/* The copyright in this software is being made available under the BSD * License, included below. This software may be subject to other third party * and contributor rights, including patent rights, and no such rights are * granted under this license. * * Copyright (c) 2010-2014, ITU/ISO/IEC * All rights reserved. * * Redistribution and use in source and binary forms, with or without * modification, are permitted provided that the following conditions are met: * * * Redistributions of source code must retain the above copyright notice, * this list of conditions and the following disclaimer. * * Redistributions in binary form must reproduce the above copyright notice, * this list of conditions and the following disclaimer in the documentation * and/or other materials provided with the distribution. * * Neither the name of the ITU/ISO/IEC nor the names of its contributors may * be used to endorse or promote products derived from this software without * specific prior written permission. * * THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS "AS IS" * AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE * IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE * ARE DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT HOLDER OR CONTRIBUTORS * BE LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR * CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF * SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS * INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN * CONTRACT, STRICT LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) * ARISING IN ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF * THE POSSIBILITY OF SUCH DAMAGE. */ /** \file TComRdCost.cpp \brief RD cost computation class */ #include #include #include "TComRom.h" #include "TComRdCost.h" //! \ingroup TLibCommon //! \{ TComRdCost::TComRdCost() { init(); } TComRdCost::~TComRdCost() { } // Calculate RD functions Double TComRdCost::calcRdCost( UInt uiBits, Distortion uiDistortion, Bool bFlag, DFunc eDFunc ) { Double dRdCost = 0.0; Double dLambda = 0.0; switch ( eDFunc ) { case DF_SSE: assert(0); break; case DF_SAD: #if RExt__HIGH_BIT_DEPTH_SUPPORT dLambda = m_dLambdaMotionSAD[0]; // 0 is valid, because for lossless blocks, the cost equation is modified to compensate. #else dLambda = (Double)m_uiLambdaMotionSAD[0]; // 0 is valid, because for lossless blocks, the cost equation is modified to compensate. #endif break; case DF_DEFAULT: dLambda = m_dLambda; break; case DF_SSE_FRAME: dLambda = m_dFrameLambda; break; default: assert (0); break; } if (bFlag) //NOTE: this "bFlag" is never true { // Intra8x8, Intra4x4 Block only... if (m_costMode != COST_STANDARD_LOSSY) { dRdCost = (Double(uiDistortion) / dLambda) + Double(uiBits); // all lossless costs would have uiDistortion=0, and therefore this cost function can be used. } else { dRdCost = (((Double)uiDistortion) + ((Double)uiBits * dLambda)); } } else { if (eDFunc == DF_SAD) { if (m_costMode != COST_STANDARD_LOSSY) { dRdCost = ((Double(uiDistortion) * 65536) / dLambda) + Double(uiBits); // all lossless costs would have uiDistortion=0, and therefore this cost function can be used. } else { dRdCost = floor(Double(uiDistortion) + (floor((Double(uiBits) * dLambda) + 0.5) / 65536.0)); } } else { if (m_costMode != COST_STANDARD_LOSSY) { dRdCost = (Double(uiDistortion) / dLambda) + Double(uiBits); // all lossless costs would have uiDistortion=0, and therefore this cost function can be used. } else { dRdCost = floor(Double(uiDistortion) + (Double(uiBits) * dLambda) + 0.5); } } } return dRdCost; } Double TComRdCost::calcRdCost64( UInt64 uiBits, UInt64 uiDistortion, Bool bFlag, DFunc eDFunc ) { Double dRdCost = 0.0; Double dLambda = 0.0; switch ( eDFunc ) { case DF_SSE: assert(0); break; case DF_SAD: #if RExt__HIGH_BIT_DEPTH_SUPPORT dLambda = m_dLambdaMotionSAD[0]; // 0 is valid, because for lossless blocks, the cost equation is modified to compensate. #else dLambda = (Double)m_uiLambdaMotionSAD[0]; // 0 is valid, because for lossless blocks, the cost equation is modified to compensate. #endif break; case DF_DEFAULT: dLambda = m_dLambda; break; case DF_SSE_FRAME: dLambda = m_dFrameLambda; break; default: assert (0); break; } if (bFlag) //NOTE: this "bFlag" is never true { // Intra8x8, Intra4x4 Block only... if (m_costMode != COST_STANDARD_LOSSY) { dRdCost = (Double(uiDistortion) / dLambda) + Double(uiBits); // all lossless costs would have uiDistortion=0, and therefore this cost function can be used. } else { dRdCost = (((Double)(Int64)uiDistortion) + ((Double)(Int64)uiBits * dLambda)); } } else { if (eDFunc == DF_SAD) { if (m_costMode != COST_STANDARD_LOSSY) { dRdCost = ((Double(uiDistortion) * 65536) / dLambda) + Double(uiBits); // all lossless costs would have uiDistortion=0, and therefore this cost function can be used. } else { dRdCost = floor(Double(uiDistortion) + (floor((Double(uiBits) * dLambda) + 0.5) / 65536.0)); } } else { if (m_costMode != COST_STANDARD_LOSSY) { dRdCost = (Double(uiDistortion) / dLambda) + Double(uiBits); // all lossless costs would have uiDistortion=0, and therefore this cost function can be used. } else { dRdCost = floor(Double(uiDistortion) + (Double(uiBits) * dLambda) + 0.5); } } } return dRdCost; } Void TComRdCost::setLambda( Double dLambda ) { m_dLambda = dLambda; m_sqrtLambda = sqrt(m_dLambda); #if RExt__HIGH_BIT_DEPTH_SUPPORT m_dLambdaMotionSAD[0] = 65536.0 * m_sqrtLambda; m_dLambdaMotionSSE[0] = 65536.0 * m_dLambda; #if FULL_NBIT dLambda = 0.57 * pow(2.0, ((LOSSLESS_AND_MIXED_LOSSLESS_RD_COST_TEST_QP_PRIME - 12) / 3.0)); #else dLambda = 0.57 * pow(2.0, ((LOSSLESS_AND_MIXED_LOSSLESS_RD_COST_TEST_QP_PRIME - 12 - 6 * (g_bitDepth[CHANNEL_TYPE_LUMA] - 8)) / 3.0)); #endif m_dLambdaMotionSAD[1] = 65536.0 * sqrt(dLambda); m_dLambdaMotionSSE[1] = 65536.0 * dLambda; #else m_uiLambdaMotionSAD[0] = (UInt)floor(65536.0 * m_sqrtLambda); m_uiLambdaMotionSSE[0] = (UInt)floor(65536.0 * m_dLambda ); #if FULL_NBIT dLambda = 0.57 * pow(2.0, ((LOSSLESS_AND_MIXED_LOSSLESS_RD_COST_TEST_QP_PRIME - 12) / 3.0)); #else dLambda = 0.57 * pow(2.0, ((LOSSLESS_AND_MIXED_LOSSLESS_RD_COST_TEST_QP_PRIME - 12 - 6 * (g_bitDepth[CHANNEL_TYPE_LUMA] - 8)) / 3.0)); #endif m_uiLambdaMotionSAD[1] = (UInt)floor(65536.0 * sqrt(dLambda)); m_uiLambdaMotionSSE[1] = (UInt)floor(65536.0 * dLambda ); #endif } // Initalize Function Pointer by [eDFunc] Void TComRdCost::init() { m_afpDistortFunc[DF_DEFAULT] = NULL; // for DF_DEFAULT m_afpDistortFunc[DF_SSE ] = TComRdCost::xGetSSE; m_afpDistortFunc[DF_SSE4 ] = TComRdCost::xGetSSE4; m_afpDistortFunc[DF_SSE8 ] = TComRdCost::xGetSSE8; m_afpDistortFunc[DF_SSE16 ] = TComRdCost::xGetSSE16; m_afpDistortFunc[DF_SSE32 ] = TComRdCost::xGetSSE32; m_afpDistortFunc[DF_SSE64 ] = TComRdCost::xGetSSE64; m_afpDistortFunc[DF_SSE16N ] = TComRdCost::xGetSSE16N; m_afpDistortFunc[DF_SAD ] = TComRdCost::xGetSAD; m_afpDistortFunc[DF_SAD4 ] = TComRdCost::xGetSAD4; m_afpDistortFunc[DF_SAD8 ] = TComRdCost::xGetSAD8; m_afpDistortFunc[DF_SAD16 ] = TComRdCost::xGetSAD16; m_afpDistortFunc[DF_SAD32 ] = TComRdCost::xGetSAD32; m_afpDistortFunc[DF_SAD64 ] = TComRdCost::xGetSAD64; m_afpDistortFunc[DF_SAD16N ] = TComRdCost::xGetSAD16N; m_afpDistortFunc[DF_SADS ] = TComRdCost::xGetSAD; m_afpDistortFunc[DF_SADS4 ] = TComRdCost::xGetSAD4; m_afpDistortFunc[DF_SADS8 ] = TComRdCost::xGetSAD8; m_afpDistortFunc[DF_SADS16 ] = TComRdCost::xGetSAD16; m_afpDistortFunc[DF_SADS32 ] = TComRdCost::xGetSAD32; m_afpDistortFunc[DF_SADS64 ] = TComRdCost::xGetSAD64; m_afpDistortFunc[DF_SADS16N] = TComRdCost::xGetSAD16N; #if AMP_SAD m_afpDistortFunc[DF_SAD12 ] = TComRdCost::xGetSAD12; m_afpDistortFunc[DF_SAD24 ] = TComRdCost::xGetSAD24; m_afpDistortFunc[DF_SAD48 ] = TComRdCost::xGetSAD48; m_afpDistortFunc[DF_SADS12 ] = TComRdCost::xGetSAD12; m_afpDistortFunc[DF_SADS24 ] = TComRdCost::xGetSAD24; m_afpDistortFunc[DF_SADS48 ] = TComRdCost::xGetSAD48; #endif m_afpDistortFunc[DF_HADS ] = TComRdCost::xGetHADs; m_afpDistortFunc[DF_HADS4 ] = TComRdCost::xGetHADs; m_afpDistortFunc[DF_HADS8 ] = TComRdCost::xGetHADs; m_afpDistortFunc[DF_HADS16 ] = TComRdCost::xGetHADs; m_afpDistortFunc[DF_HADS32 ] = TComRdCost::xGetHADs; m_afpDistortFunc[DF_HADS64 ] = TComRdCost::xGetHADs; m_afpDistortFunc[DF_HADS16N] = TComRdCost::xGetHADs; m_costMode = COST_STANDARD_LOSSY; #if RExt__HIGH_BIT_DEPTH_SUPPORT m_dCost = 0; #else m_uiCost = 0; #endif m_iCostScale = 0; } UInt TComRdCost::xGetComponentBits( Int iVal ) { UInt uiLength = 1; UInt uiTemp = ( iVal <= 0) ? (-iVal<<1)+1: (iVal<<1); assert ( uiTemp ); while ( 1 != uiTemp ) { uiTemp >>= 1; uiLength += 2; } return uiLength; } Void TComRdCost::setDistParam( UInt uiBlkWidth, UInt uiBlkHeight, DFunc eDFunc, DistParam& rcDistParam ) { // set Block Width / Height rcDistParam.iCols = uiBlkWidth; rcDistParam.iRows = uiBlkHeight; rcDistParam.DistFunc = m_afpDistortFunc[eDFunc + g_aucConvertToBit[ rcDistParam.iCols ] + 1 ]; // initialize rcDistParam.iSubShift = 0; } // Setting the Distortion Parameter for Inter (ME) Void TComRdCost::setDistParam( TComPattern* pcPatternKey, Pel* piRefY, Int iRefStride, DistParam& rcDistParam ) { // set Original & Curr Pointer / Stride rcDistParam.pOrg = pcPatternKey->getROIY(); rcDistParam.pCur = piRefY; rcDistParam.iStrideOrg = pcPatternKey->getPatternLStride(); rcDistParam.iStrideCur = iRefStride; // set Block Width / Height rcDistParam.iCols = pcPatternKey->getROIYWidth(); rcDistParam.iRows = pcPatternKey->getROIYHeight(); rcDistParam.DistFunc = m_afpDistortFunc[DF_SAD + g_aucConvertToBit[ rcDistParam.iCols ] + 1 ]; #if AMP_SAD if (rcDistParam.iCols == 12) { rcDistParam.DistFunc = m_afpDistortFunc[DF_SAD12]; } else if (rcDistParam.iCols == 24) { rcDistParam.DistFunc = m_afpDistortFunc[DF_SAD24]; } else if (rcDistParam.iCols == 48) { rcDistParam.DistFunc = m_afpDistortFunc[DF_SAD48]; } #endif // initialize rcDistParam.iSubShift = 0; } // Setting the Distortion Parameter for Inter (subpel ME with step) Void TComRdCost::setDistParam( TComPattern* pcPatternKey, Pel* piRefY, Int iRefStride, Int iStep, DistParam& rcDistParam, Bool bHADME ) { #if O0194_WEIGHTED_PREDICTION_CGS // Bug fix: The correct bit depth has not been used for weighted cost calculation rcDistParam.bitDepth = g_bitDepth[CHANNEL_TYPE_LUMA]; #endif // set Original & Curr Pointer / Stride rcDistParam.pOrg = pcPatternKey->getROIY(); rcDistParam.pCur = piRefY; rcDistParam.iStrideOrg = pcPatternKey->getPatternLStride(); rcDistParam.iStrideCur = iRefStride * iStep; // set Step for interpolated buffer rcDistParam.iStep = iStep; // set Block Width / Height rcDistParam.iCols = pcPatternKey->getROIYWidth(); rcDistParam.iRows = pcPatternKey->getROIYHeight(); // set distortion function if ( !bHADME ) { rcDistParam.DistFunc = m_afpDistortFunc[DF_SADS + g_aucConvertToBit[ rcDistParam.iCols ] + 1 ]; #if AMP_SAD if (rcDistParam.iCols == 12) { rcDistParam.DistFunc = m_afpDistortFunc[DF_SADS12]; } else if (rcDistParam.iCols == 24) { rcDistParam.DistFunc = m_afpDistortFunc[DF_SADS24]; } else if (rcDistParam.iCols == 48) { rcDistParam.DistFunc = m_afpDistortFunc[DF_SADS48]; } #endif } else { rcDistParam.DistFunc = m_afpDistortFunc[DF_HADS + g_aucConvertToBit[ rcDistParam.iCols ] + 1 ]; } // initialize rcDistParam.iSubShift = 0; } Void TComRdCost::setDistParam( DistParam& rcDP, Int bitDepth, Pel* p1, Int iStride1, Pel* p2, Int iStride2, Int iWidth, Int iHeight, Bool bHadamard ) { rcDP.pOrg = p1; rcDP.pCur = p2; rcDP.iStrideOrg = iStride1; rcDP.iStrideCur = iStride2; rcDP.iCols = iWidth; rcDP.iRows = iHeight; rcDP.iStep = 1; rcDP.iSubShift = 0; rcDP.bitDepth = bitDepth; rcDP.DistFunc = m_afpDistortFunc[ ( bHadamard ? DF_HADS : DF_SADS ) + g_aucConvertToBit[ iWidth ] + 1 ]; } Distortion TComRdCost::calcHAD( Int bitDepth, Pel* pi0, Int iStride0, Pel* pi1, Int iStride1, Int iWidth, Int iHeight ) { Distortion uiSum = 0; Int x, y; if ( ( (iWidth % 8) == 0 ) && ( (iHeight % 8) == 0 ) ) { for ( y=0; y> DISTORTION_PRECISION_ADJUSTMENT(bitDepth-8) ); } Distortion TComRdCost::getDistPart( Int bitDepth, Pel* piCur, Int iCurStride, Pel* piOrg, Int iOrgStride, UInt uiBlkWidth, UInt uiBlkHeight, const ComponentID compID, DFunc eDFunc ) { DistParam cDtParam; setDistParam( uiBlkWidth, uiBlkHeight, eDFunc, cDtParam ); cDtParam.pOrg = piOrg; cDtParam.pCur = piCur; cDtParam.iStrideOrg = iOrgStride; cDtParam.iStrideCur = iCurStride; cDtParam.iStep = 1; cDtParam.bApplyWeight = false; cDtParam.compIdx = MAX_NUM_COMPONENT; // just for assert: to be sure it was set before use cDtParam.bitDepth = bitDepth; if (isChroma(compID)) { return ((Distortion) (m_distortionWeight[compID] * cDtParam.DistFunc( &cDtParam ))); } else { return cDtParam.DistFunc( &cDtParam ); } } // ==================================================================================================================== // Distortion functions // ==================================================================================================================== // -------------------------------------------------------------------------------------------------------------------- // SAD // -------------------------------------------------------------------------------------------------------------------- Distortion TComRdCost::xGetSAD( DistParam* pcDtParam ) { if ( pcDtParam->bApplyWeight ) { return TComRdCostWeightPrediction::xGetSADw( pcDtParam ); } const Pel* piOrg = pcDtParam->pOrg; const Pel* piCur = pcDtParam->pCur; Int iRows = pcDtParam->iRows; Int iCols = pcDtParam->iCols; Int iStrideCur = pcDtParam->iStrideCur; Int iStrideOrg = pcDtParam->iStrideOrg; Distortion uiSum = 0; for( ; iRows != 0; iRows-- ) { for (Int n = 0; n < iCols; n++ ) { uiSum += abs( piOrg[n] - piCur[n] ); } piOrg += iStrideOrg; piCur += iStrideCur; } return ( uiSum >> DISTORTION_PRECISION_ADJUSTMENT(pcDtParam->bitDepth-8) ); } Distortion TComRdCost::xGetSAD4( DistParam* pcDtParam ) { if ( pcDtParam->bApplyWeight ) { return TComRdCostWeightPrediction::xGetSADw( pcDtParam ); } const Pel* piOrg = pcDtParam->pOrg; const Pel* piCur = pcDtParam->pCur; Int iRows = pcDtParam->iRows; Int iSubShift = pcDtParam->iSubShift; Int iSubStep = ( 1 << iSubShift ); Int iStrideCur = pcDtParam->iStrideCur*iSubStep; Int iStrideOrg = pcDtParam->iStrideOrg*iSubStep; Distortion uiSum = 0; for( ; iRows != 0; iRows-=iSubStep ) { uiSum += abs( piOrg[0] - piCur[0] ); uiSum += abs( piOrg[1] - piCur[1] ); uiSum += abs( piOrg[2] - piCur[2] ); uiSum += abs( piOrg[3] - piCur[3] ); piOrg += iStrideOrg; piCur += iStrideCur; } uiSum <<= iSubShift; return ( uiSum >> DISTORTION_PRECISION_ADJUSTMENT(pcDtParam->bitDepth-8) ); } Distortion TComRdCost::xGetSAD8( DistParam* pcDtParam ) { if ( pcDtParam->bApplyWeight ) { return TComRdCostWeightPrediction::xGetSADw( pcDtParam ); } const Pel* piOrg = pcDtParam->pOrg; const Pel* piCur = pcDtParam->pCur; Int iRows = pcDtParam->iRows; Int iSubShift = pcDtParam->iSubShift; Int iSubStep = ( 1 << iSubShift ); Int iStrideCur = pcDtParam->iStrideCur*iSubStep; Int iStrideOrg = pcDtParam->iStrideOrg*iSubStep; Distortion uiSum = 0; for( ; iRows != 0; iRows-=iSubStep ) { uiSum += abs( piOrg[0] - piCur[0] ); uiSum += abs( piOrg[1] - piCur[1] ); uiSum += abs( piOrg[2] - piCur[2] ); uiSum += abs( piOrg[3] - piCur[3] ); uiSum += abs( piOrg[4] - piCur[4] ); uiSum += abs( piOrg[5] - piCur[5] ); uiSum += abs( piOrg[6] - piCur[6] ); uiSum += abs( piOrg[7] - piCur[7] ); piOrg += iStrideOrg; piCur += iStrideCur; } uiSum <<= iSubShift; return ( uiSum >> DISTORTION_PRECISION_ADJUSTMENT(pcDtParam->bitDepth-8) ); } Distortion TComRdCost::xGetSAD16( DistParam* pcDtParam ) { if ( pcDtParam->bApplyWeight ) { return TComRdCostWeightPrediction::xGetSADw( pcDtParam ); } const Pel* piOrg = pcDtParam->pOrg; const Pel* piCur = pcDtParam->pCur; Int iRows = pcDtParam->iRows; Int iSubShift = pcDtParam->iSubShift; Int iSubStep = ( 1 << iSubShift ); Int iStrideCur = pcDtParam->iStrideCur*iSubStep; Int iStrideOrg = pcDtParam->iStrideOrg*iSubStep; Distortion uiSum = 0; for( ; iRows != 0; iRows-=iSubStep ) { uiSum += abs( piOrg[0] - piCur[0] ); uiSum += abs( piOrg[1] - piCur[1] ); uiSum += abs( piOrg[2] - piCur[2] ); uiSum += abs( piOrg[3] - piCur[3] ); uiSum += abs( piOrg[4] - piCur[4] ); uiSum += abs( piOrg[5] - piCur[5] ); uiSum += abs( piOrg[6] - piCur[6] ); uiSum += abs( piOrg[7] - piCur[7] ); uiSum += abs( piOrg[8] - piCur[8] ); uiSum += abs( piOrg[9] - piCur[9] ); uiSum += abs( piOrg[10] - piCur[10] ); uiSum += abs( piOrg[11] - piCur[11] ); uiSum += abs( piOrg[12] - piCur[12] ); uiSum += abs( piOrg[13] - piCur[13] ); uiSum += abs( piOrg[14] - piCur[14] ); uiSum += abs( piOrg[15] - piCur[15] ); piOrg += iStrideOrg; piCur += iStrideCur; } uiSum <<= iSubShift; return ( uiSum >> DISTORTION_PRECISION_ADJUSTMENT(pcDtParam->bitDepth-8) ); } #if AMP_SAD Distortion TComRdCost::xGetSAD12( DistParam* pcDtParam ) { if ( pcDtParam->bApplyWeight ) { return TComRdCostWeightPrediction::xGetSADw( pcDtParam ); } const Pel* piOrg = pcDtParam->pOrg; const Pel* piCur = pcDtParam->pCur; Int iRows = pcDtParam->iRows; Int iSubShift = pcDtParam->iSubShift; Int iSubStep = ( 1 << iSubShift ); Int iStrideCur = pcDtParam->iStrideCur*iSubStep; Int iStrideOrg = pcDtParam->iStrideOrg*iSubStep; Distortion uiSum = 0; for( ; iRows != 0; iRows-=iSubStep ) { uiSum += abs( piOrg[0] - piCur[0] ); uiSum += abs( piOrg[1] - piCur[1] ); uiSum += abs( piOrg[2] - piCur[2] ); uiSum += abs( piOrg[3] - piCur[3] ); uiSum += abs( piOrg[4] - piCur[4] ); uiSum += abs( piOrg[5] - piCur[5] ); uiSum += abs( piOrg[6] - piCur[6] ); uiSum += abs( piOrg[7] - piCur[7] ); uiSum += abs( piOrg[8] - piCur[8] ); uiSum += abs( piOrg[9] - piCur[9] ); uiSum += abs( piOrg[10] - piCur[10] ); uiSum += abs( piOrg[11] - piCur[11] ); piOrg += iStrideOrg; piCur += iStrideCur; } uiSum <<= iSubShift; return ( uiSum >> DISTORTION_PRECISION_ADJUSTMENT(pcDtParam->bitDepth-8) ); } #endif Distortion TComRdCost::xGetSAD16N( DistParam* pcDtParam ) { const Pel* piOrg = pcDtParam->pOrg; const Pel* piCur = pcDtParam->pCur; Int iRows = pcDtParam->iRows; Int iCols = pcDtParam->iCols; Int iSubShift = pcDtParam->iSubShift; Int iSubStep = ( 1 << iSubShift ); Int iStrideCur = pcDtParam->iStrideCur*iSubStep; Int iStrideOrg = pcDtParam->iStrideOrg*iSubStep; Distortion uiSum = 0; for( ; iRows != 0; iRows-=iSubStep ) { for (Int n = 0; n < iCols; n+=16 ) { uiSum += abs( piOrg[n+ 0] - piCur[n+ 0] ); uiSum += abs( piOrg[n+ 1] - piCur[n+ 1] ); uiSum += abs( piOrg[n+ 2] - piCur[n+ 2] ); uiSum += abs( piOrg[n+ 3] - piCur[n+ 3] ); uiSum += abs( piOrg[n+ 4] - piCur[n+ 4] ); uiSum += abs( piOrg[n+ 5] - piCur[n+ 5] ); uiSum += abs( piOrg[n+ 6] - piCur[n+ 6] ); uiSum += abs( piOrg[n+ 7] - piCur[n+ 7] ); uiSum += abs( piOrg[n+ 8] - piCur[n+ 8] ); uiSum += abs( piOrg[n+ 9] - piCur[n+ 9] ); uiSum += abs( piOrg[n+10] - piCur[n+10] ); uiSum += abs( piOrg[n+11] - piCur[n+11] ); uiSum += abs( piOrg[n+12] - piCur[n+12] ); uiSum += abs( piOrg[n+13] - piCur[n+13] ); uiSum += abs( piOrg[n+14] - piCur[n+14] ); uiSum += abs( piOrg[n+15] - piCur[n+15] ); } piOrg += iStrideOrg; piCur += iStrideCur; } uiSum <<= iSubShift; return ( uiSum >> DISTORTION_PRECISION_ADJUSTMENT(pcDtParam->bitDepth-8) ); } Distortion TComRdCost::xGetSAD32( DistParam* pcDtParam ) { if ( pcDtParam->bApplyWeight ) { return TComRdCostWeightPrediction::xGetSADw( pcDtParam ); } const Pel* piOrg = pcDtParam->pOrg; const Pel* piCur = pcDtParam->pCur; Int iRows = pcDtParam->iRows; Int iSubShift = pcDtParam->iSubShift; Int iSubStep = ( 1 << iSubShift ); Int iStrideCur = pcDtParam->iStrideCur*iSubStep; Int iStrideOrg = pcDtParam->iStrideOrg*iSubStep; Distortion uiSum = 0; for( ; iRows != 0; iRows-=iSubStep ) { uiSum += abs( piOrg[0] - piCur[0] ); uiSum += abs( piOrg[1] - piCur[1] ); uiSum += abs( piOrg[2] - piCur[2] ); uiSum += abs( piOrg[3] - piCur[3] ); uiSum += abs( piOrg[4] - piCur[4] ); uiSum += abs( piOrg[5] - piCur[5] ); uiSum += abs( piOrg[6] - piCur[6] ); uiSum += abs( piOrg[7] - piCur[7] ); uiSum += abs( piOrg[8] - piCur[8] ); uiSum += abs( piOrg[9] - piCur[9] ); uiSum += abs( piOrg[10] - piCur[10] ); uiSum += abs( piOrg[11] - piCur[11] ); uiSum += abs( piOrg[12] - piCur[12] ); uiSum += abs( piOrg[13] - piCur[13] ); uiSum += abs( piOrg[14] - piCur[14] ); uiSum += abs( piOrg[15] - piCur[15] ); uiSum += abs( piOrg[16] - piCur[16] ); uiSum += abs( piOrg[17] - piCur[17] ); uiSum += abs( piOrg[18] - piCur[18] ); uiSum += abs( piOrg[19] - piCur[19] ); uiSum += abs( piOrg[20] - piCur[20] ); uiSum += abs( piOrg[21] - piCur[21] ); uiSum += abs( piOrg[22] - piCur[22] ); uiSum += abs( piOrg[23] - piCur[23] ); uiSum += abs( piOrg[24] - piCur[24] ); uiSum += abs( piOrg[25] - piCur[25] ); uiSum += abs( piOrg[26] - piCur[26] ); uiSum += abs( piOrg[27] - piCur[27] ); uiSum += abs( piOrg[28] - piCur[28] ); uiSum += abs( piOrg[29] - piCur[29] ); uiSum += abs( piOrg[30] - piCur[30] ); uiSum += abs( piOrg[31] - piCur[31] ); piOrg += iStrideOrg; piCur += iStrideCur; } uiSum <<= iSubShift; return ( uiSum >> DISTORTION_PRECISION_ADJUSTMENT(pcDtParam->bitDepth-8) ); } #if AMP_SAD Distortion TComRdCost::xGetSAD24( DistParam* pcDtParam ) { if ( pcDtParam->bApplyWeight ) { return TComRdCostWeightPrediction::xGetSADw( pcDtParam ); } const Pel* piOrg = pcDtParam->pOrg; const Pel* piCur = pcDtParam->pCur; Int iRows = pcDtParam->iRows; Int iSubShift = pcDtParam->iSubShift; Int iSubStep = ( 1 << iSubShift ); Int iStrideCur = pcDtParam->iStrideCur*iSubStep; Int iStrideOrg = pcDtParam->iStrideOrg*iSubStep; Distortion uiSum = 0; for( ; iRows != 0; iRows-=iSubStep ) { uiSum += abs( piOrg[0] - piCur[0] ); uiSum += abs( piOrg[1] - piCur[1] ); uiSum += abs( piOrg[2] - piCur[2] ); uiSum += abs( piOrg[3] - piCur[3] ); uiSum += abs( piOrg[4] - piCur[4] ); uiSum += abs( piOrg[5] - piCur[5] ); uiSum += abs( piOrg[6] - piCur[6] ); uiSum += abs( piOrg[7] - piCur[7] ); uiSum += abs( piOrg[8] - piCur[8] ); uiSum += abs( piOrg[9] - piCur[9] ); uiSum += abs( piOrg[10] - piCur[10] ); uiSum += abs( piOrg[11] - piCur[11] ); uiSum += abs( piOrg[12] - piCur[12] ); uiSum += abs( piOrg[13] - piCur[13] ); uiSum += abs( piOrg[14] - piCur[14] ); uiSum += abs( piOrg[15] - piCur[15] ); uiSum += abs( piOrg[16] - piCur[16] ); uiSum += abs( piOrg[17] - piCur[17] ); uiSum += abs( piOrg[18] - piCur[18] ); uiSum += abs( piOrg[19] - piCur[19] ); uiSum += abs( piOrg[20] - piCur[20] ); uiSum += abs( piOrg[21] - piCur[21] ); uiSum += abs( piOrg[22] - piCur[22] ); uiSum += abs( piOrg[23] - piCur[23] ); piOrg += iStrideOrg; piCur += iStrideCur; } uiSum <<= iSubShift; return ( uiSum >> DISTORTION_PRECISION_ADJUSTMENT(pcDtParam->bitDepth-8) ); } #endif Distortion TComRdCost::xGetSAD64( DistParam* pcDtParam ) { if ( pcDtParam->bApplyWeight ) { return TComRdCostWeightPrediction::xGetSADw( pcDtParam ); } const Pel* piOrg = pcDtParam->pOrg; const Pel* piCur = pcDtParam->pCur; Int iRows = pcDtParam->iRows; Int iSubShift = pcDtParam->iSubShift; Int iSubStep = ( 1 << iSubShift ); Int iStrideCur = pcDtParam->iStrideCur*iSubStep; Int iStrideOrg = pcDtParam->iStrideOrg*iSubStep; Distortion uiSum = 0; for( ; iRows != 0; iRows-=iSubStep ) { uiSum += abs( piOrg[0] - piCur[0] ); uiSum += abs( piOrg[1] - piCur[1] ); uiSum += abs( piOrg[2] - piCur[2] ); uiSum += abs( piOrg[3] - piCur[3] ); uiSum += abs( piOrg[4] - piCur[4] ); uiSum += abs( piOrg[5] - piCur[5] ); uiSum += abs( piOrg[6] - piCur[6] ); uiSum += abs( piOrg[7] - piCur[7] ); uiSum += abs( piOrg[8] - piCur[8] ); uiSum += abs( piOrg[9] - piCur[9] ); uiSum += abs( piOrg[10] - piCur[10] ); uiSum += abs( piOrg[11] - piCur[11] ); uiSum += abs( piOrg[12] - piCur[12] ); uiSum += abs( piOrg[13] - piCur[13] ); uiSum += abs( piOrg[14] - piCur[14] ); uiSum += abs( piOrg[15] - piCur[15] ); uiSum += abs( piOrg[16] - piCur[16] ); uiSum += abs( piOrg[17] - piCur[17] ); uiSum += abs( piOrg[18] - piCur[18] ); uiSum += abs( piOrg[19] - piCur[19] ); uiSum += abs( piOrg[20] - piCur[20] ); uiSum += abs( piOrg[21] - piCur[21] ); uiSum += abs( piOrg[22] - piCur[22] ); uiSum += abs( piOrg[23] - piCur[23] ); uiSum += abs( piOrg[24] - piCur[24] ); uiSum += abs( piOrg[25] - piCur[25] ); uiSum += abs( piOrg[26] - piCur[26] ); uiSum += abs( piOrg[27] - piCur[27] ); uiSum += abs( piOrg[28] - piCur[28] ); uiSum += abs( piOrg[29] - piCur[29] ); uiSum += abs( piOrg[30] - piCur[30] ); uiSum += abs( piOrg[31] - piCur[31] ); uiSum += abs( piOrg[32] - piCur[32] ); uiSum += abs( piOrg[33] - piCur[33] ); uiSum += abs( piOrg[34] - piCur[34] ); uiSum += abs( piOrg[35] - piCur[35] ); uiSum += abs( piOrg[36] - piCur[36] ); uiSum += abs( piOrg[37] - piCur[37] ); uiSum += abs( piOrg[38] - piCur[38] ); uiSum += abs( piOrg[39] - piCur[39] ); uiSum += abs( piOrg[40] - piCur[40] ); uiSum += abs( piOrg[41] - piCur[41] ); uiSum += abs( piOrg[42] - piCur[42] ); uiSum += abs( piOrg[43] - piCur[43] ); uiSum += abs( piOrg[44] - piCur[44] ); uiSum += abs( piOrg[45] - piCur[45] ); uiSum += abs( piOrg[46] - piCur[46] ); uiSum += abs( piOrg[47] - piCur[47] ); uiSum += abs( piOrg[48] - piCur[48] ); uiSum += abs( piOrg[49] - piCur[49] ); uiSum += abs( piOrg[50] - piCur[50] ); uiSum += abs( piOrg[51] - piCur[51] ); uiSum += abs( piOrg[52] - piCur[52] ); uiSum += abs( piOrg[53] - piCur[53] ); uiSum += abs( piOrg[54] - piCur[54] ); uiSum += abs( piOrg[55] - piCur[55] ); uiSum += abs( piOrg[56] - piCur[56] ); uiSum += abs( piOrg[57] - piCur[57] ); uiSum += abs( piOrg[58] - piCur[58] ); uiSum += abs( piOrg[59] - piCur[59] ); uiSum += abs( piOrg[60] - piCur[60] ); uiSum += abs( piOrg[61] - piCur[61] ); uiSum += abs( piOrg[62] - piCur[62] ); uiSum += abs( piOrg[63] - piCur[63] ); piOrg += iStrideOrg; piCur += iStrideCur; } uiSum <<= iSubShift; return ( uiSum >> DISTORTION_PRECISION_ADJUSTMENT(pcDtParam->bitDepth-8) ); } #if AMP_SAD Distortion TComRdCost::xGetSAD48( DistParam* pcDtParam ) { if ( pcDtParam->bApplyWeight ) { return TComRdCostWeightPrediction::xGetSADw( pcDtParam ); } const Pel* piOrg = pcDtParam->pOrg; const Pel* piCur = pcDtParam->pCur; Int iRows = pcDtParam->iRows; Int iSubShift = pcDtParam->iSubShift; Int iSubStep = ( 1 << iSubShift ); Int iStrideCur = pcDtParam->iStrideCur*iSubStep; Int iStrideOrg = pcDtParam->iStrideOrg*iSubStep; Distortion uiSum = 0; for( ; iRows != 0; iRows-=iSubStep ) { uiSum += abs( piOrg[0] - piCur[0] ); uiSum += abs( piOrg[1] - piCur[1] ); uiSum += abs( piOrg[2] - piCur[2] ); uiSum += abs( piOrg[3] - piCur[3] ); uiSum += abs( piOrg[4] - piCur[4] ); uiSum += abs( piOrg[5] - piCur[5] ); uiSum += abs( piOrg[6] - piCur[6] ); uiSum += abs( piOrg[7] - piCur[7] ); uiSum += abs( piOrg[8] - piCur[8] ); uiSum += abs( piOrg[9] - piCur[9] ); uiSum += abs( piOrg[10] - piCur[10] ); uiSum += abs( piOrg[11] - piCur[11] ); uiSum += abs( piOrg[12] - piCur[12] ); uiSum += abs( piOrg[13] - piCur[13] ); uiSum += abs( piOrg[14] - piCur[14] ); uiSum += abs( piOrg[15] - piCur[15] ); uiSum += abs( piOrg[16] - piCur[16] ); uiSum += abs( piOrg[17] - piCur[17] ); uiSum += abs( piOrg[18] - piCur[18] ); uiSum += abs( piOrg[19] - piCur[19] ); uiSum += abs( piOrg[20] - piCur[20] ); uiSum += abs( piOrg[21] - piCur[21] ); uiSum += abs( piOrg[22] - piCur[22] ); uiSum += abs( piOrg[23] - piCur[23] ); uiSum += abs( piOrg[24] - piCur[24] ); uiSum += abs( piOrg[25] - piCur[25] ); uiSum += abs( piOrg[26] - piCur[26] ); uiSum += abs( piOrg[27] - piCur[27] ); uiSum += abs( piOrg[28] - piCur[28] ); uiSum += abs( piOrg[29] - piCur[29] ); uiSum += abs( piOrg[30] - piCur[30] ); uiSum += abs( piOrg[31] - piCur[31] ); uiSum += abs( piOrg[32] - piCur[32] ); uiSum += abs( piOrg[33] - piCur[33] ); uiSum += abs( piOrg[34] - piCur[34] ); uiSum += abs( piOrg[35] - piCur[35] ); uiSum += abs( piOrg[36] - piCur[36] ); uiSum += abs( piOrg[37] - piCur[37] ); uiSum += abs( piOrg[38] - piCur[38] ); uiSum += abs( piOrg[39] - piCur[39] ); uiSum += abs( piOrg[40] - piCur[40] ); uiSum += abs( piOrg[41] - piCur[41] ); uiSum += abs( piOrg[42] - piCur[42] ); uiSum += abs( piOrg[43] - piCur[43] ); uiSum += abs( piOrg[44] - piCur[44] ); uiSum += abs( piOrg[45] - piCur[45] ); uiSum += abs( piOrg[46] - piCur[46] ); uiSum += abs( piOrg[47] - piCur[47] ); piOrg += iStrideOrg; piCur += iStrideCur; } uiSum <<= iSubShift; return ( uiSum >> DISTORTION_PRECISION_ADJUSTMENT(pcDtParam->bitDepth-8) ); } #endif // -------------------------------------------------------------------------------------------------------------------- // SSE // -------------------------------------------------------------------------------------------------------------------- Distortion TComRdCost::xGetSSE( DistParam* pcDtParam ) { if ( pcDtParam->bApplyWeight ) { return TComRdCostWeightPrediction::xGetSSEw( pcDtParam ); } const Pel* piOrg = pcDtParam->pOrg; const Pel* piCur = pcDtParam->pCur; Int iRows = pcDtParam->iRows; Int iCols = pcDtParam->iCols; Int iStrideOrg = pcDtParam->iStrideOrg; Int iStrideCur = pcDtParam->iStrideCur; Distortion uiSum = 0; UInt uiShift = DISTORTION_PRECISION_ADJUSTMENT((pcDtParam->bitDepth-8) << 1); Intermediate_Int iTemp; for( ; iRows != 0; iRows-- ) { for (Int n = 0; n < iCols; n++ ) { iTemp = piOrg[n ] - piCur[n ]; uiSum += Distortion(( iTemp * iTemp ) >> uiShift); } piOrg += iStrideOrg; piCur += iStrideCur; } return ( uiSum ); } Distortion TComRdCost::xGetSSE4( DistParam* pcDtParam ) { if ( pcDtParam->bApplyWeight ) { assert( pcDtParam->iCols == 4 ); return TComRdCostWeightPrediction::xGetSSEw( pcDtParam ); } const Pel* piOrg = pcDtParam->pOrg; const Pel* piCur = pcDtParam->pCur; Int iRows = pcDtParam->iRows; Int iStrideOrg = pcDtParam->iStrideOrg; Int iStrideCur = pcDtParam->iStrideCur; Distortion uiSum = 0; UInt uiShift = DISTORTION_PRECISION_ADJUSTMENT((pcDtParam->bitDepth-8) << 1); Intermediate_Int iTemp; for( ; iRows != 0; iRows-- ) { iTemp = piOrg[0] - piCur[0]; uiSum += Distortion(( iTemp * iTemp ) >> uiShift); iTemp = piOrg[1] - piCur[1]; uiSum += Distortion(( iTemp * iTemp ) >> uiShift); iTemp = piOrg[2] - piCur[2]; uiSum += Distortion(( iTemp * iTemp ) >> uiShift); iTemp = piOrg[3] - piCur[3]; uiSum += Distortion(( iTemp * iTemp ) >> uiShift); piOrg += iStrideOrg; piCur += iStrideCur; } return ( uiSum ); } Distortion TComRdCost::xGetSSE8( DistParam* pcDtParam ) { if ( pcDtParam->bApplyWeight ) { assert( pcDtParam->iCols == 8 ); return TComRdCostWeightPrediction::xGetSSEw( pcDtParam ); } const Pel* piOrg = pcDtParam->pOrg; const Pel* piCur = pcDtParam->pCur; Int iRows = pcDtParam->iRows; Int iStrideOrg = pcDtParam->iStrideOrg; Int iStrideCur = pcDtParam->iStrideCur; Distortion uiSum = 0; UInt uiShift = DISTORTION_PRECISION_ADJUSTMENT((pcDtParam->bitDepth-8) << 1); Intermediate_Int iTemp; for( ; iRows != 0; iRows-- ) { iTemp = piOrg[0] - piCur[0]; uiSum += Distortion(( iTemp * iTemp ) >> uiShift); iTemp = piOrg[1] - piCur[1]; uiSum += Distortion(( iTemp * iTemp ) >> uiShift); iTemp = piOrg[2] - piCur[2]; uiSum += Distortion(( iTemp * iTemp ) >> uiShift); iTemp = piOrg[3] - piCur[3]; uiSum += Distortion(( iTemp * iTemp ) >> uiShift); iTemp = piOrg[4] - piCur[4]; uiSum += Distortion(( iTemp * iTemp ) >> uiShift); iTemp = piOrg[5] - piCur[5]; uiSum += Distortion(( iTemp * iTemp ) >> uiShift); iTemp = piOrg[6] - piCur[6]; uiSum += Distortion(( iTemp * iTemp ) >> uiShift); iTemp = piOrg[7] - piCur[7]; uiSum += Distortion(( iTemp * iTemp ) >> uiShift); piOrg += iStrideOrg; piCur += iStrideCur; } return ( uiSum ); } Distortion TComRdCost::xGetSSE16( DistParam* pcDtParam ) { if ( pcDtParam->bApplyWeight ) { assert( pcDtParam->iCols == 16 ); return TComRdCostWeightPrediction::xGetSSEw( pcDtParam ); } const Pel* piOrg = pcDtParam->pOrg; const Pel* piCur = pcDtParam->pCur; Int iRows = pcDtParam->iRows; Int iStrideOrg = pcDtParam->iStrideOrg; Int iStrideCur = pcDtParam->iStrideCur; Distortion uiSum = 0; UInt uiShift = DISTORTION_PRECISION_ADJUSTMENT((pcDtParam->bitDepth-8) << 1); Intermediate_Int iTemp; for( ; iRows != 0; iRows-- ) { iTemp = piOrg[ 0] - piCur[ 0]; uiSum += Distortion(( iTemp * iTemp ) >> uiShift); iTemp = piOrg[ 1] - piCur[ 1]; uiSum += Distortion(( iTemp * iTemp ) >> uiShift); iTemp = piOrg[ 2] - piCur[ 2]; uiSum += Distortion(( iTemp * iTemp ) >> uiShift); iTemp = piOrg[ 3] - piCur[ 3]; uiSum += Distortion(( iTemp * iTemp ) >> uiShift); iTemp = piOrg[ 4] - piCur[ 4]; uiSum += Distortion(( iTemp * iTemp ) >> uiShift); iTemp = piOrg[ 5] - piCur[ 5]; uiSum += Distortion(( iTemp * iTemp ) >> uiShift); iTemp = piOrg[ 6] - piCur[ 6]; uiSum += Distortion(( iTemp * iTemp ) >> uiShift); iTemp = piOrg[ 7] - piCur[ 7]; uiSum += Distortion(( iTemp * iTemp ) >> uiShift); iTemp = piOrg[ 8] - piCur[ 8]; uiSum += Distortion(( iTemp * iTemp ) >> uiShift); iTemp = piOrg[ 9] - piCur[ 9]; uiSum += Distortion(( iTemp * iTemp ) >> uiShift); iTemp = piOrg[10] - piCur[10]; uiSum += Distortion(( iTemp * iTemp ) >> uiShift); iTemp = piOrg[11] - piCur[11]; uiSum += Distortion(( iTemp * iTemp ) >> uiShift); iTemp = piOrg[12] - piCur[12]; uiSum += Distortion(( iTemp * iTemp ) >> uiShift); iTemp = piOrg[13] - piCur[13]; uiSum += Distortion(( iTemp * iTemp ) >> uiShift); iTemp = piOrg[14] - piCur[14]; uiSum += Distortion(( iTemp * iTemp ) >> uiShift); iTemp = piOrg[15] - piCur[15]; uiSum += Distortion(( iTemp * iTemp ) >> uiShift); piOrg += iStrideOrg; piCur += iStrideCur; } return ( uiSum ); } Distortion TComRdCost::xGetSSE16N( DistParam* pcDtParam ) { if ( pcDtParam->bApplyWeight ) { return TComRdCostWeightPrediction::xGetSSEw( pcDtParam ); } const Pel* piOrg = pcDtParam->pOrg; const Pel* piCur = pcDtParam->pCur; Int iRows = pcDtParam->iRows; Int iCols = pcDtParam->iCols; Int iStrideOrg = pcDtParam->iStrideOrg; Int iStrideCur = pcDtParam->iStrideCur; Distortion uiSum = 0; UInt uiShift = DISTORTION_PRECISION_ADJUSTMENT((pcDtParam->bitDepth-8) << 1); Intermediate_Int iTemp; for( ; iRows != 0; iRows-- ) { for (Int n = 0; n < iCols; n+=16 ) { iTemp = piOrg[n+ 0] - piCur[n+ 0]; uiSum += Distortion(( iTemp * iTemp ) >> uiShift); iTemp = piOrg[n+ 1] - piCur[n+ 1]; uiSum += Distortion(( iTemp * iTemp ) >> uiShift); iTemp = piOrg[n+ 2] - piCur[n+ 2]; uiSum += Distortion(( iTemp * iTemp ) >> uiShift); iTemp = piOrg[n+ 3] - piCur[n+ 3]; uiSum += Distortion(( iTemp * iTemp ) >> uiShift); iTemp = piOrg[n+ 4] - piCur[n+ 4]; uiSum += Distortion(( iTemp * iTemp ) >> uiShift); iTemp = piOrg[n+ 5] - piCur[n+ 5]; uiSum += Distortion(( iTemp * iTemp ) >> uiShift); iTemp = piOrg[n+ 6] - piCur[n+ 6]; uiSum += Distortion(( iTemp * iTemp ) >> uiShift); iTemp = piOrg[n+ 7] - piCur[n+ 7]; uiSum += Distortion(( iTemp * iTemp ) >> uiShift); iTemp = piOrg[n+ 8] - piCur[n+ 8]; uiSum += Distortion(( iTemp * iTemp ) >> uiShift); iTemp = piOrg[n+ 9] - piCur[n+ 9]; uiSum += Distortion(( iTemp * iTemp ) >> uiShift); iTemp = piOrg[n+10] - piCur[n+10]; uiSum += Distortion(( iTemp * iTemp ) >> uiShift); iTemp = piOrg[n+11] - piCur[n+11]; uiSum += Distortion(( iTemp * iTemp ) >> uiShift); iTemp = piOrg[n+12] - piCur[n+12]; uiSum += Distortion(( iTemp * iTemp ) >> uiShift); iTemp = piOrg[n+13] - piCur[n+13]; uiSum += Distortion(( iTemp * iTemp ) >> uiShift); iTemp = piOrg[n+14] - piCur[n+14]; uiSum += Distortion(( iTemp * iTemp ) >> uiShift); iTemp = piOrg[n+15] - piCur[n+15]; uiSum += Distortion(( iTemp * iTemp ) >> uiShift); } piOrg += iStrideOrg; piCur += iStrideCur; } return ( uiSum ); } Distortion TComRdCost::xGetSSE32( DistParam* pcDtParam ) { if ( pcDtParam->bApplyWeight ) { assert( pcDtParam->iCols == 32 ); return TComRdCostWeightPrediction::xGetSSEw( pcDtParam ); } const Pel* piOrg = pcDtParam->pOrg; const Pel* piCur = pcDtParam->pCur; Int iRows = pcDtParam->iRows; Int iStrideOrg = pcDtParam->iStrideOrg; Int iStrideCur = pcDtParam->iStrideCur; Distortion uiSum = 0; UInt uiShift = DISTORTION_PRECISION_ADJUSTMENT((pcDtParam->bitDepth-8) << 1); Intermediate_Int iTemp; for( ; iRows != 0; iRows-- ) { iTemp = piOrg[ 0] - piCur[ 0]; uiSum += Distortion(( iTemp * iTemp ) >> uiShift); iTemp = piOrg[ 1] - piCur[ 1]; uiSum += Distortion(( iTemp * iTemp ) >> uiShift); iTemp = piOrg[ 2] - piCur[ 2]; uiSum += Distortion(( iTemp * iTemp ) >> uiShift); iTemp = piOrg[ 3] - piCur[ 3]; uiSum += Distortion(( iTemp * iTemp ) >> uiShift); iTemp = piOrg[ 4] - piCur[ 4]; uiSum += Distortion(( iTemp * iTemp ) >> uiShift); iTemp = piOrg[ 5] - piCur[ 5]; uiSum += Distortion(( iTemp * iTemp ) >> uiShift); iTemp = piOrg[ 6] - piCur[ 6]; uiSum += Distortion(( iTemp * iTemp ) >> uiShift); iTemp = piOrg[ 7] - piCur[ 7]; uiSum += Distortion(( iTemp * iTemp ) >> uiShift); iTemp = piOrg[ 8] - piCur[ 8]; uiSum += Distortion(( iTemp * iTemp ) >> uiShift); iTemp = piOrg[ 9] - piCur[ 9]; uiSum += Distortion(( iTemp * iTemp ) >> uiShift); iTemp = piOrg[10] - piCur[10]; uiSum += Distortion(( iTemp * iTemp ) >> uiShift); iTemp = piOrg[11] - piCur[11]; uiSum += Distortion(( iTemp * iTemp ) >> uiShift); iTemp = piOrg[12] - piCur[12]; uiSum += Distortion(( iTemp * iTemp ) >> uiShift); iTemp = piOrg[13] - piCur[13]; uiSum += Distortion(( iTemp * iTemp ) >> uiShift); iTemp = piOrg[14] - piCur[14]; uiSum += Distortion(( iTemp * iTemp ) >> uiShift); iTemp = piOrg[15] - piCur[15]; uiSum += Distortion(( iTemp * iTemp ) >> uiShift); iTemp = piOrg[16] - piCur[16]; uiSum += Distortion(( iTemp * iTemp ) >> uiShift); iTemp = piOrg[17] - piCur[17]; uiSum += Distortion(( iTemp * iTemp ) >> uiShift); iTemp = piOrg[18] - piCur[18]; uiSum += Distortion(( iTemp * iTemp ) >> uiShift); iTemp = piOrg[19] - piCur[19]; uiSum += Distortion(( iTemp * iTemp ) >> uiShift); iTemp = piOrg[20] - piCur[20]; uiSum += Distortion(( iTemp * iTemp ) >> uiShift); iTemp = piOrg[21] - piCur[21]; uiSum += Distortion(( iTemp * iTemp ) >> uiShift); iTemp = piOrg[22] - piCur[22]; uiSum += Distortion(( iTemp * iTemp ) >> uiShift); iTemp = piOrg[23] - piCur[23]; uiSum += Distortion(( iTemp * iTemp ) >> uiShift); iTemp = piOrg[24] - piCur[24]; uiSum += Distortion(( iTemp * iTemp ) >> uiShift); iTemp = piOrg[25] - piCur[25]; uiSum += Distortion(( iTemp * iTemp ) >> uiShift); iTemp = piOrg[26] - piCur[26]; uiSum += Distortion(( iTemp * iTemp ) >> uiShift); iTemp = piOrg[27] - piCur[27]; uiSum += Distortion(( iTemp * iTemp ) >> uiShift); iTemp = piOrg[28] - piCur[28]; uiSum += Distortion(( iTemp * iTemp ) >> uiShift); iTemp = piOrg[29] - piCur[29]; uiSum += Distortion(( iTemp * iTemp ) >> uiShift); iTemp = piOrg[30] - piCur[30]; uiSum += Distortion(( iTemp * iTemp ) >> uiShift); iTemp = piOrg[31] - piCur[31]; uiSum += Distortion(( iTemp * iTemp ) >> uiShift); piOrg += iStrideOrg; piCur += iStrideCur; } return ( uiSum ); } Distortion TComRdCost::xGetSSE64( DistParam* pcDtParam ) { if ( pcDtParam->bApplyWeight ) { assert( pcDtParam->iCols == 64 ); return TComRdCostWeightPrediction::xGetSSEw( pcDtParam ); } const Pel* piOrg = pcDtParam->pOrg; const Pel* piCur = pcDtParam->pCur; Int iRows = pcDtParam->iRows; Int iStrideOrg = pcDtParam->iStrideOrg; Int iStrideCur = pcDtParam->iStrideCur; Distortion uiSum = 0; UInt uiShift = DISTORTION_PRECISION_ADJUSTMENT((pcDtParam->bitDepth-8) << 1); Intermediate_Int iTemp; for( ; iRows != 0; iRows-- ) { iTemp = piOrg[ 0] - piCur[ 0]; uiSum += Distortion(( iTemp * iTemp ) >> uiShift); iTemp = piOrg[ 1] - piCur[ 1]; uiSum += Distortion(( iTemp * iTemp ) >> uiShift); iTemp = piOrg[ 2] - piCur[ 2]; uiSum += Distortion(( iTemp * iTemp ) >> uiShift); iTemp = piOrg[ 3] - piCur[ 3]; uiSum += Distortion(( iTemp * iTemp ) >> uiShift); iTemp = piOrg[ 4] - piCur[ 4]; uiSum += Distortion(( iTemp * iTemp ) >> uiShift); iTemp = piOrg[ 5] - piCur[ 5]; uiSum += Distortion(( iTemp * iTemp ) >> uiShift); iTemp = piOrg[ 6] - piCur[ 6]; uiSum += Distortion(( iTemp * iTemp ) >> uiShift); iTemp = piOrg[ 7] - piCur[ 7]; uiSum += Distortion(( iTemp * iTemp ) >> uiShift); iTemp = piOrg[ 8] - piCur[ 8]; uiSum += Distortion(( iTemp * iTemp ) >> uiShift); iTemp = piOrg[ 9] - piCur[ 9]; uiSum += Distortion(( iTemp * iTemp ) >> uiShift); iTemp = piOrg[10] - piCur[10]; uiSum += Distortion(( iTemp * iTemp ) >> uiShift); iTemp = piOrg[11] - piCur[11]; uiSum += Distortion(( iTemp * iTemp ) >> uiShift); iTemp = piOrg[12] - piCur[12]; uiSum += Distortion(( iTemp * iTemp ) >> uiShift); iTemp = piOrg[13] - piCur[13]; uiSum += Distortion(( iTemp * iTemp ) >> uiShift); iTemp = piOrg[14] - piCur[14]; uiSum += Distortion(( iTemp * iTemp ) >> uiShift); iTemp = piOrg[15] - piCur[15]; uiSum += Distortion(( iTemp * iTemp ) >> uiShift); iTemp = piOrg[16] - piCur[16]; uiSum += Distortion(( iTemp * iTemp ) >> uiShift); iTemp = piOrg[17] - piCur[17]; uiSum += Distortion(( iTemp * iTemp ) >> uiShift); iTemp = piOrg[18] - piCur[18]; uiSum += Distortion(( iTemp * iTemp ) >> uiShift); iTemp = piOrg[19] - piCur[19]; uiSum += Distortion(( iTemp * iTemp ) >> uiShift); iTemp = piOrg[20] - piCur[20]; uiSum += Distortion(( iTemp * iTemp ) >> uiShift); iTemp = piOrg[21] - piCur[21]; uiSum += Distortion(( iTemp * iTemp ) >> uiShift); iTemp = piOrg[22] - piCur[22]; uiSum += Distortion(( iTemp * iTemp ) >> uiShift); iTemp = piOrg[23] - piCur[23]; uiSum += Distortion(( iTemp * iTemp ) >> uiShift); iTemp = piOrg[24] - piCur[24]; uiSum += Distortion(( iTemp * iTemp ) >> uiShift); iTemp = piOrg[25] - piCur[25]; uiSum += Distortion(( iTemp * iTemp ) >> uiShift); iTemp = piOrg[26] - piCur[26]; uiSum += Distortion(( iTemp * iTemp ) >> uiShift); iTemp = piOrg[27] - piCur[27]; uiSum += Distortion(( iTemp * iTemp ) >> uiShift); iTemp = piOrg[28] - piCur[28]; uiSum += Distortion(( iTemp * iTemp ) >> uiShift); iTemp = piOrg[29] - piCur[29]; uiSum += Distortion(( iTemp * iTemp ) >> uiShift); iTemp = piOrg[30] - piCur[30]; uiSum += Distortion(( iTemp * iTemp ) >> uiShift); iTemp = piOrg[31] - piCur[31]; uiSum += Distortion(( iTemp * iTemp ) >> uiShift); iTemp = piOrg[32] - piCur[32]; uiSum += Distortion(( iTemp * iTemp ) >> uiShift); iTemp = piOrg[33] - piCur[33]; uiSum += Distortion(( iTemp * iTemp ) >> uiShift); iTemp = piOrg[34] - piCur[34]; uiSum += Distortion(( iTemp * iTemp ) >> uiShift); iTemp = piOrg[35] - piCur[35]; uiSum += Distortion(( iTemp * iTemp ) >> uiShift); iTemp = piOrg[36] - piCur[36]; uiSum += Distortion(( iTemp * iTemp ) >> uiShift); iTemp = piOrg[37] - piCur[37]; uiSum += Distortion(( iTemp * iTemp ) >> uiShift); iTemp = piOrg[38] - piCur[38]; uiSum += Distortion(( iTemp * iTemp ) >> uiShift); iTemp = piOrg[39] - piCur[39]; uiSum += Distortion(( iTemp * iTemp ) >> uiShift); iTemp = piOrg[40] - piCur[40]; uiSum += Distortion(( iTemp * iTemp ) >> uiShift); iTemp = piOrg[41] - piCur[41]; uiSum += Distortion(( iTemp * iTemp ) >> uiShift); iTemp = piOrg[42] - piCur[42]; uiSum += Distortion(( iTemp * iTemp ) >> uiShift); iTemp = piOrg[43] - piCur[43]; uiSum += Distortion(( iTemp * iTemp ) >> uiShift); iTemp = piOrg[44] - piCur[44]; uiSum += Distortion(( iTemp * iTemp ) >> uiShift); iTemp = piOrg[45] - piCur[45]; uiSum += Distortion(( iTemp * iTemp ) >> uiShift); iTemp = piOrg[46] - piCur[46]; uiSum += Distortion(( iTemp * iTemp ) >> uiShift); iTemp = piOrg[47] - piCur[47]; uiSum += Distortion(( iTemp * iTemp ) >> uiShift); iTemp = piOrg[48] - piCur[48]; uiSum += Distortion(( iTemp * iTemp ) >> uiShift); iTemp = piOrg[49] - piCur[49]; uiSum += Distortion(( iTemp * iTemp ) >> uiShift); iTemp = piOrg[50] - piCur[50]; uiSum += Distortion(( iTemp * iTemp ) >> uiShift); iTemp = piOrg[51] - piCur[51]; uiSum += Distortion(( iTemp * iTemp ) >> uiShift); iTemp = piOrg[52] - piCur[52]; uiSum += Distortion(( iTemp * iTemp ) >> uiShift); iTemp = piOrg[53] - piCur[53]; uiSum += Distortion(( iTemp * iTemp ) >> uiShift); iTemp = piOrg[54] - piCur[54]; uiSum += Distortion(( iTemp * iTemp ) >> uiShift); iTemp = piOrg[55] - piCur[55]; uiSum += Distortion(( iTemp * iTemp ) >> uiShift); iTemp = piOrg[56] - piCur[56]; uiSum += Distortion(( iTemp * iTemp ) >> uiShift); iTemp = piOrg[57] - piCur[57]; uiSum += Distortion(( iTemp * iTemp ) >> uiShift); iTemp = piOrg[58] - piCur[58]; uiSum += Distortion(( iTemp * iTemp ) >> uiShift); iTemp = piOrg[59] - piCur[59]; uiSum += Distortion(( iTemp * iTemp ) >> uiShift); iTemp = piOrg[60] - piCur[60]; uiSum += Distortion(( iTemp * iTemp ) >> uiShift); iTemp = piOrg[61] - piCur[61]; uiSum += Distortion(( iTemp * iTemp ) >> uiShift); iTemp = piOrg[62] - piCur[62]; uiSum += Distortion(( iTemp * iTemp ) >> uiShift); iTemp = piOrg[63] - piCur[63]; uiSum += Distortion(( iTemp * iTemp ) >> uiShift); piOrg += iStrideOrg; piCur += iStrideCur; } return ( uiSum ); } // -------------------------------------------------------------------------------------------------------------------- // HADAMARD with step (used in fractional search) // -------------------------------------------------------------------------------------------------------------------- Distortion TComRdCost::xCalcHADs2x2( Pel *piOrg, Pel *piCur, Int iStrideOrg, Int iStrideCur, Int iStep ) { Distortion satd = 0; TCoeff diff[4], m[4]; assert( iStep == 1 ); diff[0] = piOrg[0 ] - piCur[0]; diff[1] = piOrg[1 ] - piCur[1]; diff[2] = piOrg[iStrideOrg ] - piCur[0 + iStrideCur]; diff[3] = piOrg[iStrideOrg + 1] - piCur[1 + iStrideCur]; m[0] = diff[0] + diff[2]; m[1] = diff[1] + diff[3]; m[2] = diff[0] - diff[2]; m[3] = diff[1] - diff[3]; satd += abs(m[0] + m[1]); satd += abs(m[0] - m[1]); satd += abs(m[2] + m[3]); satd += abs(m[2] - m[3]); return satd; } Distortion TComRdCost::xCalcHADs4x4( Pel *piOrg, Pel *piCur, Int iStrideOrg, Int iStrideCur, Int iStep ) { Int k; Distortion satd = 0; TCoeff diff[16], m[16], d[16]; assert( iStep == 1 ); for( k = 0; k < 16; k+=4 ) { diff[k+0] = piOrg[0] - piCur[0]; diff[k+1] = piOrg[1] - piCur[1]; diff[k+2] = piOrg[2] - piCur[2]; diff[k+3] = piOrg[3] - piCur[3]; piCur += iStrideCur; piOrg += iStrideOrg; } /*===== hadamard transform =====*/ m[ 0] = diff[ 0] + diff[12]; m[ 1] = diff[ 1] + diff[13]; m[ 2] = diff[ 2] + diff[14]; m[ 3] = diff[ 3] + diff[15]; m[ 4] = diff[ 4] + diff[ 8]; m[ 5] = diff[ 5] + diff[ 9]; m[ 6] = diff[ 6] + diff[10]; m[ 7] = diff[ 7] + diff[11]; m[ 8] = diff[ 4] - diff[ 8]; m[ 9] = diff[ 5] - diff[ 9]; m[10] = diff[ 6] - diff[10]; m[11] = diff[ 7] - diff[11]; m[12] = diff[ 0] - diff[12]; m[13] = diff[ 1] - diff[13]; m[14] = diff[ 2] - diff[14]; m[15] = diff[ 3] - diff[15]; d[ 0] = m[ 0] + m[ 4]; d[ 1] = m[ 1] + m[ 5]; d[ 2] = m[ 2] + m[ 6]; d[ 3] = m[ 3] + m[ 7]; d[ 4] = m[ 8] + m[12]; d[ 5] = m[ 9] + m[13]; d[ 6] = m[10] + m[14]; d[ 7] = m[11] + m[15]; d[ 8] = m[ 0] - m[ 4]; d[ 9] = m[ 1] - m[ 5]; d[10] = m[ 2] - m[ 6]; d[11] = m[ 3] - m[ 7]; d[12] = m[12] - m[ 8]; d[13] = m[13] - m[ 9]; d[14] = m[14] - m[10]; d[15] = m[15] - m[11]; m[ 0] = d[ 0] + d[ 3]; m[ 1] = d[ 1] + d[ 2]; m[ 2] = d[ 1] - d[ 2]; m[ 3] = d[ 0] - d[ 3]; m[ 4] = d[ 4] + d[ 7]; m[ 5] = d[ 5] + d[ 6]; m[ 6] = d[ 5] - d[ 6]; m[ 7] = d[ 4] - d[ 7]; m[ 8] = d[ 8] + d[11]; m[ 9] = d[ 9] + d[10]; m[10] = d[ 9] - d[10]; m[11] = d[ 8] - d[11]; m[12] = d[12] + d[15]; m[13] = d[13] + d[14]; m[14] = d[13] - d[14]; m[15] = d[12] - d[15]; d[ 0] = m[ 0] + m[ 1]; d[ 1] = m[ 0] - m[ 1]; d[ 2] = m[ 2] + m[ 3]; d[ 3] = m[ 3] - m[ 2]; d[ 4] = m[ 4] + m[ 5]; d[ 5] = m[ 4] - m[ 5]; d[ 6] = m[ 6] + m[ 7]; d[ 7] = m[ 7] - m[ 6]; d[ 8] = m[ 8] + m[ 9]; d[ 9] = m[ 8] - m[ 9]; d[10] = m[10] + m[11]; d[11] = m[11] - m[10]; d[12] = m[12] + m[13]; d[13] = m[12] - m[13]; d[14] = m[14] + m[15]; d[15] = m[15] - m[14]; for (k=0; k<16; ++k) { satd += abs(d[k]); } satd = ((satd+1)>>1); return satd; } Distortion TComRdCost::xCalcHADs8x8( Pel *piOrg, Pel *piCur, Int iStrideOrg, Int iStrideCur, Int iStep ) { Int k, i, j, jj; Distortion sad = 0; TCoeff diff[64], m1[8][8], m2[8][8], m3[8][8]; assert( iStep == 1 ); for( k = 0; k < 64; k += 8 ) { diff[k+0] = piOrg[0] - piCur[0]; diff[k+1] = piOrg[1] - piCur[1]; diff[k+2] = piOrg[2] - piCur[2]; diff[k+3] = piOrg[3] - piCur[3]; diff[k+4] = piOrg[4] - piCur[4]; diff[k+5] = piOrg[5] - piCur[5]; diff[k+6] = piOrg[6] - piCur[6]; diff[k+7] = piOrg[7] - piCur[7]; piCur += iStrideCur; piOrg += iStrideOrg; } //horizontal for (j=0; j < 8; j++) { jj = j << 3; m2[j][0] = diff[jj ] + diff[jj+4]; m2[j][1] = diff[jj+1] + diff[jj+5]; m2[j][2] = diff[jj+2] + diff[jj+6]; m2[j][3] = diff[jj+3] + diff[jj+7]; m2[j][4] = diff[jj ] - diff[jj+4]; m2[j][5] = diff[jj+1] - diff[jj+5]; m2[j][6] = diff[jj+2] - diff[jj+6]; m2[j][7] = diff[jj+3] - diff[jj+7]; m1[j][0] = m2[j][0] + m2[j][2]; m1[j][1] = m2[j][1] + m2[j][3]; m1[j][2] = m2[j][0] - m2[j][2]; m1[j][3] = m2[j][1] - m2[j][3]; m1[j][4] = m2[j][4] + m2[j][6]; m1[j][5] = m2[j][5] + m2[j][7]; m1[j][6] = m2[j][4] - m2[j][6]; m1[j][7] = m2[j][5] - m2[j][7]; m2[j][0] = m1[j][0] + m1[j][1]; m2[j][1] = m1[j][0] - m1[j][1]; m2[j][2] = m1[j][2] + m1[j][3]; m2[j][3] = m1[j][2] - m1[j][3]; m2[j][4] = m1[j][4] + m1[j][5]; m2[j][5] = m1[j][4] - m1[j][5]; m2[j][6] = m1[j][6] + m1[j][7]; m2[j][7] = m1[j][6] - m1[j][7]; } //vertical for (i=0; i < 8; i++) { m3[0][i] = m2[0][i] + m2[4][i]; m3[1][i] = m2[1][i] + m2[5][i]; m3[2][i] = m2[2][i] + m2[6][i]; m3[3][i] = m2[3][i] + m2[7][i]; m3[4][i] = m2[0][i] - m2[4][i]; m3[5][i] = m2[1][i] - m2[5][i]; m3[6][i] = m2[2][i] - m2[6][i]; m3[7][i] = m2[3][i] - m2[7][i]; m1[0][i] = m3[0][i] + m3[2][i]; m1[1][i] = m3[1][i] + m3[3][i]; m1[2][i] = m3[0][i] - m3[2][i]; m1[3][i] = m3[1][i] - m3[3][i]; m1[4][i] = m3[4][i] + m3[6][i]; m1[5][i] = m3[5][i] + m3[7][i]; m1[6][i] = m3[4][i] - m3[6][i]; m1[7][i] = m3[5][i] - m3[7][i]; m2[0][i] = m1[0][i] + m1[1][i]; m2[1][i] = m1[0][i] - m1[1][i]; m2[2][i] = m1[2][i] + m1[3][i]; m2[3][i] = m1[2][i] - m1[3][i]; m2[4][i] = m1[4][i] + m1[5][i]; m2[5][i] = m1[4][i] - m1[5][i]; m2[6][i] = m1[6][i] + m1[7][i]; m2[7][i] = m1[6][i] - m1[7][i]; } for (i = 0; i < 8; i++) { for (j = 0; j < 8; j++) { sad += abs(m2[i][j]); } } sad=((sad+2)>>2); return sad; } Distortion TComRdCost::xGetHADs( DistParam* pcDtParam ) { if ( pcDtParam->bApplyWeight ) { return TComRdCostWeightPrediction::xGetHADsw( pcDtParam ); } Pel* piOrg = pcDtParam->pOrg; Pel* piCur = pcDtParam->pCur; Int iRows = pcDtParam->iRows; Int iCols = pcDtParam->iCols; Int iStrideCur = pcDtParam->iStrideCur; Int iStrideOrg = pcDtParam->iStrideOrg; Int iStep = pcDtParam->iStep; Int x, y; Distortion uiSum = 0; if( ( iRows % 8 == 0) && (iCols % 8 == 0) ) { Int iOffsetOrg = iStrideOrg<<3; Int iOffsetCur = iStrideCur<<3; for ( y=0; y> DISTORTION_PRECISION_ADJUSTMENT(pcDtParam->bitDepth-8) ); } //! \}