Video encoding method and apparatus

Abstract

A video encoding method and apparatus are disclosed. An intra prediction method for video decoding includes determining an intra prediction mode belonging to an initial directional prediction range for an intra prediction of an MxN current block, determining whether to correct the intra prediction mode from the initial directional prediction range to a corrected directional prediction range based on a horizontal length N and a vertical length M of the MxN current block, determining a corrected intra prediction mode belonging to the corrected directional prediction range, and performing the intra prediction of the MxN current block based on the corrected intra prediction mode, the initial directional prediction range including a first prediction range, a second prediction range, and a third prediction range, the corrected directional prediction range including the first prediction range, the second prediction range, and a fourth prediction range according to the vertical length M, the intra prediction being performed by using reference regions adjacent to the current block, the reference regions including an upper reference pixel column and a left reference pixel column of the current block.

Description

[0001] This application is a divisional application of Chinese invention patent application No. 2017800283233 (international application No. PCT / KR2017 / 002578), filed on March 9, 2017, entitled "Video Coding Method and Apparatus". Technical Field

[0002] This invention relates to an image processing technology. Background Technology

[0003] With the increasing demand for high-resolution, high-quality video, efficient video compression technology, which is needed to provide next-generation video services, has become even more crucial.

[0004] In the field of video compression technology, quantization coefficient encoding and decoding technology refers to the technology of using entropy encoding technology to convert the signal obtained by performing conversion and quantization on the differential signal between the original signal and the predicted signal into a bit stream, or using entropy decoding technology to restore the bit stream generated by the above method into a differential signal. Summary of the Invention

[0005] The purpose of this invention is to provide a method and apparatus for improving coding efficiency in video compression technology.

[0006] The purpose of this invention is to provide a method and apparatus for effectively encoding / decoding video by deriving motion vector information in a decoder to improve the transmission efficiency of motion information in video encoders / decoders for high-resolution images such as FHD (Full High Definition) and UHD (Ultra High Definition).

[0007] The purpose of some embodiments of the present invention is to provide a method and apparatus for performing global motion compensation for a large area in an image.

[0008] The purpose of some embodiments of the present invention is to provide a method and apparatus for generating reference signals needed to effectively perform in-frame prediction.

[0009] The purpose of some embodiments of the present invention is to provide a method and apparatus for using in-frame prediction technology with curves in video compression technology.

[0010] However, the technical issues to be addressed in this embodiment are not limited to the aforementioned technical issues, and other technical issues may also exist.

[0011] To address the aforementioned issues, a video decoding apparatus and method according to embodiments of the present invention include: a quantization coefficient group information acquisition unit that acquires quantization coefficient group information required for performing inverse quantization; a quantization coefficient group entropy decoding unit that acquires quantization coefficients by entropy decoding of the quantization coefficient group; an inverse quantization unit that acquires conversion coefficients by performing inverse quantization on the acquired quantization coefficients; and an inverse conversion unit that acquires a differential signal by performing an inverse conversion process on the acquired conversion coefficients.

[0012] To address the aforementioned issues, a video decoding apparatus and method according to embodiments of the present invention include: a variable-size quantization coefficient group usage extraction unit, which extracts variable-size quantization coefficient group usage information from the bitstream of the current decoded bitstream; a quantization coefficient group segmentation information decoding unit, which, when the extracted variable-size quantization coefficient group usage information indicates that a variable-size quantization coefficient group is used, obtains segmentation information of the quantization coefficient group required for performing inverse quantization from the current decoding unit; and a quantization coefficient entropy decoding unit, which performs entropy decoding on the quantization coefficients.

[0013] To address the aforementioned issues, a video decoding apparatus and method according to embodiments of the present invention include: a quantization coefficient group segmentation marker extraction unit that extracts quantization coefficient group segmentation markers related to segmentation from a bitstream based on the size of a current decoding unit; a quantization coefficient group size determination unit that determines the size of the quantization coefficient group from the current decoding unit when the extracted quantization coefficient group segmentation markers indicate that segmentation is not performed; a lower-level quantization coefficient group segmentation unit that segments the current decoding unit into multiple lower-level quantization coefficient groups when the extracted quantization coefficient group segmentation markers indicate that segmentation is performed; and a quantization coefficient group entropy decoding unit that performs entropy decoding on the quantization coefficient groups.

[0014] To address the aforementioned issues, a video decoding apparatus and method according to embodiments of the present invention include: a variable-size quantization coefficient group usage extraction unit, which extracts variable-size quantization coefficient group usage information from a bitstream for the current decoded bitstream; a quantization coefficient group segmentation method determination unit, which determines a segmentation method for the variable-size quantization coefficient group when the extracted variable-size quantization coefficient group usage information indicates that the variable-size quantization coefficient group is used; and a quantization coefficient group size information acquisition unit, which acquires the size information of the quantization coefficient group required for performing inverse quantization from the current decoding unit according to the determined variable-size quantization coefficient group segmentation method.

[0015] To address the aforementioned issues, a video decoding apparatus and method according to embodiments of the present invention include: a quantization coefficient group segmentation quantity information extraction unit, which extracts quantization coefficient group segmentation quantity information related to segmentation from a bitstream based on the current size of the decoding unit; and a quantization coefficient group segmentation unit, which segments the quantization coefficient group based on the aforementioned variable-size quantization coefficient group segmentation method, the current size of the decoding unit, and the aforementioned quantization coefficient group segmentation quantity information, and uses defined segmentation information.

[0016] To address the aforementioned issues, a video decoding apparatus and method according to embodiments of the present invention include: a variable-size quantization coefficient group usage information extraction unit, which extracts variable-size quantization coefficient group usage information from the bitstream of the current decoded bitstream; a quantization coefficient group segmentation information acquisition unit, which, when the extracted variable-size quantization coefficient group usage information indicates that a variable-size quantization coefficient group is used, acquires segmentation information of the quantization coefficient group required for inverse quantization from the current decoding unit; and an entropy decoding scan order acquisition unit, which acquires the entropy decoding scan order of the quantization coefficient group based on the segmentation information of the quantization coefficient group required for inverse quantization.

[0017] In one embodiment of the present invention used to solve the above-mentioned problems, the video decoding apparatus and method include a motion information deriving device or step, which can derive motion information from the decoder without the motion vector information directly transmitted from the encoder.

[0018] As a technical means to achieve the above-mentioned problem, a video decoding apparatus and method according to one embodiment of the present invention includes: a step of obtaining information required to perform global motion compensation from a bitstream; a step of determining a global motion compensation region using the information required to perform global motion compensation; and a step of performing global motion compensation on the determined global motion compensation region.

[0019] As a technical means to achieve the above-mentioned problem, a video decoding apparatus and method according to one embodiment of the present invention includes: a step of extracting a flag from a bitstream indicating whether global motion compensation is used; and a step of extracting information from the bitstream, when the extracted flag indicates that global motion compensation is used, for determining a global motion compensation region and performing motion compensation for each determined global motion region.

[0020] As a technical means to achieve the above-mentioned problem, a video decoding apparatus and method applicable to one embodiment of the present invention includes: a step of extracting a flag from a bitstream indicating whether global motion compensation is used; and a step of performing motion compensation on a block-by-block basis when the extracted flag indicates that global motion compensation is not used.

[0021] As a technical means to achieve the above-mentioned problem, a video decoding apparatus and method according to one embodiment of the present invention includes: a step of determining the execution area of ​​global motion compensation using motion compensation region determination information obtained from a bitstream; and a step of performing motion compensation on each of the determined motion compensation execution areas.

[0022] As a technical means to achieve the above-mentioned problem, a video decoding apparatus and method applicable to one embodiment of the present invention includes the step of performing global motion compensation on a global motion compensation region that performs global motion compensation according to each global motion compensation region, using execution information of each motion compensation region obtained from the bit stream.

[0023] As a technical means to achieve the above-mentioned technical problems, the video decoding method and apparatus according to one embodiment of the present invention can generate a signal related to the unrestored area using the peripheral restored signal used as a reference during the in-frame prediction process, thereby effectively performing in-frame prediction. Furthermore, it can expand the range of the restored signal used as a reference during the in-frame prediction process, thereby referencing more restored pixels and improving in-frame prediction performance compared to existing methods.

[0024] As a technical means to achieve the above-mentioned problem, a video decoding apparatus and method according to one embodiment of the present invention includes: a step of extracting information for generating a prediction signal from a bitstream; a step of performing reference sample filling using the extracted information; a step of performing in-frame prediction using the extracted information and generating prediction samples; and a step of filtering the generated prediction samples.

[0025] As a technical means to achieve the above-mentioned problem, a video decoding apparatus and method applicable to one embodiment of the present invention includes: a step of obtaining information for generating a prediction signal from a bitstream; and a step of extracting information related to the in-frame prediction from the bitstream when the extracted in-frame prediction mode information represents in-frame prediction of a curve.

[0026] As a technical means to achieve the above-mentioned problem, a video decoding apparatus and method applicable to one embodiment of the present invention includes: determining whether to perform reference sample filling and performing reference sample filling by using information related to prediction within the curve image obtained from the bitstream and the presence or absence of reference samples in the surrounding data blocks.

[0027] As a technical means to achieve the above-mentioned problem, a video decoding apparatus and method applicable to one embodiment of the present invention includes the step of generating a prediction sample using information related to prediction within a curve image obtained from a bitstream.

[0028] As a technical means to achieve the above-mentioned problem, a video decoding apparatus and method applicable to one embodiment of the present invention includes: for the left prediction sample column and the upper prediction sample row of the generated prediction data block, when the region containing the above sample column is predicted in the horizontal direction or in the vertical direction, a step of performing filtering by utilizing the change amount of the surrounding reference samples.

[0029] The purpose of this invention is to provide a method and apparatus for using a variable-size group of quantization coefficients according to the characteristics of the signal and using a coding and decoding order that corresponds to the group of quantization coefficients in order to improve the coding efficiency of quantization coefficients.

[0030] In one embodiment of the present invention, the number of coefficients excluded during the encoding process can be increased by using a variable-size quantization coefficient group, selectable coefficient encoding, and decoding order, thereby improving the performance of quantization coefficient encoding.

[0031] Furthermore, in one embodiment of the present invention, coefficient encoding performance can be improved by using a group of quantization coefficients with variable size and variable shape, thanks to the energy concentration effect of conversion and quantization and the high-frequency component removal effect.

[0032] By means of the present invention as described above, video decoding can be performed in the decoder without directly transmitting motion vectors by means of a motion information extraction device or step, thereby improving video encoding / decoding efficiency.

[0033] The purpose of this invention is to provide a method and apparatus for performing global motion compensation on a large area of ​​an image during the motion compensation process used in existing video compression technologies in order to improve coding efficiency.

[0034] In one embodiment of the present invention, decoding performance can be improved by performing motion compensation over a large area at once and effectively transmitting information related to the global motion compensation area to the decoder.

[0035] By means of the present invention described above, the in-frame prediction performance can be improved and the overall video compression performance can be enhanced by expanding the generation and reference range of the in-frame prediction signal in the decoder.

[0036] The purpose of this invention is to provide a method and apparatus for performing curve intra-frame prediction during the intra-frame prediction process used in existing video compression technologies in order to improve encoding / decoding efficiency.

[0037] By means of the present invention as described above, the efficiency of intra-frame prediction in the encoder / decoder can be improved by in-frame prediction of curves, thereby improving the overall video compression performance. Attached Figure Description

[0038] Figure 1 This is a block diagram illustrating the configuration of a video decoding apparatus to which one embodiment of the present invention is applied.

[0039] Figure 2 This is a block diagram illustrating the decoding order of a variable-size quantization group applicable to one embodiment of the present invention.

[0040] Figure 3 This is a block diagram illustrating the sequence of determining whether to decode a variable-size quantization coefficient group and obtaining quantization coefficient group segmentation information according to one embodiment of the present invention.

[0041] Figure 4 This is a block diagram illustrating the decoding order of variable-size quantized group segmentation tags applicable to one embodiment of the present invention.

[0042] Figure 5 as well as Figure 6 This is a conceptual diagram illustrating an example of the scanning order of a 4×4 fixed-size quantization coefficient group, an 8×8 decoded data block using the aforementioned quantization coefficient group, and a 16×16 decoded data block.

[0043] Figure 7 This is a conceptual diagram illustrating an example of the use of a quadtree structure and multiple scan order related to a variable-size quantization coefficient group, applicable to one embodiment of the present invention.

[0044] Figure 8 This is a conceptual diagram illustrating an example of a quadtree segmentation of a 16×16 decoded data block and a 32×32 decoded data block associated with a variable-size quantization coefficient group, applicable to one embodiment of the present invention.

[0045] Figure 9 This is a conceptual diagram illustrating an example other than a square, based on the characteristics of the input signal when segmenting a variable-size quantization coefficient group according to one embodiment of the present invention.

[0046] Figure 10 This is a conceptual diagram illustrating a group of non-square variable-size quantization coefficients applicable to one embodiment of the present invention.

[0047] Figure 11 This is a block diagram of a decoding apparatus applicable to an embodiment of the present invention.

[0048] Figure 12 It is a sequence diagram related to motion derivation and motion compensation in the decoding unit that performs motion derivation.

[0049] Figure 13 This is an example of segmenting sub-data blocks when segmenting the decoding unit in an embodiment of the present invention.

[0050] Figure 14 It is the shape of the surrounding pixels in the instance that performs motion prediction by utilizing the surrounding pixel information of the decoding unit to perform motion.

[0051] Figure 15 Is Figure 14 An example of using two reference images in the method.

[0052] Figure 16 This is a method that uses two reference images to predict the motion of the current decoding unit by corresponding data blocks during motion export.

[0053] Figure 17 A decoding apparatus for performing global motion compensation according to one embodiment of the present invention has been illustrated.

[0054] Figure 18 This is a schematic diagram illustrating a method for global motion compensation of an execution image to which one embodiment of the present invention is applied.

[0055] Figure 19 This is a block diagram illustrating the sequence of a method for performing global motion compensation applicable to one embodiment of the present invention.

[0056] Figure 20 This is a schematic diagram illustrating a method for determining the global motion compensation region by utilizing information in the information transmitted to the decoder indicating whether the global motion compensation region is located inside or outside the determined region when performing global motion compensation according to one embodiment of the present invention.

[0057] Figure 21 This is a schematic diagram illustrating various forms of global motion compensation regions when performing global motion compensation according to one embodiment of the present invention.

[0058] Figure 22 This is a schematic diagram illustrating a method for determining the global motion compensation region according to the boundary of the coding unit when performing global motion compensation in one embodiment of the present invention.

[0059] Figure 23 This is a schematic diagram illustrating a method for determining the position of a global motion compensation region when performing global motion compensation according to one embodiment of the present invention.

[0060] Figure 24 This is a schematic diagram illustrating a method for determining the global motion compensation region by merging lattice-separated regions when performing global motion compensation, according to one embodiment of the present invention.

[0061] Figure 25This is a schematic diagram illustrating a method for determining the global motion compensation region by repeatedly segmenting an image along a vertical or horizontal direction when performing global motion compensation, according to one embodiment of the present invention.

[0062] Figure 26 This is a schematic diagram illustrating a method for determining the global motion compensation region using a warping parameter in additional information passed to the decoder when performing global motion compensation, according to one embodiment of the present invention.

[0063] Figure 27 This is a schematic diagram illustrating a method for rotating or scaling a global motion compensation region when performing global motion compensation, according to one embodiment of the present invention.

[0064] Figure 28 This is a schematic diagram illustrating the use of a Frame Rate Up Conversion (FRUC) method to increase the frame rate when performing global motion compensation, according to one embodiment of the present invention.

[0065] Figure 29 A video decoding apparatus according to one embodiment of the present invention is illustrated, which can generate an intra-frame prediction signal from intra-frame prediction information of an encoded bitstream and output a restored image using the generated intra-frame prediction signal.

[0066] Figure 30 The reference area for predictive data blocks within a picture in which embodiments of the present invention are applied is illustrated.

[0067] Figure 31 An illustration is provided of a method for performing directional intra-image prediction based on the reference pixel column length according to an embodiment of the present invention.

[0068] Figure 32 An illustration shows a method for performing directional intra-image prediction based on the length of the left pixel column according to an embodiment of the present invention.

[0069] Figure 33 The range of directional prediction applicable in the in-image prediction method according to embodiments of the present invention is illustrated.

[0070] Figure 34 An illustration shows a method for generating a prediction signal by changing the brightness of a pixel based on the pixel coordinates of an unrestored pixel region and the same tilt as the restored pixel region, according to the signal of an adjacent restored pixel region.

[0071] Figure 35 An illustration shows a method for generating a prediction signal by changing the brightness of a pixel based on the pixel coordinates of an unrestored pixel region and the same negative tilt as the restored pixel region, according to the signal of an adjacent restored pixel region.

[0072] Figure 36 Another method for generating a prediction signal in an in-image prediction method applicable to an embodiment of the present invention is illustrated, which involves changing the brightness of a pixel based on the pixel coordinates of an unrestored pixel region from the signal of an adjacent restored pixel region with the same tilt as the restored pixel region.

[0073] Figure 37 An embodiment of the present invention is illustrated for transmitting a method for determining whether a recommended in-frame prediction instruction signal needs to be transmitted using the sequence parameter set in the high-level syntax through an in-frame prediction method.

[0074] Figure 38 A method for transmitting a command signal indicating whether or not the recommended intra-frame prediction should be performed using the image parameter set in the high-level syntax, according to an embodiment of the present invention, is illustrated.

[0075] Figure 39 An embodiment of the present invention is illustrated for transmitting a method for determining whether the recommended intra-frame prediction instruction signal needs to be executed using the stripe fragment header in the high-level syntax through an intra-frame prediction method.

[0076] Figure 40 A decoding apparatus including an in-picture prediction unit, according to one embodiment of the present invention, has been illustrated.

[0077] Figure 41 The surrounding reference area during in-screen prediction is illustrated in accordance with one embodiment of the present invention.

[0078] Figure 42 A method for referencing pixels of surrounding data blocks when performing in-frame prediction, according to one embodiment of the present invention, is illustrated.

[0079] Figure 43 A method for referencing multiple pixels of a surrounding data block when performing in-frame prediction, according to one embodiment of the present invention, is illustrated.

[0080] Figure 44A method for generating non-existent reference samples of surrounding data blocks during in-screen prediction, according to one embodiment of the present invention, is illustrated.

[0081] Figure 45 An illustration is provided of a method for performing predictions in different regions of a prediction data block using reference samples from different directions, according to one embodiment of the present invention.

[0082] Figure 46 Another method for performing predictions in different regions of a prediction data block using reference samples from different directions, according to one embodiment of the present invention, is illustrated.

[0083] Figure 47 A method for filtering the leftmost prediction sample column of a prediction data block to eliminate discontinuities with surrounding data blocks during in-screen prediction, according to one embodiment of the present invention, is illustrated.

[0084] Figure 48 A method for filtering the topmost prediction sample column of a prediction data block to eliminate discontinuities with surrounding data blocks during in-screen prediction, according to one embodiment of the present invention, is illustrated.

[0085] Figure 49 The order of in-screen predictions applicable to one embodiment of the present invention is illustrated. Detailed Implementation

[0086] To address the aforementioned issues, a video decoding apparatus and method according to embodiments of the present invention include: a quantization coefficient group information acquisition unit that acquires quantization coefficient group information required for performing inverse quantization; a quantization coefficient group entropy decoding unit that acquires quantization coefficients by entropy decoding of the quantization coefficient group; an inverse quantization unit that acquires conversion coefficients by performing inverse quantization on the acquired quantization coefficients; and an inverse conversion unit that acquires a differential signal by performing an inverse conversion process on the acquired conversion coefficients.

[0087] To address the aforementioned issues, a video decoding apparatus and method according to embodiments of the present invention include: a variable-size quantization coefficient group usage extraction unit, which extracts variable-size quantization coefficient group usage information from the bitstream of the current decoded bitstream; a quantization coefficient group segmentation information decoding unit, which, when the extracted variable-size quantization coefficient group usage information indicates that a variable-size quantization coefficient group is used, obtains segmentation information of the quantization coefficient group required for performing inverse quantization from the current decoding unit; and a quantization coefficient entropy decoding unit, which performs entropy decoding on the quantization coefficients.

[0088] To address the aforementioned issues, a video decoding apparatus and method according to embodiments of the present invention include: a quantization coefficient group segmentation marker extraction unit that extracts quantization coefficient group segmentation markers related to segmentation from a bitstream based on the size of a current decoding unit; a quantization coefficient group size determination unit that determines the size of the quantization coefficient group from the current decoding unit when the extracted quantization coefficient group segmentation markers indicate that segmentation is not performed; a lower-level quantization coefficient group segmentation unit that segments the current decoding unit into multiple lower-level quantization coefficient groups when the extracted quantization coefficient group segmentation markers indicate that segmentation is performed; and a quantization coefficient group entropy decoding unit that performs entropy decoding on the quantization coefficient groups.

[0089] To address the aforementioned issues, a video decoding apparatus and method according to embodiments of the present invention include: a variable-size quantization coefficient group usage extraction unit, which extracts variable-size quantization coefficient group usage information from a bitstream for the current decoded bitstream; a quantization coefficient group segmentation method determination unit, which determines a segmentation method for the variable-size quantization coefficient group when the extracted variable-size quantization coefficient group usage information indicates that the variable-size quantization coefficient group is used; and a quantization coefficient group size information acquisition unit, which acquires the size information of the quantization coefficient group required for performing inverse quantization from the current decoding unit according to the determined variable-size quantization coefficient group segmentation method.

[0090] To address the aforementioned issues, a video decoding apparatus and method according to embodiments of the present invention include: a quantization coefficient group segmentation quantity information extraction unit, which extracts quantization coefficient group segmentation quantity information related to segmentation from a bitstream based on the current decoding unit size; and a quantization coefficient group segmentation unit, which segments the quantization coefficient group based on the aforementioned variable-size quantization coefficient group segmentation method, the current decoding unit size, and the aforementioned quantization coefficient group segmentation quantity information, and uses defined segmentation information to segment the quantization coefficient group.

[0091] To address the aforementioned issues, a video decoding apparatus and method according to embodiments of the present invention include: a variable-size quantization coefficient group usage information extraction unit, which extracts variable-size quantization coefficient group usage information from the bitstream of the current decoded bitstream; a quantization coefficient group segmentation information acquisition unit, which, when the extracted variable-size quantization coefficient group usage information indicates that a variable-size quantization coefficient group is used, acquires segmentation information of the quantization coefficient group required for inverse quantization from the current decoding unit; and an entropy decoding scan order acquisition unit, which acquires the entropy decoding scan order of the quantization coefficient group based on the segmentation information of the quantization coefficient group required for inverse quantization.

[0092] In one embodiment of the present invention used to solve the above-mentioned problems, the video decoding apparatus and method include a motion information deriving device or step, which can derive motion information from the decoder without the motion vector information directly transmitted from the encoder.

[0093] As a technical means to achieve the above-mentioned problem, a video decoding apparatus and method according to one embodiment of the present invention includes: a step of obtaining information required to perform global motion compensation from a bitstream; a step of determining a global motion compensation region using the information required to perform global motion compensation; and a step of performing global motion compensation on the determined global motion compensation region.

[0094] As a technical means to achieve the above-mentioned problem, a video decoding apparatus and method according to one embodiment of the present invention includes: a step of extracting a flag from a bitstream indicating whether global motion compensation is used; and a step of extracting information from the bitstream, when the extracted flag indicates that global motion compensation is used, for determining a global motion compensation region and performing motion compensation for each determined global motion region.

[0095] As a technical means to achieve the above-mentioned problem, a video decoding apparatus and method applicable to one embodiment of the present invention includes: a step of extracting a flag from a bitstream indicating whether global motion compensation is used; and a step of performing motion compensation on a block-by-block basis when the extracted flag indicates that global motion compensation is not used.

[0096] As a technical means to achieve the above-mentioned problem, a video decoding apparatus and method applicable to one embodiment of the present invention includes: a step of determining the execution region of global motion compensation using motion compensation region determination information obtained from a bitstream; and a step of performing motion compensation on each of the determined motion compensation execution regions.

[0097] As a technical means to achieve the above-mentioned problem, a video decoding apparatus and method applicable to one embodiment of the present invention includes: a step of performing global motion compensation on a global motion compensation region that performs global motion compensation according to each global motion compensation region, using execution information of each motion compensation region obtained from a bitstream.

[0098] As a technical means to achieve the above-mentioned technical problems, the video decoding method and apparatus according to one embodiment of the present invention can generate a signal related to the unrestored area using the peripheral restored signal used as a reference during the in-frame prediction process, thereby effectively performing in-frame prediction. Furthermore, it can expand the range of the restored signal used as a reference during the in-frame prediction process, thereby referencing more restored pixels and improving in-frame prediction performance compared to existing methods.

[0099] As a technical means to achieve the above-mentioned problem, a video decoding apparatus and method according to one embodiment of the present invention includes: a step of extracting information for generating a prediction signal from a bitstream; a step of performing reference sample filling using the extracted information; a step of performing in-frame prediction using the extracted information and generating prediction samples; and a step of filtering the generated prediction samples.

[0100] As a technical means to achieve the above-mentioned problem, a video decoding apparatus and method applicable to one embodiment of the present invention includes: a step of obtaining information for generating a prediction signal from a bitstream; and a step of extracting information related to the in-frame prediction from the bitstream when the extracted in-frame prediction mode information represents in-frame prediction of a curve.

[0101] As a technical means to achieve the above-mentioned problem, a video decoding apparatus and method applicable to one embodiment of the present invention includes: determining whether to perform reference sample filling and performing reference sample filling by using information related to prediction within the curve image obtained from the bitstream and the presence or absence of reference samples in the surrounding data blocks.

[0102] As a technical means to achieve the above-mentioned problem, a video decoding apparatus and method applicable to one embodiment of the present invention includes the step of generating a prediction sample using information related to prediction within a curve image obtained from a bitstream.

[0103] As a technical means to achieve the above-mentioned problem, a video decoding apparatus and method applicable to one embodiment of the present invention includes: for the left prediction sample column and the upper prediction sample row of the generated prediction data block, when the region containing the above sample column is predicted in the horizontal direction or in the vertical direction, a step of performing filtering by utilizing the change amount of the surrounding reference samples.

[0104] Next, to facilitate easy implementation of the invention by those skilled in the art, embodiments to which the invention applies will be described in detail with reference to the accompanying drawings. However, the invention can be implemented in many different forms and is not limited to the embodiments described herein. Furthermore, irrelevant parts have been omitted in the drawings for clarity of explanation, and similar drawing numbers have been assigned to similar parts throughout the specification.

[0105] Throughout the instruction manual, when a part is described as being "connected" to other parts, this includes not only direct connections but also cases where an electrical connection is achieved between the two parts through other components.

[0106] Furthermore, throughout the specification, when a part is described as "including" a certain constituent element, unless otherwise expressly stated to the contrary, it does not exclude the existence of other constituent elements, but rather indicates that other constituent elements can also be included.

[0107] The degree-related terms used throughout the specification, such as ~(of) step or ~ step, do not represent steps for the purpose of ~.

[0108] Furthermore, terms such as "first" and "second" may be used in describing different constituent elements, but these terms do not limit the constituent elements. These terms are merely used to distinguish one constituent element from others.

[0109] Furthermore, the constituent parts marked in the embodiments of the present invention are illustrated separately only to show different features and functions, and do not represent that each constituent part is composed of separate hardware or a single software unit. That is, the constituent parts are listed as different constituent parts only for the convenience of explanation, and it is possible to combine at least two of the constituent parts into one constituent part, or to divide one constituent part into multiple constituent parts to achieve its function. The integrated and separate embodiments of the constituent parts described above are also included within the scope of the claims of the present invention without departing from the essence of the present invention.

[0110] Furthermore, some constituent elements may not be essential for performing the essential functions of this invention, but rather optional elements used to improve performance. This invention can include only the constituent parts essential for realizing the essence of the invention, excluding those merely used to improve performance, and structures including essential constituent elements, excluding optional elements used to improve performance, are also included within the scope of the claims of this invention.

[0111] The data blocks used in this invention can be basic data block units for decoding, prediction data block units, or transformation data block units. Furthermore, the data block boundaries can be the boundaries of decoding data blocks, prediction data blocks, or transformation data blocks.

[0112] First, the terminology used in this application will be briefly explained below.

[0113] The video decoding device mentioned later can be a device contained in server terminals such as personal computers (PCs), laptops, portable multimedia players (PMPs), wireless communication terminals, smartphones, TV application servers, and server servers. It can include: user terminals such as various devices; communication devices such as communication modems for communicating with wired or wireless communication networks; memory for storing various programs and data required for decoding images or for predicting between or within frames during decoding; and microprocessors for performing calculations and control by executing programs; and so on.

[0114] Furthermore, images encoded into bitstreams by the encoder can be transmitted in real time or non-real time to the decoding device via the Internet, near-field wireless communication networks, wireless local area networks, wireless broadband access networks (Wibro), mobile communication networks, or via various types of communication interfaces such as cables and Universal Serial Bus (USB), and then decoded and restored into images for playback.

[0115] Typically, a video can be composed of a series of pictures, and each picture can be divided into coding units such as blocks. Furthermore, those skilled in the art to which this embodiment pertains should understand that the term "picture" as used below can be replaced by other terms with the same meaning, such as image or frame.

[0116] Global motion compensation refers to a method of performing motion compensation on a large area at once. The method of performing global motion compensation is called the global motion compensation method, and the area where global motion compensation is performed is called the global motion compensation area.

[0117] In the various embodiments of the invention described in this specification, "quantization coefficient group" refers to the processing unit of quantization conversion coefficients after conversion and quantization processes. It can be a collective term for all groups including converted signal groups that only perform conversion, quantized signal groups that only perform quantization processes, and signal groups that have not undergone conversion and quantization.

[0118] Next, a detailed description will be given of a video decoding apparatus and method including a variable-size quantization coefficient group proposed in one embodiment of the present invention.

[0119] Figure 1 This is a block diagram illustrating the configuration of a video decoding apparatus and method applicable to one embodiment of the present invention.

[0120] A video decoding apparatus and method according to one embodiment may include at least one of an entropy decoding unit 110, an inverse quantization unit 120, an inverse conversion unit 130, an intra-frame prediction unit 140, an inter-frame prediction unit 150, a summing unit 160, an intra-loop filtering unit 170, and a restored image buffer 180.

[0121] The entropy decoding unit 110 decodes the input bitstream 100 and outputs decoding information such as syntax elements and quantization coefficients.

[0122] The inverse quantization unit 120 and the inverse conversion unit 130 receive quantization coefficients and sequentially perform inverse quantization and inverse conversion to output a residual signal.

[0123] The in-frame prediction unit 140 generates a prediction signal by performing spatial prediction using the pixel values ​​of the decoded surrounding data blocks adjacent to the current data block being decoded.

[0124] The inter-frame prediction unit 150 generates a prediction signal by performing motion compensation using motion vectors extracted from the bitstream and the restored image stored in the restored image buffer 180.

[0125] The prediction signals output from the in-frame prediction unit 140 and the inter-frame prediction unit 150 are summed with the residual signal by the summing unit 160. At this time, the restored signal generated in units of data blocks includes the restored image.

[0126] The restored image is passed to the in-loop filtering unit 170. The restored image after filtering is saved in the restored image buffer 180 and can be used as a reference image for the inter-frame prediction unit 150.

[0127] Figure 2 This is a block diagram illustrating the decoding order of a variable-size quantization group applicable to one embodiment of the present invention.

[0128] A video decoding apparatus and method according to one embodiment includes at least one of a quantization coefficient group information decoding unit 210, a quantization coefficient entropy decoding unit 220, an inverse quantization unit 230, an inverse conversion unit 250, and a differential signal acquisition unit 260.

[0129] The quantization coefficient group information decoding unit 210 is used to extract information related to the quantization coefficient group from the bit stream.

[0130] Information related to the quantization coefficient group applicable to one embodiment includes whether a variable-size quantization group is used and its size, or the use of a variable-size quantization group, its size, and its segmentation pattern. Furthermore, the quantization coefficient group information applicable to one embodiment can be included in a sequence parameter set, image parameter set, strip header, or decoding unit, and can be transmitted through one or more units as described above. Meanwhile, the quantization coefficient group information applicable to one embodiment can be represented in a marker form, the minimum or maximum size of the quantization coefficient group, and a depth form corresponding to the available size. In this case, the minimum or maximum size can be represented in logarithmic form. The quantization coefficient group information extracted from the bitstream by the quantization coefficient group information decoding unit 210 is transmitted to the quantization coefficient entropy decoding unit 220.

[0131] The quantization coefficient entropy decoding unit 220 is executed in decoding units and is used to decode the encoded quantization coefficients from the bit stream.

[0132] Entropy decoding of quantization coefficients according to one embodiment refers to extracting the quantization coefficients corresponding to the currently decoded quantization coefficient group from the bitstream using the quantization coefficient group information extracted by the quantization coefficient group information decoding unit 210. Furthermore, when extracting quantization coefficients from the bitstream according to one embodiment, the entropy decoding scan order can use a scan order predefined in the current quantization coefficient group information, or the entropy decoding scan order related to the quantization coefficient group can be transmitted as additional information.

[0133] The inverse quantization unit 230 is used to perform inverse quantization on the quantization coefficients extracted by the quantization coefficient entropy decoding unit 220.

[0134] In one embodiment, the inverse quantization unit performs inverse quantization on the quantization coefficients extracted by the quantization coefficient entropy decoding unit 220. However, it is also possible not to perform inverse quantization when there are no extracted quantization coefficients or when the quantization execution is false.

[0135] In one embodiment, the signal extracted by the inverse quantization unit 230 is used to determine whether the conversion is performed 240. If the conversion is performed correctly, the differential signal is obtained through the inverse conversion unit 250. Conversely, if the conversion is performed incorrectly, the signal extracted by the inverse quantization unit 230 is used directly as the differential signal without going through the inverse conversion unit 250.

[0136] Figure 3 This is a block diagram illustrating the sequence of determining whether to decode a variable-size quantization coefficient group and obtaining quantization coefficient group segmentation information according to one embodiment of the present invention.

[0137] A video decoding apparatus and method according to one embodiment includes at least one of a variable-size quantization coefficient group usage extraction unit 310, a variable-size quantization coefficient group usage determination unit 320, a quantization coefficient group segmentation information decoding unit 330, and a quantization coefficient entropy decoding unit 340.

[0138] The variable-size quantization coefficient group usage extraction unit 310 is used to extract whether the variable-size quantization coefficient group is used or not.

[0139] The use or non-use of a variable-size quantization coefficient group applicable to one embodiment refers to information used to determine whether the variable-size quantization coefficient group proposed in this invention is used during the decoding process of the quantization coefficient group. This information can be represented using a marker format or by representing the segmentation format of the used variable-size quantization coefficient group as a specific value format. Furthermore, the use or non-use of a variable-size quantization coefficient group applicable to one embodiment can be included in the sequence parameter set, image parameter set, strip header, decoding unit, or quantization coefficient group, and can be transmitted through one or more units as described above.

[0140] The variable-size quantization coefficient group usage judgment unit 320 is used to determine whether the variable-size quantization coefficient group extracted by the variable-size quantization coefficient group usage extraction unit 310 is used.

[0141] The quantization coefficient group segmentation information decoding unit 330 is used to acquire quantization coefficient group segmentation information.

[0142] In one embodiment, when the variable-size quantization coefficient group is true, the quantization coefficient group segmentation information decoding unit 330 extracts information related to the quantization coefficient group from the bit stream.

[0143] Information related to the quantization coefficient group applicable to one embodiment includes the size of the quantization coefficient group, or the size of the quantization coefficient group and the segmentation pattern of the quantization coefficient group. Furthermore, the quantization coefficient group information applicable to one embodiment can be included in a sequence parameter set, image parameter set, strip header, decoding unit, or quantization coefficient group, and can be transmitted through one or more units as described above. Meanwhile, the quantization coefficient group information applicable to one embodiment can be represented in a marker form, the minimum or maximum size of the quantization coefficient group, and a depth form corresponding to the available size, etc. In this case, the minimum or maximum size can be represented in logarithmic form. The quantization coefficient group information extracted from the bitstream by the quantization coefficient group segmentation information decoding unit 330 is transmitted to the quantization coefficient entropy decoding unit 340.

[0144] The quantization coefficient entropy decoding unit 340 is executed in decoding units to decode the encoded quantization coefficients from the bit stream.

[0145] Entropy decoding of quantization coefficients according to one embodiment refers to extracting the quantization coefficients corresponding to the currently decoded quantization coefficient group from the bitstream using the quantization coefficient group information extracted by the quantization coefficient group segmentation information decoding unit 330. Furthermore, when extracting quantization coefficients from the bitstream according to one embodiment, the entropy decoding scan order can use a scan order predefined in the current quantization coefficient group information, or the entropy decoding scan order related to the quantization coefficient group can be transmitted as additional information.

[0146] Figure 4 This is a block diagram illustrating the decoding order of variable-size quantized group segmentation tags applicable to one embodiment of the present invention.

[0147] A video decoding apparatus and method according to one embodiment includes at least one of a quantization coefficient group segmentation marker extraction unit 410, a segmentation determination unit 420, a lower-level quantization coefficient group segmentation unit 430, a quantization coefficient group size determination unit 440, and a quantization coefficient group entropy decoding unit 450.

[0148] The quantization coefficient group segmentation marker extraction unit 410 is used to extract markers related to whether or not the current quantization coefficient group is segmented from the bitstream during the process of using variable-size quantization coefficients in the form of a quadtree.

[0149] A quantization coefficient group applicable to one embodiment can be segmented in the form of a quadtree, the quadtree segmentation structure of the quantization coefficient group including not segmenting the quantization coefficient group or segmenting it into a recursive segmentation structure of more than one depth.

[0150] The segmentation determination unit 420 is used to determine whether the current quantization coefficient group is segmented or not based on the markers related to the segmentation of the quantization coefficient group extracted by the quantization coefficient group segmentation marker extraction unit 410.

[0151] When the quantization coefficient group is segmented according to one embodiment, the lower-level quantization coefficient group segmentation unit 430 will be executed. At this time, the quantization coefficient group segmentation mark extraction unit 410 and the segmentation determination unit 420 will be executed recursively.

[0152] In one embodiment where the quantization coefficient group is not segmented, the quantization coefficient group size determination unit 440 determines the current data block size as the size of the quantization coefficient group and executes the quantization coefficient group entropy decoding unit 450, thereby performing entropy decoding on the quantization coefficient group.

[0153] Figure 5 as well as Figure 6 This is a conceptual diagram illustrating an example of the scanning order of a 4×4 fixed-size quantization coefficient group used in existing video decoding devices and methods, and an 8×8 decoding data block and a 16×16 decoding data block utilizing the aforementioned quantization coefficient group.

[0154] The scanning order of the 4×4 fixed-size quantization coefficient group used in existing video decoding devices and methods includes at least one of the following: zigzag scanning order 500, 600, horizontal scanning order 510, and vertical scanning order 520.

[0155] Figure 7 This is a conceptual diagram illustrating an example of the use of a quadtree structure and multiple scan order related to a variable-size quantization coefficient group, applicable to one embodiment of the present invention.

[0156] The entropy decoding process for a variable-size quantization coefficient group, applicable to one embodiment, can be included in the above content. Figure 5 as well as Figure 6 The scanning order used in existing video decoding devices and methods is explained, and the same scanning order 710 or different scanning order 720 can be used in different quantization coefficient groups within the same decoded data block.

[0157] When extracting quantization coefficients from a bitstream according to one embodiment, the entropy decoding scan order can use a scan order predefined in the current quantization coefficient group information, or the entropy decoding scan order associated with the quantization coefficient group can be transmitted as additional information.

[0158] Figure 8 This is a conceptual diagram illustrating an example of a quadtree segmentation of a 16×16 decoded data block and a 32×32 decoded data block associated with a variable-size quantization coefficient group, applicable to one embodiment of the present invention.

[0159] A video decoding apparatus and method applicable to one embodiment includes performing quantization coefficient entropy decoding using decoding data blocks 810 and 820 having a quantization coefficient group with a quarter tree segmentation.

[0160] A video decoding apparatus and method according to one embodiment includes an apparatus and method for using decoding data blocks 810 and 820 with quantization coefficient groups having quartet segmentation, and an apparatus and method for recursively segmenting the quantization coefficient groups according to quartet segmentation depth information. The 16×16 decoding data block 810 is an embodiment using 4×4 quantization coefficient groups 811 and 8×8 quantization coefficient groups 812 according to quartet segmentation, while the 32×32 decoding data block is an embodiment using 4×4 quantization coefficient groups 821, 8×9 quantization coefficient groups 822, and 16×16 quantization coefficient groups 823 according to quartet segmentation.

[0161] Figure 9 This is a conceptual diagram illustrating an example other than a square, based on the characteristics of the input signal when segmenting a variable-size quantization coefficient group according to one embodiment of the present invention.

[0162] A video decoding apparatus and method applicable to one embodiment includes diagonal quantization coefficient group segmentation 910 and L-shaped quantization coefficient group segmentation 920.

[0163] Figure 9 The method of segmenting the quantization coefficient group 910 using the diagonal is an embodiment of segmenting into low-frequency quantization coefficient group 1 911, low-frequency quantization coefficient group 2 912, high-frequency quantization coefficient group 1 913, and high-frequency quantization coefficient group 2 914.

[0164] A diagonal quantization coefficient group segmentation 910, applicable to one embodiment, can perform diagonal quantization coefficient group segmentation from a low-frequency region to a high-frequency region according to the characteristics of the input signal. The number of segments in the diagonal quantization coefficient group segmentation 910, applicable to one embodiment, can be adjusted using a fixed number or by extracting the number of segments from the bitstream.

[0165] Figure 9 The L-shaped quantization coefficient group segmentation 920 is an embodiment of segmentation into low-frequency quantization coefficient group 1 921, low-frequency quantization coefficient group 2 922, high-frequency quantization coefficient group 1 923, and high-frequency quantization coefficient group 2 924.

[0166] An L-shaped quantization coefficient group segmentation 920 according to one embodiment can perform quantization coefficient group segmentation using L-shaped lines 925 from a low-frequency region to a high-frequency region according to the characteristics of the input signal. The number of segments in the L-shaped quantization coefficient group segmentation 920 according to one embodiment can be adjusted using a fixed number or by extracting the number of segments from the bitstream.

[0167] Figure 10 This is a conceptual diagram illustrating a group of non-square variable-size quantization coefficients applicable to one embodiment of the present invention.

[0168] The non-square variable-size quantization coefficient group in a video decoding apparatus and method according to one embodiment includes at least one of non-square horizontal length information 1010 and vertical length information 1020. The non-square horizontal and vertical length information can be obtained from a higher-level square quantization coefficient group using segmentation information or extracted from the bitstream for use. When extracting and using the non-square horizontal and vertical length information from the bitstream according to one embodiment, it includes values ​​corresponding to the non-square horizontal and vertical length information, or values ​​derived using their corresponding index information and their relationship with surrounding quantization coefficients.

[0169] Figure 11 This is a block diagram of a decoder applicable to an embodiment of the present invention.

[0170] The decoder, upon receiving the bitstream from the encoder, generally performs decoding using inter-frame prediction 136-2 and intra-frame prediction 137-2. During decoding, inter-frame prediction can be performed using motion information transmitted from the encoder according to embodiments of the present invention, or using motion information derived from the decoder. When performing inter-frame prediction decoding using motion information transmitted from the encoder, the motion prediction unit 131 calculates the motion vector of the actual corresponding data block using the difference between the predicted motion vector (PMV) and the received motion vector, and performs motion compensation accordingly. When deriving motion vectors from the decoder and performing inter-frame prediction decoding using the derived motion information, the motion vectors are calculated by the motion derivation unit, and motion compensation is performed accordingly. During inter-frame prediction decoding, the method of receiving motion vectors from the encoder or deriving motion vectors from the decoder can be selectively applied; selection information and related information can be received from the encoder via syntax information.

[0171] Figure 12 This is a decoding sequence diagram for when an embodiment of the present invention is applied, in which the method of deriving motion information from the decoder or the method of receiving it from the encoder is selectively applied.

[0172] In this sequence diagram, the steps following motion compensation are omitted. The decoder derives motion derivation flag (MV_deriv_flagi,j) information from the received bitstream 201-2. Motion derivation flag 202-2 is selection information related to the motion derivation method, which the decoder uses to determine whether to perform decoding using the motion derivation method. The motion information derivation flag basically represents selection information related to the current decoding unit, but in different embodiments, it can represent selection information for motion derivation methods at different levels such as sequence, frame, frame group, stripe, stripe group, decoding unit, decoding unit group, sub-decoding unit, etc. When the motion derivation flag is 1, that is, the decoding unit that performed encoding using the motion derivation method will perform decoding using the motion derivation method, at which time the decoder will additionally decode the motion derivation information related to the current decoding unit 203-2. The motion derivation information related to the current decoding unit can include at least one of the following: depth information of the decoding unit using the motion derivation method, information related to the method of deriving motion information in the motion derivation method, information related to the shape / size / number of units or sub-units using motion derivation, and information related to the number of repetitions. At this point, the size, shape, etc. of the unit to be decoded will be defined based on one or more combinations of the information mentioned above, and motion export 204-2 will be executed. The depth information of the decoded unit allows us to understand the size information of the data block that actually needs to be processed using motion export 204-2. For example, when the size of the data block to which motion export is applied is 128×128, the depth information is 2, and the shape of the unit is square, it can be divided into... Figure 13 The sub-unit data block is shown in (a). This method can also be determined by the agreement between the encoder and decoder, and can segment the data in the decoder according to the information transmitted from the encoder, such as... Figure 13 (b) shows a data block of a specific size. After deriving the motion information of the current decoding unit through motion derivation 204-2, motion compensation 205-2 is performed using the above information.

[0173] Figure 14 This is a method for predicting motion information related to the current unit using the surrounding pixel information of the currently decoded unit or sub-unit, in one embodiment of the present invention.

[0174] This method utilizes the surrounding pixel information of a unit or sub-unit to perform motion prediction and uses the result as the motion vector value of the current unit or sub-unit. At this time, the current decoding unit can... Figure 14The previously decoded region shown in (b) is used as the region for performing motion prediction on the current decoding unit. At this time, if the current decoding unit performs motion prediction using region c-1, motion information related to the current decoding unit can be derived using the aforementioned motion prediction and decoded accordingly. While the decoding step could end here, to achieve more accurate motion prediction, motion prediction can be performed simultaneously using both the previously decoded region 402-2 and the region 401-2 used in the previous motion prediction. In this case, the repeated motion derivation step described above can be determined by the number of repetitions agreed upon between the encoder and decoder or by the number of repetitions transmitted from the encoder to the decoder. Furthermore, if the current decoding unit is divided into sub-units using the depth information of the decoding unit, motion information related to the current decoding unit can be derived using... Figure 14 The gray shading information shown in (d) is used to perform motion derivation for each subunit. The above is only one embodiment; various changes can be made to the size and shape of the gray shading used for motion prediction based on the agreement between the decoder and encoder. Related information can use fixed values ​​and shapes according to the agreement between the encoder / decoder, or it can be achieved by transmitting relevant information from the encoder to the decoder. Furthermore, the method of deriving motion information of the current data block using surrounding pixel information can also be applied to other methods such as... Figure 15 The motion vector values ​​of one or more reference images shown can be calculated using general methods used in video decoding when using multiple reference images. These general methods can calculate the motion values ​​based on the temporal difference between the reference images and the currently decoded image.

[0175] Figure 16 This is a method for deriving motion information of the currently decoded unit using the value of the corresponding data block (co-located block) of the currently decoded unit or sub-unit, in one embodiment of the present invention. Typically, motion vectors can be calculated by minimizing the error between corresponding data blocks of two or more reference images, using the currently decoded unit as a reference. This method also uses surrounding pixel information to predict motion information related to the current unit. It can be implemented in various ways by combining information related to the motion information derivation method, the shape of the unit or sub-unit to be used for motion derivation, the number of repetitions, etc.

[0176] Figure 17 A decoding apparatus for performing global motion compensation according to one embodiment of the present invention has been illustrated.

[0177] The decoding apparatus for performing global motion compensation may include at least one of the following: an entropy decoding unit 110-3, an inverse quantization unit 120-3, an inverse conversion unit 130-3, an inter-frame prediction unit 140-3, an intra-frame prediction unit 150-3, an intra-loop filtering unit 160-3, and a restored image storage unit 170-3.

[0178] The entropy decoding unit 110-3 decodes the input bitstream 100-3 and outputs decoding information such as syntax elements and quantization coefficients. The output information can include information required to perform global motion compensation.

[0179] The inverse quantization unit 120-3 and the inverse conversion unit 130-3 receive quantization coefficients and sequentially perform inverse quantization and inverse conversion to output a residual signal.

[0180] The inter-frame prediction unit 140-3 generates a prediction signal by performing motion compensation using motion vectors extracted from the bitstream and the restored image stored in the restored image storage unit 170-3. The inter-frame prediction unit 140-3 includes a step of performing global motion compensation on a global motion compensation region using information 190 required for performing global motion compensation.

[0181] The in-frame prediction unit 150-3 generates a prediction signal by performing spatial prediction using the pixel values ​​of the decoded surrounding data blocks adjacent to the current data block being decoded.

[0182] The prediction signals output from the inter-frame prediction unit 140-3 and the intra-frame prediction unit 150-3 will be summed with the residual signals, and the restored image generated by the summation will be transmitted to the in-loop filtering unit 160-3.

[0183] The restored image, after filtering in the in-loop filtering unit 160-3, is saved in the restored image storage unit 170-3 and can be used as a reference image for the inter-frame prediction unit 140-3. The restored image 180-3 can be output from the restored image storage unit 170-3.

[0184] Figure 18 This is a schematic diagram illustrating a method for global motion compensation of an execution image to which one embodiment of the present invention is applied.

[0185] It can perform global motion compensation on a large area 210-3 in image 200-3 in one go. During the image restoration process, the global motion compensation area 210-3 is determined and motion compensation is performed on the determined global motion compensation area 210-3 in one go. The global motion compensation area 210-3 can be determined based on additional information passed to the decoder.

[0186] Figure 19 This is a block diagram illustrating the sequence of a method for performing global motion compensation applicable to one embodiment of the present invention.

[0187] First, global motion compensation information 310-3 is extracted from the bitstream. Next, the extracted information is used to determine the global motion compensation region 320-3. Then, global motion compensation 330-3 is performed on the regions in the image that require global motion compensation using the determined motion compensation region information, thereby generating the restored image 350-3. In addition, decoding 340-3 is performed on data blocks for regions that do not require global motion compensation, thereby generating the restored image 350-3.

[0188] Figure 20 This is a schematic diagram illustrating a method for determining the global motion compensation region by utilizing information in the information transmitted to the decoder indicating whether the global motion compensation region is located inside or outside the determined region when performing global motion compensation according to one embodiment of the present invention.

[0189] The global motion compensation region in an image can be determined using the global motion compensation region determination information contained in the information transmitted to the decoder, and the global motion compensation region is finally determined based on information indicating whether it is located inside or outside the determined global motion compensation region. The information indicating whether it is located inside or outside the global motion compensation region can be transmitted in the form of a marker.

[0190] exist Figure 20 In the image, when the final global motion compensation region determination mark is 0, the final global motion compensation region in the image is the interior of the determined motion compensation region as shown in 410-3, 420-3. When the final global motion compensation region determination mark is 1, the final global motion compensation region in the image is the exterior of the determined motion compensation region as shown in 430-3, 440-3.

[0191] Figure 21 This is a schematic diagram illustrating various forms of global motion compensation regions when performing global motion compensation according to one embodiment of the present invention.

[0192] The global motion compensation region can use any shape as shown in 510-3, 520-3, 530-3, etc., and multiple global motion compensation regions 510-3, 520-3, 530-3 can be used within a single image.

[0193] When multiple global motion compensation regions are used within an image, the motion compensation regions can be determined using independent information transmitted for each motion compensation region, or by referring to information from other motion compensation regions.

[0194] Figure 22 This is a schematic diagram illustrating a method for determining the global motion compensation region according to the boundary of the coding unit when performing global motion compensation in one embodiment of the present invention.

[0195] The global motion compensation region can be defined according to the boundaries of the decoded data block unit as shown in 610-3 and 620-3.

[0196] Furthermore, the global motion compensation region can be composed of decoded data block units that are hierarchically segmented, as shown in the segmented data blocks in 610-3 and 620-3.

[0197] It can include global motion compensation region information in each decoded data block unit, or include information of each decoded data block in units of global motion compensation region.

[0198] Figure 23 This is a schematic diagram illustrating a method for determining the position of a global motion compensation region when performing global motion compensation according to one embodiment of the present invention.

[0199] Methods for determining the location of the global motion compensation region can include method 710-3, which uses the coordinates of the region's starting point (x, y) and ending point (x', y'), and method 720-3, which uses the coordinates of the region's starting point (x, y) and its width (Width) and height (Height).

[0200] Figure 24 This is a schematic diagram illustrating a method for determining the global motion compensation region by merging lattice-separated regions when performing global motion compensation, according to one embodiment of the present invention.

[0201] One method for determining the global motion compensation region is to divide the image into a grid of regions, then merge the separated regions with other regions to determine the global motion compensation region. Data blocks 810-3, 820-3, 830-3, 840-3, and 850-3 constitute the global motion compensation region by merging with the data blocks in the direction indicated by the arrows. Data block 850-3 references data block 840-3 and is merged; data block 840-3 references data block 830-3 and is merged; data block 830-3 references data block 820-3 and is merged; and data block 820-3 references data block 810-3 and is merged, thus forming the global motion compensation region.

[0202] Figure 25This is a schematic diagram illustrating a method for determining the global motion compensation region by repeatedly segmenting an image along a vertical or horizontal direction when performing global motion compensation, according to one embodiment of the present invention.

[0203] The global motion compensation region can be determined by repeatedly segmenting the image in the vertical or horizontal direction.

[0204] like Figure 25 The image shown can be segmented first along the horizontal boundary 910-3 and then along the vertical boundary 920-3. Region 910-2-3, segmented along the horizontal boundary 910-3, is a region that will not be further segmented, as is region 920-1-3, segmented along the vertical boundary 920-3. Next, segmentation can be performed along the horizontal boundary 950-3, the vertical boundary 970-3, and the vertical boundary 990-3. Region 950-1-3, segmented along the horizontal boundary 950-3, and region 970-1-3, segmented along the vertical boundary 970-3, are regions that will not be further segmented.

[0205] The encoder can transmit information about the segmented boundaries, while the decoder can use the transmitted boundary information to determine the global motion compensation area.

[0206] Figure 26 This is a schematic diagram illustrating a method for determining the global motion compensation region using a warping parameter in additional information passed to the decoder when performing global motion compensation, according to one embodiment of the present invention.

[0207] When an object 1010-3 in two images located at different time points is warped 1020-3, the encoder can transmit the warping parameter to the decoder, and the decoder can use the transmitted warping parameter to determine the motion compensation area.

[0208] Figure 27 This is a schematic diagram illustrating a method for rotating or scaling a global motion compensation region when performing global motion compensation, according to one embodiment of the present invention.

[0209] When an object 1110-3 in two images located at different time points is scaled 1120-3 or rotated 1130-3, the encoder can transmit scaling information or information related to the rotation area to the decoder, and the decoder can use the transmitted information to determine the motion compensation area.

[0210] Figure 28This is a schematic diagram illustrating the use of a Frame Rate Up Conversion (FRUC) method to increase the frame rate when performing global motion compensation, according to one embodiment of the present invention.

[0211] When performing the Frame Rate Conversion (FRUC) method on images with global motion compensation, it is possible to synthesize new frames between two frames not only by motion inference on a data block basis, but also by motion inference on a global motion compensation region basis.

[0212] When in Figure 28 When generating image T-1 / 2 between the previous image T-1 and the subsequent image T, it can generate data block 1220-3 synthesized using data block 1210-3 of the previous image and data block 1230-3 of the subsequent image, or generate global motion region 1250-3 synthesized using global motion region 1240-3 of the previous image and global motion region 1260-3 of the subsequent image.

[0213] Figure 29 A video decoding apparatus according to one embodiment of the present invention is illustrated, which can generate an intra-frame prediction signal from intra-frame prediction information of an encoded bitstream and output a restored image using the generated intra-frame prediction signal.

[0214] The input encoded bitstream 101-4 is decoded in the entropy decoding unit 102-4, and the residual signal is restored by the inverse quantization unit 103-4 and the inverse conversion unit 104-4. The intra-frame prediction unit 106-4 can perform intra-frame prediction using the restored signal required for intra-frame prediction generated in the prediction signal generation unit 105-4. The prediction signal generation unit 105-4 can perform a step of removing a portion of the high-frequency components by applying a low-pass filter to the restored signal required for intra-frame prediction. The motion compensation unit 107-4 can perform inter-frame prediction using the restored signal from the previous time stored in the restored image storage unit 109-4. In this way, a restored signal can be generated using the prediction signal generated by the intra-frame prediction or inter-frame prediction as described above and the residual signal. The generated restored signal is filtered by the loop filtering unit 108-4 and stored in the restored image storage unit 109-4 for reference in subsequent images. Moreover, it can be output as a restored image 110-4 in the output order of each image.

[0215] Figure 30 The reference area for predictive data blocks within a picture in which embodiments of the present invention are applied is illustrated.

[0216] To perform in-frame prediction of an in-frame prediction data block 201-4 of size M×N, a top-left reference pixel 202-4, a top reference pixel column 203-4, and a left reference pixel column 204-4 can be used. The top reference pixel column 203-4 can be greater than the horizontal length N of the in-frame prediction data block, and can be a length of n*N where n is greater than 1. The left reference pixel column 204-4 can be greater than the vertical length M of the in-frame prediction data block, and can be a length of m*N where m is greater than 1.

[0217] Figure 31 An illustration shows a method for performing directional intra-image prediction based on the length of the upper reference pixel column according to an embodiment of the present invention.

[0218] In order to perform intra-frame prediction of the current intra-frame prediction data block 301-4 with a size of M×N, when the length of the referenced upper reference pixel column 302-4 is greater than 2N, the directional intra-frame prediction of the intra-frame prediction data block 301-4 can be performed by using the directional intra-frame prediction 304-4 of the smaller angle 303-4 with a directional intra-frame prediction angle of less than 45 degrees.

[0219] Figure 32 An illustration is provided of a method for performing directional intra-image prediction based on the length of the left reference pixel column according to an embodiment of the present invention.

[0220] In order to perform intra-frame prediction for a current intra-frame prediction data block 401-4 with a size of M×N, when the length of the reference left reference pixel column 402-4 is greater than 2M, the directional intra-frame prediction of the above intra-frame prediction data block 401-4 can be performed by using the directional intra-frame prediction 404-4 of a smaller angle 403-4 with a directional intra-frame prediction angle greater than 315 degrees.

[0221] Figure 33 The range of directional prediction applicable in the in-image prediction method according to embodiments of the present invention is illustrated.

[0222] The directional prediction applicable to the proposed in-frame prediction method can be based on a range of directional prediction directions from 45 degrees to 31 degrees, including prediction ranges 1 501-4, 2 502-4, and 3 503-4. When the horizontal length of the current in-frame prediction data block is N and the length of the upper reference pixel column is greater than 2N, in-frame prediction can be performed using a directional prediction mode within the ranges of prediction ranges 1 501-4, 3 503-4, and 5 505-3. Furthermore, when the vertical length of the current in-frame prediction data block is M and the length of the left reference pixel column is greater than 2M, in-frame prediction can be performed using a directional prediction mode within the ranges of prediction ranges 1 501-4, 2 502-4, and 4 504-3. The aforementioned prediction range based on the length of the reference pixel column can be adaptively determined depending on whether the reference pixel column is decoded. Alternatively, the prediction range instruction information required to perform in-frame prediction of the current in-frame prediction data block can be transmitted through syntax elements.

[0223] Figure 34 An illustration shows a method for generating a prediction signal by changing the brightness of a pixel based on the pixel coordinates of an unrestored pixel region and the same tilt as the restored pixel region, according to the signal of an adjacent restored pixel region.

[0224] Regarding the restored pixel region 601-4 contained in the restored surrounding data block adjacent to the current predicted data block, if there exists a continuous unrestored pixel region 602-4 adjacent to the current restored pixel region, a prediction signal can be generated where the pixel value change at a position offset by a certain amount 603-4 from the starting position of the restored pixel region is the same as the pixel value change at a position offset by the same amount 604-4 from the starting position of the unrestored pixel region. Next, the signals of the restored pixel region 601-4 and the newly generated unrestored pixel region 602-4 can be used as references when performing the in-frame prediction.

[0225] Figure 35 An illustration shows a method for generating a prediction signal by changing the brightness of a pixel based on the pixel coordinates of an unrestored pixel region and the same negative tilt as the restored pixel region, according to the signal of an adjacent restored pixel region.

[0226] Regarding the restored pixel region 701-4 contained in the restored surrounding data block adjacent to the current predicted data block, if there exists a continuous unrestored pixel region 702-4 adjacent to the current restored pixel region, a prediction signal can be generated where the pixel value change at a position offset by a certain amount 703-4 from the end position of the restored pixel region is the same as the pixel value change at a position offset by the same amount 704-4 from the start position of the unrestored pixel region, but with opposite tilt signs. Next, the signals of the restored pixel region 701-4 and the newly generated unrestored pixel region 702-4 can be used as references when performing the in-frame prediction.

[0227] Figure 36 Another method for generating a prediction signal in an in-image prediction method applicable to an embodiment of the present invention is illustrated, which involves changing the brightness of a pixel based on the pixel coordinates of an unrestored pixel region from the signal of an adjacent restored pixel region with the same tilt as the restored pixel region.

[0228] Regarding the restored pixel region 801-4 contained in the restored surrounding data block adjacent to the current predicted data block, if there exists a continuous unrestored pixel region 802-4 adjacent to the current restored pixel region, a prediction signal can be generated where the pixel value change at a position offset by a certain amount 803-4 from the end position of the restored pixel region is the same as the pixel value change at a position offset by the same amount 804-4 from the start position of the unrestored pixel region. Next, the signals of the restored pixel region 801-4 and the newly generated unrestored pixel region 802-4 can be used as references when performing the in-frame prediction.

[0229] Figure 37 An embodiment of the present invention is illustrated for transmitting a method for determining whether a recommended in-frame prediction instruction signal needs to be transmitted using the sequence parameter set in the high-level syntax through an in-frame prediction method.

[0230] The proposed intra-frame prediction method can contain information on the applicability of the proposed intra-frame prediction method in the form of a 1-byte flag 902-4 of "seq_model_intra_enabled_flag" within the sequence parameter set 901 of the NAL (Network Abstraction Layer) unit in the compressed bitstream. When the value of the flag is true, the image referenced to the sequence parameter set can be decoded using the proposed intra-frame prediction method.

[0231] Figure 38A method for transmitting a command signal indicating whether or not the recommended intra-frame prediction should be performed using the image parameter set in the high-level syntax, according to an embodiment of the present invention, is illustrated.

[0232] The proposed intra-frame prediction method can include the applicability information of the aforementioned TopSky intra-frame prediction method within the image parameter set 1001-4 of the NAL (Network Abstraction Layer) unit existing in the compressed bitstream in the form of a 1-byte flag 1002-4 of "pic_model_intra_enabled_flag". When the value of the flag is true, decoding can be performed using the proposed intra-frame prediction method by referring to the stripes of the aforementioned image parameter set. Furthermore, when the value of "pic_model_intra_enabled_flag" transmitted in the aforementioned image parameter set is true, the applicability information of the proposed intra-frame prediction method in all sizes of intra-frame prediction data blocks allowed within the image can be included in the form of a 1-byte flag 1003-4 of "pic_model_intra_all_blk_sizes_flag". When the above pic_model_intra_enabled_flag is true and the above pic_model_intra_all_blk_sizes_flag is false, the minimum and maximum sizes of the data blocks in the in-picture prediction data blocks contained in the current image that are applicable to the above-proposed in-picture prediction method can be transmitted in the form of exponential Golomb coding with a base of 2, namely min_log2_model_intra_blk_size 1004-4 and max_log2_model_intra_blk_size 1005-4.

[0233] Figure 39 An embodiment of the present invention is illustrated for transmitting a method for determining whether the recommended intra-frame prediction instruction signal needs to be executed using the stripe fragment header in the high-level syntax through an intra-frame prediction method.

[0234] The proposed intra-frame prediction method can contain information on the applicability of the proposed intra-frame prediction method in the form of a 1-byte flag 1102-4 of slice_model_intra_enabled_flag inside the slice segment header 1101-4 of the NAL (Network Abstraction Layer) unit in the compressed bitstream. When the value of the above flag is true, the data block referring to the above slice segment header can be decoded using the proposed intra-frame prediction method.

[0235] Figure 40 A decoding apparatus including an in-picture prediction unit, according to one embodiment of the present invention, has been illustrated.

[0236] The decoding apparatus including an in-frame prediction unit may include at least one of an entropy decoding unit 110-5, an inverse quantization unit 120-5, an inverse conversion unit 130-5, an in-frame prediction unit 140-5, an inter-frame prediction unit 150-5, an in-loop filtering unit 160-5, and a restored image storage unit 170-5.

[0237] The entropy decoding unit 110-5 decodes the input bitstream 100-5 and outputs decoding information such as syntax elements and quantization coefficients. The output information can include information required to perform global motion compensation.

[0238] The inverse quantization unit 120-5 and the inverse conversion unit 130-5 receive quantization coefficients and sequentially perform inverse quantization and inverse conversion to output a residual signal.

[0239] The in-frame prediction unit 140-5 generates a prediction signal by performing spatial prediction using the pixel values ​​of previously decoded neighboring data blocks adjacent to the currently being decoded data block. To generate the prediction signal, neighboring pixel values ​​in the curve direction can be used.

[0240] The inter-frame prediction unit 150-5 generates a prediction signal by performing motion compensation using motion vectors extracted from the bitstream and the restored image stored in the restored image storage unit 170-5.

[0241] The prediction signals output from the in-frame prediction unit 140-5 and the inter-frame prediction unit 150-5 will be summed with the residual signals, and the restored image generated by the summation will be transmitted to the in-loop filtering unit 160-5.

[0242] The restored image after filtering in the in-loop filtering unit 160-5 is saved in the restored image storage unit 170-5 and can be used as a reference image for the inter-frame prediction unit 150-5.

[0243] Figure 41 The surrounding reference area during in-screen prediction is illustrated in accordance with one embodiment of the present invention.

[0244] In order to generate prediction samples for prediction data blocks 210-5 within the current frame with a size of M×N, the upper reference sample 220-5, the left reference sample 230-5, and the upper left reference sample 240-5 can be used.

[0245] The length of column 220-5 of the upper reference sample can be greater than the horizontal length M of the predicted data block 210-5 in the current frame. In addition, the length of column 230-5 of the left reference sample can be greater than the vertical length N of the predicted data block 210-5 in the current frame.

[0246] Figure 42 A method for referencing samples of surrounding data blocks when performing in-screen prediction, according to one embodiment of the present invention, is illustrated.

[0247] When generating prediction samples for prediction data blocks 310-5 within the current frame with a size of M×N, a prediction signal can be generated using reference samples in the direction of curve 320-5, which is expressed as an Nth power and can tend towards a straight line based on the coefficients.

[0248] The aforementioned curve information can be contained in a bitstream for transmission, and this information includes the degree or coefficients of the curve equation.

[0249] Figure 43 A method for referencing multiple samples of surrounding data blocks when performing in-screen prediction, according to one embodiment of the present invention, is illustrated.

[0250] When generating a prediction sample 420-5 of a prediction data block 410-5 with a size of M×N within the current frame, it can not only refer to the pixels of a surrounding data block in the curve direction 430-5, but also refer to two or more.

[0251] When using more than two reference samples, the weighted average of reference samples 440-5, 450, and 5 can be used to generate the predicted sample 420-5.

[0252] Figure 44 A method for generating non-existent reference samples of surrounding data blocks during in-screen prediction, according to one embodiment of the present invention, is illustrated.

[0253] When there are some reference samples around the predicted data block 510-5 in the current frame with a size of M×N, non-existent reference samples can be generated using the available reference samples 540-5 and 550-5.

[0254] When generating non-existent reference samples 520-5 and 530-5, the length of the reference samples to be generated may vary depending on the curve 560 applicable in the current prediction data block.

[0255] Figure 45 An illustration is provided of a method for performing predictions in different regions of a prediction data block using reference samples from different directions, according to one embodiment of the present invention.

[0256] When a current frame prediction data block of size M×N is divided into two regions A and B by curve 610-5, regions A and B can generate prediction samples using reference samples in the interconnected direction.

[0257] In region A or region B mentioned above, there can be more than one reference sample used to generate the prediction sample, and the more than one reference sample can be located in different directions.

[0258] Figure 46 Another method for performing predictions in different regions of a prediction data block using reference samples from different directions, according to one embodiment of the present invention, is illustrated.

[0259] When a predicted data block of size M×N within the current frame is divided into two regions A, B, C, and D by curve 710-5 and straight line 720-5 connecting the corners of the data block, regions A, B, C, and D can generate predicted samples using reference samples in the interconnected directions.

[0260] In regions A, B, C, and D, more than one reference sample can be used to generate the prediction sample, and these multiple reference samples can be located in different directions.

[0261] Figure 47 A method for filtering the leftmost prediction sample column of a prediction data block to eliminate discontinuities with surrounding data blocks during in-screen prediction, according to one embodiment of the present invention, is illustrated.

[0262] When the region used for vertical prediction is located to the left of the prediction data block in the current frame, which is divided by curves or straight lines, and the prediction sample in the leftmost column of the prediction data block can be filtered.

[0263] When filtering samples in the leftmost column of the prediction data block, the change in the left reference sample can be utilized.

[0264] For example, when the predicted data block is such as Figure 47When the curve 810-5 shown is divided into two regions, A and B, and region A is predicted using the vertical direction, the predicted sample of the leftmost column 820-5 can be filtered using the change amount 830-5 of the reference sample on the left.

[0265] Figure 48 A method for filtering the topmost prediction sample column of a prediction data block to eliminate discontinuities with surrounding data blocks during in-screen prediction, according to one embodiment of the present invention, is illustrated.

[0266] When the region used for performing horizontal prediction in the current frame, which is divided by curves or straight lines, is located at the top of the prediction block, filtering can be performed on the prediction samples in the topmost column of the prediction block.

[0267] When filtering the samples in the top column of the prediction data block, the change in the upper reference sample can be utilized.

[0268] For example, when the predicted data block is such as Figure 48 When the curve 910-5 shown is divided into two regions, A and B, and region A is predicted using the horizontal direction, the predicted sample of the top column 920-5 can be filtered using the change amount 930-5 of the upper reference sample.

[0269] Figure 49 The order of in-screen predictions applicable to one embodiment of the present invention is illustrated.

[0270] First, the prediction information within the curve frame is extracted from the bitstream 1010-5. This extracted information can include the degree or coefficients of the curve equation representing the curve. Next, the extracted curve-related information and the presence or absence of reference pixels in surrounding data blocks are used to determine whether reference sample filling is needed 1020-5. When reference sample filling is needed, a non-existent reference sample is generated using available reference samples from surrounding data blocks 1030-5. When reference sample filling is not needed, a prediction sample for the current frame's prediction data block is generated using the reference sample 1040-5. When generating the prediction sample, the previously extracted curve-related information is used to determine the reference sample. After generating the prediction data block, it is determined whether filtering of the prediction sample is needed 1050-5. When filtering of the prediction sample is needed because the left region of the prediction data block uses vertical prediction or the upper region of the prediction data block uses horizontal prediction, prediction sample filtering is performed on the surrounding reference samples using the change amount 1060-5. Filtering of the predicted samples can be performed on the leftmost sample column and the top left sample row of the predicted data block.

[0271] Industry availability

[0272] This invention is applicable to encoding / decoding video signals.

Claims

1. An in-frame prediction method for video decoding, comprising: For the current M×N block, determine the in-frame prediction mode that belongs to the initial directional prediction range. Based on the horizontal length N and vertical length M of the current M×N block, it is determined whether to correct the in-screen prediction mode from the initial directional prediction range to the corrected directional prediction range. Determine the corrected in-frame prediction mode that falls within the corrected directional prediction range; as well as Based on the modified in-frame prediction mode, perform in-frame prediction for the current M×N block. The initial directional prediction range includes a first prediction range, a second prediction range, and a third prediction range. The second prediction range is the range of horizontal prediction angles extending from the first prediction range, and the third prediction range is the range of vertical prediction angles extending from the first prediction range. The second and third prediction ranges are adjacent to the first prediction range. Specifically, based on the longitudinal length M of the current M×N block, the corrected directional prediction range is determined as a first corrected directional prediction range, and based on the lateral length N of the current M×N block, the corrected directional prediction range is determined as a second corrected directional prediction range. The first corrected directional prediction range includes the first prediction range, the second prediction range, and the fourth prediction range. The second corrected directional prediction range includes the first prediction range, the third prediction range, and the fifth prediction range. The fourth prediction range is the range of horizontal prediction angles extending from the second prediction range, and is adjacent to the second prediction range. The fifth prediction range is the range of vertical prediction angles extending from the third prediction range, and is adjacent to the third prediction range. Specifically, intra-frame prediction of the M×N current block is performed using a reference region adjacent to the current M×N block, and The reference area includes the upper reference pixel column and the left reference pixel column of the current block.

2. The method according to claim 1, wherein, The initial directional prediction range for in-frame prediction of the M×N current block is determined based on angles ranging from 45 degrees to 315 degrees.

3. An in-frame prediction method for video coding, comprising: For the current M×N block, determine the in-frame prediction mode that belongs to the initial directional prediction range. Based on the horizontal length N and vertical length M of the current M×N block, it is determined whether to correct the in-screen prediction mode from the initial directional prediction range to the corrected directional prediction range. Determine the corrected in-frame prediction mode that falls within the corrected directional prediction range; as well as Based on the modified in-frame prediction mode, perform in-frame prediction for the current M×N block. The initial directional prediction range includes a first prediction range, a second prediction range, and a third prediction range. The second prediction range is the range of horizontal prediction angles extending from the first prediction range, and the third prediction range is the range of vertical prediction angles extending from the first prediction range. The second and third prediction ranges are adjacent to the first prediction range. Specifically, based on the longitudinal length M of the current M×N block, the corrected directional prediction range is determined as a first corrected directional prediction range, and based on the lateral length N of the current M×N block, the corrected directional prediction range is determined as a second corrected directional prediction range. The first corrected directional prediction range includes the first prediction range, the second prediction range, and the fourth prediction range. The second corrected directional prediction range includes the first prediction range, the third prediction range, and the fifth prediction range. The fourth prediction range is the range of horizontal prediction angles extending from the second prediction range, and is adjacent to the second prediction range. The fifth prediction range is the range of vertical prediction angles extending from the third prediction range, and is adjacent to the third prediction range. Specifically, intra-frame prediction of the M×N current block is performed using a reference region adjacent to the current M×N block, and The reference area includes the upper reference pixel column and the left reference pixel column of the current block.

4. The method according to claim 3, wherein, The initial directional prediction range for in-frame prediction of the M×N current block is determined based on angles ranging from 45 degrees to 315 degrees.

5. A method for transmitting a bit stream, wherein, The encoding method includes: generating the bit stream by performing an encoding method and sending the bit stream. For the current M×N block, determine the in-frame prediction mode that belongs to the initial directional prediction range. Based on the horizontal length N and vertical length M of the current M×N block, it is determined whether to correct the in-screen prediction mode from the initial directional prediction range to the corrected directional prediction range. Determine the corrected in-frame prediction mode that falls within the corrected directional prediction range; and Based on the modified in-frame prediction mode, perform in-frame prediction for the current M×N block. The initial directional prediction range includes a first prediction range, a second prediction range, and a third prediction range. The second prediction range is the range of horizontal prediction angles extending from the first prediction range, and the third prediction range is the range of vertical prediction angles extending from the first prediction range. The second and third prediction ranges are adjacent to the first prediction range. Specifically, based on the longitudinal length M of the current M×N block, the corrected directional prediction range is determined as a first corrected directional prediction range, and based on the lateral length N of the current M×N block, the corrected directional prediction range is determined as a second corrected directional prediction range. The first corrected directional prediction range includes the first prediction range, the second prediction range, and the fourth prediction range. The second corrected directional prediction range includes the first prediction range, the third prediction range, and the fifth prediction range. The fourth prediction range is the range of horizontal prediction angles extending from the second prediction range, and is adjacent to the second prediction range. The fifth prediction range is the range of vertical prediction angles extending from the third prediction range, and is adjacent to the third prediction range. Specifically, intra-frame prediction of the M×N current block is performed using a reference region adjacent to the current M×N block, and The reference area includes the upper reference pixel column and the left reference pixel column of the current block.

Images