efficient video signal processing methods and devices are required.
- the purpose of the present invention is to increase the coding efficiency of video signals.
- the present invention is a unit in which Intra sub-partitions are applied, i.e., PDPC (position- Dependent intra prediction combination) and LFNST (Low-Frequency Non-Separable Transform) are proposed.
- PDPC position- Dependent intra prediction combination
- LFNST Low-Frequency Non-Separable Transform
- the present invention provides the following video signal processing apparatus and video signal processing method.
- ISP mode is applied to the current block. If applicable, dividing the current block into a plurality of horizontal or vertical rectangular transform blocks; generating prediction blocks of the transform blocks by performing intra prediction for each of the transform blocks; And Including the step of restoring the current block based on the residual block of the block and the prediction block, wherein the step of generating the prediction block, in units of transform blocks divided from the current block
- It may include performing position-dependent intra prediction sample filtering, i.e., position-dependent intra prediction sample filtering.
- the position-dependent intra prediction sample filtering is applied based on at least one of the width and height of the transform block.
- the determining step when the width of the transform block is greater than or equal to a preset reference value, and, when the height of the transform block is greater than or equal to the preset reference value, it is determined to apply the position-dependent intra prediction sampling filtering. Can be done.
- the residual block of the transform block is a second difference in units of the transform block.
- the step of determining whether a quadratic transformation is applied to the current block When the quadratic transformation is applied to the current block, inducing a quadratic transformation kernel set to be applied to the current block from among pre-defined quadratic transformation kernel sets based on the intra prediction mode of the current block; Determining a quadratic transform kernel to be applied to the current block in the kernel set; Generating a second-order inverse transformed block of the transformed block by performing a second-order inverse transform in units of the transform block; and generating a residual block of the transform block by performing a first-order inverse transform on the second-order inverse transformed block.
- a video signal processing apparatus including a processor, the processor, an intra subpartition (ISP, Intra)
- ISP intra subpartition
- Sub-Partitions) mode is applied or not, and when ISP mode is applied to the current block, the current block is divided into a plurality of horizontal or vertical rectangular transform blocks, and for each of the transform blocks
- the processor comprises: in units of transform blocks divided from the current block.
- a video signal processing apparatus which is characterized by performing position-dependent intra prediction sample filtering, i.e., position-dependent intra prediction sample filtering.
- the processor may determine whether to apply the position-dependent intra prediction sample filtering based on at least one of the width and height of the transform block.
- the processor when the width of the transform block is greater than or equal to a preset reference value, and, and the height of the transform block is greater than or equal to the preset reference value, the position-dependent intra prediction sampling filtering It can be decided by applying
- the residual block of the transform block is a second difference in units of the transform block.
- the processor determines whether or not the quadratic transformation is applied to the current block
- a quadratic transform kernel applied to the current block is determined, and a quadratic inverse transform of the transform block is generated by performing a quadratic inverse transform in units of the transform block, and with respect to the secondary inverse transformed block
- a quadratic inverse transform of the transform block may be generated.
- ISP Intra Sub-Partitions
- the current block Dividing n into a plurality of horizontal or vertical rectangular transform blocks; generating prediction blocks of the transform blocks by performing intra prediction for each of the transform blocks; and subtracting the prediction blocks from the original block Including the step of generating a residual block of the transform block, wherein the step of generating the prediction block, Position-dependent intra prediction sample filtering in units of transform blocks divided from the current block (position-dependent intra prediction sample filtering)
- a video signal processing method comprising the step of performing:
- a computer-executable component configured to run on one or more processors of a computing device is stored.
- the computer-executable component determines whether or not the Intra Sub-Partitions (ISP) mode is applied to the current block, and if the ISP mode is applied to the current block, the current block is multiplexed. Dividing into rectangular transform blocks in the horizontal or vertical direction of, and performing intra prediction for each of the transform blocks to generate prediction blocks of the transform blocks, and based on the residual block of the transform block and the prediction block, the The current block is restored, but the computer-executable component is a non-transitory, characterized by performing position-dependent intra prediction sample filtering in units of transform blocks divided from the current block.
- a computer-readable medium is provided.
- the coding efficiency of a video signal can be improved.
- the PDPC is divided into a transform block unit divided by intra sub-partitions. (Po sition-dependent intra prediction combination) and LFNST can improve the accuracy of prediction and improve compression performance.
- FIG. 1 is a schematic diagram of a video signal encoding apparatus according to an embodiment of the present invention
- FIG. 1 is a schematic diagram of a video signal decoding apparatus according to an embodiment of the present invention
- FIG. 3 shows an embodiment in which a coding tree unit is divided into coding units in a picture.
- FIG. 4 is a diagram of a method for signaling division of a quad tree and a multi-type tree
- FIG 5 and 6 more specifically illustrate an intra prediction method according to an embodiment of the present invention.
- FIG 7 shows an inter prediction method according to an embodiment of the present invention.
- FIG. 8 is a detailed illustration of a method for converting a residual signal by an encoder.
- FIG. 9 is a diagram specifically showing a method of obtaining a residual signal by inversely transforming a conversion coefficient by an encoder and a decoder.
- FIG. W is a diagram for explaining an application method of an intra prediction mode when a coding block is divided into a plurality of transform blocks according to an embodiment of the present invention.
- FIG. 11 is a diagram for explaining a method of applying a position-dependent intra prediction combination (PDPC) according to an embodiment of the present invention.
- PDPC position-dependent intra prediction combination
- FIG. 12 is a diagram illustrating a reference sample used for a PDPC according to an intra prediction mode as an embodiment of the present invention.
- FIG. 13 is a diagram illustrating a reference sample used for a PDPC according to an intra prediction mode according to an embodiment of the present invention.
- FIG. 14 is a diagram illustrating a method of applying Intra subpartitions (ISP) and position-dependent intra prediction combination (PDPC) to a coding block according to an embodiment to which the present invention is applied.
- ISP Intra subpartitions
- PDPC position-dependent intra prediction combination
- 15 is a diagram for explaining a conversion unit division processing method according to an embodiment of the present invention.
- 16 is a diagram showing a process of encoding/decoding through a primary transform and a secondary transform according to an embodiment to which the present invention is applied.
- FIG. 17 shows a conversion kernel used for a second-order conversion according to an embodiment of the present invention.
- FIG. 18 illustrates a method of applying a second-order transform in units of transform blocks according to an embodiment of the present invention.
- 19 is a diagram showing a method of applying a PDPC to a current coding block to which an intra prediction mode is applied according to an embodiment to which the present invention is applied.
- 20 is a diagram showing a video signal processing method according to an embodiment of the present invention 2020/175965 1»(:1/10 ⁇ 020/002920
- Coding can be interpreted as encoding or decoding in some cases.
- a device that performs encoding (encoding) of a video signal to generate a video signal bitstream Is referred to as an encoding device or an encoder, and a device that restores a video signal by performing decoding (decoding) of a video signal bitstream is referred to as a decoding device or a decoder.
- the video signal processing device is used as a term for a concept that includes both an encoder and a decoder.
- Information is values
- 'Unit' is a basic unit or picture of image processing. It is used as a meaning to refer to a specific position of, and refers to an image area containing at least one of a luma component and a chroma component.
- 'block' refers to the luma component and chroma components (i.e., Cb and Cr) refers to an image area containing a specific component.
- Cb and Cr chroma components
- a unit is a coding unit. It can be used as a concept that includes all of a unit, a prediction unit, and a conversion unit.
- a picture indicates a field or a frame, and the terms may be used interchangeably according to an embodiment.
- an encoding apparatus 100 of the present invention includes a conversion unit 110 and a quantization unit. 115, an inverse quantization unit 120, an inverse transform unit 125, a filtering unit 130, a prediction unit 150, and an entropy coding unit 160.
- the conversion unit (1W) converts the residual signal, which is the difference between the input video signal and the prediction signal generated by the prediction unit 150, to obtain a conversion coefficient value.
- Discrete Cosine Transformation Discrete Cosine Transform, DCT
- DST Discrete Sine Transform
- Wavelet Transform can be used.
- the input picture signal is divided into blocks to perform transform.
- the coding efficiency may vary according to the distribution and characteristics of values in the transform region.
- the quantization unit 115 transforms Output from negative (H0) 2020/175965 1»(:1 ⁇ 1 ⁇ 2020/002920
- a method of predicting a picture using a region already coded through the prediction unit 150, and obtaining a restored picture by adding a residual value between the original picture and the predicted picture to the predicted picture is used. Mismatch in the encoder and the decoder In order not to occur, the information available in the decoder must be used when performing prediction in the encoder. To this end, the encoder performs a process of restoring the encoded current block again.
- the conversion coefficient value Inverse quantization Inverse quantization
- the inverse transform unit 125 restores the residual value by using the inverse quantized transform coefficient value.
- the filtering unit 130 performs a filtering operation to improve quality and encoding efficiency of the restored picture. For example, it may include a deblocking filter, a sample adaptive offset (SA0), and an adaptive loop filter.
- SA0 sample adaptive offset
- SA0 sample adaptive offset
- the filtered picture is output or stored in a decoded picture buffer (DPB, 156) to be used as a reference picture.
- the inter prediction unit 154 may again include a motion estimation unit 154a and a motion compensation unit 154b.
- the motion vector value of the current area is obtained by referring to the restored specific area.
- the position information (reference frame, motion vector, etc.) of the reference area is transferred to the entropy coding unit ( 160) to be included in the bitstream.
- the motion compensation unit 154b performs inter-screen motion compensation using the motion vector value transmitted from the motion estimation unit 154a.
- the prediction unit 150 includes an intra prediction unit 152 and an inter prediction unit 154. Intra
- the prediction unit 152 performs intra prediction in the current picture
- the prediction unit 154 performs inter prediction for predicting the current picture using the reference picture stored in the decoded picture buffer 156.
- the intra prediction unit 152 performs intra prediction from reconstructed samples in the current picture.
- the intra-encoding information is transmitted to the entropy coding unit 160.
- the intra-encoding information may include at least one of an intra prediction mode, a Most Probable Mode (MPM) flag, and an MPM index.
- Intra-encoding information is a reference sample
- the inter prediction unit 154 may include a motion estimation unit 154a and a motion compensation unit 154b.
- the motion estimation unit 154a determines a specific region of the restored reference picture.
- the motion vector value of the current region is obtained by reference.
- the motion estimation unit 154a is a set of motion information for the reference region (reference 2020/175965 1»(:1 ⁇ 1 ⁇ 2020/002920
- the picture index, motion vector information, etc.) are transmitted to the entropy coding unit 160.
- the motion compensation unit 154b performs motion compensation by using the motion vector value transmitted from the motion estimation unit 154a.
- Inter prediction The unit 154 transmits inter-coding information including motion information for the reference region to the entropy coding unit 160.
- the prediction unit 150 may include an intra block copy (BC) prediction unit (not shown).
- the intra BC prediction unit predicts intra BC from restored samples in the current picture. And transmits the intra BC encoding information to the entropy coding unit 160.
- the intra BC prediction unit refers to a specific region in the current picture and obtains a block vector value indicating a reference region used for prediction of the current region.
- the BC prediction unit can perform intra BC prediction using the acquired block vector value.
- the intra BC prediction unit transfers the intra BC encoding information to the entropy coding unit 160.
- the intra BC encoding information may include block vector information. .
- the conversion unit 110 converts the residual value between the original picture and the predicted picture to obtain a conversion coefficient value. At this time, the conversion can be performed in a specific block unit within the picture. In addition, the size of a specific block may vary within a preset range.
- the quantization unit 115 quantizes the value of the transformation coefficient generated by the transformation unit U0 and transmits the quantization to the entropy coding unit 160.
- the entropy coding unit 160 entropy-codes information indicating a quantized conversion factor, intra-coding information, and inter-coding information to generate a video signal bitstream.
- variable length coding is performed.
- Variable Length Coding (VLC) method and arithmetic coding method can be used.
- Variable Length Coding (VLC) method converts input symbols into consecutive codewords, and the length of the codeword can be variable. For example, symbols that occur frequently are expressed as short codewords and those that do not occur frequently are expressed as long codewords.
- Context-based Adaptive Variable Length Coding (CAVLC) is a variable length coding method. ) Method can be used.
- Arithmetic coding converts consecutive data symbols into a single decimal number, where arithmetic coding can obtain the optimal fractional bits needed to represent each symbol.
- Context-based adaptive arithmetic coding (as arithmetic coding)
- Context-based Adaptive Binary Arithmetic Code (CAB AC)
- CAB AC Context-based Adaptive Binary Arithmetic Code
- the entropy coding unit 160 can binarize information representing a quantized transformation coefficient.
- the entropy coding unit 160 can be binarized.
- the bitstream can be generated by arithmetic coding the information.
- the generated bitstream is encapsulated in a basic unit of a network abstraction layer (NAL) unit.
- NAL network abstraction layer
- the NAL unit contains an integer number of coded coding tree units.
- the bitstream must first be separated into NAL unit units, and then each separated NAL unit must be decoded.
- RBSP Raw Byte Sequence Payload
- PPS Picture Parameter Set
- SPS Sequence Parameter Set
- VPS Video Parameter Set
- FIG. 1 shows the encoding apparatus 100 according to an embodiment of the present invention, and the separately displayed blocks are shown by logically discriminating elements of the encoding apparatus 100. Therefore, the above description
- the elements of the Korean encoding device 100 may be mounted as one chip or as a plurality of chips depending on the design of the device. According to one embodiment, the operation of each element of the encoding device W0 described above is a processor (not shown). Can be done by
- FIG. 2 is a schematic block diagram of a video signal decoding apparatus 200 according to an embodiment of the present invention.
- the decoding apparatus 200 of the present invention is entropy.
- It includes a decoding unit (2W), an inverse quantization unit 220, an inverse transform unit 225, a filtering unit 230 and a prediction unit 250.
- the entropy decoding unit 210 entropy decodes the video signal bitstream to extract conversion coefficient information, intra coding information, inter coding information, etc. for each region. For example, the entropy decoding unit 210 Video signal
- a binary code for conversion coefficient information of a specific region can be obtained.
- the entropy decoding unit 2W inversely binarizes the binary code to obtain a quantized conversion factor.
- the inverse quantization unit 220 is a quantization unit.
- the transform coefficient is inverse quantized, and the inverse transform unit 225 restores the residual value by using the inverse quantized transform coefficient.
- the video signal processing apparatus 200 restores the original pixel value by summing the residual value obtained by the inverse transform unit 225 with the predicted value obtained by the prediction unit 250.
- the filtering unit 230 improves image quality by performing filtering on a picture.
- This may include a deblocking filter to reduce block distortion and/or an adaptive loop filter to remove distortion of the entire picture.
- the filtered picture is output or decoded to be used as a reference picture for the next picture. picture
- the prediction unit 250 includes an intra prediction unit 252 and an inter prediction unit 254.
- the prediction unit 250 generates a prediction picture by using the encoding type decoded through the entropy decoding unit 2W described above, the transformation coefficient for each region, and intra/inter encoding information.
- the current block on which decoding is performed is performed.
- the current picture including the current block or the decoded area of other pictures can be used.
- Intra picture or I picture (or tile/slice) a picture that can perform both intra prediction, inter prediction, and intra BC prediction (or,
- a tile/slice is called an inter-picture (or tile/slice).
- a maximum of one motion is used to predict the sample values of each block in the inter-picture (or tile/slice).
- a picture (or tile/slice) using a vector and reference picture index is called a predictive picture or a P picture (or tile/slice), and a picture using up to two motion vectors and reference picture indexes (or , Tile/slice) is called a bi-predictive picture (Bi-predictive picture) or B picture (or tile/slice)
- a P picture (or tile/slice) is at most one motion information to predict each block.
- a set is used, and the B picture (or tile/slice) uses up to two sets of motion information to predict each block, where the motion information set includes one or more motion vectors and one reference picture index.
- the intra prediction unit 252 generates a block, for example, by using the intra-encoding information and the restored samples in the current picture.
- the intra-encoding information is an intra prediction mode, a Most Probable Mode (MPM) flag, At least one of the MPM indexes may be included.
- the intra prediction unit 252 predicts the sample values of the current block by using the restored samples located on the left and/or above the current block as reference samples.
- the reconstructed samples, reference samples, and samples of the current block can represent pixels. Also, sample values can represent pixel values.
- the reference samples may be samples included in a block around the current block.
- the reference samples are samples adjacent to the left boundary of the current block and/or samples adjacent to the upper boundary.
- the reference samples are samples located on a line within a preset distance from the left boundary of the current block among samples of the neighboring block of the current block and/or a line within a preset distance from the upper boundary of the current block.
- the block around the current block is the left (L) block, the upper (A) block, the lower left (BL) block, and the upper right (Above Right, AR) block.
- Blocks or Above Left (AL) blocks can contain at least one.
- the inter prediction unit 254 generates a prediction block using the reference picture and inter-coding information stored in the decoded picture buffer 256.
- the inter-coding information is a motion information set (reference picture) of the current block for the reference block. Index, motion vector information, etc.)
- Inter prediction may include L0 prediction, L1 prediction, and bi-prediction.
- L0 prediction refers to prediction using one reference picture included in the L0 picture list
- time prediction refers to prediction using one reference picture included in the L1 picture list.
- one set of motion information e.g., motion vector and reference picture index
- up to two reference areas can be used, but these two reference areas may exist in the same reference picture.
- up to two sets of motion information e.g., a motion vector and a reference picture index
- the two motion vectors are in the same reference picture index. They may be mapped to different reference picture indices.
- the reference pictures may be temporally set before or after the current picture.
- the two reference areas used in the pair prediction method may be areas selected from each of the L0 picture list and the L1 picture list.
- the inter prediction unit 254 can obtain the reference block of the current block by using the motion vector and the reference picture index.
- the reference block exists in the reference picture corresponding to the reference picture index.
- the sample value of the block specified by or its interpolated value can be used as a predictor of the current block.
- an 8-tap interpolation filter can be used for the luma signal
- a 4-tap interpolation filter can be used for the chroma signal.
- the interpolation filter for motion prediction is not limited thereto.
- the inter prediction unit 254 performs motion compensation, i.e., the texture of the current unit, from the previously restored picture. In this case, the inter prediction unit motion information Sets are available.
- the prediction unit 250 may include an intra BC prediction unit (not shown).
- the intra BC prediction unit refers to a specific area including restored samples in the current picture and refers to the current area.
- the intra BC prediction unit acquires intra BC encoding information for the current region from the entropy decoding unit 2W.
- the intra BC prediction unit obtains a block vector value of the current area indicating a specific area in the current picture.
- the intra BC prediction unit can perform intra BC prediction using the obtained block vector value.
- the intra BC encoding information is block vector information. May contain
- the residual value output from the inverse transform unit 225 is added to generate a restored video picture. That is, the video signal decoding apparatus 200 is obtained from the prediction block generated by the prediction unit 250 and the inverse transform unit 225. The current block is restored using the residual.
- FIG. 2 shows the decoding apparatus 200 according to an embodiment of the present invention, and the separately displayed blocks are shown by logically discriminating elements of the decoding apparatus 200. Therefore, the above description
- the elements of the decoding device 200 may be mounted as a single chip or a plurality of chips depending on the design of the device. According to an embodiment, the operation of each element of the decoding device 200 described above is a processor (not shown). Can be done by
- Fig. 3 is a Coding Tree Unit (CTU) encoding in the picture
- a picture can be divided into a sequence of coding tree units (CTUs).
- a coding tree unit is an NXN of luma samples. It is composed of a block and two blocks of corresponding chroma samples.
- the coding tree unit can be divided into a plurality of coding units.
- the coding tree unit may be an undivided cyclic node. 2020/175965 1»(:1 ⁇ 1 ⁇ 2020/002920
- the coding tree unit itself may be a coding unit.
- the coding unit is the basic for processing pictures in the process of processing video signals described above, i.e., intra/inter prediction, transformation, quantization and/or entropy coding. Points to the unit. In one picture, the size and shape of the coding unit may not be constant.
- the coding unit may have a square or rectangular shape.
- a rectangular coding unit (or, a rectangular block) is horizontal with a vertical coding unit (or, a vertical block). Includes a coding unit (or horizontal block).
- a vertical block is a block whose height is greater than the width
- a horizontal block is a block whose width is greater than the height.
- non-square (non-square) ) Block may refer to a rectangular block, but the present invention is not limited thereto.
- the coding tree unit first has a quad tree (QT) structure.
- one node with a size of 2NX2N can be divided into four nodes with a size of NXN.
- a quadtree is
- quadtree partitioning can be done recursively, and not all nodes need to be partitioned to the same depth.
- the leaf node of the above-described quad tree can be further divided into a multi-type tree (MTT) structure.
- MTT multi-type tree
- one The nodes of can be divided into a binary (binary) or ternary (ternary) tree structure of horizontal or vertical division, i.e., a multi-type tree structure includes vertical binary division, horizontal binary division, and vertical ternary division.
- a multi-type tree structure includes vertical binary division, horizontal binary division, and vertical ternary division.
- the width and height of the nodes in each of the above tree structures may both have a power of two.
- a node with a size of 2NX2N can be divided into two NX2N nodes by vertical binary division, and divided into two 2NXN nodes by horizontal binary division.
- a node of 2NX2N size is divided into (N/2)X2N, NX2N, and (N/2)X2N nodes by vertical ternary division, and horizontal ternary division It can be divided into nodes of 2NX(N/2), 2NXN and 2NX(N/2) by.
- This multi-type tree division can be performed recursively.
- a leaf node of a multi-type tree can be a coding unit.
- the coding unit is used as a unit of prediction and conversion without further division.
- the following parameters in the quadtree and multi-type tree described above At least one of them may be predefined or transmitted through RBSP of a higher level set such as PPS, SPS, VPS, etc.
- Minimum allowable depth of segmentation 6) Minimum BT size (MinBtSize): Minimum allowed BT leaf node size, 7) Minimum TT size (MinTtSize): Minimum allowed TT leaf node size.
- FIG. 4 is a diagram of a method for signaling division of a quad tree and a multi-type tree
- Pre-set flags may be used to signal the division of the quad tree and the multi-type tree described above.
- a flag'qt_split_flag' indicating whether or not the quad tree node is divided
- multi -type tree indicating the division whether the node flag 'mtt_split_flag', a multi-even always lag
- mtt_split_binary_flag popularly indicative of a division form of the type tree node always indicative of a dividing direction of the types of tree nodes lag 'mtt_split_vertical_flag' or multi- One can be used.
- the coding tree unit is a root node of a quad tree, and can be pre-divided into a quad tree structure.
- each node'QT_node' is qt_split_flag, and is signaled., qt_split_flag,If the value of is 1, the corresponding node
- Each quad tree leaf node'QT_leaf_node' can be further divided into a multi-type tree structure.
- 'mtt_split_flag' is signaled. If the value of'mtt_split_flag' is 1, the node is divided into a plurality of rectangular nodes, and if the value of'mtt_split_flag' is 0, the node becomes'MTT_leaf_node' of the multi-type tree. Multi-type tree node'MTT_node' When divided into'multiple rectangular nodes' (that is, the value of'mtt_split_flag' is
- 'mtt_split_binary_flag' can be signaled at the stock price. If the value of'mtt_split_vertical_flag' is 1, the vertical division of the node'MTT_node' is indicated.
- Picture prediction (motion compensation) for coding is performed on a coding unit that is no longer divided (i.e., a leaf node of the coding unit tree).
- the basic unit that performs such prediction is hereinafter referred to as a unit (prediction unit) or That is, it is called a prediction block.
- the unit may be used as a term to replace the prediction unit.
- the present invention is not limited thereto, and more broadly, it can be understood as a concept including the coding unit.
- the intra prediction unit refers to the restored samples located at the left and/or upper sides of the current block.
- the current block 's sample values 2020/175965 1»(:1 ⁇ 1 ⁇ 2020/002920
- FIG. 5 shows an embodiment of reference samples used for prediction of the current block in the intra prediction mode.
- the reference samples are samples adjacent to the left boundary of the current block and/or the upper side.
- the reference samples can be set using up to 2 +211 +1 ambient samples located on the top.
- the intra prediction unit can obtain a reference sample by performing a reference sample padding process.
- the intra prediction unit may perform a reference sample filtering process to reduce errors in intra prediction, i.e., filter the surrounding samples and/or reference samples obtained by the reference sample padding process, and filter the filtered reference samples.
- the intra prediction unit predicts the samples of the current block by using the thus obtained reference samples.
- the intra prediction unit predicts the samples of the current block by using the unfiltered reference samples or the filtered reference samples.
- the surrounding samples may contain samples on at least one reference line; for example, the surrounding samples may contain adjacent samples on a line adjacent to the boundary of the current block.
- FIG. 6 shows an embodiment of prediction modes used for intra prediction.
- intra prediction mode information indicating an intra prediction direction may be signaled.
- the intra prediction mode information indicates any one of a plurality of intra prediction modes constituting an intra prediction mode set.
- Current block In the case of this intra prediction block, the decoder receives intra prediction mode information of the current block from the bitstream.
- the intra prediction unit of the decoder performs intra prediction on the current block based on the extracted intra prediction mode information.
- the intra prediction mode set may include all intra prediction modes (eg, a total of 67 intra prediction modes) used for intra prediction. More specifically, the intra prediction mode set includes planar mode, IX: mode, and multiple (e.g.,
- Each intra prediction mode can be indicated through a preset index (ie, intra prediction mode index).
- the intra prediction mode index 0 indicates the planar mode
- the intra prediction mode index 1 indicates the 1: mode.
- the intra prediction mode indexes 2 to 66 indicate different angle modes.
- Each of the angle modes indicates different angles within a preset angle range.
- the angle mode is an angle range between 45 degrees and -135 degrees clockwise (that is, the first angle range). The angle within the range can be indicated.
- the angle mode can be defined based on the 12 o'clock position.
- the intra prediction mode index 2 is the horizontal diagonal ⁇ 03 01 1 2020/175965 1»(:1 ⁇ 1 ⁇ 2020/002920
- intra prediction mode index 34 indicates diagonal (DIA) mode
- intra prediction mode index 50 is
- the vertical (VER) mode is indicated, and the intra prediction mode index 66 indicates the vertical diagonal (VDIA) mode.
- the inter prediction method may include a general inter prediction method optimized for translation motion and an inter prediction method based on an affine model.
- the motion vector is a general inter prediction method. Depending on the method, it may contain at least one of a general motion vector for motion compensation and a control point motion vector for affine motion compensation.
- Fig. 7 shows an inter prediction method according to an embodiment of the present invention.
- the decoder can predict the current block by referring to restored samples of another decoded picture.
- the decoder obtains the reference block 702 in the reference picture 720 based on the motion information set of the current block 701.
- the motion information set may include a reference picture index and a motion vector.
- the reference picture index is In the reference picture list, a reference picture 720 containing a reference block for inter prediction of the current block is indicated.
- the reference picture list may include at least one of the L0 picture list or the L1 picture list described above.
- the motion vector represents the offset between the coordinate value of the current block 701 in the current picture (base 0) and the coordinate value of the reference block 702 in the reference picture 720.
- the decoder is in the reference block 702.
- a predictor of the current block 701 is obtained based on the sample values, and the current block 701 is restored using the predictor.
- the encoder has a restoration sequence similar to the current block in the previous pictures.
- the encoder can search for a reference block in which the sum of the difference between the current block and the sample value is the minimum within a preset search area, at this time, the current block and the reference block.
- At least one of SAD (Sum Of Absolute Difference) or SATD (Sum of Hadamard Transformed Difference) can be used to determine the similarity between the samples in the block, where SAD is the difference between the sample values contained in the two blocks, respectively. It can be the sum of all the absolute values.
- SATD is the difference between the sample values included in the two blocks.
- the current block can also be predicted using more than one reference area. As described above, the current block has two or more reference areas. Inter prediction can be made through a pair prediction method using a pair prediction method.
- the decoder can acquire two reference blocks based on two sets of motion information of the current block. In addition, the decoder can obtain two reference blocks. Based on the respective sample values, it is possible to obtain the first predictor and the second predictor of the current block. In addition, the decoder uses the first predictor and the second predictor. 2020/175965 1»(:1 ⁇ 1 ⁇ 2020/002920
- the decoder can restore the current block based on the sample-by-sample average of the first predictor and the second predictor.
- a set can be signaled.
- the similarity between a set of motion information for motion compensation for each of a plurality of blocks can be used.
- the set of motion information used for prediction of the current block is one of the other samples that have been restored. It can be derived from a set of motion information used for one prediction.
- the encoder and decoder can reduce signaling overhead.
- the decoder can generate a merge candidate list based on the plurality of candidate blocks.
- the merge candidate list is the motion information related to the motion information set of the current block among samples restored earlier than the current block. Can include candidates corresponding to samples that are likely to have been predicted based on the set.
- Encoder and decoder can construct a merge candidate list of the current block according to predefined rules.
- the merge candidate list each configured by encoder and decoder can be identical to each other. For example, encoder and decoder are within the current picture.
- the merge candidate list of the current block can be constructed based on the location of the current block.
- the location of a specific block Represents the relative position of the top-left sample of the specific block within the picture containing the specific block.
- the above-described residual signal is not coded as it is, but a method of quantizing the conversion coefficient value obtained by converting the residual signal and coding the quantized conversion coefficient can be used.
- the conversion unit can convert the residual signal to obtain a conversion coefficient value.
- the residual signal of a specific block may be distributed over the entire area of the current block. Accordingly, the frequency of the residual signal is obtained. Coding efficiency can be improved by concentrating energy in a low-frequency region through region transformation.
- a method in which the residual signal is transformed or inversely transformed will be described in detail.
- FIG. 8 is a detailed illustration of a method for converting a residual signal by an encoder.
- the residual signal in the spatial domain can be converted to the frequency domain.
- the encoder can convert the acquired residual signal to obtain the conversion coefficient.
- the encoder is the residual for the current block. At least one residual block containing a signal can be obtained.
- the residual block can be either the current block or blocks divided from the current block.
- the residual block is the residual samples of the current block. It may be referred to as a containing residual array or a residual matrix.
- a residual block may represent a transform unit or a block of the same size as the size of the transform block. 2020/175965 1»(:1 ⁇ 1 ⁇ 2020/002920
- the encoder can convert the residual block using a conversion kernel.
- the transformation kernel used for transformation for a residual block may be a transformation kernel with separable characteristics of a vertical transformation and a horizontal transformation.
- the transformation for a residual block can be performed separately into a vertical transformation and a horizontal transformation.
- the encoder can perform vertical conversion by applying the conversion kernel in the vertical direction of the residual block.
- the encoder can perform horizontal conversion by applying the conversion kernel in the horizontal direction of the residual block.
- the conversion kernel can be used as a term to designate a set of parameters used in the conversion of a residual signal, such as conversion matrix, conversion array, conversion function, conversion.
- the conversion kernel is a plurality of available kernels. It can be any one of the above.
- a transformation kernel based on a different transformation type may be used. See Figures 12 to 26 for how one of the multiple available transformation kernels is selected. It will be described later through.
- the encoder can quantize by passing the transform block transformed from the residual block to the quantization unit.
- the transform block may contain a plurality of transform coefficients.
- the transform block may be composed of a plurality of transform coefficients arranged in a two-dimensional arrangement.
- the size of the transform block is the current block or the current block, like the residual block.
- the transform coefficients passed to the quantization section can be expressed as quantized values.
- the encoder can perform additional transformations before the transformation factor is quantized. As shown in Figure 8, the above-described conversion method
- the second-order transformation may be selective for each residual block.
- the encoder may improve the coding efficiency by performing a second-order transformation for a region where it is difficult to concentrate energy in the low-frequency region only by the first-order transformation.
- a quadratic transformation may be added to a block in which residual values appear largely in a direction other than the horizontal or vertical direction of the residual block.
- the residual values of the intra-predicted block are to the residual values of the inter-predicted block.
- the probability of changing in a direction other than the horizontal or vertical direction may be high.
- the encoder can additionally perform a quadratic transformation on the residual signal of the intra-predicted block.
- the encoder can additionally perform the quadratic transformation of the inter-predicted block.
- the second order conversion can be omitted for the residual signal.
- a transformation kernel having different sizes depending on the size of the current block or residual block may be used.
- 8X8 quadratic transformation may be applied to blocks whose shorter side of the width or height is greater than or equal to the first preset length.
- the length of the shorter side of the width or height may be less than the second preset length.
- the first preset length may be a value larger than the second preset length, but this initiation is not limited to this.
- the second transform is different from the first transform, the vertical transform and the horizontal transform. It may not be performed separately as a transformation.
- This second-order transform may be referred to as a low frequency non-separable transform (LFNST).
- LNNST low frequency non-separable transform
- the high frequency band energy may not decrease even if frequency conversion is performed due to a sudden change in brightness. As a result, the compression performance due to quantization may be degraded. In addition, the residual value may be reduced. If conversion is performed for an existing region, the encoding time and the decoding time may unnecessarily increase. Accordingly, the conversion for the residual signal in a specific region may be omitted.
- Whether or not to perform transformation on the residual signal of a specific region may be determined by a syntax element related to transformation of a specific region.
- the syntax element may include transform skip information. May be a transform skip flag.
- the encoder When the transform skip information for a residual block indicates a transform skip, no transformation is performed for the corresponding residual block. In this case, the encoder does not perform transformation of the corresponding region. The remaining residual signal can be directly quantized.
- the operations of the encoder described with reference to FIG. 8 can be performed through the converter of FIG.
- the above-described conversion-related syntax elements may be information parsed from the video signal bitstream.
- the decoder can obtain conversion-related syntax elements by entropy-decoding the video signal bitstream.
- the encoder can entropy the conversion-related syntax elements. It can be coded to create a video signal bitstream.
- FIG. 9 is a diagram specifically showing a method of obtaining a residual signal by inverse conversion of a conversion factor by an encoder and a decoder.
- the inverse conversion operation is performed through the inverse conversion units of the encoder and decoder.
- the inverse transform unit may inversely transform the inverse quantized transform coefficient to obtain a residual signal.
- the inverse transform unit may detect whether an inverse transform for the corresponding region is performed from the transform-related syntax element of a specific region.
- a specific transform If the transform-related syntax element for a block indicates a transform skip, the transform for the corresponding transform block may be omitted. In this case, both the first-order inverse transform and the second-order inverse transform described above for the transform block may be omitted.
- inverse quantization The converted conversion factor can be used as a residual signal; for example, the decoder can restore the current block by using the inverse quantized conversion factor as the residual signal.
- a transform-related syntax element for a specific transform block may not indicate a transform skip.
- the inverse transform unit may determine whether to perform a second-order inverse transform for the second transform. For example, When the transform block is a transform block of an intra-predicted block, a second-order inverse transform may be performed on the transform block. Also, based on the intra prediction mode corresponding to the transform block, a second-order transform used for the transform block is used. 2020/175965 1»(:1 ⁇ 1 ⁇ 2020/002920
- the kernel can be determined. As another example, whether to perform the second-order inverse transform may be determined based on the size of the transform block.
- the second-order inverse transform can be performed after the inverse quantization process and before the first-order inverse transform is performed.
- This inverse transform unit can perform a first-order inverse transform on the inverse quantized transform coefficient or the second inverse transformed transform coefficient.
- the first-order inverse transformation it can be performed separately as a vertical transformation and a horizontal transformation, like the first-order transformation.
- the inverse transformation unit can obtain a residual block by performing a vertical inverse transformation and a horizontal inverse transformation on a transform block.
- the inverse transform unit can inverse transform the transform block based on the transform kernel used for transforming the transform block.
- the encoder explicitly specifies information indicating the transform kernel applied to the current transform block among a plurality of available transform kernels. It can be signaled implicitly.
- the decoder can select a transform kernel to be used for inverse transform of a transform block from among a plurality of available transform kernels using information representing the signaled transform kernel.
- the inverse transform section is obtained through inverse transform of the transform coefficient.
- the current block can be restored using the residual signal.
- FIG. W shows a coding block according to an embodiment of the present invention as a plurality of transform blocks.
- the intra prediction mode is a coding unit (or coding unit).
- Block (hereinafter, it may be abbreviated as a block) can be determined in units. And, the coding unit can be divided into a plurality of transform blocks. As an example, the intra prediction mode is modified (or modified) based on the shape of the coding block. Can be interpreted, decided, and improved).
- nTbW is the width of the transform block and nTbH is the height of the transform block.
- nTbW may be a variable indicating the width of the coding block
- nTbH may be a variable indicating the height of the coding block.
- whRatio is a variable representing the ratio of width and height. For example, whRatio is
- the intra prediction mode signaled from the encoder to the decoder is referred to as the first prediction mode (or the first intra prediction mode), and is modified (or The reinterpreted, determined, improved) mode can be referred to as a second prediction mode (or a second intra prediction mode).
- abs() represents an operator (or function) taking an absolute value.
- Modified intra intra prediction mode The prediction mode can be derived based on the following conditions.
- the decoder sets wideAngle to 1 when the above three conditions 1 to 3 are satisfied, 2020/175965 1»(:1 ⁇ 1 ⁇ 2020/002920
- the second prediction mode can be set to (1st prediction mode + 65).
- the second prediction mode can be set to (1st prediction mode-67).
- the intra prediction mode can be divided into a basic angle mode and an extended angle mode.
- the basic angle mode may be angle modes within the range of +-45 based on a vertical mode/horizontal mode
- the extended angle mode can be an angle mode that exceeds +-45 degrees based on the vertical mode/horizontal mode. Therefore, the signaled mode information can use the basic angle mode or the extended angle mode depending on the shape of the coding block.
- the number of available modes can be defined according to the horizontal to vertical ratio (or vertical to horizontal ratio) based on the shape of the coding block. As an example, the ratio is 2: 1, 4: 1 , 8: 1, 16: 1, etc. can be defined (or set).
- Blocks can have the form of a vertical rectangular block, where ⁇ year& is a variable representing the width of the coding block,
- the signaled intra prediction mode 2 can be reinterpreted based on the shape of the first transform block and modified to an extended angle mode. According to the above-described analysis method, it becomes (2+65), and the second prediction mode is It can be derived (or determined) by 67, i.e., the intra prediction mode determined in units of the coding block may not be used equally in units of the transform blocks. In this case, a change in performance may also occur.
- a method of applying the method of determining the wide-angle mode as follows in order to apply the intra prediction mode determined in the coding block equally to the transform block is proposed as follows.
- the encoder/decoder can set and to the peak of the coding block 53 ⁇ 43 ⁇ 4.
- the encoder/decoder determines whether to use the wide-angle mode by using the height and width of the coding block including the conversion block ( In the diagram, the case where the conversion block is divided into a horizontal rectangular shape is given as an example, but the present invention is not limited to this, i.e., the same even when divided into a vertical rectangle, a square or a combination of several shapes. Embodiments suggested below may be applied. 2020/175965 1»(:1/10 ⁇ 020/002920
- FIG. 11 is a diagram illustrating a method of applying a position-dependent intra prediction combination (PDPC) according to an embodiment of the present invention.
- PDPC can be applied to Intra Block if all of the following conditions are satisfied.
- IntraSubPartitionsSplitType (ISP split type) is ISP_NO_SPLIT or
- cldx (component index) is not equal to 0
- [111]-predModeIntra is INTRA_ANGULAR 18
- the PDPC operation can be applied according to the method described below.
- the PDPC described in the present invention is not limited to its name, and the always PDPC is a position-dependent intra, i.e., fine filtering It can also be referred to as (Position-dependent intra prediction sample filtering).
- the predicted sample pred at the (X, y) position (x, is the linearity of the reference sample according to the intra prediction mode (eg, DC, planner, directional mode) and PDPC as shown in Equation 1 below. It can be predicted using combinations.
- the intra prediction mode eg, DC, planner, directional mode
- PDPC e.g., PDPC
- the yoni and yo are the references located on the left and upper sides of the current sample ⁇ , respectively
- the weight can be referred to as the weight based on the following Equation 2) Can be calculated.
- the PDPC weight can only be calculated based on the summation and shift operation. 2020/175965 1»(:1/10 ⁇ 020/002920
- the pred(x, y) value can be calculated in a single step using Equation 1 described above.
- the additional boundary filtering is of a conventional image compression technology (eg, HEVC). It can include a DC mode boundary filter or an old filter in horizontal/vertical mode.
- FIG. 12 is a diagram illustrating a reference sample used for a PDPC according to an intra prediction mode according to an embodiment of the present invention.
- FIG. 12(a) is a case where the intra prediction mode is the prediction mode 2 12(b) assumes that the intra prediction mode is the prediction mode 66.
- FIG. 12 shows the case where the PDPC is applied to the top-right diagonal mode, reference sample R Figures R_1y and R.
- the predicted sample pred(x', y') represents the predicted sample located at (x', y') in the predicted block.
- the PDPC weight for the upper right diagonal mode is
- FIG. 13 is a diagram illustrating a reference sample used for a PDPC according to an intra prediction mode according to an embodiment of the present invention.
- FIG. 13(a) is a mode number (or mode) of an intra prediction mode. Index) is assumed to be any one of 3 to 10, and Fig. 13(b) assumes that the mode number of the intra prediction mode is any one of 58 to 65. Similar to Fig. 12, 13 is left and lower left and lower. Side diagonal
- Prediction sample P red(x', y') is (x', y') in the prediction block. )
- the PDPC weight for the lower left diagonal mode is
- the PDPC weight can be defined as in Equation 5 below.
- the PDPC weight can be defined as in Equation 6 below.
- the diagonal mode and the adjacent mode of the diagonal mode shown in FIG. May not require additional boundary filtering.
- the reference sample coordinates of the example shown in Fig. 13 can be derived based on a table defined for directional mode intra prediction. As described above, since the table can be defined in a diagonal line and its adjacent mode, in the present invention It has the advantage that an additional table is not required according to the PDPC implementation described. In addition, the multiplication operation may not be used when calculating the coordinates and. Also, in one embodiment, when a fractional sample coordinate is used, for a reference sample Linear interpolation can be performed.
- ISP Intelligent System for a coding block according to an embodiment to which the present invention is applied.
- the current coding block is a width (W) and
- FIG. 14(b) shows an example in which the current coding block is divided into 4 transform blocks in the vertical direction when the ISP mode is applied.
- FIG. 14(c) shows an example of applying the PDPC in units of each transform block divided in FIG. 14(b).
- the encoder/decoder is a transform block to which an ISP is applied in FIG. 14(b).
- the interpolation filter describes how to get the sample value from the reference sample. As an example,
- Encoder/decoder can use Cubic interpolation filter coefficient when the filter lag is 0, and Gaussian interpolation filter coefficient can be used when 1, Encoder/decoder uses the determined interpolation filter coefficient.
- the reference sample value can be determined and this value can be used as a predicted value.
- the encoder/decoder can set the filter flag to 1 for the block to which the transform block is a luma component and KP is applied.
- the encoder/decoder can be Decoder filter
- the encoder/decoder is a luma component, and for a transform block to which KP is applied, the filter flag value can be set (or determined) based on the width (W) and height (H) values of the block. In one embodiment, the encoder/decoder compares the number of samples in a block (W*H) with a predefined (or preset) specific reference value to flag a flag. 2020/175965 1»(:1 ⁇ 1 ⁇ 2020/002920
- Encoder/decoder can compare the block width and height with each reference value to set the filter flag value differently.
- the conditions for determining the filter flag value are ( ⁇ > reference value and> reference value), ( ⁇ > It can be defined as a reference value or> reference value).
- the inequality sign is not limited to greater than or equal to the reference value, but can also be defined as the same case, greater than or equal to, less than or equal to.
- the reference values applied to and may be different, and different inequality signs may be applied.
- the filter flag value may be set depending on whether it falls within a specific block size range.
- the encoder/decoder is ⁇ ! For the block to which 5 is applied, mod (: can be applied. As an example, the encoder/decoder is! For the block to which 5 is applied, Based on You can decide whether or not to apply. In one embodiment, it is based on the number of samples in the block. 5 (: Conditions for determining whether to apply or not can be defined. For example, the above conditions
- the reference value may be a preset value.
- the condition for determining whether to apply Moi 5 ( Is defined as ), ( ⁇ > reference value or
- the inequality sign is not limited to greater than the reference value, but may be defined as right, small, or less.
- the conditions for deciding whether or not are defined as ( ⁇ > reference value and 3 reference value)
- the reference value applied to and may be defined as the same value may be defined as different values, may be defined as the same sign (or inequality sign), or may be defined as different signs.
- the encoding/decoding process may be performed.
- the encoder/decoder may be applied in units of coding blocks rather than units of transform blocks in applying all 01(s.
- the encoder/decoder is a!
- feeders 5 may be performed in the coding block unit instead of the conversion block.
- the encoder/decoder can apply all 01 (only when certain conditions defined for some of the modes to which it is applied are satisfied. For example, as in the above-described embodiment, when determining whether to apply based on the number of samples or width/height, reference values for the planar mode, horizontal mode, and vertical mode may be set differently.
- the encoder/decoder is 2020/175965 1»(:1 ⁇ 1 ⁇ 2020/002920
- the filter flag indicates whether or not to apply filtering based on whether the ISP and/or PDPC are applied blocks, e.g. for blocks to which the ISP and PDPC are applied, the filter flag is 0. Or it can be set to a fixed value of 1. Alternatively, the filter length lag value can be determined by the MDIS (Mode dependent Intra smoothing) condition for the block to which the ISP and PDPC are applied, or the encoder/decoder has KP and PDPC. The filter flag value of the applied block and the filter flag value of the block to which only KP is applied can be applied differently.
- MDIS Mode dependent Intra smoothing
- the encoder/decoder may reset the intra prediction mode based on the width and height of the coding block.
- the encoder/decoder is The block to which the ISP is applied performs a re-interpretation process of the wide-angle mode based on the width and height of the coding block, and based on this, the reference sample filter flag can be set.
- the encoder/decoder can divide the block to which KP is applied/ The wide-angle mode application method can be set differently based on the size.
- a specific dividing direction is applied based on the width/height of the transform block or the width/height of the coding block, and other methods can be applied.
- the encoder/decoder can apply the wide-angle mode based on the width and height values of the divided transform blocks, for example, if the minimum of the width and height of the transform block is greater than, equal to or greater than the reference value.
- the wide-angle mode can be applied by using the height and width of the coding block, and the width and height of the transform block can be used to apply the wide-angle mode if it is equal to, smaller or smaller than the reference value.
- the width of the transform block If the minimum value of and height is greater than, equal to, or greater than the reference value, the wide-angle mode can be applied using the height and width of the conversion block, and if the minimum value is equal to, less than or less than the reference value, the wide-angle mode is applied by using the width and height of the coding block. can do.
- an encoder/decoder is a current block (coding block, coding Unit) into multiple transform blocks
- the coding block when the Intra subpartitions (ISP) mode is applied, the coding block may be divided into a plurality of transform blocks. Alternatively, if the size of the coding block is larger than the maximum transform size, the coding block may be divided into a plurality of transform blocks.
- the intra-sub-partition mode when the intra-sub-partition mode is applied, as shown in FIG. 15, the coding block is horizontally or vertically It can be divided into pseudorectangular transform blocks, and can be divided into two or four transform blocks.
- 16 is a diagram illustrating a process of encoding/decoding through a primary transform and a secondary transform according to an embodiment to which the present invention is applied.
- a coding block may be divided into a plurality of transform blocks, and an encoder/decoder may apply a transform to the divided transform block.
- FIG. 16 is a diagram in which two transforms are applied to the transform block. An example is shown.
- the Forward Primary Transform in Fig. 16 represents the first applied transform based on the encoder, that is, it may be referred to as a first order transform in the present invention.
- Forward Secondary Transform in Fig. 16 Denotes the second applied transformation on the basis of the encoder i.e. it can be referred to as a second order transformation in the present invention.
- 2nd transformation i.e. 2nd inverse transformation
- 1st transformation for the transform block inverse quantized with respect to the decoder side. I.e., first-order inverse transform
- the second-order transform is a Low Frequency Non-Separable Transform
- LFNST LFNST
- the conversion matrix (or conversion kernel, conversion type) used for the first conversion is a conversion known in conventional image compression technologies such as DCT-2, DST-7, DCT-8, etc. It can be a matrix.
- the quadratic transformation can be applied to a partial region within the transformation block.
- the partial region may be a 4x4 region or an 8x8 region.
- the position of the partial region is a coding block (or transform block).
- the width and height of the coding block are both greater than 4, it may be applied to the upper left 8x8 area, and if either side of the width and height is equal to 4, it may be applied to the upper left 4x4 area.
- the quadratic transformation can be applied to the luma component and chroma component of the block coded in intra mode.
- FIG. 17 shows a conversion kernel used for second-order conversion according to an embodiment of the present invention.
- a transformation kernel set (or transformation type set, transformation matrix set) can be determined based on the prediction mode used for intra prediction, and the table shown in Fig. 17 It can be defined in this encoder/decoder.
- the intra prediction mode can be defined from -14 to 83.
- a conversion kernel set may be determined for each grouped intra prediction mode. The same index may be applied to the luma component and the chroma component. Since the second-order conversion kernel set is determined based on the intra prediction mode, the conversion kernel set can be determined after acquiring (or determining) the intra prediction mode. This causes a dependency problem. Therefore, one embodiment of the present invention In, we explain how to get rid of this dependence.
- the encoder/decoder is in the intra prediction mode in consideration of the following.
- the encoder/decoder can determine the conversion kernel set based on the above-described items, and the above-described items can be used to determine the conversion kernel set as a combination of one or more.
- FIG. 18 illustrates a method of applying a quadratic transform in units of transform blocks according to an embodiment of the present invention.
- an encoder/decoder divides a coding block (or coding unit) into a plurality of transform blocks (or transform units), and applies a quadratic transform to each transform block. It is possible to apply a quadratic transformation to each transform block divided into a plurality of transform blocks in one coding unit.
- the size of each transform block is determined based on the segmentation method for the coding unit. ⁇ !
- the size of each transform block of the coding unit to which 5 is applied can be determined according to the vertical division or the horizontal division as shown in Fig. 15, and the size can be determined according to the number of additional divisions.
- the number of divisions in the block to which 5 is applied can be 2 or 4. In Fig.
- one coding unit is divided in the vertical direction into the block to which “all” is applied and the number of divisions is four.
- the size of the coding unit is ⁇
- the size of each transform block can be ⁇ /4 11.
- the size of the transform block can be used as the width and height to determine whether or not to apply the second transform.
- the encoder/decoder may use the same set of transform kernels for each transform block or use different sets of transform kernels.
- the segmented transform blocks have the same intra prediction mode and segmented blocks. If the size is the same, the same conversion kernel can be used. In the opposite case, the encoder/decoder can determine and use a kernel set for each conversion block.
- the luma component is converted into a plurality of conversion blocks. However, the chroma components may not be divided.
- both the luma transform block and the chroma transform block can use the same quadratic transform kernel set, and the size of the coding block to which the quadratic transform block is applied must be satisfied.
- the transform block size of and chroma may be different.
- the encoder/decoder can be applied to the 4x4 or 8x8 area according to the block size condition to which the 2nd transform block is applied.
- the encoder/decoder is the area applied to the luma transform. Chromas can also be used in the same area. Encoders/decoders can use different sets of conversion kernels, since the intra prediction modes of luma and chroma can be different.
- the size of the coding unit is larger than the maximum conversion size.
- the coding unit can be divided into a plurality of transform blocks without separate signaling.
- applying a second-order transform can reduce performance and increase complexity, thereby limiting the maximum coding block to which the second-order transform is applied.
- the block size can be the same as the maximum conversion size, or it can be used as a preset coding block size.
- the preset value can be 64, 32, 16, but is not limited thereto. It can be the length of the long side or the number of total samples.
- the encoder/decoder determines the size of the second-order transform block of the coding unit.
- Encoder/decoder is adaptive considering the ratio of horizontal to vertical/vertical to vertical ratio of the coding block.
- the area to which the quadratic transformation is applied can be determined without signaling.
- the current block is a square block. If not, the encoder/decoder can perform prediction using only the reference sample of the long side in the case of the IX: mode of intra prediction mode, in this case, the sample value of the short side may not be reflected at all in the prediction of the current coding block. In this case, the difference between the predicted value of the current block and the reference sample of the short side may be large. Therefore, the encoder/decoder can perform position-based filtering of the sample when performing intra prediction.
- the location-based filtering method of these samples can be referred to as all.! 5 ( : can be referred to as: all encoders/decoders.! 5 ( : is applied IX: in the mode, the first reference sample and the second reference sample are adjacent to the upper left. You can use the reference sample value to perform weight-based filtering. In this case, you can use Equations 7 to 12 below to derive the reference sample and/or the weight applied to each reference sample.
- predSamples[ x ][ y] clip1 Cmp( (refL[ x ][ y] * wL[ x] + refT[ x ][ y] * wT[ y]-p[ -1 ][ -1] * wTL[ x ][ y] + (64-wL[ x]-wT[ y] + wTL[ x ][ y])
- the left reference sample can be derived using Equation 7 and the right reference sample can be derived using Equation 8.
- the weight value applied to the right reference sample can be derived using Equation 9
- the weight value applied to the left reference sample can be derived using Equation W, and the weight value applied to the reference sample located at the upper left corner can be derived using Equation 11.
- the encoder/decoder can be derived using Equation 11. Based on the determined weight value, a prediction sample can be generated according to Equation 12.
- the encoder/decoder is not a square block.
- the weight value can be set differently from that of the long side.
- the weight value can be set to 16, 8, 4, etc. instead of 32. Or, used in Equations 9 and W described above.
- the scaled variable nScale can be used. Encoder/decoder can set its value according to the position of the long side and the short side.
- the encoder/decoder when using multiple reference line samples, the encoder/decoder
- PDPC can be applied in vertical mode and/or horizontal mode, or encoder/decoder can apply PDPC in vertical, horizontal, DC, PLANAR mode.
- FIG. 20 is a diagram showing a video signal processing method according to an embodiment of the present invention
- a decoder is mainly described, but the present invention is not limited thereto, and the video signal processing method according to the present embodiment can be applied in substantially the same manner to the encoder.
- the decoder is in the current block intra subpartition (ISP, Intra)
- the decoder generates prediction blocks of the transform blocks by performing intra prediction for each of the transform blocks (S2003).
- the decoder restores the current block based on the residual block of the transform block and the prediction block (S2004).
- the step of generating the prediction block includes a position-dependent intra prediction sample in units of transform blocks divided from the current block. 2020/175965 1»(:1 ⁇ 1 ⁇ 2020/002920
- 29 It may include the step of performing position-dependent intra prediction sample filtering.
- the step of generating the prediction block further includes a step of determining whether to apply the position-dependent intra prediction sample filtering based on at least one of the width and height of the transform block. can do.
- the width of the transform block is greater than or equal to a preset reference value, and the height of the transform block is If it is greater than or equal to the set reference value, it can be performed by deciding to apply the position-dependent intra prediction sample filtering.
- the residual block of the transform block can be derived by performing an inverse secondary transform and an inverse primary transform in units of the transform block.
- the step of determining whether or not the quadratic transform is applied to the current block When the quadratic transform is applied to the current block, predefined based on the intra prediction mode of the current block Deriving a quadratic transform kernel set applied to the current block from among quadratic transform kernel sets; determining a quadratic transform kernel applied to the current block in the determined quadratic transform kernel set; Inverse quadratic transform in units of the transform block.
- the method according to embodiments of the present invention is one or more ASICs (Application Specific Integrated Circuits), DSPs (Digital Signal Processors), DSPDs (Digital Signal Processing Devices), PLDs (Programmable Logic Devices), FPGAs (Field Programmable Gate Arrays), processors, controllers, microcontrollers, microprocessors, etc.
- ASICs Application Specific Integrated Circuits
- DSPs Digital Signal Processors
- DSPDs Digital Signal Processing Devices
- PLDs Programmable Logic Devices
- FPGAs Field Programmable Gate Arrays
- processors controllers, microcontrollers, microprocessors, etc.
- the method according to the embodiments of the present invention may be implemented in the form of a module, procedure, or function that performs the functions or operations described above.
- the software code is in a memory. It can be stored and driven by the processor.
- the memory can be located inside or outside the processor, and can send and receive data to and from the processor by a variety of known means.
- Some embodiments may be embodied in the form of a recording medium containing instructions executable by a computer, such as program modules executed by a computer.
- the computer-readable medium is any number of available media that can be accessed by the computer.
- Computer storage media includes both volatile and nonvolatile, removable and non-removable media implemented in any method or technology for storage of information such as computer-readable instructions, data structures, program modules or other data.
- It typically contains computer-readable instructions, data structures, or other data in a modulated data signal, such as program modules, or other transmission mechanisms, and includes any information carrier.
Landscapes
- Engineering & Computer Science (AREA)
- Multimedia (AREA)
- Signal Processing (AREA)
- Compression Or Coding Systems Of Tv Signals (AREA)
Abstract
비디오 신호를 인코딩하거나 디코딩하는 비디오 신호 처리 방법 및 장치가 개시된다. 비디오 신호 처리 방법에 있어서, 현재 블록에 인트라 서브 파티션(ISP, Intra Sub-Partitions) 모드가 적용되는지 여부를 결정하는 단계; 상기 현재 블록에 ISP 모드가 적용되는 경우, 상기 현재 블록을 복수의 수평 또는 수직 방향의 직사각형 변환 블록들로 분할하는 단계; 상기 변환 블록들 각각에 대하여 인트라 예측을 수행함으로써 상기 변환 블록들의 예측 블록을 생성하는 단계; 및 상기 변환 블록의 잔차 블록 및 상기 예측 블록에 기초하여 상기 현재 블록을 복원하는 단계를 포함할 수 있다.
Description
2020/175965 1»(:1/10公020/002920 명세서
발명의명칭:인트라예측기반비디오신호처리방법및장치 기술분야
[1] 본발명은비디오신호의처리방법및장치에관한것으로,보다상세하게는 인트라예측을기반으로비디오신호를인코딩하거나디코딩하는비디오신호 처리방법및장치에관한것이다.
배경기술
[2] 압축부호화란디지털화한정보를통신회선을통해전송하거나,저장매체에 적합한형태로저장하기위한일련의신호처리기술을의미한다.압축부호화의 대상에는음성,영상,문자등의대상이존재하며,특히영상을대상으로압축 부호화를수행하는기술을비디오영상압축이라고일컫는다.비디오신호에 대한압축부호화는공간적인상관관계 ,시간적인상관관계 ,확률적인상관관계 등을고려하여잉여정보를제거함으로써이루어진다.그러나최근의다양한 미디어및데이터전송매체의발전으로인해,더욱고효율의비디오신호처리 방법및장치가요구되고있다.
발명의상세한설명
기술적과제
[3] 본발명의목적은비디오신호의코딩효율을높이고자함에있다.구체적으로, 본발명은인트라서브파티션 (Intra sub-partitions)이적용됨에따라예즉및 복원이수행되는단위로 PDPC(position-dependent intra prediction combination)및 LFNST(Low-Frequency Non-Separable Transform)를수행하는방법을제안한다. 과제해결수단
[4] 상기와같은과제를해결하기위해 ,본발명은다음과같은비디오신호처리 장치및비디오신호처리방법을제공한다.
[5] 본발명의일실시예에따르면,비디오신호처리방법에 있어서,현재블록에 인트라서브파티션 (ISP, Intra Sub-Partitions)모드가적용되는지여부를결정하는 단계 ;상기현재블록에 ISP모드가적용되는경우,상기현재블록을복수의 수평또는수직방향의직사각형변환블록들로분할하는단계 ;상기변환 블록들각각에대하여인트라예측을수행함으로써상기변환블록들의 예측 블록을생성하는단계 ;및상기변환블록의잔차블록및상기 예측블록에 기초하여상기현재블록을복원하는단계를포함하되,상기 예측블록을 생성하는단계는,상기현재블록으로부터분할된변환블록단위로
위치-의존적인인트라예즉샘늘필터링 (position-dependent intra prediction sample filtering)을수행하는단계를포함할수있다.
[6] 실시예로서,상기예측블록을생성하는단계는,상기변환블록의너비및높이 중적어도하나에기초하여상기위치-의존적인인트라예측샘플필터링의적용
2020/175965 1»(:1^1{2020/002920
2 여부를결정하는단계를더포함할수있다.
[7] 실시예로서,상기위치-의존적인인트라예측샘플필터링의적용여부를
결정하는단계는,상기변환블록의너비가기설정된기준값보다크거나같고, 그리고,상기변환블록의높이가상기기설정된기준값보다크거나같은경우, 상기위치-의존적인인트라예측샘플필터링을적용하는것으로결정함으로써 수행될수있다.
[8] 실시예로서,상기변환블록의잔차블록은상기변환블록단위로이차
역변환 (inverse secondary transform)및일차역변환 (inverse primary仕 ansform)을 수행함으로써유도될수있다.
[9] 실시예로서,상기현재블록에이차변환이적용되는지여부를결정하는단계; 상기현재블록에상기이차변환이적용되는경우,상기현재블록의인트라 예측모드에기초하여미리정의된이차변환커널세트들중에서상기현재 블록에적용되는이차변환커널세트를유도하는단계;상기결정된이차변환 커널세트내에서상기현재블록에적용되는이차변환커널을결정하는단계; 상기변환블록단위로이차역변환을수행함으로써상기변환블록의이차 역변환된블록을생성하는단계;및상기이차역변환된블록에대하여일차 역변환을수행함으로써,상기변환블록의잔차블록을생성하는단계를포함할 수있다.
[1이 본발명의일실시예에따르면,비디오신호처리장치에 있어서,프로세서를 포함하며,상기프로세서는,현재블록에인트라서브파티션 (ISP, Intra
Sub-Partitions)모드가적용되는지여부를결정하고,상기현재블록에 ISP 모드가적용되는경우,상기현재블록을복수의수평또는수직방향의 직사각형변환블록들로분할하고,상기변환블록들각각에대하여인트라 예측을수행함으로써상기변환블록들의 예측블록을생성하고,상기변환 블록의잔차블록및상기 예측블록에기초하여상기현재블록을복원하되, 상기프로세서는,상기현재블록으로부터분할된변환블록단위로
위치-의존적인인트라예즉샘늘필터링 (position-dependent intra prediction sample filtering)을수행하는것을특징으로하는,비디오신호처리장치가제공된다.
[11] 실시예로서,상기프로세서는,상기변환블록의너비및높이중적어도하나에 기초하여상기위치-의존적인인트라예측샘플필터링의적용여부를결정할수 있다.
[12] 실시예로서,상기프로세서는,상기변환블록의너비가기설정된기준값보다 크거나같고,그리고,상기변환블록의높이가상기기설정된기준값보다 크거나같은경우,상기위치-의존적인인트라예측샘플필터링을적용하는 것으로결정할수있다.
[13] 실시예로서,상기변환블록의잔차블록은상기변환블록단위로이차
역변환 (inverse secondary transform)및일차역변환 (inverse primary仕 ansform)을 수행함으로써유도될수있다.
2020/175965 1»(:1^1{2020/002920
[14] 실시예로서,상기프로세서는,상기현재블록에이차변환이적용되는지
여부를결정하고,상기현재블록에상기이차변환이적용되는경우,상기현재 블록의인트라예측모드에기초하여미리정의된이차변환커널세트들중에서 상기현재블록에적용되는이차변환커널세트를유도하고,상기결정된이차 변환커널세트내에서상기현재블록에적용되는이차변환커널을결정하고, 상기변환블록단위로이차역변환을수행함으로써상기변환블록의이차 역변환된블록을생성하고,상기이차역변환된블록에대하여일차역변환을 수행함으로써,상기변환블록의잔차블록을생성할수있다.
[15] 본발명의일실시예에따르면,현재블록에인트라서브파티션 (ISP, Intra Sub-Partitions)모드가적용되는지여부를결정하는단계;상기현재블록에 ISP 모드가적용되는경우,상기현재블록을복수의수평또는수직방향의 직사각형변환블록들로분할하는단계 ;상기변환블록들각각에대하여인트라 예측을수행함으로써상기변환블록들의예측블록을생성하는단계 ;및원본 블록에서상기예측블록을감산함으로써상기변환블록의잔차블록을 생성하는단계를포함하되,상기예측블록을생성하는단계는,상기현재 블록으로부터분할된변환블록단위로위치-의존적인인트라예측샘플 필터링 (position-dependent intra prediction sample filtering)을수행하는단계를 포함하는,비디오신호처리방법이제공된다.
[16] 본발명의일실시예에따르면,컴퓨팅디바이스의하나이상의프로세서에서 실행하도록구성된컴퓨터실행가능한컴포넌트가저장된비
일시적 (non-仕 ansitory)컴퓨터판독가능한매체 (computer-executable
component)로서,상기컴퓨터실행가능한컴포넌트는,현재블록에인트라서브 파티션 (ISP, Intra Sub-Partitions)모드가적용되는지여부를결정하고,상기현재 블록에 ISP모드가적용되는경우,상기현재블록을복수의수평또는수직 방향의직사각형변환블록들로분할하고,상기변환블록들각각에대하여 인트라예측을수행함으로써상기변환블록들의예측블록을생성하고,상기 변환블록의잔차블록및상기예측블록에기초하여상기현재블록을 복원하되,상기컴퓨터실행가능한컴포넌트는,상기현재블록으로부터분할된 변환블록단위로위치-의존적인인트라예즉샘늘필터링 (position-dependent intra prediction sample filtering)을수행하는것을특징으로하는,비일시적 컴퓨터판독가능한매체가제공된다.
발명의효과
[17] 본발명의실시예에따르면,비디오신호의코딩효율이높아질수있다.또한, 본발명의일실시예에따르면,인트라서브파티션 (Intra sub-partitions)에의해 분할된변환블록단위로 PDPC(po sition-dependent intra prediction combination)및 LFNST를수행함으로써예측의정확도를높이고압축성능을향상시킬수있다. 도면의간단한설명
2020/175965 1»(:1^1{2020/002920
[18] 도 1은본발명의일실시예에따른비디오신호인코딩장치의개략적인
블록도이다.
[19] 도 2는본발명의일실시예에따른비디오신호디코딩장치의개략적인
블록도이다.
[2이 도 3은픽쳐내에서코딩트리유닛이코딩유닛들로분할되는실시예를
도시한다.
[21] 도 4는쿼드트리및멀티-타입트리의분할을시그널링하는방법의일
실시예를도시한다.
[22] 도 5및도 6은본발명의실시예에따른인트라예측방법을더욱구체적으로 도시한다.
[23] 도 7은본발명의일실시예에따른인터 예측방법을도시한다.
[24] 도 8은인코더가레지듀얼신호를변환하는방법을구체적으로나타내는
도면이다.
[25] 도 9은인코더및디코더가변환계수를역변환하여레지듀얼신호를획득하는 방법을구체적으로나타내는도면이다.
[26] 도 W은본발명의일실시예에따른코딩블록이복수개의변환블록으로 분할되는경우인트라예측모드의적용방법을설명하기위한도면이다.
[27] 도 11은본발명의일실시예에따른 PDPC(po sition-dependent intra prediction combination)적용방법을설명하기위한도면이다.
[28] 도 12는본발명의일실시예로서인트라예측모드에따라 PDPC에이용되는 참조샘플을예시하는도면이다.
[29] 도 13은본발명의일실시예로서인트라예측모드에따라 PDPC에이용되는 참조샘플을예시하는도면이다.
[3이 도 14는본발명이적용되는일실시예에따른코딩블록에대한 ISP(Intra subpartitions)및 PDPC(position-dependent intra prediction combination)적용방법을 예시하는도면이다.
[31] 도 15는본발명의일실시예에따른변환유닛분할처리방법을설명하기위한 도면이다.
[32] 도 16은본발명이적용되는일실시예에따른 1차변환 (primary transform)및 2차변환 (secondary transform)을거쳐인코딩/디코딩되는프로세스를나타내는 도면이다.
[33] 도 17는본발명의일실시예에따른 2차변환에이용되는변환커널을
선택하는방법을설명하기위한도면이다.
[34] 도 18은본발명의일실시예에따른변환블록단위 2차변환적용방법을
예시하는도면이다.
[35] 도 19은본발명이적용되는일실시예에따른인트라예측모드가적용된현재 코딩블록에 PDPC를적용하는방법을나타낸도면이다.
[36] 도 20은본발명의일실시예를따른비디오신호처리방법을나타내는
2020/175965 1»(:1/10公020/002920
5 흐름도이다.
발명의실시를위한형태
[37] 본명세서에서사용되는용어는본발명에서의기능을고려하면서가능한현재 널리사용되는일반적인용어를선택하였으나,이는당분야에종사하는 기술자의의도,관례또는새로운기술의출현등에따라달라질수있다.또한 특정경우는출원인이임의로선정한용어도있으며,이경우해당되는발명의 설명부분에서그의미를기재할것이다.따라서본명세서에서사용되는 용어는,단순한용어의명칭이아닌그용어가가진실질적인의미와본 명세서의전반에걸친내용을토대로해석되어야함을밝혀두고자한다.
[38] 본명세서에서일부용어들은다음과같이해석될수있다.코딩은경우에따라 인코딩또는디코딩으로해석될수있다.본명세서에서비디오신호의 인코딩 (부호화)을수행하여비디오신호비트스트림을생성하는장치는인코딩 장치또는인코더로지칭되며,비디오신호비트스트림의디코딩 (복호화)을 수행하여비디오신호를복원하는장치는디코딩장치또는디코더로지칭된다. 또한,본명세서에서비디오신호처리장치는인코더및디코더를모두 포함하는개념의용어로사용된다.정보 (information)는값 (values),
파라미터 (parameter),계수 (coefficients),성분 (elements)등을모두포함하는 용어로서,경우에따라의미는달리해석될수있으므로본발명은이에 한정되지아니한다.’유닛’은영상처리의기본단위또는픽쳐의특정위치를 지칭하는의미로사용되며 ,루마 (luma)성분및크로마 (chroma)성분중적어도 하나를포함하는이미지영역을가리킨다.또한,‘블록’은루마성분및크로마 성분들 (즉, Cb및 Cr)중특정성분을포함하는이미지영역을가리킨다.다만, 실시예에따라서‘유닛 블록’,’파티션’및’영역’등의용어는서로혼용하여 사용될수있다.또한,본명세서에서유닛은코딩유닛,예측유닛,변환유닛을 모두포함하는개념으로사용될수있다.픽쳐는필드또는프레임을가리키며 , 실시예에따라상기용어들은서로혼용하여사용될수있다.
[39] 도 1은본발명의일실시예에따른비디오신호인코딩장치 (100)의개략적인 블록도이다.도 1을참조하면,본발명의인코딩장치 (100)는변환부 (110), 양자화부 (115),역양자화부 (120),역변환부 (125),필터링부 (130),예측부 (150)및 엔트로피코딩부 (160)를포함한다.
[4이 변환부 (1 W)는입력받은비디오신호와예측부 (150)에서생성된예측신호의 차이인레지듀얼신호를변환하여변환계수값을획득한다.예를들어,이산 코사인변환 (Discrete Cosine Transform, DCT),이산사인변환 (Discrete Sine Transform, DST)또는웨이블릿변환 (Wavelet Transform)등이사용될수있다. 이산코사인변환및이산사인변환은입력된픽쳐신호를블록형태로나누어 변환을수행하게된다.변환에 있어서변환영역내의값들의분포와특성에 따라서코딩효율이달라질수있다.양자화부 (115)는변환부 (H0)에서출력된
2020/175965 1»(:1^1{2020/002920
6 변환계수값을양자화한다.
[41] 코딩효율을높이기위하여픽쳐신호를그대로코딩하는것이아니라,
예측부 (150)를통해이미코딩된영역을이용하여픽쳐를예측하고,예측된 픽쳐에원본픽쳐와예측픽쳐간의레지듀얼값을더하여복원픽쳐를획득하는 방법이사용된다.인코더와디코더에서미스매치가발생되지않도록하기위해, 인코더에서예측을수행할때에는디코더에서도사용가능한정보를사용해야 한다.이를위해,인코더에서는부호화한현재블록을다시복원하는과정을 수행한다.역양자화부 (120)에서는변환계수값을역양자화하고,
역변환부 (125)에서는역양자화된변환계수값을이용하여레지듀얼값을 복원한다.한편,필터링부 (130)는복원된픽쳐의품질개선및부호화효율 향상을위한필터링연산을수행한다.예를들어,디블록킹필터,샘플적응적 오프셋 (Sample Adaptive Offset, SA0)및적응적루프필터등이포함될수있다. 필터링을거친픽쳐는출력되거나참조픽쳐로이용하기위하여복호픽쳐 버퍼 (Decoded Picture Buffer, DPB, 156)에저장된다.
[42] 코딩효율을높이기위하여픽쳐신호를그대로코딩하는것이아니라,
예측부 (150)를통해이미코딩된영역을이용하여픽쳐를예측하고,예측된 픽쳐에원픽쳐와예측픽쳐간의레지듀얼값을더하여복원픽쳐를획득하는 방법이사용된다.인트라예측부 (152)에서는현재픽쳐내에서인트라예측을 수행하며,인터예측부 (154)에서는복호픽쳐버퍼 (156)에저장된참조픽쳐를 이용하여현재픽쳐를예측한다.인트라예측부 (152)는현재픽쳐내의복원된 영역들로부터인트라예측을수행하여 ,인트라부호화정보를엔트로피 코딩부 (160)에전달한다.인터예측부 (154)는다시모션추정부 (154a)및모션 보상부 (154b)를포함하여구성될수있다.모션추정부 (154a)에서는복원된특정 영역을참조하여현재영역의모션벡터값을획득한다.모션추정부 (154a)에서는 참조영역의위치정보 (참조프레임,모션벡터등)등을엔트로피코딩부 (160)로 전달하여비트스트림에포함될수있도록한다.모션추정부 (154a)에서전달된 모션벡터값을이용하여모션보상부 (154b)에서는화면간모션보상을수행한다.
[43] 예측부 (150)는인트라예측부 (152)와인터예측부 (154)를포함한다.인트라
예측부 (152)는현재픽쳐내에서인트라 (intra)예측을수행하며,인터
예측부 (154)는복호픽쳐버퍼 (156)에저장된참조픽쳐를이용하여현재픽쳐를 예측하는인터 (inter)예측을수행한다.인트라예측부 (152)는현재픽쳐내의 복원된샘플들로부터인트라예측을수행하여,인트라부호화정보를엔트로피 코딩부 (160)에전달한다.인트라부호화정보는인트라예측모드, MPM(Most Probable Mode)플래그, MPM인덱스중적어도하나를포함할수있다.인트라 부호화정보는참조샘플에관한정보를포함할수있다.인터예측부 (154)는 모션추정부 (154a)및모션보상부 (154b)를포함하여구성될수있다.모션 추정부 (154a)는복원된참조픽쳐의특정영역을참조하여현재영역의모션 벡터값을획득한다.모션추정부 (154a)는참조영역에대한모션정보세트 (참조
2020/175965 1»(:1^1{2020/002920
7 픽쳐인덱스,모션벡터정보등)를엔트로피코딩부 (160)로전달한다.모션 보상부 (154b)는모션추정부 (154a)에서전달된모션벡터값을이용하여모션 보상을수행한다.인터예측부 (154)는참조영역에대한모션정보를포함하는 인터부호화정보를엔트로피코딩부 (160)에전달한다.
[44] 추가적인실시예에따라,예측부 (150)는인트라블록카피 (block copy, BC) 예측부 (미도시)를포함할수있다.인트라 BC예측부는현재픽쳐내의복원된 샘플들로부터인트라 BC예측을수행하여,인트라 BC부호화정보를엔트로피 코딩부 (160)에전달한다.인트라 BC예측부는현재픽쳐내의특정영역을 참조하여현재영역의 예측에이용되는참조영역을나타내는블록벡터값을 획득한다.인트라 BC예측부는획득된블록벡터값을이용하여인트라 BC 예측을수행할수있다.인트라 BC예측부는인트라 BC부호화정보를엔트로피 코딩부 (160)로전달한다.인트라 BC부호화정보는블록벡터정보를포함할수 있다.
[45] 위와같은픽쳐예측이수행될경우,변환부 (110)는원본픽쳐와예측픽쳐간의 레지듀얼값을변환하여변환계수값을획득한다.이때,변환은픽쳐내에서 특정블록단위로수행될수있으며,특정블록의크기는기설정된범위내에서 가변할수있다.양자화부 (115)는변환부 (U0)에서생성된변환계수값을 양자화하여엔트로피코딩부 (160)로전달한다.
[46] 엔트로피코딩부 (160)는양자화된변환계수를나타내는정보,인트라부호화 정보,및인터부호화정보등을엔트로피코딩하여비디오신호비트스트림을 생성한다.엔트로피코딩부 (160)에서는가변길이코딩 (Variable Length Coding, VLC)방식과산술코딩 (arithmetic coding)방식등이사용될수있다.가변길이 코딩 (VLC)방식은입력되는심볼들을연속적인코드워드로변환하는데, 코드워드의길이는가변적일수있다.예를들어,자주발생하는심볼들을짧은 코드워드로,자주발생하지않은심볼들은긴코드워드로표현하는것이다.가변 길이코딩방식으로서컨텍스트기반적응형가변길이코딩 (Context-based Adaptive Variable Length Coding, CAVLC)방식이사용될수있다.산술코딩은 연속적인데이터심볼들을하나의소수로변환하는데,산술코딩은각심볼을 표현하기위하여필요한최적의소수비트를얻을수있다.산술코딩으로서 컨텍스트기반적응형산술부호화 (Context-based Adaptive Binary Arithmetic Code, CAB AC)가이용될수있다.예를들어,엔트로피코딩부 (160)는양자화된 변환계수를나타내는정보를이진화할수있다.또한,엔트로피코딩부 (160)는 이진화된정보를산술코딩하여비트스트림을생성할수있다.
[47] 상기생성된비트스트림은 NAL(Network Abstraction Layer)유닛을기본단위로 캡슐화된다. NAL유닛은부호화된정수개의코딩트리유닛 (coding tree unit)을 포함한다.비디오디코더에서비트스트림을복호화하기위해서는먼저 비트스트림을 NAL유닛단위로분리한후,분리된각각의 NAL유닛을 복호화해야한다.한편,비디오신호비트스트림의복호화를위해필요한
2020/175965 1»(:1^1{2020/002920
8 정보들은픽쳐파라미터세트 (Picture Parameter Set, PPS),시퀀스파라미터 세트 (Sequence Parameter Set, SPS),비디오파라미터세트 (Video Parameter Set, VPS)등과같은상위레벨세트의 RBSP(Raw Byte Sequence Payload)를통해 전송될수있다.
[48] 한편,도 1의블록도는본발명의일실시예에따른인코딩장치 (100)를나타낸 것으로서,분리하여표시된블록들은인코딩장치 (100)의엘리먼트들을 논리적으로구별하여도시한것이다.따라서전술한인코딩장치 (100)의 엘리먼트들은디바이스의설계에따라하나의칩으로또는복수의칩으로 장착될수있다.일실시예에따르면,전술한인코딩장치 (W0)의각엘리먼트의 동작은프로세서 (미도시)에의해수행될수있다.
[49] 도 2는본발명의일실시예에따른비디오신호디코딩장치 (200)의개략적인 블록도이다.도 2를참조하면본발명의디코딩장치 (200)는엔트로피
디코딩부 (2W),역양자화부 (220),역변환부 (225),필터링부 (230)및예측부 (250)를 포함한다.
[5이 엔트로피디코딩부 (210)는비디오신호비트스트림을엔트로피디코딩하여,각 영역에대한변환계수정보,인트라부호화정보,인터부호화정보등을 추출한다.예를들어,엔트로피디코딩부 (210)는비디오신호
비트스트림으로부터특정영역의변환계수정보에대한이진화코드를획득할 수있다.또한,엔트로피디코딩부 (2W)는이진화코드를역이진화하여양자화된 변환계수를획득한다.역양자화부 (220)는양자화된변환계수를역양자화하고, 역변환부 (225)는역양자화된변환계수를이용하여레지듀얼값을복원한다. 비디오신호처리장치 (200)는역변환부 (225)에서획득된레지듀얼값을 예측부 (250)에서획득된예측값과합산하여원래의화소값을복원한다.
[51] 한편,필터링부 (230)는픽쳐에대한필터링을수행하여화질을향상시킨다. 여기에는블록왜곡현상을감소시키기위한디블록킹필터및/또는픽쳐전체의 왜곡제거를위한적응적루프필터등이포함될수있다.필터링을거친픽쳐는 출력되거나다음픽쳐에대한참조픽쳐로이용하기위하여복호픽쳐
버퍼 (DPB, 256)에저장된다.
[52] 예측부 (250)는인트라예측부 (252)및인터 예측부 (254)를포함한다.
예측부 (250)는전술한엔트로피디코딩부 (2 W)를통해복호화된부호화타입,각 영역에대한변환계수,인트라/인터부호화정보등을활용하여예측픽쳐를 생성한다.복호화가수행되는현재블록을복원하기위해서,현재블록이포함된 현재픽쳐또는다른픽쳐들의복호화된영역이이용될수있다.복원에현재 픽쳐만을이용하는,즉인트라예측또는인트라 BC예측을수행하는픽쳐 (또는, 타일/슬라이스)를인트라픽쳐또는 I픽쳐 (또는,타일/슬라이스),인트라예측, 인터 예측및인트라 BC예측을모두수행할수있는픽쳐 (또는,
타일/슬라이스)를인터픽쳐 (또는,타일/슬라이스)라고한다.인터픽쳐 (또는, 타일/슬라이스)중각블록의샘플값들을예측하기위하여최대하나의모션
2020/175965 1»(:1^1{2020/002920
9 벡터 및참조픽쳐 인덱스를이용하는픽쳐 (또는,타일/슬라이스)를예측 픽쳐 (predictive picture)또는 P픽쳐 (또는,타일/슬라이스)라고하며,최대두개의 모션벡터 및참조픽쳐 인덱스를이용하는픽쳐 (또는,타일/슬라이스)를쌍예측 픽쳐 (Bi-predictive picture)또는 B픽쳐 (또는,타일/슬라이스)라고한다.다시 말해서, P픽쳐 (또는,타일/슬라이스)는각블록을예측하기 위해최대하나의 모션정보세트를이용하고, B픽쳐 (또는,타일/슬라이스)는각블록을예측하기 위해최대두개의모션정보세트를이용한다.여기서,모션정보세트는하나 이상의모션벡터와하나의참조픽쳐 인덱스를포함한다.
[53] 인트라예측부 (252)는인트라부호화정보및현재픽쳐내의복원된샘플들을 이용하여 예즉블록을생성한다.전술한바와같이 ,인트라부호화정보는 인트라예측모드, MPM(Most Probable Mode)플래그, MPM인덱스중적어도 하나를포함할수있다.인트라예측부 (252)는현재블록의좌측및/또는상측에 위치한복원된샘플들을참조샘플들로이용하여 현재블록의 샘플값들을 예측한다.본개시에서,복원된샘플들,참조샘플들및현재블록의 샘플들은 픽셀들을나타낼수있다.또한,샘플값 (sample value)들은픽셀값들을나타낼수 있다.
[54] 일실시예에 따르면,참조샘플들은현재블록의주변블록에포함된샘플들일 수있다.예를들어,참조샘플들은현재블록의좌측경계에 인접한샘플들 및/또는상측경계에 인접한샘플들일수있다.또한,참조샘플들은현재블록의 주변블록의 샘플들중현재블록의좌측경계로부터기 설정된거리 이내의 라인상에위치하는샘플들및/또는현재블록의상측경계로부터 기설정된 거리 이내의 라인상에위치하는샘플들일수있다.이때,현재블록의주변 블록은현재블록에 인접한좌측 (L)블록,상측 (A)블록,하좌측 (Below Left, BL) 블록,상우측 (Above Right, AR)블록또는상좌측 (Above Left, AL)블록중적어도 하나를포함할수있다.
[55] 인터 예측부 (254)는복호픽쳐버퍼 (256)에 저장된참조픽쳐 및 인터부호화 정보를이용하여 예측블록을생성한다.인터부호화정보는참조블록에 대한 현재블록의모션정보세트 (참조픽쳐 인덱스,모션벡터정보등)를포함할수 있다.인터 예측에는 L0예측, L1예측및쌍예측 (Bi-prediction)이 있을수있다.
L0예측은 L0픽쳐 리스트에포함된 1개의 참조픽쳐를이용한예측이고시 예측은 L1픽쳐 리스트에포함된 1개의 참조픽쳐를이용한예측을의미한다. 이를위해서는 1세트의모션정보 (예를들어,모션벡터 및참조픽쳐 인덱스)가 필요할수있다.쌍예측방식에서는최대 2개의참조영역을이용할수있는데, 이 2개의참조영역은동일한참조픽쳐에존재할수도있고,서로다른픽쳐에 각각존재할수도있다.즉,쌍예측방식에서는최대 2세트의모션정보 (예를 들어,모션벡터 및참조픽쳐 인덱스)가이용될수있는데, 2개의모션벡터가 동일한참조픽쳐 인덱스에 대응될수도있고서로다른참조픽쳐 인덱스에 대응될수도있다.이때,참조픽쳐들은시간적으로현재픽쳐 이전이나이후
2020/175965 1»(:1^1{2020/002920
10 모두에표시 (또는출력)될수있다.일실시예에따라,쌍예측방식에서는 사용되는 2개의참조영역은 L0픽쳐리스트및 L1픽쳐리스트각각에서선택된 영역일수있다.
[56] 인터예측부 (254)는모션벡터및참조픽쳐인덱스를이용하여현재블록의 참조블록을획득할수있다.상기참조블록은참조픽쳐인덱스에대응하는 참조픽쳐내에존재한다.또한,모션벡터에의해서특정된블록의샘플값또는 이의보간 (interpolation)된값이현재블록의예즉자 (predictor)로이용될수있다. 서브펠 (sub-pel)단위의픽셀정확도를갖는모션예측을위하여이를테면,루마 신호에대하여 8 -탭보간필터가,크로마신호에대하여 4 -탭보간필터가사용될 수있다.다만,서브펠단위의모션예측을위한보간필터는이에한정되지 않는다.이와같이인터 예측부 (254)는이전에복원된픽쳐로부터현재유닛의 텍스쳐를예즉하는모션보상 (motion compensation)을수행한다.이때,인터 예측부는모션정보세트를이용할수있다.
[57] 추가적인실시예에따라,예측부 (250)는인트라 BC예측부 (미도시)를포함할 수있다.인트라 BC예측부는현재픽쳐내의복원된샘플들을포함하는특정 영역을참조하여현재영역을복원할수있다.인트라 BC예측부는엔트로피 디코딩부 (2W)로부터현재영역에대한인트라 BC부호화정보를획득한다. 인트라 BC예측부는현재픽쳐내의특정영역을지시하는현재영역의블록 벡터값을획득한다.인트라 BC예측부는획득된블록벡터값을이용하여인트라 BC예측을수행할수있다.인트라 BC부호화정보는블록벡터정보를포함할 수있다.
[58] 상기인트라예측부 (252)또는인터 예측부 (254)로부터출력된예측값,및
역변환부 (225)로부터출력된레지듀얼값이더해져서복원된비디오픽쳐가 생성된다.즉,비디오신호디코딩장치 (200)는예측부 (250)에서생성된예측 블록과역변환부 (225)로부터획득된레지듀얼을이용하여현재블록을 복원한다.
[59] 한편,도 2의블록도는본발명의일실시예에따른디코딩장치 (200)를나타낸 것으로서,분리하여표시된블록들은디코딩장치 (200)의엘리먼트들을 논리적으로구별하여도시한것이다.따라서전술한디코딩장치 (200)의 엘리먼트들은디바이스의설계에따라하나의칩으로또는복수의칩으로 장착될수있다.일실시예에따르면,전술한디코딩장치 (200)의각엘리먼트의 동작은프로세서 (미도시)에의해수행될수있다.
[6이 도 3은픽쳐내에서코딩트리유닛 (Coding Tree Unit, CTU)이코딩
유닛들 (Coding Units, CUs)로분할되는실시예를도시한다.비디오신호의코딩 과정에서,픽쳐는코딩트리유닛 (CTU)들의시퀀스로분할될수있다.코딩트리 유닛은루마 (luma)샘플들의 NXN블록과,이에대응하는크로마 (chroma) 샘플들의 2개의블록들로구성된다.코딩트리유닛은복수의코딩유닛들로 분할될수있다.코딩트리유닛은분할되지않고리프노드가될수도있다.이
2020/175965 1»(:1^1{2020/002920
11 경우,코딩트리유닛자체가코딩유닛이될수있다.코딩유닛은상기에서 설명한비디오신호의 처리과정,즉인트라/인터 예측,변환,양자화및/또는 엔트로피코딩등의과정에서픽쳐를처리하기위한기본단위를가리킨다. 하나의픽쳐내에서코딩유닛의크기 및모양은일정하지 않을수있다.코딩 유닛은정사각형또는직사각형의모양을가질수있다.직사각형코딩 유닛 (또는,직사각형블록)은수직코딩유닛 (또는,수직블록)과수평코딩 유닛 (또는,수평블록)을포함한다.본명세서에서,수직블록은높이가너비보다 큰블록이며 ,수평블록은너비가높이보다큰블록이다.또한,본명세서에서 정사각형이 아닌 (non-square)블록은직사각형블록을가리킬수있지만,본 발명은이에 한정되지 않는다.
[61] 도 3을참조하면,코딩트리유닛은먼저 쿼드트리 (Quad Tree, QT)구조로
분할된다.즉,쿼드트리구조에서 2NX2N크기를가지는하나의노드는 NXN 크기를가지는네 개의노드들로분할될수있다.본명세서에서 쿼드트리는
4진 (quaternary)트리로도지칭될수있다.쿼드트리분할은재귀적으로수행될 수있으며,모든노드들이동일한깊이로분할될필요는없다.
[62] 한편,전술한쿼드트리의 리프노드 (leaf node)는멀티-타입트리 (Multi-Type Tree, MTT)구조로더욱분할될수있다.본발명의실시예에 따르면,멀티타입 트리구조에서는하나의 노드가수평또는수직분할의 2진 (binary,바이너리) 또는 3진 (ternary,터너리)트리구조로분할될수있다.즉,멀티-타입트리 구조에는수직바이너리분할,수평바이너리분할,수직터너리분할및수평 터너리분할의 4가지분할구조가존재한다.본발명의실시예에따르면,상기 각 트리구조에서노드의 너비 및높이는모두 2의거듭제곱값을가질수있다. 예를들어 ,바이너리트리 (Binary Tree, BT)구조에서 , 2NX2N크기의 노드는수직 바이너리분할에의해 2개의 NX2N노드들로분할되고,수평바이너리분할에 의해 2개의 2NXN노드들로분할될수있다.또한,터너리트리 (Ternary Tree, TT) 구조에서, 2NX2N크기의 노드는수직 터너리분할에 의해 (N/2)X2N, NX2N및 (N/2)X2N의노드들로분할되고,수평터너리분할에의해 2NX(N/2), 2NXN및 2NX(N/2)의노드들로분할될수있다.이러한멀티-타입트리분할은재귀적으로 수행될수있다.
[63] 멀티-타입트리의 리프노드는코딩유닛이될수있다.코딩유닛에 대한
분할이지시되지 않거나코딩유닛이 최대변환길이에비해크지 않은경우, 해당코딩유닛은더 이상의분할없이 예측및변환의단위로사용된다.한편, 전술한쿼드트리 및멀티-타입트리에서다음의 파라메터들중적어도하나가 사전에 정의되거나 PPS, SPS, VPS등과같은상위 레벨세트의 RBSP를통해 전송될수있다. 1) CTU크기 :쿼드트리의루트노드 (root node)크기 , 2)최소 QT 크기 (MinQtSize):허용된최소 QT리프노드크기 , 3)최대 BT크기 (MaxBtSize): 허용된최대 BT루트노드크기 , 4)최대 TT크기 (MaxTtSize):허용된최대 TT 루트노드크기, 5)최대 MTT깊이 (MaxMttDepth): QT의 리프노드로부터의 MTT
2020/175965 1»(:1^1{2020/002920
12 분할의최대허용깊이 , 6)최소 BT크기 (MinBtSize):허용된최소 BT리프노드 크기 , 7)최소 TT크기 (MinTtSize):허용된최소 TT리프노드크기 .
[64] 도 4는쿼드트리및멀티-타입트리의분할을시그널링하는방법의일
실시예를도시한다.전술한쿼드트리및멀티-타입트리의분할을시그널링하기 위해기설정된플래그들이사용될수있다.도 4를참조하면,쿼드트리노드의 분할여부를지시하는플래그’qt_split_flag’,멀티-타입트리노드의분할여부를 지시하는플래그’mtt_split_flag’,멀티-타입트리노드의분할방향을지시하는 늘래그 'mtt_split_vertical_flag'또는멀티-타입트리노드의분할모양을지시하는 늘래그 ,mtt_split_binary_flag,중적어도하나가사용될수있다.
[65] 본발명의실시예에따르면,코딩트리유닛은쿼드트리의루트노드이며,쿼드 트리구조로우선분할될수있다.쿼드트리구조에서는각각의노드 'QT_node' 별로,qt_split_flag,가시그널링된다.,qt_split_flag,의값이 1일경우해당노드는
4개의정사각형노드들로분할되며,’qt_split_flag’의값이 0일경우해당노드는 쿼드트리의리프노드’QT_leaf_node’가된다.
[66] 각각의쿼드트리리프노드’QT_leaf_node’는멀티-타입트리구조로더분할될 수있다.멀티-타입트리구조에서는각각의노드 'MTT_node'별로
'mtt_split_flag'가시그널링된다. 'mtt_split_flag'의값이 1일경우해당노드는 복수의직사각형노드들로분할되며,’ mtt_split_flag’의값이 0일경우해당노드는 멀티-타입트리의리프노드’MTT_leaf_node가된다.멀티-타입트리노드 ’MTT_node’가복수의직사각형노드들로분할될경우 (즉,’ mtt_split_flag’의값이
1일경우),노드’MTT_node’를위한’mtt_split_vertical_flag’및
'mtt_split_binary_flag’가주가로시그널링될수있다.’mtt_split_vertical_flag’의 값이 1일경우노드’ MTT_node’의수직분할이지시되며,
’mtt_split_vertical_flag’의값이 0일경우노드’MTT_node’의수평분할이 지시된다.또한, 'mtt_split_binary_flag'의값이 1일경우노드’MTT_node'는 2개의 직사각형노드들로분할되며,’ mtt_split_binary_flag’의값이 0일경우노드 ,MTT_node’는 3개의직사각형노드들로분할된다.
[67] 코딩을위한픽쳐예측 (모션보상)은더이상나누어지지않는코딩유닛 (즉 코딩유닛트리의리프노드)을대상으로이루어진다.이러한예측을수행하는 기본단위를이하에서는예즉유닛 (prediction unit)또는예즉블록 (prediction block)이라고한다.
[68] 이하,본명세서에서사용되는유닛이라는용어는예측을수행하는기본
단위인상기예측유닛을대체하는용어로사용될수있다.다만,본발명이이에 한정되는것은아니며,더욱광의적으로는상기코딩유닛을포함하는개념으로 이해될수있다.
[69] 도 5및도 6은본발명의실시예에따른인트라예측방법을더욱구체적으로 도시한다.전술한바와같이,인트라예측부는현재블록의좌측및/또는상측에 위치한복원된샘플들을참조샘플들로이용하여현재블록의샘플값들을
2020/175965 1»(:1^1{2020/002920
13 예측한다.
이 먼저,도 5는인트라예측모드에서현재블록의예측을위해사용되는참조 샘플들의일실시예를도시한다.일실시예에따르면,참조샘플들은현재 블록의좌측경계에인접한샘플들및/또는상측경계에인접한샘플들일수 있다.도 5에도시된바와같이 ,현재블록의크기가 \¥ 11이고현재블록에 인접한단일참조라인(1 句의샘플들이인트라예측에사용될경우,현재블록의 좌측및/또는상측에위치한최대 2 +211+1개의주변샘플들을사용하여참조 샘늘들이설정될수있다.
1] 또한,참조샘플로사용될적어도일부의샘플이아직복원되지않은경우, 인트라예측부는참조샘플패딩과정을수행하여참조샘플을획득할수있다. 또한,인트라예측부는인트라예측의오차를줄이기위해참조샘플필터링 과정을수행할수있다.즉,주변샘플들및/또는참조샘플패딩과정에의해 획득된참조샘플들에필터링을수행하여필터링된참조샘플들을획득할수 있다.인트라예측부는이와같이획득된참조샘플들을이용하여현재블록의 샘플들을예측한다.인트라예측부는필터링되지않은참조샘플들또는 필터링된참조샘플들을이용하여현재블록의샘플들을예측한다.본개시에서, 주변샘플들은적어도하나의참조라인상의샘플들을포함할수있다.예를 들어,주변샘플들은현재블록의경계에인접한라인상의인접샘플들을 포함할수있다.
2] 다음으로,도 6은인트라예측에사용되는예측모드들의일실시예를
도시한다.인트라예측을위해,인트라예측방향을지시하는인트라예측모드 정보가시그널링될수있다.인트라예측모드정보는인트라예측모드세트를 구성하는복수의인트라예측모드들중어느하나를지시한다.현재블록이 인트라예측블록일경우,디코더는비트스트림으로부터현재블록의인트라 예측모드정보를수신한다.디코더의인트라예측부는추출된인트라예측모드 정보에기초하여현재블록에대한인트라예측을수행한다.
3] 본발명의실시예에따르면,인트라예측모드세트는인트라예측에사용되는 모든인트라예측모드들(예,총 67개의인트라예측모드들)을포함할수있다. 더욱구체적으로,인트라예측모드세트는평면모드, IX:모드및복수의(예,
65개의)각도모드들(즉,방향모드들)을포함할수있다.각각의인트라예측 모드는기설정된인덱스(즉,인트라예측모드인덱스)를통해지시될수있다. 예를들어,도 6에도시된바와같이인트라예측모드인덱스 0은평면모드를 지시하고,인트라예측모드인덱스 1은1 :모드를지시한다.또한,인트라예측 모드인덱스 2내지 66은서로다른각도모드들을각각지시할수있다.각도 모드들은기설정된각도범위이내의서로다른각도들을각각지시한다.예를 들어,각도모드는시계방향으로 45도에서 -135도사이의각도범위(즉,제 1각도 범위)이내의각도를지시할수있다.상기각도모드는 12시방향을기준으로 정의될수있다.이때,인트라예측모드인덱스 2는수평대각田03 01 1
2020/175965 1»(:1^1{2020/002920
14
Diagonal, HDIA)모드를지시하고,인트라예측모드인덱스 18은
수평 (Horizontal, HOR)모드를지시하고,인트라예측모드인덱스 34는 대각 (Diagonal, DIA)모드를지시하고,인트라예측모드인덱스 50은
수직 (Vertical, VER)모드를지시하며,인트라예측모드인덱스 66은수직 대각 (Vertical Diagonal, VDIA)모드를지시한다.
4] 이하,도 7을참조하여본발명의일실시예에따른인터예측방법에대해
설명하도록한다.본개시에서 ,인터 예즉방법은병진운동 (translation motion)에 최적화된일반인터 예측방법및어파인 (affine)모델기반의인터 예측방법을 포함할수있다.또한,모션벡터는일반인터예측방법에따른모션보상을위한 일반모션벡터및어파인모션보상을위한컨트롤포인트모션벡터 (control point motion vector)중적어도하나를포함할수있다.
5] 도 7은본발명의일실시예에따른인터 예측방법을도시한다.전술한바와 같이,디코더는복호화된다른픽쳐의복원된샘플들을참조하여현재블록을 예측할수있다.도 7을참조하면,디코더는현재블록 (701)의모션정보세트에 기초하여참조픽쳐 (720)내의참조블록 (702)을획득한다.이때,모션정보 세트는참조픽쳐인덱스및모션벡터를포함할수있다.참조픽쳐인덱스는 참조픽쳐리스트에서현재블록의인터예측을위한참조블록이포함된참조 픽쳐 (720)를지시한다.일실시예에따라,참조픽쳐리스트는전술한 L0픽쳐 리스트또는 L1픽쳐리스트중적어도하나를포함할수있다.모션벡터는현재 픽쳐 (기 0)내에서현재블록 (701)의좌표값과참조픽쳐 (720)내에서참조 블록 (702)의좌표값간의오프셋을나타낸다.디코더는참조블록 (702)의샘플 값들에기초하여현재블록 (701)의 예측자를획득하고,상기 예측자를이용하여 현재블록 (701)을복원한다.
6] 구체적으로,인코더는복원순서가앞선픽쳐들에서현재블록과유사한
블록을탐색하여전술한참조블록을획득할수있다.예를들어 ,인코더는기 설정된탐색영역내에서현재블록과샘플값차이의합이최소가되는참조 블록을탐색할수있다.이때,현재블록과참조블록의샘플들간의유사도를 즉정하기위해 , SAD (Sum Of Absolute Difference)또는 SATD (Sum of Hadamard Transformed Difference)중적어도하나가사용될수있다.여기에서 , SAD는두 블록에포함된샘플값들의차이각각의절대값을모두더한값일수있다.또한, SATD는두블록에포함된샘플값들의차이를하다마드변환 (Hadamard
Transform)하여획득된하다마드변환계수의절대값을모두더한값일수있다.7] 한편,현재블록은하나이상의참조영역을이용하여예측될수도있다.전술한 바와같이,현재블록은 2개이상의참조영역을이용하는쌍예측방식을통해 인터 예측될수있다.일실시예에따라,디코더는현재블록의 2개의모션정보 세트에기초하여 2개의참조블록을획득할수있다.또한,디코더는획득된 2개의참조블록각각의샘플값들에기초하여현재블록의제 1예측자및제 2 예측자를획득할수있다.또한,디코더는제 1예측자및제 2예측자를이용하여
2020/175965 1»(:1^1{2020/002920
15 현재블록을복원할수있다.예를들어,디코더는제 1예측자및제 2예측자의 샘플별평균에 기초하여 현재블록을복원할수있다.
8] 전술한바와같이,현재블록의모션보상을위해,하나이상의모션정보
세트가시그널링될수있다.이때,복수의블록각각의모션보상을위한모션 정보세트간의유사성이 이용될수있다.예를들어,현재블록의 예측에 사용되는모션정보세트는기복원된다른샘플들중어느하나의 예측에 사용된모션정보세트로부터유도될수있다.이를통해,인코더 및디코더는 시그널링오버헤드를감소시킬수있다.
9] 예를들어,현재블록의모션정보세트와동일또는유사한모션정보세트에 기초하여 예측되었을가능성이 있는복수의후보블록들이존재할수있다. 디코더는해당복수의후보블록들을기초로머지후보리스트 (merge candidate list)를생성할수있다.여기에서,머지후보리스트는현재블록보다먼저복원된 샘플들중에서,현재블록의모션정보세트와관련된모션정보세트에 기초하여 예측되었을가능성이 있는샘플에 대응하는후보들을포함할수있다. 인코더와디코더는미리 정의된규칙에 따라현재블록의머지후보리스트를 구성할수있다.이때 ,인코더와디코더가각각구성한머지후보리스트는서로 동일할수있다.예를들어,인코더 및디코더는현재픽쳐내에서 현재블록의 위치에 기초하여 현재블록의머지후보리스트를구성할수있다.인코더 및 디코더가현재블록의 머지후보리스트를구성하는방법에 대해서는도 9를 통해후술하도록한다.본개시에서,특정블록의 위치는특정블록을포함하는 픽쳐 내에서특정블록의좌상단 (top-left)샘플의상대적인위치를나타낸다.
[8이 한편,코딩효율을높이기위하여 전술한레지듀얼신호를그대로코딩하는 것이 아니라,레지듀얼신호를변환하여 획득된변환계수값을양자화하고, 양자화된변환계수를코딩하는방법이사용될수있다.전술한바와같이 , 변환부는레지듀얼신호를변환하여 변환계수값을획득할수있다.이때,특정 블록의 레지듀얼신호는현재블록의 전영역에분산되어 있을수있다.이에 따라,레지듀얼신호에 대한주파수영역 변환을통해 저주파영역에 에너지를 집중시켜코딩효율을향상시킬수있다.이하에서는,레지듀얼신호가변환 또는역변환되는방법에 대해구체적으로설명하도록한다.
[81] 도 8은인코더가레지듀얼신호를변환하는방법을구체적으로나타내는
도면이다.전술한바와같이,공간영역의 레지듀얼신호는주파수영역으로 변환될수있다.인코더는획득된레지듀얼신호를변환하여 변환계수를획득할 수있다.먼저,인코더는현재블록에 대한레지듀얼신호를포함하는적어도 하나의 레지듀얼블록을획득할수있다.레지듀얼블록은현재블록또는현재 블록으로부터분할된블록들중어느하나일수있다.본개시에서,레지듀얼 블록은현재블록의 레지듀얼샘플들을포함하는레지듀얼어레이 (array)또는 레지듀얼매트릭스 (matrix)로지칭될수있다.또한,본개시에서 레지듀얼 블록은변환유닛또는변환블록의크기와동일한크기의블록을나타낼수
2020/175965 1»(:1^1{2020/002920
16 있다.
[82] 다음으로,인코더는변환커널을사용하여 레지듀얼블록을변환할수있다. 레지듀얼블록에 대한변환에사용되는변환커널은수직 변환및수평 변환의 분리 가능한특성을가지는변환커널일수있다.이경우,레지듀얼블록에 대한 변환은수직 변환및수평 변환으로분리되어수행될수있다.예를들어, 인코더는레지듀얼블록의수직방향으로변환커널을적용하여수직 변환을 수행할수있다.또한,인코더는레지듀얼블록의수평방향으로변환커널을 적용하여수평 변환을수행할수있다.본개시에서,변환커널은변환매트릭스, 변환어레이,변환함수,변환과같이 레지듀얼신호의 변환에사용되는 파라미터 세트를지칭하는용어로사용될수있다.일실시예에따라,변환 커널은복수의사용가능한커널들중어느하나일수있다.또한,수직 변환및 수평 변환각각에 대해서로다른변환타입에기반한변환커널이사용될수도 있다.복수의사용가능한변환커널들중어느하나가선택되는방법에 대해서는도 12내지도 26을통해후술하도록한다.
[83] 인코더는레지듀얼블록으로부터 변환된변환블록을양자화부로전달하여 양자화할수있다.이때,변환블록은복수의 변환계수들을포함할수있다. 구체적으로,변환블록은 2차원배열된복수의 변환계수들로구성될수있다. 변환블록의크기는레지듀얼블록과마찬가지로현재블록또는현재
블록으로부터분할된블록중어느하나와동일할수있다.양자화부로전달된 변환계수들은양자화된값으로표현될수있다.
지칭되고,주가적인변환은 2차변환 이피산때)/ 1대118&)]'111)으로지칭될수있다.
2차변환은레지듀얼블록별로선택적일수있다.일실시예에 따라,인코더는 1차변환만으로저주파영역에 에너지를집중시키기 어려운영역에 대해 2차 변환을수행하여코딩 효율을향상시킬수있다.예를들어,레지듀얼값들이 레지듀얼블록의수평또는수직 방향이외의 방향에서크게나타나는블록에 대해 2차변환이추가될수있다.인트라예측된블록의 레지듀얼값들은인터 예측된블록의 레지듀얼값들에비해수평또는수직방향이외의방향으로 변화할확률이높을수있다.이에 따라,인코더는인트라예측된블록의 레지듀얼신호에 대해 2차변환을추가적으로수행할수있다.또한,인코더는 인터 예측된블록의 레지듀얼신호에 대해 2차변환을생략할수있다.
[85] 다른예로,현재블록또는레지듀얼블록의크기에따라, 2차변환수행 여부가 결정될수있다.또한,현재블록또는레지듀얼블록의크기에따라크기가서로 다른변환커널이사용될수있다.예를들어,너비또는높이중짧은변의 길이가제 1기 설정된길이보다크거나같은블록에 대해서는 8X8 2차변환이 적용될수있다.또한,너비또는높이중짧은변의길이가제 2기 설정된길이 보다크거나같고제 1기 설정된길이보다작은블록에 대해서는 4X4 2차
2020/175965 1»(:1^1{2020/002920
17 변환이적용될수있다.이때,제 1기설정된길이는제 2기설정된길이보다큰 값일수있으나,본개시가이에제한되는것은아니다.또한, 2차변환은 1차 변환과달리수직변환및수평변환으로분리되어수행되지않을수있다.
이러한 2차변환은저대역비 -분리변환 (Low Frequency Non-Separable Transform, LFNST)으로지칭될수있다.
[86] 또한,특정영역의비디오신호의경우,급격한밝기변화로인해주파수변환을 수행하여도고주파대역에너지가줄어들지않을수있다.이에따라,양자화에 의한압축성능이저하될수있다.또한,레지듀얼값이드물게존재하는영역에 대해변환을수행하는경우,인코딩시간및디코딩시간이불필요하게증가할 수있다.이에따라,특정영역의레지듀얼신호에대한변환은생략될수있다. 특정영역의레지듀얼신호에대한변환수행여부는특정영역의변환과관련된 신택스요소에의해결정될수있다.예를들어,상기신택스요소는변환스킵 정보 (transform skip information)를포함할수있다.변환스킵정보는변환스킵 플래그 (transform skip flag)일수있다.레지듀얼블록에대한변환스킵정보가 변환스킵을나타내는경우,해당레지듀얼블록에대한변환이수행되지 않는다.이경우,인코더는해당영역의변환이수행되지않은레지듀얼신호를 곧바로양자화할수있다.도 8을참조하여설명된인코더의동작들은도 1의 변환부를통해수행될수있다.
[87] 전술한변환관련신택스요소들은비디오신호비트스트림으로부터파싱된 정보일수있다.디코더는비디오신호비트스트림을엔트로피디코딩하여변환 관련신택스요소들을획득할수있다.또한,인코더는변환관련신택스 요소들을엔트로피코딩하여비디오신호비트스트림을생성할수있다.
[88] 도 9은인코더및디코더가변환계수를역변환하여레지듀얼신호를획득하는 방법을구체적으로나타내는도면이다.이하설명의편의를위해,인코더및 디코더각각의역변환부를통해역변환동작이수행되는것으로설명한다.
역변환부는역양자화된변환계수를역변환하여레지듀얼신호를획득할수 있다.먼저,역변환부는특정영역의변환관련신택스요소로부터해당영역에 대한역변환이수행되는지검출할수있다.일실시예에따라,특정변환블록에 대한변환관련신택스요소가변환스킵을나타내는경우,해당변환블록에 대한변환이생략될수있다.이경우,변환블록에대해전술한 1차역변환및 2차역변환이모두생략될수있다.또한,역양자화된변환계수는레지듀얼 신호로사용될수있다.예를들어 ,디코더는역양자화된변환계수를레지듀얼 신호로사용하여현재블록을복원할수있다.
[89] 다른일실시예에따라,특정변환블록에대한변환관련신택스요소가변환 스킵을나타내지않을수있다.이경우,역변환부는 2차변환에대한 2차역변환 수행여부를결정할수있다.예를들어,변환블록이인트라예측된블록의변환 블록인경우,변환블록에대한 2차역변환이수행될수있다.또한,변환블록에 대응하는인트라예측모드에기초하여해당변환블록에사용되는 2차변환
2020/175965 1»(:1^1{2020/002920
18 커널이결정될수있다.다른예로,변환블록의크기에기초하여 2차역변환 수행여부가결정될수도있다. 2차역변환은역양자화과정이후 1차역변환이 수행되기전에수행될수있다.
[9이 역변환부는역양자화된변환계수또는 2차역변환된변환계수에대한 1차 역변환을수행할수있다. 1차역변환의경우, 1차변환과마찬가지로수직변환 및수평변환으로분리되어수행될수있다.예를들어,역변환부는변환블록에 대한수직역변환및수평역변환을수행하여레지듀얼블록을획득할수있다. 역변환부는변환블록의변환에사용된변환커널에기초하여변환블록을 역변환할수있다.예를들어,인코더는복수의사용가능한변환커널들중현재 변환블록에적용된변환커널을지시하는정보를명시적또는묵시적으로 시그널링할수있다.디코더는시그널링된변환커널을나타내는정보를 이용하여복수의사용가능한변환커널들중변환블록의역변환에사용될변환 커널을선택할수있다.역변환부는변환계수에대한역변환을통해획득된 레지듀얼신호를이용하여현재블록을복원할수있다.
[91] 도 W은본발명의일실시예에따른코딩블록이복수개의변환블록으로
분할되는경우인트라예측모드의적용방법을설명하기위한도면이다.본 발명의일실시예에따르면,인트라예측모드는코딩유닛(또는코딩
블록)(이하,블록으로약칭될수있음)단위로결정될수있다.그리고,코딩 유닛은복수개의변환블록으로분할될수있다.일실시예로서,인트라예측 모드는코딩블록의형태에기초하여수정(또는해석,결정,개선)될수있다.
[92] 일실시예에서 ,정방형(또는정사각형)블록이아닌경우인트라예측모드를 재해석하는방법을설명한다.도 W을참조하면, nTbW는변환블록의넓이 , nTbH는변환블록의높이를나타내는변수일수있다.또는, nTbW는코딩 블록의넓이, nTbH는코딩블록의높이를나타내는변수일수있다.또는, ISP(Intra subpartitions)가적용된블록에서 nTbW는코딩블록의넓이 , nTbH는 코딩블록의높이를나타내는변수일수있다.또한,본발명에서, whRatio는 너비와높이의비율을나타내는변수이다.일예로서, whRatio는
Abs(Log2(nTbW/nTbH))로설정(또는정의)될수있다.이하,인코더로부터 디코더로시그널링된인트라예측모드는제 1예측모드(또는제 1인트라예측 모드)로지칭되고,수정된(또는재해석된,결정된,개선된)모드는제 2예측 모드(또는제 2인트라예측모드)로지칭될수있다.이하에서, abs()는절대값을 취하는연산자(또는함수)를나타낸다.수정된인트라예측모드는아래의 조건들에기초하여유도될수있다.
[93] -제 1조건 : nTbW > nTbH
[94] -제 2조건:제 1예측모드가 2보다크거나같은지
[95] -제 3조건:제 1예측모드가 whRatio >1인경우(8+2*whRatio)보다작고,
whRatio <1인경우 8보다작은지
[96] 디코더는상기제 1내지 3의 3가지조건을만족하면 wideAngle을 1로설정하고,
2020/175965 1»(:1^1{2020/002920
광각모드가이용되는지여부를지시하는변수이다.
[98] -제 5조건:제 1예측모드가 66보다작거나같은지
[99] -제 6조건:제 1예측모드가 \¥1況止0>1인경우(60-2*\¥1況止0)보다크고, \vhRatio <1인경우 60보다큰지
설정하고,제 2예측모드를(제 1예측모드 - 67)로설정할수있다.
[101] 본발명의일실시예에따르면,인트라예측모드는기본각도모드와확장된 각도모드로구분될수있다.기본각도모드는수직모드/수평모드기준 +- 45 범위내의각도모드들일수있고,확장된각도모도는수직모드/수평모드 기준으로 +- 45도를초과하는각도모드일수있다.따라서,시그널링된모드 정보는코딩블록의형태에따라기본각도모드를사용하거나확장된각도 모드를사용될수있다.확장된각도모드는코딩블록의형태에기반한가로와 세로의비율(또는세로와가로의비율)에따라사용할수있는모드수가정의될 수있다.일예로서,상기비율은 2: 1, 4: 1, 8: 1, 16: 1등으로정의(또는설정)될수 있다.
[102] 10如에도시된바와같이,인트라예측모드 2로결정된此\¥ 出
블록은도 10(비에도시된바와같이,수평방향 2개의변환 블록으로분할될수있다. «5\¥넌110이 1½¾ !10보다크다고가정하면
는코딩블록의높이를나타내는변수이다. *\¥년 는변환블록의 너비를나타내는변수이고, 1½½ 는변환블록의높이를나타내는변수이다.
[103] 이때,시그널링된인트라예측모드 2는제 1변환블록의형태를기반으로 재해석되어확장된각도모드로수정될수있고,상술한해석방법에따라 (2+65)가되어제 2예측모드는 67로유도(또는결정)될수있다.즉,코딩블록 단위에서결정된인트라예측모드는변환블록단위에서도동일하게사용되지 않을수있다.이경우,이로인한성능변화도야기될수있다.
[104] 따라서,본발명의실시예에따르면,코딩블록에결정된인트라예측모드를 변환블록에서도동일하게적용하기위해광각모드를결정하는방법을다음과 같이적용하는방법을제안한다.실시예로서,인코더/디코더는제 2예측모드를 유도함에 있어서, 및 를코딩블록의 산5¾¾ 로설정할수 있다.인코더/디코더는변환블록이포함된코딩블록의높이및너비를 이용하여광각모드의사용여부를판단(또는결정)할수있다.도 에서,수평 직사각형형태로변환블록이분할되는경우를예로들었으나본발명이이에 한정되는것은아니다.즉,수직직사각형,정사각형또는조합된여러형태로 분할되는경우에도동일하게제안하는실시예가적용될수있다.
2020/175965 1»(:1/10公020/002920
20
[105] 도 11은본발명의일실시예에따른 PDPC(po sition-dependent intra prediction combination)적용방법을설명하기위한도면이다. PDPC는다음과같은조건을 모두만족하면인트라블록에적용될수있다.
[106] 조건 1. IntraSubPartitionsSplitType(ISP분할타입)이 ISP_NO_SPLIT이거나
cldx(컴포넌트인덱스)가 0과같지않음
[107] 조건 2. refldx(참조샘플라인인덱스)가 0과같거나 cldx가 0과같지않음
[108] 조건 3.다음조건들중하나라도해당되는경우:
[109] - predModelntra(인트라예측모드)가 INTRA_PLANAR
[110] - predModeIntra(인트라예즉모드)가 INTR A_DC
[111] - predModeIntra(인트라예측모드)가 INTRA_ANGULAR 18
[112] - predModelntra(인트라예측모드)가 INTRA_ANGULAR50
[113] - predModelntra(인트라예측모드)가 INTRA_ANGULAR10보다작거나같은 경우
[114] - predModeIntra(인트라예측모드)가 INTRA_ANGULAR58과같거나큰경우
[115] 본발명의일실시예에따르면, PDPC동작은이하에서설명하는방법에따라 적용될수있다.본발명에서설명하는 PDPC는그명칭에제한되는것이아니며, 상시 PDPC는위치의존적인인트라예즉샘늘필터링(Position-dependent intra prediction sample filtering)으로지칭될수도있다.
[116] 일실시예로서,(X, y)위치의 예측샘플 pred(x, 은다음의수학식 1과같이 인트라예측모드(예컨대, DC,플래너,방향성모드)및 PDPC에따른참조 샘플의선형조합을이용하여예측될수있다.
X 바 ;/) + 32) > > 6
[118] 여기서,요니및요 는각각현재샘플江 의좌측및상측에위치한참조
샘플을나타내고 , 11ᆻ는현재블록의좌상단 애-노的에위치한참조샘플을 나타낸다.현재블록에 IX:모드가적용되는경우,다음의수학식 2에기초하여 가중치 이 가중치로지칭될수있음)가계산될수있다.
[119] [수식 2]
wT = 32 > > ( ( y< < 1 ) > > shift ), wL = 32 > > ( ( x< < 1 ) > > shift ), wTL =
( wL> >4) +( wT> >4)
[12이 수학식 2에서 , shift는( log2( width)- 2 + log2( height) - 2 + 2) » 2로설정될수 있다.그리고,플래너모드에대하여 wTL = 0로설정될수있고,수평모드에 대하여 wTL = wT로설정될수있고,수직모드에대하여 wTL = wL으로설정될 수있다. PDPC가중치는합산및시프트연산에기초하여서만계산될수있다.
2020/175965 1»(:1/10公020/002920
21 pred(x, y)값은앞서설명한수학식 1을이용하여단일단계로계산될수있다.
[121] 만약, PDPC가 DC,플래너,수평및/또는수직모드에적용되는경우,추가적인 경계필터링이요구되지않을수있다.일예로서 ,상기추가적인경계필터링은 종래의영상압축기술 (예컨대, HEVC)의 DC모드경계필터또는수평/수직 모드의 옛지필터를포함할수있다.
[122] 도 12는본발명의일실시예로서인트라예측모드에따라 PDPC에이용되는 참조샘플을예시하는도면이다.도 12를참조하면,도 12(a)는인트라예측 모드가예측모드 2인경우를가정하고,도 12(b)는인트라예측모드가예측모드 66인경우를가정한다.구체적으로,도 12는우상측대각모드 (top-right diagonal mode)에 PDPC가적용되는경우,참조샘플 R니, R _1y및 R서를도시한다.예측 샘플 pred(x’, y’)은예측블록내 (x’, y’)에위치한예측샘플을나타낸다.참조샘플 R니의좌표 는 = ’ + / +1로주어진다.그리고,참조샘플11_1,)의좌표 는 유사하게 y = X1 + y' + 1로주어진다.
[123] 본발명의일실시예에따르면,우상측대각모드에대한 PDPC가중치는
다음의수학식 3에따라결정될수있다.
[124] [수식 3] wT = 16 >> ( (y’<<1 ) >> shift ), wL = 16 >> ( ( x’<<1 ) >> shift ), wTL =
0
[125] 도 13은본발명의일실시예로서인트라예측모드에따라 PDPC에이용되는 참조샘플을예시하는도면이다.도 13을참조하면,도 13(a)는인트라예측 모드의모드번호 (또는모드인덱스)가 3내지 10중어느하나인경우를 가정하고,도 13(b)는인트라예측모드의모드번호가 58내지 65중어느하나인 경우를가정한다.앞서설명한도 12와유사하게도 13은좌하측대각
모드 (bottom-left diagonal mode)에 PDPC가적용되는경우,참조샘늘 R니, R _1y 및 R서를도시한다.예측샘플 Pred(x’, y’)은예측블록내 (x', y’)에위치한예측 샘플을나타낸다.참조샘플 Ru의좌표 \는 = ’ + / + 1로주어진다.그리고, 참조샘플 R _1y의좌표 y는유사하게 y = X’ + y’ + 1로주어진다.
[126] 본발명의일실시예에따르면,좌하측대각모드에대한 PDPC가중치는
다음의수학식 4에따라결정될수있다.
[127] [수식 4]
wT = 16 >> ( (y'<<1 ) >> shift ), wL = 16 >> ( ( x’<<1 ) >> shift ), wTL =
0
[128] 우상측대각모드에대한도 13(a)의경우, PDPC가중치는다음의수학식 5와 같이정의될수있다.
[129]
2020/175965 1»(:1/10公020/002920
22
[수식 5]
wT = 32 > > ( ( y’< < 1 ) > > shift ), wL = 0, wTL = 0
[13이 마찬가지로,좌하측대각모드에대한도 13(b)의경우, PDPC가중치는다음의 수학식 6과같이정의될수있다.
[131] [수식 6]
wL = 32 > > ( ( x’< < 1 ) > > shift ), wT =0, wTL = 0
[132] 도 12및도 13을참조하면,실시예로서, DC,플래너,수평및/또는수직모드에 대하여 PDPC가적용되는경우와같이,대각모드및도 13에도시된대각모드의 인접모드에대해서는추가적인경계필터링이필요하지않을수있다.
[133] 도 13에도시된예시의참조샘플좌표는방향성모드인트라예측에대하여 정의된테이블에기초하여유도될수있다.상술한바와같이,대각선및그인접 모드로테이블이정의될수있기때문에본발명에서설명하는 PDPC구현에 따른추가테이블이필요하지않는다는장점을가진다.또한,좌표 와 를 계산할때는곱셈연산이사용되지않을수있다.또한,일실시예에서,분수기준 샘플좌표가이용되는경우참조샘플에대하여선형보간이수행될수있다.
[134] 도 14는본발명이적용되는일실시예에따른코딩블록에대한 ISP(Intra
subpartitions)및 PDPC(position-dependent intra prediction combination)적용방법을 예시하는도면이다.도 14(a)를참조하면,현재코딩블록은너비 (W)및
높이 (H)를이용해 W x H로표현될수있다.도 14(b)는 ISP모드가적용되는경우, 수직방향으로현재코딩블록이 4개의변환블록으로분할된예를나타낸다. 그리고,도 14(c)는도 14(b)에서분할된각각의변환블록단위로 PDPC를 적용하는예를도시한다.
[135] 일실시예에서,인코더/디코더는도 14(b)에서 ISP가적용된변환블록
단위에서보간필터 (Interpolation filter)를사용 (또는적용)할수있다.보간필터는 참조샘플로부터샘플값을취득하는방법을나타낸다.일예로서,
인코더 /디코더는필터늘래그가 0이면큐빅보간필터 (Cubic interpolation filter) 계수를사용하고, 1이면가우시안보간필터 (Gaussian interpolation filter)계수를 사용할수있다.인코더/디코더는결정된보간필터계수를이용하여참조샘플 값을결정하고이값을예측값으로사용할수있다.또한,일실시예에서, 인코더/디코더는변환블록이루마컴포넌트이고 KP가적용된블록에대하여 필터플래그를 1로설정할수있다.다른일예로,인코더/디코더는필터
플래그에기초하여보간필터의적용여부를결정할수도있다.
[136] 또한,일실시예에서,인코더/디코더는루마컴포넌트이고 KP가적용된변환 블록에대하여,블록의넓이 (W)및높이 (H)값에기초하여필터플래그값을 설정 (또는결정)할수있다.일실시예로서,인코더/디코더는블록의샘플 수 (W*H)와미리정의된 (또는미리설정된)특정기준값을비교하여플래그
2020/175965 1»(:1^1{2020/002920
23 값을설정할수있다.
인코더/디코더는블록너비및높이를각각기준값과비교하여필터플래그 값을다르게설정할수있다.일실시예로서,필터플래그값을결정하는조건은 (\¥ >기준값및 >기준값),(\¥ >기준값또는 >기준값)으로정의될수 있다.상기 예에서부등호는기준값보다크다에한정되지않으며,같은경우, 크거나같은경우,작은경우,작거나같은경우로정의될수도있다. 및 에 적용되는기준값은서로다를수도있고서로다른부등호가적용될수있다. 또는,특정블록크기범위내에속하는지여부에따라필터플래그값이설정될 수있다.
[137] 도 14( 를참조하면,본발명의일실시예에서,인코더/디코더는犯!5가적용된 블록에대하여모 (:를적용할수있다.일예로서,인코더/디코더는 !5가 적용된블록에대하여블록의
기초하여
적용여부를 결정할수있다.일실시예에서,블록의샘플수에기초하여모。!5(:의적용여부를 결정하기위한조건이정의될수있다.예를들어,상기조건은
=¾ >=기준값, =¾ <기준값, =¾ <=기준값등이될수있다.상기기준 값은기설정된값일수있다.다른일실시예로,모이5(:의적용여부를결정하기 위한조건은(
)으로정의되거나,(\¥ >기준값또는
>기준값)으로정의될수있다.이때,상기예에서부등호는기준값보다크다에 한정되지않으며 , 우,작은경우,작거나같은경우로 정의될수도있다.
여부를결정하기위한조건은(\¥ > 기준값및 ³기준값)으로정의되
)으로 정의될수있다.실시예로서 , 와 에적용되는기준값은동일한값으로 정의될수도있고,서로다른값으로정의될수도있으며,동일한부호(또는 부등호)로정의될수도있고,서로다른부호로정의될수도있다.
[138] 또한,도 14( 를참조하면,본발명의일실시예에서,犯!5가적용된블록은복수 개의직사각형변환블록으로분할되고,분할된변환블록단위로
인코딩/디코딩되는과정이수행될수있다.일실시예에서,인코더/디코더는 모01(를적용함에있어서변환블록단위가아닌코딩블록단위로적용할수도 있다.즉,인코더/디코더는 !5가적용된코딩블록에대해모이5(:는변환블록 단위대신코딩블록단위로수행할수있다.
[139] 도 14( 를참조하면,본발명의일실시예에서,犯!5가적용된블록에모이5(:를 적용함에 있어서,인코더/디코더는 가적용되는모드들중에서일부 모드들에대하여정의된특정조건을만족하는경우에한하여모01(를적용할 수있다.예를들어,상술한실시예에서와같이,샘플수또는너비/높이에 기반하여 적용여부를결정하는경우,플래너모드,수평모드,수직 모드에대한기준값이다르게설정될수있다.
[14이 또한,본발명의일실시예에따르면,인코더/디코더는참조샘플필터링의
2020/175965 1»(:1^1{2020/002920
24 적용여부를결정함에있어서, ISP및/또는 PDPC가적용된블록인지여부에 기초하여필터링적용여부를지시하는필터플래그를설정할수있다.예를 들어, ISP및 PDPC가적용된블록에대하여필터플래그는 0또는 1의고정된 값으로설정될수있다.또는, ISP및 PDPC가적용된블록에대하여 MDIS(Mode dependent Intra smoothing)조건에의해서필터늘래그값이결정될수있다.또는, 인코더/디코더는 KP및 PDPC가적용된블록의필터플래그값과 KP만적용된 블록의필터플래그값을서로다르게적용할수있다.
[141] 또한,본발명의일실시예에따르면,인코더/디코더는광각모드를결정함에 있어서,코딩블록의너비와높이에기반하여인트라예측모드를재설정할수 있다.예를들어,인코더/디코더는 ISP가적용된블록은코딩블록의너비와 높이에기반하여광각모드의재해석과정을수행하고,이에기초하여참조샘플 필터플래그를설정할수있다.또는,인코더/디코더는 KP가적용된블록의분할 방향/크기에기반하여광각모드적용방법을다르게설정할수있다.일 실시예로서,특정분할방향은변환블록의너비/높이또는코딩블록의 너비/높이에기초하여적용하고,그이외는다른방식으로적용할수있다.다른 실시예로서,인코더/디코더는분할된변환블록의너비와높이값에기반하여 광각모드를적용할수있다.예를들어,변환블록의너비와높이중최소값이 기준값보다크거나같거나크면코딩블록의높이와너비를이용하여광각 모드를적용할수있고기준값보다같거나작거나,작은경우면변환블록의 너비와높이를이용하여광각모드를적용할수있다.또는,이와반대로,변환 블록의너비와높이중최소값이기준값보다크거나같거나크면변환블록의 높이와너비를이용하여광각모드를적용할수있고기준값보다같거나 작거나,작은경우면코딩블록의너비와높이를이용하여광각모드를적용할 수있다.
[142] 이상에서 ,도 W내지도 14에서설명한실시예들은하나이상의실시예가조합 적용될수도있고,독립적으로적용될수도있다.또한,상술한실시예는 디코더와인코더에서실질적으로동일한방법으로적용될수있다.
[143] 도 15는본발명의일실시예에따른변환유닛분할처리방법을설명하기위한 도면이다.도 15를참조하면,본발명의일실시예에서,인코더/디코더는현재 블록 (코딩블록,코딩유닛)을복수개의변환블록으로분할하여
인코딩/디코딩할수있다.실시예로서,인트라서브파티션 (Intra subpartitions, ISP) 모드가적용되는경우,코딩블록은복수개의변환블록으로분할될수있다. 또는,코딩블록의크기가최대변환크기보다큰경우,코딩블록은복수개의 변환블록으로분할될수있다.인트라서브파티션모드가적용되는경우,도 15에도시된바와같이,코딩블록은수평또는수직방향의직사각형변환 블록으로분할될수있고, 2개또는 4개의변환블록으로분할될수있다.
[144] 도 16은본발명이적용되는일실시예에따른 1차변환 (primary transform)및 2차변환 (secondary transform)을거쳐인코딩/디코딩되는프로세스를나타내는
2020/175965 1»(:1^1{2020/002920
25 도면이다.전술한바와같이,코딩블록은복수개의변환블록으로분할될수 있고,분할된변환블록에대하여인코더/디코더는변환을적용할수있다.도 16은변환블록에대하여 2번의변환을적용하는예를나타낸다.도 16의순방향 일차변환(Forward Primary Transform)은인코더즉을기준으로첫번째적용되는 변환을나타내며,본발명에서 1차변환으로지칭될수있다.도 16의순방향 이차변환(Forward Secondary Transform)은인코더즉을기준으로두번째 적용되는변환을나타내며 ,본발명에서 2차변환으로지칭될수있다.디코더 측을기준으로역양자화된변환블록에대하여 2차변환(즉, 2차역변환)및 1차 변환(즉, 1차역변환)이순차적으로수행될수있다.전술한바와같이, 2차 변환은저대역비 -분리변환(Low Frequency Non-Separable Transform,
LFNST)으로지칭될수있다.
[145] 본발명의일실시예에서, 1차변환에이용되는변환매트릭스(또는변환커널, 변환타입)는 DCT-2, DST-7, DCT-8등과같은종래의영상압축기술에서공지된 변환매트릭스일수있다. 2차변환은코딩블록의크기에따라변환블록내 일부영역에적용될수있다.예를들어,상기일부영역은 4x4영역, 8x8영역일 수있다.상기일부영역의위치는코딩블록(또는변환블록)의좌상단영역일수 있다.일실시예로서,코딩블록의너비및높이가모두 4보다크면좌상단 8x8 영역에적용될수있고,너비와높이중어느한변이 4와같으면좌상단 4x4 영역에적용될수있다. 2차변환은인트라모드로코딩된블록의루마성분및 크로마성분에적용될수있다.
[146] 도 17는본발명의일실시예에따른 2차변환에이용되는변환커널을
선택하는방법을설명하기위한도면이다.도 17을참조하면,인트라예측에 사용되는예측모드에기반하여변환커널세트(또는변환타입세트,변환 매트릭스세트)가결정될수있고,도 17에도시된테이블이인코더/디코더에 정의될수있다.본실시예에서,인트라예측모드는 -14 - 83까지정의될수있다. 도 17에도시된바와같이그룹화된인트라예측모드별로변환커널세트가 결정될수있다.루마성분및크로마성분에동일한인덱스가적용될수있다. 인트라예측모드기반으로 2차변환커널세트를결정하기때문에 ,인트라예측 모드를획득(또는결정)한이후에변환커널세트를결정할수있다.이로인해 의존성문제가야기된다.따라서,본발명의일실시예에서 ,이러한의존성을 없애는방법을설명한다.
[147] 실시예로서,인코더/디코더는다음의사항을고려하여인트라예측모드에
따라현재블록에적용되는변환커널세트를결정할수있다.
[148] - CU의가로대세로비율또는세로대가로비율
[149] - CU의크기
[15이 -일차변환의종류
[151] - MTS인덱스
[152] -암시적(implicit) MTS가적용되는지여부
2020/175965 1»(:1^1{2020/002920
26
[153] 인코더/디코더는상술한사항들에기초하여변환커널세트를결정할수있고, 상술한사항들은하나또는복수의조합으로변환커널세트결정에이용될수 있다.
[154] 도 18은본발명의일실시예에따른변환블록단위 2차변환적용방법을
예시하는도면이다.본발명의일실시예에따르면,인코더/디코더는코딩 블록(또는코딩유닛)을복수개의변환블록(또는변환유닛)으로분할하고, 각각의변환블록에대하여 2차변환을적용할수있다.하나의코딩유닛에서 복수개의변환블록으로분할된각변환블록에 2차변환을적용할수있다.각 변환블록의크기는코딩유닛에대한분할방법에기초하여결정된다 .犯!5가 적용된코딩유닛의각변환블록의크기는도 15에서처럼수직방향분할혹은 수평방향분할에따라결정될수있으며,추가하여분할되는개수에따라그 크기가결정될수있다.犯!5가적용된블록에서분할되는개수는 2혹은 4가될 수있다.도 18에서하나의코딩유닛은 «모가적용된블록으로수직방향으로 분할되며분할개수는 4개인경우를예시한다.앞서설명한도 14(비처럼코딩 유닛의크기가 \¥ 이면,각변환블록의크기는 \¥/4 11가될수있다.상기 변환블록의크기는 2차변환의적용여부를결정하는너비와높이로사용될수 있다.
[155] 본발명의일실시예에따르면,인코더/디코더는각변환블록별로같은변환 커널세트를사용하거나서로다른변환커널세트를사용할수있다.분할된 변환블록은같은인트라예측모드와분할블록의크기가같으면같은변환 커널을사용할수있다.이와반대의경우인코더/디코더는각변환블록별로 커널세트를결정하여사용할수있다.인트라서브파티션모드가적용된코딩 유닛은루마성분은복수개의변환블록으로변환되지만크로마성분들은분할 되지않을수있다.이경우,루마변환블록과크로마변환블록이모두같은 2차 변환커널세트를사용할수있고, 2차변환블록이적용되는코딩블록의크기가 만족되어야한다.또한,루마와크로마의변환블록크기가다를수있다.이때 , 인코더/디코더는 2차변환블록이적용되는블록크기조건에맞춰 4x4또는 8x8 영역에적용할수있다.다른방법으로,인코더/디코더는루마변환에적용된 영역을크로마들도동일한영역에사용할수있다.인코더/디코더는루마와 크로마의인트라예측모드가서로다를수있기때문에각각다른변환커널 세트를사용할수있다.인트라예측모드기반커널세트를결정하는방법으로 설명하였으나,도 17에서기술된 2차변환커널세트를결정하는모든방법을 적용할수있다.
[156] 본발명의일실시예에따르면,코딩유닛의크기가최대변환크기보다큰
경우,코딩유닛은별도의시그널링없이복수개의변환블록으로분할될수 있다.이경우, 2차변환을적용하면성능저하및복잡도가증가할수있어 2차 변환이적용되는최대코딩블록을제한할수있다.최대코딩블록의크기는 최대변환크기와같을수있다.또는기설정된코딩블록의크기로사용할수
2020/175965 1»(:1^1{2020/002920
27 있다.기 설정된값은 64, 32, 16일수있고이에 한정하지 않는다.긴변의길이로 값또는총샘플의 개수일수있다.
[157] 또한,일실시예에서,인코더/디코더는 2차변환블록크기를코딩유닛의
좌상단 4x4영역또는 8x8영역으로한정하지 않고 2x8, 8x2, 4x16, 16x4, 2x32, 32x2으로정의할수도있다.인코더/디코더는코딩블록의가로대세로의 비율/세로대가로의 비율을고려하여 적응적/시그널링없이 2차변환이 적용되는영역을결정할수있다.
[158] 도 19은본발명이 적용되는일실시예에따른인트라예측모드가적용된현재 코딩블록에모01(를적용하는방법을나타낸도면이다.본발명의 일실시예에 따르면,현재블록이정사각형블록이아닌경우,인코더/디코더는인트라예측 모드중 IX:모드의 경우긴변의참조샘플만이용하여 예측을수행할수있다. 이 경우,짧은변의 샘플값은현재코딩블록의 예측에 전혀 반영되지 않을수 있다.이 경우,현재블록의 예측값과짧은변의참조샘플과의차이가클수 있다.따라서,인코더/디코더는인트라예측을수행함에 있어서,샘플의 위치 기반필터링을수행할수있다.전술한바와같이,본발명에서,이러한샘플의 위치 기반필터링 방법은모。!5 (:로지칭될수있다.인코더/디코더는모。!5 (:가 적용되는경우 IX:모드에서제 1참조샘플과제 2참조샘플과좌상단에 인접한 참조샘플값을사용하여 가중치에기초한필터링을수행할수있다.이때, 아래의수학식 7내지 12를이용하여참조샘플및/또는각각의참조샘플에 적용되는가중치를유도할수있다.
[159] [수식7]
田†나 X ] [乂 ] = 아 시 ][乂 ]
[16이 [수식 8] 니 ] = 에 一1 ]
[161] [수식 9]
[ ] = 32 > > ( ( < < 1 ) > > 어 )
[162] [수식 1이
[163] [수식 11]
+( [ ] > > 4)) : 0
[164]
2020/175965 1»(:1/10公020/002920
28
[수식 12]
predSamples[ x ][ y ] = clip1 Cmp( ( refL[ x ][ y ] * wL[ x ] + refT[ x ][ y ] * wT[ y ] - p[ -1 ][ -1 ] * wTL[ x ][ y ] + ( 64 - wL[ x ] - wT[ y ] + wTL[ x ][ y ] )
* predSamples[ x ][ y ] + 32 ) > >6 )
[165] 살펴보면,좌측참조샘플은수학식 7을이용하여유도될수있고,우측참조 샘플은수학식 8을이용하여유도될수있다.우측참조샘플에적용되는가중치 값은수학식 9를이용하여유도될수있고,좌측참조샘플에적용되는가중치 값은수학식 W을이용하여유도될수있고,좌상단에모서리에위치한참조 샘늘에적용되는가중치값은수학식 11을이용하여유도될수있다.그리고, 인코더/디코더는결정된가중치값에기초하여수학식 12에따라예측샘플을 생성할수있다.
[166] 본발명의일실시예에따르면,인코더/디코더는정사각형블록이아닌
비정방형블록의경우,상대적으로긴변과짧은변과의가중치값을다르게 설정할수있다.예를들어,인코더/디코더는상대적으로긴변에적용되는 가중치값을짧은변의가중치값보다작게설정할수있다.상술한수학식 9,
W에서짧은변에속하면가중치값을긴변의경우와다르게설정할수있다.일 예로서,가중치값은 32대신 16, 8, 4등으로설정될수있다.또는,상술한수학식 9및 W에사용된스케일변수 nScale을이용할수있다.인코더/디코더는긴변과 짧은변의위치에따라그값을설정할수있다.
[167] 일실시예에서,복수참조라인샘플을사용하는경우,인코더/디코더는
PDPC를수직모드및/또는수평모드에적용할수있다.또는,인코더/디코더는 PDPC를수직,수평, DC, PLANAR모드에적용할수있다.
[168] 도 20은본발명의일실시예를따른비디오신호처리방법을나타내는
흐름도이다.도 20을참조하면,설명의편의를위해디코더를위주로설명하나 본발명이이에제한되는것은아니며,본실시예에따른비디오신호처리 방법은인코더에도실질적으로동일한방법으로적용될수있다.
[169] 도 20을참조하면,디코더는현재블록에인트라서브파티션 (ISP, Intra
Sub-Partitions)모드가적용되는지여부를결정한다 (S2001).
[170] 디코더는상기현재블록에 ISP모드가적용되는경우,상기현재블록을
복수의수평또는수직방향의직사각형변환블록들로분할한다 (S2002).
[171] 디코더는상기변환블록들각각에대하여인트라예측을수행함으로써상기 변환블록들의예측블록을생성한다 (S2003).
[172] 디코더는상기변환블록의잔차블록및상기 예측블록에기초하여상기현재 블록을복원한다 (S2004).
[173] 전술한바와같이,상기예측블록을생성하는단계는,상기현재블록으로부터 분할된변환블록단위로위치-의존적인인트라예측샘플
2020/175965 1»(:1^1{2020/002920
29 필터링 (position-dependent intra prediction sample filtering)을수행하는단계를 포함할수있다.
[174] 또한,전술한바와같이,상기 예측블록을생성하는단계는,상기 변환블록의 너비 및높이중적어도하나에기초하여상기위치-의존적인인트라예측샘플 필터링의 적용여부를결정하는단계를더포함할수있다.
[175] 또한,전술한바와같이,상기위치-의존적인인트라예측샘플필터링의 적용 여부를결정하는단계는,상기 변환블록의 너비가기 설정된기준값보다 크거나같고,그리고,상기 변환블록의높이가상기기 설정된기준값보다 크거나같은경우,상기 위치-의존적인인트라예측샘플필터링을적용하는 것으로결정함으로써수행될수있다.
[176] 또한,전술한바와같이,상기 변환블록의잔차블록은상기 변환블록단위로 이차역변환 (inverse secondary transform)및 일차역변환 (inverse primary transform)을수행함으로써유도될수있다.
[177] 또한,전술한바와같이,상기 현재블록에 이차변환이 적용되는지 여부를 결정하는단계;상기 현재블록에상기 이차변환이 적용되는경우,상기 현재 블록의 인트라예측모드에기초하여 미리정의된이차변환커널세트들중에서 상기 현재블록에 적용되는이차변환커널세트를유도하는단계;상기 결정된 이차변환커널세트내에서상기 현재블록에 적용되는이차변환커널을 결정하는단계;상기 변환블록단위로이차역변환을수행함으로써상기 변환 블록의 이차역변환된블록을생성하는단계;및상기 이차역변환된블록에 대하여 일차역변환을수행함으로써,상기 변환블록의잔차블록을생성하는 단계를포함할수있다.
[178] 상술한본발명의실시예들은다양한수단을통해구현될수있다.예를들어, 본발명의실시예들은하드웨어,펌웨어 (firmware),소프트웨어또는그것들의 결합등에의해구현될수있다.
[179] 하드웨어에의한구현의경우,본발명의실시예들에 따른방법은하나또는그 이상의 ASICs(Application Specific Integrated Circuits), DSPs(Digital Signal Processors), DSPDs(Digital Signal Processing Devices), PLDs(Programmable Logic Devices), FPGAs(Field Programmable Gate Arrays),프로세서 ,컨트롤러 ,마이크로 컨트롤러,마이크로프로세서등에 의해구현될수있다.
[180] 펌웨어나소프트웨어에 의한구현의 경우,본발명의실시예들에따른방법은 이상에서 설명된기능또는동작들을수행하는모듈,절차또는함수등의 형태로구현될수있다.소프트웨어코드는메모리에 저장되어프로세서에의해 구동될수있다.상기 메모리는프로세서의 내부또는외부에위치할수있으며, 이미 공지된다양한수단에의해프로세서와데이터를주고받을수있다.
[181] 일부실시예는컴퓨터에의해실행되는프로그램모듈과같은컴퓨터에의해 실행가능한명령어를포함하는기록매체의 형태로도구현될수있다.컴퓨터 판독가능매체는컴퓨터에의해 액세스될수있는임의의 가용매체일수있고,
2020/175965 1»(:1^1{2020/002920
30 휘발성 및비휘발성 매체 ,분리형 및비분리형 매체를모두포함한다.또한, 컴퓨터판독가능매체는컴퓨터 저장매체 및통신매체를모두포함할수있다. 컴퓨터 저장매체는컴퓨터판독가능명령어 ,데이터구조,프로그램모듈또는 기타데이터와같은정보의 저장을위한임의의 방법또는기술로구현된휘발성 및비휘발성,분리형 및비분리형 매체를모두포함한다.통신매체는
전형적으로컴퓨터판독가능명령어,데이터구조또는프로그램모듈과같은 변조된데이터신호의 기타데이터 ,또는기타전송메커니즘을포함하며 , 임의의 정보전달매체를포함한다.
[182] 전술한본발명의설명은예시를위한것이며,본발명이속하는기술분야의 통상의지식을가진자는본발명의기술적사상이나필수적인특징을변경하지 않고서다른구체적인형태로쉽게 변형이가능하다는것을이해할수있을 것이다.그러므로이상에서기술한실시예들은모든면에서 예시적인것이며 한정적인것이아는것으로해석해야한다.예를들어,단일형으로설명되어 있는각구성요소는분산되어실시될수도있으며,마찬가지로분산된것으로 설명되어 있는구성요소들도결합된형태로실시될수있다.
[183] 본발명의 범위는상기상세한설명보다는후술하는특허청구범위에 의하여 나타내어지며,특허청구범위의 의미 및범위그리고그균등개념으로부터 도출되는모든변경또는변형된형태가본발명의범위에포함되는것으로 해석되어야한다.
산업상이용가능성
[184] 이상,전술한본발명의바람직한실시예는,예시의목적을위해개시된것으로, 당업자라면이하첨부된특허청구범위에 개시된본발명의 기술적사상과그 기술적 범위내에서,다양한다른실시예들을개량,변경,대체또는부가등이 가능할것이다.
Claims
[청구항 1] 비디오신호처리방법에있어서,
현재블록에인트라서브파티션 !5, Intra Sub-Partitions)모드가 적용되는지여부를결정하는단계 ;
상기현재블록에 KP모드가적용되는경우,상기현재블록을복수의 수평또는수직방향의직사각형변환블록들로분할하는단계 ;
상기변환블록들각각에대하여인트라예측을수행함으로써상기변환 블록들의 예측블록을생성하는단계 ;및
상기변환블록의잔차블록및상기 예측블록에기초하여상기현재 블록을복원하는단계를포함하되,
상기 예측블록을생성하는단계는,
상기현재블록으로부터분할된변환블록단위로위치-의존적인인트라 예즉샘늘필터링 (position-dependent intra prediction sample filtering)을 수행하는단계를포함하는,비디오신호처리방법.
[청구항 2] 제 1항에 있어서,
상기 예측블록을생성하는단계는,
상기변환블록의너비및높이중적어도하나에기초하여상기 위치-의존적인인트라예측샘플필터링의적용여부를결정하는단계를 더포함하는,비디오신호처리방법.
[청구항 3] 제 2항에 있어서,
상기위치-의존적인인트라예측샘플필터링의적용여부를결정하는 단계는,
상기변환블록의너비가기설정된기준값보다크거나같고,그리고, 상기변환블록의높이가상기기설정된기준값보다크거나같은경우, 상기위치-의존적인인트라예측샘플필터링을적용하는것으로 결정함으로써수행되는,비디오신호처리방법 .
[청구항 4] 제 1항에 있어서,
상기변환블록의잔차블록은상기변환블록단위로이차역변환 (inverse secondary transform)및일차역변환 (inverse primary transform)을
수행함으로써유도되는,비디오신호처리방법.
[청구항 5] 제 1항에 있어서,
상기현재블록에이차변환이적용되는지여부를결정하는단계;
상기현재블록에상기이차변환이적용되는경우,상기현재블록의 인트라예측모드에기초하여미리정의된이차변환커널세트들중에서 상기현재블록에적용되는이차변환커널세트를유도하는단계;
상기결정된이차변환커널세트내에서상기현재블록에적용되는이차 변환커널을결정하는단계 ;
2020/175965 1»(:1^1{2020/002920
32 상기변환블록단위로이차역변환을수행함으로써상기변환블록의 이차역변환된블록을생성하는단계;및
상기이차역변환된블록에대하여일차역변환을수행함으로써,상기 변환블록의잔차블록을생성하는단계를포함하는,비디오신호처리 방법.
[청구항 6] 비디오신호처리장치에있어서,
프로세서를포함하며 ,
상기프로세서는,
현재블록에인트라서브파티션 !5, Intra Sub-Partitions)모드가 적용되는지여부를결정하고,
상기현재블록에 KP모드가적용되는경우,상기현재블록을복수의 수평또는수직방향의직사각형변환블록들로분할하고, 상기변환블록들각각에대하여인트라예측을수행함으로써상기변환 블록들의예측블록을생성하고,
상기변환블록의잔차블록및상기예측블록에기초하여상기현재 블록을복원하되,
상기프로세서는,
상기현재블록으로부터분할된변환블록단위로위치-의존적인인트라 예즉샘늘필터링 (position-dependent intra prediction sample filtering)을 수행하는것을특징으로하는,비디오신호처리장치 .
[청구항 7] 제 6항에있어서,
상기프로세서는,
상기변환블록의너비및높이중적어도하나에기초하여상기 위치-의존적인인트라예측샘플필터링의적용여부를결정하는,비디오 신호처리장치 .
[청구항 8] 제 7항에있어서,
상기프로세서는,
상기변환블록의너비가기설정된기준값보다크거나같고,그리고, 상기변환블록의높이가상기기설정된기준값보다크거나같은경우, 상기위치-의존적인인트라예측샘플필터링을적용하는것으로 결정하는,비디오신호처리장치 .
[청구항 9] 제 6항에있어서,
상기변환블록의잔차블록은상기변환블록단위로이차역변환 (inverse secondary transform)및일차역변환 (inverse primary transform)을
수행함으로써유도되는,비디오신호처리장치.
[청구항 10] 제 6항에있어서,
상기프로세서는,
상기현재블록에이차변환이적용되는지여부를결정하고,
2020/175965 1»(:1^1{2020/002920
33 상기 현재블록에상기 이차변환이 적용되는경우,상기 현재블록의 인트라예측모드에 기초하여미리 정의된이차변환커널세트들중에서 상기 현재블록에 적용되는이차변환커널세트를유도하고, 상기 결정된이차변환커널세트내에서상기 현재블록에 적용되는이차 변환커널을결정하고,
상기 변환블록단위로이차역변환을수행함으로써상기 변환블록의 이차역변환된블록을생성하고,
상기 이차역변환된블록에 대하여 일차역변환을수행함으로써,상기 변환블록의잔차블록을생성하는,비디오신호처리장치 .
[청구항 11] 비디오신호처리 방법에 있어서 ,
현재블록에 인트라서브파티션 !5, Intra Sub-Partitions)모드가 적용되는지 여부를결정하는단계 ;
상기 현재블록에 KP모드가적용되는경우,상기 현재블록을복수의 수평또는수직 방향의직사각형 변환블록들로분할하는단계 ;
상기 변환블록들각각에 대하여 인트라예측을수행함으로써상기 변환 블록들의 예측블록을생성하는단계 ;및
원본블록에서상기 예측블록을감산함으로써상기 변환블록의잔차 블록을생성하는단계를포함하되 ,
상기 예측블록을생성하는단계는,
상기 현재블록으로부터분할된변환블록단위로위치-의존적인인트라 예즉샘늘필터링 (position-dependent intra prediction sample filtering)을 수행하는단계를포함하는,비디오신호처리 방법.
[청구항 12] 컴퓨팅 디바이스의하나이상의프로세서에서실행하도록구성된컴퓨터 실행가능한컴포넌트가저장된비 일시적 (non-transitory)컴퓨터판독 가능한매체 (computer-executable component)로서 ,상기 컴퓨터실행 가능한컴포넌트는,
현재블록에 인트라서브파티션 !5, Intra Sub-Partitions)모드가 적용되는지 여부를결정하고,
상기 현재블록에 KP모드가적용되는경우,상기 현재블록을복수의 수평또는수직 방향의직사각형 변환블록들로분할하고, 상기 변환블록들각각에 대하여 인트라예측을수행함으로써상기 변환 블록들의 예측블록을생성하고,
상기 변환블록의잔차블록및상기 예측블록에 기초하여상기 현재 블록을복원하되,
상기 컴퓨터실행가능한컴포넌트는,
상기 현재블록으로부터분할된변환블록단위로위치-의존적인인트라 예즉샘늘필터링 (position-dependent intra prediction sample filtering)을 수행하는것을특징으로하는,비 일시적 컴퓨터판독가능한매체.
Priority Applications (4)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
JP2021550241A JP7293376B2 (ja) | 2019-02-28 | 2020-02-28 | イントラ予測ベースのビデオ信号処理方法及び装置 |
CN202080016690.3A CN113491116B (zh) | 2019-02-28 | 2020-02-28 | 基于帧内预测的视频信号处理方法和装置 |
KR1020217025729A KR20210122797A (ko) | 2019-02-28 | 2020-02-28 | 인트라 예측 기반 비디오 신호 처리 방법 및 장치 |
US17/401,923 US11979554B2 (en) | 2019-02-28 | 2021-08-13 | Intra prediction-based video signal processing method and device |
Applications Claiming Priority (4)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
KR10-2019-0024368 | 2019-02-28 | ||
KR20190024368 | 2019-02-28 | ||
KR10-2019-0035438 | 2019-03-27 | ||
KR20190035438 | 2019-03-27 |
Related Child Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
US17/401,923 Continuation US11979554B2 (en) | 2019-02-28 | 2021-08-13 | Intra prediction-based video signal processing method and device |
Publications (1)
Publication Number | Publication Date |
---|---|
WO2020175965A1 true WO2020175965A1 (ko) | 2020-09-03 |
Family
ID=72238632
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
PCT/KR2020/002920 WO2020175965A1 (ko) | 2019-02-28 | 2020-02-28 | 인트라 예측 기반 비디오 신호 처리 방법 및 장치 |
Country Status (4)
Country | Link |
---|---|
US (1) | US11979554B2 (ko) |
JP (2) | JP7293376B2 (ko) |
KR (1) | KR20210122797A (ko) |
WO (1) | WO2020175965A1 (ko) |
Families Citing this family (9)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
KR20240027865A (ko) * | 2019-03-11 | 2024-03-04 | 닛폰 호소 교카이 | 화상 부호화 장치, 화상 복호 장치 및 프로그램 |
US20220182616A1 (en) * | 2019-03-12 | 2022-06-09 | Lg Electronics Inc. | Method and apparatus for encoding/decoding video and method for transmitting bitstream |
WO2020192793A1 (en) * | 2019-03-28 | 2020-10-01 | Huawei Technologies Co., Ltd. | Method and apparatus for intra smoothing |
MX2021014605A (es) * | 2019-06-18 | 2022-01-18 | Electronics & Telecommunications Res Inst | Metodo y aparato de codificacion/decodificacion de video, y medio de grabacion que almacena un flujo de bits. |
WO2020260313A1 (en) * | 2019-06-25 | 2020-12-30 | Fraunhofer-Gesellschaft zur Förderung der angewandten Forschung e.V. | Decoder, encoder and methods comprising a coding for intra subpartitions |
WO2021054799A1 (ko) * | 2019-09-21 | 2021-03-25 | 엘지전자 주식회사 | 변환에 기반한 영상 코딩 방법 및 그 장치 |
WO2021060827A1 (ko) * | 2019-09-23 | 2021-04-01 | 엘지전자 주식회사 | 변환에 기반한 영상 코딩 방법 및 그 장치 |
US11818395B2 (en) | 2021-04-22 | 2023-11-14 | Electronics And Telecommunications Research Institute | Immersive video decoding method and immersive video encoding method |
WO2024146511A1 (en) * | 2023-01-03 | 2024-07-11 | Mediatek Inc. | Representative prediction mode of a block of pixels |
Citations (5)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
KR20180040126A (ko) * | 2010-04-13 | 2018-04-19 | 삼성전자주식회사 | 트리 구조의 부호화 단위에 기초한 예측 단위를 이용하는 비디오 부호화 방법과 그 장치, 및 비디오 복호화 방법 및 그 장치 |
KR20180063201A (ko) * | 2015-09-29 | 2018-06-11 | 퀄컴 인코포레이티드 | 비디오 코딩을 위한 위치-의존적인 예측 결합을 이용한 향상된 비디오 인트라-예측 |
US20180176587A1 (en) * | 2016-12-21 | 2018-06-21 | Arris Enterprises Llc | Constrained position dependent intra prediction combination (pdpc) |
KR20180122362A (ko) * | 2016-03-21 | 2018-11-12 | 퀄컴 인코포레이티드 | 비디오 코딩에서 비-정사각형 블록들에 대한 예측 파라미터들의 결정 |
US20180332284A1 (en) * | 2017-05-09 | 2018-11-15 | Futurewei Technologies, Inc. | Intra-Prediction With Multiple Reference Lines |
Family Cites Families (17)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
KR20100095992A (ko) | 2009-02-23 | 2010-09-01 | 한국과학기술원 | 비디오 부호화에서의 분할 블록 부호화 방법, 비디오 복호화에서의 분할 블록 복호화 방법 및 이를 구현하는 기록매체 |
CN108712652A (zh) | 2012-06-29 | 2018-10-26 | 韩国电子通信研究院 | 视频编码方法和计算机可读介质 |
US10382792B2 (en) | 2015-02-13 | 2019-08-13 | Lg Electronics Inc. | Method and apparatus for encoding and decoding video signal by means of transform-domain prediction |
KR102600756B1 (ko) | 2015-03-06 | 2023-11-10 | 한국과학기술원 | 저 복잡도 변환에 기반한 영상 부호화 및 복호화 방법 및 이를 이용하는 장치 |
US10123044B2 (en) | 2015-07-16 | 2018-11-06 | Mediatek Inc. | Partial decoding circuit of video encoder/decoder for dealing with inverse second transform and partial encoding circuit of video encoder for dealing with second transform |
WO2018038554A1 (ko) | 2016-08-24 | 2018-03-01 | 엘지전자(주) | 이차 변환을 이용한 비디오 신호의 인코딩/디코딩 방법 및 장치 |
CN110024399B (zh) | 2016-11-28 | 2024-05-17 | 韩国电子通信研究院 | 对图像编码/解码的方法和设备及存储比特流的记录介质 |
US11405645B2 (en) | 2017-02-28 | 2022-08-02 | Google Llc | Transform kernel selection and entropy coding |
US10855997B2 (en) | 2017-04-14 | 2020-12-01 | Mediatek Inc. | Secondary transform kernel size selection |
JP7223014B2 (ja) | 2018-06-03 | 2023-02-15 | エルジー エレクトロニクス インコーポレイティド | 縮小された変換を用いて、ビデオ信号を処理する方法及び装置 |
CN113225559B (zh) * | 2018-10-12 | 2023-05-16 | Oppo广东移动通信有限公司 | 视频信号的编码方法和解码方法及其装置 |
JP7200379B2 (ja) | 2018-12-06 | 2023-01-06 | エルジー エレクトロニクス インコーポレイティド | 二次変換に基づく映像コーディング方法、及びその装置 |
US11570436B2 (en) * | 2019-01-28 | 2023-01-31 | Apple Inc. | Video signal encoding/decoding method and device therefor |
CN113302942B (zh) | 2019-02-01 | 2024-02-13 | Lg电子株式会社 | 基于二次变换的图像编码方法及其设备 |
KR20210114386A (ko) | 2019-02-08 | 2021-09-23 | 주식회사 윌러스표준기술연구소 | 이차 변환을 이용하는 비디오 신호 처리 방법 및 장치 |
US20220132109A1 (en) * | 2019-02-21 | 2022-04-28 | Lg Electronics Inc. | Image decoding method and apparatus using intra prediction in image coding system |
WO2020260313A1 (en) | 2019-06-25 | 2020-12-30 | Fraunhofer-Gesellschaft zur Förderung der angewandten Forschung e.V. | Decoder, encoder and methods comprising a coding for intra subpartitions |
-
2020
- 2020-02-28 JP JP2021550241A patent/JP7293376B2/ja active Active
- 2020-02-28 WO PCT/KR2020/002920 patent/WO2020175965A1/ko active Application Filing
- 2020-02-28 KR KR1020217025729A patent/KR20210122797A/ko active Search and Examination
-
2021
- 2021-08-13 US US17/401,923 patent/US11979554B2/en active Active
-
2023
- 2023-06-06 JP JP2023093523A patent/JP2023105114A/ja active Pending
Patent Citations (5)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
KR20180040126A (ko) * | 2010-04-13 | 2018-04-19 | 삼성전자주식회사 | 트리 구조의 부호화 단위에 기초한 예측 단위를 이용하는 비디오 부호화 방법과 그 장치, 및 비디오 복호화 방법 및 그 장치 |
KR20180063201A (ko) * | 2015-09-29 | 2018-06-11 | 퀄컴 인코포레이티드 | 비디오 코딩을 위한 위치-의존적인 예측 결합을 이용한 향상된 비디오 인트라-예측 |
KR20180122362A (ko) * | 2016-03-21 | 2018-11-12 | 퀄컴 인코포레이티드 | 비디오 코딩에서 비-정사각형 블록들에 대한 예측 파라미터들의 결정 |
US20180176587A1 (en) * | 2016-12-21 | 2018-06-21 | Arris Enterprises Llc | Constrained position dependent intra prediction combination (pdpc) |
US20180332284A1 (en) * | 2017-05-09 | 2018-11-15 | Futurewei Technologies, Inc. | Intra-Prediction With Multiple Reference Lines |
Also Published As
Publication number | Publication date |
---|---|
JP2023105114A (ja) | 2023-07-28 |
US11979554B2 (en) | 2024-05-07 |
KR20210122797A (ko) | 2021-10-12 |
CN113491116A (zh) | 2021-10-08 |
JP2022522013A (ja) | 2022-04-13 |
US20210377519A1 (en) | 2021-12-02 |
JP7293376B2 (ja) | 2023-06-19 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN112438050B (zh) | 使用二次变换的视频信号处理方法及装置 | |
KR102402045B1 (ko) | 비디오 신호 처리 방법 및 장치 | |
CN109661819B (zh) | 用于处理视频信号的方法和设备 | |
WO2020175965A1 (ko) | 인트라 예측 기반 비디오 신호 처리 방법 및 장치 | |
KR102555352B1 (ko) | 인트라 예측 방법과 이를 이용한 부호화 장치 및 복호화 장치 | |
CN113196780A (zh) | 使用多变换核处理视频信号的方法和设备 | |
KR20210148321A (ko) | 인트라 예측 기반 비디오 신호 처리 방법 및 장치 | |
KR102344190B1 (ko) | 참조 샘플을 이용하는 비디오 신호 처리 방법 및 장치 | |
KR20180005120A (ko) | 비디오 신호 처리 방법 및 장치 | |
CN113395517B (zh) | 视频信号编码/解码方法及其设备 | |
KR20210084567A (ko) | 화면 내 예측 필터링을 이용한 비디오 신호 처리 방법 및 장치 | |
KR102606291B1 (ko) | 교차성분 선형 모델을 이용한 비디오 신호 처리 방법 및 장치 | |
JP2022521911A (ja) | イントラ予測ベースのビデオ信号処理方法及び装置 | |
CN113225560A (zh) | 视频编码/解码方法、视频编码/解码装置和存储介质 | |
CN114521326B (zh) | 使用缩放处理的视频信号处理方法及装置 | |
CN113491116B (zh) | 基于帧内预测的视频信号处理方法和装置 | |
US20240259557A1 (en) | Intra prediction-based video signal processing method and device | |
KR20230137825A (ko) | 인접 정보 기반 팔레트 모드를 이용하는 비디오 코딩 방법 및 장치 | |
KR20230139780A (ko) | 적응적 다중변환선택을 이용하는 비디오 코딩방법 및 장치 | |
KR20240026180A (ko) | 인트라 예측을 이용한 비디오 신호 처리 방법 및 이를 위한 장치 | |
KR20240031030A (ko) | 1차 변환 커널에 적응적인 분리 불가능한 2차 변환 기반 비디오 코딩방법 및 장치 |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
121 | Ep: the epo has been informed by wipo that ep was designated in this application |
Ref document number: 20762541 Country of ref document: EP Kind code of ref document: A1 |
|
ENP | Entry into the national phase |
Ref document number: 2021550241 Country of ref document: JP Kind code of ref document: A |
|
NENP | Non-entry into the national phase |
Ref country code: DE |
|
122 | Ep: pct application non-entry in european phase |
Ref document number: 20762541 Country of ref document: EP Kind code of ref document: A1 |