method and apparatus for encoding / decoding video signal using secondary transform

专利摘要:
the present invention relates to a method and apparatus for encoding / decoding a video signal. specifically, a method for decoding a video signal may include: generating a quantized transform block by performing entropy decoding for the video signal; generate an unquantified transform block by executing decanting for the quantized transform block; determine whether to apply a secondary inverse transform based on information related to a non-zero coefficient in the unquantified transform block; and performing the secondary reverse transform for the decantified transform block using a secondary reverse transform core applied to the decantized transform block.
公开号:BR112019013834A2
申请号:R112019013834
申请日:2017-12-27
公开日:2020-01-28
发明作者:Jang Hyeongmoon；Lim Jaehyun；Nam Junghak；Kim Seunghwan
申请人:Lg Electronics Inc；
IPC主号:

专利说明:

“METHOD AND APPARATUS FOR ENCODING / DECODING VIDEO SIGNAL USING SECONDARY TRANSFORMED”
FIELD OF TECHNIQUE [001] The present invention relates to an apparatus and apparatus for encoding / decoding a video signal and, more particularly, a method for encoding / decoding a video signal using secondary transform and apparatus for endure the same.
BACKGROUND OF THE INVENTION [002] Compression encoding means a series of signal processing techniques for transmitting digitized information over a communication line or techniques for storing information in a form suitable for a storage medium. The medium, including a picture, an image, audio, etc., can be a target for compression encoding and, particularly, a technique for performing compression encoding on an image is called video image compression.
[003] The next generation video content must have the characteristics of high spatial resolution, high frame rate and high dimensionality in the representation of the scene. In order to process such content, it results in a dramatic increase in memory storage, the rate of access to memory and processing capacity.
[004] Thus, it is necessary to design an encoding tool to efficiently process next generation video content.
SUMMARY OF THE INVENTION [005] Technical Problem [006] One embodiment of the present invention provides a method for applying the secondary transform to a signal from a transform region that is primarily transformed.
Petition 870190065456, of 7/11/2019, p. 8/70
2/48 [007] In addition, one embodiment of the present invention provides a method for efficiently dimensioning the size of a transform core used for the secondary transform.
[008] In addition, one embodiment of the present invention proposes a method to adaptively select a transform core according to a block size and perform a secondary transform using the selected transform core.
[009] Furthermore, one embodiment of the present invention provides a method for adaptively selecting a core adapted to the secondary transform, transmitting the size of a transform core.
[010] In addition, one embodiment of the present invention provides a method for determining whether to apply the secondary transform or an application range using a residual signal.
[011] In addition, one embodiment of the present invention provides a method for deriving a size of a transform core applied to the secondary transform using a residual signal.
[012] The objectives of the present invention are not limited to the technical objectives described above, and other techniques that are objectives not mentioned here can be understood by those skilled in the art from the description below.
[013] Technical Solution [014] In one aspect of the present invention, a method for decoding a video signal may include: generating a quantized transform block by performing entropy decoding for the video signal; generate an unquantified transform block by executing decanting for the quantized transform block; determine whether to apply a secondary inverse transform based on information related to a non-zero coefficient in the decantified transformed block; and perform the secondary reverse transform for the block of
Petition 870190065456, of 7/11/2019, p. 9/70
3/48 deconstructed transform using a secondary reverse transform core applied to the deconstructed transform block.
[015] Preferably, determining whether to apply the secondary inverse transform may include checking whether one or more non-zero coefficients exist in a specific upper-left region of the unquantified transform block, and whether there are one or more non-zero coefficients in the region specific, the secondary reverse transform can be applied to the unquantified transform block.
[016] Preferably, determining whether to apply the secondary inverse transform may include checking the number of non-zero coefficients in the specific upper-left region of the unquantified transform block, and if the number of non-zero coefficients in the specific region exceeds a specific limit , the secondary reverse transform can be applied to the unquantified transform block.
[017] Preferably, determining whether to apply the secondary inverse transform may include partitioning the unquantified transform block into sub-blocks of a specific size, and determining whether to apply the secondary inverse transform in units of the sub-block.
[018] Preferably, determining whether to apply the secondary inverse transform in units of the sub-block may include checking whether there are one or more non-zero coefficients in a current sub-block, and whether one or more non-zero coefficients exist in the current sub-block, the secondary inverse transform can be applied to the current sub-block.
[019] Preferably, determining whether to apply the secondary inverse transform in units of the sub-block may include checking the number of non-zero coefficients in the current sub-block, and if the number of non-zero coefficients in the current sub-block exceeds a specific limit, the secondary inverse transform can be applied to the current sub-block.
Petition 870190065456, of 7/11/2019, p. 10/70
4/48 [020] Preferably, the method may also include determining a size of the secondary reverse transform core applied to the unquantified transform block based on information related to the non-zero coefficient in the unquantified transform block.
[021] Preferably, the size of the secondary reverse transform core can be determined as the size of the smallest secondary reverse transform core among secondary reverse transform cores including the non-zero coefficients that exist in a region that has a specific size in size specific left upper part of the transform block decanted.
[022] Preferably, the method may also include: if the size of the unquantified transform block is larger than a block having a predetermined minimum size, extract a syntax indicating the size of the secondary reverse transform core of the video signal; and determining the size of the secondary reverse transform core applied to the transform block unquantified based on syntax.
[023] Preferably, the syntax that indicates the size of the secondary reverse transform core can be transmitted in units of a sequence, an image, a slice, a coding block, or a transform block.
[024] In another aspect of the present invention, an apparatus for decoding a video signal may include: an entropy decoding unit generating a quantized transform block performing entropy decoding for the video signal; a decanting unit generating a decantized transform block executing decanting for the quantized transform block; a secondary reverse transform determination unit determining whether to apply secondary reverse transform based on information related to a non-zero coefficient in the unquantified transform block; and a unit
Petition 870190065456, of 7/11/2019, p. 11/70
5/48 secondary reverse transform by executing the secondary reverse transform for the decantized transform block using a secondary reverse transform core applied to the decantized transform block.
ADVANTAGEOUS EFFECTS [025] According to one embodiment of the present invention, the compression performance can be further improved by performing a secondary transform for a transform domain signal that is transformed primarily and the amount of residual signal data signaled for a decoder can be efficiently reduced.
[026] In addition, according to one embodiment of the present invention, by applying the secondary transform, it is possible to increase the compression efficiency by determining cores of suitable sizes for blocks of various sizes.
[027] In addition, according to one embodiment of the present invention, transform cores of various sizes can be applied signaling an optimized size of core information for a decoder, regardless of the size of a block, thereby improving the compression performance.
[028] The technical effects of the present invention are not limited to the technical effects described above, and other technical effects not mentioned here can be understood by those skilled in the art from the description below.
BRIEF DESCRIPTION OF THE DRAWINGS [029] The attached drawings, which are included here as part of the description to help understand the present invention, provide embodiments of the present invention and describe the technical features of the present invention with the description below.
[030] Figure 1 illustrates a schematic block diagram of an encoder
Petition 870190065456, of 7/11/2019, p. 12/70
6/48 in which the encoding of a still image or video signal is performed, as a modality to which the present invention is applied.
[031] Figure 2 illustrates a schematic block diagram of a decoder in which the decoding of a still image or video signal is performed, as a modality to which the present invention is applied.
[032] Figure 3 is a diagram for describing a division structure of a coding unit that can be applied to the present invention.
[033] Figure 4 is a diagram to describe a prediction unit that can be applied to the present invention.
[034] Figure 5 is a diagram to describe a method for determining a size of a transform core used for the secondary transform based on the size of a block as a modality to which the present invention is applied.
[035] Figures 6 and 7 are diagrams to describe a method for determining a size of a transform core used for the secondary transform based on a block width and height as a modality to which the present invention is applied.
[036] Figure 8 represents a case in which the non-separable transform is applied for the secondary transform to be assumed.
[037] Figure 9 is a flow chart showing a method for determining a size of a transform core used for secondary transform as a modality to which the present invention is applied.
[038] Figure 10 is a flow chart illustrating a method for determining a size of a transform core used for the secondary transform as a modality to which the present invention is applied.
[039] Figure 11 is a diagram illustrating a method for determining whether to apply a secondary transform using a residual signal according to
Petition 870190065456, of 7/11/2019, p. 13/70
7/48 is an embodiment of the present invention.
[040] Figures 12 and 13 are diagrams that illustrate a method for determining whether to apply the secondary transform using a residual signal according to an embodiment of the present invention.
[041] Figure 14 is a diagram illustrating a method for determining a size of a secondary transform core using a residual signal in accordance with an embodiment of the present invention.
[042] Figure 15 is a diagram illustrating a method of decoding a video signal according to an embodiment of the present invention.
[043] Figure 16 is a diagram illustrating a video signal decoding apparatus according to an embodiment of the present invention.
DETAILED DESCRIPTION [044] In the following, a preferred embodiment of the present invention will be described with reference to the accompanying drawings. The description that will be made below with the accompanying drawings is to describe exemplary embodiments of the present invention, and is not intended to describe the only embodiment to which the present invention can be implemented. The description below includes particular details to provide a perfect understanding of the present invention. However, it is understood that the present invention can be carried out without the particular details for those skilled in the art.
[045] In some cases, in order to prevent the technical concept of the present invention from being unclear, structures or devices that are publicly known may be omitted, or may be represented as a block diagram centering on the central functions of the structures or structures. devices.
[046] Furthermore, although the general terms widely used today are selected as the terms in the present invention as much as possible, a term that is arbitrarily selected by the applicant is used in one case
Petition 870190065456, of 7/11/2019, p. 14/70
Specific 8/48. Once the meaning of the term is clearly described in the corresponding part of the description in such a case, it is understood that the present invention will not simply be interpreted by the terms used only in the description of the present invention, but the meaning of the terms must be discovered.
[047] Specific terminologies used in the description below can be provided to assist in understanding the present invention. In addition, the specific terminology can be modified to other forms within the scope of the technical concept of the present invention. For example, a signal, data, a sample, an image, a frame, a block, etc., can be properly replaced and interpreted in each coding process.
[048] Next, in this specification, a 'processing unit' means a unit on which an encoding / decoding processing process, such as prediction, transform and / or quantization, is performed. In the following, for convenience of description, the processing unit may be referred to as a 'processing block' or a 'block'.
[049] The processing unit can be interpreted as including a unit for a luma component and a unit for a chroma component. For example, the processing unit may correspond to a Coding Tree Unit (CTU), a Coding Unit (CU), a Prediction Unit (PU), or a Transform Unit (TU).
[050] In addition, the processing unit can be interpreted as the unit for the luma component and the unit for the chroma component. For example, the processing unit can correspond to a Coding Tree Block (CTB), a Coding Block (CB), a Prediction Block (PB) or a Transform Block (TB) for the luma component. Alternatively, the processing unit can correspond to the Coding Tree Block (CTB), the Coding Block (CB), the Prediction Block (PB) or the Transform Block
Petition 870190065456, of 7/11/2019, p. 15/70
9/48 (TB) for the chroma component. Furthermore, the present invention is not limited to this and the processing unit can be interpreted as including the unit for the luma component and the unit for the chroma component.
[051] In addition, the processing unit is not particularly limited to a square block, but can be configured as a polygonal shape having three or more vertices.
[052] In addition, hereinafter, in this specification, a pixel and the like will collectively be called a sample. In addition, using the sample can mean using a pixel value and the like.
[053] Figure 1 illustrates a schematic block diagram of an encoder in which the encoding of a still image or video signal is performed, as a modality to which the present invention is applied.
[054] With reference to Figure 1, encoder 100 can include a video splitting unit 110, a subtractor 115, a transform unit 120, a quantization unit 130, a decanting unit 140, a reverse transform unit 150 , a filter unit 160, a decoded image buffer (DPB) 170, a prediction unit 180 and an entropy coding unit 190. In addition, the prediction unit 180 may include an inter prediction unit 181 and an intra prediction unit 182.
[055] The video splitting unit 110 divides an incoming video signal (or image or frame), inserted in the encoder 100, into one or more processing units.
[056] Subtractor 115 generates a residual signal (or residual block) by subtracting a prediction signal (or prediction block), emitted by the prediction unit 180 (ie, by the inter prediction unit 181 or by the intra prediction unit 182 ), from the input video signal. The generated residual signal (or residual block) is transmitted to the transform unit 120.
Petition 870190065456, of 7/11/2019, p. 16/70
10/48 [057] Transform unit 120 generates transform coefficients by applying a transform scheme (for example, discrete cosine transform (DCT), discrete sine transform (DST), graph-based transform (GBT) or transform of Karhunen-Loeve (KLT)) to the residual signal (or residual block). In this case, the transform unit 120 can generate transform coefficients by performing the transform using a prediction mode applied to the residual block and a transform scheme determined based on the size of the residual block.
[058] The quantization unit 130 quantifies the transform coefficient and transmits it to the entropy coding unit 190, and the entropy coding unit 190 performs an entropy coding operation of the quantized signal and transmits it as a bit stream.
[059] However, the quantized signal emitted by the quantization unit 130 can be used to generate a prediction signal. For example, a residual signal can be reconstructed by applying decanting and reverse transform to the quantized signal through the decanting unit 140 and the reverse transform unit 150. A reconstructed signal can be generated by adding the reconstructed residual signal to the prediction signal emitted by the inter prediction 181 or the intra prediction unit 182.
[060] Meanwhile, during a compression process, neighboring blocks are quantized by different quantization parameters. Thus, an artifact in which a block boundary is shown can occur. Such a phenomenon is called a blocking artifact, which is one of the important factors in evaluating image quality. In order to decrease this artifact, a filtering process can be performed. Through this filtering process, the blocking artifact is removed and the error of a current image is reduced at the same time, thus improving the quality of the image.
Petition 870190065456, of 7/11/2019, p. 17/70
11/48 [061] Filter unit 160 applies filtering to the reconstructed signal, and sends it through a playback device or transmits it to the decoded image buffer 170. The filtered signal transmitted to the decoded image buffer 170 can be used as a reference image in the inter prediction unit 181. As described above, an encoding rate, as well as the image quality, can be improved using the filtered image as a reference image in an inter prediction mode. .
[062] The decoded image buffer 170 can store the filtered image for use as a reference image in the inter prediction unit 181.
[063] The inter 181 prediction unit performs temporal prediction and / or spatial prediction with reference to the reconstructed image, in order to remove temporal redundancy and / or spatial redundancy. In this case, a blocking artifact or touch artifact can occur because a reference image used to perform the prediction is a transformed signal that experiences quantization or decanting in a block unit when it is previously encoded / decoded.
[064] Consequently, in order to resolve the performance degradation attributable to the discontinuity of such a signal or quantization, the signals between the pixels can be interpellated in a subpixel unit by applying a low-pass filter to the inter 181 prediction unit. , the subpixel means a virtual pixel generated by applying an interpolation filter, and an entire pixel means a real pixel that is present in a reconstructed image. A linear interpolation, a bilinear interpolation, a wiener filter, and the like can be applied as an interpolation method.
[065] The interpolation filter can be applied to the reconstructed image, and can improve the accuracy of the prediction. For example, the inter 181 prediction unit can
Petition 870190065456, of 7/11/2019, p. 18/70
12/48 make prediction by generating an interpolation pixel by applying the interpellation filter to the entire pixel and using the interpolated block including interpolated pixels as a prediction block.
[066] The intra-182 prediction unit predicts a current block with reference to samples neighboring the block that is now being coded. The intra prediction unit 182 can perform the following procedure to perform intra prediction. First, the intra prediction unit 182 can prepare a reference sample needed to generate a prediction signal. In addition, the intra-182 prediction unit can generate a prediction signal using the prepared reference sample. Then, the intra prediction unit 182 can encode a prediction mode. In this case, the reference sample can be prepared by filling in the reference sample and / or filtering the reference sample. A quantization error may be present because the reference sample experiences the prediction and the reconstruction process. Consequently, in order to reduce such an error, a reference sample filtering process can be performed in each prediction mode used for intra prediction.
[067] The prediction signal (or prediction block) generated through the inter prediction unit 181 or the intra prediction unit 182 can be used to generate a reconstructed signal (or reconstructed block) or can be used to generate a residual signal (or residual block).
[068] Figure 2 illustrates a schematic block diagram of a decoder in which the decoding of a still image or video signal is performed, as a modality to which the present invention is applied.
[069] With reference to Figure 2, the decoder 200 may include an entropy decoding unit 210, a decanting unit 220, a reverse transform unit 230, an adder 235, a filtering unit 240, a temporary image store decoded (DPB) 250 and a
Petition 870190065456, of 7/11/2019, p. 19/70
13/48 prediction 260. In addition, the prediction unit 260 may include an inter prediction unit 261 and an intra prediction unit 262.
[070] In addition, a reconstructed video signal emitted through the decoder 200 can be reproduced through a reproduction device.
[071] Decoder 200 receives a signal (i.e., bit stream) emitted by encoder 100 shown in Figure 1. Entropy decoding unit 210 performs an entropy decoding operation on the received signal.
[072] The decanting unit 220 obtains transform coefficients from the entropy decoded signal using quantization step size information.
[073] The reverse transform unit 230 obtains a residual signal (or residual block) by inversely transforming the transform coefficients by applying an inverse transform scheme.
[074] Adder 235 adds the obtained residual signal (or residual block) to the prediction signal (or prediction block) emitted by the prediction unit 260 (that is, the inter prediction unit 261 or the intra prediction unit 262) , thereby generating a reconstructed signal (or reconstructed block).
[075] Filter unit 240 applies filtering to the reconstructed signal (or reconstructed block) and outputs the filtered signal to a playback device or transmits the filtered signal to the decoded image buffer 250. The filtered signal transmitted to the storage temporary decoded image 250 can be used as a reference image in the inter prediction unit 261.
[076] In this specification, the modalities described in the filter unit 160, in the inter prediction unit 181 and in the intra prediction unit 182 of the encoder 100 can be applied in the same way to the filter unit 240, to the inter prediction unit 261 and decoder intra 262 prediction unit,
Petition 870190065456, of 7/11/2019, p. 20/70
14/48 respectively.
[077] Processing Unit Division Structure [078] In general, a block-based image compression method is used in the compression technique (for example, HEVC) of a still image or a video. The block-based image compression method is a method of processing an image by dividing it into specific block units, and can decrease memory usage and computational load.
[079] Figure 3 is a diagram to describe a division structure of a coding unit that can be applied to the present invention.
[080] An encoder divides a single image (or figure) into encoding tree units (CTUs) in a quadrangular form, and sequentially encodes the CTUs one by one, according to the raster scan order.
[081] In HEVC, a CTU size can be determined as one of 64 χ 64, 32 χ 32 and 16 χ 16. The encoder can select and use the size of a CTU based on the resolution of an input video signal or the characteristics of the input video signal. The CTU includes a coding tree block (CTB) for one luma component and the CTB for two chroma components that correspond to it.
[082] A CTU can be divided into a quad-tree structure. That is, a CTU can be divided into four units, each having a square shape and having half the horizontal size and half the vertical size, thus being able to generate coding units (CUs). Such division of the quad-tree structure can be performed recursively. That is, the CUs are hierarchically separated from a CTU in the quad-tree structure.
[083] A CU means a basic unit for the process of processing an incoming video signal, for example, encoding in which the intra / inter prediction is performed. A CU includes a coding block (CB) for
Petition 870190065456, of 7/11/2019, p. 21/70
15/48 a luma component and a CB for two chroma components corresponding to the luma component. In HEVC, a CU size can be determined as one of 64 χ 64, 32 χ 32, 16 χ 16 and 8 χ 8.
[084] With reference to Figure 3, the root node of a quad-tree is related to a CTU. The quad-tree is split until a leaf node is reached. The leaf node corresponds to a CU.
[085] This is described in more detail. The CTU corresponds to the root node and has the lowest depth value (that is, depth = 0). A CTU may not be divided depending on the characteristics of an incoming video signal. In this case, the CTU corresponds to a CU.
[086] A CTU can be divided into a quad-tree format. As a result, the lower nodes, that is, a depth 1 (depth = 1), are generated. In addition, a node (ie, leaf node) that belongs to the lower nodes with a depth of 1 and that is no longer divided, corresponds to a CU. For example, in Figure 3 (b), a CU (a), a CU (b) and a CU (j) corresponding to nodes a, b and j have been divided from the CTU, and have a depth of 1.
[087] At least one of the nodes with a depth of 1 can be divided into a quad-tree format. As a result, lower nodes with depth 1 (that is, depth = 2) are generated. In addition, a node (that is, a leaf node) that belongs to the lower nodes with a depth of 2 and that is no longer divided, corresponds to a CU. For example, in Figure 3 (b), a CU (c), a CU (h) and a CU (i) corresponding to nodes c, h and i have been divided twice from the CTU, and have a depth of 2.
[088] In addition, at least one of the nodes with a depth of 2 can again be divided into a quad-tree format. As a result, lower nodes with a depth of 3 (ie, depth = 3) are generated. In addition, a node (that is, leaf node) that belongs to the lower nodes with a depth of 3 and that is no longer
Petition 870190065456, of 7/11/2019, p. 22/70
16/48 divided, corresponds to a CU. For example, in Figure 3 (b), a CU (d), a CU (e), a CU (f) and a CU (g) corresponding to nodes d, e, f and g were divided three times from the CTU, and have a depth of 3.
[089] In the encoder, a maximum or minimum size of a CU can be determined based on the characteristics of a video image (for example, resolution) or considering the encoding rate. In addition, information about the maximum or minimum size or information capable of deriving the information can be included in a bit stream. A CU having a maximum size is called the largest coding unit (LCU), and a CU having a minimum size is called the smallest coding unit (SCU).
[090] In addition, a CU with a tree structure can be divided hierarchically with information of predetermined maximum depth (or information of maximum level). In addition, each split CU can have depth information. Since the depth information represents a division count and / or the degree of a CU, it can include information about the size of a CU.
[091] Since the LCU is divided into a quad-tree format, the SCU size can be obtained using an LCU size and the maximum depth information. Or, conversely, the size of the LCU can be obtained using a size of the SCU and the maximum depth information of the tree.
[092] For a single CU, the information (for example, a CU split flag (split_cu_flag)) that represents whether the corresponding CU is split can be forwarded to the decoder. This division information is included in all CUs, except the SCU. For example, when the value of the flag representing dividing is 'Γ, the corresponding CU is divided into four CUs, and when the value of the flag representing dividing is Ό', the corresponding CU is no longer divided, and the process processing for the corresponding CU can be
Petition 870190065456, of 7/11/2019, p. 23/70
17/48 executed.
[093] As described above, a CU is a basic unit of coding in which the intra or inter prediction is performed. HEVC splits the CU into a prediction unit (PU) to encode an incoming video signal more efficiently.
[094] A PU is a basic unit for generating a prediction block and, even in a single CU, the prediction block can be generated differently by a PU unit. However, intra prediction and inter prediction are not used together for PUs that belong to a single CU, and PUs that belong to a single CU are encoded by the same prediction method (that is, intra prediction or prediction inter).
[095] A PU is not divided into the quad-tree structure, but is divided once into a single CU in a predetermined form. This will be described with reference to the drawing below.
[096] Figure 4 is a diagram to describe a prediction unit that can be applied to the present invention.
[097] A PU is divided differently depending on whether the intra prediction mode is used or whether the inter prediction mode is used as the CU encoding mode to which the PU belongs.
[098] Figure 4 (a) illustrates a PU if the intra prediction mode is used, and Figure 4 (b) illustrates a PU if the inter prediction mode is used.
[099] With reference to Figure 4 (a), assuming the size of a single CU is 2N x 2N (N = 4, 8, 16 and 32), the single CU can be divided into two types (ie 2N x 2N or N x N).
[0100] In this case, if a single CU is divided into the PU in a 2N χ 2N way, it means that only one PU is present in a single CU.
[0101] However, if a single CU is divided into the PU in an N χ N form, a
Petition 870190065456, of 7/11/2019, p. 24/70
18/48 a single CU is divided into four PUs, and different prediction blocks are generated for each PU unit. However, this PU split can be performed only if the CB size for the CU luma component is the minimum size (that is, the case in which a CU is a SCU).
[0102] With reference to Figure 4 (b), assuming the size of a single CU is 2N χ 2N (N = 4, 8, 16 and 32), a single CU can be divided into eight types of PU (ie , 2N χ 2N, N χ N, 2N χ N, N χ 2N, nL χ 2N, nR χ 2N, 2N χ nU and 2N χ nD).
[0103] As in the intra prediction, PU division of NxN form can be performed only if the CB size for the CU luma component is the minimum size (that is, the case in which a CU is a SCU).
[0104] The inter prediction supports the division of PU in the form of 2N χ N which is divided in a horizontal direction and in the form of N χ 2N which is divided in a vertical direction.
[0105] In addition, inter prediction supports the division of PU in the form of nL χ 2N, nR χ 2N, 2N χ nU and 2N χ nD, which is an asymmetric movement division (AMP). In this case, ‘n’ means 1/4 of the value of 2N. However, AMP cannot be used if the CU to which the PU belongs is the CU of minimum size.
[0106] In order to efficiently encode the incoming video signal into a single CTU, the optimal division structure of the encoding unit (CU), the prediction unit (PU) and the transform unit (TU) can be determined based on a minimum distortion rate value through the processing process as follows. For example, for the optimal CU division process on a 64 χ 64 CTU, the cost of rate distortion can be calculated by dividing a 64 χ 64 size CU to an 8x8 size CU. The detailed process is as follows.
[0107] 1) The ideal division structure of a PU and TU that generates the minimum distortion rate value is determined by performing inter / intra, transformed /
Petition 870190065456, of 7/11/2019, p. 25/70
19/48 quantization, decanting / inverse transform and entropy coding in CU size 64 χ 64.
[0108] 2) The optimal division structure of a PU and a TU is determined to divide the CU 64 χ 64 into four CUs of size 32 χ 32 and generate the minimum distortion rate value for each CU 32 χ 32.
[0109] 3) The optimal division structure of a PU and TU is determined to further divide the CU 32 χ 32 into four CUs of size 16 χ 16 and generate the minimum distortion rate value for each CU 16 χ 16.
[0110] 4) The ideal split structure of a PU and a TU is determined to further divide the CU 16 χ 16 into four CUs of size 8 χ 8 and generate the minimum distortion rate value for each 8x8 CU.
[0111] 5) The ideal division structure of a CU in the 16 χ 16 block is determined by comparing the 16 χ 16 CU distortion rate value obtained in process 3) with the addition of the distortion rate distortion value of the four 8x8 CUs obtained in process 4). This process is also performed for the remaining three 16 χ 16 CUs in the same way.
[0112] 6) The optimal CU split structure in the 32 χ 32 block is determined by comparing the CU 32 χ 32 distortion rate value obtained in process 2) with the addition of the rate distortion value of the four CUs 16 χ 16 that is obtained in process 5). This process is also performed for the remaining three 32 χ 32 CUs in the same way.
[0113] 7) Finally, the optimal division structure of CU in the 64 χ 64 block is determined by comparing the distortion rate value of CU 64 χ 64 obtained in process 1) with the addition of the distortion rate value of the four 32 χ 32 CUs obtained in process 6).
[0114] In intra prediction mode, a prediction mode is selected as a PU unit, and prediction and reconstruction are performed in prediction mode
Petition 870190065456, of 7/11/2019, p. 26/70
20/48 selected on a real TU unit.
[0115] A TU means a basic unit in which actual prediction and reconstruction are performed. A TU includes a transform block (TB) for a luma component and a TB for two chroma components corresponding to the luma component.
[0116] In the example in Figure 3, as in an example in which a CTU is divided into the quad-tree structure to generate a CU, a TU is divided hierarchically from a CU to be coded in the quad-tree structure.
[0117] The division of TUs from a CU can be divided into smaller and lower TUs, because a TU is divided into the quad-tree structure. In HEVC, the size of a TU can be determined to be 32 χ 32, 16 χ 16, 8 χ 8 and 4 χ 4.
[0118] With reference again to Figure 3, the root node of a quad-tree is considered to be related to a CU. The quad-tree is divided until a leaf node is reached and the leaf node corresponds to a TU.
[0119] This is described in more detail. A CU corresponds to a root node and has the lowest depth value (that is, depth = 0). A CU may not be divided depending on the characteristics of an input image. In this case, the CU corresponds to a TU.
[0120] A CU can be divided into a quad-tree shape. As a result, lower nodes with depth 1 (depth = 1) are generated. In addition, a node (that is, the leaf node) that belongs to the lower nodes having a depth of 1 and that is no longer divided, corresponds to a TU. For example, in Figure 3 (b), a TU (a), a TU (b) and a TU (j) corresponding to nodes a, b and j are divided once from a CU and have a depth of 1.
[0121] At least one of the nodes with a depth of 1 can be divided into a quad-tree shape again. As a result, lower nodes having a depth of 2 (that is, depth = 2) are generated. In addition, a node (that is,
Petition 870190065456, of 7/11/2019, p. 27/70
21/48 leaf node) which belongs to the lower nodes having a depth of 2 and which is no longer divided, corresponds to a TU. For example, in Figure 3 (b), a TU (c), a TU (h) and a TU (i) corresponding to node c, h and I have been divided twice from the CU and have a depth of 2.
[0122] In addition, at least one of the nodes having a depth of 2 can again be divided into a quad-tree shape. As a result, lower nodes with a depth of 3 (that is, depth = 3) are generated. In addition, a node (that is, the leaf node) that belongs to the lower nodes with a depth of 3 and that is no longer divided, corresponds to a TU. For example, in Figure 3 (b), a TU (d), a TU (e), a TU (f) and a TU (g) corresponding to nodes d, e, f and g were divided three times from the CU and have a depth of 3.
[0123] A TU with quad-tree structure can be divided hierarchically with information of predetermined maximum depth (or information of maximum level). In addition, each TU can have depth information. Depth information can include information about the size of the TU because it indicates the division number and / or the grade of the TU.
[0124] The information (for example, a TU split flag 'split_transform_flag') indicating whether a corresponding TU has been split in relation to a TU can be transferred to the decoder. The split information is included in all TUs other than a minimum size TU. For example, if the value of the flag that indicates whether a TU has been divided is ‘T, the corresponding TU will be divided into four TUs. If the value of the flag indicating whether a TU has been divided is Ό ’, the corresponding TU will no longer be divided.
[0125] In an existing image compression encoding / decoding technique, an encoder generates a prediction block (or a current processing block) from a current block by inter prediction or intra prediction and subtracts the prediction block from an image original (or input image) (or block
Petition 870190065456, of 7/11/2019, p. 28/70
Residual 22/48). The encoder performs a transform for the generated residual signal, quantifies the transformed residual signal, and performs entropy coding for a quantized coefficient. The decoder receives a signal emitted from the encoder and performs entropy decoding for the signal. The decoder generates a residual signal performing decanting and an inverse transform for the entropy decoded signal. In addition, the decoder generates the prediction block of the current block and reconstructs the current block by adding the residual signal.
[0126] That is, in the existing image compression encoding / decoding technique, the encoder performs a primary transform (or a core transform) for the signal of a pixel region to generate a signal from the transform domain and performs quantization for the signal of the transform domain. However, if the signal from the transform domain is transformed (that is, a secondary transform is performed) into the signal from the transform domain, the compression performance can be further improved compared to the existing technique and the amount of data residual signal signaled to the decoder can be reduced.
[0127] The present invention proposes a method for efficiently dimensioning the size of a transform core used for such a secondary transform.
[0128] In addition, the present invention proposes a method for executing the secondary transform using the transform core selected adaptively according to the block size.
[0129] Furthermore, the present invention provides a method for adaptively selecting the core adapted for the secondary transform, transmitting the size of the transform core.
[0130] In addition, the present invention provides a method for determining whether to apply the secondary transform or an application range using the signal
Petition 870190065456, of 7/11/2019, p. 29/70
Residual 23/48.
[0131] In addition, the present invention provides a method for deriving the size of the transform core applied to the secondary transform using the residual signal.
[0132] Mode 1 [0133] In one embodiment of the present invention, the encoder / decoder can adaptively select the size of the transform core to be used for the secondary transform according to the size of the processing block and execute the transform secondary using the selected transform core.
[0134] Here, the processing block can refer to a unit (or block) in which an encoding / decoding processing process such as prediction, transform and / or quantization is performed and can be called as the block, the unit processing, and the like for convenience of description. In addition, when the processing block is used as a unit on which the transform process is performed, the processing block can be called the coding block, the coding unit, the transform block, the transform unit, etc.
[0135] According to the modality, the encoder can perform the primary transform and then perform the secondary transform for the residual signal primarily transformed before performing the quantization. In this case, the decoder can perform the decanting for the residual signal received from the encoder and execute the secondary transform before executing the primary reverse transform for the decantized residual signal.
[0136] In an image compression technique in which the transform is performed in various block sizes, such as a Quadtree plus Binarytree (QTBT) structure, the encoder / decoder can apply the transform
Petition 870190065456, of 7/11/2019, p. 30/70
Secondary 24/48 using a secondary transform (or secondary transform) core suitable for the block size.
[0137] Figure 5 is a diagram for describing a method for determining a size of a transform core used for secondary transform based on the size of a block as a modality to which the present invention is applied.
[0138] The method for determining the size of the transform core based on the block size described in the present invention can be applied to both the encoder and the decoder and is described based on the decoder for convenience of description.
[0139] With reference to Figure 5, a case in which a non-separable transform is applied to the secondary transform is assumed.
[0140] The decoder extracts the quantized transform coefficient from the bit stream received from the encoder and performs the decanting for the extracted quantized transform coefficient (S501). In this case, the decoder can perform entropy decoding for the bit stream received from the encoder in order to extract the quantized transform coefficient.
[0141] The decoder determines the size of the secondary transform core applied to the current block using the size (that is, a width and height of the current block) of the current block (S502). For example, the decoder can determine the secondary transform core determined according to the current block size (or mapped to the current block size) among the predetermined secondary transform cores with sizes of 4 χ 4, 8 χ 8, 16 χ 16 32 χ 32 and 64 χ
64.
[0142] For example, when the width or height of the current block is less than 8, the decoder can perform the secondary transform using the secondary transform core having the size of 4 x 4 for the current block. Instead,
Petition 870190065456, of 7/11/2019, p. 31/70
25/48 when the width or height of the current block is equal to or greater than 8, the decoder can perform the secondary transform using the secondary transform core with the size of 8 x 8.
[0143] The decoder performs the secondary inverse transform for the decoupled transform block using the secondary transform core determined in step S502 (S503).
[0144] The unquantified transform block represents a 2D arrangement of the unquantified transform coefficient extracted in step S501 above.
[0145] In addition, when the size of the secondary reverse transform core applied to the current block is smaller than the size of the current block, the decoder can perform the secondary reverse transform only for an upper left region (that is, a domain of the lower frequency) of the current block, partition the current block into sub-blocks of a unit the size of the secondary reverse transform core, and apply the secondary reverse transform into bits of the sub-block. In other words, the decoder can apply the secondary reverse transform only to the upper left region of the secondary reverse transform core in the current block or apply the secondary reverse transform to the entire current block in units of the secondary reverse transform core size.
[0146] The decoder can generate the residual block of the current block, executing the primary reverse transform for the current block that is subjected to the secondary reverse transform.
[0147] Figures 6 and 7 are diagrams to describe a method for determining a size of a transform core used for a secondary transform based on a block width and height as a modality to which the present invention is applied.
[0148] With respect to Figures 6 and 7, the case in which the transform does not
Petition 870190065456, of 7/11/2019, p. 32/70
26/48 separable is applied to the secondary transform is assumed.
[0149] The decoder extracts the quantized transform coefficient from the bit stream received from the encoder and performs the decanting for the extracted quantized transform coefficient (S601 and S701). In this case, the decoder can perform entropy decoding for the bit stream received from the encoder in order to extract the quantized transform coefficient.
[0150] The decoder determines the size of the secondary transform core applied to a horizontal direction of the current block using the current block width (S602 and S702). In addition, the decoder determines the size of the secondary transform core applied to a vertical direction of the current block using the height of the current block (S603 and S703). For example, the decoder can determine each secondary transform core determined according to the current block width or height (or mapped to the current block width or height) between the predetermined secondary transform cores with sizes of 4, 8 , 16 32 and 64.
[0151] Here, Figure 6 illustrates a case where the transform core applied to the horizontal direction and the transform core applied to the vertical direction are not distinguished and Figure 7 illustrates a case where the transform core applied to the horizontal direction and the transform core applied to the vertical direction are distinguished.
[0152] The decoder performs the secondary inverse transform for the decoupled transform block using the secondary transform core determined in steps S602, S603, S702 and S703 (S604 and S704).
[0153] In addition, when the size of the secondary reverse transform core applied to the current block is less than the width or height of the current block, the decoder can apply the secondary reverse transform only to the upper left region (i.e., the lowest frequency domain) of the current block and
Petition 870190065456, of 7/11/2019, p. 33/70
27/48 partition the width or height of the current block into units of the secondary reverse transform core size and perform the secondary reverse transform for the partitioned blocks.
[0154] The decoder can generate the residual block of the current block, executing the primary reverse transform for the transform block that is subjected to the primary reverse transform.
[01551Modality 2 [0156] In one embodiment of the present invention, the encoder can selectively and adaptively select the size of the transform core applied to the processing block and transmit the size information of the transform core to the decoder.
[0157] In the example of Modality 1 described above, when the current block size is large, a transform core of a relatively larger size can be selected. However, even if the current block size is relatively larger, applying a smaller transform core can be advantageous in terms of compression performance. Consequently, transform cores with various sizes can be applied by signaling the core information having an optimized size for the current block to be signaled to the decoder regardless of the current block size, thereby increasing the compression performance.
[0158] Figure 8 is a diagram to describe a method for executing the secondary transform using a syntax that indicates the size of the transform core used for the secondary transform as a modality to which the present invention is applied.
[0159] The method for determining the size of the transform core described in the present invention can be applied to both the encoder and the decoder and is described based on the decoder for convenience of description.
Petition 870190065456, of 7/11/2019, p. 34/70
28/48 [0160] With reference to Figure 8, the case in which the non-separable transform is applied to the secondary transform is assumed.
[0161] Step S801 can be performed similarly to step S501 in Figure 5.
[0162] The decoder determines the size of the transform core applied to the current block using the syntax that indicates the size of the transform core used for the secondary transform (S802). In this case, a step of parsing a syntax with a transform core size from the bit stream can be added before step S802.
[0163] Step S803 can be performed similarly to step S503 in Figure 5.
[0164] Furthermore, even when a separable transform is applied to the secondary transform as the case illustrated in Figures 6 and 7 above, the method proposed in the modality can be applied in a method that is the same as the method described in Figure 8.
[Q165] Mode 2-1 [0166] In the mode of the present invention, a method for transmitting the size of the transform core applied to the secondary transform in a method of compressing a block structure in which the transform block and the block of coding are equal to each other is proposed.
[0167] For example, in the case of a QTBT structure in which prediction, transform and quantization are performed in the same block unit (coding block and coding unit), additional partitioning in the transform block (or transform unit ) in a transform procedure may not be performed. In this case, the encoder can select the size of the secondary transform core applied to the secondary transform and signal to the decoder the selected size of the secondary transform core.
Petition 870190065456, of 7/11/2019, p. 35/70
29/48 in units of the coding block or a higher level (for example, sequence, image, slice, or CTU).
[0168] Figure 9 is a flow chart showing a method for determining a size of a transform core used for a secondary transform as a modality to which the present invention is applied.
[0169] With reference to Figure 9, it is assumed that the method for determining the size of the transform core described in the modality is applied to the block structure in which the transform block and the coding block are determined to be equal to one other.
[0170] The decoder checks whether the width and height of the current block are greater than 4 (S901).
[0171] When the width and height of the current block are greater than 4 as a result of the determination in step S901, the decoder analyzes the syntax indicating the size of the secondary transform core (S902) and checks the size of the transform core secondary applied to the current block (S903). For example, when a non-separable secondary transform (NSST) is applied, the syntax can be a syntax indicating the size of the NSST kernel.
[0172] When the syntax indicates a 4x4 core as a result of verification in step S903 or if the width or height of the current block is equal to or less than 4 as the result of determination in step S901, the decoder applies the secondary transform for the current block using the transform core with size 4x4 (S904).
[0173] As the result of the verification in step S903, when the syntax indicates the core 8 x 8, the decoder applies the secondary transform to the current block using the transform core having the size 8x8 (S905).
[0174] That is, even if the width and height of the current block are greater than 4, if the syntax transmitted from the encoder indicates the size of 4 χ 4, the
Petition 870190065456, of 7/11/2019, p. 36/70
The decoder can perform the secondary transform for the block region having the size 4 x 4, the block region having the size 8 x 8, or an entire region of the current block using the secondary transform core having the size 4x4.
[0175] In Figure 9, the method is described assuming that the transform cores of size 4 χ 4 and size 8x8 are applied to the secondary transform, but the present invention is not limited to these. That is, a method proposed in the modality can be applied using the transform cores having various sizes, as well as the transform cores of size 4 x 4 and size 8x8. In addition, the current block is larger than a block having a minimum size to which the secondary transform is applied, the decoder can determine the size of the secondary transform core applied to the current block based on the syntax information transmitted from the encoder .
[0176] In one embodiment, the syntax indicating the secondary transform core can be transmitted from the encoder to the decoder in units of the sequence, image, slice, encoding tree unit (CTU), or the block of coding.
[Q177] Mode 2-2 [0178] In the mode of the present invention, a method for transmitting the size of the transform core applied to the secondary transform in a method of compressing a block structure in which the transform block and the block of coding are not equal to each other is proposed.
[0179] The method proposed in the modality can be applied to a block structure in which the transform block and the coding block can be determined to be different from each other, unlike the method in Mode 2-1 described above. For example, the transform block (or transform unit) representing the unit on which the transform and quantization are performed can be a block that is partitioned from the coding block.
Petition 870190065456, of 7/11/2019, p. 37/70
31/48 [0180] In one embodiment, the encoder can transmit to the decoder a syntax indicating the size of the transform core in units of the coding block. In that case, the decoder can select the core size based on the syntax transmitted from the encoder in units of the transform block in the coding block and apply the secondary transform. In this case, the decoder can perform the secondary transform using the minimum size secondary transform core, regardless of the syntax received from the encoder, when the transform block is the same size as the minimum size block to which the secondary transform is applied. .
[0181] Figure 10 is a flow chart showing a method for determining a size of a transform core used for secondary transform as a modality to which the present invention is applied.
[0182] With reference to Figure 10, it is assumed that the method of determining the size of the transform core described in the modality is applied to the block structure in which the transform block and the coding block are determined individually (or hierarchically) . In addition, it is assumed that the syntax indicating the secondary transform core is transmitted in units of the coding block.
[0183] The decoder checks whether the width and height of the current coding block are greater than 4 (S1001).
[0184] When the width and height of the current coding block are greater than 4 as the result of the determination in step S1001, the decoder analyzes the syntax indicating the size of the secondary transform core (S1002).
[0185] Subsequently, the decoder can determine the size of the secondary transform core during the cycle in units of the transform block (or transform unit) in the current coding block.
[0186] Specifically, the decoder checks whether the transform block
Petition 870190065456, of 7/11/2019, p. 38/70
32/48 current in the current coding block is a last transform block (S1003).
[0187] As the result of the verification in step S1003, the decoder checks whether the width and height of the current transform block are greater than 4 until the current transform block becomes the last transform block (S1004). When the width and height of the current block are greater than 4 as a result of the determination in step S1004, the decoder checks the secondary transform core applied to the current transform block using the syntax analyzed in step S1002 (S1005). For example, when the non-separable secondary transform (NSST) is applied, the syntax can be the syntax indicating the size of the NSST kernel.
[0188] When the syntax indicates the 4x4 core as the result of the verification in step S1005, when the width or height of the current coding block is equal to or less than 4 as the result of the determination in step S1001, or when the width or the height of the current transform block is equal to or less than 4 as the result of the determination in step S1004, the decoder applies the secondary transform to the current block using the transform core having the size 4x4 (S1006).
[0189] As the result of the verification in step S1005, when the syntax indicates the 8 x 8 core, the decoder applies the secondary transform to the current transform block using the 8x8 size transform core (S1007).
[0190] That is, the decoder can analyze the syntax indicating the transform core in units of the coding block. The decoder can then determine the size of the secondary transform core using the syntax analyzed in units of the transform block within the coding block. That is, even when the width and height of the current block are greater than 4, if the syntax transmitted from the encoder indicates the size 4 x 4, the decoder can perform the secondary transform for the block region with size 4x4,
Petition 870190065456, of 7/11/2019, p. 39/70
33/48 the block region having the size 8 x 8, or an entire region of the current transform block using the secondary transform core having the size 4 χ 4.
[0191] In Figure 10, the method is described assuming that the transform cores of size 4 χ 4 and size 8x8 are applied to the secondary transform, but the present invention is not limited to these. That is, a method proposed in the modality can be applied using the transform cores having various sizes, as well as the transform cores of size 4 x 4 and size 8x8. In addition, the current transform block is larger than a block having a minimum size to which the secondary transform is applied, the decoder can determine the size of the secondary transform core applied to the current transform block based on the syntax information transmitted. from the encoder.
[0192] In one embodiment, the syntax indicating the secondary transform core can be transmitted from the encoder to the decoder in units of the sequence, image, slice, CTU or coding block in addition to the coding block unit .
[0193] Furthermore, in an embodiment of the present invention, when a block structure in which the luma component and the chroma component are different is provided, the encoder can signal the transform core size information to the decoder as the examples from Tables 1 to 4 below.
[0194] In the examples in Tables 1 to 4 below, it is assumed that the block structures of the luma component (or a luma channel) and the chroma component (or a chroma channel) are determined to be different in the case of slice I and the block structures of the luma component and the chroma component are determined to be equal in the case of slice B. In the case of slice B, since the block structures of the luma component and the chroma component are determined to be equal,
Petition 870190065456, of 7/11/2019, p. 40/70
34/48 the encoder can transmit to the decoder a flag indicating the size of the transform core for the luma component only. In addition, it is assumed that the current block size is larger than the 8x8 block size. When the size of the current block is not larger than the 4 x 4 block or when the current block is the minimum size block, the transform core having the size 4x4 can be applied to the current block.
[0195] [Table 1]
Fal ia 1 Slice B Luma Chroma Luminance Chroma Flare gun O X O4x4 core O X O X Core 8x8 O O O O
[0196] With respect to Table 1, in slice I, the encoder may not signal a decoder to the decoder indicating the size of the transform core for the chroma component. In this case, the encoder / decoder can apply the 8 x 8 size transform core to a chroma component block having the size 8 x 8 or more, regardless of the luma component flag information Even for the slice B chroma component, the encoder / decoder can apply the 8 x 8 size transform core to the chroma component block having the size 8 x 8 or more, regardless of the luma component flag information.
[0197] [Table2]
Slice 1 Slice B Luma Chroma Luma Chroma Flare gun O X O4x4 core O X O O Core 8x8 O O O O
[0198] With respect to Table 2, in slice I, the encoder may not signal the decoder to the flag indicating the size of the transform core for the chroma component. In this case, the encoder / decoder can apply the 8 x 8 size transform core to a chroma component block having the
Petition 870190065456, of 7/11/2019, p. 41/70
35/48 size 8 χ 8 or more, regardless of the luma component flag information. However, in the case of slice B, since the block structures can be determined to be equal, the encoder / decoder can apply the transform core of size 4 x 4 or 8 x 8 to the chroma component block having the size 8 x 8 or more according to the luma component flag information.
[0199] [Table 3]
Fal ia 1 Slice B Luma Chroma Luma Chroma Flare gun O O O4x4 core O O O O Core 8x8 O O O O
[0200] With respect to Table 3, in slice I, the encoder can signal the decoder to the flag indicating the size of the transform core for the chroma component. In this case, the encoder / decoder can determine the size of the transform core and apply the secondary transform using the signaled information for each component. However, in the case of slice B, since the block structures can be determined to be equal, the encoder / decoder can apply the transform core of size 4 x 4 or 8 x 8 to the component chroma block with size 8 x 8 or more according to the luma component flag information.
[0201] [Table 4]
Fal ia 1 Slice B Luma Chroma Luma Chroma Flare gun O X O4x4 core O O O O Core 8x8 O 0 O O
[0202] With respect to Table 4, in slice I, the encoder may not signal the decoder to the flag indicating the size of the transform core for the chroma component. In this case, the encoder / decoder can apply the transform core of size 4 x 4 or 8 x 8 to the chroma component block having the
Petition 870190065456, of 7/11/2019, p. 42/70
36/48 size 8 χ 8 or more according to the luma component flag information. In addition, in the case of slice B, since the block structures can be determined to be equal, the encoder / decoder can apply the transform core of size 4 x 4 or 8 x 8 to the component chroma block with size 8 x 8 or more according to the signaling information of the luma component.
[0203] Mode 3 [0204] In one embodiment of the present invention, the encoder / decoder can determine whether to apply the secondary transform or adaptively select the size of the transform core using characteristics of the residual signal of the block signal in the application of the secondary transformed. In the mode 2 described above, the decoder can receive from the encoder the syntax that indicates the size of the transform core, while in the mode, the decoder can derive the size of the transform core using the characteristics in the block without receiving an additional syntax.
[0205] Mode 3-1 [0206] In one embodiment of the present invention, the encoder / decoder can determine whether to apply the secondary transform considering the characteristics of the residual signal in the block. For example, the encoder / decoder can determine whether to apply the secondary transform using a degree of residual signal distribution, the number of residual signals, or the size of the residual signal.
[0207] It is possible to save the bits used in the flag to indicate the secondary transform and improve the compression performance, determining whether to apply the secondary transform based on the residual signal.
[0208] In the following, the present invention will be described assuming that the non-separable secondary transform (NSST) is used as a secondary transform, but the present invention is not limited to it. Other transformations
Petition 870190065456, of 7/11/2019, p. 43/70
37/48 can be applied as the secondary transform.
[0209] Figure 11 is a diagram illustrating a method for determining whether to apply a secondary transform using a residual signal according to an embodiment of the present invention.
[0210] With reference to Figure 11, it is assumed that the size of the current block is 16 x 16 and the NSST is applied to block 8 χ 8 1101 in the upper left end.
[0211] As shown in Figure 11, if the residual signal is not distributed in the upper left block 8 χ 8 1101, there is no difference between the case of applying the NSST and the case of not applying the NSST and an unnecessary bit can be used to signal whether to apply the NSST or the core size.
[0212] Therefore, in an embodiment of the present invention, a method for determining whether to apply the secondary transform based on the residual signal distribution is proposed to solve this problem.
[0213] The encoder / decoder can determine whether to apply the secondary transform according to whether the residual signal is distributed in the upper left region 8 x 8 1101 of the current block.
[0214] In one embodiment, the encoder / decoder may not execute the secondary transform when there is no residual signal in the upper left 8 x 8 1101 region to which the secondary transform is applied. In other words, the secondary transform can be applied to the corresponding region when there are one or more residual signs in the upper left region 8x81101.
[0215] Alternatively, the encoder / decoder can apply the secondary transform when the number of residual signals that exist in the upper left region 8 x 8 1101 is greater than a specific limit. When the number of residual signals is equal to or less than the specific limit, the encoder / decoder may not apply the secondary transform.
[Q216] Mode 3-2
Petition 870190065456, of 7/11/2019, p. 44/70
38/48 [0217] In an embodiment of the present invention, the encoder / decoder can determine the range of application of the secondary transform considering the characteristics of the residual signal in the block. When the secondary transform is applied to the entire block, the larger the block size, the greater the complexity and the compression performance can be degraded.
[0218] Consequently, in order to solve this problem, the present invention proposes a method to determine whether to apply the secondary transform in units of sub-regions in the block. In accordance with the embodiment of the present invention, complexity can be reduced by applying the secondary transform to a block in which one or more residual signals or a specific number or more of residual signals exist.
[0219] Figures 12 and 13 are diagrams that illustrate a method for determining whether to apply the secondary transform using a residual signal according to an embodiment of the present invention.
[0220] With reference to Figure 12, the encoder / decoder can partition the current block (or the current residual block) into sub-blocks (or sub-regions) and determine whether to apply the secondary transform based on the residual signals in units partitioned sub-blocks.
[0221] The encoder may or may not transmit the flag indicating whether to apply the secondary transform to the decoder or vice versa. When the flag indicating whether to apply the secondary transform is signaled by the encoder, the decoder can determine whether to apply the secondary transform by the received flag. In addition, when the received flag indicates the application of the secondary transform, the decoder can partition the current block into a plurality of sub-blocks and determine whether to apply the secondary transform in units of the sub-block based on the residual signal in each sub -block.
[0222] When the flag indicating whether to apply the transform
Petition 870190065456, of 7/11/2019, p. 45/70
39/48 secondary is not signaled by the encoder, the decoder can partition the current block into the plurality of sub-blocks and determine whether to apply the secondary transform in units of the sub-block based on the residual signal in each sub-block.
[0223] When there is no residual signal in the partitioned sub-block, the encoder / decoder can determine that the secondary transform is not applied and when there is one or more residual signals in the partitioned sub-block, the encoder / decoder can determine the application of the secondary transformed.
[0224] For example, when the current block size is 16 x 16, the encoder / decoder can partition the current block into 8x8 size sub-blocks. When there is no residual signal in an upper left sub-block 1201 as illustrated in Figure 12, the encoder / decoder may not apply the secondary transform to the upper left sub-block 1201.
[0225] Conversely, as illustrated in Figure 12, when there is a residual signal in an upper right sub-block 1202, a lower left sub-block 1203, and a lower right sub-block 1204, the encoder / decoder can apply the secondary to the corresponding sub-blocks 1202, 1203 and 1204.
[0226] With reference to Figure 13, when there are residual signs of a specific number (or limit) or more in the partitioned sub-block, the encoder / decoder can determine that the secondary transform is not applied and when there are residual signs of a number specific or more in the partitioned sub-block, the encoder / decoder can determine to apply the secondary transform.
[0227] For example, when the current block size is 16 x 16, the encoder / decoder can partition the current block into 8x8 size sub-blocks. When a specific number or more of residual signals does not exist in the upper left sub-block, upper right sub-block and lower left sub-block as shown in Figure 13, the encoder / decoder may not apply the
Petition 870190065456, of 7/11/2019, p. 46/70
40/48 secondary transform in the upper left sub-block, the upper right sub-block and the lower left sub-block. Here, a specific number representing a limit to determine whether to apply the secondary transform can have a predetermined value. In the modality, it is assumed that the specific number is 2.
[0228] Conversely, when there is a specific number or more of residual signals in a lower right sub-block 1301 illustrated in Figure 13, the encoder / decoder can apply the secondary transform to the lower right sub-block 1301.
[Q229] Mode 3-3 [0230] In one embodiment of the present invention, the encoder / decoder can determine the size of the secondary transform core considering the characteristics of the residual signal in the block. When the secondary transform is performed using a secondary transform core of a predetermined size, regardless of a residual signal distribution range, the secondary transform can be applied to an unnecessary region when the residual signal is distributed only in a relatively small region, and as a result, the compression efficiency can be reduced.
[0231] The encoder / decoder can adaptively determine the size of the secondary transform core based on the degree of distribution of the residual signal.
[0232] Figure 14 is a diagram illustrating a method for determining a size of a secondary transform core using a residual signal according to an embodiment of the present invention.
[0233] With reference to Figure 14, the encoder / decoder can check the distribution region of the residual signal in the current block. For example, the encoder / decoder can check whether the residual signal exists for each applicable size of the secondary transform core. The encoder / decoder can determine the
Petition 870190065456, of 7/11/2019, p. 47/70
41/48 minimum size of the transform core, which includes the existing region of the residual signal between the applicable transform cores, according to the transform core applied to the secondary transform of the current block.
[0234] For example, when size 4 x 4 and size 8x8 transform cores exist as the applicable secondary transform core and the residual signal exists only in the size 4x4 region as illustrated in Figure 14, the encoder / decoder can determine the size of the transform core applied to the secondary transform of the current block as the 4x4 size.
[0235] Regarding the modalities described above, the respective modalities can be applied independently and several modalities can be used in combination.
[0236] Figure 15 is a diagram illustrating a method of decoding a video signal according to an embodiment of the present invention.
[0237] In the following, the decoder will first be described for convenience of description in the description of the modality, but the method of decoding the video signal according to the present invention can be performed on the encoder and decoder in the same way.
[0238] The decoder generates a quantized transform block by executing entropy decoding for the video signal (S1501). Specifically, the decoder can extract quantized transform coefficients by entropy decoding the bit stream received from the encoder to extract the quantized transform coefficients. In addition, the decoder can generate the quantized transform blocks from a 2D array, with the quantized transform coefficients according to a predetermined scanning order.
[0239] The decoder generates an unquantified transform block by executing the decanting for the quantized transform block (S1502).
Petition 870190065456, of 7/11/2019, p. 48/70
42/48 [0240] The decoder determines whether to apply the secondary inverse transform based on information related to a non-zero coefficient in the unquantified transform block (S1503).
[0241] As described above in Figure 11, step S1503 can include checking if there are one or more non-zero coefficients in the specific upper left region of the current block (that is, the unquantified transform block). When one or more non-zero coefficients exist in the specific region, the secondary inverse transform can be applied to the current block.
[0242] In addition, step S1503 may include checking the number of non-zero coefficients in the specific upper left region of the current block. When the number of non-zero coefficients in the specific region exceeds a specific limit, the decoder can apply the secondary reverse transform to the current block.
[0243] In addition, as described in Figures 12 and 13, the decoder can partition the current block into sub-blocks of a specific size and determine whether to apply the secondary reverse transform in units of the partitioned sub-blocks. In this case, the decoder can check if there are one or more non-zero coefficients in a current sub-block. As a result of the verification, when one or more non-zero coefficients exist in the current sub-block, the decoder can apply the secondary reverse transform to the current sub-block.
[0244] Alternatively, the decoder can check the number of non-zero coefficients in the current sub-block. When the number of non-zero coefficients in the specific region exceeds a specific limit, the decoder can apply the secondary reverse transform to the current sub-block.
[0245] In addition, as described in Figure 14 above, the decoder can determine the size of the secondary reverse transform core applied to the current block based on information related to the non-zero coefficient in the current block. Specifically, the size of the secondary reverse transform core
Petition 870190065456, of 7/11/2019, p. 49/70
43/48 can be determined as the size of the smallest secondary reverse transform core among the secondary reverse transform cores including the non-null coefficients that exist in a region with a specific size in a specific upper left size of the current block.
[0246] In addition, as described in Figures 8 to 10, when the current block size is larger than that of a predetermined block having a minimum size, the decoder can extract a syntax indicating the size of the secondary reverse transform core of the video signal. In addition, the decoder can determine the size of the secondary reverse transform core applied to the current block based on the syntax. The syntax that indicates the size of the secondary reverse transform core can be transmitted in units of a sequence, an image, a slice, a coding block or a transform block.
[0247] The decoder performs the secondary reverse transform for the unquantified transform block using the secondary reverse transform core applied to the unquantized transform block. For example, the secondary inverse transform can be performed using either a discrete cosine transform (DCT), a discrete sine transform (DST), a Karhunen-Loeve transform, a graph-based transform, a non-separable secondary transform ( NSST).
[0248] In addition, the decoder can generate the residual block by executing the primary reverse transform for the transform block that is subjected to the secondary reverse transform.
[0249] Figure 16 is a diagram illustrating a video signal decoding apparatus according to an embodiment of the present invention.
[0250] In Figure 16, an entropy decoding unit 1601, a decanting unit 1602, a transform determination unit
Petition 870190065456, of 7/11/2019, p. 50/70
44/48 secondary reverse 1603 and a secondary reverse transform unit 1604 are shown as separate blocks, respectively, but can be implemented as components included in the encoder and / or decoder.
[0251] With reference to Figure 16, the decoding apparatus implements the functions, procedures and / or methods proposed in Figures 5 to 15 above. Specifically, the decoding apparatus can be configured to include an entropy decoding unit 1602, a decanting unit 1602, a secondary reverse transform determination unit 1603 and a secondary reverse transform unit 1604. The entropy decoding unit 1601 and the decanting unit 1602 can be included in the entropy decoding unit 210 (in Figure 2) and in the decanting unit 220 (in Figure 2) described in Figure 2 above, respectively. In addition, the secondary reverse transform determination unit 1603 and / or the secondary reverse transform unit 1604 can be included in the reverse transform unit 230 (in Figure 2) described in Figure 2.
[0252] Entropy decoding unit 1601 generates a quantized transform block by executing entropy decoding for the video signal. Specifically, the entropy decoding unit 1601 can extract quantized transform coefficients by entropy decoding from the bit stream received from the encoder. In addition, the entropy decoding unit 1601 can generate the quantized transform blocks of the 2D matrix, with the quantized transform coefficients arranged according to a predetermined scan order.
[0253] The decanting unit 1602 generates a decantized transform block by executing decanting for the quantized transform block.
[0254] The secondary reverse transform determination unit 1603 determines whether to apply the secondary reverse transform based on
Petition 870190065456, of 7/11/2019, p. 51/70
45/48 information related to a non-zero coefficient in the unquantified transform block.
[0255] As described above in Figure 11, the secondary reverse transform determination unit 1603 can check whether there are one or more non-zero coefficients in the specific upper left region of the current block (i.e., the unquantified transform block). When one or more non-zero coefficients exist in the specific region, the secondary reverse transform determination unit 1603 can determine that the secondary reverse transform is applied to the current block.
[0256] In addition, the secondary reverse transform determination unit 1603 can check the number of non-zero coefficients in the specific upper left region of the current block. When the number of non-zero coefficients in the specific region exceeds a specific limit, the secondary reverse transform determination unit 1603 can determine that the secondary reverse transform is applied to the current block.
[0257] In addition, as described above in Figures 12 and 13, the secondary reverse transform determination unit 1603 can partition the current block into sub-blocks of a specific size and determine whether to apply the secondary reverse transform in units of the sub - partitioned blocks. In this case, the secondary reverse transform determination unit 1603 can check for one or more non-zero coefficients in a current sub-block. As a result of the verification, when one or more non-zero coefficients exist in the current sub-block, the secondary reverse transform determination unit 1603 can determine that the secondary reverse transform is applied to the current sub-block.
[0258] Alternatively, the secondary reverse transform determination unit 1603 can check the number of non-zero coefficients in the current sub-block. When the number of non-zero coefficients in the current sub-block exceeds a limit
Petition 870190065456, of 7/11/2019, p. 52/70
46/48, the secondary reverse transform determination unit 1603 can determine that the secondary reverse transform is applied to the current sub-block.
[0259] Furthermore, as described above in Figure 14, the secondary reverse transform determination unit 1603 can determine the size of the secondary reverse transform core applied to the current block based on information related to the non-zero coefficient in the current block. Specifically, the size of the secondary reverse transform core can be determined as the size of the smallest secondary reverse transform core among the secondary reverse transform cores including the non-zero coefficients that exist in a region with a specific size in a specific upper left size. the current block.
[0260] In addition, as described in Figures 8 to 10, when the current block size is larger than that of a predetermined block having a minimum size, the decoder can extract a syntax indicating the size of the secondary reverse transform core of the video signal. In addition, the secondary reverse transform determination unit 1603 can determine the size of the secondary reverse transform core applied to the current block based on the syntax. The syntax that indicates the size of the secondary reverse transform core can be transmitted in units of a sequence, an image, a slice, a coding block or a transform block.
[0261] The secondary reverse transform unit 1604 performs the secondary reverse transform for the unquantified transform block using the secondary reverse transform core applied to the unquantified transform block (S1504). For example, the secondary inverse transform can be performed using either a discrete cosine transform (DCT), a discrete sine transform (DST), a Karhunen-Loeve transform, a graph-based transform, a non-separable secondary transform
Petition 870190065456, of 7/11/2019, p. 53/70
47/48 (NSST).
[0262] In addition, the decoder can generate the residual block by executing the primary reverse transform for the transform block that is subjected to the secondary reverse transform.
[0263] In the embodiments described above, the components and features of the present invention are combined in a predetermined form. Each component or feature should be considered as an option, unless expressly stated otherwise. Each component or feature can be implemented to not be associated with other components or features. In addition, the modality of the present invention can be configured by associating some components and / or resources. The order of operations described in the modalities of the present invention can be changed. Some components or resources of any modality can be included in another modality or replaced by the component and resource corresponding to another modality. It is evident that claims that are not expressly cited in the claims are combined to form a modality or be included in a new claim for an amendment after the request.
[0264] The modalities of the present invention can be implemented by hardware, changeable software, software or combinations thereof. In the case of hardware implementation, according to the hardware implementation, the exemplary modality described here can be implemented using one or more application specific integrated circuits (ASICs), digital signal processors (DSPs), digital signal processing devices (DSPDs), programmable logic devices (PLDs), field programmable port arrangements (FPGAs), processors, controllers, microcontrollers, microprocessors and the like.
[0265] In the case of implementation by unalterable software or software, the modality of the present invention can be implemented in the form of a module,
Petition 870190065456, of 7/11/2019, p. 54/70
48/48 a procedure, function and the like to perform the functions or operations described above. A software code can be stored in memory and executed by the processor. The memory can be positioned inside or outside the processor and can transmit and receive data to / from the processor by various means.
[0266] It is evident to those skilled in the art that the present invention can be incorporated in other specific forms without abandoning the essential characteristics of the present invention. Consequently, the detailed description mentioned above should not be interpreted as restrictive in all terms and should be considered by way of example. The scope of the present invention is to be determined by rational construction of the appended claims and all modifications within an equivalent scope of the present invention are included in the scope of the present invention.
[0267] [Industrial Applicability] [0268] Here above, the preferred modalities of the present invention are described for illustrative purposes and then modifications, alterations, substitutions or additions of various other modalities will be made within the technical spirit and technical scope of the present invention described in the appended claims by those skilled in the art.

权利要求:
Claims (11)
[1]
1. Method for decoding a video signal, CHARACTERIZED by the fact that it comprises:
generate a quantized transform block by executing entropy decoding for the video signal;
generate an unquantified transform block by executing the disquantization for the quantized transform block;
determine whether to apply a secondary reverse transform based on information related to a non-zero coefficient in the decantized transform block; and carrying out the secondary reverse transform for the unquantified transform block using a secondary reverse transform core applied to the unquantized transform block.
[2]
2. Method, according to claim 1, CHARACTERIZED by the fact that the determination to apply the secondary inverse transform comprises verifying if one or more non-zero coefficients exist in a specific upper left region of the decantified transform block, and if one or more non-zero coefficients exist in the specific region, the secondary inverse transform is applied to the unquantized transform block.
[3]
3. Method, according to claim 1, CHARACTERIZED by the fact that the determination to apply the second inverse transform comprises verifying the number of non-zero coefficients in the specific upper left region of the transform block, and if the number of coefficients nonzero in the specific region exceeds a specific limit, the secondary inverse transform is applied to the unquantized transform block.
[4]
4. Method, according to claim 1, CHARACTERIZED by the fact that
Petition 870190062105, of 7/3/2019, p. 12/15
2/3 that the determination to apply the second inverse transform comprises:
partition the unquantified transform block into sub-blocks having a specific size, and determine whether to apply the secondary reverse transformation in units of the sub-block.
[5]
5. Method, according to claim 4, CHARACTERIZED by the fact that the determination to apply the secondary inverse transform in units of the sub-block comprises verifying if there is one or more non-zero coefficients in a current sub-block, and if there is one or more non-zero coefficients in the current sub-block, the secondary inverse transform is applied to the current sub-block.
[6]
6. Method, according to claim 4, CHARACTERIZED by the fact that the determination to apply the second inverse transform in units of the sub-block comprises verifying a number of non-zero coefficients in the current sub-block, and if the number of non-zero coefficients in the current sub-block exceed a specific limit, the secondary inverse transform is applied to the current sub-block.
[7]
7. Method, according to claim 1, further CHARACTERIZED by the fact that it comprises:
determine a size of the secondary reverse transform core applied to the decantized transform block based on information related to the non-zero coefficient in the decantized transform block.
[8]
8. Method for decoding a video signal according to claim 7, CHARACTERIZED by the fact that the size of the secondary reverse transform core is determined as a size of a minor secondary reverse transform core among the secondary reverse transform cores including non-zero coefficients that exist in a region having a specific size
Petition 870190062105, of 7/3/2019, p. 13/15
3/3 in a specific upper left size of the transform block unquantified.
[9]
9. Method, according to claim 1, further CHARACTERIZED by the fact that it comprises:
if the size of the unquantified transform block is larger than a block having a predetermined minimum size, extract a syntax indicating the size of the secondary reverse transform core from the video signal; and determining the size of the secondary reverse transform core applied to the transform block unquantified based on syntax.
[10]
10. Method for decoding a video signal, according to claim 9, CHARACTERIZED by the fact that the syntax that indicates the size of the secondary reverse transform core is transmitted in units of a sequence, an image, a slice, a block coding, or a transform block.
[11]
11. Apparatus for decoding a video signal, FEATURED by the fact that it comprises:
an entropy decoding unit that generates a quantized transform block by performing entropy decoding for the video signal;
a decanting unit that generates an unquantified transform block by executing decanting for the quantized transform block;
a secondary reverse transform determination unit determining whether to apply secondary reverse transform based on information related to a non-zero coefficient in the unquantified transform block; and a secondary reverse transform unit carrying out the secondary reverse transform for the decantified transform block using a secondary reverse transform core applied to the decantized transform block.

类似技术:

---

公开号 | 公开日 | 专利标题

---

BR112019013834A2|2020-01-28|method and apparatus for encoding / decoding video signal using secondary transform

---

US10917639B2|2021-02-09|Intra-prediction mode-based image processing method and device therefor

---

BR112019013831A2|2020-01-28|image processing method and apparatus for such purpose

---

CN107409207B|2020-07-28|Method and apparatus for processing image based on intra prediction mode

---

BR112019018464A2|2020-04-14|intra filtering indicator in video encoding

---

AU2018282523A1|2019-12-05|Intra filtering applied together with transform processing in video coding

---

JP2019534629A|2019-11-28|Image processing method and apparatus therefor

---

US20190200011A1|2019-06-27|Intra-prediction mode-based image processing method and apparatus therefor

---

BR112020019715A2|2021-02-09|combination of extended position-dependent intraprediction with angular modes

---

KR20190042732A|2019-04-24|Intra prediction mode based image processing method and apparatus therefor

---

RU2746706C1|2021-04-19|Method and device for encoding / decoding of images

---

US20200288146A1|2020-09-10|Intra-prediction mode-based image processing method and apparatus therefor

---

KR20210158432A|2021-12-30|Video signal processing method and device using reference sample

---

BR112021000667A2|2021-04-13|VIDEO PROCESSING METHOD AND APPLIANCE IN A VIDEO ENCODING SYSTEM TO ENCODE OR DECODE VIDEO IMAGES WITH A PARTITION RESTRICTION AND MEDIA READABLE BY NON-TRANSITIONAL COMPUTER

---

KR101642978B1|2016-07-26|Method and apparatus for video decoding by individual parsing or decoding in data unit level, and method and apparatus for video encoding for individual parsing or decoding in data unit level

---

KR20180009048A|2018-01-25|Method and apparatus for encoding / decoding image

---

BR122021003526B1|2022-02-15|VIDEO DECODING METHOD PERFORMED BY A DECODING DEVICE, VIDEO ENCODING METHOD PERFORMED BY AN ENCODING DEVICE, AND COMPUTER READable STORAGE MEDIA

---

BR112019022971B1|2022-02-15|VIDEO DECODING METHOD PERFORMED BY A DECODING DEVICE, VIDEO ENCODING METHOD PERFORMED BY AN ENCODING DEVICE, AND COMPUTER READable STORAGE MEDIA

---

KR20150051963A|2015-05-13|Method and apparatus for video decoding by individual parsing or decoding in data unit level, and method and apparatus for video encoding for individual parsing or decoding in data unit level

---

KR20180009358A|2018-01-26|Method and apparatus for encoding / decoding image

---

KR20150051962A|2015-05-13|Method and apparatus for video decoding by individual parsing or decoding in data unit level, and method and apparatus for video encoding for individual parsing or decoding in data unit level

---

KR20140140002A|2014-12-08|Method and apparatus for video decoding by individual parsing or decoding in data unit level, and method and apparatus for video encoding for individual parsing or decoding in data unit level

---

KR20140071298A|2014-06-11|Method and apparatus for video decoding by individual parsing or decoding in data unit level, and method and apparatus for video encoding for individual parsing or decoding in data unit level

同族专利:

---

公开号 | 公开日

---

JP6870096B2|2021-05-12|

---

JP2020504526A|2020-02-06|

---

KR102287305B1|2021-08-06|

---

US20190356915A1|2019-11-21|

---

EP3567858A1|2019-11-13|

---

WO2018128323A1|2018-07-12|

---

KR20190090866A|2019-08-02|

---

CN110326294A|2019-10-11|

---

EP3567858A4|2020-06-17|

引用文献:

---

公开号 | 申请日 | 公开日 | 申请人 | 专利标题

---

---

SG171883A1|2008-12-03|2011-07-28|Nokia Corp|Switching between dct coefficient coding modes|

---

US9661338B2|2010-07-09|2017-05-23|Qualcomm Incorporated|Coding syntax elements for adaptive scans of transform coefficients for video coding|

---

US9344722B2|2011-11-18|2016-05-17|Futurewei Technologies, Inc.|Scanning of prediction residuals in high efficiency video coding|

---

DK3361732T3|2012-07-02|2020-01-02|Samsung Electronics Co Ltd|Entropy coding of a video and entropy decoding of a video|

---

US9532051B2|2012-10-05|2016-12-27|Futurewei Technologies, Inc.|Architecture for hybrid video codec|

---

US20140254661A1|2013-03-08|2014-09-11|Samsung Electronics Co., Ltd.|Method and apparatus for applying secondary transforms on enhancement-layer residuals|

---

CN105516730B|2014-09-24|2018-04-24|晨星半导体股份有限公司|Video coding device and video decoded device and its coding and coding/decoding method|

---

US20210297701A1|2016-05-13|2021-09-23|Sony Corporation|Apparatus and method for image processing|

---

US10855997B2|2017-04-14|2020-12-01|Mediatek Inc.|Secondary transform kernel size selection|CA3014332A1|2016-02-11|2017-08-17|Thomson Licensing|Method and device for encoding/decoding an image unit comprising image data represented by a luminance channel and at least one chrominance channel|

---

US10855997B2|2017-04-14|2020-12-01|Mediatek Inc.|Secondary transform kernel size selection|

---

EP3761642A4|2018-04-01|2021-04-14|LG Electronics Inc.|Method and apparatus for processing video signal by applying secondary transform to partitioned block|

---

JP2021517796A|2018-04-01|2021-07-26|エルジー エレクトロニクス インコーポレイティド|Methods and equipment for processing video signals using reduced quadratic transformation|

---

US20190313108A1|2018-04-05|2019-10-10|Qualcomm Incorporated|Non-square blocks in video coding|

---

US11166044B2|2018-07-31|2021-11-02|Tencent America LLC|Method and apparatus for improved compound orthonormal transform|

---

CN111742555A|2018-09-05|2020-10-02|Lg电子株式会社|Method of encoding/decoding video signal and apparatus therefor|

---

KR20210046002A|2018-09-07|2021-04-27|주식회사 윌러스표준기술연구소|Video signal processing method and apparatus using multi-transformation kernel|

---

US11190794B2|2019-02-12|2021-11-30|Tencent America LLC|Method and apparatus for video coding|

---

CN111669582A|2019-03-09|2020-09-15|杭州海康威视数字技术股份有限公司|Method, encoding end, decoding end and system for encoding and decoding|

---

US11172211B2|2019-04-04|2021-11-09|Tencent America LLC|Method and apparatus for video coding|

---

CN113812154A|2019-05-10|2021-12-17|北京字节跳动网络技术有限公司|Multiple quadratic transform matrices for video processing|

---

GB2585030A|2019-06-25|2020-12-30|British Broadcasting Corp|Method of signalling in a video codec|

---

WO2021054787A1|2019-09-21|2021-03-25|엘지전자 주식회사|Transform-based image coding method and device therefor|

---

WO2021060827A1|2019-09-23|2021-04-01|엘지전자 주식회사|Image coding method based on transform, and device therefor|

---

WO2021110018A1|2019-12-02|2021-06-10|Beijing Bytedance Network Technology Co., Ltd.|Separable secondary transform processing of coded video|

---

CN111741302B|2020-08-07|2021-05-14|腾讯科技（深圳）有限公司|Data processing method and device, computer readable medium and electronic equipment|

法律状态:
2021-10-13| B350| Update of information on the portal [chapter 15.35 patent gazette]|

---

2021-10-19| B08F| Application dismissed because of non-payment of annual fees [chapter 8.6 patent gazette]|Free format text: REFERENTE A 4A ANUIDADE. |

---

2022-02-08| B08K| Patent lapsed as no evidence of payment of the annual fee has been furnished to inpi [chapter 8.11 patent gazette]|Free format text: EM VIRTUDE DO ARQUIVAMENTO PUBLICADO NA RPI 2650 DE 19-10-2021 E CONSIDERANDO AUSENCIA DE MANIFESTACAO DENTRO DOS PRAZOS LEGAIS, INFORMO QUE CABE SER MANTIDO O ARQUIVAMENTO DO PEDIDO DE PATENTE, CONFORME O DISPOSTO NO ARTIGO 12, DA RESOLUCAO 113/2013. |

优先权:

---

申请号 | 申请日 | 专利标题

---

US201762441587P| true| 2017-01-03|2017-01-03|

---

PCT/KR2017/015591|WO2018128323A1|2017-01-03|2017-12-27|Method and device for encoding/decoding video signal using secondary transform|

---