[go: up one dir, main page]

WO2025151003A1 - Procédé, dispositif et support d'enregistrement pour le codage/décodage d'image - Google Patents

Procédé, dispositif et support d'enregistrement pour le codage/décodage d'image

Info

Publication number
WO2025151003A1
WO2025151003A1 PCT/KR2025/000673 KR2025000673W WO2025151003A1 WO 2025151003 A1 WO2025151003 A1 WO 2025151003A1 KR 2025000673 W KR2025000673 W KR 2025000673W WO 2025151003 A1 WO2025151003 A1 WO 2025151003A1
Authority
WO
WIPO (PCT)
Prior art keywords
block
filter
information
prediction
ibc
Prior art date
Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
Pending
Application number
PCT/KR2025/000673
Other languages
English (en)
Korean (ko)
Inventor
임웅
김동현
김종호
임성창
최진수
강정원
최해철
김지영
강재하
한희지
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Electronics and Telecommunications Research Institute ETRI
Industry Academic Cooperation Foundation of Hanbat National University
Original Assignee
Electronics and Telecommunications Research Institute ETRI
Industry Academic Cooperation Foundation of Hanbat National University
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by Electronics and Telecommunications Research Institute ETRI, Industry Academic Cooperation Foundation of Hanbat National University filed Critical Electronics and Telecommunications Research Institute ETRI
Priority claimed from KR1020250004268A external-priority patent/KR20250109639A/ko
Publication of WO2025151003A1 publication Critical patent/WO2025151003A1/fr
Pending legal-status Critical Current
Anticipated expiration legal-status Critical

Links

Classifications

    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04N—PICTORIAL COMMUNICATION, e.g. TELEVISION
    • H04N19/00—Methods or arrangements for coding, decoding, compressing or decompressing digital video signals
    • H04N19/10—Methods or arrangements for coding, decoding, compressing or decompressing digital video signals using adaptive coding
    • H04N19/102—Methods or arrangements for coding, decoding, compressing or decompressing digital video signals using adaptive coding characterised by the element, parameter or selection affected or controlled by the adaptive coding
    • H04N19/103—Selection of coding mode or of prediction mode
    • H04N19/105—Selection of the reference unit for prediction within a chosen coding or prediction mode, e.g. adaptive choice of position and number of pixels used for prediction
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04N—PICTORIAL COMMUNICATION, e.g. TELEVISION
    • H04N19/00—Methods or arrangements for coding, decoding, compressing or decompressing digital video signals
    • H04N19/10—Methods or arrangements for coding, decoding, compressing or decompressing digital video signals using adaptive coding
    • H04N19/102—Methods or arrangements for coding, decoding, compressing or decompressing digital video signals using adaptive coding characterised by the element, parameter or selection affected or controlled by the adaptive coding
    • H04N19/117—Filters, e.g. for pre-processing or post-processing
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04N—PICTORIAL COMMUNICATION, e.g. TELEVISION
    • H04N19/00—Methods or arrangements for coding, decoding, compressing or decompressing digital video signals
    • H04N19/10—Methods or arrangements for coding, decoding, compressing or decompressing digital video signals using adaptive coding
    • H04N19/134—Methods or arrangements for coding, decoding, compressing or decompressing digital video signals using adaptive coding characterised by the element, parameter or criterion affecting or controlling the adaptive coding
    • H04N19/157—Assigned coding mode, i.e. the coding mode being predefined or preselected to be further used for selection of another element or parameter
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04N—PICTORIAL COMMUNICATION, e.g. TELEVISION
    • H04N19/00—Methods or arrangements for coding, decoding, compressing or decompressing digital video signals
    • H04N19/10—Methods or arrangements for coding, decoding, compressing or decompressing digital video signals using adaptive coding
    • H04N19/169—Methods or arrangements for coding, decoding, compressing or decompressing digital video signals using adaptive coding characterised by the coding unit, i.e. the structural portion or semantic portion of the video signal being the object or the subject of the adaptive coding
    • H04N19/17—Methods or arrangements for coding, decoding, compressing or decompressing digital video signals using adaptive coding characterised by the coding unit, i.e. the structural portion or semantic portion of the video signal being the object or the subject of the adaptive coding the unit being an image region, e.g. an object
    • H04N19/176—Methods or arrangements for coding, decoding, compressing or decompressing digital video signals using adaptive coding characterised by the coding unit, i.e. the structural portion or semantic portion of the video signal being the object or the subject of the adaptive coding the unit being an image region, e.g. an object the region being a block, e.g. a macroblock
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04N—PICTORIAL COMMUNICATION, e.g. TELEVISION
    • H04N19/00—Methods or arrangements for coding, decoding, compressing or decompressing digital video signals
    • H04N19/10—Methods or arrangements for coding, decoding, compressing or decompressing digital video signals using adaptive coding
    • H04N19/169—Methods or arrangements for coding, decoding, compressing or decompressing digital video signals using adaptive coding characterised by the coding unit, i.e. the structural portion or semantic portion of the video signal being the object or the subject of the adaptive coding
    • H04N19/186—Methods or arrangements for coding, decoding, compressing or decompressing digital video signals using adaptive coding characterised by the coding unit, i.e. the structural portion or semantic portion of the video signal being the object or the subject of the adaptive coding the unit being a colour or a chrominance component
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04N—PICTORIAL COMMUNICATION, e.g. TELEVISION
    • H04N19/00—Methods or arrangements for coding, decoding, compressing or decompressing digital video signals
    • H04N19/50—Methods or arrangements for coding, decoding, compressing or decompressing digital video signals using predictive coding
    • H04N19/593—Methods or arrangements for coding, decoding, compressing or decompressing digital video signals using predictive coding involving spatial prediction techniques
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04N—PICTORIAL COMMUNICATION, e.g. TELEVISION
    • H04N19/00—Methods or arrangements for coding, decoding, compressing or decompressing digital video signals
    • H04N19/70—Methods or arrangements for coding, decoding, compressing or decompressing digital video signals characterised by syntax aspects related to video coding, e.g. related to compression standards

Definitions

  • the present invention relates to a method, an apparatus and a recording medium for video encoding/decoding. Specifically, the present invention discloses a method, an apparatus and a recording medium for video encoding/decoding using filtering.
  • a decoding method including a step of obtaining prediction signal filtering information; a step of generating a prediction signal; and a step of performing filtering on the prediction signal based on the prediction signal filtering information.
  • the shape of the filter template can be derived.
  • the above prediction signal may be an Intra Block Copy (IBC) prediction signal generated by IBC.
  • IBC Intra Block Copy
  • the filter template can be selected from a plurality of filter templates having different shapes.
  • a computer-readable recording medium for storing a bitstream generated by the above encoding method.
  • a computer-readable recording medium storing a bitstream for image decoding, wherein the bitstream includes prediction signal filtering information, a prediction signal is generated, and filtering is performed on the prediction signal based on the prediction signal filtering information.
  • the filtering of the prediction signal performed based on the above prediction signal filtering information may include specifying a filter model, obtaining filtering coefficients according to the filter model, and performing the filtering of the prediction signal based on the filtering coefficients.
  • the above filtering can be performed using a filter template.
  • the filter template can be selected from a plurality of filter templates having different shapes.
  • Figure 2 shows an image segmentation structure according to one embodiment.
  • Figure 3 illustrates the structure of intra prediction according to one embodiment.
  • Figure 4 shows the structure of inter prediction for explaining the inter prediction process according to one embodiment.
  • Figure 5 shows the order in which spatial candidates are added to the candidate list according to one embodiment.
  • Figure 6 illustrates multiple in-loop filters according to an example.
  • Figure 7 shows the structure of entropy encoding and entropy decoding according to an example.
  • FIG. 8 is a flowchart of a method for predicting a target block and a method for generating a bitstream according to one embodiment.
  • FIG. 9 is a flowchart of a method for predicting a target block using a bitstream according to one embodiment.
  • Fig. 10 is a flowchart of a coding method according to an example.
  • the coding method may include an encoding method and/or a decoding method.
  • FIGS 11, 12 and 13 illustrate forms of filter templates according to examples.
  • Figure 15 illustrates the use of a 1x1 rectangular filter according to an example.
  • Figure 21 shows binarization of ibc_lic_efibc_index according to another example.
  • FIG. 23 is a first block diagram illustrating signaling of information for an IBC filter mode according to an example.
  • FIG. 24 is a second block diagram illustrating signaling of information for an IBC filter mode according to an example.
  • Figure 25 illustrates a case where the values of the entire block are used to calculate the difference according to an example.
  • Figure 26 illustrates a first case where only a region of a specific size is used for calculating the difference according to an example.
  • Figure 27 illustrates a second case where only a region of a certain size is used for calculating the difference according to an example.
  • Figure 28 illustrates an intra block copy according to an example.
  • Figure 33 shows a reference region used to derive filter coefficients according to an example.
  • the first component transmitting (or providing) information to the second component may mean that the first component directly transmits information to the second component, or may mean that the first component transmits information to the second component via another third component.
  • the information that the second component receives (or obtains) may be information transmitted by the first component, or information generated by applying a specific processing to the information transmitted by the first component.
  • the components of the embodiments may be depicted independently to represent different characteristic functions, and it does not mean that each component corresponds to a separate hardware or software configuration unit. That is, the components of the embodiments may be distinguished and listed for convenience of description. Two or more components described in the embodiments may be regarded as one component. In addition, one component described in the embodiments may be separated into a plurality of components that perform the functions of the above-mentioned component by dividing them. Embodiments in which these components are integrated and embodiments in which the components are separated are also included in the scope of the present invention as long as they do not depart from the essence of the present invention.
  • a unit can be determined for specific processing in coding.
  • a unit can be information about a specific area within an image.
  • an image can be recursively divided into multiple parts.
  • a unit can represent an area to which specific processing is applied and information about the area.
  • the type of a unit may indicate a specific processing applied to the unit. Depending on the type of the unit, a specific processing may be applied to the unit.
  • a 'specific' unit may be a unit for processing named 'specific' in coding.
  • the unit may be at least one of an original unit, a CTU, a coding unit, a prediction unit, a residual unit, a reconstructed residual unit, a transform unit, and a reconstructed unit.
  • a unit may include samples having a two-dimensional form or arrangement.
  • a 'unit' may also mean a 'block'.
  • a block may be at least one of an original block, a CTB, a Coding Block (CB), a Prediction Block (PB), a Residual Block, a Reconstructed Residual Block, a Transform Block (TB), and a Reconstructed Block.
  • a division of a unit may mean a division of a block corresponding to the unit.
  • the shape of a block can be one or more of a tetragon, a rectangular block, a square, a rectangle whose width is different from its height (that is, an oblong), a trapezoid, a triangle, a right-angled triangle, and a pentagon.
  • the width and height of the rectangle can be different from each other.
  • the shape of the block can include other geometric figures that can be expressed in two dimensions.
  • the shape of the block may be a square or a pentagon defined by excluding the area of a right triangle from the area of a rectangle, wherein the right vertex of the right triangle may be one of the vertices of the rectangle.
  • the shape of the block may be a combination of two or more of the aforementioned shapes.
  • the shape of the block may be the remainder of one of the aforementioned shapes with the other shape excluded.
  • a block may be limited to at least one of a vertically oriented block and a horizontally oriented block.
  • a vertically oriented block may mean a block whose vertical length is greater than its horizontal length.
  • a horizontally oriented block may mean a block whose horizontal length is greater than its vertical length.
  • a unit may include a luma component block (i.e., a Y block) and two chroma component blocks (i.e., at least one of a Cb block and a Cr block), and may include information for each block.
  • the information may include syntax elements.
  • Unit information may include unit type, unit size, unit depth, unit encoding order, and unit decoding order.
  • the target unit may be a block, an encoding target unit that is a target of encoding, and/or a decoding target unit that is a target of decoding.
  • the target unit may be a specific area within a target picture to which one or more specific processing of coding is to be applied.
  • a unit of a specific type may be generated by applying a specific processing to the target unit.
  • the target unit may represent a unit having a specific type for a specific processing of coding.
  • the depth of a block can represent the level of the node corresponding to the block when the blocks that make up the image are expressed in a tree structure. Or, the depth of a block can represent the number of divisions applied until the block is determined. The depth of a block can increase by 1 as the block is further divided.
  • the level of the root node may be considered to be the smallest, and the level of the leaf node may be considered to be the largest.
  • the root node may be the topmost node of the tree structure and may correspond to the first undivided block.
  • the level of the root node may be 0 or 1.
  • a node having a level of 1 may represent a block determined by the first block being divided once.
  • a node having a level of n may represent a block determined by the first block being divided n times.
  • a leaf node may be the lowest node of the tree structure.
  • a leaf node may be a node that cannot be further divided.
  • the depth of a leaf node may be a predefined maximum depth. For example, the maximum depth may be a positive integer such as 3.
  • the root node may mean a CTU.
  • a leaf node may mean at least one of a CU, a PU, and a TU.
  • PU may denote a basic unit for processing related to prediction.
  • processing related to prediction may include inter prediction, intra prediction, intra block copy (IBC) prediction, intra compensation, and motion compensation.
  • IBC intra block copy
  • a PU can be divided into multiple sub-PUs with smaller sizes than the PU.
  • Multiple sub-PUs can also be basic units for processing related to prediction.
  • a prediction unit partition generated by dividing a prediction unit can also be a prediction unit.
  • the parameter set may include at least one of a video parameter set (VPS), a sequence parameter set (SPS), a picture parameter set (PPS), an adaptation parameter set (APS), and a decoding parameter set (DPS).
  • VPS video parameter set
  • SPS sequence parameter set
  • PPS picture parameter set
  • APS adaptation parameter set
  • DPS decoding parameter set
  • Information signaled through a parameter set can be applied to pictures referencing the parameter set.
  • information in a VPS can be applied to pictures referencing the VPS.
  • Information in an SPS can be applied to pictures referencing the SPS.
  • Information in a PPS can be applied to pictures referencing the PPS.
  • a parameter set can refer to a higher-level parameter set.
  • a PPS can refer to an SPS.
  • An SPS can refer to a VPS.
  • the parameter set may include tile group information, slice header information, and tile header information.
  • a tile group may mean a group or slice including a plurality of tiles.
  • One or more MPMs can be determined based on the intra prediction mode of the reference block. There can be multiple reference blocks. One or more different MPMs can be determined depending on which intra prediction modes are used for the one or more reference blocks.
  • the reference blocks can include spatial neighboring blocks.
  • the MPM usage directive can indicate whether the MPM list is used for prediction on the target block.
  • the prediction mode may be information indicating a prediction method for a target block, such as a mode used for intra prediction or a mode used for inter prediction.
  • the prediction mode may indicate one of the modes related to prediction described in the embodiments.
  • the prediction mode may include at least one of an intra mode, an inter mode, and an intra block copy mode.
  • One or more reference image lists may be used for inter prediction for the target block. Parts such as 'L0' and 'L1' in the names of information related to inter prediction may refer to reference image lists related to the information.
  • Reference image index may be an index indicating one reference image among one or more reference images in the reference image list used for prediction of the target block.
  • a reference block may be a block referenced for encoding/decoding a target block, such as prediction and filtering.
  • a reference block may include a reference sample referenced for deriving a prediction sample, and may mean a block providing information used for decoding a target block.
  • the inter prediction indicator can indicate the direction of inter prediction for the target block.
  • the inter prediction can be one of uni-directional prediction and bi-directional prediction.
  • the inter prediction indicator can indicate the number of reference pictures used when generating a prediction block of the target block.
  • the inter prediction indicator can indicate the number of prediction blocks used for inter prediction for the target block.
  • the reference direction can mean the inter prediction indicator.
  • the inter prediction indicator can indicate one of uni-directional and bi-directional.
  • the inter prediction indicator can have a first value of '0', for an inter mode that uses only reference pictures in the L1 reference picture list, the inter prediction indicator can have a second value of '1', and for an inter mode that uses at least two of the reference pictures in the L0 reference picture list and the reference pictures in the L1 reference picture list, the inter prediction indicator can have a third value of '2'.
  • the prediction list utilization flag for a specific reference image list can indicate whether at least one reference image in the specific reference image list is used to generate a prediction block of the target block. For example, a value of the prediction list utilization flag for a specific reference image list of '0' can indicate that a prediction block is not generated using a reference image in the specific reference image list. A value of the prediction list utilization flag for a specific reference image list of '1' can indicate that a prediction block is generated using a reference image in the specific reference image list.
  • An inter prediction indicator can be derived using a prediction list utilization flag.
  • a prediction list utilization flag can be derived using an inter prediction indicator.
  • an inter prediction indicator can be derived using prediction list utilization flags for a plurality of reference image lists. If an inter prediction indicator indicates that specific reference lists among a plurality of reference image lists are used, prediction list utilization flags of specific reference lists indicated by the inter prediction indicator among prediction list utilization flags of the plurality of reference image lists can be set to '1', and prediction list utilization flags of the remaining reference image lists not indicated by the inter prediction indicator can be set to '0'.
  • Reference direction can point to a reference image list used for prediction of the target block.
  • the reference direction can point to one or more of the reference image list L0 and the reference image list L1.
  • a unidirectional reference direction may mean that one reference image list is used.
  • a bidirectional reference direction may mean that two reference image lists are used.
  • the reference direction may indicate one of the following: only reference image list L0 is used, only reference image list L1 is used, and two reference image lists are used. Additionally, the reference direction may be indicated by an inter prediction indicator.
  • POC Picture Order Count
  • Motion information may be information used to specify a reference block.
  • the motion information may include information used for inter prediction, such as a motion vector (MV), a reference picture index, a reference picture, an inter prediction indicator, a prediction list utilization flag, and the like.
  • the motion information may include information used in a specific inter prediction mode, such as an MV candidate, an MV candidate index, a merge candidate, and a merge index.
  • multiple motion information for multiple reference image lists can be used respectively.
  • Motion information for a specific reference image list can be used for prediction using a specific reference image list.
  • Multiple (intermediate) prediction blocks can be derived respectively by the multiple motion information.
  • a (final) prediction block for the target block can be generated using statistical values for the multiple (intermediate) prediction blocks.
  • MV can be a two-dimensional vector used in inter prediction. MV can mean the offset between the target block and the reference block. Or, MV can represent the difference between the location of the target block and the location of the reference block.
  • MV can be expressed in the form of (mv x , mv y ).
  • mv x can represent the horizontal component
  • mv y can represent the vertical component.
  • the zero vector can be (0, 0) MV.
  • Block Vector can be a two-dimensional vector used in intra block copy prediction.
  • BV can mean an offset between a target block in a target image and a reference block in the target image.
  • BV can represent displacement between a target block and a reference block in the target image.
  • BV can be expressed in a form similar to MV, such as (bv x , bv y ).
  • bv x can represent the horizontal component
  • bv y can represent the vertical component.
  • the zero vector can be (0, 0) BV.
  • Motion information candidate In a specific prediction, the motion information of the target block can be selected from motion information candidates determined by a specific method.
  • the motion information candidate may mean the motion information of the reference block, or may mean the reference block itself having the motion information.
  • the reference block may be a block determined by a specific method for selecting the motion information candidate.
  • the candidate list may be a list including one or more candidates.
  • the candidate list may include a motion information candidate list, a merge candidate list, an MV candidate list, an MPM list, etc.
  • the candidate list may be generated in the same manner by the encoding device and the decoding device.
  • the candidate list used by the encoding device and the candidate list used by the decoding device may be the same, and the same candidate list may be shared by the encoding device and the decoding device.
  • the encoding device may select a candidate used for processing the target block among the candidates in the candidate list.
  • An indicator indicating the selected candidate may be signaled from the encoding device to the decoding device.
  • the decoding device may specify a candidate used for processing the target block among the candidates in the candidate list by using the indicator.
  • the encoding device and the decoding device may specify a candidate used for processing the target block among the candidates in the candidate list by the same rule.
  • Motion information candidate list may mean a list constructed using one or more motion information candidates.
  • Motion information candidate index may be an identifier or indicator that points to a motion information candidate used for prediction of a target block among the motion information candidates in the motion information candidate list.
  • motion information of other reconstructed blocks may be used to derive motion information of the target block.
  • the other blocks may include neighboring blocks.
  • the motion information for the target block itself is not individually signaled, but other information used to derive motion information of the target block based on the motion information of other reconstructed blocks may be signaled.
  • the other information may include information indicating which block among the other reconstructed blocks' motion information is used to derive motion information of the target block, such as a motion information candidate index.
  • these inter prediction modes may include AMVP mode, merge mode, and skip mode.
  • the motion information candidate index may be a merge index or an MV candidate index.
  • MV may be a part of motion information.
  • information about motion information such as motion information candidate, motion information candidate list and motion information candidate index may be replaced with information about MV such as MV candidate, MV candidate list and MV candidate index, and the description about motion information may also be applied to MV.
  • Merge can mean merging motion information for multiple blocks, and can mean applying motion information of another block to the target block as well.
  • merge mode can mean a mode in which motion information of the target block is derived from motion information of neighboring blocks.
  • a merge candidate may mean a specific (restored) block used for merging the target block, or may mean motion information of a specific block.
  • a merge candidate may include motion information of a specific block.
  • - Merge candidates for the target block may include a spatial merge candidate, a temporal merge candidate, a history-based candidate, an average candidate based on the average of two merge candidates, and a zero merge candidate.
  • a merge candidate list can be a list constructed using one or more merge candidates.
  • the merge index may be an indicator that indicates a merge candidate used for prediction of a target block among the merge candidates in the merge candidate list.
  • the motion information of the merge candidate indicated by the merge index among the merge candidates in the merge candidate list may be used as the motion information of the target block.
  • a neighbor block may refer to a block adjacent to a target block.
  • a neighbor block may include a spatial neighbor block and a temporal neighbor block.
  • a neighbor block may also refer to a reconstructed neighbor block within a reference image.
  • Spatial neighboring blocks can be blocks that are spatially adjacent to the target block.
  • the target block and spatial neighboring blocks can be included in the target image.
  • a spatial neighboring block may include a block whose boundary is at least partially adjacent to a boundary of the target block.
  • a spatial neighboring block may include a block whose distance from the target block is less than a specific value.
  • a spatial neighboring block may include a block diagonally adjacent to a vertex of the target block.
  • Temporal neighboring blocks can be blocks that are temporally adjacent to the target block.
  • a temporal neighboring block may include a collocated block (COL block).
  • a collocated block may be a block in a reconstructed picture within a reference picture buffer.
  • a collocated picture (col picture) may refer to a picture that includes a collocated block.
  • a collocated picture may be a picture included in a reference picture list.
  • a call block can be determined based on the location of the target block within the target image. Two blocks being 'temporally adjacent' can mean that the locations of the two blocks satisfy certain conditions.
  • the position of the call block in the call image may be the same as the position of the target block in the target image.
  • the position of the call block in the call image may correspond to the position of the target block in the target image.
  • the correspondence of the positions of the blocks may mean that the areas of the blocks are the same, may mean that the area of one block is included in the area of another block, or may mean that one block occupies a specific position of another block.
  • a temporal neighboring block may be a block that is temporally adjacent to a spatial neighboring block of the target block.
  • Quantized level can be an integer quantity that is used as input to dequantization.
  • QP Quantization Parameter
  • QP may refer to an argument used when generating quantized levels for transform coefficients in quantization.
  • QP may refer to an argument used when generating (restored) transform coefficients for quantized levels in dequantization.
  • QP may be a value mapped to a quantization step size.
  • Quantization matrix coefficients can be each element within a quantization matrix.
  • Non-zero transform coefficient can mean a transform coefficient with a non-zero value or a quantized level with a non-zero value.
  • a bitstream may mean a sequence of bits containing encoded information generated by encoding an image.
  • a bitstream may contain information according to a specific syntax element.
  • information may contain a syntax element.
  • An encoding device may generate a bitstream containing information according to a specific syntax element.
  • a decoding device may obtain information from a bitstream according to a specific syntax element.
  • Signaling of information may indicate that information is transmitted from an encoding device to a decoding device via a bitstream.
  • the information may include a syntax element.
  • signaling may mean that the encoding device includes the information in the bitstream.
  • the information signaled by the encoding device may be used by the decoding device.
  • the bitstream may be transmitted via a network and may be included in a recording medium.
  • the description that information is signaled may include: 1) for signaling of information, the encoding device determines and generates information, 2) the encoding device performs encoding on the information to generate encoded information, 3) (encoded) information is transmitted from the encoding device to the decoding device via a bitstream, 4) the decoding device performs decoding on the encoded information to obtain the information, and 5) through signaling of the information, the decoding device determines and generates the information via signaling.
  • An encoding device can perform encoding on information to generate encoded information.
  • the encoded information can be signaled via a bitstream.
  • a decoding device can perform decoding on the encoded information to obtain information.
  • That information is signaled for a specific target may mean that the information is used for each specific target, and that the processing indicated by the information is applied to each specific target.
  • that information is signaled at a specific unit level may mean that the information is used/processed for each specific unit.
  • the information being signaled may include one or more sub-information. That specific information is signaled may mean that each piece of information of one or more sub-information included in the specific information is signaled.
  • Selective signaling Signaling of information may be performed selectively. Selective signaling of information may mean that an encoding device selectively includes information in the bitstream (under certain conditions). Selective signaling of information may mean that a decoding device selectively obtains information from the bitstream (under certain conditions).
  • Omission of signaling Signaling for information may be omitted. Omission of signaling for information may mean that an encoding device (under certain conditions) does not include the information in the bitstream. Omission of signaling for information may mean that a decoding device (under certain conditions) does not obtain the information from the bitstream. The decoding device may derive the information for which signaling is omitted using other information of the embodiments.
  • Symbol It can mean at least one piece of information of the target unit, such as a syntactic element of the target unit or target block, a coding parameter, a quantized level, and a transform coefficient.
  • the symbol can mean a target of entropy encoding or a result of entropy decoding.
  • Entropy encoding can assign a small number of bits to symbols with a high occurrence probability, and a large number of bits to symbols with a low occurrence probability. As symbols are expressed through this assignment, the size of the bitstream representing the symbols can be reduced.
  • VLC Variable Length Coding
  • CABAC Context-Adaptive Binary Arithmetic Coding
  • entropy coding can be performed using a variable length table.
  • CABAC Context-Adaptive Binary Arithmetic Coding
  • a binarization method for symbols and a probability model of symbols/bins can be derived for entropy coding, and arithmetic coding using context can be performed.
  • Entropy decoding In entropy decoding, the processes performed in entropy encoding can be performed in reverse. Symbols can be generated by entropy decoding the bitstream.
  • the values of information related to the specific entity(ies) described in the embodiments can be used as inputs of a specific operation.
  • the statistical value can be a value derived by a specific operation on the values related to the specific entity(ies).
  • the statistical value for the specific information can be one or more of an average value, a weighted average value, a weighted sum, a minimum value, a maximum value, a mode, a median value, an interpolated value, a sum of products, and a product of sums of values of the specific information.
  • information of the embodiment having a specific value determined by the operation such as constants, variables, and coding parameters, can have a specific statistical value according to the embodiment.
  • the coding parameters may be information required for coding.
  • the coding parameters may include information signaled from an encoding device to a decoding device, may include information calculated/derived during the coding process described in the embodiments, and may include information used for the coding process described in the embodiments.
  • the coding parameters include a size of a CTU, a size of a unit, a form of a unit, a shape of a unit, a depth of a unit, a minimum unit size, a maximum unit size, a maximum unit depth, a minimum unit depth, partition information of a unit, QT partition information, BT partition information, a partition direction of a BT partition, a partition shape of a BT partition, TT partition information, a partition direction of a TT partition, a partition shape of a TT partition, MTT partition information, a combination of MTT partitions, a partition direction of an MTT partition, a partition shape of an MTT partition, a prediction mode, an intra prediction mode, a luma intra prediction mode, a chroma intra prediction mode, an intra partition information, an inter partition information, coding block partition information, prediction block partition information, transform block partition information, a reference sample line index, a reference sample filtering method, a reference sample filter tap, a reference
  • MVD Motion Vector
  • BVD Block Vector
  • BVD Block Vector
  • BVD resolution BV size, BV representation accuracy, BV candidate, BV candidate index, BV candidate list, filter tap of interpolation filter, filter coefficient of interpolation filter, transformation type, transformation size, transformation selection information, primary transformation usage information, secondary transformation usage information, primary transformation selection information, secondary transformation selection information, residual block existence information, coded block pattern, coded block flag flag), QP, delta QP, quantization matrix, deblocking filter usage information, coefficients of the deblocking filter, filter taps of the deblocking filter, strength of the deblocking filter, shape/shape of
  • the coding parameter may further include 1) a value of information that may be included in the coding parameter, 2) a combination of multiple pieces of information that may be included in the coding parameter, 3) a statistical value for the information that may be included in the coding parameter, 4) information related to the coding parameter, 5) information used to calculate/derive the coding parameter, and 6) information calculated/derived using the coding parameter.
  • the "X usage information" may be "information indicating whether X is used/applied/performed.”
  • the "X usage information” may be "information indicating whether X is available.”
  • the "specific mode usage information” may be information indicating whether a specific mode is used.
  • the mode information may indicate a mode used for a target block among the modes described in the embodiments.
  • the specific mode usage information may be replaced with the mode information, and the description of the specific mode usage information may also be applied to the mode information.
  • the "X usage information” and the "X indicator” may be used interchangeably.
  • coding parameters and syntax elements may correspond to each other.
  • syntax elements of an embodiment may be used as coding parameters, and coding parameters may be signaled as syntax elements.
  • X presence information may be considered as “information indicating whether X exists” or “information indicating whether information indicating X exists in the bitstream.”
  • the “X selection information” may be information indicating one of the candidates or methods for X.
  • the “X selection information” may be considered an “X index.”
  • the splitting form of a particular tree may represent one of symmetric splitting and asymmetric splitting, and may represent one of QT, BT, TT, and non-split.
  • the splitting direction of a particular tree may represent one of horizontal and vertical directions.
  • coding parameter when a coding parameter has one of multiple values, "coding parameter" may be replaced with "whether the coding parameter has a specific value among the multiple values available to the coding parameter.”
  • coding parameter when a coding parameter points to one of a plurality of objects, "coding parameter" may be replaced with "whether the coding parameter points to a specific object among the plurality of objects.”
  • Each of the encoding device (110) and the decoding device (150) may be a computer or an electronic apparatus.
  • the encoding device (110) may include a processor (120), storage (140), and a communicator (149).
  • the processor (120), storage (140), and communication device (149) can be connected through a bus.
  • the processor (120) can perform generation and processing of information input to the encoding device (110), output from the encoding device (110), or used within the encoding device (110) in the embodiments, and can perform comparisons and judgments related to such information.
  • the processor (120) may include a plurality of components.
  • the plurality of components may include a partitioner (122), a subtractor (124), a transformer (125), a quantizer (126), an inverse quantizer (127), an inverse transformer (128), an adder (129), a filter (130), and an entropy encoder (139).
  • the program modules may be included in the encoding device (110) in the form of an operating system, an application, and other program modules.
  • the program modules may be instructions or computer executable codes stored in the storage (140) and executed by the processor (120).
  • the communication device (149) can perform a function related to communication of information in the encoding device (110). For example, the communication device (149) can transmit a bitstream to the decoding device (150).
  • the storage (140) may also be named a storage unit.
  • the encoding device (110) can sequentially encode one or more images of a video.
  • the storage (140) can store the original image.
  • the original image can be used as a target image in the encoding device (110).
  • the processor (120) can generate a bitstream including encoded information by performing encoding on the target image, and can store the generated bitstream in the storage (140).
  • the generated bitstream can be stored in a computer-readable recording medium, and can be transmitted to the communication device (189) of the decoding device (150) via a wired and/or wireless transmission medium by the communication device (149).
  • the segmenter (122) can determine a target block by performing segmentation on the target image.
  • the predictor (123) can determine the prediction mode of the target block.
  • the predictor (123) can generate a prediction block of the target block by performing prediction according to the prediction mode.
  • the prediction mode of the target block can be one of the available prediction modes.
  • the available prediction modes can include intra prediction, inter prediction, and IBC prediction.
  • the predictor (123) can perform IBC prediction on the target block to generate a prediction block of the target block.
  • the subtractor (124) can generate a residual block of the target block.
  • the residual block can be the difference between the original block and the predicted block.
  • the original block can be an area pointed to by the target block in the original image.
  • the residual block can mean a block generated by applying one or more of transformation and quantization to the difference between the original block and the predicted block.
  • the transformer (125) can perform a transformation on the residual block to generate transformation coefficients.
  • the converter (125) can perform the conversion using one of a plurality of conversion methods.
  • the multiple transform methods may include a Discrete Cosine Transform (DCT), a Discrete Sine Transform (DST), a Karhunen-Loeve Transform (KLT), and transforms based on each transform.
  • DCT Discrete Cosine Transform
  • DST Discrete Sine Transform
  • KLT Karhunen-Loeve Transform
  • Information for decoding an image may include quantized levels and syntax elements produced by a quantizer (126).
  • the entropy encoder (139) can change the quantized levels in the form of two-dimensional blocks into the form of one-dimensional vectors by using scanning to perform encoding on the quantized levels.
  • scanning which scan among the upper right diagonal scan, the vertical scan, and the horizontal scan will be used can be determined based on coding parameters such as the size of the block and the intra prediction mode of the block.
  • the predictor (123) uses a reference image/block for prediction.
  • the encoded target image/block can be used as a reference image/block for other images/blocks to be processed later.
  • the processor (120) can perform restoration on the encoded target block, and store a restored image including the restored target block generated by the restoration as a reference image in the reference picture buffer (141). Inverse quantization and inverse transformation can be performed on the encoded target block for restoration.
  • the inverse quantizer (127) can generate inverse quantized transform coefficients by performing inverse quantization on the quantized level.
  • the inverse transformer (128) can generate inverse quantized and inversely transformed coefficients by performing inverse transformation on the inverse quantized transform coefficients.
  • the inverse quantized and/or inversely transformed coefficients can mean coefficients to which at least one of inverse quantization and inverse transformation has been applied.
  • the inverse quantized and inversely transformed coefficients can be a reconstructed residual block.
  • An adder (129) can generate a restored block by combining a predicted block and a restored residual block.
  • the restoration block may pass through a filter (130).
  • the filter (130) may apply one or more of a plurality of filters to the target.
  • Each filter of the plurality of filters may be an in-loop filter.
  • the target may be a restoration sample, a restoration block, or a restoration image.
  • the reference picture buffer (141) can store a restoration block/image provided from the filter (130).
  • the restoration image can be an image including a restoration block.
  • the restoration image can be an image composed of restoration blocks.
  • the reference picture buffer (141) can provide a stored restored image as a reference image to the predictor (123).
  • the reference picture buffer (141) may also be referred to as a decoded picture buffer (DPB).
  • the decryption device (150) may include a processor (160), a storage (180), and a communication device (189).
  • the description of the processor (120), storage (140) and communication device (149) related to the encoding device (110) can also be applied to the processor (160), storage (180) and communication device (189) related to the decoding device (150). Duplicate descriptions are omitted.
  • the processor (160) may include a plurality of components.
  • the plurality of components may include an entropy decoder (161), a divider (162), a predictor (163), an inverse quantizer (167), an inverse transformer (168), an adder (169), and a filter (170).
  • the storage (180) may include a reference picture buffer (181).
  • the communication device (149) of the encoding device (110) can transmit the bitstream generated by the encoding device (100) to the decoding device (150).
  • a computer-readable recording medium storing the bitstream can transmit the bitstream generated by the encoding device (100) to the decoding device (150).
  • the communication device (189) can receive a bitstream from the encoding device (110) via a wired and/or wireless transmission medium.
  • the received bitstream can be stored in the storage (180).
  • a bitstream may contain encoded information.
  • An entropy decoder (161) can generate information for decoding an image by performing entropy decoding based on a probability distribution on encoded information of a bitstream.
  • the entropy decoder (161) can change the quantized levels in the form of a one-dimensional vector into the form of a two-dimensional block by using scanning to perform decoding on the quantized levels.
  • scanning which scan among the upper right diagonal scan, the vertical scan, and the horizontal scan will be used can be determined based on coding parameters such as the size of the block and the intra prediction mode of the block.
  • the descriptions of the divider (122), the predictor (123), the inverse quantizer (127), the inverse transformer (128), the adder (129), the filter (130), and the reference picture buffer (141) of the encoding device (110) disclosed in the embodiments may also be applied to the divider (162), the predictor (163), the inverse quantizer (167), the inverse transformer (168), the adder (169), the filter (170), and the reference picture buffer (181) of the decoding device (150). Duplicate descriptions are omitted.
  • each of the divider (122), the predictor (123), the inverse quantizer (127), the inverse transformer (128), the adder (129), and the filter (130) of the encoding device (110) can generate syntactic element information specifying processing for the target.
  • Each of the divider (162), the predictor (163), the inverse quantizer (167), the inverse transformer (168), the adder (169), and the filter (170) of the decoding device (150) can perform processing for the target (same as that performed in the encoding device (110)) using the syntactic element information.
  • the processor may represent the processor (120) of the encoding device (110) and/or the processor (160) of the decoding device (150).
  • the processor may represent a predictor (123), a subtractor (124) and an adder (129), and may represent a predictor (163) and an adder (169).
  • the processing unit may represent a transformer (125) and an inverse transformer (128), and may represent an inverse transformer (168).
  • the processor may represent a quantizer (126) and an inverse quantizer (127), and may represent an inverse quantizer (167).
  • Figure 2 can schematically represent an example in which one unit is divided into multiple sub-units.
  • the attributes of a block may include the size of the block. Specific processing described in the embodiments may be applied/performed when specific conditions regarding the size of the block are met.
  • Each picture composing the video can be classified into an I picture (i.e., an intra picture), a P picture (i.e., a uni-prediction picture), and a B picture (i.e., a bi-prediction picture) according to the coding type. Coding can be performed for each picture according to the coding type of each picture.
  • coding for the target picture can be performed using information within the target picture without inter prediction referring to other images.
  • coding for the I picture can be performed using intra prediction and/or IBC prediction.
  • Coding for P pictures and B pictures can be performed by at least one of intra prediction, IBC prediction, and inter prediction using a reference picture.
  • coding for the target picture can be performed using unidirectional inter prediction using one reference picture list.
  • the target block can be a prediction block or a split prediction block.
  • Inter prediction can be performed using reference images and motion information.
  • a reference image can be selected using a reference image index, and a reference block corresponding to a target block within the reference image can be determined using motion information.
  • a prediction block for the target block can be generated using the determined reference block.
  • Motion information can be derived using coding parameters, etc.
  • motion information can be derived using motion information of a reconstructed neighboring block, motion information of a called block, and/or motion information of a block adjacent to a called block.
  • a candidate list may be used for inter prediction.
  • the candidate list may include a plurality of candidates.
  • An index indicating a candidate used for inter prediction for a target block among the candidates in the candidate list may be signaled.
  • the candidate list may be derived in the same manner based on the same information in the encoding device (110) and the decoding device (150).
  • the same information may include a restored image and a restored block.
  • the order of the candidates in the candidate list may have to be constant.
  • prediction for a target block can be performed by using motion information of a spatial candidate or a temporal candidate as motion information of the target block.
  • Motion information of a spatial candidate can be referred to as spatial motion information.
  • Motion information of a temporal candidate can be referred to as temporal motion information.
  • a spatial candidate may be a restored spatial neighboring block that is spatially adjacent to the target block.
  • a spatial candidate can be a block that 1) exists within the target image, 2) has already been restored through decryption, and 3) is adjacent to the target block.
  • a temporal candidate may be a reconstructed temporal neighboring block corresponding to a target block in a reconstructed COL image.
  • the motion information of the spatial candidate may be motion information of a block including the spatial candidate.
  • the motion information of the temporal candidate may be motion information of a block including the temporal candidate.
  • a call block may include a first call block and a second call block.
  • the first call block may be a block occupying coordinates (xP + nPSW, yP + nPSH).
  • the second call block may be a block occupying coordinates (xP + (nPSW >> 1), yP + (nPSH >> 1)).
  • the second call block may be optionally used as a call block when the first call block is unavailable.
  • the MV of the target block can be determined based on the MV of the call block. Scaling can be performed on the MV of the call block.
  • the scaled MV of the call block can be used as the MV or the prediction MV of the target block.
  • the MV of the temporal candidate stored in the candidate list related to inter prediction can be a scaled MV.
  • the ratio of the scaled MV and the MV of the call block may be equal to the ratio of the first temporal distance and the second temporal distance.
  • the first temporal distance may be the distance between the reference image and the target image of the target block.
  • the second temporal distance may be the distance between the reference image and the call image of the call block.
  • the method by which motion information is derived can be determined by the inter prediction mode of the target block.
  • the inter prediction mode AMVP mode, merge mode, skip mode, merge mode with MVD, sub-block merge mode, GPM, Combined Inter Intra Prediction (CIIP) mode, and affine inter mode can be used.
  • CIIP Combined Inter Intra Prediction
  • an MV candidate list including one or more MV candidates can be generated using the MV of the spatial candidate, the MV of the temporal candidate, the history-based MV candidate, and the zero vector. At least one of the MV of the spatial candidate, the MV of the temporal candidate, and the zero vector can be determined and used as an MV candidate.
  • the encoding device (110) can determine an MV to be used for encoding the target block within the search range using the MV candidate list.
  • the maximum number of MV candidates in the MV candidate list can be predefined.
  • N can represent a predefined maximum number.
  • the maximum number of the candidates can be signaled from the encoding device to the decoding device, or derived from the decoding device.
  • the encoding device (110) can determine an MV candidate to be used as a prediction MV of the target block among the MV candidates in the MV candidate list.
  • the MV to be used for encoding the target block can be an MV that can be encoded at the minimum cost.
  • the encoding device (110) can determine whether to use the AMVP mode in encoding the target block, and can generate AMVP mode usage information indicating whether the AMVP mode is used.
  • Inter prediction information may include 1) AMVP mode usage information, 2) MV candidate index, 3) MVD, 4) MVD resolution information, 5) reference direction, and 6) reference image index, and may include a residual block.
  • Inter prediction information may be signaled from an encoding device (110) to a decoding device (150) in the form of a bitstream.
  • the MVD can represent the difference between the MV to be actually used for inter prediction of the target block and the predicted MV.
  • the encoding device (110) can derive a predicted MV that is close to the MV to be actually used for inter prediction of the target block in order to use an MVD of a size as small as possible.
  • the decoding device (150) can derive the MV of the target block by combining the MVD and the predicted MV. In other words, the MV of the target block derived by the decoding device (150) can be the sum of the MVD and the predicted MV candidate.
  • the encoding device (110) can generate MVD resolution information.
  • the MVD resolution information can be information used to adjust the resolution of the MVD.
  • the decoding device (150) can adjust the resolution of the MVD using the MVD resolution information.
  • the encoding device (110) can calculate the MVD based on the affine model.
  • the affine control point MV of the target block can be derived based on the sum of the affine control point MV candidates and the MVD.
  • the MV of each subblock within the target block can be derived using the affine control point MV.
  • a history-based merge candidate may be motion information within a list that includes motion information of other blocks that were encoded/decoded prior to encoding/decoding of the target block.
  • An average merge candidate may be a merge candidate generated based on the average of two merge candidates in the merge candidate list.
  • a zero merge candidate may be zero vector motion information.
  • Zero vector motion information may be motion information whose MV is a zero vector.
  • Merge candidates can be added to the merge candidate list according to a predefined manner and a predefined order so that the merge candidate list has a set number of merge candidates.
  • the same merge candidate list can be constructed in the encoding device (110) and the decoding device (150) through the predefined manner and the predefined order.
  • the encoding device (110) can select a merge candidate to be used for encoding a target block from among the merge candidates in the merge candidate list.
  • the encoding device (110) can determine whether to use a merge mode in encoding a target block, and can generate merge mode usage information indicating whether the merge mode is used.
  • Inter prediction information may include 1) merge mode usage information, 2) merge index, and 3) correction information, and may include a residual block.
  • Inter prediction information may be signaled from an encoding device (110) in the form of a bitstream to a decoding device (150) in the form of a bitstream.
  • the encoding device (110) can select an optimal merge candidate from among the merge candidates included in the merge candidate list, and set the value of the merge index to point to the selected merge candidate.
  • the correction information may be information used for correction of the MV.
  • the encoding device (110) may generate the correction information.
  • the decoding device (150) may perform correction on the MV of the merge candidate selected by the merge index based on the correction information, thereby deriving a corrected MV.
  • the corrected MV may be used as the MV of the target block.
  • the correction information may include an MVD.
  • the correction information may include one or more of correction usage information, correction direction information, and correction size information.
  • the correction usage information may indicate whether correction is used for the MV.
  • a merge mode that performs correction for the MV based on the correction information may be referred to as a merge mode having an MVD.
  • prediction for a target block can be performed using a merge candidate pointed to by a merge index among the merge candidates included in the merge candidate list.
  • Motion information of the target block can be derived from 1) MV of the merge candidate pointed to by the merge index, 2) reference image index, and 3) reference direction.
  • the merge candidates of the merge candidate list may be specific modes that derive inter prediction information.
  • the merge candidate may be information indicating a specific mode that derives inter prediction information.
  • Inter prediction information of the target block may be derived according to the specific mode indicated by the merge candidate.
  • the specific mode may be regarded as a specific inter prediction information deriving mode or a specific motion information deriving mode.
  • the specific mode may include a series of processes that derive inter prediction information.
  • Inter prediction information of a target block can be derived according to a specific mode indicated by a merge candidate selected by a merge index among the merge candidates in the merge candidate list.
  • the specific modes can include a sub-block unit motion information derivation mode and an affine motion information derivation mode, and can include other modes for deriving motion information described in the embodiments.
  • Skip mode may be a mode that does not use residual blocks. That is, when skip mode is used, the reconstructed block may be the same as the predicted block.
  • the description of the merge mode of the embodiments may also be applied to the skip mode.
  • the difference between the merge mode and the skip mode may be whether the residual block is signaled and used. That is, the skip mode may be similar to the merge mode except that the residual block is not transmitted/used, and the description of the merge mode may also be applied to the skip mode.
  • a first prediction block and a second prediction block can be generated using two pieces of motion information for a target block.
  • a final prediction sample of a final prediction block can be generated using a weighted sum of a first prediction sample of the first prediction block and a second prediction sample of the second prediction block.
  • the value of the final prediction sample of the final prediction block can be determined using a weighted sum of the first prediction sample of the first prediction block and the second prediction sample of the second prediction block. If the distance between the final prediction sample and the boundary is greater than the reference value, one of the first weight and the second weight can be 1 and the other can be 0.
  • a separable transformation or a 2-dimensional (2D) non-separable transformation can be performed on the residual block.
  • the separable transformation can be a transformation that performs 1-dimensional (1D) transformations on the residual block in each of the horizontal and vertical directions.
  • the secondary transform may be a transform for improving the energy concentration of the transform coefficients generated by the primary transform.
  • the secondary transform may be 1) a separable transform like the primary transform, or 2) a 2D non-separable transform.
  • the 2D non-separable transform may mean a low frequency non-separable transform (LFNST) or a non-separable primary transform (NSPT).
  • NSPT can be applied to specific block sizes such as 4x4, 4x8, 8x4, 4x16, 16x4, 8x8, 8x16, and 16x8 for intra coding.
  • a set of transformations can also be defined in the second-order transformation.
  • the methods for deriving and/or determining the set of transformations of the embodiments can be applied to the second-order transformation as well as the first-order transformation.
  • a primary transformation and/or a secondary transformation may be determined for a specific target.
  • the transformation selection information may include transformation target information.
  • the transformation target information may indicate a target to which the primary transformation and/or the secondary transformation is applied.
  • a first-order transform and/or a second-order transform may be applied to one or more of the signal components, including the luma component and the chroma component.
  • the transform selection information may include primary transform usage information and secondary transform usage information.
  • the primary transform usage information may indicate whether the primary transform is applied to the residual block of the target block.
  • the secondary transform usage information may indicate whether the secondary transform is applied to the residual block of the target block.
  • the description of the transformation described above can also be applied to the inverse transformation.
  • the reverse processing of the processing described for the transformation can be performed in the inverse transformation.
  • the "transformation" in the name related to the transformation can be changed to "inverse transformation.”
  • the input of the transformation can be considered as the output of the inverse transformation.
  • the output of the transformation can be considered as the input of the inverse transformation.
  • the decoding device (150) can obtain information related to the transformation, such as the transformation selection information, and can perform the reverse processing of the processing related to the transformation indicated by the information related to the transformation using the information related to the transformation.
  • the target block may include a plurality of subblocks. Each subblock may be defined according to a minimum block size or a minimum block shape.
  • the target block may be divided into a plurality of subblocks, and each subblock may include coefficients such as 4x4, 2x8, and 8x2.
  • the target block may be a transform block.
  • the transform coefficients or quantized levels may be expressed in the form of a block.
  • the transform coefficients may be quantized transform coefficients.
  • the transform coefficients or quantized levels can be scanned according to at least one of scanning types, such as diagonal scanning, vertical scanning and horizontal scanning.
  • the diagonal scanning can be right-upper diagonal scanning or left-lower diagonal scanning.
  • coefficients can be changed or arranged in the form of a one-dimensional vector by scanning the coefficients of a block using diagonal scanning.
  • Vertical scanning can be scanning coefficients in the form of a two-dimensional block in the column direction.
  • Horizontal scanning can be scanning coefficients in the form of a two-dimensional block in the row direction.
  • Scanning for each scanning type can start at a specific starting point and end at a specific ending point.
  • the scanning order according to the scanning type can be applied first between subblocks.
  • the scanning order according to the scanning type can be applied to the transform coefficients or quantized levels within the subblock.
  • the encoding device (110) can perform entropy encoding on transform coefficients or quantized levels to generate a bitstream including entropy-encoded transform coefficients or entropy-encoded quantized levels.
  • the decoding device (150) can obtain entropy-encoded transform coefficients or entropy-encoded quantized levels from a bitstream and perform entropy decoding to generate transform coefficients or quantized levels.
  • the coefficients can be arranged in the form of a two-dimensional block through inverse scanning.
  • the arrangement of the inverse scanning can be a rearrangement opposite to the arrangement of the scanning.
  • Inverse scanning of coefficients can generate inversely scanned transform coefficients or inversely scanned quantized levels.
  • the inverse scanning types of the inverse scanning can include diagonal scan, vertical scan, and horizontal scan, and the inverse scanning type of the inverse transformation corresponding to the scanning type of the transform can be selected.
  • inverse quantization can be performed on (inversely scanned) coefficients.
  • a second inverse transform can be performed on a result generated by performing inverse quantization.
  • a first inverse transform can be performed on a result generated by performing the second inverse transform.
  • a restored residual block can be generated by selectively performing the second inverse transform and the first inverse transform on the coefficients.
  • the target sample may be one of the samples described in the embodiments.
  • the target sample may be one or more of the samples described in the embodiments, such as a prediction sample, a reference sample, a residual sample, a restored sample, and a restored sample with filtering applied.
  • the target sample can be a sample within one or more of a target picture, a target slice, a target CTB, a target block, a reference sample line, and a template.
  • the target block can be one of the blocks described in the embodiments.
  • the target block can be one or more of the blocks described in the embodiments, such as a transform block, a prediction block, a reference block, a residual block, and a reconstruction block.
  • the filtering processing described as being applied to one object may also be applied to other objects.
  • the filtering processing described in a specific in-loop filtering may also be applied to transform blocks, prediction blocks, reference blocks, and residual blocks.
  • the filter tab can indicate the number of input samples used for the filter.
  • the input samples can include the target sample.
  • the input samples can include a specific value determined for the target sample.
  • the input samples can include one or more reference samples.
  • the one or more reference samples can be determined based on an attribute of the target block described in the embodiments.
  • the attribute can include a coding parameter.
  • an attribute of the target sample can include a position of the target sample.
  • One or more reference samples can be specified based on a relative position with respect to the position of the target sample.
  • the filter shape can represent the shape that the input samples form.
  • a specific value determined for a target sample can be considered as a target sample.
  • the target sample can also be considered as forming a filter shape.
  • the number of samples whose values are determined by filtering may be plural.
  • the filter strength may indicate the range of samples whose values are determined by filtering.
  • the filter strength may be one of a strong filtering strength and a weak filtering strength.
  • the number of samples whose values are determined by the strong filtering strength may be larger than the number of samples whose values are determined by the weak filtering strength.
  • the filter strength may indicate the range of values that are changed by filtering.
  • the range of sample values that are changed by the strong filtering strength may be wider than the range of sample values that are changed by the weak filtering strength.
  • the reference sample may include one or more of the upper left reference sample, the upper reference sample, the upper right reference sample, the left reference sample, and the lower left reference sample. Filtering on the prediction sample may be performed by applying specific weights to the prediction sample, the left reference sample, the upper reference sample, and/or the upper left reference sample, respectively.
  • Luma signal mapping can perform codeword redistribution for the luma signal.
  • Luma signal mapping may include forward mapping and backward mapping.
  • forward mapping an existing dynamic range may be divided into multiple intervals.
  • a mapped dynamic range may be determined by performing codeword redistribution for an input image using a linear model for each interval.
  • backward mapping reverse mapping is performed from the mapped dynamic range to the existing dynamic range.
  • Chroma scaling can correct chroma signals based on the correlation between a luma signal and a corresponding chroma signal.
  • Forward mapping can be performed between inter prediction for a luma signal and restoration for a luma signal, and between inter prediction for a luma signal and chroma scaling.
  • Backward mapping can be performed between restoration for a luma signal and in-loop filtering for a luma signal.
  • Chroma scaling can be performed between inverse transformation and restoration for a chroma signal.
  • inverse quantizations for the luma signal and the chroma signal can be performed within the mapped dynamic range.
  • In-loop filterings for the luma signal and the chroma signal, inter predictions for the luma signal and the chroma signal, intra prediction for the chroma signal and restoration for the chroma signal can be performed within the existing dynamic range.
  • the deblocking filter can remove block distortion occurring at boundaries between blocks in a restored image.
  • the blocks may be transform blocks.
  • the blocks may be subblocks of a specific block described in the embodiments.
  • the boundaries between blocks may mean samples adjacent to the boundaries between blocks.
  • Deblocking filters can be applied to vertical boundaries and horizontal boundaries between blocks. After filtering is performed on the vertical boundaries of blocks, filtering can be performed again on the horizontal boundaries of the filtered blocks.
  • the filter to be applied may be determined based on the strength of the required deblocking filtering. That is, a filter determined based on the strength of the deblocking filtering among a plurality of other filters may be applied to the target block.
  • the plurality of filters may include one of a long-tap filter, a strong filter, a weak filter, and a Gaussian filter.
  • the maximum length of the deblocking filter can be determined based on the properties of the target block, such as the size of the target block, components of the target block, and coding parameters.
  • SAO can compensate for distortion between the original image and the restored image on a sample-by-sample basis. For compensation, SAO can apply an appropriate offset to the sample value of the sample. That is, the offset can be added to the sample value.
  • An offset can be determined for the target block. For example, an offset can be determined for each component of the CTB. The determined offset can be applied to samples within a specific component of the CTB.
  • SAO may include SAO using Edge Offset (EO) and SAO using Band Offset (BO). Depending on the characteristics of samples within a specific block such as a CTU, whether SAO using EO is performed or SAO using BO is performed may be determined respectively.
  • EO Edge Offset
  • BO Band Offset
  • Pattern classes of the EO can include a horizontal pattern, a vertical pattern, a 135 degree diagonal pattern, and a 45 degree diagonal pattern.
  • information indicating a pattern class applied to the target block and a plurality of offsets of the pattern class can be signaled. The number of offsets can be four.
  • adjacent samples of the target sample can be determined according to the direction of the pattern class.
  • An offset to be applied to the target sample can be determined by the pattern of the adjacent samples.
  • correction for distortion of a sample can be performed by classifying the brightness values of samples within a target block into specific bands.
  • the bit depth of an input image can be divided into m sections.
  • m can be 32.
  • the specific bands can be n consecutive sections among the m sections.
  • n can be 4.
  • N offsets for the n sections can be signaled.
  • information indicating a first section selected as n sections among the m sections can be signaled.
  • An offset of a section to which a target sample corresponds can be added to a sample value of a target sample of a target unit.
  • the context modeler can perform context updates to apply the current probability information to the entropy encoding of the bins of the syntactic elements of the target block.
  • the updated context can be stored in the context memory.
  • the updated context corresponding to the syntactic elements of the target block (or the bins in the syntactic elements of the target block) can be derived by the context modeler.
  • the updated context can be used for entropy encoding of syntactic elements of the target block.
  • the updated context can be used for entropy decoding of syntactic elements of the target block.
  • the entropy decoding unit can generate bins for the delimiting elements of the target block by performing entropy decoding on the encoded information of the bitstream based on the updated context.
  • the entropy decoding unit can use at least one of an arithmetic decoding method and a bypass decoding method.
  • Information about syntactic elements and bins can be provided from the de-binarization unit to the context selection unit.
  • a syntax element may be one of the coding parameters described in the embodiments.
  • one or more of the binarization methods, inverse binarization methods, entropy encoding methods and entropy decoding methods listed below may be used to perform signaling for specific information.
  • FIG. 8 is a flowchart of a method for predicting a target block and a method for generating a bitstream according to one embodiment.
  • the prediction method and bitstream generation method of the target block of the embodiment can be performed by the encoding device (110).
  • the embodiment can be a part of the encoding method of the target block or the video encoding method.
  • the processor (120) can determine prediction information to be applied to encoding the target block.
  • the prediction signal may be an IBC prediction signal.
  • the filter may be an IBC filter.
  • prediction signal filtering information can be obtained.
  • the prediction signal filtering information can be IBC prediction signal filtering information.
  • a bitstream may be transmitted from an encoding device (110) to a decoding device (150).
  • the bitstream may include prediction signal filtering information.
  • the bitstream may include encoded prediction signal filtering information.
  • Prediction signal filtering information can be obtained from a bitstream.
  • prediction signal filtering information can be derived by performing decoding on encoded prediction signal filtering information of a bitstream.
  • the prediction signal filtering information may include at least one of: information indicating whether filtering is performed; information indicating the type of the filter; information indicating the type of the filter template; the number of filter coefficient sets; a filter model; and a filter mode.
  • the prediction signal filtering information may include at least one of: information indicating whether filtering is performed; information indicating the type of the filter; information indicating the type of the filter template; the number of filter coefficient sets; a filter model; and a filter mode.
  • the prediction signal filtering information may include at least one of: information indicating whether filtering is performed; information indicating a type of a filter; information indicating a type of a filter template; the number of filter coefficient sets; a filter model; and a filter mode.
  • the prediction signal filtering information may include at least one of: information indicating whether IBC filtering is performed; information indicating a type of an IBC filter; information indicating a type of an IBC filter template; the number of IBC filter coefficient sets; an IBC filter model; and an IBC filter mode.
  • the prediction signal filtering information may be determined based on at least one of: a template T t of a target block; a filtering flag; a filter type; a filter template type; the number of filter sets; a prediction mode flag; a general merge flag; a merge index; a DBV mode; motion information; a second filter template type; an average of prediction signals; an average of filter templates of the target block; a histogram distribution of the prediction signal; a histogram distribution of the filter templates of the target block; a coding parameter of the target block and/or a coding parameter of a unit including the target block; and a coding parameter of a reference block and/or a coding parameter of a unit including the reference block.
  • different pieces of information described in the embodiments may be used to derive different pieces of information included in the prediction signal filtering information.
  • the coding parameters may include information described in the embodiments, information signaled via a bitstream described in the embodiments, and syntax elements described in the embodiments.
  • the prediction signal filtering information may be determined based on at least one of: a template T t of a target block; ibc_filtering_flag; ibc_filter_shape; ibc_filter_tpl_shape; ibc_filter_set_num; pred_mode_ibc_flag; general_merge_flag; merge_idx, DBV mode; block vector BV (BV x , BV y ); ibc_filter_tpl_shape 2 ; an average of IBC prediction signals; an average of filter templates of the target block; a histogram distribution of IBC prediction signals; a histogram distribution of filter templates of the target block; a coding parameter of the target block and/or a coding parameter of a unit including the target block; and a coding parameter of a reference block and/or a coding parameter of a unit including the reference block.
  • different pieces of information described in the embodiments may be
  • a prediction signal may be generated.
  • the prediction signal may be an IBC prediction signal.
  • filtering may be performed on the prediction signal based on the prediction signal filtering information.
  • a filtered prediction signal may be generated by filtering the prediction signal.
  • the prediction signal and the prediction block in the embodiments may mean a filtered prediction signal.
  • filtering may be performed on the prediction signal based on a filtering coefficient.
  • the prediction signal may be an IBC prediction signal.
  • a prediction signal on which filtering is performed based on at least one of a filter coefficient set (FilterSet); a number of filter sets; a prediction signal; a filter model; a coding parameter of a target block and/or a coding parameter of a unit including the target block; and a coding parameter of a reference block and/or a coding parameter of a unit including the reference block may be obtained.
  • an IBC prediction signal P' t on which filtering is performed based on at least one of a filter coefficient set (FilterSet), ibc_filter_set_num, an IBC prediction signal P t ; a filter model, coding parameters of a target block and/or coding parameters of a unit including the target block; and coding parameters of a reference block and/or coding parameters of a unit including the reference block; can be obtained.
  • Merge mode can be a prediction mode that uses block vectors of surrounding blocks and transmits residual signals.
  • the prediction block vector can be a motion vector of a temporal neighboring block or a spatial neighboring block of the target block.
  • Target block vector The target block vector can be a motion vector pointing to a predicted block of the target block.
  • Residual block vector can be the difference vector between the predicted block vector and the target block vector.
  • the target block vector can be the sum of the predicted block vector and the residual block vector.
  • Block vector information may be information consisting of a predicted block vector and a residual block vector.
  • the reference image index can indicate an image referenced by the target block.
  • the reference image index can also indicate a slice or tile referenced by the target block. If the reference image index indicates an image including the target block, it can indicate that the encoding/decoding mode of the target block is a block vector prediction mode. If the reference image index indicates an image not including the target block, it can indicate that the encoding/decoding mode of the target block is an inter prediction mode.
  • Neighbor pixels A neighbor pixel of an object can be one of the pixels within certain relative positions to the object's position.
  • the surrounding pixels of a target pixel may be pixels adjacent to the target pixel.
  • the surrounding pixels of the target may be pixels whose distance from the target is 1.
  • the distance may be a horizontal distance; a vertical distance; or a maximum of the horizontal distance and the vertical distance.
  • the surrounding pixels of the target may be adjacent pixels of the target.
  • the surrounding pixels may be limited to available pixels (e.g., pixels that have already been restored).
  • the surrounding pixels may be one of: a pixel diagonally adjacent to the top-leftmost corner of the target; a pixel vertically adjacent to the top of the target; and a pixel horizontally adjacent to the left of the target.
  • the surrounding pixels of the target may be pixels whose distance from the target is less than or equal to N.
  • the distance may be a horizontal distance; a vertical distance; or a maximum of the horizontal distance and the vertical distance; N may be one of integers greater than or equal to 1.
  • the surrounding pixels may be limited to available pixels (e.g., pixels that have already been reconstructed).
  • the surrounding pixels may be pixels within a specific region that includes an area of the target block.
  • the area of the target block may be rectangular.
  • the specific region may be rectangular.
  • the position of the bottom-most rightmost pixel of the target block and the position of the bottom-most rightmost pixel of the specific region may be the same.
  • a first difference between the width of the specific region and the width of the region of the target block and a second difference between the height of the specific region and the height of the region of the target block may be the same.
  • the surrounding pixels of the target block may be pixels whose minimum value among the horizontal distance from the target block; and the vertical distance from the target block; is less than or equal to size t .
  • the surrounding pixels may be pixels whose 1) x-coordinate is greater than or equal to x - size t and less than or equal to x + h - 1; 2) y-coordinate is greater than or equal to y - size t and less than or equal to y + h - 1, and 3) is not included in the target block.
  • the size t may be an integer greater than or equal to 1.
  • the surrounding pixels of the target block can be pixels whose vertical distance from the target block is less than or equal to size t .
  • the surrounding pixels can be pixels that 1) have an x-coordinate greater than or equal to x and less than or equal to x + h - 1; 2) have a y-coordinate greater than or equal to y - size t and less than or equal to y - 1, and 3) are not included in the target block.
  • the size t can be an integer greater than or equal to 1.
  • the surrounding pixels of the target block can be pixels whose horizontal distance from the target block is less than or equal to size t .
  • the surrounding pixels can be pixels that 1) have an x-coordinate greater than or equal to x - size t and less than or equal to x - 1; 2) have a y-coordinate greater than or equal to y and less than or equal to y + h - 1, and 3) are not included in the target block.
  • the size t can be an integer greater than or equal to 1.
  • FIGS 11, 12 and 13 illustrate forms of filter templates according to examples.
  • Figure 11 illustrates an L-shaped template according to an example.
  • Figure 12 illustrates an upper template according to an example.
  • Figure 13 illustrates a left template according to an example.
  • IBC-Local Illumination Compensation (IBC-LIC)
  • IBC-LIC may be a technique for applying linear filtering to an IBC prediction signal (or an IBC prediction block) generated by prediction using IBC.
  • P(x, y) can represent an IBC prediction signal (or IBC prediction value) before filtering is applied to the pixel whose coordinates are (x, y).
  • P'(x, y) can represent an IBC prediction signal (or IBC prediction value) to which filtering is applied to the pixel whose coordinates are (x, y).
  • a and b can be filtering coefficients.
  • the left 1/2w ⁇ h portion (horizontal mode) or the upper w ⁇ 1/2h portion (vertical mode) of the final prediction block can be set as the intra-prediction signal.
  • a higher level flag such as SPS level
  • the flag at SPS level may indicate whether the blending method of the embodiments is applied for IBC merge mode at SPS level.
  • the intra prediction mode candidate list can be configured to include inter prediction and intra prediction for intra prediction.
  • the size of the IPM candidate list can be predefined as 3.
  • the 48 geometric partition modes can be divided into two sets of geometric partition modes.
  • IBC-GPM When IBC-GPM is used, a flag indicating which of the two sets is selected and an index for the selected set can be signaled. Additionally, a flag indicating whether intra prediction mode is used for the first subpartition can be signaled. If intra prediction is used for the subpartition, an intra prediction mode index can be signaled. If IBC is used for the subpartition, a merge index can be signaled.
  • IBC-LIC may be a coding tool to compensate for local illumination variation within a picture between a current block coded using IBC and a prediction block having a linear model.
  • the parameters of the linear model can be derived in the same way as LIC for inter prediction, except that the reference templates are generated using the block vectors of IBC-LIC.
  • IBC-LIC can be applied to IBC AMVP mode and IBC merge mode.
  • the IBC-LIC flag may be signaled to indicate the use of IBC-LIC.
  • the first two modes may be related to the choice of template form.
  • top-only templates left-only templates, or L-shaped templates can be used.
  • the signaling of the signal can be as listed below.
  • the value "0" can indicate not IBC-LIC mode.
  • the value "100" can indicate the default IBC-LIC mode.
  • the filter shape can be a square of a given size or a cross of a given size.
  • the filter shape may be a 3x3 square.
  • the filter shape may have other shapes and sizes as described in the embodiments.
  • the filter shape may be a p-gon of size nxm.
  • n, m, and p may be an integer greater than or equal to 1.
  • the template form may mean the form of a filter template.
  • a left template of a block (T Left ); a top template of a block (T Top ); and an L-shaped template (T L-shape ) that uses both the left and top templates of a block may be used. Filtering may be performed using at least one of T Left , T Top and T L-shape .
  • the filter coefficient set may mean a set of filter coefficients used in the filter model. In embodiments, the number of filter coefficient sets may be 1 or more.
  • one or more sets of filter coefficients may be used.
  • the filter model may refer to a formula used to perform IBC prediction signal filtering.
  • the formula can be constructed using at least one of an IBC prediction pixel to which filtering is to be applied; surrounding pixels of the IBC prediction pixel; a bias; and a filter coefficient.
  • the formula of the filter model can be [Formula 1] described above.
  • the formula of the filter model can be [Formula 6] described above.
  • the formula of the filter model can be [Formula 7] below.
  • the formula of the filter model can be another formula deriving P'(x, y) as described in the embodiments.
  • Information can be obtained about whether filtering is performed on the IBC prediction signal.
  • the ibc_filtering_flag syntax element can indicate whether filtering is performed.
  • Filtering may be performed on the IBC prediction block when ibc_filtering_flag is the first value, and filtering may not be performed when ibc_filtering_flag is the second value.
  • filtering can be performed using the pixel value of the p 0 pixel.
  • filtering can be performed using the pixel values of the p 0 pixel, the p 1 pixel, the p 2 pixel, the p 3 pixel, and the p 4 pixel together.
  • the first value could be 1.
  • the second value could be 0.
  • Figure 17 illustrates the use of a 3x3 rectangular filter according to an example.
  • the ibc_filter_shape syntax element can indicate a filter shape.
  • IBC prediction signal filtering can be applied using a 3x3 rectangular filter.
  • IBC prediction signal filtering can be applied using a 3x3 cross filter.
  • filtering can be performed using pixel values of pixel p 0 , pixel p 1 , pixel p 2 , pixel p 3 , pixel p 4 , pixel p 5 , pixel p 6 , pixel p 7 , and pixel p 8 together.
  • filtering can be performed using the pixel values of the p 0 pixel, the p 1 pixel, the p 2 pixel, the p 3 pixel, and the p 4 pixel together.
  • the first value could be 1.
  • the second value could be 0.
  • the ibc_filter_tpl_shape syntax element can specify the shape of a filter template.
  • the template shape can be one of T Left , T Top and T L-shape .
  • an L-shape template T L-shape can be used as the filter template shape.
  • the top template T Top of the block can be used as the filter template shape.
  • the left template T Left of the block can be used as the filter template shape.
  • the first value could be “0”.
  • the second value could be “1”.
  • the third value could be “2”.
  • a pre-defined template shape can be defined.
  • T L-shape can be used as a pre-defined template shape.
  • T Top can be used as a pre-defined template form.
  • T Left can be used as a pre-defined template form.
  • a filter coefficient set can be selected based on two threshold values Thres 0 and Thres 1 , and the selected filter coefficient set can be used.
  • P t '(x, y) F(P t (x, y), FilterSet 1 ) if, P t (x, y) ⁇ Thres 0
  • P t '(x, y) F(P t (x, y), FilterSet 2 ) if, Thres 0 ⁇ P t (x, y) ⁇ Thres 1
  • ibc_filter_set_num when ibc_filter_set_num is the first value, it can be indicated that one filter coefficient set is used. When ibc_filter_set_num is the second value, it can be indicated that two filter coefficient sets are used. When ibc_filter_set_num is the third value, it can be indicated that three filter coefficient sets are used.
  • P t (x, y) can be filtered using one set of filter coefficients.
  • the certain criteria can include a combination of one or more coding parameters of the embodiments.
  • a filter coefficient set can be selected based on Thres, and the selected filter coefficient set can be used.
  • Thres can be an average value of pixels included in the filter template.
  • Thres can be another statistical value of pixels.
  • the first filter coefficient set FilterSet 1 can be used, and if the value of P t (x, y) is greater than Thres, the second filter coefficient set FilterSet 2 can be used.
  • P t '(x, y) F(P t (x, y), FilterSet 1 ) if, P t (x, y) ⁇ Thres
  • the certain criteria can include a combination of one or more coding parameters of the embodiments.
  • the IBC filter mode can be determined by the coding parameters of the embodiments.
  • the IBC filter mode can be defined by at least one of the signaled syntax elements ibc_filter_flag, ibc_filter_shape, ibc_filter_tpl_shape and ibc_filter_set_num.
  • a syntax element for indicating an IBC filter mode can be signaled, as will be described later in the second embodiment for the IBC filter mode below.
  • the IBC filter mode may be defined by at least one of the signaled ibc_filter_flag, ibc_filter_shape, ibc_filter_tpl_shape1, ibc_filter_tpl_shape2 and ibc_filter_set_num syntax elements.
  • - L-shaped IBC-LIC mode can be a mode that applies filtering using T L-shape , rectangular filter and one set of filter coefficients.
  • Top-only IBC-LIC mode can be a mode that applies filtering using T Top , a rectangular filter, and one set of filter coefficients.
  • - Left-only IBC-LIC mode can be a mode that applies filtering using T Left , a rectangular filter, and one set of filter coefficients.
  • Top-only FIBC mode can be a mode that applies filtering using T Top , a cross-shaped filter, and one set of filter coefficients.
  • - Left-only FIBC mode can be a mode that applies filtering using T Left , a cross filter, and one set of filter coefficients.
  • the IBC filter mode may be determined by at least one of the signaled ibc_filter_flag, ibc_filter_shape, ibc_filter_tpl_shape, ibc_filter_set_num.

Landscapes

  • Engineering & Computer Science (AREA)
  • Multimedia (AREA)
  • Signal Processing (AREA)
  • Compression Or Coding Systems Of Tv Signals (AREA)

Abstract

La divulgation concerne un procédé, un dispositif et un support d'enregistrement pour le codage/décodage d'image. Le procédé de codage/décodage d'image peut comprendre les étapes consistant à : acquérir des informations de filtrage de signal de prédiction ; générer un signal de prédiction ; et filtrer le signal de prédiction sur la base des informations de filtrage de signal de prédiction. L'étape de filtrage du signal de prédiction peut comprendre les étapes consistant à : identifier un modèle de filtre ; acquérir des coefficients de filtrage selon le modèle de filtre ; et filtrer le signal de prédiction sur la base des coefficients de filtrage. Le signal de prédiction peut être un signal de prédiction de copie intra-bloc généré par copie intra-bloc.
PCT/KR2025/000673 2024-01-10 2025-01-10 Procédé, dispositif et support d'enregistrement pour le codage/décodage d'image Pending WO2025151003A1 (fr)

Applications Claiming Priority (8)

Application Number Priority Date Filing Date Title
KR20240004445 2024-01-10
KR10-2024-0004445 2024-01-10
KR20240006993 2024-01-16
KR10-2024-0006993 2024-01-16
KR10-2024-0008229 2024-01-18
KR20240008229 2024-01-18
KR10-2025-0004268 2025-01-10
KR1020250004268A KR20250109639A (ko) 2024-01-10 2025-01-10 영상 부호화/복호화를 위한 방법, 장치 및 기록 매체

Publications (1)

Publication Number Publication Date
WO2025151003A1 true WO2025151003A1 (fr) 2025-07-17

Family

ID=96387247

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/KR2025/000673 Pending WO2025151003A1 (fr) 2024-01-10 2025-01-10 Procédé, dispositif et support d'enregistrement pour le codage/décodage d'image

Country Status (1)

Country Link
WO (1) WO2025151003A1 (fr)

Citations (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
KR101757947B1 (ko) * 2009-10-05 2017-07-13 톰슨 라이센싱 비디오 인코딩 및 디코딩에서의 크로마 성분에 대한 예측 픽셀의 적응성 필터링을 위한 방법 및 장치
KR20180093950A (ko) * 2016-01-11 2018-08-22 삼성전자주식회사 영상 부호화 방법 및 장치와 영상 복호화 방법 및 장치
KR20190106804A (ko) * 2018-03-09 2019-09-18 한국전자통신연구원 샘플 필터링을 이용한 영상 부호화/복호화 방법 및 장치
KR102160667B1 (ko) * 2015-09-10 2020-09-28 엘지전자 주식회사 비디오 코딩 시스템에서 인트라 예측 방법 및 장치

Patent Citations (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
KR101757947B1 (ko) * 2009-10-05 2017-07-13 톰슨 라이센싱 비디오 인코딩 및 디코딩에서의 크로마 성분에 대한 예측 픽셀의 적응성 필터링을 위한 방법 및 장치
KR102160667B1 (ko) * 2015-09-10 2020-09-28 엘지전자 주식회사 비디오 코딩 시스템에서 인트라 예측 방법 및 장치
KR20180093950A (ko) * 2016-01-11 2018-08-22 삼성전자주식회사 영상 부호화 방법 및 장치와 영상 복호화 방법 및 장치
KR20190106804A (ko) * 2018-03-09 2019-09-18 한국전자통신연구원 샘플 필터링을 이용한 영상 부호화/복호화 방법 및 장치

Non-Patent Citations (1)

* Cited by examiner, † Cited by third party
Title
H.-J. JHU (KWAI), X. XIU, C.-W. KUO, W. CHEN, N. YAN, C. MA, X. WANG (KWAI): "Non-EE2: Filtered Intra Block Copy (FIBC)", 30. JVET MEETING; 20230421 - 20230428; ANTALYA; (THE JOINT VIDEO EXPLORATION TEAM OF ISO/IEC JTC1/SC29/WG11 AND ITU-T SG.16 ), 21 April 2023 (2023-04-21), pages 1 - 3, XP030309003 *

Similar Documents

Publication Publication Date Title
WO2021101345A1 (fr) Procédé et dispositif de filtrage en boucle adaptative
WO2019177354A1 (fr) Dispositif et procédé de codage/décodage d'image et support d'enregistrement ayant un train de bits stocké en son sein
WO2016195453A1 (fr) Procédé de codage et de décodage d'image et dispositif de décodage d'image
WO2021015581A1 (fr) Procédé, appareil et support d'enregistrement pour coder/décoder une image à l'aide d'un partitionnement géométrique
WO2019107927A1 (fr) Procédé et appareil de prédiction intra bidirectionnelle
WO2018174617A1 (fr) Procédé de prédiction basé sur une forme de bloc et dispositif associé
WO2020060316A1 (fr) Procédé et dispositif de codage/décodage d'images et support d'enregistrement stockant un flux binaire
WO2020050600A1 (fr) Procédé et dispositif de codage/décodage vidéo, et support d'enregistrement pour stockage de flux binaire
WO2020060242A1 (fr) Procédé et dispositif d'encodage/de décodage de mode d'intraprédiction, et support d'enregistrement dans lequel est stocké un train de bits
WO2019074273A1 (fr) Procédé et dispositif utilisant des informations d'inter-prédiction
WO2020256466A1 (fr) Mode de prédiction intra, et procédé et dispositif de codage/décodage entropique
WO2020060317A1 (fr) Procédé et appareil de codage/décodage d'image, et support d'enregistrement pour stocker un flux binaire
WO2020226359A1 (fr) Procédé et dispositif de codage/décodage d'image et support d'enregistrement mémorisant un flux binaire
WO2021112652A1 (fr) Procédé, appareil et support d'enregistrement pour codage/décodage d'image différentielle basée sur une zone
WO2020017873A1 (fr) Procédé et appareil de codage/décodage d'image, et support d'enregistrement sur lequel un flux binaire est stocké
WO2020256324A1 (fr) Procédé et appareil de codage/décodage vidéo et support d'enregistrement contenant en mémoire un flux binaire
WO2021071342A1 (fr) Procédé et dispositif de codage/décodage d'informations par transformée et support de stockage de flux binaire
WO2018174618A1 (fr) Procédé et dispositif de prédiction à l'aide d'un bloc de référence
WO2019117645A1 (fr) Procédé et dispositif de codage et de décodage d'image utilisant un réseau de prédiction
WO2024147679A1 (fr) Procédé, appareil et support d'enregistrement permettant un codage/décodage d'image
WO2024076220A1 (fr) Procédé, dispositif et support d'enregistrement pour codage/décodage d'image
WO2019147067A1 (fr) Méthode et appareil de codage et de décodage d'image à l'aide d'informations de mouvement temporelles
WO2021101356A1 (fr) Procédé et dispositif de traitement vidéo utilisant une résolution d'image de référence, et support d'enregistrement
WO2020256495A1 (fr) Procédé, dispositif et support d'enregistrement pour coder/décoder une image à l'aide d'une représentation de référence
WO2023055199A1 (fr) Procédé, appareil et support d'enregistrement pour coder/décoder une image

Legal Events

Date Code Title Description
121 Ep: the epo has been informed by wipo that ep was designated in this application

Ref document number: 25739163

Country of ref document: EP

Kind code of ref document: A1