VVC代码学习1：xCompressCU函数

一、函数简述

xCompressCU这个函数通过递归的调用自己来完成对整个CTU的划分和每个CU的模式选择。与HEVC中该函数不同的是，VVC中加入了很多新的模式，比如帧间Affine模式，帧内IBC模式等。

亮度块最大允许尺寸为128×128，并且划分方式采用多类型树结构，CU可以进行四叉划分，水平二叉划分，垂直二叉划分，水平三叉划分，垂直三叉划分。最优划分模式的确定在xCheckModeSplit函数中，该函数会继续调用xCompressCU对划分的小CU块进行模式的选择和进一步划分。所以模式的选择和CU的划分会一直在这两个函数之间跳转。看的时候很容易懵逼= =！

二、函数流程

void EncCu::xCompressCU( CodingStructure *&tempCS, CodingStructure *&bestCS, Partitioner &partitioner )
{
  if (m_shareState == NO_SHARE)
  {
    tempCS->sharedBndPos = tempCS->area.Y().lumaPos();//CU块的位置信息
    tempCS->sharedBndSize.width = tempCS->area.lwidth();//CU的宽
    tempCS->sharedBndSize.height = tempCS->area.lheight();//CU的高
    bestCS->sharedBndPos = bestCS->area.Y().lumaPos();
    bestCS->sharedBndSize.width = bestCS->area.lwidth();
    bestCS->sharedBndSize.height = bestCS->area.lheight();
  }
#if ENABLE_SPLIT_PARALLELISM//并行划分
  CHECK( m_dataId != tempCS->picture->scheduler.getDataId(), "Working in the wrong dataId!" );

  if( m_pcEncCfg->getNumSplitThreads() != 1 && tempCS->picture->scheduler.getSplitJobId() == 0 )
  {
    if( m_modeCtrl->isParallelSplit( *tempCS, partitioner ) )
    {
      m_modeCtrl->setParallelSplit( true );
      xCompressCUParallel( tempCS, bestCS, partitioner );
      return;
    }
  }

#endif

  Slice&   slice      = *tempCS->slice;//slice  
  const PPS &pps      = *tempCS->pps;//图像参数集
  const SPS &sps      = *tempCS->sps;//序列参数集
  const uint32_t uiLPelX  = tempCS->area.Y().lumaPos().x;//获取当前块的左上x坐标
  const uint32_t uiTPelY  = tempCS->area.Y().lumaPos().y;//获取当前块的左上y坐标

  const UnitArea currCsArea = clipArea( CS::getArea( *bestCS, bestCS->area, partitioner.chType ), *tempCS->picture );//当前区域

  m_modeCtrl->initCULevel( partitioner, *tempCS );//CU的初始化，添加各种模式
  if( partitioner.currQtDepth == 0 && partitioner.currMtDepth == 0 && !tempCS->slice->isIntra() && ( sps.getUseSBT() || sps.getUseInterMTS() ) )
  {
    auto slsSbt = dynamic_cast<SaveLoadEncInfoSbt*>( m_modeCtrl );
    int maxSLSize = sps.getUseSBT() ? tempCS->slice->getSPS()->getMaxSbtSize() : MTS_INTER_MAX_CU_SIZE;
    slsSbt->resetSaveloadSbt( maxSLSize );
#if ENABLE_SPLIT_PARALLELISM
    CHECK( tempCS->picture->scheduler.getSplitJobId() != 0, "The SBT search reset need to happen in sequential region." );
    if (m_pcEncCfg->getNumSplitThreads() > 1)
    {
      for (int jId = 1; jId < NUM_RESERVERD_SPLIT_JOBS; jId++)
      {
        auto slsSbt = dynamic_cast<SaveLoadEncInfoSbt *>(m_pcEncLib->getCuEncoder(jId)->m_modeCtrl);
        slsSbt->resetSaveloadSbt(maxSLSize);
      }
    }
#endif
  }
  m_sbtCostSave[0] = m_sbtCostSave[1] = MAX_DOUBLE;

  m_CurrCtx->start = m_CABACEstimator->getCtx();

  m_cuChromaQpOffsetIdxPlus1 = 0;
  //色度QP调整
  if( slice.getUseChromaQpAdj() )
  {
    // TODO M0133 : double check encoder decisions with respect to chroma QG detection and actual encode
    int lgMinCuSize = sps.getLog2MinCodingBlockSize() +
      std::max<int>( 0, sps.getLog2DiffMaxMinCodingBlockSize() - int( pps.getPpsRangeExtension().getCuChromaQpOffsetSubdiv()/2 ) );
    m_cuChromaQpOffsetIdxPlus1 = ( ( uiLPelX >> lgMinCuSize ) + ( uiTPelY >> lgMinCuSize ) ) % ( pps.getPpsRangeExtension().getChromaQpOffsetListLen() + 1 );
  }

  if( !m_modeCtrl->anyMode() )//测试模式为空集则停止对该CU的划分和模式的选择
  {
    m_modeCtrl->finishCULevel( partitioner );
    return;
  }

  DTRACE_UPDATE( g_trace_ctx, std::make_pair( "cux", uiLPelX ) );
  DTRACE_UPDATE( g_trace_ctx, std::make_pair( "cuy", uiTPelY ) );
  DTRACE_UPDATE( g_trace_ctx, std::make_pair( "cuw", tempCS->area.lwidth() ) );
  DTRACE_UPDATE( g_trace_ctx, std::make_pair( "cuh", tempCS->area.lheight() ) );
  DTRACE( g_trace_ctx, D_COMMON, "@(%4d,%4d) [%2dx%2d]\n", tempCS->area.lx(), tempCS->area.ly(), tempCS->area.lwidth(), tempCS->area.lheight() );


  int startShareThisLevel = 0;
  m_pcInterSearch->resetSavedAffineMotion();
  //开始进行各种模式的检测，包括对CU的划分
  do
  {
    EncTestMode currTestMode = m_modeCtrl->currTestMode();//当前测试模式

    if (pps.getUseDQP() && CS::isDualITree(*tempCS) && isChroma(partitioner.chType))//如果使用DQP，亮度色度分离，是色度通道
    {
      const Position chromaCentral(tempCS->area.Cb().chromaPos().offset(tempCS->area.Cb().chromaSize().width >> 1, tempCS->area.Cb().chromaSize().height >> 1));//色度中心
      const Position lumaRefPos(chromaCentral.x << getComponentScaleX(COMPONENT_Cb, tempCS->area.chromaFormat), chromaCentral.y << getComponentScaleY(COMPONENT_Cb, tempCS->area.chromaFormat));
      const CodingStructure* baseCS = bestCS->picture->cs;
      const CodingUnit* colLumaCu = baseCS->getCU(lumaRefPos, CHANNEL_TYPE_LUMA);

      if (colLumaCu)
      {
        currTestMode.qp = colLumaCu->qp;
      }
    }

#if SHARP_LUMA_DELTA_QP || ENABLE_QPA_SUB_CTU
    if (partitioner.currQgEnable() && (
#if SHARP_LUMA_DELTA_QP
        (m_pcEncCfg->getLumaLevelToDeltaQPMapping().isEnabled()) ||
#endif
#if ENABLE_QPA_SUB_CTU
        (m_pcEncCfg->getUsePerceptQPA() && !m_pcEncCfg->getUseRateCtrl() && pps.getUseDQP())
#else
        false
#endif
      ))
    {
#if ENABLE_SPLIT_PARALLELISM
      CHECK( tempCS->picture->scheduler.getSplitJobId() > 0, "Changing lambda is only allowed in the master thread!" );
#endif
      if (currTestMode.qp >= 0)
      {
        updateLambda (&slice, currTestMode.qp, CS::isDualITree (*tempCS) || (partitioner.currDepth == 0));//更新lambda
      }
    }
#endif

    if( currTestMode.type == ETM_INTER_ME )//帧间运动估计
    {
      if( ( currTestMode.opts & ETO_IMV ) != 0 )
      {
        tempCS->bestCS = bestCS;
        xCheckRDCostInterIMV( tempCS, bestCS, partitioner, currTestMode );
        tempCS->bestCS = nullptr;
      }
      else
      {
        tempCS->bestCS = bestCS;
        xCheckRDCostInter( tempCS, bestCS, partitioner, currTestMode );
        tempCS->bestCS = nullptr;
      }

    }
    else if (currTestMode.type == ETM_HASH_INTER)//帧间哈希
    {
      xCheckRDCostHashInter( tempCS, bestCS, partitioner, currTestMode );
    }
    else if( currTestMode.type == ETM_AFFINE )//仿射模式
    {
      xCheckRDCostAffineMerge2Nx2N( tempCS, bestCS, partitioner, currTestMode );
    }
#if REUSE_CU_RESULTS
    else if( currTestMode.type == ETM_RECO_CACHED )
    {
      xReuseCachedResult( tempCS, bestCS, partitioner );
    }
#endif
    else if( currTestMode.type == ETM_MERGE_SKIP )//Merge模式
    {
      xCheckRDCostMerge2Nx2N( tempCS, bestCS, partitioner, currTestMode );
      CodingUnit* cu = bestCS->getCU(partitioner.chType);
      if (cu)
      cu->mmvdSkip = cu->skip == false ? false : cu->mmvdSkip;
    }
    else if( currTestMode.type == ETM_MERGE_TRIANGLE )
    {
      xCheckRDCostMergeTriangle2Nx2N( tempCS, bestCS, partitioner, currTestMode );
    }
    else if( currTestMode.type == ETM_INTRA )
    {
      xCheckRDCostIntra( tempCS, bestCS, partitioner, currTestMode );//帧内入口函数
    }
    else if( currTestMode.type == ETM_IPCM )
    {
      xCheckIntraPCM( tempCS, bestCS, partitioner, currTestMode );
    }
    else if (currTestMode.type == ETM_IBC)//IBC模式
    {
      xCheckRDCostIBCMode(tempCS, bestCS, partitioner, currTestMode);
    }
    else if (currTestMode.type == ETM_IBC_MERGE)//IBC合并模式
    {
      xCheckRDCostIBCModeMerge2Nx2N(tempCS, bestCS, partitioner, currTestMode);
    }
    else if( isModeSplit( currTestMode ) )
    {

      xCheckModeSplit( tempCS, bestCS, partitioner, currTestMode );//此函数完成CU的划分，并且内部会递归的调用xCompressCU
    }
    else
    {
      THROW( "Don't know how to handle mode: type = " << currTestMode.type << ", options = " << currTestMode.opts );
    }
  } while( m_modeCtrl->nextMode( *tempCS, partitioner ) );//检测下一个模式

  if(startShareThisLevel == 1)
  {
    m_shareState = NO_SHARE;
    m_pcInterSearch->setShareState(m_shareState);
    setShareStateDec(m_shareState);
  }

  //
  // Finishing CU
#if ENABLE_SPLIT_PARALLELISM
  if( bestCS->cus.empty() )
  {
    CHECK( bestCS->cost != MAX_DOUBLE, "Cost should be maximal if no encoding found" );
    CHECK( bestCS->picture->scheduler.getSplitJobId() == 0, "Should always get a result in serial case" );

    m_modeCtrl->finishCULevel( partitioner );
    return;
  }

#endif
  // set context states 设置上下文状态
  m_CABACEstimator->getCtx() = m_CurrCtx->best;

  // QP from last processed CU for further processing
  bestCS->prevQP[partitioner.chType] = bestCS->cus.back()->qp;
  if ((!slice.isIntra() || slice.getSPS()->getIBCFlag())
    && partitioner.chType == CHANNEL_TYPE_LUMA
    && bestCS->cus.size() == 1 && (bestCS->cus.back()->predMode == MODE_INTER || bestCS->cus.back()->predMode == MODE_IBC)
    && bestCS->area.Y() == (*bestCS->cus.back()).Y()
    )
  {
    const CodingUnit&     cu = *bestCS->cus.front();
    const PredictionUnit& pu = *cu.firstPU;

    if (!cu.affine && !cu.triangle)
    {
      MotionInfo mi = pu.getMotionInfo();
      mi.GBiIdx = (mi.interDir == 3) ? cu.GBiIdx : GBI_DEFAULT;
      cu.cs->addMiToLut(CU::isIBC(cu) ? cu.cs->motionLut.lutIbc : cu.cs->motionLut.lut, mi);
    }
  }
  bestCS->picture->getPredBuf(currCsArea).copyFrom(bestCS->getPredBuf(currCsArea));
  bestCS->picture->getRecoBuf( currCsArea ).copyFrom( bestCS->getRecoBuf( currCsArea ) );
  m_modeCtrl->finishCULevel( partitioner );

#if ENABLE_SPLIT_PARALLELISM
  if( tempCS->picture->scheduler.getSplitJobId() == 0 && m_pcEncCfg->getNumSplitThreads() != 1 )
  {
    tempCS->picture->finishParallelPart( currCsArea );
  }

#endif
  // Assert if Best prediction mode is NONE
  // Selected mode's RD-cost must be not MAX_DOUBLE.
  CHECK( bestCS->cus.empty()                                   , "No possible encoding found" );
  CHECK( bestCS->cus[0]->predMode == NUMBER_OF_PREDICTION_MODES, "No possible encoding found" );
  CHECK( bestCS->cost             == MAX_DOUBLE                , "No possible encoding found" );
}