Video encoding method and video encoding apparatus and video decoding method and video decoding apparatus, which perform deblocking filtering based on tree-structure encoding units.
Abstract
Disclosed is a video encoding method which involves performing deblocking filtering based on encoding units. The method of the present invention involves: determining a filtering boundary for which deblocking filtering is to be performed on the basis of at least one data unit of encoding units hierarchically configured in accordance with the depth which indicates the number of spatial partitions performed on a maximum encoding unit, prediction units for prediction of the encoding units, and transformation units for transformation of the encoding units; determining a filtering strength at the filtering boundary on the basis of a prediction mode of an encoding unit to which pixels adjacent to the determined filtering boundary belong and on the basis of transform coefficient values of pixels adjacent to the filtering boundary; and performing deblocking filtering on the basis of the determined filtering strength.

Term
4.6 yearsleft in the term
Expires 13 April 2031.
- Priority
- Filed
- Granted
- Today
- Expires
4 claims: 4 independent, 0 dependent
- 1CLAIMS REIVINDICACIONES Habiéndose descrito la invención como antecede, se reclama como propiedad lo contenido en las siguientes reivindicaciones:Having described the invention as above, the content of the following claims is claimed as property: 1. Un método de decodificación de video al utilizar filtrado de desbloqueo, el cual caracterizado porque comprende: one. A video decoding method using unblocking filtering, which is characterized in that it comprises: receiving a bit stream including image data, information about a maximum encoding unit size, and division information;recibir un flujo de bits incluyendo datos de una imagen, información sobre un tamaño de una unidad de codificación máxima, e información de división;dividing the image into a plurality of maximum encoding units using the information on the size of the maximum encoding unit;dividir la imagen en una pluralidad de unidades de codificación máxima utilizando la información sobre el tamaño de la unidad de codificación máxima;hierarchically dividing the maximum encoding unit into one or more encoding units based on the division information;jerárquicamente dividir la unidad de codificación máxima en una o más unidades de codificación con base en la información de división;determinar una o más unidades de predicción en la unidad de codificación utilizando información de tipo de partición, en donde la información de tipo de partición indica uno de un tipo simétrico y un tipo asimétrico;determining one or more prediction units in the encoding unit using partition type information, wherein the partition type information indicates one of a symmetric type and an asymmetric type;determinar una o más unidades de transformación en la unidad de codificación utilizando información de tamaño de una unidad de transformación, en donde la unidad de transformación es rectangular con un tamaño horizontal y un determine one or more transformation units in the encoding unit using information on the size of a transformation unit, where the transformation unit is rectangular with a horizontal size and a VIΡ ϊ tamaño vertical indicado por la información VIΡ ϊ vertical size indicated by the information DE LA PROPIEDAD T ·» OF THE PROPERTY T · » INDUSTRIAL 'W »» carry out prediction in the prediction unit in the coding unit and inverse transformation in the transformation unit in the coding unit, in order to generate a reconstructed coding unit;INDUSTRIAL 'W »» llevar a cabo predicción en la unidad de predicción en la unidad de codificación y transformación inversa en la unidad de transformación en la unidad de codificación, con el fin de generar una unidad de codificación reconstruida;When a limit included in the reconstructed coding unit corresponds to at least one of a limit of the prediction unit and a limit of the transformation unit, determine a limit force limit based on at least one of the transformation coefficients nonzero, prediction mode, a motion vector, and a reference index;cuando un límite incluido en la unidad de codificación reconstruida corresponde a por lo menos uno de un límite de la unidad de predicción y un límite de la unidad de transformación, determinar una fuerza límite del límite con base en al menos uno de los coeficientes de transformación no cero, modo de predicción, un vector de movimiento, y un índice de referencia;determinar un método de filtrado por desbloqueo incluyendo al menos uno de un número de golpes de filtro y ubicación de pixeles para ser filtrados por desbloqueo, con base en al menos uno de la fuerza límite y pixeles adyacentes adjuntos al límite;y llevar a cabo filtrado por desbloqueo en los pixeles adyacentes de conformidad con el método de filtrado por desbloqueo, con el fin de generar una unidad de codificación filtrada incluyendo los pixeles filtrados por desbloqueo;determining an unlock filtering method including at least one of a number of filter strokes and pixel locations to be unlocked filtered, based on at least one of the limit force and adjacent pixels attached to the limit;and performing unlock filtering on the adjacent pixels in accordance with the unlock filtering method, in order to generate a filtered encoding unit including the unlocked filtered pixels;en donde: where: cuando la información de división indica una división para una profundidad actual, una unidad de when the division information indicates a division for a current depth, a unit of ..,. . ' . , . . . . . INSTITUTO MEXICANO codificación de una profundidad actual se aivajae'Titeo unidades de codificación de una profundidad inferior, independientemente de unidades de codificación vecinas, y cuando la información de división indica una no división de una profundidad actual, se obtiene al menos una unidad de predicción desde la unidad de codificación de la profundidad actual y se obtienen la una o más unidades de transformación desde la unidad de codificación de la profundidad actual. ..,. . '. ,. . . . . INSTITUTO MEXICANO coding of a current depth aitajae'Titeo coding units of a lower depth, independently of neighboring coding units, and when the division information indicates a non-division of a current depth, at least one prediction unit is obtained from the current depth coding unit and the one or more transformation units are obtained from the current depth coding unit.
- 2The method according to claim 2. El método de conformidad con la reivindicación 1, caracterizado porque el límite incluido en la unidad de codificación reconstruida corresponde a por lo menos uno de un límite de la unidad de predicción y un límite de la unidad de transformación incluida en la unidad de codificación reconstruida corresponde a por lo menos uno de un límite de la unidad de predicción y un límite de la unidad de transformación. 1, characterized in that the limit included in the reconstructed coding unit corresponds to at least one of a limit of the prediction unit and a limit of the transformation unit included in the reconstructed coding unit corresponds to at least one of a prediction unit limit and a transformation unit limit.
- 3The method according to claim 3. El método de conformidad con la reivindicación 1, caracterizado porque la determinación de la fuerza de límite comprende determinar la fuerza de límite en función de si el modo de predicción de la unidad de codificación a la que los pixeles adyacentes pertenecen con base en el límite es un modo intra o un modo Ínter y si los valores de coeficientes de transformación de los pixeles adyacentes son 1, characterized in that the determination of the limit force comprises determining the limit force as a function of whether the prediction mode of the encoding unit to which the adjacent pixels belong based on the limit is an intra mode or an Inter mode. and if the values of transformation coefficients of the adjacent pixels are 0. 0. IMPI IMPI INSTITUTO XEKfCANO XEKfCANO INSTITUTE
- 4The method in accordance with the relViaSrí'áfe. 4 . El método de conformidad con la relViaSrí'áfe. 1, caracterizado porque la determinación i fusr^a de límite comprende determinar la fuerza de límite con base en un marco de referencia y un vector de movimiento utilizado para predicción de movimiento de una unidad de predicción a la que pertenecen los pixeles adyacentes, cuando un modo de predicción de la unidad de codificación a la que pertenecen los pixeles es un modo Ínter. 1, characterized in that the determination and limit fusion includes determining the limit force based on a reference frame and a motion vector used for motion prediction of a prediction unit to which the adjacent pixels belong, when a prediction mode of the encoding unit to which the pixels belong is an Inter mode. * OXS «nSkS.aStl ^. *OXS«nSkS.aStl^.
Independent claims4
466 paragraphs in 38 sections, as filed
(54) Title: VIDEO CODING METHOD AND VIDEO CODING DEVICE AND VIDEO DECODING METHOD AND VIDEO DECODING DEVICE, WHICH PERFORM UNLOCKING FILTRATION BASED ON TREE STRUCTURE CODING UNITS.
(54) Title: VIDEO ENCODING METHOD AND VIDEO ENCODING APPARATUS AND VIDEO DECODING METHOD AND VIDEO DECODING APPARATUS, WHICH PERFORM DEBLOCKING FILTERING BASED ON TREE-STRUCTURE ENCODING UNITS.
(57) Summary
A method of encoding a video is described, which performs unlock filtering based on the encoding units. The method includes determining a filtering limit on which the unlocking filtering is to be performed, based on at least one data unit among the encoding units that are hierarchically configured according to depths that indicate a number of times in which at least one maximum encoding unit is spatially divided, and the prediction units and transformation units respectively for the prediction and transformation of the coding units, determining the strength of the filtration at the filtration limit based on a prediction mode of a coding unit to which the adjacent pixels, based on the filtration limit, and the transformation coefficient values of the pixels adjacent to the filtration limit, and performing the deblocking filtration based on the determined filtration force.
(57) Abstract
Disclosed is a video encoding method which involves performing deblocking filtering based on encoding units. The method of the present invention involves: determining a filtering boundary for which deblocking filtering is to be performed on the basis of at least one data unit of encoding units hierarchically configured in accordance with the depth which indicates the number of spatial partitions performed on a maximum encoding unit, prediction units for prediction of the encoding units, and transformation units for transformation of the encoding units; determining a filtering strength at the filtering boundary on the basis of a prediction mode of an encoding unit to which pixels adjacent to the determined filtering boundary belong and on the basis of transform coefficient values of pixels adjacent to the filtering boundary; and performing deblocking filtering on the basis of the determined filtering strength.
<img file="MX341305B_D0001.tif" />
PATENT TITLE NO. 341305 _SE_ «CMTAA Μ (« ΜΟΜΜ
Institute
Mexican Property
Industrial
Owner (s): SAMSUNG ELECTRONICS CO., LTD.
Address: 129, Samsung-ro, Yeongtong-gu, Suwon-s¡, Gyeongg¡-do, 443-742, REPÚBLICA
FROM KOREA
Name: VIDEO CODING METHOD AND VIDEO CODING DEVICE AND VIDEO DECODING METHOD AND VIDEO DECODING DEVICE, WHICH PERFORM UNLOCKING FILTRATION BASED ON TREE STRUCTURE UNITS.
Classification: lnt.CI.8: H04N19 / 117; H04N19 / 50
In
WOO-JI KOO Kl 4 | \ NLE
Number:
MX / a / 2015/014848
<img file="MX341305B_D0002.tif" />
N; ELENA ALSHINA, ALEXANDER ALSHIN; ILKQLAY SHtYAKHOV T / ¡
SUN-IL LEE; MIN-SU CHEON;
N.lfcDiMSEREGTN. ϊΌΟΝ-ΜΙ HONG
REQUEST
International filing date:
April 2011 nal of Patent Number: 334296
Ifeb, W- “1!
Country:
US
PRIORITY
Date:
April 2010
Number:
61/323,449
Ifeem
F> cha cia: Twenty years of Expiration 13 of abid
<img file="MX341305B_D0003.tif" />
The reference patent is partied with Andamento e 'os a4 eums 1 dB in accordance with article 23 of the Property Law taken as of the date of filing of the solfwtuc m · rights 2 f
X 3Í
Q ^ In this section, this title is based on
PÚpiedad Industrial (DianoiOfldai dele Federación (DOF) 27/06 / TW1, γβΜιιΒ | ΒΜ (ιι on <8S © 8 / tS94, 10/25/1996, 12/28/1997, 01/28/2004, 06/16 / 2005, 25§1 / 2006, # 05 / 2009,06 / 01/2010, 06/18/2010, 2B / O6 / ¿O10, 01/27/2012 and 04/09/2012); Articles 1 °, 3 'ΙΠΙΒΒΒ ·· ^^ and, —- * -'— * -' ·· - * 'section V, 6 ° section lll, and 5Θ of the Industrial Property Law, this patent has a validity of twenty years non-extendable, and will be subject to the page of the tanfa to keep in force bis 2 of the Law of 705/1999, section V and lll dM ^ HHMnMMMH ^ aiiaaa «^ B« RnwaaanMMtMI ^ M ^ MtlMMWMfcrmado 01/07 / 2002, 07/15/2004, 07/28/2004 and 09/07/2007); Articles 1, 3, 4, 5, section V, subsection a), 16 sections I and III and 30 of the Organic Statute of the Mexican Institute of Industrial Property (DOF) 12/27/1999, amended on 10/10/2002, 07/29/2004, 08/04/2004 and 09/13/2007); 1, 3 and 5 Clause 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. (DOF 12/15/1999, amended on 02/04/2000, 07/29/2004, 08/04/2004 and 09/13/2007).
<img file="MX341305B_D0004.tif" />
Arenal No. 550, Floor 1,
Santa María Tepepan town. Xochimiíco, CP. 16020.
Mexico City
Tel (55) 63 34 07 00 -> vv> tripigobnix
Issue Date: August 15, 2016
DIVISIONAL DIRECTOR OF PATENTS
<img file="MX341305B_D0005.tif" />
NAHANNY CANAL REYES
<img file="MX341305B_D0006.tif" />
<img file="MX341305B_D0007.tif" />
<img file="MX341305B_D0008.tif" />
MX / 2016/64841 iso'S<sup>-</sup> ή
<img file="MX341305B_D0009.tif" />
<img file="MX341305B_D0010.tif" />
MEXICAN INSTITUTE
VIDEO CODING METHOD AND COD DEVICE? Wíí8B8ifó
VIDEO AND VIDEO DECODING METHOD OF_
DECODING OF VIDEO, THAT PERFORM THE FILTRATION OF
UNLOCKING BASED ON CODE UNITS IN TREE STRUCTURE
Field of the Invention
The present invention relates to the encoding and decoding of a video.
Background of the Invention
As the physical equipment (hardware) to reproduce and store high resolution or high quality video content is being developed and supplied, the need for a video encoder / decoder (codee) to effectively encode or decode the video is increasing. high-resolution or high-quality video content. In a conventional video codec, a codec is encoded according to a limited macroblock-based encoding method that has a predetermined size.
A macroblock based prediction coding method can generate a blocking effect due to discontinuous pixel values at the block boundaries. Consequently, in a video codec, unlock filtering is performed to improve the compressibility of the video and the quality of a restored image.
Ref: 261386
<img file="MX341305B_D0011.tif" />
Brief Description of the Invention
TECHNICAL PROBLEM
<img file="MX341305B_D0012.tif" />
The present invention provides a 'method' for performing unlock filtering on a video codec by using an encoding unit that exceeds the limits of a conventional macroblock-based encoding method and has a new tree structure .
TECHNICAL SOLUTION
The present invention provides a method and apparatus for performing unlock filtering, which reduces an unlock effect generated in a boundary region of the encoding units in an encoded video based on a tree structure encoding unit.
ADVANTAGING EFFECTS
According to the embodiments of the present invention, the quality of a compressed and restored video can be remarkably improved by eliminating an unlocking effect from the restored compressed video, based on a tree structure encoding unit.
Brief Description of the Figures
Figure 1 is a block diagram of an apparatus for encoding a video, which carries out the filtering of unblocking based on the encoding units according to a tree structure, according to an embodiment
ΙΜΡΪ
MEXICAN INSTITUTE OF INDUSTRIAL PROPERTY of the present invention;
FIG. 2 is a block diagram of an apparatus for decoding a video, which performs unlock filtering based on the encoding units according to a tree structure, according to an embodiment of the present invention;
Figure 3 is a diagram for describing a concept of encoding units according to an embodiment of the present invention;
Figure 4 is a block diagram of an image encoder based on the encoding units according to an embodiment of the present invention;
Figure 5 is a block diagram of an image decoder based on the encoding units according to an embodiment of the present invention,
Figure 6 is a diagram illustrating the deepest encoding units according to depths, and the prediction units according to an embodiment of the present invention;
Figure 7 is a diagram for describing a relationship between a coding unit and transformation units, according to an embodiment of the present invention;
<img file="MX341305B_D0013.tif" />
Figure 8 is a diagram paaálVdeAciSi
MEXICAN INSTITUTE OF PROPERTY encoding information for units corresponding to a coded depth, —OS ÁCUéT'dó 5 an embodiment of the present invention;
Figure 9 is a diagram of deeper encoding units according to depths, according to one embodiment of the present invention;
Figures 10 to 12 are diagrams for describing a relationship between encoding units, prediction units, and transformation units, according to an embodiment of the present invention;
Figure 13 is a diagram for describing a relationship between a coding unit, a prediction unit, and a transformation unit, according to the coding mode information in Table 1;
Figure 14 illustrates the maximum encoding units, the encoding units according to a tree structure obtained by subdividing each maximum encoding unit, the prediction units obtained by subdividing each encoding unit, and the data units including the transformation units;
Figures 15 through 17 each illustrate the filtering limits determined based on the limits of the data units having a predetermined size or above, with respect to the data units of Figure <sub>14</sub>. IMPI 'MEXICAN INSTITUTE
OF INDUSTRIAL PROPERTY
Figure 18 is a reference diagram for describing an unlocking filtration process according to an embodiment of the present invention, based on the <sup>5</sup> filtration limits of Figure 17;
Figures 19A and 19B illustrate the pixels located above the filtration limits to describe unblocking filtration according to an embodiment of the present invention;
Figures 20A and 20B are reference tables for determining threshold values a and β, according to an embodiment of the present invention;
Figure 21 is a reference diagram for describing a process of adjusting an offset value 15 with respect to a threshold value a, in accordance with an embodiment of the present invention;
Figures 22A and 22B are reference tables used to determine a predetermined intermediate value. <sub>2</sub>θ used during an unblocking filtration process, according to an embodiment of the present invention;
Figure 23 is a flowchart illustrating a method of encoding a video based on the encoding units, in accordance with an embodiment of the present invention; and
<img file="MX341305B_D0014.tif" />
Figure 24 is a flow chart
<img file="MX341305B_D0015.tif" />
decoding method of a video based on the encoding units, according to an embodiment of the present invention.
Detailed description of the invention
In accordance with one aspect of the present invention, a video encoding method is provided, which performs unlocking filtering based on the encoding units, the method includes: dividing an image into at least one maximum encoding which is a unit of data that has a maximum size; determining the encoding units that are hierarchically configured according to depths, indicating a number of times that at least one maximum encoding unit is spatially divided, and the prediction units and transformation units respectively for prediction and transformation of the coding units; determining a filtering limit on which the unblocking filtering is to be performed based on at least one data unit from among the encoding units, the prediction units, and the transformation units; the determination of the filtration force at the filtration limit, based on a prediction mode of a coding unit to which the adjacent pixels belong, based on the filtration limit,
<img file="MX341305B_D0016.tif" />
and the values of the transition coefficient pixels adjacent to the filtration limit; and performing the deblocking filtration based on the determined filtration force.
In accordance with yet another aspect of the present invention, a method of decoding a video is provided, which performs unlock filtering based on the encoding units, the method includes: extracting the encoded image data according to the encoding units, the encoding mode information regarding the encoding units according to a tree structure, and the information regarding unlocking filtering on an encoding unit maximum, according to the encoding units according to the tree structure included in each maximum encoding unit, obtained by dividing a current image, by recognizing a received bitstream; the determination of the prediction units and the transformation units for the prediction and the transformation according to the encoding units, and the decoding of the encoded image data, based on the information of the encoding mode with respect to the units of coding according to the tree structure; the determination of a filtration limit at which the unblocking filtration between the
<img file="MX341305B_D0017.tif" />
limits of at least one input data unit<sub>s</sub>.¿Rt ^<sub>M</sub>& A¿t6la \ coding according to the tree structure the prediction units, and the transformation units, by using the information regarding the unlocking filtering; Determination of the filtration force of the filtration limit, based on a prediction mode of a coding unit to which the adjacent pixels belong based on the determined filtration limit, and the values of the transformation coefficient of the adjacent pixels. to the limit of filtration; and performing the unlock filtering on the decoded image data based on the determined filtering force.
In accordance with yet another aspect of the present invention, there is provided an apparatus for encoding a video, which performs unlock filtering based on the encoding units, the apparatus includes: a coding unit determiner to determine the coding units that are hierarchically configured according to depths, indicating a number of times that at least one maximum coding unit is spatially divided, where the maximum coding unit is one unit of data that has a maximum size that is divided to encode an image, and the prediction units and transformation units
<img file="MX341305B_D0018.tif" />
nL
<img file="MX341305B_D0019.tif" />
respectively, for prediction and tra:
OF THE PROPERTY
INDUSTRIAL _ the coding units; an unblocking filtering unit for determining a filtering limit on which the unblocking filtering is to be performed based on at least one data unit among the encoding units, the prediction units, and the transformation units, determining the filtering force at the filtering limit, based on a prediction mode of a coding unit to which the adjacent pixels belong based on the filtering limit, and the values of the transformation coefficients of the pixels adjacent the filtration limit, and the performance of the unblocking filtration based on the determined filtration force; and a transmitter for encoding the information regarding the unlock filtering and for transmitting the information with the encoding data of an image and the encoding mode information regarding the encoding units, according to the tree structure.
In accordance with yet another aspect of the present invention, an apparatus for decoding a video is provided, which performs the unlocking filtering based on the encoding units, the apparatus includes: a receiving and extracting unit for extracting the image data encoded according to the encoding units, the information of the encoding mode regarding the encoding units of
MEXICAN INSTITUTE OF PROPERTY tree structure, and information regarding the unlocking f'ITtTá'cioTi in a maximum coding unit, according to the coding units, according to the tree structure included in each coding unit maximum obtained by dividing a current image, by recognizing a received bit stream; a decoder for determining the prediction units and transformation units for prediction and transformation according to the encoding units, and decoding the decoded image data, based on the encoding mode information regarding the units coding according to the tree structure; and an unblocking filtering unit to determine a filtering limit at which unblocking filtering is to be carried out between the limits of at least one data unit among the encoding units, according to the tree structure, the prediction units, and transformation units, by using the information regarding the unlock filter, the determination of the filtration force of the filtration limit based on a prediction mode of a coding unit to which the adjacent pixels belong based on the determined filtration limit, and the values of the transformation coefficients of the pixels
ÍsMpP I adjacent to the filtration limit, and the real
MEXICAN INSTITUTE OF PROPERTY unlocking leak on 'TfiíStfen data
<img file="MX341305B_D0020.tif" />
decoded, determined.
based on ion filter strength ··
MODALITY OF THE INVENTION
FIG. 1 is a block diagram of a video encoding apparatus 100 that performs unlock filtering based on the encoding units according to a tree structure, according to an embodiment of the present invention.
The video encoding apparatus 100 includes a encoding unit determiner 110, an unlock filter unit 130 and a transmitter 120.
The determining encoding unit 110 receives the image data of an image from a video and divides the image into at least one maximum encoding unit which is a data unit having a maximum size. The maximum encoding unit according to an embodiment of the present invention may be a data unit having a size of 32 x 32, 64 x 63, 128 x 128, 256 x 256, etc., where a form of the data unit is a square that has a width and length in boxes of 2 that is greater than 8.
Coding unit 110 determines the coding units that have a hierarchical structure according to the regions especially divided by maximum coding units.
<img file="MX341305B_D0021.tif" />
The
MEXICAN INSTITUTE OF PROPERTY coding can be expressed based<sup>1NDUÍ</sup>éft<sup>L </sup>depth indicating a number of times that the -unielad of<sup>1 </sup>Maximum encoding is spatially divided. In detail, the coding units according to a tree structure include coding units corresponding to a depth determined by being a coded depth, out of all the deepest coding units according to the depths included in the maximum coding unit. . A coding unit of a coded depth can be hierarchically determined according to the depths of the same region of the maximum coding unit, and can be independently determined in the different regions.
Encoding unit 110 can encode for each deepest encoding unit included in a current maximum encoding unit, and determine an encoding unit to output an optimal encoding result and an encoded depth that is a corresponding depth by comparison. of the coding results of the coding units of a higher depth and a lower depth according to the regions. Also, a coded depth of the current region can be independently determined from a coded depth of yet another region.
<img file="MX341305B_D0022.tif" />
Consequently, the unit of co
INSTITUTO MEXICANO DE LA PROPIEDAD can determine the encoding units according to a tree structure at an independently encoded depth determined according to the regions per maximum encoding unit. Also, the encoding unit 110 performs the prediction encoding while determining the encoding units of the encoded depth. Coding unit 110 may determine a prediction unit or a division which is a data unit for performing prediction coding, to output an optimal coding result in the coding unit of the coded depth. For example, examples of a division type with respect to a coding unit that is 2N x 2N in size (where N is a positive integer) can include divisions that are 2N x 2N, 2N x N, N x 2N or N x N. Examples of the type of division include symmetric divisions that are obtained by symmetrically dividing a height or width of the encoding unit, divisions obtained by asymmetrically dividing the height or width of the encoding unit, such as ln on: l, the divisions that are obtained by dividing the prediction unit geometrically, and the divisions that have arbitrary shapes. Also, a division type prediction mode may be an Inter mode, an intra mode, a jump mode, or the like.
<img file="MX341305B_D0023.tif" />
a
A coding unit according to an embodiment of the present invention can be characterized by a maximum size and depth. Depth denotes a number of times the coding unit is hierarchically divided from the maximum coding unit, and as the depth becomes greater, deeper coding units according to depths can be divided from one maximum encoding unit to a minimum encoding unit. A maximum code unit depth is a higher depth and a minimum code unit depth is a lower depth. Since a size of a coding unit corresponding to each depth decreases as the depth of the maximum coding unit becomes larger, a coding unit corresponding to a greater depth may include a plurality of coding units corresponding to the lower depths.
A maximum depth according to an embodiment of the present invention is an index related to the number of times of division from a maximum encoding unit to a minimum encoding unit. A first maximum depth according to an embodiment of the present invention can denote the total number of times of division from the maximum encoding unit
<img file="MX341305B_D0024.tif" />
minimal coding. A second maximum depth according to the embodiment of the present invention can denote the total number of depth levels from the maximum coding unit to the minimum coding unit. For example, when a depth of the maximum coding unit is 0, a depth of one coding unit, in which the maximum coding unit is divided once, can be set to, and a depth of one coding unit, in which the maximum encoding unit is divided twice, can be set to 2. Here, if the minimum coding unit is a coding unit in which the maximum coding unit is divided four times, there are 5 depth depth levels of 0, 1, 2, 3, and 4, and thus the first maximum depth can be adjusted to 4, and the second maximum depth can be adjusted to 5.
Coding units according to a tree structure in a maximum coding unit and a method of determining a division, according to the embodiments of the present invention, will be described in detail below with reference to Figures 3 to.
The unlocking filter unit 130 determines a filtering limit at which the
IMPI unlock filtering based on at least one<sup>s</sup>aDEumí (Se ^ · * INDUSTRIAL data from among the coding units, the units of
<img file="MX341305B_D0025.tif" />
prediction and transformation units, and determines the filtration strength at the filtration limit based on a prediction mode of a coding unit to which the adjacent pixels belong, based on the determined filtration limit and the values of the transformation coefficient of the pixels adjacent to the filtration limit, and performs the unblocking filtrations based on the strength of the filtration. For example, when the encoding units, prediction units, and transformation units are determined, as will be described later, the unlock filter unit 130 can determine a limit of the data units having a predetermined size. or above as the filtering limit at which the unlocking filtering will be performed based on the sizes of the encoding units, the prediction units, and the transformation units, and to perform the unblocking filtering on the pixels adjacent to the filtering limit.
Transmitter 120 may encode the information regarding the unlock filter determined by the unlock filter unit 13 0, and transmit the information along with the encoded image data and the
ΐ) you ....... ....
V 'ί *> Α' 'νΛ encoding mode information regarding
INDUSTRIAL coding according to the tree structure of the maximum coding unit. The information regarding the unlock filtering may include the filtering limit determination information, such as a size of a data unit to determine a data unit to perform the unlock filtering, from the limits of laq, units of data, such as encoding units, prediction units, and transformation units.
Transmitter 120 can insert and transmit information regarding unlock filtering within a Sequence Parameter Group (SPS) or Image Parameter Group (PPS) of the image.
A process of determining a filtration limit for unblocking filtration and an unblocking filtration process according to the embodiments of the present invention, will be described in detail below with reference to Figures 14 to 24.
Coding unit 110 can determine a coding unit that has an optimal shape and size per maximum coding unit, based on a maximum size and depth of a given maximum coding unit considering the characteristics of a current image. Also, since the 'modification can be made
<img file="MX341305B_D0026.tif" />
<img file="MX341305B_D0027.tif" />
by using any of the diyei<sup>F</sup>Sl * ^ ™<sub>p</sub><sup>N</sup>Jg ^ g {gs' INDUSTRIAL prediction and transformation methods per maximum encoding unit, an optimal encoding mode can be determined considering the image characteristics of the encoding units of the various image sizes.
If an image that has a high resolution or a large amount of data is encoded in a conventional macroblock unit that has a fixed size of 16x16 or 8x8, a number of macroblocks per image is excessively increased. Consequently, a number of pieces of compressed information generated for each macroblock are increased, and thus it is difficult to transmit the compressed information and the efficiency of data compression decreases. However, by using the encoding unit 110, the final compression efficiency of a video can be increased since an encoding unit is adjusted while considering the characteristics of an image, while increasing its maximum image size. one encoding unit, while an image size is considered.
Also, prediction encoding that has reduced error with an original image can be performed by using a reference image that is filtered by unlock, via unlock filtration, or based on encoding units according to a ® MEXICAN INSTITUTE: OF PROPERTY
INDUSTRIAL tree.
FIG. 2 is a block diagram of a video decoding apparatus 200 that performs unlock filtering based on the encoding units according to a tree structure, according to an embodiment of the present invention.
The video decoding apparatus 200 includes a receiving and extracting unit 210, a decoder 220 and an unlocking filtering unit 230.
The receiving and extracting unit 210 extracts the image data encoded according to the encoding units according to a tree structure, the information of the encoding mode relative to the encoding units, and the information regarding unlock filtering , according to the maximum encoding units, for the reception and recognition of a bit stream with respect to a video. The receive and extract unit 210 can extract the information regarding unlock filtering from SPS or PPS of an image.
Decoder 220 decodes the encoded image data according to the encoding units, based on the encoding mode information regarding the encoding units according to the tree structure extracted by the receiving and extracting unit 210.
IMPIOUS
Decoder 220 can determimaiiruTaiinacArunKtóa
INDUSTRIAL PROPERTY encoding of a encoded depth included in a maximum encoding unit, and a division type, a prediction mode, and transformation unit of the encoding unit, based on the encoding mode information regarding encoding units according to the tree structure according to the maximum encoding units.
Decoder 220 can decode the encoded image data of a maximum encoding unit by decoding the encoded image data, based on the determined type of division, a prediction mode, and the transformation unit per encoding unit of encoding units according to the tree structure included in the maximum encoding unit.
The image data decoded by the decoder 220 and the information regarding the unlocking filtering extracted by the receiving and extracting unit 210, are input to the unlocking filtering unit 230.
Unlock filtering unit 230 determines a filtering limit for which unblocking filtering is to be performed from within the limits of at least one data unit from among the encoding units of
<img file="MX341305B_D0028.tif" />
prediction, and the transformation units by using the information regarding the unlocking filtering, determines the strength of the filtering at the filtering limit based on a prediction mode of a coding unit to which the adjacent pixels belong , based on the filtration limit and the transformation coefficient values of the pixels adjacent to the filtration limit, and performs unlock filtering on the decoded image data based on the strength of the filtering.
By using the unlock filter unit 230, an error between a restored image and an original image can be reduced since the prediction decoding is performed on a next image by reference to a reference image to which the image is performed. Unlock Filtration.
Figure 3 is a diagram for describing a concept of the encoding units according to an embodiment of the present invention.
A size of a coding unit can be expressed in width x height, and can be 64 x 64, 32 x 32, 16 x 16, and 8 x 8. A 64 x 64 coding unit can be divided into units of 64 x 64, 64 x 32, 32 x 64 6 32 x 32 prediction, and a 32 x encoding unit
IMPI can be divided into prediction units <sup>, n</sup>F&<sup>or</sup>i3r & h ^ a ^
INDUSTRIAL x 16, 16 x 32 or 16 x 16, a 16 x 16 encoding unit can be divided into prediction units of 16 x 16, 16 x 8, 8xl6u8x8, and an 8 x 8 encoding unit can be divided in prediction units of 8 x 8, 8x4 or 4 x 4.
In video data 310, a resolution is
1920 x 1080, a maximum size of one encoding unit is 64, and a maximum depth of 2. In 320 video data, a resolution is 1920 x 1080, a maximum size of one encoding unit is 64, and a maximum depth is 3. In 330 video data, a resolution is 352 x 288, a maximum size of one encoding unit is 16, and a maximum depth is 1. The maximum depth shown in Figure 3 denotes a total number of divisions from a maximum encoding unit to a minimum decoding unit.
If a resolution is high or a data amount is large, a maximum size of one encoding unit may be large not only to increase encoding efficiency but also to accurately reflect the characteristics of an image. Accordingly, the maximum size of the encoding unit for video data 310 and 320 that have higher resolution than video data 330, can be 64.
<img file="MX341305B_D0029.tif" />
, MEXICAN JTITUTE
Since the maximum depth of loé. dah »**<sup>$</sup>
I My P>
XIC.
-.NofcY<sup>1</sup>
310 is 2, the encoding units 315 of the video data 310 may include a maximum encoding unit that has a long axis size of 64, and encoding units that have long axis sizes of 32 and 16, since depths are increased to two layers by dividing the maximum coding unit twice.
Meanwhile, since the maximum depth of the video data 330 is 1, the encoding units 335 of the video data 330 may include a maximum encoding unit having a long axis size of 16, and the units of Encodes that have a long axis size of 8 since depths are increased to one layer per division of the maximum encoding unit only once.
Since the maximum depth of the video data
320 is 3, the encoding units 325 of the video data 320 can include a maximum encoding unit that has a long axis size of 64, and the encoding units that have long axis sizes of 32, 16, and 8 already that depths are increased to 3 layers by division of the maximum coding unit three times. As depth increases, detailed information can be expressed accurately.
Figure 4 is a block diagram of an image encoder 400 based on the units of
<img file="MX341305B_D0030.tif" />
coding, invention.
according to a modality
<img file="MX341305B_D0031.tif" />
OF INDUSTRIAL PROPERTY
<img file="MX341305B_D0032.tif" />
Image encoder 400 may correspond to video encoding apparatus 100. In other words, an intra predictor 410 performs the intra prediction on the encoding units in an intra mode, from among a current structure 405, and motion estimator 420 and a motion compensator 425 performs the inter estimate and the motion compensation on the encoding units in an inter mode between the current structure
405 by using the current structure 405, and a reference structure 495.
Data sent from intra-predictor 410, motion estimator 420, and motion compensator 425, is sent as a quantized transformation coefficient through transformer 430 and quantizer 440. The quantized transform coefficient is restored as data in a spatial domain through an inverse quantizer 460 and an inverse transformer 470, and the restored data in the spatial domain is output as the reference structure 495 after being pos t- processed through a 480 unlock unit and a 490 loop filter unit. The quantized transform coefficient can be sent as a 455 bit stream through an encoder
<img file="MX341305B_D0033.tif" />
of entropy 450.
Intra 410 predictor, 420 motion estimator, 425 motion compensator, 430 transformer, 440 quantizer, 450 entropy encoder, 460 reverse quantizer, 470 reverse transformer, 480 unlock unit, and unity Image filtering loop filtering 490 400 can operate considering the encoding units according to the tree structure, according to the maximum encoding units.
Specifically, the unlocking unit 480 determines a filtering limit at which the unlocking filtering is to be performed based on a maximum size of one encoding unit and the encoding units according to a tree structure, determines the strength of filtering at the filtering limit, based on a prediction mode of a coding unit to which the adjacent pixels belong, based on the filtration limit and the transformation coefficient values of the pixels adjacent to the filtration limit, and performs the unblocking filtration based on the filtration strength.
FIG. 5 is a block diagram of an image decoder 500 based on encoding units, according to an embodiment of the present invention.
<img file="MX341305B_D0034.tif" />
Recognizer 510 recognizes datwj-srindeExickmá
OF THE INDUSTRIAL PROPERTY encoded to be decoded, and the information regarding the encoding required for decoding a 505 bit stream. The encoded image data is output as reverse quantized data through an entropy decoder 520 and an inverse quantizer 530, and the inverse quantized data are restored to image data in a spatial domain through an inverse transformer 540.
An intra predictor 550 performs the intra prediction on the encoding units in an intra mode with respect to the image data in the spatial domain, and a motion compensator 560 performs motion compensation on the encoding units in an inter mode by the use of a 585 reference structure.
The image data in the spatial domain, which is passed through the intra 550 predictor and the 560 motion compensator, can be sent as a restored structure 595 after being post-processed through a 57 0 unlocking unit and a Loop Filtering Unit 580. Also, the image data that is post-processed through Unlock Unit 570 and Loop Filtering Unit 580 can be sent as reference frame 585.
In order to decode the image data in
<img file="MX341305B_D0035.tif" />
the 23 0 image decoder of the ^ sTnO & pa & a & c
FROM THE INDUSTRIAL PROPERTY video decoding 200, the image decoder 500 can perform the operations that are performed after the recognizer 510.
Since image decoder 500 corresponds to video decoding apparatus, recognizer 510, entropy decoder 520, inverse quantizer 530, inverse transformer 540, intra predictor 550, motion compensator 560, unlocking unit 570, and the loop filter unit 580 of the image decoder 500 perform operations based on the encoding units having a tree structure for each maximum encoding unit.
Specifically, the unlocking unit 570 determines a filtering limit at which the unlocking filtering is to be performed between the limits of at least one data unit among the encoding units according to a tree structure, the units of prediction, and the transformation units by using the recognized information regarding unlock filtering, determines the strength of the leak at the leak limit based on a prediction mode of a coding unit to which the adjacent pixels belong, based on the leak limit and the transformation coefficient values of the pixels adjacent to the . IMPI filtration limit, and performs the filtration V ^^^ y ^^
INDUSTRIAL regarding decoded image data based on the strength of filtering. Detailed operations regarding unblocking filtration will be described in detail later with reference to Figure 14.
Figure 6 is a diagram illustrating the deepest encoding units according to depths, and the prediction units according to an embodiment of the present invention.
The video encoding apparatus 100 and the video decoding apparatus 200 use the encoding units according to a tree structure, which are independently determined according to the regions, to consider the characteristics of an image. A maximum height, a maximum width, and a maximum depth of the encoding units can be adaptively determined according to the image characteristics, or can be adjusted differently by a user. The sizes of the deepest coding units according to the depths can be determined according to the predetermined maximum size of the coding unit.
In a hierarchical structure 600 of the coding units, according to an embodiment of the present invention, the maximum height and maximum width of the coding units are each 64, and the depth
<img file="MX341305B_D0036.tif" />
IMPI, maximum is 5. The maximum depth shown is W'Tg ^ MF ^ Í ^ N ^ i
INDUSTRIAL
6- denotes a total number of depth levels
<img file="MX341305B_D0037.tif" />
coming from a maximum encoding unit to a minimum encoding unit.
Since a depth becomes deeper along a vertical axis of hierarchical structure 600, a height and width of the deepest coding unit are each divided. Also, a prediction unit, or the divisions that are the basis for the prediction coding of each deeper coding unit, are shown along a horizontal axis of hierarchical structure 600.
In other words, an encoding unit 610 is a maximum encoding unit in hierarchical structure 600, where a depth is 0, and a size, for example height by width, is 64 x 64. The depth is deepened along the vertical axis, and a 620 encoder unit having a size of 32 x 32 and a depth of 1, a 630 encoder unit having a size of 16 x 16 and a depth of 2, a 640 encoding unit having a size of 8 x 8 and a depth of 3, and a 650 encoding unit having a size of 4 x 4 and a depth of 4, exist. Coding unit 650 which is 4 x 4 in size and depth of 4 is a minimum coding unit.
Divisions are accommodated
<img file="MX341305B_D0038.tif" />
comosTiTunadade
OF INDUSTRIAL PROPERTY
<img file="MX341305B_D0039.tif" />
prediction of the coding units, along the horizontal axis according to each depth. In other words, a prediction unit of the coding unit 610 is 64 x 64 in size and the depth of 0 includes a division 610 is 64 x 64 in size, the
<td>divisions</td><td> 612</td><td>than</td><td>they have the size</td><td>of</td><td> 64</td><td>x 32,</td>
<td>divisions</td><td> 614</td><td>than</td><td>they have the size</td><td>of 36</td><td>X</td><td>64, or</td>
<td>divisions</td><td> 616</td><td>than</td><td>they are the size of</td><td>32 x</td><td> 32 .</td><td>In others</td>
<td>words,</td><td colspan="2">unit</td><td>610 encoding</td><td>can</td><td>to be</td><td>a unit</td>
of square data that has a minimum size that includes divisions 610, 612, 614, and 616.
Similarly, a 620 encoding unit prediction unit that is 32 x 32 in size and depth of 1, may include a 620 division that is 32 x 32 in size, 622 divisions that are 32 x in size 16, 624 divisions that are 16 x 32 in size, and 624 divisions that are 16 x 16 in size.
Similarly, a prediction unit of encoding unit 630 having the size of 16 x 16 and depth of 2, may include a division having a size of 16 x 16 included in encoding unit 630, divisions 632 which they are 16 x 8 in size, the 634 divisions are 8 x 16 in size, and the 636 divisions are 8 x 8 in size.
<img file="MX341305B_D0040.tif" />
, λι;
<img file="MX341305B_D0041.tif" />
Similarly, a unit of predicts
OF THE INDUSTRIAL PROPHECY coding 640 which has' the size of 8 x 8
<td>depth of</td><td>3, can</td><td>include</td><td>a division</td><td colspan="2">who has a</td>
<td>8 x size</td><td>8 included</td><td>in the</td><td colspan="2">coding unit</td><td> 64 0,</td>
<td>the divisions</td><td>642 that</td><td>have</td><td>a size of</td><td>8x4,</td><td>the</td>
<td>divisions 644</td><td colspan="2">they have a</td><td>size 4</td><td>x 8, y</td><td>the</td>
<td>divisions 646</td><td>that have</td><td colspan="2">one size 4x4.</td><td></td><td></td>
The 650 encoding unit having the size of 4x4 and the depth of 4 is the minimum encoding unit and one encoding unit of the lowest depth. A coding unit prediction unit 650 may include a division 650 that is 4 x 4 in size, divisions 652 that are 4 x 2 in size, divisions 654 that are 2 x 4 in size, and 656 divisions that are 2 x 2 in size.
In order to determine at least one encoded depth of the encoding units constituting the maximum encoding unit 610, the encoder unit 120 determiner of the video encoding apparatus 100 performs the encoding for the corresponding encoding units at each depth included in the maximum coding unit 610.
A number of deeper encoding units according to depths that include data in the same interval and the same size increases as the
<img file="MX341305B_D0042.tif" />
¡1232 «fclS depth becomes greater. For example, cu,
<img file="MX341305B_D0043.tif" />
coding that correspond to a depth of 2 are included in the data that is included in a coding unit corresponding to a depth of 1. Consequently, in order to compare the coding results of the same data according to the depths, the coding unit corresponding to the depth of 1 and four coding units corresponding to the depth of 2 are each coded.
In order to perform coding for current depth from among the depths, at least one coding error may be selected for the current depth for performing coding for each prediction unit in the coding units corresponding to the current depth , along the horizontal axis of hierarchical structure 600. Alternatively, the minimum coding error can be sought by comparing the minimum coding errors according to depths, by performing coding for each depth as the depth becomes greater along the vertical axis of hierarchical structure 600. A depth and a prediction unit that has the minimum coding error in the maximum coding unit 610 can be selected as the coded depth and a division type of the maximum coding unit 610. I-IMI-Ρ ΐ> INSTITUTO MEXICANO '• i' OF PROPERTY í! INDUSTRIAL _
Figure 7 is an even diagram * ®. · Describe a
<img file="MX341305B_D0044.tif" />
relationship between a coding unit 710 and transformation units 720, according to an embodiment of the present invention.
The video encoding apparatus 100 or the video decoding apparatus 200 encodes or decodes an image according to the encoding units having sizes smaller than or equal to a maximum encoding unit for each maximum encoding unit. Transformation unit sizes for transformation during encoding can be selected based on data units that are no larger than a corresponding encoding unit.
For example, in the video encoding apparatus 100 or in the video decoding apparatus 200, if a size of the encoding unit 710 is 64 x 64, the transformation can be performed by using the transformation units 720 they are 32 x size.
Also, data from the 710 encoding unit that is 64 x 64 in size can be encoded by performing the transformation on each of the transformation units that are 32 x 32, 16x16, 8x8, and 4x4 in size, which are smaller than 64 x 64, y. IMPI then a transformation unit that has ®4τιτπτβ ^ ®ιχΝ «
INDUSTRIAL coding, can be selected.
<img file="MX341305B_D0045.tif" />
Figure 8 is a diagram for describing the coding information of the coding units corresponding to a coded depth, according to an embodiment of the present invention.
Video encoding apparatus 100 can encode and transmit information 800 regarding a type of division, information 810 regarding a prediction mode, and information 820 regarding a size of a transformation unit for each encoding unit that corresponds to a coded depth, such as the encoding mode information regarding the encoding units according to a tree structure.
Information 800 indicates information regarding a type of current encoding unit that is divided as a prediction unit (division) for a prediction encoding of the current encoding unit. For example, a current CU_0 encoding unit having a size of 2N x 2N and a depth of 0 can be used as a prediction unit after being divided into any one of an 802 division having a size of 2N x 2N, a division 804 having a size of 2N x N, a division 806 having a size of N x 2N, and a division 8 08 having a size of N x N. Here, information 800 regarding a type of
<img file="MX341305B_D0046.tif" />
IMPI '·' INSTITUTO MEXICANO division is adjusted to indicate one of the divisSÓVj ^ SíK has a size of 2N x N, division 806 that has a size of N x 2N, and division 808 that has a size of N x N.
Information 810 indicates a prediction mode of each encoding unit. For example, information 810 may indicate a prediction coding mode performed on a prediction unit indicated by information 800, i.e., an intra 812 mode, an inter 814 mode, or a jump mode 816.
<td>The</td><td>information</td><td> 820</td><td>indicates</td><td>a</td><td>Unit</td><td>of</td>
<td>transformation</td><td>in which</td><td>going to</td><td>to be</td><td>based</td><td>when</td><td>the</td>
<td>t rans f ormac i on</td><td>is done</td><td>on</td><td colspan="2">a unit of</td><td colspan="2">coding</td>
current. For example, the transformation unit may have one of a first size 822 and a second size 824 in an intra mode, and a first size 826 and a second size 828 in an inter mode.
The receiving and extracting unit 220 of the video decoding apparatus 200 can extract the information 800, 810 and 820 according to each deeper encoding unit, and the decoder 220 can use the information 800, 810 and 920 for the decoding.
Figure 9 is a diagram of deeper encoding units according to depths, according to one embodiment of the present invention.
<img file="MX341305B_D0047.tif" />
<img file="MX341305B_D0048.tif" />
The division information can be ^ íi'g ^ O ^ Sfe.
INDUSTRIAL indicate a change of a depth. The division information indicates whether a coding unit of a current depth is divided into coding units of a lower depth.
A prediction unit 910 for predicting the coding of a coding unit 900 having a depth of 0 and a size of 2N_0 x 2N_0 may include the divisions of a division type 912 having a size of 2N_0 x 2N_0, a division type 914 having a size of 2N_0 x N_0, a division type 916 having a size of
N_0 x 2N_0 y, a type of division 918 that has a size of
N_0 x N_0. Figure 9 illustrates only the division types 912 to 918 that are obtained by symmetrically dividing the prediction unit 910, but a division type is not limited thereto.
Prediction encoding is repeatedly performed on one prediction unit that is 2N_0 x N_0 in size, two prediction units that are 2N_0 x N_0 in size, two prediction units that are N_0 x 2N_0 in size, and four units of prediction that have a size of N_0 x N_0, according to each type of division. Prediction encoding in an intra mode and an inter mode can be performed on the prediction units having sizes of 2N_0 x 2N_0, N_0 x 2N_0,
IMPI □ ΛΗΒίβΒίβϊΙ »
INDUSTRIAL
<img file="MX341305B_D0049.tif" />
£
2N_0 x N_0, and N_0 x N_0. Jump mode encoding is performed only on the prediction unit that is 2N_0 x 2N_0 in size.
If a coding error is smaller in one of the division types 912 to 916, the prediction unit 910 may not be divided into a lower depth.
If an encoding error is the smallest in division type 918, a depth is changed from 0 to 1 to divide division type 918 in operation 920, and encoding is repeatedly performed on the 930 encoding units that have a depth of 2 and a size of N_0 x N_0 to look for a minimal encoding error.
A prediction unit 940 for the prediction coding of the coding unit 930 having a depth of 1 and a size of 2N_1 x 2N_1 (= N_0 x N_ =) may include divisions of a division type 942 having a size of 2N_1 x 2N_1, a division type 944 that has a size of 2N_ x N_l, a division type 946 that has a size of N_1 x 2N_1, and a division type 94 8 that has a size of N_1 x N_1. If an encoding error is the smallest in division type 948, a depth is changed from 1 to 2 to divide division type 948 in step 950, and encoding is repeatedly performed on the 960 encoding units, which they have one
IMPI
MEXICAN INSTITUTE for *
Στ lunircTRiAf.
<img file="MX341305B_D0050.tif" />
depth of 2 and a size of N_2 x N_2 minimal coding error.
When a maximum depth is d, the division operation according to each depth can be performed up to when a depth becomes d-1, and the division information can be coded as up to when a depth is one from 0 to d-2. . In other words, when encoding is performed up to when the depth is d-1 after the encoding unit corresponding to a depth of d-2 is divided in step 970, a prediction unit 990 for the prediction encoding of a 98 0 encoding unit that has a depth of d-1 and a size of 2N_ (dl) x 2N_ (dl) can include partitions of a partition type 992 that has a size of 2N_ (dl) x 2N_ (dl) , a 994 partition type that has a size of 2N_ (dl) x N_ (d-1), a 996 partition type that has a size of N_ (d-1) x 2N_ (dl), and a 998 partition type which has a size of N_ (d-1) x N_ (d-1).
Prediction coding can be repeatedly performed in a prediction unit that is 2N_ (dl) x 2N_ (dl) in size, two prediction units that are 2N_ (dl) x N_ (d-1) in size, two prediction units that have a size of N_ (d-1) x 2N_ (dl), four prediction units that have a size of N_ (d-1) x N_ (d-1) of the partition types 992 to 998 for
IMPI
MEXICAN INSTITUTE find a partition type that has uff <sup>THE</sup>i £ ^ ft $ r minimum encoding. .....
Even when partition type 998 has the minimum encoding error, since a maximum depth is d, a CU_ (dl) encoding unit that has a depth of d-1 is no longer divided to a lower depth, and a encoded depth for the encoding units constituting a current maximum 900 encoding unit is determined to be d-1 and a partition type of the current maximum 900 encoding unit can be determined as N_ (d-1) x N_ (d -one) . Also, since the maximum depth is d and a minimum 980 encoding unit that has a lower depth of d-1 is no longer divided to a lower depth, the division information for the 980 minimum encoding unit is not established.
A 999 data unit can be a minimum unit for the current maximum encoding unit. A minimum unit according to an embodiment of the present invention can be a square data unit obtained by dividing a minimum coding unit 980 by 4, that is, it can be a square data unit that has a maximum size that can be included in the coding units of all coded depths, the prediction units, and the transformation units
<img file="MX341305B_D0051.tif" />
IMPI, · ί INSTITUTO M'X'CANO included in the maximum encoding unit E. ^^ lgjgr performing encoding repeatedly, the video encoding apparatus 100_ can select a depth that has the least encoding error when comparing the encoding errors. coding according to the depths of the coding unit 900 to determine a coded depth, and setting a corresponding partition type and a prediction mode as a coded depth encoding mode.
As such, the minimum coding errors according to depths are comparable at all depths from 1 ad, and a depth having the least coding error can be determined as a coding depth. The coded depth and a predicted coded depth can be coded and transmitted as information regarding a coding mode. Also, since a coding unit is divided from a depth of 0 to a coded depth, only the coded depth division information is set to 0, and the depth division information excluding the coded depth is set to 1 .
The receiving and extracting unit 210 of the video decoding or apparatus 200 can extract and use the information regarding the encoded depth and the
CSX.
<img file="MX341305B_D0052.tif" />
prediction unit of the decoding unit partition 912. The video decoding apparatus 200 can determine a depth, at which the division information is 0, as a depth encoded by using the division information according to the depths, and using the information regarding a corresponding depth encoding mode to decode the encoded data from the corresponding encoding unit.
Figures 10 through 12 are diagrams for describing a relationship between encoding units 1010, prediction units 1060, and transformation units 1070, according to one embodiment of the present invention.
The encoding units 1010 are encoding units according to a tree structure determined by the video encoding apparatus 100 in a current maximum encoding unit. Prediction units 1060 are prediction units of the encoding units of each encoded depth in encoding units 1010, and transformation units 1070 are transformation units of each of the encoding units.
1010.
When a depth of a maximum encoding unit is 0 in the 1010 encoding units, the 1010 encoding units include the units
<img file="MX341305B_D0053.tif" />
Coding units 1012 and 1054 that have a coding units 1014, 1016, 1018, 1028, 1050 and 1052 that have a depth of 2, coding units 1020, 1022, 1024, 1026, 1030, 1032, and 1048 that they have a depth of 3 and the 1040, 1042, 1044 and 1046 encoding units that have a depth of 4.
In 1060 prediction units, some 1014, 1016, 1022, 1032, 1048, 1050 encoding units,
1052 and 1054 are obtained by dividing the encoding units into the 1010 encoding units. In other words, the partition types in the 1014, 1022, 1050, and 1054 encoding units are 2N x N in size, the partition types in encoding units 1016, 1048, and 1052 are N x 2N in size, and a partition type of encoding unit 1032 is N x N. In other words, the prediction units are smaller than or equal to each encoding unit.
The transformation or reverse transformation is performed on the image data from the encoding unit 1052 in the transformation units 1070 into a data unit that is smaller than the encoding unit 1052. Also, the encoding units 1014, 1016 , 1022, 1032, 1048, 1050, and 1052 in transformation units 1070 are data units that have different sizes or shapes from those in prediction units 1060.
In other words,
<img file="MX341305B_D0054.tif" />
INDUSTRIAL transformation and the prediction units of a coding unit are independently determined. Accordingly, video encoding and decoding apparatus 100 and 200 can perform intra-prediction, motion estimation, motion compensation, transformation, and inverse transformation individually in one data unit in the same data unit. coding.
Consequently, encoding is performed recursively on each of the encoding units that have a hierarchical structure in each region in each maximum encoding unit to determine an optimal encoding unit, and thus encoding units that have a structure. recursive tree can be obtained.
The encoding information may include the division information regarding a coding unit, the information regarding a partition type, the information regarding a prediction mode, and the information regarding a size of a transformation unit.
Table 1 shows the encoding information that can be set by the video encoding and decoding apparatus 100 and
200.
IMPI
MEXICAN INSTITUTE OF INDUSTRIAL PROPERTY
<img file="MX341305B_D0055.tif" />
Table 1
<td></td><td colspan="2">Information of</td><td>Division 0</td><td></td><td>information</td>
<td colspan="3">(Codification on the Unit of</td><td colspan="2">Coding that has Tstmáño</td><td>from DivisTefr *</td>
<td></td><td colspan="4">2N x 2N and Current Depth of d)</td><td> 1</td>
<td>Mode of</td><td>Kind of</td><td>Partition</td><td colspan="2">Unit Size</td><td>Encode</td>
<td>Prediction</td><td></td><td></td><td colspan="2">Transformation</td><td>Repeatedly</td>
<td></td><td>Kind of</td><td>Kind of</td><td>information</td><td>information</td><td>and the</td>
<td>Intra</td><td>Participated</td><td>Partition</td><td>Division 0</td><td>Division 1</td><td>Units of</td>
<td>Inter</td><td>n</td><td>Asymmetric</td><td>of the Unit</td><td>of the Unit</td><td>Coding</td>
<td></td><td>Symmetric</td><td>to</td><td>of</td><td>of</td><td>that have</td>
<td></td><td>to</td><td></td><td>Transformation</td><td>Transformation</td><td>Depth</td>
<td></td><td></td><td></td><td>n</td><td>n</td><td>s Lower</td>
<td>Jump</td><td>2N x 2N</td><td>2N X nU</td><td></td><td>N x N</td><td>from d + 1</td>
<td>(Onicament</td><td>2N x N</td><td>2N x nD</td><td></td><td>(Kind</td><td></td>
<td>e 2N X 2N)</td><td>N x 2N</td><td>nL X 2N</td><td>2N X 2N</td><td>Symmetrical)</td><td></td>
<td></td><td>NxN</td><td>nR x 2N</td><td></td><td>N / 2 x N / 2</td><td></td>
<td></td><td></td><td></td><td></td><td>(Kind</td><td></td>
<td></td><td></td><td></td><td></td><td>Asymmetric)</td><td></td>
The transmitter 120 of the video encoding apparatus 100 can output the encoding information regarding the encoding units having a tree structure, and the receiving and extracting unit 210 of the video decoding apparatus 200 can extract the information encoding with respect to encoding units that have a tree structure from a received bitstream.
The division information indicates whether a current encoding unit is divided into encoding units of a lower depth. If the division information of a current depth d is 0, a depth, at which
Mexican INSTITUTE a current coding unit is no longer divf & i, ^^<sup>1</sup>^^ Mgfrgr'gj lowest depth, is a depth cod-i fj nothing. and thus, information regarding a partition type, prediction mode, and a size of a transformation unit can be defined for the coded depth. If the current encoding unit is further divided according to the division information, the encoding is independently performed on four
<td>units</td><td>of</td><td>split encoding of</td><td>a</td><td>depth</td><td>plus</td>
<td>low.</td><td>A</td><td>prediction mode can</td><td>to be</td><td>one of a</td><td>mode</td>
<td>intra, a</td><td colspan="2">inter mode, and a jump mode</td><td>The</td><td>intra mode</td><td>and the</td>
Inter mode can be defined on all partition types, and jump mode is defined only on a partition type that is 2N x 2N in size.
Information regarding partition type can indicate symmetric partition types that have sizes of 2N x 2N, 2N x N, N x 2N, and N x N, and asymmetric partition types that have sizes of 2N x nU, 2N x nD, nL x 2N, and nR x 2N, which are obtained by asymmetrically dividing the height or width of the prediction unit. The asymmetric partition types having the sizes of 2N x nU and 2N x nD can be respectively obtained by dividing the height of the prediction unit into 1: 3 and 3: 1 and the types of
IMPI • INSTITUTO MEXICANO ¢ / asymmetric partition that have the sizes of nIP<sup>AND</sup>k<sup>AND</sup>^.^
2N can be respectively obtained by dividing the width of the prediction unit into 1: 3 and 3: 1.
The size of the transformation unit can be adjusted to be of two types according to the division information of the transformation unit. In other words, if the transformation unit division unit is 0, the transformation unit having a size of 2N χ 2N can be adjusted to be the size of the current encoding unit. If the division information of the transformation unit is 1, the transformation units can be obtained by dividing the current encoding unit. Also, if a current encoding unit partition type that is 2N χ 2N in size is a symmetric partition type, a transformation unit size can be N χ N, and if the unit partition type is current encoding is a type of asymmetric partition, the transformation unit size can be N / 2 χ N / 2.
The coding information regarding the coding units having a tree structure can include at least one of a coding unit corresponding to a coded depth, a prediction unit and a minimum unit. Coding unit
IΜ Ρ ϊ
INSTITUTE A 'EX ICANO í corresponding to the coded depth can be less than one of a prediction unit and a nnid ^. minimum gu<sub>P</sub> it contains the same encoding information.
Accordingly, it is determined whether or not the adjacent data units are included in the same encoding unit corresponding to the encoded unit, by comparing the encoding information of the adjacent data units. Also, a corresponding encoding unit, which corresponds to an encoded depth, is determined by using the encoding information of a data unit and thus a distribution of the encoded depths in a maximum encoding unit can be determined. .
Accordingly, if a current encoding unit is predicted based on the encoding information of the adjacent data units, the encoding information of the data units in the deepest encoding units adjacent to the current encoding unit may be directly referred to and used.
Alternatively, if a current encoding unit is predicted based on the encoding information of the adjacent data units, the data units adjacent to the current encoding unit are searched using encoded information from the data units, and
ΜΡ1
INSTITUTO MEXICANO the searched adjacent coding units<sup>13</sup>^^^^ 5¾¾¾¾ 'referenced to predict current encoding unit.
Figure 13 is a diagram for describing a relationship between a coding unit, a prediction unit, or a partition, and a transformation unit, according to the coding mode information of the
Table l.
A maximum 1300 encoding unit includes encoding units 1302, 1304, 1306, 1312, 1314, 1316 and
1318 from the coded depths. Here, since the coding unit 1318 is a coding unit of a coded depth, the division information can be set to 0. The information regarding a partition type of the coding unit 1318 having a size of 2N x 2N can be set to be one of a 1322 partition type that has a size of 2N x 2N, a type of
<td>partition</td><td> 1324</td><td>than</td><td>has</td><td>a</td><td>size</td><td>of</td><td>2N</td><td>X</td><td>N,</td><td>a</td><td>type</td><td>of</td>
<td>partition</td><td> 1326</td><td>than</td><td>has</td><td>a</td><td>size</td><td>of</td><td>N</td><td>X</td><td>2N,</td><td>a</td><td>type</td><td>of</td>
<td>partition</td><td> 1328</td><td>than</td><td>has</td><td>a</td><td>size</td><td>of</td><td>N</td><td>X</td><td>N,</td><td>a</td><td>type</td><td>of</td>
<td>partition</td><td> 1332</td><td>than</td><td>has</td><td>a</td><td>size</td><td>of</td><td>2N</td><td>X</td><td>wildebeest,</td><td>a</td><td>type</td><td>of</td>
<td>partition</td><td> 1334</td><td>than</td><td>has</td><td>a</td><td>size</td><td>of</td><td>2N</td><td>X</td><td>nD,</td><td>a</td><td>type</td><td>of</td>
partition 1336 that has a size of nL x 2N, and a partition type 1338 that has a size of nR x N.
When the partition type is set to be symmetrical, that is, the partition type 1322, 1324, 1326 or
<img file="MX341305B_D0056.tif" />
<img file="MX341305B_D0057.tif" />
<img file="MX341305B_D0058.tif" />
1328, a transformation unit 1342 what
INDUSTRIAL 2N χ 2N is adjusted if the division information (TU size flag) of a transformation unit is 0, and a 1344 transformation unit having a size of N χ N is adjusted to a TU size flag is one.
When the partition type is set to be asymmetric, that is, the partition type 1332, 1334, 1336 or
1338, a 1352 transformation unit that has a size of 2N tamaño 2N is set if a flag of TU size is 0, and a 1354 transformation unit that has a size of N / 2 χ N / 2 is adjusted if a flag TU size is 1.
Figure 14 illustrates the maximum encoding units, the encoding units according to a tree structure obtained by subdividing each maximum encoding unit, the prediction units obtained by subdividing each encoding unit, and the data units including the transformation units. In other words, Figure 14 illustrates the data units 1400 including the encoding units, the prediction units, and the transformation units of Figures 10 through 12 described above, by overlapping the encoding units, the prediction units, and transformation units. In Figure 14, it is assumed that a maximum encoding unit size is 32 x 32, and the encoding units of the encoded depths
INSTITUTO MEXÍC / ϊΝΟ DE LA PROPERTY are shown in solid lines and the units of ^ TeELtccM and the transformation units obtained from 51 divide — tsar encoding units from the encoded units, are shown in long and short dashed lines alternated in the units of coding. Also, the encoded depths of encoding units that have a tree structure are 0, 1, and 2, where a maximum depth is 3.
Referring to Figure 14, the 1400 data units include the encoded depth encoding units of 9 maximum encoding units that are 3 2 x 32 in size. As described above, an optimal coding unit is determined by recursively performing coding on coding units that have a hierarchical structure classified according to depths, for each maximum coding unit, while an optimal prediction unit and a Optimal transformation unit for prediction and transformation can be determined according to each coding unit. Unlock filtering unit 130 determines a filtering limit at which unlocking filtering is to be performed with at least one data unit 1400 from among the encoding units, prediction units, and transformation units shown in the Figure 14.
IMPIOS
Mexican institute
In detail, the filter unit of
0 You can determine a filter limit bathed in the limits of data units that are predetermined or larger in size between encoding units, prediction units, and transformation units. In other words, referring to Figure 14, when a current image is divided into maximum encoding units having a size of 32 x 32, each maximum encoding unit is again divided into encoding units having a hierarchical structure classified according to to the depths, and each coding unit is again divided into prediction units and transformation units, which are smaller in size than the coding units, According to the depths for prediction and transformation, only a limit of the data units that have a predetermined size or greater between the limits of the encoding units, the prediction units and the transformation units can be determined for be a filtration limit at which the unlocking filtration will be performed.
Figures 15 through 17 each illustrate the filtering limits determined based on the limits of the data units having a predetermined or larger size, relative to the 1400 data units of the
IMPI
MEXICAN INSTITUTE DR THE PROPERTY
<img file="MX341305B_D0059.tif" />
Figure 14.
Referring to Figures 14 and 15, when the encoding units, prediction units, and transformation units of Figure 14 are determined, the unlock filter unit 130 can determine only a limit of the data units having a predetermined or larger size within the limits of the encoding units, the prediction units, and the transformation units as a filtration limit at which the unblocking filtration is to be performed. For example, the unlock filter unit 130 may determine only the limits of the encoding units, the prediction units, and the transformation units, which are equal to or greater than 32 x 32 in size, as the filter limits at which will be carried out the unblocking filtration, as shown in Figure 15, determining only the limits of the coding units, the prediction units, and transformation units having a size equal to or greater than 16 x 16, such as the filtration limits at which the unblocking filtration is to be performed, as shown in Figure 16, or to determine only the limits of the encoding units, prediction units, and transformation units, which have a size equal to or greater than 8 x 8, such as the limits of
<img file="MX341305B_D0060.tif" />
filtration to which the leT ^ aSOásiíLó will be performed
INDUSTRIAL unlocking as shown in Figure 17. As such, when only the limits of the data units having a predetermined size or greater are determined to be the filtering limits at which the unlocking filtering is to be performed, a The filtration limit at which the filtration will be carried out is changed with respect to the same form of division. For example, when unlock filtering is performed only over the boundaries of data units that are 32 x 32 or larger in size as shown in Figure 15, a. Inner boundary is not considered as a filtering boundary, except for a boundary portion that overlaps with a maximum encoding unit 1510 that is 32 x 32 in size from the boundaries of the encoding units, transformation units, and the prediction units obtained by dividing the maximum encoding unit 1510.
On the other hand, when unlock filtering is performed only over the limit of data units that are 16 x 16 in size or above, as shown in Figure 16, the internal limits of encoding units 1611 to the 1614 obtained by dividing a maximum coding unit 1610 corresponding to the maximum coding unit 1510 of Figure 15, are also determined as the filtration limits.
Meanwhile, a
<img file="MX341305B_D0061.tif" />
. ,,, 'BlSriIUTO MEXICANO unit f íaEWaoapem
INDUSTRIAL
<img file="MX341305B_D0062.tif" />
Unlock 130 does not determine a limit on data units that are predetermined or larger in size, such as a filter limit if the limit is a frame limit. In other words, unlock filtering according to an embodiment of the present invention is not performed at an outermost boundary corresponding to an edge of an image.
Figure 18 is a reference diagram for describing an unlock filtration process according to an embodiment of the present invention, based on the filtration limits of Figure 17.
When the filtering limits at which the unlocking filtering is to be performed are determined based on the limits of the data units having a predetermined size or above, the unlocking filtering unit 130 determines the strength of the filtration at the limits of the filtration, based on a prediction mode of a coding unit to which the adjacent pixels belong, based on the filtration limit and the transformation coefficient values of the pixels adjacent to the filtration limit.
Hereinafter, a process of performing deblocking filtration based on the filtration limits, such as a filtration limit of, will be described. IMPIí
MEXICAN INSTITUTE '
OF THE PROPERTY horizontal direction 1810 and a limit of f ilti'PáÓ'iÓn vertical direction 1920 of Figure 18. .......
Figures 19A and 19B illustrate the pixels located above the filtration limits, to describe unblocking filtration according to an embodiment of the present invention.
Referring to Figure 19A, the pixel values of the 1910 pixels adjacent to the top and bottom of a horizontal direction filtering boundary before unlock filtering are defined to be pO to p4 and qO to q4. . Also, the pixel values of pixels 1920 adjacent to the top and bottom of the horizontal direction filtering limit after unlock filtering, are defined to be pO 'to p4' and q0 'to q4'. Similarly, referring to Figure 19B, the pixel values of the 193 0 pixels adjacent to the left and right of a vertical direction filtering boundary before unlock filtering are defined to be pO to p4 and qO to q4. Also, the values of pixels 1940 adjacent to the left and right of the vertical direction filtering limit after unlock filtering are defined to be pO 'to p4' and qO 'to q4'. Unlock filtering operations based on horizontal and vertical direction filtering limits, are
<img file="MX341305B_D0063.tif" />
or> ''
MEXICAN INSTITUTE Ά <sub>B</sub> PROPERTY OR »». '., / Ζ, Α identical except for a difference of address. <sup>1NDUSTWAL</sup>
The filtering unit unblocked 'TJU dététftiftá' the filtering force based on whether the prediction mode of the encoding unit to which the adjacent pixels belong, based on the filtering limit, is either an intra mode or a Inter mode, and if the values of the transformation coefficient of the pixels adjacent to the filtration limit are 0. When the limit force (Bs) denotes the filtration force, Bs can be classified in 5 stages from 0 to 4 . A size of Bs is proportional to the filtration force. In other words, when Bs = 4, the filtration force is the strongest, and when Bs = 0, the filtration force is the weakest. Here, unlock filtering may not be performed when Bs = 0.
In detail, when pO and qO denote pixels that are adjacent to the filtration limit and are divided based on the filtration limit, the unlocking filtration unit 130 can determine the filtration force to have a value of Bs = 4 when a prediction mode of at least one coding unit to which pO belong and qO is in an intra mode and the filtering limit is a limit of the coding units. For example, when the deblocking filtration is performed based on the horizontal direction filtering limit 1810 of Figure 18, the deblocking filtration unit 130 can determine the
<img file="MX341305B_D0064.tif" />
IMPI
INSTITUTE and. PROPERTY EX ICAN Filtration force to have a value of Bs = 4, dtfSSgf predicted mode of at least one of mine - '' racted - decoding 184 0 to which belongs pO, and a coding unit 1830 to which It belongs to qO, which are the closest pixels based on the 1810 horizontal direction filtering limit , it is in an intra mode and a current filtering limit is a limit of the encoding units. Here, if the horizontal direction filtering limit 1810 is a limit of the encoding units, this can be determined based on whether the encoding unit 184 0 to which pO belongs and the encoding unit 183 0 to which it belongs qO are data units based on the deepest encoding units, and whether or not the horizontal direction filtering limit 1810 corresponds to a limit of the prediction units and transformation units of Figures 11 and 12 obtained by dividing the coding units for prediction and transformation.
Alternatively, the deblocking filter unit 130 determines the strength of the filtering to have a value of Bs = 3, when the prediction mode of at least one of the encoding units to which pO and qO belong is in an intra mode and the filtration limit is not a limit of the encoding units.
Alternatively the filter unit of
<img file="MX341305B_D0065.tif" />
,,,,, INSTITUTO MEXICANO unlocking 13 0 determines the strength of the f to have a value of Bs = 2, when the prediction modes of the coding units to which pO and qO belong
<td>are not the ways</td><td>intra</td><td>and a value</td><td>of the</td><td>coefficient</td><td>of</td>
<td>transformation of</td><td colspan="2">at least one of</td><td>the</td><td>units</td><td>of</td>
<td>transformation at</td><td>which</td><td>they belong to</td><td>and what</td><td>, is not 0.</td><td></td>
<td colspan="2">Alternatively,</td><td>unit</td><td>of</td><td>filtration</td><td>of</td>
<td colspan="2">unlock 130 determines</td><td>the force of</td><td colspan="3">filtration for</td>
have a value of Bs = 1, when the prediction modes of the encoding units to which pO and qO belong are not in intra modes, the values of the transformation coefficient of the transformation units to which pO and qO belong, they are 0, and either of a reference structure and a motion vector used for predicting the motion of the prediction units to which pO and qO belong, is different from each other.
Alternatively, the unlock filter unit 130 determines the strength of the filter to have a value of Bs = 0 when the prediction modes of the encoding units to which pO and qO belong are not in intra modes, the coefficient values of transformation of the transformation units to which pO and qO belong, are 0, and the reference structure and the motion vector used to predict the movement of the prediction units at which JKSXS
PI. INSTITUTO MEXICANt belong to PO and qO, they are the same.<sup>OF</sup> Industry
Meanwhile, the unlock filter unit 130 can determine whether the unlock filter is performed over a filter limit based on the filter strength, and a result of comparing a predetermined threshold value, and a difference between the absolute values of the pixel values of a predetermined number of adjacent pixels based on the limit of the filtration. In detail, the unlock filter unit 130 determines the unlock filter performance only when an absolute value of a difference between the pixel values of the pixels adjacent the filter limit and divided based on the filter limit and a value absolute of a difference between the pixel values of the adjacent pixels on the same side based on the filtering limit, they are smaller than a predetermined threshold value, determined according to a parameter of quantification of the transformation units to which the pixels belong, and the filtration force is not the weakest. For example, the unlock filter unit 130 can perform the unlock filter over a filter limit only when i) the filter force Bs is not 0 and ii) a condition of | p0-g0 | <a; | pl-gO | <β; | gl-gO | <β is satisfied. Here, the threshold values can be
<img file="MX341305B_D0066.tif" />
predetermined based on the dáge »parameters used during the quantification of the transformation units to which pO and qO belong.
Figures 20A and 20B are reference tables for determining threshold values α and β, according to an embodiment of the present invention.
With reference to Figures 20A and 20B, the threshold values α and β are determined based on indexA and indexB, where indexA and indexB are values determined according to the equations indexA = Clip3 (0, 51, qPav + FilterOffsetA) and indexB = Clip3 (0, 51, qPav + FilterOffsetB). Here, qPav denotes an average value of the quantization parameters of the transformation units to which pO and qO belong.
clip3 (a, b, c) denotes a truncation operation of a value of c such that a <c <b. For example, the equation indexA = Clip3 (0,
51, qPav + FilterOffsetA) is 0 when a value of qPav + FilterOffsetA is less than or equal to 0 and is 51 when the value of qPav + FilterOf fsetA is equal to or greater than 51. Also, FilterOffsetA and FilterOffsetB denote default offset values. adjusted respectively with respect to threshold values α and β.
Figure 21 is a reference diagram for describing a process of adjusting an offset value with respect to a threshold value a, according to a
The embodiment of the present invention. By axis »n®2uiP / 'Exouanac<sup>c J r</sup>OF LArkOI lEDAD V *
INDUSTRIAL threshold value a is adjusted proportionally to a quantization parameter such as a curve shown in Figure 21, an indexA value is increased as a FilterOffsetA value is increased, and as a result, a curve adjusted as a default is moved in a direction to the right (direction b) as shown in Figure 21. When the FilterOffsetA value is decreased, the indexA value is decreased, and thus the fit of the curve as an omission is moved in a left direction (direction A) as shown in Figure 21. As such, FilterOffsetA and FilterOffsetB can be used to adjust the strength of the filtering.
As such, the unlock filter unit 130 determines whether the unlock filter is performed over a filter limit based on the quantization parameters of the transformation units to which pO and qO belong and the predetermined offset values.
With respect to a limit at which the blocking filtration is to be carried out, the unlocking filtration unit 130 determines a number and the filter derivation coefficients of the pixels to be filtered adjacent to a filtration limit, based on the force Filtering, an absolute value of a difference between the pixel values of adjacent pixels
INDUSTRIAL filtering and divided based on the filtering limit, and an absolute value of a difference between the pixel values of the adjacent pixels on the same side based on the filtering limit. Also, the unlock filter unit 130 performs filtering by changing the pixel values of the pixels to be filtered via a weighted sum based on the filter bypass coefficients.
In detail, when the filtration force of a current filtration limit is Bs <4, the unlock filter unit 130 generates pl ', p0', q0 ', and ql' by using a finite impulse response filter 4-lead (FIR) using pl, pO, qO, and ql as inputs. Unlock filter unit 130 generates a delta value Δ according to an equation A = clip3 [-tc, tc ((((q0-p0) «2) + plql) * 4))) 3)). Here, it can be determined based on | p2-p0 |, | q2-q0 | and a threshold value β.
Figures 22A and 22B are reference tables used to determine a predetermined intermediate value (te) used during an unblocking filtration process, according to an embodiment of the present invention. As shown in Figures 22A and 22B, it can be determined by using predetermined values in a table according to the indexA and the filtering force bS.
. IMPI
MEXICAN INSTITUTE
Also, the deH & $ §g $$ o filtration unit generates the pixel values of pO and qO that are closest to a filtration limit and filtered by unlocking according to the equations ρΟ '= ρΟ + Δ and q0' = q0 + á. The pixel values of pl and ql that are adjacent to the filtration limit after pO and qO are changed according to the equations pl '= pl + Á / 2 and ql' = ql + A / 2.
Meanwhile, when the filter force has a value of Bs = 4 at a current filter limit, the unlock filter unit 130 determines a number and the filter bypass coefficients of the pixels that are filtered adjacent to the filter limit , based on an absolute value of a difference between the pixel values of the pixels adjacent to the filter limit and divided based on the filter limit and an absolute value of a difference between the pixel values of the pixels adjacent to the same side , based on the filtration limit. In detail, when | ρ2-ρθ | <β; | <p0-q0 | <round (a / 4), the unlock filter unit 130 sets the input pixel values to be p2, pl, pO, qO and ql with respect to the pO closest to the filter limit, and generates pO 'which is Filtered by using a 5-lead filter that has a filter bypass coefficient of {1,2,2,2,1}.
<img file="MX341305B_D0067.tif" />
Regarding nearest pl
<img file="MX341305B_D0068.tif" />
riTUTdnWirt ^ Ak ^
OF INDUSTRIAL PROPERTY
<img file="MX341305B_D0069.tif" />
After pO filtering, the unlock filter unit 130 adjusts the input pixel values to be p2 pO and ql, and generates pl 'filtering by using a 4-bypass filter having a filter bypass coefficient of { 1,1,1,1}.
With respect to p2 closest to the filter limit after pl, the unlock filter unit 130 adjusts the input pixel values to be p3, p2, pl, pO, and ql, and generates filtered p2 'by using a 5-lead filter that has a filter lead coefficient of {2,3,1,1,1}.
Similarly, when | q2-q0 | <β; | pO-qO | <round (a / 4), the unlock filter unit 130 adjusts the input pixel values to be q2, ql, qO, pO, and pl with respect to qO closest to the filter limit, and generates qO 'filter by using a 5-tap filter that has a filter tap coefficient of {1,2,2,21}.
With respect to ql closest to the filter limit after qO, the unlock filter unit 130 adjusts the input pixel values to be q2, ql, qO, and pl and general ql 'filtered by using a 4 filter. leads that has a filter lead coefficient of {1,1,1,1}.
<img file="MX341305B_D0070.tif" />
130
Regarding q2 closest to the limit<sup>, Ns</sup>^ L ° áí ^ 'í ^ £ al
INDUSTRIAL after ql, the unlock filter unit adjusts the input pixel values to be q3, q2, ql and pO, and generates q2 'filtered by using a 5-lead filter that has a filter bypass coefficient of {2,3,1,1,1}.
The unlocking filtering unit 230 of the video decoding apparatus 200 according to one embodiment determines a filtering limit at which the unlocking filtering will be carried out starting from the limits of at least one data unit among the data units. coding according to a tree structure, the prediction units and the transformation units, Determines the strength of the filtration at the filtration limit based on a prediction mode of the encoding units to which the adjacent pixels belong based on the filtration limit and the values of the transformation coefficient of the pixels adjacent to the limit filter, and performs unlock filtering on the decoded image data, based on the strength of the filtering, by using the information regarding recognized unlock filtering from a bit stream. Since the operations of the 230 Unlock Filtration Unit of the 200 Unlock Filtration Unit are similar to those of the
<img file="MX341305B_D0071.tif" />
<img file="MX341305B_D0072.tif" />
unlocking filter unit 12 0 deiJsTiTuapñciEafeo
OF LATEO! INDUSTRIAL AGE video encoding 100, the detailed descriptions of them will not be repeated.
Figure 23 is a flowchart illustrating a video encoding method, based on the encoding units, according to an embodiment of the present invention.
Referring to Figure 23, at step 2310, the encoder unit determiner 110 divides an image into at least one maximum encoding unit which is a data unit having a maximum size. Then, in step 2320, the encoder unit determiner 110 determines the encoder units that are hierarchically configured according to depths indicating a number of times that at least one maximum encoding unit is spatially divided, and the prediction units and transformation units respectively for prediction and transformation of encoding units.
In step 2330, the unlock filter unit 130 determines a filter limit on which the unlock filter is to be performed based on at least one data unit among the encoding units, the prediction units and the transformation units. As previously described, the filtration limit can be determined based on u:
<img file="MX341305B_D0073.tif" />
<img file="MX341305B_D0074.tif" />
INDUSTRIAL PROPERTY CANC data units that have a predetermined size or above.
The unlocking filter unit 130 determines the strength of the filtering at the filtering limit based on a prediction mode in a coding unit to which the adjacent pixels belong, based on the filtering limit, and the values of the transformation coefficient of the pixels adjacent to the filtering limit in step 2340, and performs the unlock filtering based on the determined filtering force, in step 2350.
Figure 24 is a flowchart illustrating a method of decoding a video, based on the encoding units, in accordance with an embodiment of the present invention.
Referring to Figure 24, in step 2410, the receiving and extracting unit 210 extracts the encoded image data according to the encoding units, the encoding mode information relative to the encoding units according to a structure of tree, and the information regarding the unlocking filtration in a maximum coding unit, according to the coding units according to the tree structure included in each unit of
<img file="MX341305B_D0075.tif" />
encoding obtained by dividing a * 4Llh * gfen ^ a '
MEXICAN INSTITUTE OF PROPERTY for recognition of a current of recent bits<sup>1</sup>'.
At step 2420, the decoder '' 220 'determines the prediction units and transformation units for prediction and transformation according to the encoding units, and decodes the encoded image data, based on the information from the coding mode with respect to the coding units according to the tree structure.
In step 2430, the unlock filter unit 230 determines a filter limit at which the unlock filter will be performed from within the limits of at least one data unit from among the encoding units according to the tree structure , the prediction units and the transformation units, by using the information regarding the unlocking filtration.
In step 2440, the unlock filter unit 230 determines the filter strength of the filter limit based on a prediction mode of an encoding unit to which the adjacent pixels based on the determined filter limit belong, and the values of the transformation coefficient of the pixels adjacent to the filtration limit.
In operation 2450, the IMPI filter unit
MEXICAN INSTITUTE: OF PROPERTY
Γ INDUSTRIAL unlocking 230 performs unlocking filtering on the decoded image data, based on the determined filtering force.
The embodiments of the present invention can be written as computer programs and can be implemented in general-purpose digital computers that run the programs using a computer-readable recording medium. Examples of computer-readable recording media include magnetic storage media (eg ROM, floppy disk, hard drive, etc.) and optical recording media. (for example, CD-ROMs, or DVDs).
While this invention has been particularly shown and described with reference to the preferred embodiments thereof, it may be understood by those of ordinary skill in the art that various changes in form and detail may be made therein without departing of the spirit and scope of the invention as defined by the appended claims. Preferred modalities should be considered in a descriptive sense only and not for purposes of limitation. Therefore, the scope of the invention is defined not by the detailed description of the invention, but by the appended claims, and all differences within the scope will be considered as included herein.
<img file="MX341305B_D0076.tif" />
MEXICAN INSTITUTE
OF THE PROPERTY
INDUSTRIAL invention.
<img file="MX341305B_D0077.tif" />
It is noted that in relation to this date, the best method known by the applicant to put the aforementioned invention into practice is the one that is clear from the present description of the invention.
Contents38
97 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 Sheet 78 Sheet 79 Sheet 80 Sheet 81 Sheet 82 Sheet 83 Sheet 84 Sheet 85 Sheet 86 Sheet 87 Sheet 88 Sheet 89 Sheet 90 Sheet 91 Sheet 92 Sheet 93 Sheet 94 Sheet 95 Sheet 96 Sheet 97
163 members in 13 offices
Priority claims9
| Document | Office | Kind | Date |
|---|---|---|---|
| 32344910 | United States of America | P | |
| 32344910 | United States of America | P | |
| 61323449 | United States of America | – | |
| 2011002647 | Republic of Korea | W | |
| 2011002647 | Republic of Korea | W | |
| 61323449 | – | – | – |
| PCTKR2011002647 | – | – | – |
| US20100323449P | – | – | – |
| WO2011KR02647 | – | – | – |
Members163
| Document | Office | Kind | |
|---|---|---|---|
| KR20110114495A | Republic of Korea | A | |
| KR20110114496A | Republic of Korea | A | |
| KR20110114498A | Republic of Korea | A | |
| CA2796203A1 | Canada | A1 | |
| CA2796364A1 | Canada | A1 | |
| CA2796368A1 | Canada | A1 | |
| CA2922690A1 | Canada | A1 | |
| CA2922693A1 | Canada | A1 | |
| CA2962290A1 | Canada | A1 | |
| CA2992235A1 | Canada | A1 | |
| CA3007527A1 | Canada | A1 | |
| CA3007544A1 | Canada | A1 | |
| WO2011129619A2 | World Intellectual Property Organization (WIPO) | A2 | |
| WO2011129620A2 | World Intellectual Property Organization (WIPO) | A2 | |
| WO2011129621A2 | World Intellectual Property Organization (WIPO) | A2 | |
| WO2011129619A3 | World Intellectual Property Organization (WIPO) | A3 | |
| WO2011129620A3 | World Intellectual Property Organization (WIPO) | A3 | |
| WO2011129621A3 | World Intellectual Property Organization (WIPO) | A3 | |
| AU2011241282A1 | Australia | A1 | |
| AU2011241283A1 | Australia | A1 | |
| AU2011241284A1 | Australia | A1 | |
| MX2012011846A | Mexico | A | |
| MX2012011772A | Mexico | A | |
| EP2547107A2 | European Patent Office (EPO) | A2 | |
| EP2547108A2 | European Patent Office (EPO) | A2 | |
| MX2012012017A | Mexico | A | |
| US2013028331A1 | United States of America | A1 | |
| US2013034155A1 | United States of America | A1 | |
| CN102934436A | China | A | |
| EP2560387A2 | European Patent Office (EPO) | A2 | |
| CN102948145A | China | A | |
| CN102948146A | China | A | |
| US2013077682A1 | United States of America | A1 | |
| JP2013524715A | Japan | A | |
| JP2013524716A | Japan | A | |
| JP2013524717A | Japan | A | |
| RU2518996C1 | Russian Federation | C1 | |
| EP2547107A4 | European Patent Office (EPO) | A4 | |
| EP2547108A4 | European Patent Office (EPO) | A4 | |
| EP2560387A4 | European Patent Office (EPO) | A4 | |
| RU2012148135A | Russian Federation | A | |
| RU2012148138A | Russian Federation | A | |
| RU2528132C2 | Russian Federation | C2 | |
| AU2011241282B2 | Australia | B2 | |
| KR20150000855A | Republic of Korea | A | |
| KR20150009501A | Republic of Korea | A | |
| KR20150009502A | Republic of Korea | A | |
| AU2011241283B2 | Australia | B2 | |
| AU2011241284B2 | Australia | B2 | |
| RU2014105979A | Russian Federation | A | |
| RU2014105981A | Russian Federation | A | |
| JP5832519B2 | Japan | B2 | |
| US9247247B2 | United States of America | B2 | |
| RU2014127502A | Russian Federation | A | |
| RU2014128147A | Russian Federation | A | |
| KR101602873B1 | Republic of Korea | B1 | |
| KR101604031B1 | Republic of Korea | B1 | |
| KR101604036B1 | Republic of Korea | B1 | |
| KR20160030153A | Republic of Korea | A | |
| KR101604032B1 | Republic of Korea | B1 | |
| US2016094845A1 | United States of America | A1 | |
| JP2016048929A | Japan | A | |
| CA2796203C | Canada | C | |
| CA2796368C | Canada | C | |
| CN102934436B | China | B | |
| CN105721867A | China | A | |
| JP2016129363A | Japan | A | |
| CN105812801A | China | A | |
| CN105847799A | China | A | |
| MX341305BThis record | Mexico | B | |
| CN105898298A | China | A | |
| RU2597471C2 | Russian Federation | C2 | |
| RU2598586C2 | Russian Federation | C2 | |
| KR20160119441A | Republic of Korea | A | |
| CN102948146B | China | B | |
| KR101665031B1 | Republic of Korea | B1 | |
| US9485511B2 | United States of America | B2 | |
| CN106131549A | China | A | |
| KR101676790B1 | Republic of Korea | B1 | |
| US2016337658A1 | United States of America | A1 | |
| CN106162178A | China | A | |
| CN106162179A | China | A | |
| KR20160134613A | Republic of Korea | A | |
| CN106231342A | China | A | |
| RU2608672C2 | Russian Federation | C2 | |
| US9565438B2 | United States of America | B2 | |
| RU2612576C2 | Russian Federation | C2 | |
| CN102948145B | China | B | |
| JP2017079486A | Japan | A | |
| US2017118469A1 | United States of America | A1 | |
| US9654790B2 | United States of America | B2 | |
| CN106686379A | China | A | |
| KR101743247B1 | Republic of Korea | B1 | |
| CN106803957A | China | A | |
| KR20170063495A | Republic of Korea | A | |
| CA2796364C | Canada | C | |
| BR112012026191A2 | Brazil | A2 | |
| US9712822B2 | United States of America | B2 | |
| US9712823B2 | United States of America | B2 | |
| EP3200454A1 | European Patent Office (EPO) | A1 |
Numbers
- Publication
- 341305
- Publication, DOCDB
- 341305
- Publication, EPODOC
- MX341305
- Application
- 2015014848
- Application, DOCDB
- 2015014848
- Application, EPODOC
- MX20150014848
Titles
- Spanish
- METODO DE CODIFICACION DE VIDEO Y APARATO DE CODIFICACION DE VIDEO Y METODO DE DECODIFICACION DE VIDEO Y APARATO DE DECODIFICACION DE VIDEO, QUE REALIZAN LA FILTRACION DE DESBLOQUEO CON BASE EN UNIDADES DE CODIFICACION EN ESTRUCTURA DE ARBOL.
Classification
- CPC, 31
- H04N19/117
- H04N19/119
- H04N19/86
- H04N19/44
- H04N19/59
- H04N19/139
- H04N19/147
- H04N19/159
- H04N19/176
- H04N19/61
- H04N19/82
- H04N19/96
- H04N19/197
- H04N19/50
- H04N19/513
- H04N19/52
- H04N19/593
- H04N19/169
- H04N19/60
- H04N19/70
- H04N19/103
- H04N19/134
- H04N19/172
- H04N19/196
- H04N19/46
- H04N19/85
- G06T9/40
- H04N19/146
- H04N19/80
- H04N19/184
- H04N19/30
- IPC, 2
- H04N19 50
- H04N19 117