Method and apparatus for encoding/decoding image.
Abstract
A method for decoding an image, according to the present invention, comprises the steps of: receiving image information that corresponds to a block to be decoded; performing entropy-decoding with respect to the image information that is received; deciding a transform skip mode of the block to be decoded from a plurality of transform skip mode candidates, based on the image information that is entropy-decoded; and reverse-transforming the block to be decoded based on the transform skip mode that is decided.

Term
7.6 yearsleft in the term
Expires 16 April 2034.
- Priority
- Filed
- Granted
- Today
- Expires
4 claims: 1 independent, 3 dependent
- 1CLAIMS REIVINDICACIONES IMPI Dt LA RI »IPO £ D5F; • NrXISTRlAi IMPI Dt LA RI»OPI£D5f; •NrXISTRlAi 1. Un método para decodificar una señal de vídeo, el método comprende:one. A method of decoding a video signal, the method comprises: get residual coefficients that relate to a current block of a bit stream;obtener coeficientes residuales que se relacionan con un bloque actual de un flujo de bits;obtain the inverse residual coefficients quantified by inverse quantification of the residual coefficients;obtener los coeficientes residuales inversos cuantificados por cuantificación inversa de los coeficientes residuales;determinar, en base a un índice del modo de transformación de salto que especifica un modo de transformación de salto en relación con el bloque actual, el modo de salto en relación con el bloque actual a partir de un candidato del modo de transformar el salto, en donde el candidato del modo transformación del salto incluye al menos una de un modo de transformación 2D, un modo de transformación horizontal, un modo de transformación vertical o un modo de no transformación, y en donde un número del candidato del modo de transformación de salto es diferente de conformidad con ya sea un tamaño del bloque actual o una forma del bloque actual;y obtener, en base al modo determinado de transformación del modo de salto, muestras residuales que se relacionan con el bloque actual a partir de los coeficientes residuales inversos en donde, cuantificados, cuando el modo determinado de determine, based on a jump transform mode index specifying a jump transform mode relative to the current block, the jump mode relative to the current block from a hop transform mode candidate, wherein the jump transformation mode candidate includes at least one of a 2D transformation mode, a horizontal transformation mode, a vertical transformation mode, or a non-transformation mode, and wherein a hop transform mode candidate number is different in accordance with either a current block size or a current block shape;and obtain, based on the determined mode of transformation of the jump mode, residual samples that are related to the current block from the inverse residual coefficients where, quantified, when the determined mode of ΪΜΡΙ : ί»5Τΐη” »* ΜΐθΜ <ο! ELAMenEDAU 'NOUSTWIAI transformation is the non-transformed mode, the residual samples are obtained by scaling the residual quantized inverse coefficients with a pre-determined value. ΪΜΡΙ :ί»5Τΐη”»*ΜΐθΜ<ο !ELAMenEDAU ’NOUSTWIAI transformación es el modo no transformado, las muestras residuales se obtienen escalando los coeficientes inversos cuantificados residuales con un valor pre-determinado.
427 paragraphs in 71 sections, as filed
(54) Title: METHOD AND APPARATUS FOR CODING / DECOFICING IMAGES. (54) Title: METHOD AND APPARATUS FOR ENCODING / DECODING IMAGE.
(57) Summary
A method of decoding an image, in accordance with the present invention, comprises the steps of receiving image information corresponding to a block to be decoded, performing entropy decoding with respect to the image information received, deciding a mode to skip the transformation of the block to be decoded from a plurality of candidates so to skip the transformation, based on the image information that is decoded by entropy; and the inverse transformation of the block that will be decoded based on the mode to skip the transformation that is decided.
(57) Abstract
A method for decoding an image, according to the present invention, comprises the steps of: receiving image information that corresponds to a block to be decoded; performing entropy-decoding with respect to the image Information that is received; deciding a transform skip mode of the block to be decoded from a plurality of transform skip mode candidates, based on the image information that is entropy-decoded; and reverse-transforming the block to be decoded based on the transform skip mode that is decided.
I KNOW
Institute
Mexican Property
Industrial
<img file="MX339392B_D0001.tif" />
PATENT TITLE NO. 339392
Owner (s): KT CORPORATION
Address: 90 Buljeong-ro, Bundang-gu, Seongnam-city, Kyeongg¡-do, 463-711, KOREA
Name: METHOD AND APPARATUS FOR CODING / DECODING IMAGES
Classification: lnt.CI.8; H04N19 / 124; H04N19 / 503; H04N19 / 513; H04N19 / 60; H04N19 / 61;
H04N19 / 91
Inventor (s): BAE KEUN LEE; JAE CHEOL KWON; JOO YOUNG KIM
REQUEST
Number: International filing date:
MX / a / 2015/014510 October 17, 2012
Divisional Patent Number: 334355
PRIORITY
Country: Date: Number:
KR October 17, 2011 10-2011-016107
Validity: Twenty years
Expiration Date: October 17, 2032
The reference patent is granted based on articles 1, 2 fraction V, 6 fraction III, and 59 of the Industrial Property Law.
In accordance with article 23 of the Industrial Property Law, this patent is valid for twenty years without extension, counted from the date of filing of the international partnership and will be subject to the payment of the fee to maintain the rights in force. .
Whoever subscribes to this title does so based on the provisions of articles 6 sections III and 7 bis 2 of the Industrial Property Law (Official Gazette of the Federation (DOF, 06/27/1991, amended on 02 / 08/1994, 10/25/1996, 12/26/1997, 05/17/1999, 01/26/2004, 06/16/2005, 01/25/2006, 06/05/2009 / 06/01 / 2010, 06/18/2010, 06/28/2010, 01/27/2012 and 04/09/2012); Articles 1, 3, fraction V, subsection a), 4, and 12, sections l and lll of the Regulations of the Mexican Institute of Industrial Property (DO.F. 07/2004, 07/28/2004 and 09/07/2007); articles 1, 3, 4, 5, traction v inosoa), 16 fractions l and lll and 30 of the Organic Statute of the Mexican Institute of Industrial Property (DOF 12/27/1999, amended on 10 / 1Q / 2Q02, 07/29/2004, 08/04/2004 and 09/13/2007); 1, 3 and 5 subsection a) of the Agreement that delegates powers to the Deputy Directors General, Coordinator, Divisional Directors, Holders of the Regional Offices, Divisional Deputy Directors, Departmental Coordinators and other subordinates of the Mexican Institute of Industrial Property. (D OF 12/15/1999, amended on 02/04/2000, 07/29/2004, 08/04/2004 and 09/13/2007).
Issue Date: May 25, 2016
<img file="MX339392B_D0002.tif" />
<img file="MX339392B_D0003.tif" />
33W _Zbl5 // y & O
METHOD AND APPARATUS FOR CODING / DECODING IMAGES
The present invention relates to image processing, and more particularly, to a method for transformation and apparatus for transformation.
PREVIOUS TECHNIQUE
Recently, the demands for high-resolution, high-quality videos such as high-definition (HD) and ultra-high-definition (UHD) videos have increased.
To provide videos with higher resolution and higher quality, the amount of video data increases. Accordingly, the costs of transferring and storing video data are increased in order to provide high-quality videos compared to conventional video data processing methods. In order to solve these problems that occur with increased resolution and quality of video data, highly efficient video compression techniques can be used.
As a video data compression technology, various schemes are used as inter prediction that depends on the image data elements different from the current image, intra prediction that is derived from only data elements of the same decoded slice, and the entropy of
<img file="MX339392B_D0004.tif" />
<img file="MX339392B_D0005.tif" />
IMPI <sup>NiT</sup>S7 '?, Í<sup>, fX, CAN</sup><>
The PROPERTY • ndustrim encoding / decoding of shortest signaling codes that occur or appear frequently.
DESCRIPTION
Technical problem
One aspect of the present invention is to provide a video encoding method and a video encoding apparatus that are capable of increasing video encoding performance.
Another aspect of the present invention is to provide a video decoding method and a video decoding apparatus that is capable of increasing the performance of video decoding.
Still another aspect of the present invention is to provide a transformation method and transformation apparatus that are capable of increasing video encoding performance.
However, another aspect of the present invention is to provide a reverse transform method and reverse transform apparatus that are capable of increasing video decoding performance.
However, another aspect of the present invention is to provide a scanning method and scanning apparatus that are capable of increasing video encoding performance.
<img file="MX339392B_D0006.tif" />
is to provide a reverse scanning method and a reverse scanning apparatus that are capable of increasing the performance of video decoding.
Technical solution
An embodiment of the present invention provides a method of video decoding. The method may include receiving information about an image corresponding to a decoding destination block, determining a transform jump mode (TSM) for the decoding destination block among a plurality of TSM candidates based on the entropy decoded information about the image, and the inverse transformation of the decoding destination block based on the determined TSM. Herein, TSM candidates may include at least one of a 2-way (2D) transformation mode to perform both horizontal transformation and vertical transformation, a horizontal transformation mode to perform horizontal transformation, a vertical transform to perform vertical transform and a no transform mode · that does not perform the transform.
The image information may include information in a prediction mode corresponding to the decoding destination block and a type of a unit.
<img file="MX339392B_D0007.tif" />
prediction (PU) corresponding to the decoding destination block.
When the prediction mode corresponding to the decoding destination block is an Inter mode and the 5 PU type corresponding to the decoding destination block is Nx2N, where N is a natural number, the vertical transformation mode can assign a codeword shorter than horizontal transformation mode.
When the prediction mode corresponding to the decoding destination block is an inter mode and the PU type corresponding to the decoding destination block is 2NxN, where N is a natural number, the TSM candidates can include the 2D transformation mode , the horizontal transformation mode and the no transformation mode except for the vertical transformation mode.
When the prediction mode corresponding to the decoding destination block is an inter mode and the PU type corresponding to the decoding destination block is 1Nx2N, where N is a natural number, the TSM candidates can include the 2D transformation mode , vertical transformation mode and no transformation mode except for horizontal transformation mode.
<img file="MX339392B_D0008.tif" />
<sup>IEU</sup>| ND?<sub>)</sub><sup>F</sup>''<sup>FD</sup>''<sup>r</sup>’ '
When the prediction mode runs sponcíi'éht decoding destination block is a remote intra prediction mode (SDIP) and the PU type corresponding to the decoding destination block is 2NX (1/2) N, N being a number Natural that is 2 or greater, TSM candidates can include 2D transform mode, horizontal transform mode, and no transform mode except for vertical transform mode.
When the prediction mode corresponding to the decoding destination block is an SDIP mode and the PU type corresponding to the decoding destination block is (1/2) Nx2N, where N is a natural number that is 2 or greater, the TSM candidates can include 2D transformation mode, vertical transformation mode, and no transformation mode except for horizontal transformation mode.
The image information may include information about a prediction mode corresponding to the decoding destination block and prediction address of a PU corresponding to the decoding destination block.
When the prediction mode corresponding to the decoding destination block is an intra mode and the PU prediction direction corresponding to the block
IMPI
MEXICAN INSTITUTE OF INDUSTRIAL PROPERTY
<img file="MX339392B_D0009.tif" />
decoding target is a vertical address, vertical transformation mode, a shorter codeword can be assigned than horizontal transformation mode.
The video decoding method may further include determining a scanning mode for the decoding destination block based on TSM determination, and inverse of scanning the decoding destination block based on the determined scanning mode.
The determination of the scanning mode can determine a vertical scanning mode than the scanning mode when the determined TSM is the horizontal transformation mode.
The determination of the scanning mode can determine a horizontal scanning mode, such as the scanning mode when the determined TSM is the vertical transformation mode.
Another embodiment of the present invention provides a video decoding apparatus. The apparatus may include an entropy decoding module for receiving information in an image corresponding to a decoding destination block and entropy decoding the information in the image, and an inverse transformation module for determining a TSM for the destination block for decoding among a plurality of TSM candidates
<img file="MX339392B_D0010.tif" />
MEXICAN INSTITUTE OF INDUSTRIAL PROPERTY based on the entropy-encoded information on the image and to invert-transform the decoding destination block based on the determined SST. Herein, TSM candidates comprise at least one of a 2D transformation mode to carry out both horizontal transformation and vertical transformation, a horizontal transformation mode to carry out horizontal transformation, a vertical transformation mode to carry out perform vertical transformation and a no transformation mode of not performing the transformation.
Still another embodiment of the present invention provides a video encoding method. The method may include generating a residual block corresponding to a coding destination block, determining a TSM for the coding destination block among a plurality of TSM candidates; and transform the residual block based on the determined SST.
Herein, TSM candidates may include at least one of a 2D transformation mode to carry out both horizontal transformation and vertical transformation, a horizontal transformation mode to carry out horizontal transformation, a vertical transformation mode to carry out perform vertical transformation and a no transformation mode to not perform transform.
<img file="MX339392B_D0011.tif" />
IMPI
MEXICAN INSTITUTE OE INDUSTRIAL PROPERTY
A prediction mode corresponding to the coding destination block may be an inter-mode, and TSM determination may determine TSM based on a type of a PU corresponding to the coding destination block.
A prediction mode corresponding to the encoding destination block may be an SDIP mode, and TSM determination may determine TSM based on a type of a PU corresponding to the encoding destination block.
A prediction mode corresponding to the coding destination block may be an intra mode, and the determination of the TSM may determine the TSM based on the prediction intra mode address of a PU corresponding to the coding destination block.
The video encoding method may further include determining a scanning mode for the encoding destination block based on TSM determination, and examining the encoding destination block based on the determined scanning mode.
However, another embodiment of the present invention provides a video encoding apparatus. The apparatus may include a residual block generation module for generating a residual block corresponding to a coding destination block, and a
ΙΜΡΙ
INSTITUTO MEXICANO DE LA ÍROflIDA · INDUSTRIAL
<img file="MX339392B_D0012.tif" />
transformation to determine the TSM for the encoding destination block among a plurality of determined TSMs and transform the residual block based on determined TSM. Herein, TSM candidates may include at least one of a 2D transformation mode to carry out both horizontal transformation and vertical transformation, a horizontal transformation mode to carry out horizontal transformation, a vertical transformation mode to carry out perform vertical transformation and a no transformation mode that does not perform transformation.
Advantageous effects
In accordance with a video encoding method of the present invention, the video encoding performance can be improved.
In accordance with a video decoding method of the present invention, video decoding performance can be improved.
In accordance with a reverse transformation / transformation method of the present invention, the video encoding / decoding performance can be improved.
<img file="MX339392B_D0013.tif" />
sweep / reverse of the present invention, the video encoding / decoding performance can be improved.
DESCRIPTION OF THE DRAWINGS
FIG. 1 is a block diagram illustrating a configuration of a video encoding apparatus in accordance with an illustrative embodiment of the present invention.
FIG. 2 is a block diagram illustrating a configuration of a video decoding apparatus in accordance with an illustrative embodiment of the present invention.
Figure 3 schematically illustrates a transformation method based on a transformation mode in accordance with an illustrative embodiment of the present invention.
Figure 4 is a flow diagram schematically illustrating a process of transforming the coding apparatus in accordance with an illustrative embodiment of the present invention.
FIG. 5 is a flow diagram schematically illustrating a reverse process of the transformation decoding apparatus in accordance with an illustrative embodiment of the present invention.
IMPI
INSTRrUTO MEXICANO ΠΕ LA ERORIEDAD
INDUSTRIAL
<img file="MX339392B_D0014.tif" />
Figure 6 illustrates a method for determining a candidate jump transformation mode and a method of assigning a codeword to a transformation jump mode according to a PU shape in an intermode.
Figure 7 illustrates a method for determining a given transformation jump mode and a method of assigning a codeword to a transformation jump mode according to a PU form in SDIP.
Figure 8 illustrates a method of assigning a codeword to a transformation jump mode according to the instructions of the intra prediction mode.
Figure 9 schematically illustrates a transformation coefficient scanning method based on a transformation jump mode in accordance with an illustrative embodiment of the present invention.
FIG. 10 is a flow diagram schematically illustrating a coding method in accordance with an illustrative embodiment of the present invention.
FIG. 11 is a flow chart schematically illustrating a decoding method in accordance with an illustrative embodiment of the present invention.
IMPI
<img file="MX339392B_D0015.tif" />
Mode of the invention
Although the elements illustrated in the drawings are shown independently in order to represent different distinctive functions in a video encoding apparatus / decoding apparatus, such as such a configuration does not indicate that each element is constructed of a component of separate hardware or software constituent. That is, the elements are independently arranged for the convenience of the description, where at least two elements can be combined into a single element, or a single element can be divided into a plurality of elements to perform functions. It is to be noted that the modalities in which some elements are integrated into a combined element and / or an element is divided into multiple separate elements are included within the scope of the present invention without departing from the essence of the present invention.
Hereinafter, exemplary embodiments of the invention will be described in detail with reference to the accompanying drawings. Reference numbers in the drawings are
<td>refer to</td><td>similar items,</td><td>and</td><td>the descriptions</td>
<td>redundant</td><td colspan="2">of how elements will be</td><td>omitted in this</td>
<td>document.</td><td></td><td></td><td></td>
<td>The</td><td>Figure 1 is a diagram</td><td>of</td><td>blocks illustrating</td>
a configuration of a video encoding apparatus of
IMPI
MEXICAN INSTITUTE OF INDUSTRIAL PROPERTY
<img file="MX339392B_D0016.tif" />
in accordance with an illustrative embodiment of the present invention. Referring to Figure 1, the video encoding apparatus may include an image slice module 110, an Inter prediction module 120, an intra prediction module 125, a transformation module 130, a quantization module 135, a dequantization module
140, a reverse transformation module 145, a filter module 150, a memory 155, a rearrangement module 160, and an entropy encoding module 165.
Image slice module 110 can divide an input image into one or more encoding units.
A coding unit (CU) is a coding unit carried out by the video coding apparatus and can be recursively subdivided with depth information based on a quaternary tree structure. A CU can have different sizes of 8 x 8, 16 x 16, 32 x 32, and 64 x 64. A CU with a maximum size is known as a larger unit of encoding (LCU), and a CU with a minimum size such as smallest coding unit (SCU).
Image slice module 110 can divide a CU to generate a prediction unit (PU) and a transformation unit (TU). A PU may be less than or equal to a CU, and may not necessarily be a square block, but is a rectangular block.
<img file="MX339392B_D0017.tif" />
IMPI
MEXICAN INSTITUTE OF THE INDUSTRIAL AGE ROM
Generally, intra prediction can be performed by 2N * 2N or N * N blocks. Herein, N is a natural number, representing a number of pixels, and 2N * 2N or N * N can represent a PU size (and / or cut mode). However, at a short intra prediction distance (SDIP), not only a 2N * 2N PU but a subdivided PU with a size of hN * 2N / 2N * hN (in this case, h = 1/2) can also be used to increase efficiency in intra prediction. When a hN * 2N / 2N * hN of the PU is used, the directivity of a limit in a block can be reflected more, and consequently the energy of a prediction error signal can be decreased to reduce the number of necessary bits for encoding, thereby increasing encoding efficiency.
The inter prediction can be done by blocks of 2N * 2N, 2N * N, N * 2 N or N * N.
Herein, N is a natural number, representing a number of pixels, and 2N * 2N, 2N * N, N * 2 N or N * N can represent a PU size (and / or cut mode) . Furthermore, between the prediction can be made by
2NxnU, 2NxnD, nLx2N or nRx2N PU, in addition to the 2N * 2N, 2N * N,
N * 2NoN * N PU, in order to improve efficiency in inter prediction. Here, 2NxnU, 2NxnD, nLx2N, or nR.x2N can represent a PU (and / or break mode. In 2NxnU and 2NxnD break modes, a PU can have a size
IMPI
<img file="MX339392B_D0018.tif" />
of 2NX (L / 2) N or 2NX ($ / 2) N, while e ~ ñ '' m¿> dT) 'S de-short<sup>1 </sup>nLx2N and nRx2N, a PU can have a size of (1/2) Nx2N or (3/2) Nx2N.
In an inter prediction mode, the inter prediction module 120 can perform motion estimation (ME) and motion compensation (MC). The inter prediction module 120 can generate a prediction block based on the information on at least one of the previous and subsequent images of the current image.
The inter prediction module 120 can perform motion estimation based on a divided prediction target block and at least one reference block stored in memory 155. The inter prediction module 120 can generate motion information including a vector motion (MV), a reference block index, and a prediction mode as a result of motion estimation.
Furthermore, the inter prediction module 120 can perform motion compensation using the motion information and the reference block. In this case, the inter prediction module 120 can generate and output from a corresponding prediction block an input block of the reference block.
In an intra prediction mode, the intra prediction module 125 can generate a prediction block based
IMPI
MEXICAN INSTITUTE OF INDUSTRIAL PROPERTY
<img file="MX339392B_D0019.tif" />
in the information of a pixel in the current image. In the intra prediction mode, the intra prediction module 125 can carry out the prediction of a current block, based on a prediction target block and a previously reconstructed reconstructed block through transformation and quantization. Herein, the reconstructed block may be a reconstructed image that has not been subjected to filter module 150.
In the inter prediction mode or intra prediction mode described above, the prediction can be performed on a prediction target block to generate a prediction block. Herein, a residual block can be generated by differentiation between the prediction target block and the generated prediction block.
The transformation module 130 can transform a residual block by a TU to generate a transformation coefficient. A TU can have a tree structure within maximum and minimum sizes. It is possible to indicate by means of a flag if a current block is divided into sub-blocks for each TU. The transform module 13 0 can
<td>perform</td><td>the</td><td>transformation</td><td>based on a</td><td>transformation</td>
<td>discreet</td><td>of the</td><td>cosine (DCT)</td><td>and / or sinusoidal</td><td>transformation</td>
<td>discreet</td><td>(DST</td><td> ) .</td><td></td><td></td>
<td></td><td>The</td><td colspan="3">quantization module 130 can quantify</td>
values transformed by the transform module 130. A
IMPI
MEXICAN INSTITUTE »E LA RRORIBDAD
INDUSTRIAL
<img file="MX339392B_D0020.tif" />
Quantization coefficient can change based on a_block or the importance of an image. The quantized transformation coefficient can be provided to the rearrangement module 160 and the dequantization module 140.
The rearrangement module 160 can organize a two-dimensional (2D) block of the transform coefficients quantized into a one-dimensional (ID) vector of transform coefficients in the examinations in order to improve the efficiency of entropy coding. The rearrangement module 160 can change the scanning order based on stochastic statistics to improve the entropy coding efficiency.
The entropy encoding module 165 can entropy to encode the values obtained by the reordering module 160. In entropy encoding, a syntax element value occur more frequently a code word can be assigned from smaller bit numbers , while a less frequently occurring syntax element value can assign a code word of more bit numbers. Therefore, a bit string size for the symbols to be encoded can be reduced to improve the compression performance of video encoding. Various encoding methods, such as Exponential Golomb encoding, Context Adaptive Variable Length Coding (CAVLC) and / or
ΙΜΡΙ
MEXICAN INSTITUTE OF INDUSTRIAL PROFFKDAD
<img file="MX339392B_D0021.tif" />
Adaptive binary arithmetic encoding (CABAC) can be used for entropy encoding. The encoded information can be formed in a compressed bitstream and is transferred or stored through a network abstraction layer (NAL).
The dequantization module 140 can dequantize the transformation coefficients quantized by the quantization module 135, and the inverse transformation module 145 can generate a reconstructed residual block for inverse transformation of the dequantized transformation coefficients. The reconstructed residual block can be merged with the prediction block generated by the inter prediction module 120 or the intra prediction module 125 to generate a reconstructed block. The reconstructed block can be provided to the intra prediction module 125 and the filter module 150.
Filter module 150 can filter the reconstructed residual block using an unlock filter, a sample adaptive offset (SAO) and / or an adaptive loop filter (ALF). The unlock filter can filter the reconstructed block in order to eliminate a distortion in the boundaries between blocks that occur in encoding and decoding. SAO is a loop filtration process to be performed on the residual block through the unlocking filter to rebuild a
IMPI
MEXICAN INSTITUTE OE IA MOR INDUSTRIAL AGE
<img file="MX339392B_D0022.tif" />
offset difference of an original image by one pixel. A scroll band and an edge scroll can be used as the ODS. Band offset can divide a pixel into 32 bands according to intensity and apply offsets to two divided groups of 16 bands in an edge area and 16 bands in a downtown area. The ALF can filter to minimize an error between the prediction target block and the finally rebuilt block. The ALF can perform filtering based on a value obtained by comparing the reconstructed block filtered by the unlock filter with the current target prediction block, and the filter of the coefficient information in the ALF can be loaded into a section header and transferred from the encoding apparatus to the decoding apparatus.
The memory 155 can store the finally reconstructed block through the filter module 150, and the finally reconstructed block can be provided to the inter prediction module by carrying out the inter prediction.
FIG. 2 is a block diagram illustrating a configuration of a video decoding apparatus in accordance with an illustrative embodiment of the present invention. Referring to Figure 2, the video decoding apparatus may include an entropy decoding module 210, a reordering module
IMPI <sup>, NST</sup>MEXICAN MTO OF THE INDUSTRIAL MOR
<img file="MX339392B_D0023.tif" />
215, a dequantization module 220, an inverse transformation module 225, an inter prediction module 230, an intra prediction module 235, a filter module 240 and a memory 245.
The entropy decoding module 210 can receive a compressed bit stream from a NAL. The modulo 210 entropy decoding can entropy decode the received bitstream, and also entropy decode a prediction mode and motion vector information if the bitstream includes the prediction mode and motion vector information. When using entropy decoding, a syntax element value occurs more frequently which can assign a code word of smaller bit numbers, while a smaller syntax element value occurs more frequently than one. code word with more bit numbers. Therefore, a bit string size for the symbols to be encoded can be reduced to improve the compression performance of video encoding.
An entropy encoding transformation coefficient or residual signal can be provided to the rearrangement module 215. The rearrangement module 215 can examine the transformation coefficient of
IMPI Mexican Institute OF IA INDUSTRIAL PROPERTY
<img file="MX339392B_D0024.tif" />
inverse decoding or residual signal to generate a block of 2D transformation coefficients.
The dequantization module 220 can dequantize the rearranged transformation coefficients. The inverse transform modulo 225 can reverse transform the transform dequantization coefficients to generate a residual block.
The residual block can be combined with a prediction block generated by the inter prediction module 230 or the intra prediction module 235 to generate a reconstructed block. The reconstructed block can be provided to the intra prediction module 235 and the filter module 240. The inter prediction module 230 and the intra prediction module 235 performs operations of the same or equivalent to those of the inter prediction module 120 and the intra prediction module 125 of the video encoding apparatus, and therefore the descriptions of the they will be omitted herein.
Filter module 240 can filter the rebuilt block using an unlock filter, an SAO, and / or an ALF. The unlock filter can filter the rebuilt block to eliminate distortion at a boundary between blocks that occurs in encoding and decoding. SAO can be applied to the rebuilt block filtered by
IMPI
MEXICAN INSTITUTE Of. INDUSTRIAL PROPERTY the one pixel unlock filter to reduce a
<img file="MX339392B_D0025.tif" />
difference from an original image. The ALF can filter the rebuilt block through the SAO to minimize an error between the prediction target block and the finally rebuilt block.
The memory 245 can store the finally reconstructed block obtained through the filter module 240, and the stored reconstructed block can be provided to the inter prediction module 230 to perform the inter prediction.
Hereinafter, a block may refer to a video encoding and decoding unit. Therefore, in this description, a block can mean a CU, PU, TU, and the like. Furthermore, an encoding / decoding destination block may collectively include a transformation / reverse transformation target block, if the transformation / reverse transformation is performed; a prediction target block, if prediction is carried out; and the like.
As described above with reference to Figures 1 and 2, the encoding apparatus can perform transformation into a residual block using a TU, and the decoding apparatus can reverse transform dequantized transformation coefficients to generate a reconstructed residual block.
IMPI
MEXICAN INSTITUTE OF INDUSTRIAL PROPERTY
<img file="MX339392B_D0026.tif" />
In the following description, invorous transform can also be referred to as convenience transformation as required, which will be easily understood by a person having ordinary skill in the art.
The encoding apparatus and the decoding apparatus can perform bidirectional (2D) transformation including vertical transformation and horizontal transformation. However, when vertical and horizontal signals have markedly different characteristics, vertical transformation or horizontal transformation can be omitted. Also, the entire process of transforming can be skipped for a poor signal. Such transformation methods can reduce complexity in the decoding apparatus and improve encoding efficiency.
Hereafter, a transformation mode involving horizontal transformation and vertical transformation is referred to as a 2D transformation mode. A transformation mode that involves horizontal transformation only without vertical transformation is called a horizontal transformation mode, and a transformation mode that involves vertical transformation only without horizontal transformation is called a vertical transformation mode. Also, a transformation mode that involves neither horizontal transformation nor vertical transformation is
IMPI
<img file="MX339392B_D0027.tif" />
known as a transformationless mode.
<img file="MX339392B_D0028.tif" />
Non-transform mode is also called as a transformation bypass mode.
Figure 3 schematically illustrates a transformation method based on a transformation mode in accordance with an illustrative embodiment of the present invention.
Quadrant blocks 310-340 shown in
Figure 3 are destination transformation blocks. In this case, the transformation destination blocks can be TU and / or CU. Also, the arrows marked on blocks 310-330 can indicate transformation directions.
A target transform block 310 can undergo both vertical transformation and horizontal transformation. Therefore, a transform mode for block 310 may correspond to the 2D transform mode. A transformation target block 320 can be subjected to horizontal transformation only without vertical transformation. Therefore, a transformation mode for block 320 can correspond to the horizontal transformation mode. In this case, since the transformation is done in rows, not columns, a method of transformation in horizontal transformation mode can also be called as transformation in only rows. A destination transform block 330 can undergo vertical transformation alone without
IMPI
MEXICAN INSTITUTE OF INDUSTRIAL PROPERTY
<img file="MX339392B_D0029.tif" />
horizontal transformation. In the prpRpnt-p. nn transformation mode for block 330 corresponds to vertical transformation mode. In this case, the transformation is done on columns, not on rows, a transformation method in vertical transformation mode can also be called as transformation only on columns. A destination transform block 340 cannot be transformed. Therefore, a transformation mode for block 340 is the no transformation mode.
In the above transformation modes, the vertical transformation and / or horizontal transformation may or may not be omitted. Therefore, these transformation modes can also be referred to as a transformation jump mode (TSM). That is, the transformation jump mode can include 2D transformation mode, horizontal transformation mode, vertical transformation mode, and no transformation mode. 2D transform mode, horizontal transform mode, vertical transform mode, and / or no transform mode can be used as candidates for transform jump mode for a transformation target block.
In an illustrative embodiment, at least one of the 2D transformation mode, the horizontal transformation mode,
<img file="MX339392B_D0030.tif" />
IMPI
The Mexican Institute of Industrial Property vertical transformation mode and —non-transformation mode can be used as a transformation jump mode candidate for an objective transformation block. Herein, the selected transformation jump mode from a plurality of transformation jump mode candidates can be applied to a target block transformation. The coding apparatus may select a transform jump mode with a lower cost value, in view of the rate distortion optimization (RDO) among the transformation jump mode candidates. Herein, the encoding apparatus can transform the destination transform block based on the selected transform jump mode. That is, you can apply the encoding apparatus of a selected transform jump mode from 2D transform mode, horizontal transform mode, vertical transform mode, and / or no transform mode to the target transform block.
Furthermore, the encoding apparatus can encode information about the selected transform jump mode and transmit the information to the decoding apparatus. The transformation jump mode can be determined by a CU or TU. Herein, when the transformation jump mode is determined by a CU, the
<img file="MX339392B_D0031.tif" />
IMPI
INSTITUTO MEXICANO DE LA PROPIEDAD INDUSTRIAL information may be transmitted by a CU. When the transformation jump mode is determined by a TU, the information can be transmitted by a TU.
For example, the information about the transformation jump mode can be transmitted to the decoding apparatus through a transformation jump mode index. The transformation jump mode index can be an index indicating the transformation jump mode to be applied to the transformation target block among the transformation jump mode candidates. The transformation jump mode index can be assigned an index value based on the transformation jump mode. Herein, 2D transformation mode, horizontal transformation mode and vertical transformation mode can correspond to different transformation index values.
The decoding apparatus may decode the information about the transformation jump mode (eg, the encoded transformation jump mode index) that is received from the apparatus that performs the encoding. Herein, the decoding apparatus may derive the transformation jump mode to be applied to the transformation destination block based on the decoded information. The decoding apparatus can transform the destination transformation block according
<img file="MX339392B_D0032.tif" />
IMPI
INSTITUTO MEXICANO de la peopieí ao INDUSTRIAL with the derived transformation jump mode. That is, the decoding apparatus can apply a transformation jump mode derived from 2D transformation mode, horizontal transformation mode, vertical transformation mode, and / or no transformation mode to transform the transformation target block.
Figure 4 is a flow diagram schematically illustrating a transformation process of the coding apparatus in accordance with an illustrative embodiment of the present invention.
Referring to Figure 4, the encoding apparatus can determine a transformation jump mode for a transformation destination block among a plurality of transformation jump mode candidates (S410). Herein, transformation jump mode candidates may include at least one of 2D transformation mode, horizontal transformation mode, vertical transformation mode, and no transformation mode. In this case, the encoding apparatus may select a transform jump mode that has a lower cost value, in view of RDO among the candidates for the transformation jump mode. A method of determining a candidate transformation jump mode in accordance with an illustrative embodiment will be described in detail.
IMPI
<img file="MX339392B_D0033.tif" />
Again referencing
<img file="MX339392B_D0034.tif" />
encoding apparatus can transform the target block according to the determined transform jump mode (S420). That is, you can apply the encoding apparatus of a transform jump mode selected from 2D transform mode, horizontal transform mode, vertical transform mode, and no transform mode to the transformation target block .
Furthermore, the encoding apparatus can encode the information in the transformation hop mode applied to the transformation destination block and transmit the information to the decoding apparatus. For example, the information can be transmitted to the decoding apparatus through a transformation jump mode index. Herein, as described above, taking into account the transformation sato mode probabilities, the coding apparatus may assign a short code word to a probable transformation jump mode and a long code word to a transformation mode. transformation jump less likely.
FIG. 5 is a flow diagram schematically illustrating a reverse process of the transformation decoding apparatus in accordance with an illustrative embodiment of the present invention.
IMPI
MEXICAN INSTITUTE Ut LA PEONE n<sub>TO</sub>, ¡ '<sup>n</sup>l> 'J5TKIal
<img file="MX339392B_D0035.tif" />
The decoding apparatus may___dgcpdify a bit stream including the information about the transform jump mode (eg, the encoded transform jump mode index) that is received from the encoding apparatus. In the bit stream received from the encoding apparatus, a short code word can be assigned to a more probable transformation jump mode and a long code word can be assigned to a less probable transformation jump mode. A method of assigning a codeword for a transformation jump mode according to an illustrative embodiment 5 will be described in detail.
Referring to Figure 5, the decoding apparatus can derive a jump mode transformation for a destination inverse transformation block among a plurality of transformation jump mode candidates (S510). Herein, transformation jump mode candidates may include at least one of 2D transformation mode, horizontal transformation mode, vertical transformation mode, and no transformation mode. The decoding apparatus may use the same transformation hop mode candidate as used in the coding apparatus. Herein, the decoding apparatus can derive the transformation jump mode target block for inverse transformation on
IMPI
MEXICAN INSTITUTE OE INDUSTRIAL PROPERTY
<img file="MX339392B_D0036.tif" />
the base of the decoded information (the information about the transformation jump mode, for example, the decoded index of transformation jump mode). A method of determining a candidate transformation jump mode in accordance with an illustrative embodiment will be described in detail.
Referring again to Figure 5, the decoding apparatus can reverse transform the destination reverse transform block according to the derived transform jump mode (S520). That is, you can apply the decoding apparatus of a selected transformation jump mode to 2D transformation mode, horizontal transformation mode, vertical transformation mode, and / or non-transformation mode to the inverse transformation target block.
Meanwhile, in the modalities illustrated in Figures 4 and 5, the encoding apparatus and the decoding apparatus can utilize the entire 2D transform mode, the horizontal transform mode, the vertical transform mode and / or the no transform mode. as transformation jump mode candidates. Herein, the 2D transformation mode (and / or a transformation jump mode index corresponding to the 2D transformation mode), the horizontal transformation mode (and / or
IMPI
INSTITUTO MEXICANO Ot LA PRORfEOAO a transformation jump mode index corresponding to the horizontal transformation mode), the vertical transformation mode (and / or a transformation jump mode index corresponding to the vertical transformation mode) and / or the No transform mode (and / or transform jump mode index corresponding to no transform mode) Different codewords can be assigned, respectively. In this case, as described above, the coding apparatus may assign a short code word to a more likely transformation jump mode and a long code word to a less likely transformation jump mode taking into account the probabilities. of transformation jump mode. Table 1 illustrates a method of assigning a codeword for a transformation jump mode according to an illustrative embodiment.
IMPI
INSTITUTO MEXICANO IM LA NKJMEDAD INDUSTRIAL
<img file="MX339392B_D0037.tif" />
Table 1
<td>TSM</td><td>Transformation row</td><td>Transformation column</td><td>Word of code (CABAC me CAVLC)</td><td>Note</td>
<td>TSO</td><td> 0</td><td> 0</td><td> 1</td><td>Transformation 2D</td>
<td>TS1</td><td> 0</td><td> -</td><td> 01</td><td>Transformation ID</td>
<td>TS2</td><td> -</td><td> 0</td><td> 001</td><td>Transformation ID</td>
<td>TS3</td><td> -</td><td> -</td><td> 000</td><td>Without transformation</td>
In transformation transformation transformation transformation transformation
<td>the board</td><td> 1,</td><td>TSO represents the</td><td>mode</td><td>of</td>
<td>2D.</td><td></td><td>TS1 represents the</td><td>mode</td><td>of</td>
<td colspan="2">horizontal,</td><td>and TS2 represents the</td><td>mode</td><td>of</td>
<td>vertical.</td><td></td><td>TS3 represents the</td><td>mode</td><td>without</td>
<td>In the</td><td colspan="2">present both in the</td><td>mode</td><td>of</td>
Horizontal as vertical transformation mode can correspond to ID transformation mode.
ΙΜΡΙ
MEXICAN INSTITUTE • ΐ IA INDUSTRIAL PROPERTY
For example, referring to Table 1, if the mode
<img file="MX339392B_D0038.tif" />
2D transformation occurs more frequently, 2D transformation mode can be assigned a code word 1.
Likewise, according to the frequency, the horizontal transformation mode can be assigned a code word 01, the vertical transformation mode is assigned a code word 001, and the non-transformation mode is assigned a code word 000 .
Even when vertical transformation and / or horizontal transformation are omitted depending on the transformation jump modes, the same quantization matrix can be applied as in the 2D transformation mode. In this case, the coding apparatus and the decoding apparatus can carry out the enlargement in the values in rows and / or columns not subjected to transformation, which can be represented by the
Equation 1.
Equation 1 y = (x * Scale + deviation) >> displacement Here, x can be an element in a row and / or column without transformation, and can be an increased value. Scale can be a scale factor. Deviation can be a deviation value that is applied on the scale, and displacement can be a value of
<img file="MX339392B_D0039.tif" />
IMPI
INSTITUTO MÍMICA NO OE LA PR «PIE» A »
INDUSTRIAL bit shift applied on the scale. Here, offset and offset can have the same values as an offset value and an applied bit transfer value when the transformation is not skipped, for example, in 2D transformation mode.
Furthermore, in Equation 1, the scale factor applied to the encoding apparatus and the decoding apparatus can be determined as a function of a size of
YOU. In an illustrative embodiment, the scale factor according to TU size can be adjusted as indicated in Table 2.
Table 2
<td>N</td><td> 4</td><td> 8</td><td> 16</td><td> 32</td>
<td>Scale</td><td> 128</td><td> 181</td><td> 256</td><td> 362</td>
Herein, N (and / or NxN) can be of a TU size, and scale can be a scale factor. Referring to Figure 2, when a TU is 8x8 in size, a scale factor value of 181 can be applied.
As mentioned above, one of the PU may not necessarily have a square shape, but has a rectangular shape. For example, in the inter mode, a PU
IMPI
MEXICAN INSTITUTE OF THE PKOEIFDAI)
INnuíTRIAL
<img file="MX339392B_D0040.tif" />
it can have a size (and / or shape) of 2N * N, N * 2N, 2NxnU, 2NxnD, nLx2N or nRx2N. In SDIP, a PU can have a size (and / or shape) of 2N * (L / 2) N or (112) N * 2N. In this case, since a transformation jump mode is less likely to happen, the encoding apparatus and decoding apparatus may not use the less probable transformation jump mode as a transformation jump mode candidate, thus improving encoding / decoding performance. Alternatively, the encoding apparatus may assign a short codeword to the least likely transformation hop mode, thereby improving encoding / decoding performance. Consequently, a method for determining a transformation jump mode candidate and a method for assigning a codeword for a transformation jump mode according to a size (and / or shape) of PU can be provided.
Figure 6 illustrates a method of determining a jump candidate transformation mode and a method of assigning a codeword to a transformation jump mode according to a PU mode in the Intermode.
Figure 6 schematically shows a size (and / or shape) of the PU in the intermode. Referring to Figure 6, a UC 610 can be divided into different PU sizes according to the properties of an image and
IMPI
<img file="MX339392B_D0041.tif" />
Similar. Figure 6 shows that a CU 610 cTIvrdé err a plurality of PU 620 units in inter prediction. In the inter mode, the PU can have sizes (and / or shapes) of 2N *
2N 621, 2 N * N 622, N * 2 N 623, N * N 624, 2NxnU 625, 2NxnD
626, nLx2N 627 or nRx2N 628. Herein, a PU with a size (and / or shape) N * N 624 can be used only for one
SCU as a minimum CU to avoid redundant calculations to calculate prediction costs.
Meanwhile, in the inter mode, the probabilities of the horizontal transformation mode and the vertical transformation mode may vary in the PU shapes. Therefore, different code words can be assigned to transform jump modes (and / or transform jump mode indices) depending on the PU shapes. That is, the code words assigned to the transform jump modes / and / or transform jump mode indices) can be determined based on the PU shapes.
In an illustrative embodiment, when one of the PU has a shape of N * 2N 623, the horizontal transformation energy compaction effect may be less than the vertical transformation compaction effect energy. Therefore, the vertical transformation mode may have a higher probability than the horizontal transformation mode. In Table 1, the transformation mode
IMPI
<img file="MX339392B_D0042.tif" />
horizontal is assigned the code word ¿i ¿ii Hp vertical transformation is assigned the code word 001, that is, more likely a transformation codeword is assigned a time code word.
Therefore, on the PU in the form of N * 2N 623, the codeword for horizontal transform mode and the codeword for vertical transform mode are reset, thus improving encoding performance. Table 3 illustrates a codeword allocation method for transformation jump modes in the PU in the form of N * 2N 623 according to an illustrative embodiment.
Table 3
IMPI institut »Mexican OF EA INDUSTRIAL PROPERTY
<td>TSM</td><td>Transformation row</td><td>Transformation column</td><td>Word of code (CABAC me CAVLC)</td><td>Note</td>
<td>TSO</td><td> 0</td><td> 0</td><td> 1</td><td>Transformation 2D</td>
<td>TS1</td><td> 0</td><td> -</td><td> 01</td><td>Transformation ID</td>
<td>TS2</td><td> -</td><td> 0</td><td> 001</td><td>Transformation ID</td>
<td>TS3</td><td> -</td><td> -</td><td> 000</td><td>Without transformation</td>
<td>In</td><td>the board</td><td>3, TSO represents</td><td>the</td><td>mode</td><td>of</td>
<td>transformation</td><td>2D.</td><td>TS1 represents</td><td>the</td><td>mode</td><td>of</td>
<td>transformation</td><td colspan="2">horizontal, and TS2 represents</td><td>the</td><td>mode</td><td>of</td>
<td>transformation</td><td>vertical.</td><td>TS3 represents the</td><td colspan="2">modality</td><td>without</td>
<td>transformation</td><td>In the</td><td>present both in</td><td>the</td><td>mode</td><td>of</td>
<td>transformation</td><td>horizontal</td><td>like the way</td><td colspan="3">transformation</td>
vertical can correspond to an ID transformation mode.
<img file="MX339392B_D0043.tif" />
•> · --TfT,, T> MEXICAN Γ · '.Λ MOPIEDAD «I<sup>1</sup>* ISTRIAL
Referring to Table 3, a.1— Hp horizontal transformation mode can be assigned a code word 001, and vertical transformation mode can be assigned a code word 01. As previously described, in the PU with the shape of N * 2N 623, the vertical transformation mode can have a higher probability than the horizontal transformation mode, and therefore the vertical transformation mode can be assigned a more code shorter than horizontal transformation mode.
<td></td><td>Even if</td><td>the</td><td>Table 3</td><td>I know</td><td>describes on the basis</td><td>of the</td>
<td>PU with</td><td>shape</td><td>of</td><td>N * 2N</td><td> 623,</td><td>the present invention</td><td>I dont know</td>
<td>limited to</td><td colspan="2">it. By</td><td>example,</td><td>in</td><td>a PU with a fofll!</td><td>L of</td>
<td>nLx2N 62 7</td><td>or 628</td><td colspan="3">nRx2N in addition</td><td colspan="2">of N * 2 N 623, the mode of</td>
Vertical transformation may also have a higher probability than the horizontal transformation mode. Accordingly, the vertical transform mode can be assigned a shorter code than the horizontal transform mode.
On the other hand, in PUs with 2N * N 622, 2NxnU 625 and 626 2NxnD shapes, the horizontal transformation mode may have a higher probability than the vertical transformation mode. Consequently, the horizontal transformation mode can be assigned a shorter code than the vertical transformation mode. For example, in the PU with the forms 2N * N 622, 2NxnU 625 and 2NxnD
<img file="MX339392B_D0044.tif" />
IMPI
MEXICAN INSTITUTE OF INDUSTRIAL PROPERTY
626, the same codeword allocation method can be used as in Table 1.
While, as described above, since the probabilities of the horizontal transformation mode and the vertical transformation mode in the inter-mode may vary in PU shapes, a number of transformation jump mode candidates may be determined differently. on . the base of PU shapes. That is, the transformation hop mode candidates of a transformation destination block may be determined based on a PU shape corresponding to the transformation destination block.
In an illustrative embodiment, when one of the PU has a 2N * N 622 shape, the vertical transformation energy compaction effect may be less than the horizontal transformation energy compaction effect, and therefore the transformation mode vertical may have a lower probability than horizontal transformation mode. Therefore, in the PU with the form 2 N * N 622, the 2D transformation mode, the horizontal transformation mode and the no transformation mode can be used as transformation jump mode candidates of a transformation target block , excluding the vertical transformation mode. In this case, a jump mode transfer between the
<img file="MX339392B_D0045.tif" />
IMPI
MEXICAN INSTITUTE OF INDUSTRIAL PROPERTY 2D transformation, horizontal transformation mode and no transformation mode can be applied to the transformation destination block. Table 4 illustrates a method of assigning code words to transformation jump mode when using 2D transformation mode, horizontal transformation mode, and no transformation mode as transformation jump mode candidates according to a illustrative modality.
Table 4
<td>TSM</td><td>Transformation row</td><td>Transformation column</td><td>Word of code (CABAC me CAVLC)</td><td>Note</td>
<td>TSO</td><td> 0</td><td> 0</td><td> 0</td><td>Transformation 2D</td>
<td>TS1</td><td> 0</td><td> -</td><td> 10</td><td>Transformation ID</td>
<td>TS3</td><td> -</td><td> -</td><td> 11</td><td>Without transformation</td>
IMPI
<img file="MX339392B_D0046.tif" />
In Table 4, TSO represents the 2D transformation mode, TS1 represents the horizontal transformation mode and TS3 represents the untransformed mode. Herein, the horizontal transformation mode can be used may correspond to an ID transformation mode. With reference to Table 4, in the PU with the form 2N * N
622, 2D transform mode, horizontal transform mode, and no transform mode as transformation jump mode candidates.
Although Table 4 is described based on PU in the form of 2N * N 622, the present invention is not limited thereto. For example, in PU with shapes of 2NxnU 625 and 2NxnD
626 In addition to 2N * N 622, the vertical transformation mode may also have a lower probability than the horizontal transformation mode. Consequently, 2D transformation mode, horizontal transformation mode, and no transformation mode can be used as transformation jump mode candidates for a transformation target block, excluding vertical transformation mode.
Alternatively, in the PU with the shape of N * 2N
623, since the horizontal transformation energy compaction effect may be less than the vertical transformation energy compaction effect, the horizontal transformation mode may have a
IMPI
<img file="MX339392B_D0047.tif" />
probability lower than transformation mode I poured
Therefore, in the PU with the shape of N * 2 N 623, the 2D mode, the vertical transformation mode and the no transformation mode can be used as candidates for
<td colspan="2">5 transformation jump for</td><td>a block</td><td>of</td><td>destination</td><td>of</td>
<td>transformation</td><td>transform,</td><td>excluding</td><td>the</td><td>mode</td><td>of</td>
<td>transformation</td><td colspan="2">horizontal. In this case, a</td><td>mode</td><td>of jump</td><td>of</td>
<td>transformation</td><td>between the mode of</td><td colspan="2">2D transformation,</td><td>The mode</td><td>of</td>
vertical transformation and no transformation mode can be applied to the transformation target block. Table 5 illustrates a method of assigning code words to transformation jump modes when 2D transformation mode, vertical transformation mode, and no transformation mode are used as transformation jump mode candidates according to a illustrative modality.
IMPI
<img file="MX339392B_D0048.tif" />
Table 5
<td>TSM</td><td>Transformation row</td><td>Transformation column</td><td>Word of code (CABAC me CAVLC)</td><td>Note</td>
<td>TSO</td><td> 0</td><td> 0</td><td> 0</td><td>Transformation 2D</td>
<td>TS2</td><td> -</td><td> -</td><td> 10</td><td>Transformation ID</td>
<td>TS3</td><td> -</td><td> -</td><td> 11</td><td>Without transformation</td>
In Table 5, TSO represents the 2D transformation mode, TS2 represents the vertical transformation mode, and TS3 represents the no transformation mode. Herein, the vertical transformation mode may correspond to an ID transformation mode. Referring to Table 5, in the PU in the form of N * 2 N
623, 2D transformation mode, vertical transformation mode, and no transformation mode can be used as transformation jump mode candidates.
ΙΜΡΙ
MEXICAN INSTITUTE Dt THE INDUSTRIAL PROPERTY
<img file="MX339392B_D0049.tif" />
Although Table 5 is described on the basis of leF PU in the form of N * 2N 623, the present invention is not limited thereto. For example, in a PU with a shape of nLx2N 627 or nRx2N 628 in addition to N * 2 N 623, the horizontal transformation mode may also have a lower probability than the vertical transformation mode. Consequently, 2D transformation mode, vertical transformation mode, and no transformation mode can also be used as transformation jump mode candidates for a transformation target block, excluding horizontal transformation mode.
<td>In the</td><td colspan="5">previous illustrated modalities in</td><td>the</td>
<td>Tables 3 to 5,</td><td>I know</td><td>can reduce</td><td colspan="2">the numbers</td><td>of</td><td>bits</td>
<td>Used for</td><td>the</td><td>coding</td><td>modes</td><td>of</td><td>jump</td><td>of</td>
<td>transformation</td><td>(me</td><td>the indices of</td><td>mode</td><td>of</td><td>jump</td><td>of</td>
<td>transformation).</td><td colspan="2">Consequently,</td><td>can</td><td colspan="2">to get well</td><td>the</td>
encoding / decoding performance.
<td></td><td>Fight.</td><td>Figure 7</td><td>illustrates a</td><td>method</td><td>for</td><td colspan="2">decide</td><td>a</td>
<td>way of</td><td colspan="3">transformation jump</td><td colspan="2">candidate and</td><td>a</td><td>method</td><td>of</td>
<td>to assign</td><td>a</td><td>word</td><td>code</td><td>yet</td><td>mode</td><td>of</td><td>jump</td><td>of</td>
transformation according to a form of PU in SDIP.
Figure 7 schematically shows a size (and / or shape) of PU in SDIP. Referring to Figure 7, a
UC 710 can be divided into different PU sizes according to the properties of an image and the like. The
ΙΜΡΙ
<img file="MX339392B_D0050.tif" />
<img file="MX339392B_D0051.tif" />
industrial
<img file="MX339392B_D0052.tif" />
of PU 720 units in SDIP. In SDIP, U's can have sizes (and / or shapes) of 2N * 2N 721, N * N 723, (1/2) N * 2N 725, or 2N * (1/2) N 727. Herein, a PU with a size (and / or shape) N * N 723 can be used only for one SCU as a minimum CU to avoid redundant calculations for prediction cost calculation.
In SDIP, since the probabilities of horizontal transformation mode and vertical transformation mode can vary in PU shapes, a number of transformation jump mode candidates can be determined differently based on shapes of
PU. That is, transformation hop mode candidates of a transformation destination block may be determined based on a shape of the PU corresponding to the transformation destination block.
In an exemplary embodiment, when one of the PU has a 2N * (1/2) N 727 shape, the energy compaction effect of the transformation vertical may be less than the transformation energy compaction effect horizontal, and therefore the vertical transformation mode may have a lower probability than the horizontal transformation mode. Therefore, in the
PU with the form 2N * (L / 2) N 727, the transformation mode
2D, horizontal transformation mode and without mode
<img file="MX339392B_D0053.tif" />
IMPI
MEXICAN INSTITUTE • f THE INDUSTRIAL PROPERTY transformation can be used as transformation jump mode candidates for a transformation destination block, excluding vertical transformation mode. In this case, a jump mode transfer between 2D transform mode, horizontal transform mode, and no transform mode can be applied to the transform target block. A codeword mapping method for transform jump modes when 2D transform mode, horizontal transform mode, and no transform mode are used since the transform jump mode candidates have been previously described in Table 4, and therefore a description thereof is omitted herein.
Alternatively, in a PU with a shape of (1/2) N * 2N 725, since horizontal transformation energy compaction effect may be less than the vertical transformation energy compaction effect, the horizontal transformation mode may have a lower probability than the vertical transformation mode.
Therefore, in the PU with the form (1/2) N * 2N 725, the 2D transformation mode, the vertical transformation mode and the no transformation mode can be used as transformation jump mode candidates for a block transformation target, excluding transformation mode
ΪΜΡΙ
Mexican Institute of Industrial Property
<img file="MX339392B_D0054.tif" />
horizontal. In this case, the transformation jump mode enters the 2D transformation mode. vertical transformation mode, and no transformation mode can be applied to the transformation target block. A codeword mapping method of transformation jump modes when 2D transformation mode, vertical transformation mode, and no transformation mode are used as transformation jump candidates which has been described above in Table 5 , and therefore a description thereof is omitted herein.
In the above embodiments, the number of bits used for encoding transform jump modes (and / or transform jump mode indices) may be reduced. Consequently, the encoding / decoding performance can be improved.
<td></td><td>The</td><td>Figure 8</td><td>illustrates</td><td>a</td><td>method</td><td>of</td><td>assignment of a</td>
<td>word</td><td>of</td><td>code a</td><td>a way</td><td>of</td><td>jump</td><td>of</td><td>transformation of</td>
<td>agreement</td><td>with</td><td colspan="2">an address of</td><td colspan="2">prediction</td><td>in</td><td>the intra mode.</td>
As described above with reference to Figures 1 and 2, the encoding apparatus and decoding apparatus can generate a prediction block using the intra prediction mode based on the information of a pixel within a current image. Intra prediction can be performed according to an intra mode
IMPI
MEXICAN INSTITUTE Say THE INDUSTRIAL PROPERTY
<img file="MX339392B_D0055.tif" />
prediction for a prediction target block. The intra prediction mode can include a DC mode, a flat mode, a vertical mode, a horizontal mode, and an angular mode. DC mode and flat mode are non-directional modes and the other modes are directional modes. Herein, the angular mode may be a directional prediction mode other than portrait mode and landscape mode.
Figure 8 illustrates a prediction direction, an intra prediction mode and a mode value assigned to each prediction direction. In Figure 8, each intra prediction mode has a different prediction direction. The numbers assigned to each of the intra prediction modes can be referred to as mode values.
Referring to Figure 8, an intra prediction mode with a mode value of 0 can be referred to as a flat mode. In plane mode, the reference pixels used for predicting a pixel value of a prediction target pixel can be determined based on a location of the prediction target pixel in a prediction target block. A prediction value of the prediction target pixel can be derived based on the determined reference pixels. An intra prediction mode with a mode value of 1 can be referred to as a DC mode, where a prediction block can be generated using an average pixel value
IMPI
MEXICAN INSTITUTE D € THE INDUSTRIAL PROPERTY
<img file="MX339392B_D0056.tif" />
of neighboring pixels for the prediction block Hpatino. In an intra prediction mode with a mode value of 26, the prediction can be carried out in the vertical direction based on the pixel values of neighboring blocks. Therefore, the intra prediction mode with the mode value of 26 can also be referred to as the vertical mode. In an intra prediction mode with a mode value of 10 (landscape mode, prediction can be carried out in the horizontal direction based on the pixel values of neighboring blocks. Therefore, the intra prediction mode with the mode value of 10 can also be referred to as the horizontal mode In the other modes, the prediction can be made based on pixel values of neighboring blocks according to corresponding angles.
The probabilities of the horizontal transformation mode and the vertical transformation mode can vary in an intra prediction mode (and / or the prediction direction) of one of the PU that corresponds to a transformation destination block. Therefore, a different codeword can be assigned to a transformation jump mode (and / or transformation jump mode index) based on the intra prediction mode (and / or the prediction direction) of the PU . That is, a codeword assigned to a jump transform mode (and / or
IMPI
MEXICAN INSTITUTE OF INDUSTRIAL PROPERTY
<img file="MX339392B_D0057.tif" />
skip mode mode index) nc puaele<sup>1 </sup>determine based on the intra prediction mode (and / or the prediction direction) of the PU corresponding to the transformation destination block.
In an exemplary embodiment, when the intra prediction mode of the PU is the vertical mode, the horizontal transformation energy compaction effect may be less than the compaction effect energy of the vertical transformation. Therefore, in this case, the vertical transformation mode may have a greater probability than the horizontal transformation mode. In the mode illustrated with reference to Table 1, the horizontal transformation mode is assigned the code word 01 and the vertical transformation mode is assigned the code word 001, that is, a transformation jump mode with the more likely you are assigned a code word. Therefore, when the PU intra prediction mode is vertical transform mode, the codeword for horizontal transform mode and the codeword for vertical transform mode are reset, thereby improving encoding performance. That is, when the PU intra prediction mode is the vertical transformation mode, the vertical transformation mode may have a higher probability than the horizontal transformation mode, to the vertical transformation mode
IMPI
MEXICAN INSTITUTE AND THE INDUSTRIAL PROFTSBAD
<img file="MX339392B_D0058.tif" />
it can be assigned a shorter code than horizontal transform mode. A shorter codeword assignment mode for vertical transformation mode than for horizontal transformation mode is similar to the modality illustrated in Table 3, and therefore a description thereof is omitted herein. document.
Alternatively, when the PU intra prediction mode corresponding to the target block of is the horizontal transformation mode, the horizontal transformation mode may have a higher probability than the vertical transformation mode. Therefore, in this case, the horizontal transformation mode can be assigned a shorter code than the vertical transformation mode. For example, when the PU intra prediction mode for the transformation target block is landscape mode, the same codeword allocation method can be used as in Table 1.
Figure 9 schematically illustrates a transformation coefficient scanning method based on a transformation jump mode in accordance with an illustrative embodiment of the present invention.
Figure 9 shows horizontal scan 910, vertical scan 920 and zigzag scan 930 according to an illustrative embodiment. Although Figure 9
IMPI
INSTITUT · MEXICANC OE LA MOWEDAP INDUSTRIAL illustrates a method of exploration (and / or order of exploration)
<img file="MX339392B_D0059.tif" />
For a 4x4 block only, such a method can be applied regardless of, but not limited to, block sizes.
In the embodiment of Figure 9, the reverse scan may also be referred to as a sweep for convenience of description, as needed, which will be readily understood by a person having ordinary skill in the art.
As previously described in Figure 1, the coding apparatus can perform two-dimensional (2D) block scanning of transformation coefficients quantized into a one-dimensional (ID) vector of the transformation coefficient in order to improve efficiency in entropy coding. Also, as described above in Figure 2, the decoding apparatus can generate a 2D transform coefficient block by scanning a vector of decoded identification transform coefficients.
Herein, the encoding apparatus and the decoding apparatus may determine a scanning method (and / or scanning order) based on a transformation jump mode. That is, according to illustrative embodiments of the present invention, different
IMPI
INSTITUTO MEXICANO DF LA PROPIEDAD INDUSTRIAL
<img file="MX339392B_D0060.tif" />
Drill methods (and / or Drill Orders) can be used based on a transformation jump mode to transform a target block.
In an illustrative embodiment, when the transformation jump mode is the horizontal transformation mode, the residual signals are more likely to stay in the vertical direction. Therefore, when the transformation jump mode to transform the transformation target block, the horizontal transformation mode, vertical scan 920 can be used for the transformation target block. When transform jump mode is vertical transform mode, the residual signal is more likely to stay in the horizontal direction. Therefore, when the transform jump mode for transforming the target block is vertical transform mode, 910 horizontal scan can be used for the transformation target block. Transformation jump mode other than horizontal transformation mode and vertical transformation mode, zigzag 930 scanning can be used to perform scanning.
FIG. 10 is a flow diagram schematically illustrating a coding method in accordance with an illustrative embodiment of the present invention.
• nstitutomexicano BE THE PROPERTY
INDUSTRIAL
Referring to Figure 10, the coding apparatus can generate a residual block corresponding to a current block (S1010). As described above, the coding apparatus can perform inter prediction and / or intra prediction in the current block, thereby generating a prediction block corresponding to the current block. Herein, the coding apparatus can generate a residual signal, ie, the residual block, differentiated by one pixel between a pixel value of the current block and a pixel value of the prediction block.
In Figure 10, the coding apparatus can transform the residual signal, ie the residual block (S1020). The encoding apparatus can encode the residual signal by applying a transformation core, and a transcoding core can be 2 * 2, 4 * 4, * 8, 16 * 16, 32 * 32 or 64 * 64 in size. In an illustrative embodiment, a transformation coefficient C during an N * N block can be calculated by Equation 2.
Equation 2
<td>C (n, n)</td><td>= T (n, n) x B (n,</td><td>n)</td><td>x T</td><td>(n,</td><td>n)<sup>T</sup></td>
<td>In the</td><td>present, C (n,</td><td>n)</td><td>is</td><td colspan="2">an array of</td>
<td>transformation of</td><td>coefficients n * n,</td><td>T</td><td>(n,</td><td>n)</td><td>is a matrix</td>
<td colspan="2">transformation core n * n, and</td><td>B</td><td>(n,</td><td>n)</td><td>is a matrix</td>
of a residual block of n * n.
+
IMPI
M EXICANO INSTITUTE • E INDUSTRIAL PROPERTY
<img file="MX339392B_D0061.tif" />
When a transformation coefficient is generated through transformation, the encoding apparatus can quantify the generated transformation coefficient.
It can be determined through RDO that is transmitted between the residual block and the transformation coefficient. When the prediction is done correctly, the residual block, that is, the residual signal, can be transmitted without transcoding. The coding apparatus can compare the cost functions before / after transcoding and select a method that involves minimal costs. Herein, the encoding apparatus may transmit information about a type of a transmission signal (residual signal or transformation coefficient) relative to the current block for the decoding apparatus.
The transformation processes have been illustrated in the previous modalities, and therefore their descriptions are omitted in the present one.
Referring again to Figure 10, the coding apparatus can examine the transformation coefficient (S1030). Herein, as described above, the coding apparatus can determine a scanning method (and / or scanning order) based on a transformation jump mode. A method of determining a scan order based on a transformation jump mode has been described above, and therefore a description of the
IMPI
MEXICAN INSTITUTE • E LA MONEDAD INDUSTRIAL
<img file="MX339392B_D0062.tif" />
They are omitted herein.
When the scan is carried out, the entropy coding apparatus can code the scanned coefficient of transformation and the lateral information (eg, information on an inter-prediction mode of the current block) (S1040). The encoded information can be formed in a compressed bitstream and is transferred or stored through a NAL.
Although the coding method is described in a series of steps based on the flowchart in Figure 10, the present invention is not limited thereto. Some steps in Figure 10 can be carried out in a different order from that described above or in parallel. Furthermore, additional steps may be included among the stages of the flow chart, or one or more steps may be removed from that of the flow chart of Figure 10 within the scope of the present invention.
FIG. 11 is a flow chart schematically illustrating a decoding method in accordance with an illustrative embodiment of the present invention.
Referring to Figure 11, the entropy decoding apparatus can decode a bit stream received from the encoding apparatus (S1110). For example, the decoding apparatus may derive a mode
<img file="MX339392B_D0063.tif" />
IMPI · η; Τ3 MEXICAN Dñ NtONÍD * »INDUSTRIAL prediction and a residual signal of a block current based on a variable length table encoding (VLC) and / or CABAC. The decoding apparatus can obtain information on whether a received signal with respect to the current block is the residual signal or a transformation coefficient and obtain the residual signal or a transformation coefficient identification vector for the current block. When the received bitstream includes lateral information necessary for decoding, both the bitstream and the supplemental information may be decoded by entropy.
In Figure 11, the decoding apparatus can reverse examine the residual signal decoded by entropy or transformation coefficients to generate a 2D block (S1120). Herein, a residual block can be generated in the case of the residual signal, and a 2D transform coefficient block can be generated in the case of the transform coefficients. When the transform coefficients are generated, the decoding apparatus can dequantize the generated transform coefficients.
As described above, in reverse scanning, the decoding apparatus may determine a scanning method (and / or scanning order) based on a transformation jump mode. Previously
IMPI
I '' TUTO MEXICANC
V PROPERTY '"niimiAi described a method of determining a scanning order based on a transformation jump mode and therefore a description thereof is omitted herein.
Referring again to Figure 11, the decoding apparatus can reverse transform the dequantized transform coefficients, thereby generating a residual block (S1130). The inverse transformation can be represented by Equation 3.
<img file="MX339392B_D0064.tif" />
Equation 3
B (η, η) = T (n, n) x C (η, η) x T (n, n)<sup>T</sup>
The reverse transformation has been described above and therefore a description thereof is omitted herein.
When the residual block is generated, the decoding apparatus can generate a reconstructed block based on the generated residual block (S1140). As described above, the decoding apparatus may perform inter prediction and / or intra prediction on a decoding destination block to generate a prediction block corresponding to the decoding destination block.
Herein, the decoding apparatus may merge a prediction block pixel value and a
IMPI
MSTITUTO MEXICANO nt the ntoniOAO INDUSTRIAL pixel value of the residual block by one pixel, thereby generating the reconstructed block.
Although the decoding method is described in a series of steps based on the flowchart in Figure 11, the present invention is not limited thereto. Some steps in Figure 11 can be carried out in a different order from that described above or in parallel. Furthermore, additional steps may be included among the stages of the flow chart, or one or more steps may be removed from the flow chart of Figure 11 within the scope of the present invention.
Although the methods have been described with a series of steps or blocks based on flow charts in the aforementioned embodiments, the present invention is not limited to the above sequence of steps. Some
<img file="MX339392B_D0065.tif" />
<td colspan="6">stages can be carried out in different order than described</td>
<td>previously</td><td>or in</td><td>parallel</td><td>to the</td><td>Same time.</td><td>Also I know</td>
<td>will understand by</td><td>the</td><td>experts</td><td>in</td><td>the technique that</td><td>the stages</td>
<td>illustrated in</td><td>the</td><td>diagrams</td><td>of</td><td>flow are not</td><td>exclusive,</td>
<td colspan="3">can be included in the</td><td colspan="2">Flowchart</td><td>the stages</td>
Additional, or one or more steps can be removed from the flow charts without affecting the scope of the present invention.
<B f,> Λ '' '/ υ <sup>τ</sup>'> *'. ALCANO <./ '«οη« Μΐ>' Nousnut
<img file="MX339392B_D0066.tif" />
The present invention has Yes Hq-Hpsrite with reference to illustrative embodiments and the above embodiments include various exemplary aspects.
Although all possible combinations cannot be mentioned to illustrate various aspects, it will be appreciated by those skilled in the art that changes, modifications and alternatives can be made in these illustrative embodiments without departing from the principles and spirit of the invention, the scope of which is defined in the appended claims and their equivalents.
Contents71
77 sheets
Sheet 1 Sheet 2 Sheet 3 Sheet 4 Sheet 5 Sheet 6 Sheet 7 Sheet 8 Sheet 9 Sheet 10 Sheet 11 Sheet 12 Sheet 13 Sheet 14 Sheet 15 Sheet 16 Sheet 17 Sheet 18 Sheet 19 Sheet 20 Sheet 21 Sheet 22 Sheet 23 Sheet 24 Sheet 25 Sheet 26 Sheet 27 Sheet 28 Sheet 29 Sheet 30 Sheet 31 Sheet 32 Sheet 33 Sheet 34 Sheet 35 Sheet 36 Sheet 37 Sheet 38 Sheet 39 Sheet 40 Sheet 41 Sheet 42 Sheet 43 Sheet 44 Sheet 45 Sheet 46 Sheet 47 Sheet 48 Sheet 49 Sheet 50 Sheet 51 Sheet 52 Sheet 53 Sheet 54 Sheet 55 Sheet 56 Sheet 57 Sheet 58 Sheet 59 Sheet 60 Sheet 61 Sheet 62 Sheet 63 Sheet 64 Sheet 65 Sheet 66 Sheet 67 Sheet 68 Sheet 69 Sheet 70 Sheet 71 Sheet 72 Sheet 73 Sheet 74 Sheet 75 Sheet 76 Sheet 77
94 members in 12 offices
Priority claims5
| Document | Office | Kind | Date |
|---|---|---|---|
| 102011016107 | Republic of Korea | – | |
| 20110106107 | Republic of Korea | A | |
| 1020110106107 | – | – | – |
| 102011016107 | – | – | – |
| KR20110106107 | – | – | – |
Members94
| Document | Office | Kind | |
|---|---|---|---|
| CA2856198A1 | Canada | A1 | |
| WO2013058542A1 | World Intellectual Property Organization (WIPO) | A1 | |
| AU2012326873A1 | Australia | A1 | |
| GB201407661D0 | United Kingdom | D0 | |
| KR20140077928A | Republic of Korea | A | |
| GB2510078A | United Kingdom | A | |
| MX2014004777A | Mexico | A | |
| KR20140111042A | Republic of Korea | A | |
| KR20140111043A | Republic of Korea | A | |
| US2014269915A1 | United States of America | A1 | |
| CN104081775A | China | A | |
| KR20140116973A | Republic of Korea | A | |
| KR20140116974A | Republic of Korea | A | |
| KR20140135243A | Republic of Korea | A | |
| CN104378640A | China | A | |
| IN3098DEN2014A | India | A | |
| US2015139306A1 | United States of America | A1 | |
| KR101533720B1 | Republic of Korea | B1 | |
| PL408822A1 | Poland | A1 | |
| KR20150091430A | Republic of Korea | A | |
| KR20150091431A | Republic of Korea | A | |
| KR20150091432A | Republic of Korea | A | |
| KR20150091433A | Republic of Korea | A | |
| KR101550723B1 | Republic of Korea | B1 | |
| KR101550724B1 | Republic of Korea | B1 | |
| KR101550725B1 | Republic of Korea | B1 | |
| KR101550726B1 | Republic of Korea | B1 | |
| CN105100805A | China | A | |
| RU2014117487A | Russian Federation | A | |
| AU2012326873B2 | Australia | B2 | |
| AU2016201699A1 | Australia | A1 | |
| AU2016201713A1 | Australia | A1 | |
| MX339392BThis record | Mexico | B | |
| AU2016247083A1 | Australia | A1 | |
| AU2016247085A1 | Australia | A1 | |
| RU2606066C2 | Russian Federation | C2 | |
| US9560384B2 | United States of America | B2 | |
| US9560385B2 | United States of America | B2 | |
| KR101718953B1 | Republic of Korea | B1 | |
| KR101718954B1 | Republic of Korea | B1 | |
| US2017099502A1 | United States of America | A1 | |
| US2017099503A1 | United States of America | A1 | |
| US2017099504A1 | United States of America | A1 | |
| BR112014009403A2 | Brazil | A2 | |
| US9661346B2 | United States of America | B2 | |
| US9661352B2 | United States of America | B2 | |
| US9661354B2 | United States of America | B2 | |
| AU2016201699B2 | Australia | B2 | |
| AU2016201713B2 | Australia | B2 | |
| US2017223380A1 | United States of America | A1 | |
| CA2856198C | Canada | C | |
| CN104081775B | China | B | |
| GB201713539D0 | United Kingdom | D0 | |
| GB201713541D0 | United Kingdom | D0 | |
| GB201713548D0 | United Kingdom | D0 | |
| GB201713557D0 | United Kingdom | D0 | |
| CN107257457A | China | A | |
| CN107257475A | China | A | |
| CN107343204A | China | A | |
| CN107360421A | China | A | |
| US9826251B2 | United States of America | B2 | |
| CN104378640B | China | B | |
| GB2551086A | United Kingdom | A | |
| GB2551087A | United Kingdom | A | |
| GB2551088A | United Kingdom | A | |
| GB2551290A | United Kingdom | A | |
| CN107483930A | China | A | |
| CN107493479A | China | A | |
| RU2646307C1 | Russian Federation | C1 | |
| RU2648605C1 | Russian Federation | C1 | |
| RU2648607C1 | Russian Federation | C1 | |
| GB2510078B | United Kingdom | B | |
| CN105100805B | China | B | |
| KR101857109B1 | Republic of Korea | B1 | |
| KR101857110B1 | Republic of Korea | B1 | |
| KR101880642B1 | Republic of Korea | B1 | |
| GB2551086B | United Kingdom | B | |
| GB2551087B | United Kingdom | B | |
| GB2551088B | United Kingdom | B | |
| GB2551290B | United Kingdom | B | |
| AU2016247083B2 | Australia | B2 | |
| AU2016247085B2 | Australia | B2 | |
| RU2016140545A | Russian Federation | A | |
| PL230821B1 | Poland | B1 | |
| CN107343204B | China | B | |
| GB2510078C | United Kingdom | C | |
| RU2016140545A3 | Russian Federation | A3 | |
| RU2715031C2 | Russian Federation | C2 | |
| CN107360421B | China | B | |
| CN107483930B | China | B | |
| CN107257457B | China | B | |
| CN107257475B | China | B | |
| CN107493479B | China | B | |
| BR112014009403B1 | Brazil | B1 |
Numbers
- Publication
- 339392
- Publication, DOCDB
- 339392
- Publication, EPODOC
- MX339392
- Application
- 2015014510
- Application, DOCDB
- 2015014510
- Application, EPODOC
- MX20150014510
Titles
- Spanish
- METODO Y APARATO PARA CODIFICAR/DECOFICAR IMAGENES.
Classification
- CPC, 22
- H04N19/11
- H04N19/12
- H04N19/122
- H04N19/46
- H04N19/503
- H04N19/44
- H04N19/60
- H04N19/186
- H04N19/91
- H04N19/124
- H04N19/159
- H04N19/176
- H04N19/61
- H04N19/147
- H04N19/103
- H04N19/513
- H04N19/70
- H04N19/90
- H04N19/117
- H04N19/86
- H04N19/13
- H04N19/172
- IPC, 6
- H04N19 60
- H04N19 124
- H04N19 503
- H04N19 513
- H04N19 61
- H04N19 91