KR20130105402A - Method of multi-view video coding and decoding based on local illumination and contrast compensation of reference frames without extra bitrate overhead - Google Patents

Abstract

PURPOSE: A multiview video coding/decoding method is provided to improve the compression rate. CONSTITUTION: A pixel value of a reference block within a reference frame and a pixel value of a current block within an encoded frame are received. A pixel adjacent to the reference block of the reference frame and a pixel adjacent to the current block of currently encoded frame are decoded and a recovered value is received. A relationship between the value of the pixel adjacent to the reference block and the recovered value of the pixel adjacent to the current block and the value of the pixel adjacent to the reference block and the value of the pixel adjacent to the reference value are determined. Difference compensation lighting, contrast compensation lighting and contrast compensation parameter between the reference block and the encoded block are determined based on the value of the pixel adjacent to the reference block, the recovered value of the pixel adjacent to the current block and the determined relationships. The lighting and contrast compensation between the reference block and encoded block using the determined lighting and contrast compensation parameter are performed. [Reference numerals] (101) Original data view; (102) Decoded view; (103) Decoded depth; (104) View synthesis; (105) Hybrid multi view video encoder; (106) Reference image management; (107) Inter estimation; (108) Intra estimation; (109) Compensation; (110) Space conversion; (111) Rate distortion optimization; (112) Entropy coding

Description

METHODO OF MULTI-VIEW VIDEO CODING AND DECODING BASED ON LOCAL ILLUMINATION AND CONTRAST COMPENSATION OF REFERENCE FRAMES WITHOUT EXTRA BITRATE OVERHEAD}

The following embodiments relate to an illumination and contrast compensation method applied to a frame including a multi-view video sequence, and more specifically, to an illumination and multi-view video coding method. A contrast compensation method.

One of the multi-view video coding methods uses a frame from an adjacent view or a frame synthesized from an adjacent view as a reference frame for predictive coding. In predictive coding, the displacement of an object between a currently coded frame and one of the reference frames is compensated. "Displacement" refers to the motion of an object or the difference in position of an object in a frame synthesized for a current coded view or a frame from an adjacent view. The difference is coded in subsequent encoding steps (e.g., the difference is coded, quantized and transformed by an entropy coder).

There is a difference in the light flux registered by a particular camera and in the parameters of the camera used for multi-view video sequences capturing. The presence of these differences results in lighting and contrast differences between frames from adjacent views. Moreover, this illumination and contrast difference affects the properties of the synthesized frames. This can seriously degrade the compression efficiency of the predictive coding of multi-view video sequences.

To solve this problem, the H.264 standard uses a weighted prediction technique. The weighted prediction technique was originally developed to compensate for fade-up, fade-down, flickering, or scene change effects in single video sequence coding. . This weighted prediction technique suppresses illumination chages between the coding frame and the reference frame. Weighted prediction is applied to motion and displacement compensation at the macroblock level. The weighting factors are the same for all macroblocks of a particular slice, so they can be thought of as global. The weight factor can be determined and stored as a bit stream ("explicit" weighted prediction) or calculated during decoding ("implicit" weighted prediction). However, for multi-view video sequences, there are local illumination and contrast changes that make this technique inefficient.

Another approach to solving this problem is the adaptive block-wise illumination compensation technique. One variation of these techniques applied to multi-view video coding is called Multi view One-Step Affine Illumination Compensation (MOSAIC). This method combines inter prediction techniques and block-by-block illumination compensation techniques described in the H.264 standard. In each step of the modified inter prediction procedure, an average value for the block currently being coded and the reference block is calculated. Then, mean-removed versions for mentioned above blocks are constructed. Next, the sum of absolute differences (MRSAD) for mean-removed blocks is calculated. The result of the inter prediction is the relative coordinates (displacement vector) of the reference block that minimizes the value of the encoding cost. The calculation of the encoding cost is based on the MRSAD value and the estimation of side information that must be transmitted for further decoding. In addition to the displacement vector, the side information also includes the difference between the mean and mean values of the current and reference block (called DVIC (Difference Value of Illumination Compensation)). In the so-called "P Skip" coding mode (one of the modes used to realize this coding method and one of the modes used for encoding), the DVIC value is the DVIC value of an adjacent macroblock already encoded without any additional information. Can be obtained from Nevertheless, this method cannot avoid transmitting additional side information (DVIC) for decoding.

The parameters of illumination and contrast compensation can be obtained from the restored (already encoded and decoded) areas of the frame. This can reduce the amount of side information that is encoded and explicitly transmitted in the bit stream. This technique is realized by the method of Weighted Prediction using Neighboring Pixels (WPNP). This method uses pixels surrounding the reference block and pixels surrounding the current block to estimate pixel-wise illumination changes. In this case, the illumination change of two neighbor pixels is weighted and added to the estimated illumination change for the pixels of the current and reference block. Weighted coefficients are defined for each pixel of the current block. The value of the weighting coefficient depends on the distance between the pixels of the current block and the neighboring pixels. The main disadvantage of this method is that the reduction of side information is achieved by the potential degradation of the illumination change prediction quality. The reason for the quality reduction is that the illumination change for the neighboring pixels may be different than the illumination change for the pixels of the current and reference block.

Another method of realizing illumination and contrast parameter estimation by analyzing a restored (already coded / decoded) region of a frame is described in US patent application US 2011/0286678. The method of coding a multi-view video sequence includes an illumination compensation step during predictive coding. The illumination compensation (correction) parameter is estimated from the reference region (block) and the currently coded adjacent region. Since adjacent regions can be obtained at the decoding side, there is no need to encode the illumination compensation parameters. The obtained parameter is applied to the illumination compensation of the reference area (block). The reliability of illumination change estimation, using the estimated illumination compensation parameters, corrects adjacent regions of the reference region, which are already coded / decoded neighboring regions of the current coded region (block). Is predicted by comparison with The disadvantage of this method is that reliability is determined only by analysis of adjacent areas. The information contained in the reference area is not used for the analysis of illumination compensation reliability. Therefore, an error may occur during lighting compensation.

Another example is described in US patent application US 2008/0304760. Here, a method for compensating illumination and contrast of a reference block includes receiving an input of a recovered value of a pixel neighboring a reference block and a recovered value of a pixel neighboring a current block; Predicting an average value of pixels of the reference block and the current block based on an input recovered value of a pixel neighboring the reference block and an input recovered value of a pixel neighboring the current block; Based on a predicted mean value of pixels of the current block, a predicted mean value of pixels of the reference block, and values of pixels of the reference block and the current block, lighting compensation of the reference block is performed. Determining an illumination compensation parameter for the; And performing illumination compensation of the reference block by using the determined illumination compensation parameter.

The disadvantages of this prototype are: The recovered values of the reference block and the pixels neighboring the current block are used only for prediction of the mean value. Because of this restriction, information contained in neighboring areas cannot be used. Moreover, no analysis of the relationship between the value of the pixel neighboring the reference block and the value of the pixel from the reference block is performed. As such, this prototype does not take into account possible differences in lighting and contrast parameters between neighboring regions and currently coded blocks. This degrades the reliability of the lighting change compensation and adversely affects the compression efficiency of the predictive coding.

According to the prototype, a method of encoding an image based on lighting compensation includes: determining a reference block to be used to generate a predicted block of a current block; Determining an illumination compensation parameter for illumination compensation of the determined reference block; Performing illumination compensation of the determined reference block using the determined illumination compensation parameter; Generating a prediction block of the current block by using the illumination-compensated reference block and generating a bit stream by encoding a difference value between the current block and the generated prediction block; And storing information about the determined illumination compensation parameter in the determined region of the generated bit stream. The disadvantage of this method is the need to store the illumination compensation parameters when generating the bit stream.

According to one embodiment, the steps of lighting and contrast parameter estimation and compensation can be more reliably processed and a method and apparatus for improving multi-view coding using a hybrid approach can be provided. This improvement can be achieved by using more information for the estimation of the illumination and contrast change parameters. More specifically, embodiments may use the relationship between the recovered value of a pixel neighboring a reference block and the value of a pixel of the reference block and the relationship between the recovered values of pixels neighboring a current and reference block, respectively.

A method for multi-view video encoding and decoding based on illumination and contrast compensation according to one embodiment uses a value of a pixel from an area where illumination change estimation is available on the encoding side as well as on the decoding side, Since the illumination compensation parameters can be accurately reconstructed without transmitting information on the separate information about the illumination compensation parameters, the compression efficiency can be improved.

According to one embodiment, illumination and contrast differences between a reference block and an encoded block in a predicting stage of a multi-view coding process The method for local compensating of (discrepancy)

Receiving a value of a pixel of a reference block in a reference frame and a value of a pixel of a current block in an encoded frame;

Receiving a value of a pixel neighboring the reference block of the reference frame and an already decoded and restored value of a pixel neighboring a current block of a currently coded frame;

Determining a relationship between a value of a pixel neighboring the reference block and a recovered value of a pixel neighboring the current block and a relationship between a value of a pixel neighboring the reference block and a value of a pixel of the reference block;

Based on the value of the pixel neighboring the reference block, the recovered value of the pixel neighboring the current block, the value of the pixel of the reference block, and the determined relationships, the difference between the reference block and the encoded block Determining illumination and contrast compensaton parameters for illumination and contrast compensation of discretion (mismatch) compensation; And

 The method may include performing lighting and contrast compensation of a difference (mismatch) between the reference block and the encoded block by using the determined illumination and contrast compensation parameter.

According to one embodiment, determining the relationship to the pixel of the reference frame and the current coded frame and determining the illumination and contrast compensation parameters,

Calculating statistical characteristics of the value of the recovered pixel neighboring the current block, statistical characteristics of the value of the pixel of the reference block, and statistical characteristics of the value of the pixel neighboring the reference block;

Determining a relationship between the recovered value of a pixel neighboring the reference block and a statistical feature on the value of the pixel of the reference block; And

Determining an illumination and contrast compensation parameter for illumination and contrast compensation of the reference block based on the relationship and the calculated statistical feature.

According to one embodiment, calculating the statistical feature, determining a relationship to the statistical features, and determining an illumination and contrast compensation parameter,

A mean value for a recovered pixel located to the left of the current block and neighboring the current block, a mean value for a recovered pixel located above the current block and neighboring the current block, a pixel of the reference block Calculating an average value for, an average value of pixels located to the left of the reference block and neighboring the reference block, and an average value of pixels located above the reference block and neighboring the reference block;

If there is a restored pixel located to the left of the current block and neighboring the current block and there is a pixel located to the left of the reference block and neighboring to the reference block, located to the left of the reference block and the reference Compute a ratio value between an average value of pixels neighboring a block and an average value of pixels of the reference block, and calculate an average value of the recovered pixels located to the left of the current block and neighbor the current block and the ratio Calculating a product of the values and determining the illumination and contrast compensation parameter as a ratio between the calculated product and an average value for the pixels of the reference block;

If there is a recovered pixel located above the current block and neighboring the current block and there is a pixel located above the reference block and neighboring the reference block, the average value of the pixels of the reference block and above the reference block Calculate a ratio value between an average value of pixels located and neighboring the reference block, calculate a product of the average value and the ratio value of the recovered pixels located above the current block and neighboring the current block, and calculate the calculated product Determining the illumination and contrast compensation parameter as a ratio between an average value for pixels of the reference block; And

Otherwise, calculate an estimate for the mean value of the current block using median adaptive prediction (MAP), and estimate the estimated mean value of pixels of the current block and the reference block. Determining the illumination and contrast compensation parameter as a ratio between the mean values for the pixels.

According to one embodiment, a method for compensating lighting and contrast of a reference block during multi-view coding,

Receiving an input of a value of a pixel of a reference block of a reference frame and a value of a pixel of a current block of a current coded frame;

Input of a value of a pixel neighboring the reference block of the reference frame and a restored (encoded and then decoded) value of a pixel neighboring the current block of the current coded frame Receiving;

을 계산하는 단계 - 상기 제1 추정 값

은 상기 현재 블록에 이웃하는 픽셀의 복구된 값

의 선형 결합(linear combination)의 함수이고, k = 0, ..., N-1이고, N은 상기 참조 블록 및 상기 현재 블록에 이웃하는 픽셀의 총수(amount)임 -; A first estimated value for each pixel position (i, j) in the reference block

To calculate the first estimated value

Is the recovered value of the pixel neighboring the current block

Is a function of a linear combination of k, 0, ..., N-1, where N is the total number of pixels neighboring the reference block and the current block;

을 계산하는 단계 - 상기 제2 추정 값

은 상기 참조 블록에 이웃하는 픽셀의 값

의 선형 결합(linear combination)의 함수이고, k = 0, ..., N-1임 -; A second estimated value for each pixel position (i, j) in the reference block

To calculate the second estimated value

Is the value of the pixel neighboring the reference block

Is a function of the linear combination of k = 0, ..., N-1-;

, 상기 제2 추정 값

, 상기 참조 블록의 픽셀의 값

, 상기 현재 블록에 이웃하는 픽셀의 복구된 값

및 상기 참조 블록에 이웃하는 픽셀의 값

에 기초하여 상기 참조 블록 내 각 픽셀 위치에 대한 조명 및 콘트라스트 보상을 위한 상기 조명 및 콘트라스트 보상 파라미터로 결정하는 단계; 및 The first estimated value

, The second estimated value

The pixel value of the reference block

A recovered value of a pixel neighboring the current block

And a value of a pixel neighboring the reference block

Determining based on the illumination and contrast compensation parameters for illumination and contrast compensation for each pixel position in the reference block based on; And

And performing illumination and contrast compensation for each pixel position in the reference block by using the determined illumination and contrast compensation parameter.

According to an embodiment, the method may further include calculating the first estimate value and the second estimate value for each pixel position in the reference block, and determining the illumination and contrast compensation parameter for each pixel position in the reference block. The steps are,

을 수학식 1

로 계산하는 단계 -

는 미리 정해진 가중치 계수(weighted coefficients)이고,

은 현재 블록에 이웃하는 픽셀의 복구된 값이고, N은 상기 참조 블록 및 상기 현재 블록에 이웃하는 픽셀의 총수임 -; The first estimated value

To Equation 1

Steps to calculate-

Is predetermined weighted coefficients,

Is a recovered value of a pixel neighboring a current block, and N is a total number of pixels neighboring the reference block and the current block;

을 수학식 2로 계산하는 단계 -

는 미리 정해진 가중치 계수(weighted coefficients)이고,

은 참조 블록에 이웃하는 픽셀의 값이고, N은 상기 참조 블록 및 상기 현재 블록에 이웃하는 픽셀의 총수임 -; 및 The second estimated value

To calculate to the equation (2)-

Is predetermined weighted coefficients,

Is a value of a pixel neighboring a reference block, and N is a total number of pixels neighboring the reference block and the current block; And

를

의 비율로 결정하는 단계를 포함할 수 있다.
The illumination and contrast compensation parameter for illumination and contrast compensation for each pixel position in the reference block

To

Determining the ratio of may include.

및 상기 제2 추정 값

에 대한 가중치 계수

를 계산하는 단계를 포함할 수 있다. 상기 참조 블록 내 각 픽셀 위치 (i, j)에 대하여, 상기 가중치 계수

는 대응하는 절대차

의 감소/증가에 따라

가 역비례(inverse proportional) 증가/감소하는 절대차

의 비증가 함수(non-increasing function)일 수 있다. 여기서,

는 상기 참조 블록의 픽셀의 값이고,

은 상기 참조 블록에 이웃하는 픽셀의 값이고, N은 상기 참조 블록 및 상기 현재 블록에 이웃하는 픽셀의 총수(amount)일 수 있다.According to an embodiment, the calculating of the first estimated value and the second estimated value for each pixel position in the reference block may include: the first estimated value

And the second estimated value

Weighting factor for

And a step of calculating For each pixel position (i, j) in the reference block, the weighting coefficient

Corresponds to the absolute difference

On decrease / increase in

Absolute difference in inverse proportional increase / decrease

It may be a non-increasing function of. here,

Is the value of the pixel of the reference block,

Is a value of a pixel neighboring the reference block, and N may be a total number of pixels neighboring the reference block and the current block.

및 상기 제2 추정 값

에 대한 가중치 계수

를 계산하는 단계를 포함할 수 있다. 상기 참조 블록 내 각 픽셀 위치 (i, j)에 대하여, 상기 가중치 계수

는,

인 경우, 대응하는 절대차

의 감소/증가에 따라

가 역비례(inverse proportional) 증가/감소하는 절대차

의 비증가 함수(non-increasing function)일 수 있고, 그 외의 경우,

는 0일 수 있다. 여기서, Thr은 미리 정해진 스레시홀드(threshold)이고,

는 상기 참조 블록의 픽셀의 값이고,

은 상기 참조 블록에 이웃하는 픽셀의 값일 수 있다.According to an embodiment, the calculating of the first estimated value and the second estimated value for each position of the pixel in the reference block may include: the first estimated value

And the second estimated value

Weighting factor for

And a step of calculating For each pixel position (i, j) in the reference block, the weighting coefficient

Quot;

If, the corresponding absolute difference

On decrease / increase in

Absolute difference in inverse proportional increase / decrease

May be a non-increasing function of, otherwise,

May be zero. Where Thr is a predetermined threshold,

Is the value of the pixel of the reference block,

May be a value of a pixel neighboring the reference block.

및 상기 제2 추정 값

에 대한 가중치 계수

를 계산하는 단계를 포함할 수 있다. 상기 참조 블록 내 각 픽셀 위치 (i, j)에 대하여, 상기 가중치 계수

는,According to an embodiment, the calculating of the first estimated value and the second estimated value for each position of the pixel in the reference block may include: the first estimated value

And the second estimated value

Weighting factor for

And a step of calculating For each pixel position (i, j) in the reference block, the weighting coefficient

Quot;

및

And

인 경우,

Quot;

의 감소/증가에 따라

가 역비례(inverse proportional) 증가/감소하는 절대차

의 비증가 함수(non-increasing function)이고, The corresponding absolute difference

On decrease / increase in

Absolute difference in inverse proportional increase / decrease

Is a non-increasing function of,

은 상기 현재 블록에 이웃하는 픽셀의 값이고, Thr1은 제1 미리 정해진 스레시홀드이고, Thr2는 제2 미리 정해진 스레시홀드이고,

Is a value of a pixel neighboring the current block, Thr1 is a first predetermined threshold, Thr2 is a second predetermined threshold,

는 0이고, Otherwise,

Is 0,

는 상기 참조 블록의 픽셀의 값이고,

은 상기 참조 블록에 이웃하는 픽셀의 값이고, N은 상기 참조 블록 및 상기 현재 블록에 이웃하는 픽셀의 총수(amount)일 수 있다.

Is the value of the pixel of the reference block,

Is a value of a pixel neighboring the reference block, and N may be a total number of pixels neighboring the reference block and the current block.

및 상기 제2 추정 값

에 대해 가중치 계수

를 계산하는 단계를 포함할 수 있다.According to an embodiment, the calculating of the first estimated value and the second estimated value for each position of the pixel in the reference block may include: the first estimated value

And the second estimated value

Weighting factor for

And a step of calculating

For each pixel position (i, j) in the reference block,

인 경우,

Quot;

는The weighting factor

The

ego,

C is a predetermined constant greater than zero,

ego,

는 0이고, Otherwise,

Is 0,

는 상기 참조 블록의 픽셀의 값이고,

은 상기 참조 블록에 이웃하는 픽셀의 값일 수 있다.Thr is a predetermined threshold,

Is the value of the pixel of the reference block,

May be a value of a pixel neighboring the reference block.

및 상기 제2 추정 값

에 대한 미리 정해진 가중치 계수

를 계산하는 단계를 포함할 수 있다.According to an embodiment, the calculating of the first estimated value and the second estimated value for each position of the pixel in the reference block may include: the first estimated value

And the second estimated value

Predetermined weighting factor for

And a step of calculating

For each pixel position (i, j) in the reference block,

및

And

인 경우,

Quot;

는The weighting factor

The

ego,

는 0이고, Otherwise,

Is 0,

C is a predetermined constant greater than zero,

ego,

Is the value of the pixel of the reference block,

은 상기 참조 블록에 이웃하는 픽셀의 값이고,

Is a value of a pixel neighboring the reference block,

은 상기 현재 블록에 이웃하는 픽셀의 값이고,

Is a value of a pixel neighboring the current block,

Thr1 is the first predetermined threshold,

Thr2 may be a second predetermined threshold.

According to an embodiment, the method may include determining a position value of a pixel neighboring the reference block and a position of a recovered value of the pixel neighboring the current block. This position may be adaptively determined instead of the corresponding pixel having a preset neighbor in the reference block of the reference frame.

According to one embodiment, a multi-view video encoding method based on lighting and contrast compensation,

Determining a reference block used to generate a predicted block for the current block;

Determining an illumination and contrast compensation parameter for illumination and contrast compensation of the discrepancy (mismatch) between the reference blocks during or after determining the reference block;

Performing illumination and contrast compensation of the determined reference block using the determined illumination and contrast parameters;

Generating the prediction block for the current block using the illumination and contrast compensated reference block; And

Encoding the current block by using the generated prediction block without encoding the determined illumination and contrast compensation parameter, and encoding information on the position of the reference block when necessary for decoding.

According to one embodiment, the step of determining the illumination and contrast compensation parameters,

Receiving a value of a pixel neighboring the reference block and a reconstructed (ie, encoded and decoded) value of a pixel neighboring the current block;

A relationship between a value of a pixel of the reference block and a value of a pixel neighboring the reference block and a relationship between a value of a pixel neighboring the reference block and a restored value of a pixel neighboring the current block ( determining relations; And

Illumination and contrast for illumination and contrast compensation of the reference block based on the determined relationship, a value of a pixel of the reference block, a recovered value of a pixel neighboring the current block, and a value of a pixel neighboring the reference block Determining a compensation parameter.

According to one embodiment, a multi-view video decoding method based on lighting and contrast compensation,

Determining a reference block, decoding information about an existing reference block as necessary for determining the reference block of the current block;

Determining an illumination and contrast compensation parameter for illumination and contrast compensation of the determined reference block;

Performing illumination and contrast compensation of the determined reference block using the determined illumination and contrast compensation parameter;

Generating a prediction block for the current block using the illumination and contrast compensated reference block;

And decoding the current block by using the determined illumination and contrast compensation parameter and the generated prediction block.

According to one embodiment, the step of determining the illumination and contrast compensation parameters,

Receiving a value of a pixel neighboring the reference block and a reconstructed (already decoded) value of a pixel neighboring the current block;

A relationship between a value of a pixel of the reference block and a value of a pixel neighboring the reference block and a relationship between a value of a pixel neighboring the reference block and a restored value of a pixel neighboring the current block ( determining relations; And

Illumination and contrast for illumination and contrast compensation of the reference block based on the determined relationship, a value of a pixel of the reference block, a recovered value of a pixel neighboring the current block, and a value of a pixel neighboring the reference block Determining a compensation parameter.

1 illustrates a configuration of a high-level structure and a multi-view coding environment of a hybrid multi-view coding environment according to an embodiment.
2 is a partial conceptual diagram of a hybrid video encoder implementing an embodiment included in predictive coding.
3 is a diagram for describing a method of compensating for lighting and contrast of a reference block, according to an exemplary embodiment.
4 is a flowchart illustrating a method of compensating lighting and contrast of a reference block according to an embodiment.
5 is a diagram for describing a process of selecting an input block in a current frame during calculation of an illumination and contrast compensation parameter, according to an exemplary embodiment.
6 is a diagram for describing a lighting and contrast compensation method of a reference block, according to an embodiment.
7 is a flowchart illustrating a method of compensating illumination and contrast of a pixel-wise reference block according to an embodiment.
8 is a diagram for describing a lighting and contrast compensation method of a reference block, according to an exemplary embodiment.
9 is a flowchart illustrating a multi-view video encoding method based on lighting and contrast compensation, according to an exemplary embodiment.
10 is a flowchart for describing a multi-view video decoding method based on lighting and contrast compensation, according to an exemplary embodiment.

Hereinafter, embodiments will be described in detail with reference to the accompanying drawings.

1 shows a structure of a hybrid multi-view coding device. The input data of the hybrid multi-view video encoder 105 is the already coded / decoded views 102 which are encoded multi-view video data and the original. It includes a view 101. The already coded / decoded depth 102 and the already coded / decoded depth sequences 103 are synthesized for the original view by view synthesis 104. Used to generate synthesized views. The generated synthesized view video sequence also becomes input data for the hybrid multi view video encoder 105.

The hybrid multi view video encoder 105 includes the following tools that can be used for encoding the original view: reference picture management 106, inter-frame prediction 107, intra Intra-frame prediction 108, inter and intra-frame compensation 109, spatial transform 110, rate-distortion optimization 111 ), And entropy coding 112. More detailed information on the above tools is described in Richardson I.E. "The H.264 Advanced Video Compression Standard. Second Edition. 2010". Embodiments may be implemented inside of inter frame prediction 107.

2 shows a partial conceptual diagram of a hybrid video encoder implementing in embodiments included in predictive coding. The hybrid encoder includes a subtraction unit 201, a transform and quantization unit 202, an entropy encoding unit 203, an inverse transform and inverse quantization unit quantization unit 204, displacement and illumination / contrast change compensation unit 205, view synthesis unit 206, addition unit 207 Reference pictures and depths buffer unit 208, prediction of compensation parameters unit 209, displacement and illumination / contrast change estimation unit 210, and a macroblock mode decision unit 211. Units 201-204, 207-209, and 211 are described in Richardson IE's "The H.264 Advanced Video Compression Standard." Second Edition. It is a standard encoding unit used in the basic hybrid coding method of 2010. A view synthesis unit 206 is a specific unit for multi-view coding. Synthesize additional reference frames from the decoded frame and depth data.

The method may be implemented inside of units 205 and 210. This unit realizes a block-wise predictive coding technique comprising the following steps.

For a current block of a currently coded frame, a search for a reference block that minimizes Equation 3 is performed.

는 참조 블록 및 현재 블록 간의 조명 및 콘트라스트 변화를 어느 정도 보상하는 함수를 의미한다. 이 기술은 유닛(210)에서 실현된다. 얻어진 DV와 함께 조명 및 콘트라스트 보상의 결정된 파라미터는 유닛(205) 및 유닛(209)에 전송된다.Here, I (m, n) represents the illumination value of the pixel at the position (m, n) of the current block having M × N size. (i, j) specifies a displacement vector that points to the reference block R within a predetermined search area.

Denotes a function that compensates to some extent the illumination and contrast change between the reference block and the current block. This technique is realized in unit 210. The determined parameters of illumination and contrast compensation together with the resulting DVs are sent to units 205 and 209.

The selected reference block is modified according to the determined illumination and contrast compensation parameters (unit 205). Then, a residual block is generated by the unit 201. The residual block is then transformed by a discrete cosine transform (DCT), quantized (unit 202) and encoded by an entropy encoder (unit ( 203)). Side information (SI) necessary for decoding is also encoded by the entropy encoder (unit 203).

는 다음과 같이 정의된다.
3 is a diagram for describing a lighting and contrast compensation method of a reference block, according to an exemplary embodiment. According to FIG. 3, while a search for a reference block is in progress, a displacement vector (DV) 320 is assigned to the current block 311 of the currently encoded frame 310 repeatedly. (assign) The displacement vector DV points to the reference block 301 of the reference frame 300. Lighting and Contrast Compensation Function

Is defined as

The illumination and contrast compensation parameter α is expressed by Equation 5.

refMX is the mean value of the reference block 301 with the coordinates of the left-top corner (i, j). S means the reference frame 300. The value denoted by estMX is an estimation (approximation) of the mean value for the reference block 311.

4 is a flowchart illustrating a method for compensating for lighting and contrast of a reference block according to one embodiment. The method includes the following steps.

1. Receiving input of pixel values of blocks 301, 302, 303, 311, 312, 313, 314 (FIGS. 4, 401).

2. Step of calculating the average value below (FIGS. 4, 402):

Calculating an average value encMX_L of block 312;

Where DI (p, q) represents the restored (already decoded) illumination value of the pixel at position (p, q) of block 312. The size of block 312 is P x Q.

Calculating a mean value encMX_A of block 313;

Where DI (u, v) is the restored (encoded and then decoded) illumination value of the pixel at position (u, v) of block 313. Indicates. The size of block 313 is U x V.

Calculating an average value refMX of the reference block 301;

Calculating an average value refMX_L of block 302;

The size of block 302 is equal to the size of block 312.

Calculating an average value refMX_A of block 303;

The size of block 303 is equal to the size of block 313.

3. Checking condition 1 (FIGS. 4, 403): If block 302 and block 312 are available (ie, blocks 302 and 312 are located inside frame boundaries, and If the reference block is a synthesized frame, all pixels belonging to the block 302 are available (not occluded), and the estimation of the estMX value according to Equation 10 (Figs. 4 and 405). Proceed to).

Otherwise, proceed to the check step of condition 2 (FIGS. 4, 404).

4. Checking condition 2 (FIG. 4, 404): If block 303 and block 313 are available (ie, blocks 303 and 313 are located inside frame boundaries, and If the reference block is a synthesized frame, all pixels belonging to the block 303 are available (not occluded), and the estimation of the estMX value according to Equation 11 (Figs. 4 and 407). Proceed to).

Otherwise, proceed to the estimation of the estMX value according to equation (12) (FIGS. 4, 406).

Here, MAP (x, y, z) is a Median Adaptive Predictor (MAP) described in Martucci SA's "Reversible compression of HDTV images using median adaptive prediction and arithmetic coding" (IEEE Int. Symp. On Circuits and Systems, 1990). EncMX_LA is the mean value of block 314.

The size of block 314 is U x Q and is equal to the corresponding size of blocks 312 and 313.

5. calculating the illumination and contrast compensation parameter α by using the obtained values of estMX and refMX (FIGS. 4, 408);

6. performing illumination and contrast compensation of the reference block 301 by using the calculated parameter α (FIGS. 4, 409);

Reference frames 300 with blocks 301, 302, 303 and restored (already coded / decoded) blocks 312, 313, 314 are available during encoding and decoding. Do. 5 illustrates a geometric relationship between areas in a current frame 500. The region 501 of the current frame 500 is available during encoding and decoding of the currently coded block. Region 501 includes blocks 312, 313, 314. Area 501 is called a "template area". Region 503 is not present during decoding of current block 502 and does not include blocks 312, 313, 314. Thus, the embodiments described above can be applied simultaneously to both the encoder and the decoder, and no additional information needs to be placed in the encoded bit stream.

According to an embodiment, the pixel may include pixel-wise illumination and contrast compensation of the reference block during predictive coding. Pixel-wise estimation of illumination and contrast parameter compensation is performed based on the recovered values of pixels neighboring the current block, the values of the pixels of the reference frame, and their similarities. 6 illustrates one embodiment.

으로 표시된다. 현재 코드된 프레임(currently coded frame)의 복구된 픽셀에 대응하는 참조 블록에 이웃하는 픽셀(블록(602 및 603))의 값은

으로 표시된다. 블록(612, 613) 및 블록(602, 603) 내 픽셀의 총수(total amount)는 같다.According to FIG. 6, while the search of the reference block is in progress, a displacement vector (DV) 620 may be assigned to the current block 611 belonging to the currently encoded frame 610. Can be. The displacement vector DV points to the reference block 601 of the reference frame 600. The current block 611 contains the values of the pixels represented by A00 to A33. Reference block 601 includes pixels indicated by R00 to R33. The restored value of the pixel neighboring the current block (blocks 612 and 613) is

. The value of the pixel (blocks 602 and 603) neighboring the reference block corresponding to the recovered pixel of the currently coded frame is

. The total amount of pixels in blocks 612 and 613 and blocks 602 and 603 is equal.

For each pixel position (i, j) in reference block 601, illumination and contrast compensation is performed according to equation (14).

Here, the pixel-by-pixel illumination and contrast compensation parameter is represented by Equation 15.

는 참조 블록 내 픽셀 위치 (i, j)에 대한 제1 추정 값(first estimation value)이다.

는 참조 블록 내 픽셀 위치 (i, j)에 대한 제2 추정 값(second estimation value)이다.here,

Is a first estimation value for pixel position (i, j) in the reference block.

Is a second estimation value for pixel position (i, j) in the reference block.

Pixel-by-pixel illumination and contrast compensation of the reference block is shown in FIG. 7. This embodiment includes the following steps.

1. From blocks 612, 613, block 611, and reference frames 600 from template area of a currently coded frame 610, block 601, 602, 603. Receiving an input of a value of a pixel (operation 701);

를 계산하는 단계(동작(702)).2. Weighted coefficients for each pixel position (i, j) in reference block 601

Calculating (operation 702).

는 수학식 16으로 표현될 수 있다.
weight

May be represented by Equation 16.

가

에 더 비슷할수록 조명 및 콘트라스트 파라미터에 더 많은 기여를 준다는 사실을 반영한다.[sigma] is greater than zero, defines a smoothness degree, and is determined experimentally. Where N is the total amount of pixels in blocks 612 and 613 (or blocks 602 and 603). By default, the weighting factor is a value

end

More closely reflects the fact that it contributes more to the lighting and contrast parameters.

의 값을 계산하는 단계(동작(703)).
3. For each pixel position (i, j) in the reference block 601 according to equation (17)

Calculating the value of (act 703).

와 다른 참조 블록에 이웃하는 픽셀의 값을 제외(excluding)하는 데 사용된다.Thr1 and Thr2 are predetermined threshold values. The threshold value is different from the value of the corresponding pixel neighboring the reference block or the current block and is the value of the reference block.

It is used to exclude values of pixels neighboring to other reference blocks.

의 값을 계산하는 단계(동작(704)).
4. For each pixel position (i, j) in the reference block 601 according to equation (18)

Calculating the value of (operation 704).

의 계산에 있어서 같다.The predetermined threshold values Thr1 and Thr2

Is the same in the calculation of.

및

의 얻어진 값을 사용하여 참조 블록(601) 내 각 픽셀 위치 (i, j)에 대하여 조명 및 콘트라스트 보상 파라미터(illumination and contrast compensation parameter)

을 계산하는 단계(동작(705)).5.

And

Illumination and contrast compensation parameter for each pixel position (i, j) in reference block 601 using the obtained value of

Calculating (operation 705).

를 사용하여, 참조 블록(601)의 조명 및 콘트라스트 보상을 수행하는 단계(동작(706)).6. Calculated Parameters

To perform illumination and contrast compensation of reference block 601 (operation 706).

According to another embodiment, the value of a pixel, usually adjacent to a reference block, is defined as a group of pixels immediately adjacent to the reference block. In practice, the displacement vector estimation procedure uses a motion / displacement vector such that the value of the pixel neighboring the reference block is not similar to the recovered value of the pixel neighboring the current block. Can be selected. One example is a fairly uniform or periodic area bounded by contrast areas. In this case, pixels neighboring the reference block appear in the contrast area, while recovered pixels neighboring the current block are located in the smooth area. A more complex case is an area with a large change in displacement or intensity of the area in the reference frame relative to the corresponding area in the encoded frame defined by the group of pixels adjacent to the encoded reference block. In this case, since the values of the corresponding pixels of the reference frame and the currently coded frame are very different, the illumination and contrast compensation may operate incorrectly.

In order to solve this problem, the use of a "float" position (relative to the reference block) of a group of pixels neighboring the reference block is proposed. 8 illustrates a method according to one embodiment. Referring to FIG. 8, during a search of a reference block, a displacement vector (DV) 820 is assigned to the current block 811 of a currently coded frame 810. The displacement vector DV points to the reference block 801 of the reference frame 800. Floating is determined by an additional refinement displacement vector 804 that indicates the position of the pixel of the reference frame. The improved displacement vector 804 is the result of a displacement estimation procedure. The estimation process involves finding a displacement vector (DV) 804 that defines the minimum value of an error function that defines the similarity degree of blocks 802, 803 and blocks 812, 813, respectively. Is over. One of ordinary skill in the art, for example, any type of similarity functions, such as Means Square Error (MSE), Sum of Absolute Differences (SAD), Mean Removed Sum of Absolute Differences (MRSAD), etc. You can also use The vector 804 may be implicitly determined during the encoding and decoding process without sending any additional information in the output bit stream.

9 is a diagram for describing a method for multi-view video encoding based on lighting and contrast compensation, according to an embodiment. In step 901, a reference block used for generation of a predicted block is determined. In step 902, illumination and contrast compensation parameters for illumination and contrast compensation of the determined reference block are determined. The determination of the illumination and contrast compensation parameter includes the following steps.

Receiving a value of a pixel neighboring the reference block and a reconstructed (already decoded) value of the pixel neighboring the current block;

A relation between a value of a pixel of the reference block and a value of a pixel neighboring the reference block and a relationship between a value of a pixel neighboring the reference block and a restored value of a pixel neighboring the current block determining relatives; And

For illumination and contrast compensation of the reference block based on the determined relationships, a value of a pixel of the reference block, a recovered value of a pixel neighboring the current block, and a value of a pixel neighboring the reference block. Determining illumination and contrast compensation parameters.

In step 903, the illumination and contrast compensation of the reference block is performed using the determined illumination and contrast compensation parameters. In step 904, a predicted block for the current block is generated using the illumination and contrast compensated reference block. In step 905, the generated block is encoded using the generated prediction block. In particular, when information about a reference block is necessary for decoding, this information is also encoded. At the same time, the determined illumination and contrast compensation parameters are not encoded.

10 is a diagram for describing a multi-view video decoding method based on lighting and contrast compensation, according to an exemplary embodiment. According to FIG. 10, information about a reference block is decoded when it is necessary to decode. The decoded information may be used at step 1001 for the determination of the reference block. In step 1002, illumination and contrast compensation parameters for illumination and contrast compensation of the reference block are determined. The determination of the illumination and contrast compensation parameter includes the following steps.

Receiving a value of a pixel neighboring the reference block and a reconstructed (already decoded) value of the pixel neighboring the current block;

A relationship between a value of a pixel of the reference block and a value of a pixel neighboring the reference block and a relationship between a value of a pixel neighboring the reference block and a restored value of a pixel neighboring the current block ( determining relations; And

For illumination and contrast compensation of the reference block based on the determined relationships, a value of a pixel of the reference block, a recovered value of a pixel neighboring the current block, and a value of a pixel neighboring the reference block. Determining illumination and contrast compensation parameters.

In step 1003, the determined illumination and contrast compensation parameters are used to perform illumination and contrast compensation of the reference block. In step 1004, the predictive block for the current block is generated by using the illumination and contrast compensated reference block. In step 1005, the current block is decoded using the generated prediction block.

Embodiments may be implemented to encode and decode multi-view video sequences.

The method according to embodiments may be implemented in the form of a program instruction that may be executed through various computer means and recorded in a computer-readable medium. The computer readable medium may include program instructions, data files, data structures, etc. alone or in combination. The program instructions recorded on the media may be those specially designed and constructed for the purposes of the embodiments, or they may be of the kind well-known and available to those having skill in the computer software arts. Examples of computer-readable recording media include magnetic media such as hard disks, floppy disks, and magnetic tape, optical media such as CD-ROMs, DVDs, and magnetic disks, such as floppy disks. Magneto-optical media, and hardware devices specifically configured to store and execute program instructions, such as ROM, RAM, flash memory, and the like. Examples of program instructions include not only machine code generated by a compiler, but also high-level language code that can be executed by a computer using an interpreter or the like. The hardware devices described above may be configured to operate as one or more software modules to perform the operations of the embodiments, and vice versa.

While the present invention has been particularly shown and described with reference to exemplary embodiments thereof, it is to be understood that the invention is not limited to the disclosed exemplary embodiments. For example, it is to be understood that the techniques described may be performed in a different order than the described methods, and / or that components of the described systems, structures, devices, circuits, Lt; / RTI > or equivalents, even if it is replaced or replaced.

Therefore, other implementations, other embodiments, and equivalents to the claims are also within the scope of the following claims.

Claims (16)

In the local compensation method of the lighting and contrast difference between the reference block and the encoded block in the prediction step of the multi-view coding process,
Receiving a value of a pixel of a reference block in a reference frame and a value of a pixel of a current block in an encoded frame;
Receiving a value of a pixel neighboring the reference block of the reference frame and an already decoded and recovered value of a pixel neighboring the current block of a currently encoded frame;
Determining a relationship between a value of a pixel neighboring the reference block and a recovered value of a pixel neighboring the current block and a relationship between a value of a pixel neighboring the reference block and a value of a pixel of the reference block;
Based on the value of the pixel neighboring the reference block, the recovered value of the pixel neighboring the current block, the value of the pixel of the reference block, and the determined relationships, the difference compensation between the reference block and the encoded block ( determining an illumination and contrast compensation parameter for illumination and contrast compensation of discrepancy (mismatch) compensation; And
Performing illumination and contrast compensation of the difference between the reference block and the encoded block using the determined illumination and contrast compensation parameter.
And a local compensation method of illumination and contrast differences between the reference block and the encoded block in the prediction step of the multi-view coding process comprising a. The method of claim 1,
Determining the relationship and determining the illumination and contrast compensation parameters include:
Calculating statistical characteristics of the value of the recovered pixel neighboring the current block, statistical characteristics of the value of the pixel of the reference block, and statistical characteristics of the value of the pixel neighboring the reference block;
Determining a relationship between the recovered value of a pixel neighboring the reference block and a statistical feature on the value of the pixel of the reference block; And
Calculating an illumination and contrast compensation parameter for illumination and contrast compensation of the reference block based on the statistical feature of the reference block and the received statistical feature for the current block.
And a local compensation method of illumination and contrast differences between the reference block and the encoded block in the prediction step of the multi-view coding process comprising a. The method of claim 1,
Determining the relationship and determining the illumination and contrast compensation parameters include:
An average value for a recovered pixel located to the left of the current block and neighboring the current block, an average value for a recovered pixel located above the current block and neighboring the current block, an average value for a pixel of the reference block Calculating an average value of pixels located to the left of the reference block and neighboring the reference block, and an average value of pixels located above the reference block and neighboring the reference block;
If there is a recovered pixel located to the left of the current block and neighboring the current block and there is a pixel located to the left of the reference block and neighboring to the reference block, located to the left of the reference block and neighboring to the reference block A ratio value is calculated between an average value of a pixel and an average value of a pixel of the reference block, and a product of the average value and the ratio value of a recovered pixel located to the left of the current block and neighboring the current block is calculated. Calculating and determining the illumination and contrast compensation parameter as a ratio between the calculated product and an average value for the pixels of the reference block;
If there is a recovered pixel located above the current block and neighboring the current block and there is a pixel located above the reference block and neighboring the reference block, the average value of the pixels of the reference block and above the reference block Calculate a ratio value between an average value of pixels located and neighboring the reference block, calculate a product of the average value and the ratio value of the recovered pixels located above the current block and neighboring the current block, and calculate the calculated product Determining the illumination and contrast compensation parameter as a ratio between an average value for pixels of the reference block; And
Otherwise, median adaptive prediction (MAP) is used to calculate an estimate for the average value of the current block, and the estimated average value of the pixels of the current block and the average value for the pixels of the reference block. Determining the illumination and contrast compensation parameter as a ratio of
And a local compensation method of illumination and contrast differences between the reference block and the encoded block in the prediction step of the multi-view coding process comprising a. The method of claim 1,
Determining the relationship and determining the illumination and contrast compensation parameters include:
A first estimated value for each pixel position (i, j) in the reference block

To calculate the first estimated value

Is the recovered value of the pixel neighboring the current block

Is a function of a linear combination of k, 0, ..., N-1, where N is the total number of pixels neighboring the reference block and the current block;
A second estimated value for each pixel position (i, j) in the reference block

To calculate the second estimated value

Is the value of the pixel neighboring the reference block

Is a function of the linear combination of k = 0, ..., N-1-;
The first estimated value

, The second estimated value

The pixel value of the reference block

A recovered value of a pixel neighboring the current block

And a value of a pixel neighboring the reference block

Determining based on the illumination and contrast compensation parameters for illumination and contrast compensation for each pixel position in the reference block based on; And
Performing illumination and contrast compensation for each pixel position in the reference block using the determined illumination and contrast compensation parameters.
And a local compensation method of illumination and contrast differences between the reference block and the encoded block in the prediction step of the multi-view coding process comprising a. 5. The method of claim 4,
Determining the relationship, determining the illumination and contrast compensation parameters, and performing the illumination and contrast compensation,
The first estimated value

of

Step to calculate-

Is predetermined weighted coefficients,

Is a recovered value of a pixel neighboring a current block, and N is a total number of pixels neighboring the reference block and the current block;
The second estimated value

of

Step to calculate-

Is predetermined weighted coefficients,

Is a value of a pixel neighboring a reference block, and N is a total number of pixels neighboring the reference block and the current block;
The second estimated value

Is not 0, the illumination and contrast compensation parameter for illumination and contrast compensation for each pixel position in the reference block

To

Determined by the ratio of the second estimated value

Is 0, the illumination and contrast compensation parameter

Is set to 1; And
The value of each pixel in the reference block

The corresponding compensation parameter to

Performing illumination and contrast compensation of the reference block by multiplying
And a local compensation method of illumination and contrast differences between the reference block and the encoded block in the prediction step of the multi-view coding process comprising a. The method of claim 5,
Computing the first estimate value and the second estimate value for each position of a pixel in the reference block,
The first estimated value

And the second estimated value

Weighting factor for

Step to calculate
For each pixel position (i, j) in the reference block, the weighting factor

Corresponds to the absolute difference

On decrease / increase in

Absolute difference in inverse proportional increase / decrease

Is a non-increasing function of,

Is the value of the pixel of the reference block,

Is the value of the pixel neighboring the reference block, and N is the total number of pixels neighboring the reference block and the current block
And a local compensation method of illumination and contrast differences between the reference block and the encoded block in the prediction step of the multi-view coding process comprising a. The method of claim 5,
Computing the first estimate value and the second estimate value for each position of a pixel in the reference block,
The first estimated value

And the second estimated value

Weighting factor for

Step to calculate
For each pixel position (i, j) in the reference block, the weighting factor

Quot;

If, the corresponding absolute difference

On decrease / increase in

Absolute difference in inverse proportional increase / decrease

Is a non-increasing function of,
Otherwise,

Is 0,
Thr is a predetermined threshold,

Is the value of the pixel of the reference block,

Is the value of the pixel neighboring the reference block −
And a local compensation method of illumination and contrast differences between the reference block and the encoded block in the prediction step of the multi-view coding process comprising a. The method of claim 5,
Computing the first estimate value and the second estimate value for each position of a pixel in the reference block,
The first estimated value

And the second estimated value

Weighting factor for

Step to calculate
For each pixel position (i, j) in the reference block, the weighting factor

Quot;

And

If, the corresponding absolute difference

On decrease / increase in

Absolute difference in inverse proportional increase / decrease

Is a non-increasing function of,

Is a value of a pixel neighboring the current block, Thr1 is a first predetermined threshold, Thr2 is a second predetermined threshold,
Otherwise,

Is 0,

Is the value of the pixel of the reference block,

Is the value of the pixel neighboring the reference block −
And a local compensation method of illumination and contrast differences between the reference block and the encoded block in the prediction step of the multi-view coding process comprising a. The method of claim 5,
Computing the first estimate value and the second estimate value for each position of a pixel in the reference block,
The first estimated value

And the second estimated value

Predetermined weighting factor for

Step to calculate
For each pixel position (i, j) in the reference block,

If, the weighting factor

The

ego,
C is a predetermined constant greater than zero,

ego,
Otherwise,

Is 0,
Thr is a predetermined threshold,

Is the value of the pixel of the reference block,

Is the value of the pixel neighboring the reference block −
And a local compensation method of illumination and contrast differences between the reference block and the encoded block in the prediction step of the multi-view coding process comprising a. The method of claim 5,
Computing the first estimate value and the second estimate value for each position of a pixel in the reference block,
The first estimated value

And the second estimated value

Predetermined weighting factor for

Step to calculate
For each pixel position (i, j) in the reference block,

And

Quot;
The weighting factor

The

ego,
Otherwise,

Is 0,
C is a predetermined constant greater than zero,

ego,

Is the value of the pixel of the reference block,

Is a value of a pixel neighboring the reference block,

Is a value of a pixel neighboring the current block,
Thr1 is the first predetermined threshold,
Thr2 is the second predetermined threshold-
And a local compensation method of illumination and contrast differences between the reference block and the encoded block in the prediction step of the multi-view coding process comprising a. The method of claim 1,
The position of the value of the pixel neighboring the reference block and the recovered value of the pixel neighboring the currently encoded block is adaptively determined instead of the corresponding pixel occupying a predetermined position.
A method for local compensation of lighting and contrast differences between reference blocks and encoded blocks in the prediction phase of a multi-view coding process. In the multi-view video encoding method based on lighting and contrast compensation,
Determining a reference block used to generate a predicted block for the current block;
Determining an illumination and contrast compensation parameter for illumination and contrast compensation of the reference block during or after determining the reference block;
Performing illumination and contrast compensation of the determined reference block using the determined illumination and contrast parameters;
Generating the prediction block for the current block using the illumination and contrast compensated reference block; And
Encoding the current block using the generated prediction block without encoding the determined illumination and contrast compensation parameters, encoding information about the position of the reference block as needed for decoding
Multi-view video encoding method based on lighting and contrast compensation comprising. The method of claim 12,
Determining the illumination and contrast compensation parameters,
Receiving a value of a pixel neighboring the reference block and a reconstructed value of a pixel neighboring the current block;
Between mathematical ratios between a value of a pixel of the reference block and a value of a pixel neighboring the reference block and between a value of a pixel neighboring the reference block and a restored value of a pixel neighboring the current block Determining a mathematical ratio; And
Illumination for contrast and illumination compensation of the reference block based on the determined mathematical ratio, the value of the pixel of the reference block, the recovered value of the pixel neighboring the current block, and the value of the pixel neighboring the reference block; Determining Contrast Compensation Parameters
Multi-view video encoding method based on lighting and contrast compensation comprising. A multi-view video decoding method based on lighting and contrast compensation,
Determining a reference block, decoding information about the reference block as needed to determine the reference block of a current block;
Determining an illumination and contrast compensation parameter for illumination and contrast compensation of the determined reference block;
Performing illumination and contrast compensation of the determined reference block using the determined illumination and contrast compensation parameter;
Generating a prediction block for the current block using the illumination and contrast compensated reference block;
Decoding the current block using the determined illumination and contrast compensation parameter and the generated prediction block
Multi-view video decoding method based on lighting and contrast compensation comprising a. 15. The method of claim 14,
Determining the illumination and contrast compensation parameters,
Receiving a value of a pixel neighboring the reference block and a reconstructed value of a pixel neighboring the current block;
Between mathematical ratios between a value of a pixel of the reference block and a value of a pixel neighboring the reference block and between a value of a pixel neighboring the reference block and a restored value of a pixel neighboring the current block Determining a mathematical ratio; And
Illumination for contrast and illumination compensation of the reference block based on the determined mathematical ratio, the value of the pixel of the reference block, the recovered value of the pixel neighboring the current block, and the value of the pixel neighboring the reference block; Determining Contrast Compensation Parameters
Multi-view video decoding method based on lighting and contrast compensation comprising a. A non-transitory computer-readable recording medium having recorded thereon a program for executing the method of claim 1.

Images