WO2019146811A1 - Décodeur vidéo et son procédé de commande - Google Patents

Décodeur vidéo et son procédé de commande Download PDF

Info

Publication number
WO2019146811A1
WO2019146811A1 PCT/KR2018/001112 KR2018001112W WO2019146811A1 WO 2019146811 A1 WO2019146811 A1 WO 2019146811A1 KR 2018001112 W KR2018001112 W KR 2018001112W WO 2019146811 A1 WO2019146811 A1 WO 2019146811A1
Authority
WO
WIPO (PCT)
Prior art keywords
block
signal
size
transform
subblock
Prior art date
Application number
PCT/KR2018/001112
Other languages
English (en)
Inventor
Kyungyong Kim
Donggyu Sim
Hongsuk JEONG
Seungchul JANG
Seanae Park
Juntaek PARK
Original Assignee
Lg Electronics Inc.
Kwangwoon University Industry-Academic Collaboration Foundation
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by Lg Electronics Inc., Kwangwoon University Industry-Academic Collaboration Foundation filed Critical Lg Electronics Inc.
Priority to EP18901916.9A priority Critical patent/EP3744093A4/fr
Priority to PCT/KR2018/001112 priority patent/WO2019146811A1/fr
Publication of WO2019146811A1 publication Critical patent/WO2019146811A1/fr

Links

Images

Classifications

    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04N—PICTORIAL COMMUNICATION, e.g. TELEVISION
    • H04N19/00—Methods or arrangements for coding, decoding, compressing or decompressing digital video signals
    • H04N19/10—Methods or arrangements for coding, decoding, compressing or decompressing digital video signals using adaptive coding
    • H04N19/169—Methods or arrangements for coding, decoding, compressing or decompressing digital video signals using adaptive coding characterised by the coding unit, i.e. the structural portion or semantic portion of the video signal being the object or the subject of the adaptive coding
    • H04N19/18—Methods or arrangements for coding, decoding, compressing or decompressing digital video signals using adaptive coding characterised by the coding unit, i.e. the structural portion or semantic portion of the video signal being the object or the subject of the adaptive coding the unit being a set of transform coefficients
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04N—PICTORIAL COMMUNICATION, e.g. TELEVISION
    • H04N19/00—Methods or arrangements for coding, decoding, compressing or decompressing digital video signals
    • H04N19/10—Methods or arrangements for coding, decoding, compressing or decompressing digital video signals using adaptive coding
    • H04N19/102—Methods or arrangements for coding, decoding, compressing or decompressing digital video signals using adaptive coding characterised by the element, parameter or selection affected or controlled by the adaptive coding
    • H04N19/103—Selection of coding mode or of prediction mode
    • H04N19/105—Selection of the reference unit for prediction within a chosen coding or prediction mode, e.g. adaptive choice of position and number of pixels used for prediction
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04N—PICTORIAL COMMUNICATION, e.g. TELEVISION
    • H04N19/00—Methods or arrangements for coding, decoding, compressing or decompressing digital video signals
    • H04N19/10—Methods or arrangements for coding, decoding, compressing or decompressing digital video signals using adaptive coding
    • H04N19/102—Methods or arrangements for coding, decoding, compressing or decompressing digital video signals using adaptive coding characterised by the element, parameter or selection affected or controlled by the adaptive coding
    • H04N19/132—Sampling, masking or truncation of coding units, e.g. adaptive resampling, frame skipping, frame interpolation or high-frequency transform coefficient masking
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04N—PICTORIAL COMMUNICATION, e.g. TELEVISION
    • H04N19/00—Methods or arrangements for coding, decoding, compressing or decompressing digital video signals
    • H04N19/10—Methods or arrangements for coding, decoding, compressing or decompressing digital video signals using adaptive coding
    • H04N19/169—Methods or arrangements for coding, decoding, compressing or decompressing digital video signals using adaptive coding characterised by the coding unit, i.e. the structural portion or semantic portion of the video signal being the object or the subject of the adaptive coding
    • H04N19/17—Methods or arrangements for coding, decoding, compressing or decompressing digital video signals using adaptive coding characterised by the coding unit, i.e. the structural portion or semantic portion of the video signal being the object or the subject of the adaptive coding the unit being an image region, e.g. an object
    • H04N19/176—Methods or arrangements for coding, decoding, compressing or decompressing digital video signals using adaptive coding characterised by the coding unit, i.e. the structural portion or semantic portion of the video signal being the object or the subject of the adaptive coding the unit being an image region, e.g. an object the region being a block, e.g. a macroblock
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04N—PICTORIAL COMMUNICATION, e.g. TELEVISION
    • H04N19/00—Methods or arrangements for coding, decoding, compressing or decompressing digital video signals
    • H04N19/44—Decoders specially adapted therefor, e.g. video decoders which are asymmetric with respect to the encoder
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04N—PICTORIAL COMMUNICATION, e.g. TELEVISION
    • H04N19/00—Methods or arrangements for coding, decoding, compressing or decompressing digital video signals
    • H04N19/50—Methods or arrangements for coding, decoding, compressing or decompressing digital video signals using predictive coding
    • H04N19/59—Methods or arrangements for coding, decoding, compressing or decompressing digital video signals using predictive coding involving spatial sub-sampling or interpolation, e.g. alteration of picture size or resolution

Definitions

  • the present invention relates to a video decoder, and more particularly, to a video decoder and controlling method thereof.
  • the present invention is suitable for a wide scope of applications, it is particularly suitable for reducing an operation quantity by selecting prescribed subblocks from a transform block only and then decoding the selected subblocks only.
  • HEVC high efficiency video coding
  • the present invention is directed to a video decoder and controlling method thereof that substantially obviate one or more of the problems due to limitations and disadvantages of the related art.
  • An object of the present invention is to provide a video decoder and controlling method thereof, which can reduce an operation quantity by decoding a specific partial block only instead of a whole block in performing video decoding.
  • Another object of the present invention is to provide a video decoder and controlling method thereof, which can reduce an operation quantity, if the number of coefficients within a subblock is equal to or greater than a preset reference value in performing video decoding, by decoding the corresponding subblock only.
  • Further object of the present invention is to provide a video decoder and controlling method thereof, which can reduce an operation quantity, if a location value of a quantization coefficient in a current subblock is equal to or smaller than a half of a current transform block size, by performing dequantization on the current subblock.
  • Another further object of the present invention is to provide a video decoder and controlling method thereof, which can reduce an operation quantity, if a current transform block size is equal to or greater than a first block size, by performing inverse transform based on a corresponding subblock to be decoded in a current block and then performing linear interpolation on the inverse-transformed corresponding subblock.
  • a video decoder includes a reconstruction signal selecting unit selecting a signal to be reconstructed for a bitstream, an entropy decoding unit obtaining a quantization coefficient of at least one block unit by entropy-decoding the selected signal to be reconstructed, a dequantization unit obtaining a transform coefficient through dequantization performed on the obtained quantization coefficient of the at least one block unit, an inverse transform unit obtaining a residual signal through inverse transform using a specific transform base suitable for a block size of the obtained transform coefficient, an intra picture prediction unit obtaining a predicted signal by referring to reference samples for a current block to be decoded, a residual signal compensating unit scaling a block of the obtained residual signal based on a block size of the predicted signal, and an adding-up unit generating a reconstructed signal by adding the scaled residual signal and the predicted signal together.
  • a method of decoding a video in a device includes selecting a signal to be reconstructed for a bitstream, obtaining a quantization coefficient of at least one block unit by entropy-decoding the selected signal to be reconstructed, if a preset condition is met, dequantizing specific partial blocks, and outputting a decoded video based on a result from dequantizing the partial block, wherein the preset condition is determined according to at least one selected from the group consisting of a chroma signal, a size of a transform block, and a location value of a coefficient.
  • an operation quantity can be reduced and a video decoding execution speed can be improved, whereby user convenience is enhanced.
  • Decoding used in the present specification may be performed in order reverse to that of an encoding process.
  • an operation quantity can be reduced by decoding the corresponding subblock only and a video decoding execution speed can be improved, whereby user convenience can be enhanced.
  • a location value of a quantization coefficient in a current subblock is equal to or smaller than a half of a current transform block size
  • an operation quantity can be reduced by performing dequantization on the current subblock and a video decoding execution speed can be improved, whereby user convenience can be enhanced.
  • an operation quantity can be reduced by performing inverse transform based on a corresponding subblock to be decoded in a current block and then performing linear interpolation on the inverse-transformed corresponding subblock and a video decoding execution speed can be improved, whereby user convenience can be enhanced.
  • FIG. 1 is a schematic diagram showing an overall configuration of a device according to one embodiment of the present invention
  • FIG. 2 is a diagram showing details of prescribed components shown in FIG. 1 according to one embodiment of the present invention.
  • FIG. 3 is a flowchart for a method of controlling a video decoder according to one embodiment of the present invention
  • FIG. 4 is a diagram showing an embodiment of a method for selecting a signal to be reconstructed in a reconstruction signal selecting unit 210 shown in FIG. 2;
  • FIG. 5 is an emphasized diagram showing a partial block of a signal processed by an entropy decoding unit 220, a dequantization unit 230 and an inverse transform unit 240 according to one embodiment of the present invention
  • FIG. 6 is a diagram to describe a transform base processed by an inverse transform unit 240 according to one embodiment of the present invention.
  • FIG. 7 is a diagram to describe a scaling process of a residual signal compensating unit 260 according to one embodiment of the present invention.
  • FIG. 8 is a flowchart for selectively performing dequantization according to one embodiment of the present invention.
  • FIG. 9 is a diagram showing an example of 8 x 8 subblock according to one embodiment of the present invention.
  • FIG. 10 is a diagram showing an example of 4 x 4 transform block in HEVC standard according to one embodiment of the present invention.
  • FIG. 11 is a detailed flowchart of a process of an inverse transform unit 240 according to one embodiment of the present invention.
  • Terminologies including ordinal numbers such as first, second and the like may be used to describe various components, by which the components may be non-limited. And, the terminologies are used for the purpose of discriminating one component from other components only.
  • the former component may be connected to accesses the latter component in direct. Yet, it is understood that a different component may be present in-between. On the other hand, if one component is mentioned as ‘directly connected to’ or ‘directly accessing’ another component, it is understood that a different component may is not present in-between.
  • Singular expression may include plural expressions unless having a clear meaning in the context.
  • FIG. 1 is a schematic diagram showing an overall configuration of a device according to one embodiment of the present invention.
  • a device 1000 shown in FIG. 1 may include any device capable of performing video decoding and refer to a video thumbnail extractor in pursuit of study if focused on functions in the present specification.
  • the device 1000 may include a thumbnail selecting unit 100, a decoding unit 200, a downsampling unit 300 and a filtering unit 400.
  • the thumbnail selecting unit 100 selects an image to be outputted as a thumbnail in a whole video from an input bitstream 10.
  • the decoding unit 200 decodes the image selected by the thumbnail selecting unit 100.
  • the downsampling unit 300 reduces a size of the decoded image into a size of a thumbnail to be used.
  • the filtering unit 400 filters the reduced image for image quality enhancement and outputs the filtered image as a thumbnail 20.
  • FIG. 2 is a diagram showing details of prescribed components shown in FIG. 1 according to one embodiment of the present invention. Particularly, although the functions performed by the decoding unit 200 shown in FIG. 1 are illustrated as the respective modules in FIG. 2, merging to design prescribed modules into a single module pertains to the scope of a right of the present invention.
  • the decoding unit 200 includes a reconstruction signal selecting unit 210, an entropy decoding unit 220, a dequantization unit 230, an inverse transform unit 240, an intra picture prediction unit 250, a residual signal compensating unit 260, an adding-up unit 270 and the like.
  • the reconstruction signal selecting unit 210 determines a signal to be reconstructed through a size ratio of an image size of an inputted bitstream 10 to a size of a thumbnail to be generated, a signal to be reconstructed, an amount of information of the signal to be reconstructed, and a block size of the signal to be reconstructed.
  • the entropy decoding unit 220 outputs at least one of a syntax element and a quantized coefficient to be reconstructed by decoding a signal to be determined as the signal to be reconstructed in an inputted bitstream 10.
  • the outputted information may be named decoding information.
  • the entropy decoding unit 220 is designed to vary a block size of the quantization coefficient obtained according to a transform block size of the selected signal to be reconstructed.
  • a transform block size of a signal to be reconstructed is 16 x 16 block
  • a block size of the quantization coefficient to be obtained may become 8 x 8 block.
  • a transform block size of a signal to be reconstructed is 32 x 32 block
  • a block size of the quantization coefficient to be obtained may become 16 x 16 block.
  • the dequantization unit 230 receives the partially quantized coefficient to be reconstructed from the entropy decoding unit 220, performs dequantization, and outputs a transform coefficient.
  • the inverse transform unit 240 outputs the residual signal as a result from receiving the partially transform coefficient to be reconstructed and then performing inverse transform using a portion of a transform base only.
  • the intra picture prediction unit 250 generates a predicted signal by performing spatial prediction based on a pixel value of a previously decoded neighbor block adjacent to a current block to be decoded, i.e., a reference sample.
  • a reference sample means a previously encoded or decoded sample within a current frame.
  • the residual signal compensating unit 260 scales the block size of the residual signal based on the block size of the predicted signal. Namely, the residual signal compensating unit 260 scales the block size of the residual signal so that the block size of the residual signal and the block size of the predicted signal are made to become equal to each other.
  • the adding-up unit 270 generates a reconstructed signal by a block unit in a manner of adding the predicted signal and the scaled residual signal together.
  • the reconstructed signal contains a reconstructed image.
  • a block size of a predicted signal is 16 x 16 block and a block size of a residual signal is 8 x 8 block
  • a block of the residual signal is scaled into 16 x 16 block based on the block size of the predicted signal and the adding-up unit 270 generates a reconstructed signal by 16 x 16 block unit in a manner of adding the predicted signal and the scaled residual signal together.
  • FIG. 1 the elements described in Figures 1 and 2 are included in a video processor, a CPU (central processing unit), graphics processor or any controller.
  • FIG. 3 is a flowchart for a method of controlling a video decoder according to one embodiment of the present invention.
  • the reconstruction signal selecting unit 210 selects a signal to be reconstructed for a bitstream [S310].
  • the entropy decoding unit 220 obtains a quantization coefficient of a block unit by entropy-decoding the selected signal to be reconstructed [S320].
  • the dequantization unit 230 obtains a transform coefficient by performing dequantization on the obtained quantization coefficients of the block unit [S330].
  • the inverse transform unit 240 obtains a residual signal through an inverse transform process using a specific transform base suitable for a block size of the obtained transform coefficient [S340].
  • the intra picture prediction unit 250 obtains a predicted signal by referring to reference samples for a current block to be decoded [S350].
  • the residual signal compensating unit 260 scales a block of the obtained residual signal to become equal to a block size of the predicted signal based on the block size of the predicted signal [S360].
  • the adding-up unit 270 generates a reconstructed signal by block unit in a manner of adding the scaled residual signal and the predicted signal together [S370].
  • the technical feature of one embodiment of the present invention includes a method of reducing or reinforcing a decoding step selectively within a minimum error range.
  • prescribed subblocks among the 64 subblocks can be selectively decoded according to priority only.
  • 16 subblocks close to a DC value among the 64 subblocks can be decoded only.
  • Dequantization and inverse transform may be performed on prescribed subblocks in two ways as follows.
  • the random value may include 0. Yet, the random value may be limited to other numerical values, which pertains to the scope of the right of the present invention.
  • an output image decoded in the inverse transform process can become a reconstructed block in 32 x 32 size after experiencing inverse transform by 32 x 32 unit that is a size the preset transform block.
  • 16 prescribed subblocks in 32 x 32 transform block can be dequantized and inverse-transformed.
  • a decoded output image can become a reconstructed block in 16 x 16 size configured with the 16 prescribed subblocks. Therefore, since it is not necessary to maintain a memory for the whole 32 x 32 block, it is efficient in aspects of memory and calculation amount.
  • FIG. 4 is a diagram showing an embodiment of a method for selecting a signal to be reconstructed in the reconstruction signal selecting unit 210 shown in FIG. 2.
  • an image size 400 of an inputted bitstream is 1920 x 1080, and a size 410 of a thumbnail to be created is 480 x 270.
  • a ratio of the two images is 16:1, and a relative ratio of a block size of a reconstructed signal to a block size of an input signal can be determined as 1:4 for the thumbnail creation.
  • a relative ratio of a block size of a reconstructed signal to a block size of an input signal can be determined as 1:4, a 4 x 4 quantization coefficient block 430 including DC frequency information and low frequency information in an inputted 8 x 8 quantization coefficient block 420 is decoded and reconstructed. Furthermore, the DC frequency information and the low frequency information are assumed as containing important substance of image information required for a video decoding process for example.
  • FIG. 5 is an emphasized diagram showing a partial block of a signal processed by the entropy decoding unit 220, the dequantization unit 230 and the inverse transform unit 240 according to one embodiment of the present invention.
  • a prescribed block of a signal used by the entropy decoding unit 220, the dequantization unit 230 and the inverse transform unit 240 is a block 510 including DC frequency information and low frequency information in N x M size corresponding to a portion of a transform coefficient block 500.
  • N and M are 4 and 4, respectively, if the transform coefficient block 500 is 8 x 8 block, a prescribed block of a signal may become 4 x 4 block.
  • FIG. 6 is a diagram to describe a transform base processed by the inverse transform unit 240 according to one embodiment of the present invention.
  • a transform base used by the inverse transform unit 240 is a transform base required for reconstructing a portion of a signal used by the entropy decoding unit 220, the dequantization unit 230 and the inverse transform unit 240, and is a block 410 including a DC frequency base and a low frequency base as a K x L transform base block 610, which corresponds to a partial block of a transform base block 600 required for reconstructing all transform coefficient signals.
  • the transform base partial block 610 may become 4 x 4 block.
  • FIG. 7 is a diagram to describe a scaling process of a residual signal compensating unit 260 according to one embodiment of the present invention.
  • the residual signal compensating unit 260 outputs a scaled residual signal 720 by scaling the received residual signal 700 by linear interpolation.
  • the residual signal compensating unit 260 scales the block size of the residual signal 700 to twice in width and twice in length by linear interpolation.
  • the block size of the residual signal 720 is scaled to be equal to that of the predicted signal 710 and then outputted.
  • FIG. 8 is a flowchart for selectively performing dequantization according to one embodiment of the present invention.
  • a method of selectively performing dequantization may be performed based on various embodiments and conditions as follows.
  • the method can selectively apply for a random block size.
  • the method applies to 32 x 32 block size only or is applicable to sizes smaller or greater than the 32 x 32 block size.
  • the method is applicable to at least one of a luminance signal and a chroma signal Cb and Cr. According to further embodiment of the present invention, the method is applicable to at least one of red (R), green (G) and blue (B) signals.
  • a method newly proposed by the present invention may be selectively applicable deepening on depth of a coding block (CB).
  • source code in ffmpeg (https://www.ffmpeg.org/), which is media framework open source, can be implemented by being modified as follows.
  • a process for the entropy decoding unit 220 to select a block to be decoded from random block unit quantization coefficients can be implemented by modifying a ‘ff_hevc_hls_residual_coding’ function within “libavcodec/hevc_cabac.c” source as follows.
  • the specific conditions correspond to a case 1) of a chroma signal, a case 2) that a transform block size is equal to or smaller than 8 x 8, and a case 3) that a location value of a current coefficient is equal to or smaller than a half of a current transform block size.
  • a current subblock contains high priority information of a whole block and that a sufficiently identifiable image can be reconstructed by dequantizing the current subblock.
  • a chroma signal means a signal having chroma information only without having information on brightness and also means a signal excluding luminance signal (Y) information from each color signal (R, G, B).
  • a luminance signal means a signal that represents video image brightness as voltage waveform.
  • a chroma signal Compared to a luminance signal, a chroma signal has a relatively small information size. Although the present invention applies to a chroma signal, an effect of reducing an operation quantity is insignificant. Hence, dequantization is applied to a chroma signal like the existing method.
  • the second condition i.e., transform block size
  • the second condition is described as follows. First of all, if a size of a transform block is equal to or smaller than 8 x 8, since high priority information is contained, dequantization is applied like the existing method. On the other hand, if a size of a transform block is greater than 8 x 8, prescribed subblocks are dequantized through the third condition (i.e., coefficient value) only.
  • the third condition i.e., coefficient value
  • the third condition shall be described in detail with reference to FIG. 9 later.
  • the dequantization unit 230 performs dequantization on the current subblock [S820].
  • the dequantization unit 230 substitutes 0 for a dequantization coefficient of the current subblock [S830].
  • the routine goes to the step S810 of checking whether the specific condition is met.
  • FIG. 9 is a diagram showing an example of 8 x 8 subblock according to one embodiment of the present invention.
  • the present invention has the technical effect on a method of decoding a prescribed subblock only. And, a method of selecting a subblock to decode is described as follows.
  • 8 x 8 block includes 4 4 x 4 subblocks.
  • the 8 x 8 block 900 includes a first subblock 910, a second subblock 920, a third subblock 930 and a fourth subblock 940.
  • the number of coefficients within each subblock is equal to or greater than or smaller than a preset reference value, it is able to decode the corresponding subblock.
  • Subblocks failing to meet the corresponding condition may be substituted with a random value without being decoded.
  • the random value may include 0.
  • the number of coefficients of the first subblock 910 is 4, the number of coefficients of the second subblock 920 is 1, the number of coefficients of the third subblock 930 is 0, and the number of coefficients of the fourth subblock 940 is 2. If a preset reference value is 3, the first subblock 910 meets the corresponding condition only, whereas the second to fourth subblocks 920, 930 and 940 fail to meet the corresponding condition.
  • the entropy decoding unit 220 decodes the first subblock 910 only and substitutes the rest of the subblocks, i.e., the second to fourth subblocks 920, 930 and 940 with 0 without decoding the second to fourth subblocks 920, 930 and 940.
  • the number of coefficients within each subblock can be inferred.
  • the current transform block 900 is 8 x 8 block that includes the first to fourth subblocks 910, 920, 930 and 940.
  • a location value 950 of a current quantization coefficient is (2, 3) in x-y coordinates
  • the location value 950 of the current quantization coefficient is included in 4 x 4 block corresponding to a half of 8 x 8 block corresponding to a current transform block size.
  • ‘1’ means that a coefficient exists.
  • a current subblock becomes the first subblock 910.
  • the dequantization unit 230 selects the first subblock 910 only, performs dequantization on the first subblock 910, and substitutes the rest of the subblocks, i.e., the second to fourth subblocks 920, 930 and 940 with 0 instead of performing dequantization thereon.
  • FIG. 10 is a diagram showing an example of 4 x 4 transform block in HEVC standard according to one embodiment of the present invention.
  • a 4x4 transform block 1010 includes coefficients of 9, -1, -5, 3, and 1.
  • the number of coefficients is 5.
  • a significant_coeff_flag value 1020 is checked, since the number of 1 is 5, it can be observed that the number of coefficients is 5.
  • ‘1’ indicates that a coefficient exists. If a coefficient exists, a significant_coeff_flag value becomes 1. If a coefficient does not exist, a significant_coeff_flag value becomes 0.
  • the entropy decoding unit 220 can decode a corresponding subblock.
  • Subblocks failing to meet the corresponding condition can be substituted with a random value instead of being decoded.
  • the random value may include 0.
  • 4 x 4 transform block 1010 includes coefficients of 9, -1, -5, 3, and 1. If a reference value is 2, regarding 9, -5, and 3 among the coefficients, an absolute value of a corresponding coefficient becomes equal to or greater than the reference value. And, the entropy decoding unit 220 can decode the 4 x 4 transform block 1010.
  • coeff_abs_level_greater1_flag the number of coefficients greater than 1 in coeff_abs_level_greater1_flag is 3.
  • a single coeff_abs_level_greater2_flag exists per subblock to the maximum.
  • coeff_abs_level_greater2_flag means a diagonal scan in FIG. 10.
  • it is able to know a location of a coefficient greater than 2 that appears first.
  • coeff_abs_level_greater1_flag and coeff_abs_level_greater2_flag it is able to derive a basic value (3, 2, 2) of coefficients greater than 1.
  • (3, 5, 9) can be derived by adding the basic value (3, 2, 2) of the coefficients greater than 1 derived through coeff_abs_level_greater1_flag and coeff_abs_level_greater2_flag and the coeff_abs_level_remaining value (0, 3, 7) together.
  • the absolute value of coefficients in each subblock can be inferred as 9, 5, 3.
  • a diagonal scan is performed from a right side to a left side or from a top right end to a bottom left end.
  • xy coordinates of a coefficient value is found.
  • the coordinates become (0, 0).
  • the coordinates become (3, 0).
  • the coordinates become (0, 1).
  • the coordinates become (0, 2).
  • the coordinates become (1, 2).
  • last_sig_coeff_x becomes 3 and last_sig_coeff_y becomes 0, ‘-1’ corresponding to (3, 0) becomes a last coefficient in a transform block.
  • a location of a last coefficient exists at a randomly determined section, e.g., locations of 0, 1, 2, 3, 4, 5, 6, 7, 8, and 9 in the diagonal scan order shown in FIG. 10, it is able to decode a corresponding subblock.
  • FIG. 11 is a detailed flowchart of a process of the inverse transform unit 240 according to one embodiment of the present invention.
  • a process for obtaining a residual signal in a manner of obtaining a transform coefficient by dequantizing a selected random block unit quantization coefficient in the dequantization unit 230 and performing inverse transform using a specific transform base suitable for a block size of the obtained transform block size in the inverse transform unit 240 can be implemented by modifying the ‘ff_hevc_hls_residual_coding’ function within the “libavcodec/hevc_cabac.c” source as follows.
  • the ‘ff_hevcdsp_init_neon’ function within the “libavcodec/arm/hevcdsp_init_neon.c” source can be implemented by being modified as follows.
  • the “libavcodec/hevcdsp.c” source can be implemented by being modified as follows.
  • the “libavcodec/hevcdsp_template.c” source can be implemented by being modified as follows.
  • the source code control logic is described as follows.
  • the proposed method is selectively applicable depending on a size of a transform block.
  • inverse transform can be performed by block units of 4 x 4, 8 x 8, 16 x 16, and 32 x 32.
  • the proposed method is applicable to a block on which inverse transform of a block unit of 16 x 16 or 32 x 32 among 4 x 4, 8 x 8, 16 x 16, and 32 x 32 is performed only. In case of a block unit of 4 x 4 or 8 x 8, all blocks can be decoded.
  • the reconstruction signal selecting unit 210 checks whether a transform block size is 4 x 4 [S1110].
  • the inverse transform unit 240 executes 4 x 4 inverse transform [S1112].
  • the adding-up unit 270 reconstructs 4 x 4 block [S1114].
  • the reconstruction signal selecting unit 210 checks whether a transform block size is 8 x 8 [S1120].
  • the inverse transform unit 240 executes 8 x 8 inverse transform [S1122].
  • the adding-up unit 270 reconstructs 8 x 8 block [S1124].
  • the reconstruction signal selecting unit 210 checks whether a transform block size is 16 x 16 [S1130].
  • the reconstructing signal selecting unit 210 selects 8 x 8 partial block only according to a priority in the 16 x 16 transform block.
  • the inverse transform unit 240 performs 8 x 8 inverse transform on the partial block [S1132]. As the priority is described in detail with reference to FIG. 8, its details are omitted.
  • the residual signal compensating unit 260 performs linear interpolation, i.e., scaling on the 8 x 8 block [S1134].
  • the residual signal compensating unit 260 reconstructs the 8 x 8 block into 16 x 16 block [S1136].
  • the reconstruction signal selecting unit 210 checks whether a transform block size is 32 x 32 [S1140].
  • the reconstruction signal selecting unit 210 selects 16 x 16 partial block only according to a priority in the 32 x 32 transform block.
  • the inverse transform unit 240 performs 16 x 16 inverse transform on the partial block [S1142].
  • the residual signal compensating unit 260 performs linear interpolation, i.e., scaling on the 16 x 16 block [S1144].
  • the residual signal compensating unit 260 reconstructs the 16 x 16 block into 32 x 32 block [S1146].
  • 32 x 32 transform block is divided into 4 subblocks of 16 x 16 unit.
  • a random one of the 4 subblocks can be selectively decoded according to a priority.
  • a single subblock close to a DC value among the 4 subblocks can be decoded only.
  • a prescribed subblock in the 32 x 32 transform block is decoded only, it means that the prescribed subblock is dequantized only and that the rest of subblocks are substituted with a random value instead of performing inverse quantization.
  • the random value may include 0.
  • an output image decoded in the inverse transform process may become a reconstructed block in 32 x 32 size corresponding to a value resulting from performing inverse transform by 32 x 32 unit.
  • a prescribed subblock in the 32 x 32 transform block is decoded only, it means that the prescribed subblock is dequantized and inverse-transformed.
  • a size of a decoded output image may become a size of the prescribed subblock.
  • a decoded output image may include a reconstructed block in 16 x 16 size configured with 4 prescribed subblocks.
  • a system need not maintain a memory for the whole 32 x 32 block, it is efficient in aspects of memory and operation quantity.
  • an inverse transform process for a prescribed subblock may need to be redesigned.
  • a transform block size is 16 x 16 or 32 x 32
  • a prescribed block is selected.
  • inverse transform of a block unit can be performed on the selected prescribed block only.
  • the present invention has an industrial applicability, because the present invention can be applied to any digital device (ex : smart TV, mobile device and so on) including a video decoder.

Landscapes

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

Abstract

La présente invention concerne un décodeur vidéo et son procédé de commande. En particulier, la présente invention est caractérisée en ce qu'elle divise un bloc en sous-blocs d'une unité prescrite, en sélectionnant des sous-blocs prescrits conformément à une priorité, et en décodant les sous-blocs prescrits sélectionnés.
PCT/KR2018/001112 2018-01-25 2018-01-25 Décodeur vidéo et son procédé de commande WO2019146811A1 (fr)

Priority Applications (2)

Application Number Priority Date Filing Date Title
EP18901916.9A EP3744093A4 (fr) 2018-01-25 2018-01-25 Décodeur vidéo et son procédé de commande
PCT/KR2018/001112 WO2019146811A1 (fr) 2018-01-25 2018-01-25 Décodeur vidéo et son procédé de commande

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
PCT/KR2018/001112 WO2019146811A1 (fr) 2018-01-25 2018-01-25 Décodeur vidéo et son procédé de commande

Publications (1)

Publication Number Publication Date
WO2019146811A1 true WO2019146811A1 (fr) 2019-08-01

Family

ID=67396160

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/KR2018/001112 WO2019146811A1 (fr) 2018-01-25 2018-01-25 Décodeur vidéo et son procédé de commande

Country Status (2)

Country Link
EP (1) EP3744093A4 (fr)
WO (1) WO2019146811A1 (fr)

Citations (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20110134999A1 (en) * 2009-12-09 2011-06-09 Samsung Electronics Co., Ltd. Method and apparatus for encoding video, and method and apparatus for decoding video
US20130016774A1 (en) * 2010-07-31 2013-01-17 Soo-Mi Oh Intra prediction decoding apparatus
US20140140410A1 (en) * 2012-06-29 2014-05-22 Wenhao Zhang Systems, methods, and computer program products for scalable video coding based on coefficient sampling
US20140152767A1 (en) * 2012-12-04 2014-06-05 Samsung Electronics Co., Ltd. Method and apparatus for processing video data
US20160142716A1 (en) * 2014-11-17 2016-05-19 Vixs Systems, Inc. Video coder with simplified rate distortion optimization and methods for use therewith

Family Cites Families (7)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US5262854A (en) * 1992-02-21 1993-11-16 Rca Thomson Licensing Corporation Lower resolution HDTV receivers
JP2002164790A (ja) * 2000-11-28 2002-06-07 Canon Inc 圧縮ストリーム復号化装置及び方法並びに記憶媒体
EP1730695A2 (fr) * 2004-03-09 2006-12-13 Thomson Research Funding Corporation Mode de mise a jour de resolution reduite destine a un codage video avance
CN1939066B (zh) * 2004-04-02 2012-05-23 汤姆森许可贸易公司 用于复杂度可伸缩视频解码器的方法和设备
US8781238B2 (en) * 2011-09-08 2014-07-15 Dolby Laboratories Licensing Corporation Efficient decoding and post-processing of high dynamic range images
CN107071420B (zh) * 2012-04-13 2021-07-16 佳能株式会社 视频数据的变换单位的子集的编解码的方法、设备和系统
US10257524B2 (en) * 2015-07-01 2019-04-09 Mediatek Inc. Residual up-sampling apparatus for performing transform block up-sampling and residual down-sampling apparatus for performing transform block down-sampling

Patent Citations (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20110134999A1 (en) * 2009-12-09 2011-06-09 Samsung Electronics Co., Ltd. Method and apparatus for encoding video, and method and apparatus for decoding video
US20130016774A1 (en) * 2010-07-31 2013-01-17 Soo-Mi Oh Intra prediction decoding apparatus
US20140140410A1 (en) * 2012-06-29 2014-05-22 Wenhao Zhang Systems, methods, and computer program products for scalable video coding based on coefficient sampling
US20140152767A1 (en) * 2012-12-04 2014-06-05 Samsung Electronics Co., Ltd. Method and apparatus for processing video data
US20160142716A1 (en) * 2014-11-17 2016-05-19 Vixs Systems, Inc. Video coder with simplified rate distortion optimization and methods for use therewith

Also Published As

Publication number Publication date
EP3744093A1 (fr) 2020-12-02
EP3744093A4 (fr) 2022-01-26

Similar Documents

Publication Publication Date Title
WO2011034372A2 (fr) Procédé et appareil de codage et décodage d'informations de mode
WO2017057953A1 (fr) Procédé et dispositif de codage de signal résiduel dans un système de codage vidéo
WO2012148138A2 (fr) Procédé de prédiction intra, et codeur et décodeur l'utilisant
WO2011152635A2 (fr) Signalisation de modes d'intra-prédiction améliorés
WO2014098456A2 (fr) Procédé de codage/décodage d'une image, et dispositif l'utilisant
WO2017052118A1 (fr) Procédé et dispositif de prédiction intra dans un système de codage d'image
WO2019240448A1 (fr) Procédé et dispositif de traitement de signal vidéo sur la base d'une référence entre des composants
WO2012023763A2 (fr) Procédé de codage d'inter-prédictions
WO2011087323A2 (fr) Procédé et appareil pour encoder et décoder une image en utilisant une unité de transformation importante
WO2018212569A1 (fr) Procédé de traitement d'image basé sur un mode de prédiction intra et appareil associé
WO2015012600A1 (fr) Procédé et appareil de codage/décodage d'une image
WO2013162249A1 (fr) Procédé de codage vidéo, procédé de décodage vidéo et appareil les mettant en œuvre
WO2021015537A1 (fr) Procédé et dispositif de codage/décodage d'image permettant de signaler des informations de prédiction de composante de chrominance en fonction de l'applicabilité d'un mode palette et procédé de transmission de flux binaire
WO2019143103A1 (fr) Procédé et dispositif de codage vidéo à l'aide de diverses techniques de transformation
WO2011126274A2 (fr) Procédés et appareils destinés à coder et à décoder une image sur la base de segments
WO2021118265A1 (fr) Codage vidéo ou d'image utilisant un filtre de boucle adaptatif
WO2019146811A1 (fr) Décodeur vidéo et son procédé de commande
WO2014007514A1 (fr) Procédé permettant de décoder une image et appareil utilisant ce procédé
WO2021118296A1 (fr) Dispositif de codage d'image et procédé de commande de filtrage en boucle
WO2017078450A1 (fr) Procédé et appareil de décodage d'images dans un système de codage d'images
WO2021118297A1 (fr) Appareil et procédé pour coder une image sur la base d'une signalisation d'informations de filtrage
WO2020185027A1 (fr) Procédé et dispositif pour appliquer efficacement un mode de saut de transformation à un bloc de données
WO2018222020A1 (fr) Procédé et appareil de traitement de signal vidéo par modification de zone cible
WO2018174657A1 (fr) Procédé et dispositif de décodage d'image en fonction d'une prédiction intra dans un système de codage d'image
WO2017065490A1 (fr) Procédé destiné au codage/décodage d'une image et appareil s'y rapportant

Legal Events

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

Ref document number: 18901916

Country of ref document: EP

Kind code of ref document: A1

NENP Non-entry into the national phase

Ref country code: DE

ENP Entry into the national phase

Ref document number: 2018901916

Country of ref document: EP

Effective date: 20200825