帧间预测样点解码过程

帧间预测样点解码过程 8.4.2 Decoding process for Inter prediction samples 帧间预测样点解码过程 Inputs to this process are 本过程输入的是: - a macroblock partition mbPartIdx, - a sub-macroblock partition subMbPartIdx. - variables specifying partition width and height, partWidth and partHeight - luma motion vectors mvL0 and mvL1 and chroma motion vectors mvCL0 and mvCL1 - reference indices refIdxL0 and refIdxL1 - prediction list utilization flags, predFlagL0 and predFlagL1 - 一个宏块划分mbPartIdx, - 一个子宏块划分subMbPartIdx。 - 标明划分宽度和高度的变量partWidth和partHeight - 亮度运动矢量mvL0和mvL1，色度运动矢量mvCL0和mvCL1 - 参考索引refIdxL0和refIdxL1 - 预测 表 关于同志近三年现实表现材料材料类招标技术评分表图表与交易pdf视力表打印pdf用图表说话 pdf 使用标志位predFlagL0 和predFlagL1 Outputs of this process are 本过程的输出是: - the Inter prediction samples predPart, which are a (partWidth)x(partHeight) array predPart of Lprediction luma samples, and two (partWidth/2)x(partHeight/2) arrays predPart, predPart of CbCrprediction chroma samples, one for each of the chroma components Cb and Cr. - 帧间预测样点predPart，包括一个(partWidth)x(partHeight)的亮度预测矩阵predPart,两个L(partWidth/2)x(partHeight/2)的色度样点预测矩阵，分别是Cb预测矩阵predPart和CrCb 预测矩阵predPart。 Cr Let predPartL0 and predPartL1 be (partWidth)x(partHeight) arrays of predicted luma sample LL values and predPartL0, predPartL1, predPartL0, and predPartL1 be CbCbCrCr(partWidth/2)x(partHeight/2) arrays of predicted chroma sample values. 令predPartL0和predPartL1为(partWidth)x(partHeight)的亮度预测矩阵，predPartL0、LLCbpredPartL1predPartL0和predPartL1为(partWidth/2)x(partHeight/2)的色度预测矩阵。、CbCrCr For LX being replaced by either L0 or L1 in the variables predFlagLX, RefPicListX, refIdxLX, refPicLX, predPartLX, the following is specified. 用L0或L1替代变量predFlagLX、RefPicListX、 refIdxLX、 refPicLX和 predPartLX 中的LX，根据predFlagLX执行下面的操作。 When predFlagLX is equal to 1, the following applies. 1 当predFlagLX等于1时，执行下面的步骤。 - The reference frame consisting of an ordered two-dimensional array refPicLX of luma L samples and two ordered two-dimensional arrays refPicLX and refPicLX of chroma CbCr samples is derived by invoking the process specified in subclause 8.4.2.1 with refIdxLX and RefPicListX given as input. - 调用8.4.2.1节叙述的过程,译注:选择参考图像,，用refIdxLX和RefPicListX作为输 入，得到由一个二维亮度矩阵refPicLX和两个二维色度矩阵refPicLX、refPicLX组LCbCr 成的参考帧。 [译注:原文在矩阵前加了定语 ordered，看不出为什么。] - The arrays predPartLX, predPartLX, and predPartLX are derived by invoking the process LCbCr specified in subclause 8.4.2.2 with the current partition specified by mbPartIdx\subMbPartIdx, the motion vectors mvLX, mvCLX, and the reference arrays with refPicLX, refPicLX, and LCb refPicLX given as input. Cr - 调用8.4.2.2节中叙述的过程,译注:样点内插。,，用当前划分mbPartIdx\subMbPartIdx， 运动矢量mvLX、mvCLX和参考矩阵refPicLX,、refPicLX、 refPicLX作为输入，得LCbCr 到矩阵predPartLX、predPartLX和predPartLX。 LCbCr For C being replaced by L, Cb, or Cr, the array predPart of the prediction samples of component C C is derived by invoking the process specified in subclause 8.4.2.3 with the current partition specified by mbPartIdx and subMbPartIdx and the array predPartL0 and predPartL1 as well as CC predFlagL0 and predFlagL1 given as input. 用L、Cb和Cr替代C，调用8.4.2.3中叙述的过程,译注:加权预测，将多个源合并成一个 目标,，用当前划分(由mbPartIdx和subMbPartIdx指定的)，和矩阵predPartL0predPartL1、CC 以及predFlagL0、predFlagL1作为输入，分别得出预测矩阵predPart(即分别得出predPart、CL predPart和 predPart)。 CbCr 8.4.2.1 Reference picture selection process 参考图像选择过程 Input to this process is a reference index refIdxLX. 本过程输入的是一个参考索引refIdxLX Output of this process is a reference picture consisting of a two-dimensional array of luma samples refPicLX and two two-dimensional arrays of chroma samples refPicLX and LCb refPicLX. Cr 本过程输出的是由一个二维亮度矩阵refPicLX和两个二维色度矩阵refPicLXrefPicLX、LCbCr 组成的参考图像。 Reference picture list RefPicListX is a list of variables PicNum (for short-term reference pictures) and LongTermPicNum (for long-term reference pictures) of previously decoded reference frames, complementary reference field pairs, or non-paired reference fields that have been marked as “used for reference” as specified in subclause 错误～未找到引用源。8.2.5. 参考图像列表RefPicListX是已解码的参考帧、互补参考场对或不成对参考场(这些帧或场在 错误～未找到引用源。8.2.5节中标识为“用来作参考”)的图像编号变量PicNum (短时参考 图像)和LongTermPicNum(长时参考图像)的列表。 Depending on field_pic_flag, the meaning of PicNum and LongTermPicNum is specified as follows. PicNum和LongTermPicNum的含义根据field_pic_flag 规定 关于下班后关闭电源的规定党章中关于入党时间的规定公务员考核规定下载规定办法文件下载宁波关于闷顶的规定 如下。 2 - If field_pic_flag is equal to 1, all entries of the RefPicListX are variables PicNum and LongTermPicNum of decoded reference fields or fields of decoded reference frames. - 如果field_pic_flag等于1，RefPicListX中的所有元素是参考场或参考帧中的场的变量 PicNum和LongTermPicNum。 - Otherwise (field_pic_flag is equal to 0), all entries of RefPicListX are variables PicNum and LongTermPicNum of decoded reference frames or complementary reference field pairs. - 否则(field_pic_flag等于0)，RefPicListX中的所有元素是参考帧或互补参考场对的变量 PicNum和LongTermPicNum。 The reference picture list RefPicListX is derived as specified in subclause 错误～未找到引用源。 8.2.4. 调用错误～未找到引用源。8.2.4节中过程以得到参考图像列表RefPicListX。 ,译注:原文在这里不是很明确，到底是现在调用的还是已经调用完了。如果是已经调用完 了，应该 说明 关于失联党员情况说明岗位说明总经理岗位说明书会计岗位说明书行政主管岗位说明书 用它当输入。如果现在调用，写法应该更明确些。, For the derivation of the reference picture, the following applies. 下面的步骤用于得出参考图像: - If field_pic_flag is equal to 1, the reference field or field of a reference frame referred by PicNum = RefPicListX[ refIdxLX ] or LongTermPicNum = RefPicListX[ refIdxLX ] shall be the output. The output reference field or field of a reference frame consists of a (PicWidthInSamples)x(PicHeightInSamples) array of luma samples refPicLX and two LLL (PicWidthInSamples)x(PicHeightInSamples) arrays of chroma samples refPicLX and CCCb refPicLX. Cr - 如果field_pic_flag等于1，则由PicNum = RefPicListX[ refIdxLX ]或LongTermPicNum , RefPicListX[ refIdxLX ]指定的参考场或参考帧中的场作为输出。输出参考场或参考帧中 的场由一个(PicWidthInSamples)x(PicHeightInSamples)亮度矩阵refPicLX和两个LLL (PicWidthInSamples)x(PicHeightInSamples)色度矩阵refPicLX、refPicLX组成。CCCbCr - Otherwise (field_pic_flag is equal to 0), the following applies. - 否则(field_pic_flag等于0)，使用下面的步骤得出参考图像: - If the current macroblock is a frame macroblock, the reference frame or complementary reference field pair referred by PicNum = RefPicListX[ refIdxLX ] or LongTermPicNum = RefPicListX[ refIdxLX ] shall be the output. The output reference frame or complementary reference field pair consists of a (PicWidthInSamples)x(PicHeightInSamples) array of luma samples refPicLX and two LLL (PicWidthInSamples)x(PicHeightInSamples) arrays of chroma samples refPicLX and CCCb refPicLX. Cr - 如果当前宏块是帧宏块，由PicNum = RefPicListX[ refIdxLX ] 或LongTermPicNum , RefPicListX[ refIdxLX ]指定的参考帧或互补参考场对作为输出。参考帧或互补参考场 对由一个(PicWidthInSamples)x(PicHeightInSamples)的亮度矩阵refPicLX和两个LLL (PicWidthInSamples)x(PicHeightInSamples)的色度矩阵refPicLX、refPicLX组成。CCCbCr 3 - Otherwise (the current macroblock is a field macroblock), the following applies. - 否则(当前宏块为场宏块)，使用下面的步骤得出参考图像: - Let refFrame be the reference frame or complementary reference field pair that is referred by PicNum = RefPicListX[ refIdxLX / 2 ] or LongTermPicNum = RefPicListX[ refIdxLX / 2 ]. - 令refFrame作为由PicNum = RefPicListX[ refIdxLX / 2 ]或LongTermPicNum, RefPicListX[ refIdxLX / 2 ]指定的参考帧或互补参考场对。 ,译注: refIdxLX 居然可以除以2，说明当图像是帧编码但宏块是场编码时，refIdxLX 要将奇偶性(差)作为最低位。, - The field of refFrame is selected as follows. - If refIdxLX % 2 is equal to 0, the field of refFrame that has the same parity as the current macroblock shall be the output. - Otherwise (refIdxLX % 2 is equal to 1), the field of refFrame that has the opposite parity as the current macroblock shall be the output. - refFrame的场按照下面规定进行选择。 - 如果refIdxLX % 2等于0，输出和当前宏块具有相同奇偶性的refFrame的场。 - 否则(refIdxLX % 2等于1)，输出和当前宏块具有相反奇偶性的refFrame的 场。 - The output reference field or field of a reference frame consists of a (PicWidthInSamples)x(PicHeightInSamples/2) array of luma samples refPicLX and LLL two (PicWidthInSamples)x(PicHeightInSamples/2) arrays of chroma samples CC refPicLX and refPicLX. CbCr - 输出参考场或参考帧中的场由一个(PicWidthInSamples)x(PicHeightInSamples/2)的LL 亮度矩阵refPicLX和两个(PicWidthInSamples)x(PicHeightInSamples/2)色度矩阵LCC refPicLX、 refPicLX组成。 CbCr The reference picture sample arrays refPicLX, refPicLX, refPicLX correspond to decoded LCbCr sample arrays S, S, S derived in subclause 错误～未找到引用源。8.7 for previous decoded LCbCr pictures. 参考图像refPicLX、refPicLX、refPicLX对应于错误～未找到引用源。8.7节中已解码图CbCr L 像S、S、S。 LCbCr ,译注:此句的潜在意义待研究。, 8.4.2.2 Fractional sample interpolation process 分象素插值过程 Inputs to this process are 本过程输入的是 - the current partition given by its partition index mbPartIdx and its sub-macroblock partition index subMbPartIdx, - the width and height partWidth, partHeight of this partition in luma-sample units, - a luma motion vector mvLX given in quarter-luma-sample units, - a chroma motion vector mvCLX given in eighth-chroma-sample units, and 4 - the selected reference picture sample arrays refPicLX, refPicLX, and refPicLXLCbCb - 由宏块划分索引mbPartIdx和子宏块划分索引subMbPartIdx给出的当前划分， - 以亮度样点为单位的当前划分的宽partWidth和高partHeight， - 以1/4亮度样点为单位的亮度运动矢量mvLX， - 以1/8色度样点为单位的色度运动矢量mvCLX， - 选定的参考图像(样点矩阵)refPicLXrefPicLX 和 refPicLX。、LCbCb ,译注:以下，在不致引起混淆时，将“以亮度样点为单位的”简称为“整象素”，将“以1/4(1/8，分数)亮度样点为单位的”简称为“1/4(1/8，分)像素”。, Outputs of this process are 本过程输出的是 - a (partWidth)x(partHeight) array predPartLX of prediction luma sample values and L - two (partWidth/2)x(partHeight/2) arrays predPartLX, and predPartLX of prediction CbCr chroma sample values. - 一个(partWidth)x(partHeight)的亮度预测矩阵predPartLX L - 两个(partWidth/2)x(partHeight/2)的色度预测矩阵predPartLX和 predPartLX。CbCr Let ( xA, yA ) be the location given in full-sample units of the upper-left luma sample of the LL current partition given by mbPartIdx\subMbPartIdx relative to the upper-left luma sample location of the given two-dimensional array of luma samples. 令( xA, yA )为当前划分中左上角(亮度样点)相对于给定的(二维亮度)矩阵的左上角(亮LL 度样点)的整象素位置，当前划分由mbPartIdx\subMbPartIdx标明。 ,译注:简单地说，( xA, yA ) 是当前划分的左上角相对于图像原点的位置。,LL Let ( xInt, yInt ) be a luma location given in full-sample units and ( xFrac, yFrac ) be an offset LLLLgiven in quarter-sample units. These variables are used only inside this subclause for specifying general fractional-sample locations inside the reference sample arrays refPicLX, refPicLX, and LCbrefPicLX. Cr 令 ( xInt, yInt ) 为整象素亮度位置，( xFrac, yFrac ) 为1/4象素偏移量。这些变量是为了LLLL 说明在参考矩阵refPicLX、refPicLX 和 refPicLX中的分数样点位置并且只在本节中使LCbCr 用。 For each luma sample location (0 <= x < partWidth, 0 <= y < partHeight) inside the prediction LL luma sample array predLX, the corresponding predicted luma sample value predLX[ x, y ] is LLLLderived as follows: 根据亮度样点在亮度预测矩阵predLX中的位置(0 <= x < partWidth, 0 <= y < partHeight)，LLL如下得出亮度预测样点值predLX[ x, y ]: LLL xInt = xA + ( mvLX[ 0 ] >> 2 ) + x (错误～文档中没有指定样式的文字。8-1) LLL yInt = yA + ( mvLX[ 1 ] >> 2 ) + y (错误～文档中没有指定样式的文字。8-2) LLL 5 xFrac = mvLX[ 0 ] & 3 (错误～文档中没有指定样式的文字。8-3) L yFrac = mvLX[ 1 ] & 3 (错误～文档中没有指定样式的文字。8-4) L - The prediction sample value predLX[ x, y ] is derived by invoking the process specified in LLL subclause 8.4.2.2.1 with ( xInt, yInt ), ( xFrac, yFrac ) and refPicLX given as input.LLLLL - 调用8.4.2.2.1节叙述的过程，用( xInt, yInt )、( xFrac, yFrac ) 和refPicLX作为输入，LLLLL 得到亮度预测值predLX[ x, y ]。 LLL Let ( xInt, yInt ) be a chroma location given in full-sample units and ( xFrac, yFrac ) be an CCCCoffset given in one-eighth sample units. These variables are used only inside this subclause for specifying general fractional-sample locations inside the reference sample arrays refPicLX, and CbrefPicLX. Cr 令 ( xInt, yInt ) 为整象素色度(样点)的位置，( xFrac, yFrac ) 为1/8象素偏移量。这些变CCCC 量是为了说明在参考矩阵refPicLX 和 refPicLX中的分数样点位置并且只在本节中使CbCr 用。 For each chroma sample location (0 <= x < partWidth/2, 0 <= y < partHeight/2) inside the CC prediction chroma sample arrays predPartLX and predPartLX, the corresponding prediction CbCr chroma sample values predPartLX[ x y ] and predPartLX[ x y ] are derived as follows:CbC,CCrC,C 由色度样点在色度预测矩阵predPartLX和predPartLX中的位置(0 <= x < partWidth/2, CbCrC0 <= y > 1 ) + ( mvCLX[ 0 ] >> 3 ) + x (错误～文档中没有指定样式的文字。8-5) CLC yInt = ( yA >> 1 ) + ( mvCLX[ 1 ] >> 3 ) + y (错误～文档中没有指定样式的文字。8-6) CLC xFrac = mvCLX[ 0 ] & 7 (错误～文档中没有指定样式的文字。8-7) C yFrac = mvCLX[ 1 ] & 7 (错误～文档中没有指定样式的文字。8-8) C - The prediction sample value predPartLX[ x y ] is derived by invoking the process CbC,C specified in subclause 8.4.2.2.2 with ( xInt, yInt ), ( xFrac, yFrac ) and refPicLX given CCCCCb as input. - 调用8.4.2.2.2节中叙述的过程，用( xInt, yInt )、( xFrac, yFrac )和refPicLX作为输CCCCCb 入，得到色度预测值predPartLX[ x y ]。 CbC,C - The prediction sample value predPartLX[ x y ] is derived by invoking the process CrC,C specified in subclause 8.4.2.2.2 with ( xInt, yInt ), ( xFrac, yFrac ) and refPicLX given CCCCCr as input. - 调用8.4.2.2.2节中叙述的过程，用( xInt, yInt )、( xFrac, yFrac )和 refPicLX作为输CCCCCr 入，得到色度预测值predPartLX[ x y ]。 CrC,C 8.4.2.2.1 Luma sample interpolation process 亮度样点插值过程 Inputs to this process are 本过程输入的是 - a luma location in full-sample units ( xInt, yInt ), LL - a luma location offset in fractional-sample units ( xFrac, yFrac ), and LL 6 - the luma sample array of the selected reference picture refPicLX L - 整象素亮度位置( xInt, yInt ), LL - 分象素亮度位置偏移量( xFrac, yFrac ), LL - 选定的参考图像的亮度矩阵refPicLX L Output of this process is a predicted luma sample value predPartLX[ x, y ]. LLL 本过程输出的是一个亮度预测矩阵predPartLX[ x, y ]。 LLL AaaB CbbD EFGabcHIJ defg ccddhijkmeeff npqr KLMsNPQ RggS ThhU 图 错误～文档中没有指定样式的文字。8-1 – Integer samples (shaded blocks with upper-case letters) and fractional sample positions (un-shaded blocks with lower-case letters) for quarter sample luma interpolation. 1/4像素亮度插值中的整数样点(带阴影，大写字母)和分数样点位置(不带阴影，小写字母) The variable refPicHeightEffective, which is the height of the effective reference picture luma L array, is derived as follows. 由下面得出有效参考图像亮度矩阵的高refPicHeightEffective。 L – If MbaffFrameFlag is equal to 0 or mb_field_decoding_flag is equal to 0, refPicHeightEffective is L set equal to PicHeightInSamples. L – 如果MbaffFrameFlag等于0或mb_field_decoding_flag等于0，refPicHeightEffective等于L PicHeightInSamples。 L 7 – Otherwise (MbaffFrameFlag is equal to 1 and mb_field_decoding_flag is equal to 1), refPicHeightEffective is set equal to PicHeightInSamples / 2. LL – 否则(MbaffFrameFlag和mb_field_decoding_flag都等于1)，refPicHeightEffective等于L PicHeightInSamples / 2。 L In 图 错误～文档中没有指定样式的文字。8-1, the positions labelled with upper-case letters within shaded blocks represent luma samples at full-sample locations inside the given two-dimensional array refPicLX of luma samples. These samples may be used for generating the L predicted luma sample value predPartLX[ x, y ]. The locations ( xZ, yZ ) for each of the LLLLL corresponding luma samples Z, where Z may be A, B, C, D, E, F, G, H, I, J, K, L, M, N, P, Q, R, S, T, or U, inside the given array refPicLX of luma samples are derived as follows:L 在图 错误～文档中没有指定样式的文字。8-1中用大写字母标明的阴影块位置表示在给定的 二维亮度矩阵refPicLX中的整象素亮度样点位置。这些样点用来生成亮度预测样点L predPartLX[ x, y ]。在给定的亮度矩阵refPicLX中，亮度样点Z的位置( xZ, yZ ) 由下LLLLLL 式得出(Z可以是A、B、C、D、E、F、G、H、I、J、K、L、M、N、P、Q、R、S、T或 U),: xZ = Clip3( 0, PicWidthInSamples – 1, xInt + xDZ ) LLLL yZ = Clip3( 0, refPicHeightEffective – 1, yInt + yDZ )(错误～文档中没有指定样式的文字。8-9) LLLL 表 错误～文档中没有指定样式的文字。8-1 specifies ( xDZ, yDZ ) for different replacements LL of Z. 表 错误～文档中没有指定样式的文字。8-1中根据Z代表的不同象素点规定了( xDZ, yDZ )。LL 表 错误～文档中没有指定样式的文字。8-1 – Differential full-sample luma locations 整象素亮度 位置差值 Z A B C D E F G H I J K L M N P Q R S T U xDZ 0 1 0 1 -2 -1 0 1 2 3 -2 -1 0 1 2 3 0 1 0 1 L yDZ -2 -2 -1 -1 0 0 0 0 0 0 1 1 1 1 1 1 2 2 3 3 L Given the luma samples „A? to „U? at full-sample locations ( xA, yA ) to ( xU, yU ), the luma LLLL samples „a? to „s? at fractional sample positions are derived by the following rules. The luma prediction values at half sample positions shall be derived by applying a 6-tap filter with tap values ( 1, -5, 20, 20, -5, 1 ). The luma prediction values at quarter sample positions shall be derived by averaging samples at full and half sample positions. The process for each fractional position is described below. 给定亮度样点„A?,„U?(其整象素位置是( xA, yA ),( xU, yU )),由下面的 规则 编码规则下载淘宝规则下载天猫规则下载麻将竞赛规则pdf麻将竞赛规则pdf 得出分象LLLL 素位置的亮度样点„a?, „s?。应用权(抽头)值为( 1, -5, 20, 20, -5, 1 )的六阶滤波器得出半象 素位置的亮度预测样点。通过对整象素位置和半象素位置的亮度样点取平均值得出1/4象素 位置的亮度预测样点。下面叙述了如何得出每个分数象素位置的亮度样点。 8 – The samples at half sample positions labelled b shall be derived by first calculating intermediate values denoted as b by applying the 6-tap filter to the nearest integer position samples in the 1 horizontal direction. The samples at half sample positions labelled h shall be derived by first calculating intermediate values denoted as h by applying the 6-tap filter to the nearest integer 1 position samples in the vertical direction: – 先通过对离半象素位置b水平方向最近的整象素位置样点使用6阶滤波器得出中间值b，对离1 半象素位置h垂直方向最近的整象素位置样点使用6阶滤波器得出中间值h:1 b = ( E – 5 * F + 20 * G + 20 * H – 5 * I + J ) (错误～文档中没有指定样式的文字。8-10) 1 h = ( A – 5 * C + 20 * G + 20 * M – 5 * R + T ) (错误～文档中没有指定样式的文字。8-11) 1 The final prediction values b and h shall be derived using: 使用下面两式得出最终的预测样点值b和h: b = Clip1( ( b + 16 ) >> 5 ) (错误～文档中没有指定样式的文字。8-12) 1 h = Clip1( ( h + 16 ) >> 5 ) (错误～文档中没有指定样式的文字。8-13) 1 – The samples at half sample position labelled as j shall be derived by first calculating intermediate value denoted as j by applying the 6-tap filter to the intermediate values of the closest half sample 1 positions in either the horizontal or vertical direction because these yield an equal result. 先通过对离半象素位置j的最近的水平或垂直方向半象素位置样点使用6阶滤波器得出中间值 j(使用水平或垂直方向半象素位置样点得出同样的结果)。 1 j = cc – 5 * dd + 20 * h + 20 * m – 5 * ee + ff, 或 (错误～文档中没有指定样式的文字。8-14) 111 j = aa – 5 * bb + 20 * b + 20 * s – 5 * gg + hh (错误～文档中没有指定样式的文字。8-15) 111 where intermediate values denoted as aa, bb, gg, s and hh shall be derived by applying the 6-tap 1 filter horizontally in the same manner as the derivation of b and intermediate values denoted as cc, 1 dd, ee, m and ff shall be derived by applying the 6-tap filter vertically in the same manner as the 1 derivation of h. The final prediction value j shall be derived using: 1 式中一些中间值的计算方法:按照上面得出b的方法，在水平方向使用6阶滤波器得出中间值1 aa、bb、gg、s和hh;按照上面得出h的方法，在垂直方向使用6阶滤波器得出中间值cc、dd、11 e、m和ff。 1 使用下式得出最终的预测样点值j: j = Clip1( ( j + 512 ) >> 10 ) (错误～文档中没有指定样式的文字。8-16) 1 – The final prediction values s and m shall be derived from s and m in the same manner as the 11 derivation of b and h, as given by: – 用得出b和h的方法，由s和m得出s 和m : 1 1 s = Clip1( ( s + 16 ) >> 5 ) (错误～文档中没有指定样式的文字。8-17) 1 m = Clip1( ( m + 16 ) >> 5 ) (错误～文档中没有指定样式的文字。8-18) 1 9 – The samples at quarter sample positions labelled as a, c, d, n, f, i, k, and q shall be derived by averaging with upward rounding of the two nearest samples at integer and half sample positions using: – 通过对离1/4象素最近的两个整象素和半象素位置的样点求均值并向上取整，得出1/4象素位 a、c、d、n、f、i、k 和q 的样点值: 置 a = ( G + b + 1 ) >> 1 (错误～文档中没有指定样式的文字。8-19) -20) c = ( H + b + 1 ) >> 1 (错误～文档中没有指定样式的文字。8 d = ( G + h + 1 ) >> 1 (错误～文档中没有指定样式的文字。8-21) n = ( M + h + 1 ) >> 1 (错误～文档中没有指定样式的文字。8-22) f = ( b + j + 1 ) >> 1 (错误～文档中没有指定样式的文字。8-23) i = ( h + j + 1 ) >> 1 (错误～文档中没有指定样式的文字。8-24) k = ( j + m + 1 ) >> 1 (错误～文档中没有指定样式的文字。8-25) q = ( j + s + 1 ) >> 1. (错误～文档中没有指定样式的文字。8-26) – The samples at quarter sample positions labelled as e, g, p, and r shall be derived by averaging with upward rounding of the two nearest samples at half sample positions in the diagonal direction using – 通过对对角线方向两个最近的半象素位置的样点求均值并向上取整得出1/4象素位置e、g、p、 和r 的样点值: e = ( b + h + 1 ) >> 1 (错误～文档中没有指定样式的文字。8-27) g = ( b + m + 1 ) >> 1 (错误～文档中没有指定样式的文字。8-28) p = ( h + s + 1 ) >> 1 (错误～文档中没有指定样式的文字。8-29) r = ( m + s + 1 ) >> 1. (错误～文档中没有指定样式的文字。8-30) ,译注:对角线有两个方向，它只取了一个方向(包围中心的方向，而不是中心向外的方 向)。, The luma location offset in fractional-sample units ( xFrac, yFrac ) specifies which of the LL generated luma samples at full-sample and fractional-sample locations is assigned to the predicted luma sample value. This assignment is done according to 表 错误～文档中没有指定样式的文 字。8-2. The value of predPartLX[ x, y ] shall be the output. LLL 按照表 错误～文档中没有指定样式的文字。8-2，根据分象素位置偏移量( xFrac, yFrac ) 将LL 上面求出的某个值赋给亮度预测值predPartLX[ x, y ]。predPartLX[ x, y ]作为输出。LLLLLL 表 错误～文档中没有指定样式的文字。8-2 – Assignment of the luma prediction sample predPartLX[ x, y ] LLL 亮度预测值 predPartLX[ x, y ] 与分像素位置偏移量的关系 LLL xFrac 0 0 0 0 1 1 1 1 2 2 2 2 3 3 3 3 L yFrac 0 1 2 3 0 1 2 3 0 1 2 3 0 1 2 3 L predPartLX[ x, y ] G D h n a E i p b f j q c g k r LLL 10 8.4.2.2.2 Chroma sample interpolation process 色度样点插值过程 Inputs to this process are 本过程输入的是 - a chroma location in full-sample units ( xInt, yInt ), CC - a chroma location offset in fractional-sample units ( xFrac, yFrac ), and CC - chroma component samples from the selected reference picture refPicLX.C - 一个整象素色度样点位置( xInt, yInt ), CC - 一个分象素色度样点位置偏移量( xFrac, yFrac ), CC - 选定参考图像中的色度矩阵refPicLX。 C Output of this process is a predicted chroma sample value predPartLX[ x, y ].CCC 本过程输出的是色度预测值predPartLX[ x, y ] CCC In 图 错误～文档中没有指定样式的文字。8-2, the positions labelled with A, B, C, and D represent chroma samples at full-sample locations inside the given two-dimensional array refPicLX of chroma samples. C 图 错误～文档中没有指定样式的文字。8-2中标有A、B、C和D的位置代表色度整象素样点在给定的二维色度矩阵refPicLX中的位置。 C BA yFracC xFrac8-xFracCC 8-yFracC DC 图 错误～文档中没有指定样式的文字。8-2 – Fractional sample position dependent variables in chroma interpolation and surrounding integer position samples A, B, C, and D. 色度插值中的与分像素位置相关的变量以及周围的整数位置样点A, B, C 和 D The variable refPicHeightEffective, which is the height of the effective reference picture C chroma array, is derived as follows. 由下面得出有效参考图像色度矩阵的高refPicHeightEffective。 C – If MbaffFrameFlag is equal to 0 or mb_field_decoding_flag is equal to 0, refPicHeightEffective C is set equal to PicHeightInSamples. C 11 – 如果MbaffFrameFlag等于0或mb_field_decoding_flag等于0，refPicHeightEffective等于C PicHeightInSamples。 C – Otherwise (MbaffFrameFlag is equal to 1 and mb_field_decoding_flag is equal to 1), refPicHeightEffective is set equal to PicHeightInSamples / 2. CC – 否则(MbaffFrameFlag和 mb_field_decoding_flag都等于1)，refPicHeightEffective等于C PicHeightInSamples / 2。 C These samples may be used for generating the predicted chroma sample value predPartLX[ x, y ]. CCC 下面这些样点用来生成色度预测样点predPartLX[ x, y ]。 CCC ) (错误～文档中没有指定样式的文字。8-31) xA = Clip3( 0, PicWidthInSamples – 1, xIntCCC xB = Clip3( 0, PicWidthInSamples – 1, xInt + 1 ) (错误～文档中没有指定样式的文字。8-32)CCC xC = Clip3( 0, PicWidthInSamples – 1, xInt ) (错误～文档中没有指定样式的文字。8-33)CCC xD = Clip3( 0, PicWidthInSamples – 1, xInt + 1 ) (错误～文档中没有指定样式的文字。8-34) CCC yA = Clip3( 0, refPicHeightEffective – 1, yInt ) (错误～文档中没有指定样式的文字。8-35) CCC yB = Clip3( 0, refPicHeightEffective – 1, yInt ) (错误～文档中没有指定样式的文字。8-36)CCC yC = Clip3( 0, refPicHeightEffective – 1, yInt + 1 ) (错误～文档中没有指定样式的文字。8-37)CCC yD = Clip3( 0, refPicHeightEffective – 1, yInt + 1 ) (错误～文档中没有指定样式的文字。8-38) CCC Given the chroma samples at full-sample locations, the predicted chroma sample value predPartLX[ x, y ] is derived as follows: CCC 通过给定的色度样点A、B、C和D，如下得出色度预测值predPartLX[ x, y ]:CCC predPartLX[ x, y ] = ( ( 8 – xFrac ) * ( 8 – yFrac ) * A + xFrac * ( 8 – yFrac ) * B + CCCCCCC ( 8 – xFrac ) * yFrac * C + xFrac * yFrac * D + 32 ) >> 6(错误～文档中没有指定样式的文字。8-39) CCCC 8.4.2.3 Weighted sample prediction process 加权样点预测过程 Inputs to this process are 本过程输入的是: - mbPartIdx: the current partition given by the partition index - subMbPartIdx: the sub-macroblock partition index - predFlagL0 and predFlagL1: prediction list utilization flags - predPartLX: a (partWidth)x(partHeight) array of prediction luma samples (with LX being L replaced by L0 or L1 depending on predFlagL0 and predFlagL1) - predPartLX and predPartLX: (partWidth/2)x(partHeight/2) arrays of prediction chroma CbCr samples, one for each of the chroma components Cb and Cr (with LX being replaced by L0 or L1 depending on predFlagL0 and predFlagL1) - mbPartIdx:由划分索引给定的当前划分， - subMbPartIdx:由子宏块划分索引给定的子宏块划分 - predFlagL0 和predFlagL1:预测列表使用标志位， - predPartLX:一个(partWidth)x(partHeight)的亮度预测矩阵 (根据用到predFlagL0还是L 12 predFlagL1分别用L0或L1替代LX)， - predPartLX和predPartLX:(partWidth/2)x(partHeight/2)的色度矩阵 (根据用到CbCr predFlagL0还是predFlagL1分别用L0或L1替代LX)，Cb和Cr各有一个。 Outputs of this process are 本过程输出的是: - predPart: a (partWidth)x(partHeight) array of prediction luma samples and L - predPart, and predPart: (partWidth/2)x(partHeight/2) arrays of prediction chroma samples, CbCr one for each of the chroma components Cb and Cr. - predPart:一个(partWidth)x(partHeight)的亮度预测矩阵， L - predPart和predPart:(partWidth/2)x(partHeight/2)的色度预测矩阵，Cb和Cr各一个。CbCr For macroblocks or partitions with predFlagL0 equal to 1 in P and SP slices, the following applies. 对于在P和SP slices中predFlagL0等于1的宏块或划分，执行下面操作: - If weighted_pred_flag is equal to 0, the default weighted sample prediction process as described in subclause 8.4.2.3.1 is invoked with the same inputs and outputs as the process described in this subclause. - 如果weighted_pred_flag等于0，调用8.4.2.3.1节叙述的默认加权预测过程，输入和输出 和本节中是一样的。 - Otherwise (weighted_pred_flag is equal to 1), the explicit weighted prediction process as described in subclause 8.4.2.3.2 is invoked with the same inputs and outputs as the process described in this subclause. - 否则(weighted_pred_flag等于1),调用8.4.2.3.2节中叙述的显式加权预测过程，输入和输 出和本节中是一样的。 For macroblocks or partitions with predFlagL0 or predFlagL1 equal to 1 in B slices, the following applies. 对于在B slices中predFlagL0或predFlagL1等于1的宏块和划分执行下面操作: - If weighted_bipred_idc is equal to 0, the default weighted sample prediction process as described in subclause 8.4.2.3.1 is invoked with the same inputs and outputs as the process described in this subclause. - 如果weighted_bipred_idc等于0，调用8.4.2.3.1节叙述的默认加权预测过程，输入和输 出和本节中是一样的。 - Otherwise, if weighted_bipred_idc is equal to 1, the explicit weighted sample prediction process as described in subclause 8.4.2.3.2, for macroblocks or partitions with predFlagL0 or predFlagL1 equal to 1 with the same inputs and outputs as the process described in this subclause. - 否则，如果weighted_bipred_idc等于1, 则调用8.4.2.3.2节中叙述的显式加权预测过程， 输入和输出和本节中是一样的。 - Otherwise (weighted_bipred_idc is equal to 2), the following applies. - 否则(weighted_bipred_idc等于2),执行下面操作。 13 - If predFlagL0 is equal to 1 and predFlagL1 is equal to 1, the implicit weighted sample prediction as described in subclause 8.4.2.3.2 is invoked with the same inputs and outputs as the process described in this subclause. - 如果predFlagL0和predFlagL1都等于1，调用8.4.2.3.2节中叙述的隐式加权预测过 程，输入和输出和本节中是一样的。 - Otherwise (predFlagL0 or predFlagL1 are equal to 1 but not both), the default weighted sample prediction process as described in subclause 8.4.2.3.1 is invoked with the same inputs and outputs as the process described in this subclause. - 否则(predFlagL0或predFlagL1等于1，但不都等于1)，调用8.4.2.3.1节叙述的默认 加权预测过程，输入和输出和本节中是一样的。 8.4.2.3.1 Default weighted sample prediction process 默认的加权预测过程 Input to this process are the same as specified in subclause 8.4.2.3. Output of this process are the same as specified in subclause 8.4.2.3. 本过程的输入和输出与8.4.2.3描述的输入输出是相同的。 [译注:默认过程的简单描述为:如果只用一个参考图像，则预测值为这个参考图像的值; 如果用了两个参考图象，则预测值为两个图像的值的平均。] Depending on the component for which the prediction block is derived, the following applies. 根据要得到是亮度还是色度样点进行下面处理。 - If the luma sample prediction values predPart[ x, y ] are derived, the following applies with L C set equal to L, x set equal to 0 .. partWidth - 1, and y set equal to 0 .. partHeight - 1. - 如果要得到的是亮度预测值predPart[ x, y ]，使C等于L，x取值范围0 , partWidth – 1，L y取值范围0 , partHeight – 1。 - Otherwise, if the chroma Cb component sample prediction values predPart[ x, y ] are Cb derived, the following applies with C set equal to Cb, x set equal to 0 .. partWidth / 2 - 1, and y set equal to 0 .. partHeight / 2 - 1. - 否则，如果要得到的是色度Cb预测值predPart[ x, y ]，使C等于Cb，x取值范围Cb 0 , partWidth /2– 1，y取值范围0 , partHeight/2 – 1。 - Otherwise (the chroma Cr component sample prediction values predPart[ x, y ] are derived), Cr the following applies with C set equal to Cr, x set equal to 0 .. partWidth / 2 - 1, and y set equal to 0 .. partHeight / 2 - 1. - 否则(要得到的是色度Cr预测值predPart[ x, y ])，使C等于Cr，x取值范围Cr 0 , partWidth /2– 1，y取值范围0 , partHeight/2 – 1。 The prediction sample values are derived as follows. 由下面步骤得出预测值。 - If predFlagL0 is equal to 1 and predFlagL1 is equal to 0 for the current partition - 如果当前划分的predFlagL0等于1且predFlagL1等于0 predPart[ x, y ] = predPartL0[ x, y ] (错误～文档中没有指定样式的文字。8-40) CC - Otherwise, if predFlagL0 is equal to 0 and predFlagL1 is equal to 1 for the current partition 14 - 否则，如果当前划分的predFlagL0等于0且predFlagL1等于1 predPart[ x, y ]= predPartL1[ x, y ] (错误～文档中没有指定样式的文字。8-41) CC - Otherwise (predFlagL0 and predFlagL1 are equal to 1 for the current partition), - 否则(当前划分的predFlagL0和predFlagL1都等于1)， predPart[ x, y ] = ( predPartL0[ x, y ] + predPartL1[ x, y ] + 1 ) >> 1. (错误～文档中没有指定样式的文字。8-42) CCC 8.4.2.3.2 Weighted sample prediction process 加权预测过程 ,译注:其中包含隐式(implicit)和显式(explicit)加权模式。原文中未明确指明什么是 隐式，但使用了“下面的显式”来指明“显式”。故可以认为不是显式的就是隐式。, Input to this process are the same as specified in subclause 8.4.2.3. Output of this process are the same as specified in subclause 8.4.2.3. 本过程的输入和输出与8.4.2.3描述的输入输出是相同的。 Depending on the component for which the prediction block is derived, the following applies. 根据要得到的是亮度还是色度样点进行下面处理。 - If the luma sample prediction values predPart[ x, y ] are derived, the following applies with L C set equal to L, x set equal to 0 .. partWidth - 1, and y set equal to 0 .. partHeight - 1. - 如果要得到的是亮度预测值predPart[ x, y ]，使C等于L，x取值范围0 , partWidth – 1，L y取值范围0 , partHeight – 1。 - Otherwise, if the chroma Cb component sample prediction values predPart[ x, y ] are Cb derived, the following applies with C set equal to Cb, x set equal to 0 .. partWidth / 2 - 1, and y set equal to 0 .. partHeight / 2 - 1. - 否则，如果要得到的是色度Cb预测值predPart[ x, y ]，使C等于Cb，x取值范围Cb 0 , partWidth /2– 1，y取值范围0 , partHeight/2 – 1。 - Otherwise (the chroma Cr component sample prediction values predPart[ x, y ] are derived), Cr the following applies with C set equal to Cr, x set equal to 0 .. partWidth / 2 - 1, and y set equal to 0 .. partHeight / 2 - 1. - 否则(得到的是色度Cr预测值predPart[ x, y ])，使C等于Cr，x取值范围Cr 0 , partWidth /2– 1，y取值范围0 , partHeight/2 – 1。 The prediction sample values are derived as follows 由下面步骤得出预测样点值 - If the partition mbPartIdx\subMbPartIdx has predFlagL0 equal to 1 and predFlagL1 equal to 0, the final predicted sample values predPart[ x, y ] are derived by C - 如果划分mbPartIdx\subMbPartIdx的predFlagL0等于1且predFlagL1等于0，由下面公 式得出最终的预测值predPart[ x, y ] C if( logWD >= 1 ) logWD - 1 predPart[ x, y ] = Clip1( ( ( predPartL0[ x, y ] * w + 2 ) >> logWD ) + o ) CC00 15 else (错误～文档中没有指定样式的文字。8-43) predPart[ x, y ] = Clip1( predPartL0[ x, y ] * w + o ) CC00 - Otherwise, if the partition mbPartIdx?subMbPartIdx has predFlagL0 equal to 0 and predFlagL1 equal to 1, the final predicted sample values predPart[ x, y ] are derived by C - 否则，如果划分mbPartIdx\subMbPartIdx的predFlagL0等于0且predFlagL1等于1，由 下面公式得出最终的预测值predPart[ x, y ] C if( logWD >= 1 ) logWD - 1 predPart[ x, y ] = Clip1( ( ( predPartL1[ x, y ] * w + 2 ) >> logWD ) + o ) CC11 -44) else (错误～文档中没有指定样式的文字。8 predPart[ x, y ] = Clip1( predPartL1[ x, y ] * w + o ) CC11 - Otherwise (the partition mbPartIdx?subMbPartIdx has both predFlagL0 and predFlagL1 equal to 1), the final predicted sample values predPart[ x, y ] are derived by C - 否则(划分mbPartIdx\subMbPartIdx的predFlagL0和predFlagL1都等于1)，由下面公式 得出最终的预测值predPart[ x, y ] C logWDpredPart[ x, y ] = Clip1( ( ( predPartL0[ x, y ] * w + predPartL1[ x, y ] * w + 2 ) >> CC0C1 ( logWD + 1 ) ) + ( ( o + o + 1 ) >> 1 ) )(错误～文档中没有指定样式的文字。8-45) 01 The variables in the above derivation for the prediction samples are derived as follows. 上式中用于得出预测值的变量是用下面的方法得出的:[译注:它的实现在 fill_wp_params( ) 中] If weighted_bipred_idc is equal to 2 and the slice_type is equal to B, - 如果weighted_bipred_idc等于2且slice_type等于B， logWD = 5 (错误～文档中没有指定样式的文字。8-46) o = 0 (错误～文档中没有指定样式的文字。8-47) 0 o = 0 (错误～文档中没有指定样式的文字。8-48) 1 and w0 and w1 are derived as follows. 由下面步骤得到w 和 w。 01 - If DiffPicOrderCnt( picA, picB ) is equal to 0 with picA being the picture referred by RefPicList1[ refIdxL1 ] and picB being the picture referred by RefPicList0[ refIdxL0 ] or one or both reference pictures is a long-term reference picture or (DistScaleFactor >> 2) < -64 or (DistScaleFactor >> 2) > 128 where is specified in subclause 错误～未找到引用源。 8.4.1.2.3 - 令picA和picB分别是RefPicList1[ refIdxL1 ]和RefPicList0[ refIdxL0 ]引用的图像，如 果DiffPicOrderCnt( picA, picB )等于0，或者1个或2个参考帧是长时参考帧，或者 (DistScaleFactor >> 2) < -64 或 (DistScaleFactor >> 2) > 128(错误～未找到引用源。 16 8.4.1.2.3节中叙述了DistScaleFactor[译注:它是运动向量的放缩系数]) w = 32 (错误～文档中没有指定样式的文字。8-49) 0 w = 32 (错误～文档中没有指定样式的文字。8-50) 1 - Otherwise, - 否则， w = 64 – (DistScaleFactor >> 2) (错误～文档中没有指定样式的文字。8-51) 0 w = DistScaleFactor >> 2 (错误～文档中没有指定样式的文字。8-52)1 - Otherwise (weighted_pred_flag is equal to 1 in P or SP slices or weighted_bipred_idc equal to 1 in B slices), explicit mode weighted prediction is used as follows. - 否则(P或SP slices中的weighted_pred_flag等于1或B slices中的weighted_bipred_idc等于1)，应用下面的显式加权预测过程。 - The variables refIdxL0WP and refIdxL1WP are derived as follows. - 如下得到变量refIdxL0WP和refIdxL1WP。 - If MbaffFrameFlag is equal to 1 and the current macroblock is a field macroblock - 如果MbaffFrameFlag等于1且当前宏块是场宏块 refIdxL0WP = refIdxL0 >> 1 (错误～文档中没有指定样式的文字。8-53) refIdxL1WP = refIdxL1 >> 1 (错误～文档中没有指定样式的文字。8-54) - Otherwise (MbaffFrameFlag is equal to 0 or the current macroblock is a frame macroblock), - 否则，(MbaffFrameFlag等于0或当前宏块是帧宏块) refIdxL0WP = refIdxL0 (错误～文档中没有指定样式的文字。8-55) refIdxL1WP = refIdxL1 (错误～文档中没有指定样式的文字。8-56) - The variables logWD, w, w, o, and o are derived as follows. 0101 - 如下得到变量logWD、w、w、o和o。 0101 - If C in predPart[ x, y ] is replaced by L for luma samples C - 如果对于亮度样点(即用L替代predPart[ x, y ]中的C) C logWD = luma_log2_weight_denom (错误～文档中没有指定样式的文字。8-57) w = luma_weight_l0[ refIdxL0WP ] (错误～文档中没有指定样式的文字。8-58) 0 w = luma_weight_l1[ refIdxL1WP ] (错误～文档中没有指定样式的文字。8-59) 1 o = luma_offset_l0[ refIdxL0WP ] (错误～文档中没有指定样式的文字。8-60) 0 17 o = luma_offset_l1[ refIdxL1WP ] (错误～文档中没有指定样式的文字。8-61) 1 - Otherwise (C in predPart[ x, y ] is replaced by Cb or Cr for chroma samples, with C iCbCr = 0 for Cb, iCbCr = 1 for Cr), - 否则(对于色度样点，即用Cb或Cr替代predPart[ x, y ]中的C。对于Cb，iCbCr = 0;C 对于Cr，iCbCr = 1), logWD = chroma_log2_weight_denom (错误～文档中没有指定样式的文字。8-62) w = chroma_weight_l0[ refIdxL0WP ][ iCbCr ](错误～文档中没有指定样式的文字。8-63) 0 w = chroma_weight_l1[ refIdxL1WP ][ iCbCr ](错误～文档中没有指定样式的文字。8-64)1 o = chroma_offset_l0[ refIdxL0WP ][ iCbCr ](错误～文档中没有指定样式的文字。8-65) 0 o = chroma_offset_l1[ refIdxL1WP ][ iCbCr ](错误～文档中没有指定样式的文字。8-66) 1 When in explicit mode weighted prediction mode and predFlagL0 equal to 1 and predFlagL1 equal to 1, the following constraints shall be obeyed 当使用显式加权预测模式且predFlagL0和 predFlagL1都等于1时，有下述限制: -128 <= w + w <= 127 (错误～文档中没有指定样式的文字。8-67) 01 NOTE – For implicit mode weighted prediction, weights are guaranteed to be in the range is -64 <= w, w <= 01128. 注:对于隐式加权预测模式，加权值一定在-64 <= w, w <= 128之内。 01 18