Method of applying edge offset.
Abstract
Provided is a methodgenerates an edge index of a current sample, and applies an edge offset corresponding to the edge index to the current sample. The edge index is generated using the differences between a current sample and two neighboring samples determined by an edge offset type. Accordingly, the difference between original samples and reconstructed samples are effectively reduced by generating the optimum edge index. Also, the quantity of bits required for reducing the differences are reduced by fixing the sign of offset to positive or negative.

Term
6.3 yearsleft in the term
Expires 8 January 2033.
- Priority
- Filed
- Granted
- Today
- Expires
10 claims: 2 independent, 8 dependent
- 1REIVINDICACIONES IMPI INSTITUTO MEXICANO DE LA PROPIEDAD INDUSTRIAL 1. Un método de obtención de una imagen reconstruida, el método comprende:derivar un bloque de predicción en base a un modo de predicción;escanear inversamente información de coeficientes cuantificados para generar un bloque cuantificado;cuantificar inversamente el bloque cuantificado para generar un bloque de transformación usando un parámetro de cuantificación;transformar inversamente el bloque de transformación para generar bloque residual;generar la imagen reconstruida utilizando el bloque de predicción y el bloque residual;determinar una intensidad de límite para cada borde de 4 muestras en la imagen reconstruida, que es un borde de predicción o un borde de transformación y cae en rejilla de muestra de 8x8;determinar si el filtrado de desbloqueo se aplica sobre el borde de 4 muestras o no usando la intensidad de límite y un parámetro de cuantificación de límite;filtrar el borde de 4 muestras si el filtrado de desbloqueo se aplica sobre el borde de 4 muestras;y aplicar una compensación de borde si un tipo de compensación de adaptación de muestra (SAO) indica una IMPI INSTITUTO MEXICANO DE LA PROPIEDAD compensación de borde, en donde la etapa de apiWsfí r compensación de borde comprende: generar un indi ce. 'borde de una muestra actual;y aplicar una compensación de borde correspondiente al índice de borde para la muestra actual, en donde el índice de borde se genera utilizando la siguiente ecuación, bordeldx=2+signo3(reclmagen(x)-reclmagen(x-1)) +signo3(reclmagen(x)-reclmagen(x+1)) igual a 0, y en donde el recPicture variable (x) representa el valor de muestra actual, las variables recPicture (x-1) y recPicture (x + 1) representan los dos valores de muestra vecinos, en donde el parámetro de cuantificación se genera utilizando un predictor de parámetro de cuantificación y un parámetro de cuantificación diferencial, cuando parámetros de cuantificación izquierdo y arriba están disponibles, el predictor de parámetro de cuantificación se genera utilizando parámetros de cuantificación izquierdo y arriba.
- 2El método de la reivindicación 1, en donde las dos muestras adyacentes se determinan por el tipo de compensación de borde de un área actual.
- 3El método de la reivindicación 1, en donde la compensación de borde se establece como negativa o positiva con base en el índice de borde.
- 4El método de la reivindicación en dónde,' la compensación de borde se establece como positiva si el índice de borde es igual a 0 o 1.
- 5El método de la reivindicación 3, en donde la compensación de borde se establece como negativa si el índice de borde es igual a 3 o 4.
- 6El método de la reivindicación 1, en donde la compensación de borde se establece en 0 si el índice de borde es igual a 2.
- 7El método de la reivindicación 1, en donde si una de las dos muestras adyacentes pertenece a otra LCU, la compensación de borde no se aplica a la muestra actual.
- 8El método de la reivindicación 1, en donde si una de las dos muestras adyacentes pertenece a otra LCU, la muestra adyacente que pertenece a otra LCU se reemplaza con otra muestra dentro de la LCU.
- 9El método de la reivindicación 1, en donde el tipo SAO se especifica mediante una LCU cuando el SAO está disponible.
- 10El método de la reivindicación 1, en donde, cuando sólo uno de los parámetros de cuantificación izquierdo y arriba se encuentra disponible, el predictor de parámetro de cuantificación se genera utilizando el parámetro de cuantificación disponible y un parámetro de cuantificación anterior. IMPI INSTITUTO MEXICANO DI LA PROPIEDAD industrial
Independent claims10
284 paragraphs in 47 sections, as filed
(54) Title: METHOD FOR APPLYING EDGE COMPENSATION. (54) Title: METHOD OF APPLYING EDGE OFFSET.
(57) Summary
The provided method generates the edge index of a current sample, and applies the edge compensation corresponding to the edge index to the current sample. The edge index is generated using the differences between a current sample and two adjacent samples determined by the type of edge compensation. Consequently, the difference between the original sample and the reconstructed samples is effectively reduced by generating the optimal edge index. Also, the number of bits required to reduce the differences is reduced, by setting the offset sign to positive or negative.
(57) Abstract
Provided is a methodgenerates an edge index of a current sample, and applies an edge offset corresponding to the edge index to the current sample. The edge index is generated using the differences between a current sample and two neighboring samples determined by an edge offset type. Accordingly, the difference between original samples and reconstructed samples are effectively reduced by generating the optimum edge index. Also, the quantity of bits required for reducing the differences are reduced by fixing the sign of offset to positive or negative.
_SE_ «CRfTAKU ¡X ΜΧΜβΜίΑ
<img file="MX340366B_D0001.tif" />
Mexican Institute of Industrial Property
<img file="MX340366B_D0002.tif" />
PATENT TITLE NO. 340366
Holder (s): INFOBRIDGE PTE. LTD.
Address: 10 Anson Road # 23-140 International Plaza, Singapore, 079903, SINGAPORE
Name: METHOD FOR APPLYING EDGE COMPENSATION.
Classification: lnt.CI.8: H04N19 / 117; H04N19 / 14; H04N19 / 176
Inventor (s): MIN JANG
REQUEST
Number: International filing date:
MX / a / 2015/009937 January 8, 2013
Divisional Patent Number: 332102
PRIORITY
Country: Date: Number:
KR January 17, 2012 10-2012-0005334
Validity: Twenty years
Expiration Date: January 8, 2033
The reference patent is granted based on articles 1, 2 fraction V, 6 fraction III, and 58 of the dgffc Industrial Property Law.
In accordance with article 23 of the Industrial Property Law, this patent has a non-renewable term of twenty years, counted from the filing date of the international application and will be subject to the payment of the fee to keep the rights in force. .
-Those who subscribe to the present title do so based on the provisions of articles 8 “sections M and 7 ° bis 2 of the Industrial Property Law (Official Federation Document (DOF) 06/27/1991, amended to 08/02/1884,25 / 10/1996, 12/25/1387, 05/17/1999, 01/26/2004, 06/16/2003, 01/25/2006, 06/05/2009 / 08 / 01/2010, 08/18/2010, 28/06 / 2n-0, 01/27/2012 and 09> O4 / 2O12), Articles I °, 3rd section V subsection a), subsection iii) 4th and 12 * fractions i and lll ofMhgiainento of the Menieane Institute of Industrial Property (DO F. 12/14/1999, amended on 07/01/2008, 07/15/2004, 07/28/2004 and 09/07/2807); articles 1 ·, 3f>, 4 ·, 8 section V subsection a), sub subsection iii), 16 sections I and lll and 30 of the Organic Statute of the Mexican Institute of Industrial Property (DO F 27/12/1999, reformed the 10/10/2002, 07/29/2004, 04/08/2004 and 09/13/2007); 1st, 3rd and 5th paragraph a) and antepenultimate paragraph of the Agreement that delegates powers to the General Appointed Directors, Coordinator, Divisional Directors, Titulars of. the Rggmiales Offices, Divisional Deputy Directors, Departmental Coordinators and other subordinates of the Mexican Institute of Industrial Property. (D: O * F. 12/15/1999, Tétórmadb on 02/04/2000, 07/29/2004, 08/04/2004 and 09/13/2007).
Issue Date: July 7, 2016
DIVISIONAL DEPUTY DIRECTOR OF EXAMINATION OF PATENT FUND, MECHANICAL, ELECTRICAL AREA AND REGISTRY OF INDUSTRIAL DESIGNS AND
<img file="MX340366B_D0003.tif" />
<img file="MX340366B_D0004.tif" />
<img file="MX340366B_D0005.tif" />
IMPI. "
MEXICAN INSTITUTE \
Dt THE PROPERTY
- 'INDUSTMIAL
METHOD FOR APPLYING EDGE COMPENSATION
<img file="MX340366B_D0006.tif" />
TECHNICAL FIELD
The present invention relates to a method for adaptive compensation of samples to reduce the difference between the original samples and the reconstructed samples, and more particularly, to a method for adaptively adding the compensation to reconstructed samples based on the difference between the sample current and adjacent samples.
BACKGROUND OF THE INVENTION
For the compression of video data, a plurality of video standards have been developed. Such video standards are, for example, MPEG-2, MPEG-4, and H.264 / MPEG-4AVC. As the successor to H.264 / MPEG4 AVC. Video Encoding of
High Efficiency (HEVC) is currently under joint development by the ISO / IEC Moving Image Expert Group (MPEG) and the Expert Group on
ITU-T Video Coding (VCEG).
According to HEVC, an image is divided into larger encoding units (LCUs), one or more encoding units from each LCU are encoded by generating a prediction block using either inter prediction or intra prediction. The difference between an original block and the prediction blocks is transformed to generate a transformed block, and the transformed block is quantized using a parameter of
<img file="MX340366B_D0007.tif" />
ΙΜΡΪ
INSTITUTO MEX1CANC Dt LA PROPIEDAD, _. . ,. . INDUSTRIAL.
quantification and one of a plurality of predetermined quantization matri. The quantized coefficients of the quantized block are scanned by means of a predetermined scan type and then subjected to entropic coding. The quantized coefficients are inversely quantized and inversely transformed to generate a residual block which is combined with the prediction block to generate the reconstructed image. The reconstructed image is adaptively filtered using one or more unlock filters to remove the lock objects.
But, the unlock filter technique described in
H.264 and HEVC under development deteriorates the performance of decoding devices since the technique is too complicated. Also, even if unlock filtering is applied to the edges of the blocks, the differences between the original samples and the filtered samples still remain. But, according to the current SAO process, the differences between the original samples and the filtered samples occasionally increase since the optimal edge index should not be determined.
Therefore, a new technique is required to reduce post-processing complexity to improve post-processing performance.
IMPI
BRIEF DESCRIPTION OF THE INVENTION<sup>1</sup>
MEXICAN INSTITUTE nxr<sup>OF</sup> * · * INOUSTWAL eeupiety
<img file="MX340366B_D0008.tif" />
Technical problem
The present invention is directed to a method for edge compensation to reduce the difference between the original samples and the reconstructed samples and to reduce the number of bits required to compensate for the differences.
Technical Solution
An aspect of the present invention provides a method of applying edge compensation, comprising: generating an index of the current edge of a current sample, and applying edge compensation corresponding to the edge index to the current sample. The edge index is generated using the differences between a current sample and two adjacent samples determined by the type of edge compensation.
Advantageous Effects
The method according to the present invention generates an edge index of a current sample, and applies the edge compensation corresponding to the edge index, to the current sample. The edge index is generated using the differences between a current sample and two adjacent samples determined by the type of edge compensation. Consequently, the difference between the original samples and the reconstructed samples is effectively reduced by generating the optimal edge index. Also, the amount of bits required to reduce the differences, is redrawn when setting the
IMPI
MEXICAN INSTITUTE. PROPERTY sign of compensation positive or negative. industrial
<img file="MX340366B_D0009.tif" />
DESCRIPTION OF THE DRAWINGS
FIG. 1 is a block diagram illustrating an apparatus for encoding moving images in accordance with the present invention.
<td>The</td><td>FIG. 2 is a diagram</td><td>of blocks that</td><td>illustrates</td><td>a</td>
<td>apparatus</td><td>for decoding</td><td colspan="2">moving images</td><td>of</td>
<td>agreement</td><td>with the present invention.</td><td></td><td></td><td></td>
<td>The</td><td>FIG. 3 is a diagram of</td><td>illustrating flow</td><td colspan="2">a process</td>
deblocking filter according to the present invention.
FIG. 4 is a conceptual diagram illustrating a method for determining the intensity of the limit according to the present invention.
FIG. 5 is a conceptual diagram illustrating the edge of 4 samples in accordance with the present invention.
FIG. 6 is a conceptual diagram illustrating a method of dividing an image into multiple areas in accordance with the present invention.
FIG. 7 is a conceptual diagram illustrating edge types in accordance with the present invention.
FIG. 8 is a conceptual diagram illustrating edge indices in accordance with the present invention.
DETAILED DESCRIPTION OF THE INVENTION
From here, several will be described in detail
MEXICAN INSTITUTE
SAY THE PROPERTY - <sub>T</sub>-F*
INDUSTRIAL, modalities of the present invention, with reference to the attached drawings. However, the present invention is not limited to the exemplary embodiments described below, but can be implemented in various types. Therefore, many other modifications and variations of the present invention are possible, and it should be understood that within the scope of the described concept, the present invention may be practiced in any other way than as specifically described.
A moving image encoding apparatus and a moving image decoding apparatus according to the present invention may be a user terminal, such as a personal computer, a personal mobile terminal, a mobile multimedia player, a smartphone , or wireless communication terminals. The image encoding device and the image decoding device can be included in a communication unit to communicate with various devices, a memo to store various programs and data used to encode or decode images.
FIG. 1 is a block diagram of an image encoding apparatus 1000 in accordance with the present invention.
Referring to FIG. 1, the image decoding apparatus 1000 includes a dividing unit 1010
TMPI .'νιτυτο MEXICAN INDUSTRIAL PROPERTY
<img file="MX340366B_D0010.tif" />
imaging, an intra prediction unit 1020, a quantization unit 1030, a scanning unit 1040, an entropy encoding unit 1050, an intra prediction unit 1060, an interpolation unit 1070, a reverse quantification unit 1080 , a reverse transformation unit 1090, a post-processing unit 1100, an image storage unit 1110, a subtraction unit 1120 and an addition unit 1130.
The image division unit 1010 divides an image or a section into several larger encoding units (LCUs) and divides each LCU into one or more encoding units. LCUs can be 32x32, 64x64 or
128x128. The image division unit 1010 determines the prediction mode and the division mode of each encoding unit.
An LCU includes one or more encoding units. The
LCUs have a recursive quaternary tree structure to specify the division structure of LCUs. The parameters for specifying the maximum size and minimum size of the encoding units are included in a sequence parameter set. The division structure is specified by one or more division encoding unit flags. The size of a coding unit is 2Nx2N. If the size of the LCUs is 64x64 and the size of a smaller encoding unit (SCU) is
MEXICAN INSTITUTE
D £ THE PROPERTY ζ
8x8, the size of the p-ttede'ALse encoding units<sup>1</sup>
<img file="MX340366B_D0011.tif" />
64x64, 32x32, 16x16 or 8x8.
Coding units include one or more prediction units. In intra prediction, the size of the prediction units is 2Nx2N or NxN. In inter prediction, the size of the prediction units is specified by the division mode. The division mode is one of 2Nx2N, 2NxN, Nx2N and NxN if the encoding units are symmetrically divided. The division mode is one of 2NxnU, 2NxnD, nLx2N and nRx2N if the units are divided asymmetrically.
Coding units include one or more transformation units. Transformation units have a recursive quaternary tree structure to specify the division structure of the encoding units. The division structure is specified by one or more transformation unit flags per division. The parameters for specifying the maximum size and minimum size of transformation units are included in a sequence parameter set.
Transformation unit 1020 transforms the residual signals to generate a transformed block. The residual signals are transformed on a transformation unit basis. The residual signals are derived by subtracting a prediction block which is generated by the intra prediction unit 106 0 or the inter prediction unit 1070, from a
IMPI
MEXICAN INSTITUTE OF INDUSTRIAL PROPERTY
<img file="MX340366B_D0012.tif" />
original block.
The difference transformation matrix can be used according to the prediction mode (the intra prediction mode or the inter prediction mode). Also, in the intra prediction mode, the transformation matrix can be adaptively determined based on the intra prediction mode. The transformation units are transformed using two one-dimensional transformation matrices (the horizontal matrix and the vertical matrix).
For example, in intra horizontal prediction mode of intra prediction, a DCT-based integer matrix is applied to the vertical direction, and a DST-based or KLT-based integer matrix is applied to the horizontal direction since the signals residuals can have vertical directionality. In intra-vertical prediction mode of intra-prediction, an ECT-based integer matrix is applied to the vertical direction, and a DST or KLT-based integer matrix is applied to the vertical direction.
Alternatively, the type of the transformation matrix is determined based on the size of the transformation units.
The quantization unit 1030 determines the quantization parameter to quantize the transformed block. The quantization parameter is the quantization step size. The quantization parameter is determined
IMPIAS
MEXICAN INSTITUTE OF PROPERTY according to the quantification unit. The quantitative unit is a unit of quantification greater than or equal to a predetermined size. The default size is called the minimum size of the units of quantification. Quantification units that have the minimum size are called minimum quantization units. If the size of the encoding units is equal to or greater than a minimum size of the quantization units, the encoding units become the quantization units. A plurality of encoding units can be included in the minimum quantization units. The minimum units of quantification can be 8x8 blocks or 16x16 blocks. The minimum size can be determined by image.
The quantization unit 1030 generates a predictor of the quantization parameter and generates a differential quantization parameter by subtracting the predictor of the quantization parameter from the quantization parameter.
The differential quantification parameter is subjected to entropic coding.
The predictor of the quantization parameter is generated as follows.
First modality
The quantization parameters of a coding unit on the left, a top coding unit and a coding unit at the top left are
<img file="MX340366B_D0013.tif" />
Quantification parameters available. For example, the first available quantization parameter is set as the predictor of the quantization parameter. Or the average of the first two available quantization parameters is set as the predictor of the quantization parameter, and if only one quantization parameter is available, the available quantization parameter is set as the predictor of the quantization parameter.
Second modality
There may be none of the encoder unit on the left, the upper encoder unit, and the encoder unit at the top left of the current encoder unit. On the other hand there may be a previous encoding unit of the current encoding unit in the order of the encoding. Therefore, the quantization parameters of the encoding units adjacent to the current encoding unit and the previous encoding unit can be used to generate the predictor of the quantization parameter. The quantification parameters are recovered in the following order; 1) the quantization parameter of a coding unit on the left, 2) the quantization parameter of a superior adjacent coding unit, 3) the parameter of
IMPIAS
<img file="MX340366B_D0014.tif" />
MEXICAN INSTITUTE »*
OF THE MOHEDAL; Industrial C '•' í. '”__ quantization of an adjacent encoding unit at the top left, 4) the quantization parameter of the previous encoding unit.
Alternatively, the quantization parameters are recovered in the following order; 1) the quantization parameter of a left adjacent encoding unit, 2) the quantization parameter of a higher adjacent encoding unit, and 3) the quantization parameter of the previous encoding unit.
The average of the first two quantization parameters is set as the predictor of the quantization parameter when two or more quantization parameters are available, and when only one quantization parameter is available, the available quantization parameter is set as the predictor of the parameter. of quantification. For example, if the quantization parameters of the left and upper coding units are available, the average of the left and upper quantization parameters is set as the predictor of the quantization parameter. If only one of the quantization parameters of the encoding units to the left and top is available, the average of the available quantization parameter and the quantization parameter of the previous encoding unit is set as the predictor of the
IMPI
<img file="MX340366B_D0015.tif" />
MEXICAN INSTITUTE. _. . , „. »E THE PROtFEDAD quantification. If the quantization pastwtetr of the encoding units -a i-aq ^ yord and above are available, the quantization parameter of the previous encoding unit is set as the predictor of the quantization parameter. The average is rounded.
The quantization unit 1030 quantizes the transformed blocks using a quantization matrix and the quantization parameter to generate a quantized block. The quantized block is provided to the reverse quantization unit 1080 and the scanning unit 1040.
Scanning unit 1040 scans the quantized coefficients and transforms the quantized coefficients into components of the one-dimensional quantized coefficients by applying a quantized block scan pattern.
In the intra prediction mode, the distribution of the quantified coefficients varies according to the intra prediction mode and the size of the transformation units. Therefore, the scanning pattern is determined based on the intra prediction mode and the size of the transformation units. The scan pattern can be selected from a zigzag scan pattern, vertical scan, and horizontal scan. The zigzag scan can be replaced with a diagonal scan.
IMPI
<img file="MX340366B_D0016.tif" />
MEXICAN INSTITUTE i
OF THE PROPERTY '
For example, if the size of the 'hassii? DhdeS-2 transformation is equal to or less than 8x8 ~ the horizontal scan is selected for vertical mode and the default number of intra-adjacent prediction modes for vertical mode and vertical scan is selected for landscape mode and the predetermined number of intra-adjacent prediction modes of landscape mode, and zigzag or diagonal scan is selected for the other intra prediction modes. When the size of the transformation units is greater than 8x8, the
<td>exploration</td><td>zigzag or the</td><td>exploration</td><td>in</td><td>diagonal</td><td>I know</td>
<td>select</td><td>for all modes</td><td colspan="3">intra prediction.</td><td></td>
<td>At</td><td>prediction mode</td><td>inter, se</td><td>uses</td><td>a pattern</td><td>of</td>
<td>exploration</td><td>predetermined.</td><td>The boss</td><td>of</td><td colspan="2">exploration</td>
<td colspan="2">default may be</td><td>exploration</td><td>in</td><td>zigzag or</td><td>the</td>
<td>exploration</td><td>in diagonal.</td><td></td><td></td><td></td><td></td>
<td>When</td><td>the size of the</td><td>units of</td><td colspan="2">transformation</td><td>is</td>
<td>greater than</td><td colspan="2">a predetermined size,</td><td>the</td><td colspan="2">coefficients</td>
quantized are divided into a plurality of subsets and then explored. The default size can be
4x4. The scanning pattern for exploring subsets is equal to the scanning pattern for exploring quantized coefficients within each subset. The quantized coefficients within each subset are explored in the reverse direction. The subsets are also
<img file="MX340366B_D0017.tif" />
IMPI
MEXICAN INSTITUTE. OF PROPERTY explore in the reverse direction. industrial
A parameter indicating the last position dlefeintaH.-de. zero is encoded and transmitted to the decoder. The last non-zero position specifies the position of the last non-zero quantized coefficient within the transformation unit. A parameter indicating the position of the last nonzero quantized coefficient within each subset is also transmitted to the decoding apparatus.
The inverse quantization unit 1080 inversely quantizes the quantized coefficients. The inverse transformation unit 1090 inversely transforms the coefficients with inverse quantization to generate residual signals.
The aggregation unit 1130 aggregates the residual signals generated by the inverse transformation unit 1090 and the prediction signals generated by the intra prediction unit 1060 or the inter prediction unit 1070. The subtraction unit 1120 subtracts the prediction samples from the original samples to generate the residual signals.
The post-processing unit 1100 performs the unlock filtering process, an adaptive sample compensation process, and an adaptive loop filtering process.
The unlock filtering process is carried out to
IMPI
MEXICAN INSTITUTE OF PROPERTY
<img file="MX340366B_D0018.tif" />
remove the lock objects which ap ^ Fé'flen reconstructed image. '----- ------ The adaptive sample compensation process is carried out after the unlock filter process has been carried out to reduce the difference between an original sample and a reconstructed sample. It is determined by image or section whether the adaptive compensation process is carried out or not. The image or section can be divided into a plurality of offset areas, and the type of offset can be determined by each area. There are four types of edge compensation and two types of band compensation. If the offset type is one of the edge offset types, the edge type is determined for each swatch within the offset area, and the offset corresponding to the edge type is added to each swatch. The type of the border is determined by comparing the current sample with two adjacent samples.
The adaptive loop filtering process can be carried out by comparing the reconstructed image and the original image to obtain the filter coefficients. The filter coefficients apply to all samples within the 4x4 block or the 8x8 block. Whether or not adaptive loop filtering is carried out is determined by encoding unit. Therefore, the size and coefficients of the loop filter can be changed based on a unit.
IMPIá
MEXICAN INSTITUTE OF THE MOPIF.VAD INDUSTRIAL
<img file="MX340366B_D0019.tif" />
coding.
The lffiágéñé storage unit 1110 I received the<sup>-</sup> reconstructed images from the post-processing unit 1100 and stores them in memory. The image is a box-based image or a field-based image.
Intermediate prediction unit 1070 performs motion estimation using one or more images stored in image storage unit 1110, and determines one more reference image indexes specifying one or more reference images and one or more motion vectors. The Inter prediction unit 1070 generates a prediction block using one or more indices of the reference images and one or more vectors.
Intra prediction unit 1060 determines the intra prediction mode of a current prediction unit and generates a prediction block using the intra prediction mode.
The entropic encoding unit 1050 applies the entropic encoding to the components of the quantized coefficients received from the scanning unit 1040, the intra prediction information received from the unit
1060 prediction unit, the motion information received from the inter prediction unit 1070.
FIG. 2 is a block diagram illustrating an apparatus 2000 for decoding motion pictures in accordance with the present invention.
ΙΜΡΙ
MEXICAN INSTITUTE
OF THE «INDUSTRIAL OTIEDAD
<img file="MX340366B_D0020.tif" />
As shown in FIG. 2, the motion picture decoding apparatus 2000 includes an entropic decoding unit 2010, a reverse scan unit 2020, a reverse quantization unit 2030, a reverse transform unit 2040, a unit
2050 intra prediction unit, an inter prediction unit 2060, a post processing unit 2070, an image storage unit 2089, and an add unit 2090.
The entropic decoding unit 2010 extracts and applies the entropic decoding to the intra prediction information, the inter prediction information and the quantized coefficient components of a received bit stream. The entropic decoding unit 2010 transmits the Inter prediction information to the Inter prediction unit 2060, transmits the intra prediction information to the intra prediction unit 2050, and transmits the quantized coefficient coefficients to the reverse scan unit 2020.
The reverse scan unit 2020 transforms the components of the quantized coefficient into two-dimensional quantized blocks using an inverse scan pattern.
In the intra prediction mode, the inverse scan pattern is selected based on the intra prediction mode and the size of the transformation units. The pattern of
IMPI
INSTITUTO MEXICANO DELA PROPIEDAD reverse exploration can be selected <sup>, N0</sup>^ TTÚte ^ - + Sn5 zigzag scan, the 'VtifLlual-y' scan - a horizontal scan. The zigzag scan can be replaced with a diagonal scan.
For example if the size of the transformation units is equal to or less than 8x8, the horizontal expiration is selected for the vertical mode and a predetermined number of intra-adjacent prediction modes of the vertical mode, the vertical scan is selected for the horizontal mode and the default number of horizontal mode intra-adjacent prediction modes, and zigzag or diagonal scan are selected for the other intra prediction modes. When the size of the transformation units is greater than 8x8, the
<td>exploration</td><td>zigzag or the</td><td>exploration</td><td>in</td><td>diagonal</td><td>I know</td>
<td>select</td><td>for all modes</td><td colspan="3">intra prediction.</td><td></td>
<td>At</td><td>prediction mode</td><td>Inter, it</td><td>uses</td><td>a pattern</td><td>of</td>
<td>exploration</td><td>predetermined.</td><td>The boss</td><td>of</td><td colspan="2">exploration</td>
<td colspan="2">default may be</td><td>exploration</td><td>in</td><td>zigzag or</td><td>the</td>
<td>exploration</td><td>in diagonal.</td><td></td><td></td><td></td><td></td>
If the size of the current transformation unit is larger than a predetermined size, the quantized coefficient components are inversely scanned on a subset basis to construct the quantized block. The subset is the default size. The size
TMPI ^^ 'TUTO MEXICANO
THE DEFAULT V7 PROPERTY may be 4x4. If the size of<sup>, N</sup>lMS &<sup>TO</sup>hín? t transformation keys equals the default size, ICTS<sup></sup>Components of the quantized coefficient of transformation units are explored inversely to construct the transformation units. When the components of the quantized coefficient are scanned inversely on a subset basis, the same inverse scan pattern applies to the components of the quantized coefficient of each subset.
The multiple subsets are scanned inversely in the reverse direction. The components of the quantized coefficient are also scanned inversely in the reverse direction. The inverse scan pattern applied to the components of the coefficient quantized to construct a subset is equal to the inverse scan pattern applied to the multiple subsets constructed. The reverse scan unit 2020 performs the reverse scan using the parameter indicating the position of the last non-zero quantized coefficient of the transformation units.
The reverse quantization unit 2030 receives the differential quantization parameter from the entropic decoding unit 2010 and generates a predictor of the quantization parameter to obtain a quantization parameter of a current encoding unit.
IMPI
IWC'HTI Ι'ΓΑ t> JL * VI ΓΆ klfl
MEXICAN INSTITUTE OF INDUSTRIAL PROPERTY
<img file="MX340366B_D0021.tif" />
The predictor of the quantization parameter is generated as follows.
First modality
The quantization parameters of a left coding unit, a higher coding unit, and a top left coding unit are retrieved sequentially in this order. The quantization parameter predictor is generated using one or two available quantization parameters. For example, the first available quantization parameter is set as the predictor of the quantization parameter. 0 the average of the first two available quantization parameters is set as the predictor of the quantization parameter, and if only one quantization parameter is available, the quantization parameter is set as the predictor of the quantization parameter.
Second modality
There may be no encoder unit on the left, a higher encoder unit, and an encoder unit at the top left of the current encoder unit. On the other hand, there may be a previous encoding unit of the current encoding unit in the order of the encoding. Therefore, the quantization parameters of the adjacent encoding units, adjacent to the current encoding unit and the unit of
IMPI NJT1TUTO MEXICANO Dt IA PXOPIÉOAL »
INDUSTRIAL precoding can be used to generate the predictor of the quantization parameter. The quantification parameters are recovered as in the following order; 1) the quantization parameter of an adjacent left encoding unit, 2) the quantization parameter of an upper adjacent encoding unit, 3) the quantization parameter of an upper left adjacent encoding unit, and 4) the quantification parameter of the previous encoding unit.
Alternatively, the quantization parameters are recovered in the following order; 1) the quantization parameter of a left adjacent encoding unit, 2) the quantization parameter of a higher adjacent encoding unit, and 3) the quantization parameter of the previous encoding unit.
The average of the first two available quantization parameters is set as the predictor of the quantization parameter when two or more quantization parameters are available, and when a quantization parameter is available, the available quantization parameter is set as the predictor of the quantification parameter. For example, if the quantization parameters of the left and upper coding units are available, the average of the left and upper quantization parameters is set as the
<img file="MX340366B_D0022.tif" />
I
MEXICAN INSTITUTE Di 1A PROPERTY
INDUSTRIAL
<img file="MX340366B_D0023.tif" />
predictor of the quantification parameter. If only one of the quantization parameters of the left and top encoding units is available, the average of the available quantization parameter and the quantization parameter of the previous encoding unit is set as the predictor of the quantization parameter. If the quantization parameters of the left and top encoding units are available, the quantization parameter of the previous encoding unit is set as the predictor of the quantization parameter. The average is rounded.
The reverse quantization unit 2030 generates the quantization parameter of the current encoding unit by adding the differential quantization parameter and the predictor of the quantization parameter. If the differential quantization parameter for the current encoding unit is not transmitted from the encoding side, the differential quantization parameter is set to zero.
The quantization parameter is generated per unit of quantification.
The reverse quantization unit 2030 inversely quantizes the quantized block.
Inverse transformation unit 2040 inversely transforms the block with inverse quantization, to generate a residual block. The type of the array
IMPI
MEXICAN INSTITUTE OF PROPERTY
INDUSTRIAL __ inverse transformation is determined based on the prediction mode (the intra prediction mode or the Inter prediction mode) and the size of the transformation units.
The addition unit 2090 generates the reconstructed samples by adding the residual block and the prediction block.
The intra prediction unit 2050 retrieves the intra prediction mode from the current prediction unit based on the intra prediction information received from the entropic decoding unit 2010, and generates a prediction block according to the intra prediction mode.
The inter prediction unit 2060 recovers one or more reference image indices and one or more motion vectors, based on the inter prediction information received from the entropic decoding unit 2010, and generates a prediction block using one or more reference images and one or more motion vectors.
The operation of the post-processing unit 2070 is the same as that of the post-processing unit 1100 of FIG.
1.
The image storage unit 2080 stores which images were post-processed by the post-processing unit 2070.
FIG. 3 is a flow chart illustrating the process
IMPI
MEXICAN INSTITUTE OF PROPERTY
<img file="MX340366B_D0024.tif" />
filter unlock according to iNí> g? TRiA £<sub>r</sub>invention
The unlocking filtering process is carried out by the post-processing unit 1100 of the moving image encoding apparatus 100 shown in FIG. 1 and by the post-processing unit 2070 of the motion picture decoding apparatus 2,000 shown in FIG.
.
When it is determined that the unlock filter is carried out on a section, the unlock filter process is applied to the section. The motion picture decoding apparatus uses a 'disable_block_filter_flag' flag received from a bit stream to determine whether or not sectional filtering is performed.
Unlock filtering is performed on each encoding unit. The vertical edges are filtered starting with the edge from the left side of the encoder unit towards the right side of the encoder unit. The horizontal edges are then filtered starting with the edge at the top of the encoder unit towards the bottom of the encoder unit.
The unlock filter is applied only to the edges of the prediction units and the edges of the prediction units
IMPI
MEXICANÍ INSTITUTE> OR THE PROPERTY
<img file="MX340366B_D0025.tif" />
transformation. If the width or height of the prediction units or transformation units is less than<sup></sup>length of 8 samples, the deblocking filter is applied only to the edges lying on the 8x8 sample mesh.
The intensity of the limit is determined on each sample edge lying on the 8x8 sample mesh (S110).
FIG. 4 is a conceptual diagram illustrating a method for determining the intensity of the limit according to the present invention.
As shown in FIG. 4, the intensity of the limit is determined on each edge of 4 samples that lies on the 8x8 sample mesh. The limit intensity is then determined on the edges of the 8x8 blocks using two consecutive limit intensities.
Accordingly, the computational complexity required to determine the limit intensity according to the present invention is reduced by 50% when compared to the HEVC under development. Also, the present invention reduces the memory capacity and bandwidth required to determine the limit intensity to 50%. Therefore, the present invention reduces the complexity of the equipment and the programs without deterioration of the image quality.
FIG. 5 is a conceptual diagram illustrating the edge
IMPI
MEXICAN INSTITUTE OF PROPERTY
<img file="MX340366B_D0026.tif" />
of 4 samples according to the present invention ™<sup>AI</sup>Coi ^ 'shown in FIG. 5, the edge of 4 samples is located between a P block that contains the pO sample and a Q block that contains the qO sample. Sample pO corresponds to one of samples pO<sub>0</sub> ~ p0<sub>3</sub>, and the sample qO corresponds to one of the samples qO<sub>or</sub> ~ q0<sub>3</sub>. Block P and Q is a prediction unit or a transformation unit.
The intensity of the limit is determined as follows. The intensity of the limit is determined by edge of 4 samples,
If the prediction unit containing the pO sample or the prediction unit containing the qO sample are subjected to intracoding, the intensity of the limit of the edge of 4 samples is set equal to 2. The edge of 4 samples is an edge of the prediction unit. That is, if block P and block Q undergo intercoding, the intensity of the limit is set equal to 0 or 1.
If one or more of the following conditions is satisfied, the intensity of the limit is set equal to 1.
1) The edge of 4 samples is the edge of a transformation unit, so the transformation unit that contains the sample pO or the transformation unit that contains the sample qO contains one or more non-zero transformation coefficients.
2) the edge of 4 samples is the edge of a unit of
ΙΜΡΙ
MEXICAN INSTITUTE OF PROPERTY
<img file="MX340366B_D0027.tif" />
prediction, the prediction unit containing the 'ÍWÍÍ ^ tra ^ ff and the prediction unit containing 1 <T “UllltiyLia — qQ additive to intercoding, and the prediction unit containing the pO sample or the prediction unit that contains the sample that have different reference images or a different number of motion vectors.
3) The prediction unit containing the pO sample and the prediction unit containing the qO sample are inter-coded, the prediction unit containing the pO sample and the prediction unit containing the qO sample have a vector of motion, and the absolute difference between the horizontal and vertical component of the motion vectors is greater than or equal to and a predetermined value (for example, 1 sample). The border is not part of the horizontal limit of the LCUs.
4) The prediction unit containing the pO sample and the prediction unit containing the qO sample undergo intercoding, the prediction unit containing the pO sample and the prediction unit containing the qO sample have two vectors of motion, the prediction unit containing the pO sample and the prediction unit containing the qO sample have at least the same reference image, and the absolute difference between the horizontal or vertical component of two motion vectors corresponding to the same reference image is greater or
IMPI
MEXICAN INSTITUTE OF INDUSTRIAL PROPERTY
<img file="MX340366B_D0028.tif" />
equal to the default value. The border is not part of the horizontal limit of the LCUs.
As described above, if the edge of 4 samples does not lie on the 8x8 sample mesh, the intensity of the limit is set equal to 0.
On the other hand, when the edge is a horizontal edge of the LCUs and a prediction unit containing the pO sample is located above the horizontal edge of the LCUs, the movement information of the prediction unit containing the pO sample can be replaced with information on the movement of a prediction unit adjacent to the left or to the right of the prediction unit containing the pO sample, based on the size and / or location of the prediction unit containing the pO sample.
Next, it is determined whether or not the deblocking filtering is carried out on the edge of 4 samples (S120).
For the edge of 4 samples, deblocking filtering is carried out if the following two conditions are satisfied.
1) bS> 0
2) d <p bS represents the intensity of the limit. The available β value is determined based on the QP limit quantification parameter<sub>B</sub>.
Variable d is defined as follows.
IMPI
MEXICAN INSTITUTE OF INDUSTRIAL PROPERTY
1) d = dpo + dqo + dp3 <sup>+</sup> dq3
2) d<sub>pk</sub>= | p2<sub>k</sub> - 2-pl<sub>k</sub>+ pO<sub>k</sub>| yd<sub>qk</sub>= | q2<sub>k</sub> - 2-ql<sub>k</sub> + qO<sub>k</sub> |
Next, if it is determined that the unlock filter is applied to the edge of 4 samples, an unlock filter is selected between a strong filter and a weak filter. But, if it is determined that the unlock filter is not applied to the edge of 4 samples, the unlock filter process ends for that edge. As shown in FIG. 5, a filter is selected for each edge of 4 samples.
If the following conditions are met, the strong filter is selected.
1) d <(β >> 2)
2) | p3i-p0i | + | q3i-q0i | <(β >> 3) for each i, i = 0, 3
3) | p0i-q0i | <(5 * t<sub>c</sub>+ l) >> 1 for each i, i = 0, 3
OR
1) άι <(β >> 1) for each i, i = 0, 3
2) | p3i-p0i | + | q3i-q0i | <(β >> 3) for each i, i = 0, 3
3) | p0j.-q0i | <(5 * t<sub>c</sub>+ l) >> 1 for each i, i = 0, 3
Otherwise, the weak filter is selected. The value of the variable t<sub>c</sub> is determined based on the quantification parameter of the QP limit<sub>B</sub>Next, if the unlock filter is selected, the edge is filtered using the unlock filter (S14Q).
The strong filter is as follows.
<img file="MX340366B_D0029.tif" />
Po '= (pz + 2 * Pi + 2 * p<sub>0</sub> + 2 * q<sub>0</sub> + qi + 4) >> 3)
<img file="MX340366B_D0030.tif" />
Ρι '= (p2 + Pi + Po + qo + 2) >> 2 p<sub>2</sub>'= (2 * P3 + 3 * p<sub>2</sub>+ Pi + Po + qo + 4) >> 3 qo '= (pi + 2 * p<sub>0</sub>+ 2 * q<sub>0</sub>+ 2 * qi + q<sub>2</sub>+4) >> 3 qi '= (p<sub>or</sub>+ qo + qi + q2 + 2) >> 2 q<sub>2</sub>'= (po + qo + qi + 3 * q<sub>2</sub>+ 2 * q<sub>3</sub>+4) >>3
The weak filter is as follows á = Segment3 (-t<sub>c</sub>, t<sub>c</sub>, Δ)
Po '= Segmentol (p<sub>0</sub> + Δ) q<sub>0</sub>'= Segmentol (q<sub>0</sub>-á) áp = Segment3 (- (t<sub>c</sub>>> l), t<sub>c</sub>>> l, (((ρ<sub>2</sub>+ Ρο + 1) -ρι + Δ) >> 1)
Pi '= Segment (ρι + Δρ) áq = Segment3 (- (t<sub>c</sub>>> l), t<sub>c</sub>>> l, (((q<sub>2</sub>+ q<sub>0</sub>+ l) >> 1) -qi-Δ) >> 1) qi '= Segmentol (qi + áq) the variables β and t<sub>c</sub> are determined by the QP limit quantization parameter<sub>B</sub>, and increases monotonically when the quantization parameter QP<sub>B</sub> increases. The relationship between the parameters β and t<sub>c</sub>, and the quantization parameter are defined as a table.
The quantization parameter of the QP limit<sub>B</sub> is the average of the quantization parameter QP<sub>P</sub> of the block P containing the sample pO and QP<sub>what</sub> of the block Q that contains the sample qO. The average is the rounded value. If at least one of block P and block Q undergoes intracoding, the parameter t<sub>c</sub> increases by 0, 1, or 2 when QP<sub>B</sub> increases in
IMPI
MEXICAN INSTITUTE OE INDUSTRIAL PROPERTY
<img file="MX340366B_D0031.tif" />
Now, the adaptive compensation process of the samples according to the present invention is described. The adaptive sample compensation process is carried out by the post-processing unit 1100 of the moving image encoding apparatus 1000 shown in the
FIG. 1 and by the apparatus post-processing unit 2070
2000 for decoding motion pictures shown in FIG. 2.
FIG. 6 is a conceptual diagram illustrating a method of dividing an image into multiple areas in accordance with the present invention. The SAO type is defined by area. As shown in FIG. 6, the areas are generated by dividing an image into a quaternary tree structure. The area can be an LCU. There are three types of ODS types. If the SAO type is the first type (deactivated), the SAO process does not take place over the corresponding area. If the ODS type indicates the band offset (BO), a band offset is added to each sample within the area. If the SAO type indicates the edge compensation (EO), the edge compensation determined by the edge index is added to each sample within the area.
FIG. 7 is a conceptual diagram illustrating edge types in accordance with the present invention.
As shown in FIG. 7, There are four kinds of bodes in edge compensation. The edge guy is
IMPI
<img file="MX340366B_D0032.tif" />
INSTITUTO MEXICANO of the property determines by the positions of the adjacent samples ^^ tisa to derive the edge index. The first tipu<sup>1</sup> ± '~ ± rürde ·· indicates the 0 degree ID edge type, the second type indicates the 90 degree ID edge type, the third type indicates the 135 degree ID edge type, and the fourth type indicates the 90 degree ID border type. Sample C represents a current sample and the two shaded samples represent two adjacent samples determined by the type of the border.
The adaptive sample compensation process is carried out as follows when the adaptive sample compensation type indicates one of the edge compensation types according to the present invention.
First, the edge index is derived using the differences between a current sample and two adjacent samples. The two adjacent samples are determined by the type of edge compensation for a current area. The edge index is derived by sample within the current area. The edge index is derived as follows.
edgeIdx = 2 + sign3 (recImage (x) -recImage (x-1)) + sign3 (recImage (x) -reclmagen (x + 1))
The sign3 (y) function is equal to 1 if y is greater than 0, the sign3 (y) function is equal to -1 if y is less than 0, and the sign3 (y) function is equal to 0 if y is equal to 0.
The variable reclmagen (x) represents the value of the
<img file="MX340366B_D0033.tif" />
current sample, the variables recImage (x-Γ} and recImage (x + 1) represent the two values of adjacent samples.The two adjacent samples are determined by the type of edge compensation of the current area.
FIG. 8 is a conceptual diagram illustrating edge indices in accordance with the present invention. In the
FIG. 8, the horizontal axis represents the position of the sample and the vertical axis represents the value of the sample.
As shown in FIG. 8, the edge index is set to 0 if both values of two adjacent samples are greater than the value of the current sample, the edge index is set to 1 if one of the values of two adjacent samples is greater than the current sample and the other is equal to the value of the current sample, the edge index is set to whether one of the values of two adjacent samples is greater than the current sample and the other is less than the value of the current sample, the edge index is set to 3 if one of the values of two adjacent samples is less than the current sample and the other is equal to the value of the current sample, and the edge index is set to 4 if both values of two samples adjacent are less than the current sample. The edge index is also set to 3 if both values of two adjacent samples are equal to the current sample.
Meanwhile, when one of the two adjacent samples
<img file="MX340366B_D0034.tif" />
ΙΜΡΙ
INSTITUTO MEXICANO DE LA TRONE DAD,, _ industrial belongs to another LCU, edge compensation may not apply to the current sample or another adjacent sample within the LCU is used instead of the adjacent sample that belongs to the other LCU.
Next, the edge offset is added to the current sample as follows.
recSaolmagen (x) = reclmagen (x) + Edge_offset [edgeldx] Edge offset is determined based on the edge index. In motion picture decoding apparatus 2000, edge compensation is obtained from a bit stream transmitted from motion picture encoding apparatus 1000. Motion picture encoding apparatus 1000 can transmit 4 or 5 edge offsets. If 4 edge offsets are transmitted, the 4 edge offsets correspond to the edge indices 0, 1, 3, 4 respectively, the edge offset is considered as 0.
Edge compensation can be a positive value or a negative value. The number of bits required to transmit the 4 edge offsets increases when the area is larger. The method for reducing the amount of bits according to the present invention is as follows.
First modality
A positive offset is applied to the edge index 0 and a negative offset is applied to the edge index 4.
IMPI '^ ΕΚψ'ί ^ ΑΝΟ W industrial property _
That is, only the absolute values of the two compensations
<img file="MX340366B_D0035.tif" />
from the edge are transmitted to reduce the number of bits. For edge indices 1 and 3, the absolute value and sign of the edge offset are transmitted.
Second modality
A positive offset applies to edge indices 0 and 1, and a negative offset applies to edge indices 3 and 4. That is, only the absolute values of the four edge offsets are transmitted to reduce the number of bits.
Also, compensation is not added to the current sample if the difference between the current sample and a negative sample is greater than a threshold. For example, if the absolute value of the difference between the current sample and an adjacent sample is greater than the threshold, the offset value is set to 0. On the other hand, either negative offset or positive offset is used.
Although the invention has been shown and described with reference to certain exemplary embodiments thereof, those skilled in the art will understand that various changes may be made in the form and details thereof without departing from the spirit and scope of the invention. as defined by the appended claims.
Contents47
43 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
140 members in 17 offices
Priority claims9
| Document | Office | Kind | Date |
|---|---|---|---|
| 1020120005334 | Republic of Korea | – | |
| 20120005334 | Republic of Korea | A | |
| 20120005334 | Republic of Korea | A | |
| 2013070222 | China | W | |
| 2013070222 | China | W | |
| 1020120005334 | – | – | – |
| KR20120005334 | – | – | – |
| PCTCN2013070222 | – | – | – |
| WO2013CN70222 | – | – | – |
Members140
| Document | Office | Kind | |
|---|---|---|---|
| CN103209324A | China | A | |
| CA2849038A1 | Canada | A1 | |
| CA2996158A1 | Canada | A1 | |
| CA3090780A1 | Canada | A1 | |
| CA3206389A1 | Canada | A1 | |
| WO2013107314A1 | World Intellectual Property Organization (WIPO) | A1 | |
| TW201345264A | Taiwan Province of China | A | |
| AU2013211395A1 | Australia | A1 | |
| SG11201400670XA | Singapore | A | |
| KR20140057391A | Republic of Korea | A | |
| IL231708A0 | Israel | A0 | |
| IL231708D0 | Israel | D0 | |
| KR20140067133A | Republic of Korea | A | |
| PH12014500959A1 | Philippines | A1 | |
| EP2749028A1 | European Patent Office (EPO) | A1 | |
| KR20140103903A | Republic of Korea | A | |
| MX2014004103A | Mexico | A | |
| JP2015502715A | Japan | A | |
| US2015103930A1 | United States of America | A1 | |
| US2015110179A1 | United States of America | A1 | |
| AU2013211395B2 | Australia | B2 | |
| IN3867CHN2014A | India | A | |
| US9172961B2 | United States of America | B2 | |
| AU2015255239A1 | Australia | A1 | |
| AU2015255240A1 | Australia | A1 | |
| AU2015255241A1 | Australia | A1 | |
| AU2015255242A1 | Australia | A1 | |
| JP5828967B2 | Japan | B2 | |
| US2015365667A1 | United States of America | A1 | |
| US2015365668A1 | United States of America | A1 | |
| US2015365704A1 | United States of America | A1 | |
| US2015365705A1 | United States of America | A1 | |
| EP2749028A4 | European Patent Office (EPO) | A4 | |
| JP2016012942A | Japan | A | |
| JP2016012943A | Japan | A | |
| MX336661B | Mexico | B | |
| JP2016028524A | Japan | A | |
| JP2016029829A | Japan | A | |
| RU2014113084A | Russian Federation | A | |
| MX340366BThis record | Mexico | B | |
| MX340367B | Mexico | B | |
| MX340368B | Mexico | B | |
| US9485522B2 | United States of America | B2 | |
| US9491488B2 | United States of America | B2 | |
| US9491489B2 | United States of America | B2 | |
| TWI558212B | Taiwan Province of China | B | |
| RU2602986C2 | Russian Federation | C2 | |
| US9503758B2 | United States of America | B2 | |
| US9516348B2 | United States of America | B2 | |
| SG10201608492RA | Singapore | A | |
| TW201701652A | Taiwan Province of China | A | |
| JP6073442B2 | Japan | B2 | |
| JP6073443B2 | Japan | B2 | |
| JP6073444B2 | Japan | B2 | |
| JP6073445B2 | Japan | B2 | |
| US2017064307A1 | United States of America | A1 | |
| PH12016502086A1 | Philippines | A1 | |
| PH12016502086B1 | Philippines | B1 | |
| PH12016502087A1 | Philippines | A1 | |
| PH12016502087B1 | Philippines | B1 | |
| PH12016502088A1 | Philippines | A1 | |
| PH12016502088B1 | Philippines | B1 | |
| PH12016502089A1 | Philippines | A1 | |
| PH12016502089B1 | Philippines | B1 | |
| BR112014008256A2 | Brazil | A2 | |
| JP2017092975A | Japan | A | |
| IL252379A0 | Israel | A0 | |
| IL252379D0 | Israel | D0 | |
| AU2015255239B2 | Australia | B2 | |
| AU2015255240B2 | Australia | B2 | |
| AU2015255241B2 | Australia | B2 | |
| AU2015255242B2 | Australia | B2 | |
| MX350560B | Mexico | B | |
| PH12014500959B1 | Philippines | B1 | |
| TWI604733B | Taiwan Province of China | B | |
| IL231708A | Israel | A | |
| CN103209324B | China | B | |
| BR112014008256A8 | Brazil | A8 | |
| CN107743229A | China | A | |
| CN107835425A | China | A | |
| JP6302577B2 | Japan | B2 | |
| CA2849038C | Canada | C | |
| CN107888918A | China | A | |
| CN107888919A | China | A | |
| CN107888920A | China | A | |
| CN107888921A | China | A | |
| RU2016140522A | Russian Federation | A | |
| MY166250A | Malaysia | A | |
| IL259929D0 | Israel | D0 | |
| US10063859B2 | United States of America | B2 | |
| IL252379A | Israel | A | |
| IL252379B | Israel | B | |
| RU2016140521A | Russian Federation | A | |
| RU2016140524A | Russian Federation | A | |
| RU2016140523A | Russian Federation | A | |
| RU2016140521A3 | Russian Federation | A3 | |
| RU2016140522A3 | Russian Federation | A3 | |
| RU2016140523A3 | Russian Federation | A3 | |
| RU2016140524A3 | Russian Federation | A3 | |
| IL259929A | Israel | A |
Numbers
- Publication
- 340366
- Publication, DOCDB
- 340366
- Publication, EPODOC
- MX340366
- Application
- 2015009937
- Application, DOCDB
- 2015009937
- Application, EPODOC
- MX20150009937
Titles2
- Spanish
- METODO PARA APLICAR COMPENSACION DEL BORDE.
- English
- METHOD OF APPLYING EDGE OFFSET.
Classification
- CPC, 11
- H04N19/14
- H04N19/117
- H04N19/86
- H04N19/114
- H04N19/124
- H04N19/174
- H04N19/176
- H04N19/196
- H04N19/50
- H04N19/513
- H04N19/593
- IPC, 3
- H04N19 117
- H04N19 14
- H04N19 176