www.pudn.com > wm2.5.zip > rdopt.c
/*
***********************************************************************
* COPYRIGHT AND WARRANTY INFORMATION
*
* Copyright 2004, Advanced Audio Video Coding Standard, Part II
*
* DISCLAIMER OF WARRANTY
*
* These software programs are available to the users without any
* license fee or royalty on an "as is" basis. The AVS disclaims
* any and all warranties, whether express, implied, or statutory,
* including any implied warranties of merchantability or of fitness
* for a particular purpose. In no event shall the contributors or
* the AVS be liable for any incidental, punitive, or consequential
* damages of any kind whatsoever arising from the use of this program.
*
* This disclaimer of warranty extends to the user of this program
* and user's customers, employees, agents, transferees, successors,
* and assigns.
*
* The AVS does not represent or warrant that the program furnished
* hereunder are free of infringement of any third-party patents.
* Commercial implementations of AVS, including shareware, may be
* subject to royalty fees to patent holders. Information regarding
* the AVS patent policy is available from the AVS Web site at
* http://www.avs.org.cn
*
* THIS IS NOT A GRANT OF PATENT RIGHTS - SEE THE AVS PATENT POLICY.
************************************************************************
*/
#include
#include "global.h"
#include "interpred.h"
#include "intrapred.h"
#include "macroblock.h"
#include "block.h"
#include "rdopt.h"
#include "ratectl.h" // LX 0409
extern MacroblockHeader MbHeader; // LX 0409
/*
*************************************************************************
* Function:Save RD cost state
* Input:
* Output:
* Return:
* Attention:
*************************************************************************
*/
void Save_RD_State()
{
pgRDStream->bits_to_go = pgcurrBitStream->bits_to_go;
pgRDStream->byte_buf = pgcurrBitStream->byte_buf;
pgRDStream->byte_pos = pgcurrBitStream->byte_pos;
}
/*
*************************************************************************
* Function:Restore RD cost state
* Input:
* Output:
* Return:
* Attention:
*************************************************************************
*/
void Restore_RD_State()
{
pgcurrBitStream->bits_to_go = pgRDStream->bits_to_go;
pgcurrBitStream->byte_buf = pgRDStream->byte_buf;
pgcurrBitStream->byte_pos = pgRDStream->byte_pos;
}
/*
*************************************************************************
* Function:Calculate RD cost for a macroblock
* Input:
* Output:
* Return:
* Attention:
*************************************************************************
*/
double RD_Cost_Macrobock()
{
double distortion;
double rdcost;
int rate;
int i, j, uv;
int pic_x = pgMbData->mb_x<<4;
int pic_y = pgMbData->mb_y<<4;
int pic_c_x = pic_x>>1;
int pic_c_y = pic_y>>1;
write_one_macroblock();
rate = (pgcurrBitStream->byte_pos - pgRDStream->byte_pos)*8 + (pgRDStream->bits_to_go-pgcurrBitStream->bits_to_go);
distortion = 0;
for(j=0; j<16; j++)
for(i=0; i<16; i++)
distortion += quadarr[pgMbData->org_luma[j][i]-imgY[pic_y+j][pic_x+i]];
for(uv=0; uv<2; uv++)
for(j=0; j<8; j++)
for(i=0; i<8; i++)
distortion += quadarr[pgMbData->org_chroma[uv][j][i]-imgUV[uv][pic_c_y+j][pic_c_x+i]];
rdcost = distortion + pgMbData->lamda_mode * rate;
return rdcost;
}
/*
*************************************************************************
* Function:Calculate RD cost for Inter macroblock
* Input:
* Output:
* Return:
* Attention:
*************************************************************************
*/
int RD_Cost_Inter_Macrobock()
{
int blocktype;
int b8, b4;
int i, j, k;
int pic_block_x, pic_block_y;
int step_h0, step_v0;
int img_x = (pgMbData->mb_x<<4);
int img_y = (pgMbData->mb_y<<4);
int img_c_x = img_x>>1;
int img_c_y = img_y>>1;
double rdcost, intra_rdcost;
int min_cost = 1<<30;//WJP
int coeff_cost;
int best_mode=1;//WJP
int dummy;
int old_skip_counter;
int best_abt_size, abt_size_flag=0;
//cbzhu 12-15
#ifdef _ISOLATED_REGION_
int iregallow=1;
int m,n;
#endif
for(blocktype=1; blocktype<=4; blocktype++)
{
//cbzhu 12-15
#ifdef _ISOLATED_REGION_
iregallow=1;
if(input.IREGEnable)
{
if(pCurrFrmIREGMap[pgImage->mb_y][pgImage->mb_x] != 0)
{
for(m=0;m<4;m++)
{
for(n=0;n<4;n++)
{
if(iIREGReliableFrmFlag==1 || iIREGState==2*input.IREGRate || input.num_reference_frames==1)
if(typemap[0][blocktype-1][m][n]==0)
{
iregallow=0;
break;
}
else if(typemap[0][blocktype-1][m][n]==0 && typemap[1][blocktype-1][m][n]==0)
{
iregallow=0;
break;
}
}
if(iregallow==0)break;
}
}
if(iregallow==0)
{
rdcost=1<<30;
continue;
}
}
#endif //_ISOLATED_REGION_
Restore_Motion_Vector(blocktype);
//restore reference index
for(b8=0; b8<4; b8++)
for(b4=0; b4<4; b4++)
{
refFrArr[pgImage->block_y+((b8>>1)<<1) + (b4>>1)][pgImage->block_x +((b8 &1)<<1) + (b4 &1)] = sInter_Ref[blocktype][b8][b4];
}
//WJP 040824N
if(input.ABT_RDO)
abt_size_flag=!abt_size_flag; //qhg for abt
else if (blocktype<4)
{
abt_size_flag=1;
}
//WJP 040824N
pgMbData->best_mode = blocktype; // P8x8 used when blocktype == 4
pgMbHeader->mb_type = blocktype;
if(pgMbHeader->mb_type == 4) //qwang 2004-3-22
{
for(i=0; i<4; i++)
pgMbHeader->subblk_type[i] = pgMbData->best_mode_subblk[i];
abt_size_flag=0;
}
//--- Set the numbers of MV ---
pgMbHeader->mv_num = (pgMbData->best_mode == 3) ? 2 : pgMbData->best_mode;
//--- Update left ref ---
pgMbData->left_ref = 0;
//--- Update Intra line buffer ---
for (b8=0;b8<4;b8++)
for (b4=0;b4<4;b4++)
{
pic_block_x=(img_x/4 + 2*(b8%2)+(b4%2));
pic_block_y=(img_y/4 + 2*(b8/2)+(b4/2));
ipredmode[pic_block_x+1][img_y/4+1] = (active_pps->constrained_intra_pred_flag ? -1 : 10);//MZ
/* if(active_pps->constrained_intra_pred_flag)
c_ipredmode[img_x/16+1][img_y/16+1] = -1;//zhangnan
else
c_ipredmode[img_x/16+1][img_y/16+1] = 0; //MZ 04-03
*/
}
//--- Get RefIndex info --- //zhangnan
step_h0 = (csiInter_Block_Size[pgMbHeader->mb_type-1][0] >> 2);
step_v0 = (csiInter_Block_Size[pgMbHeader->mb_type-1][1] >> 2); //??? zhangnan
for (j=0; j<4; j+=step_v0)
for (i=0; i<4; i+=step_h0)
{
k=j+(i/2);
pgMbHeader->refIndex[k] = refFrArr[pgMbData->mb_y*4+j][pgMbData->mb_x*4+i];
}
//--- Get MVD info ---
for(b8=0; b8<4; b8++)
for(b4=0; b4<4; b4++)
{
pgMbHeader->mvd[b8][b4][0] = pgMbData->mvb[b8][b4][0] - pgMbData->pred_mv[b8][b4][0];
pgMbHeader->mvd[b8][b4][1] = pgMbData->mvb[b8][b4][1] - pgMbData->pred_mv[b8][b4][1];
}
for(b8=0; b8<4; b8++)
for(b4=0; b4<4; b4++)
{
tmp_mv[0][pgImage->block_y +((b8>>1)<<1) + (b4>>1)][pgImage->block_x +((b8 &1)<<1) + (b4 &1)+4] = pgMbData->mvb[b8][b4][0];
tmp_mv[1][pgImage->block_y +((b8>>1)<<1) + (b4>>1)][pgImage->block_x +((b8 &1)<<1) + (b4 &1)+4] = pgMbData->mvb[b8][b4][1];
}
//--- Reset cbp ---
pgMbHeader->cbp = 0;
// dongjie
for(i=0;i<6;i++)
pgMbHeader->cbp_4x4[i] = 0;
// end
Inter_ReLoad_Residual();
for (b8=0; b8<4; b8++) //zhangnan
{
coeff_cost = 0;
coeff_cost_4x4[0]=coeff_cost_4x4[1]=coeff_cost_4x4[2]=coeff_cost_4x4[3] = 0; // dongjie
//WJP 040903
if(pgMbHeader->mb_type==4 && pgMbHeader->subblk_type[b8]==1 && (!input.ABT_RDO) && input.ABT)
{
abt_size_flag=1;
}
//WJP END
if(abt_size_flag==0||!input.ABT) //qhg for abt
{
for (b4=0; b4<4; b4++)
{
Transform_B4(b8, b4);
// LX 0409
for( i = 0; i < BLOCK_SIZE; i++)
for( j = 0; j < BLOCK_SIZE; j++)
// block_x = 0 4 8 12
g_nbestDCTcoef[ b8*4 + i ][ b8*4 + j ]= MbHeader.pred_residual[b8][b4][j][i];
// LX 0409
Quant_B4(pgMbData->qp_mb, b8, b4, &coeff_cost);
}
// dongjie
if(pgMbHeader->cbp_4x4[b8]==0)
pgMbHeader->cbp &= (~(1<cbp |= (1<qp_mb, b8, &coeff_cost);
}
if(coeff_cost <= (b8<4 ? LUMACOEFFCOST:CHROMACOEFFCOST))
{
pgMbHeader->cbp &= (63 - (1<pred_residual[b8][b4][j][i] = 0;
}
// dongjie
else
{
for(b4=0;b4<4;b4++)
{
if(coeff_cost_4x4[b4] <= LUMACOEFFCOST4x4)
{
pgMbHeader->cbp_4x4[b8] &= (15 - (1<pred_residual[b8][b4][j][i] = 0;
}
}
}
// end
if(abt_size_flag==0||!input.ABT) //qhg for abt
{
for (b4=0; b4<4; b4++)
{
Dequant_B4(pgMbData->qp_mb, b8, b4);
Inv_Transform_B4(b8, b4);
}
}
else
{
abt_dequant_B8(pgMbData->qp_mb,b8);
abt_Itransform_B8(b8);
}
//WJP 040903
if(pgMbHeader->mb_type==4 && pgMbHeader->subblk_type[b8]==1 && (!input.ABT_RDO) && input.ABT)
{
abt_size_flag=0;
}
//WJP END
}
for(b8=4; b8<6; b8++)
{
coeff_cost = 0;
for(b4=0; b4<4; b4++)
Transform_B4(b8, b4);
//TranformChromaDC(b8);
for(b4=0; b4<4; b4++)
Quant_B4(pgMbData->qp_mb, b8, b4, &coeff_cost);
// dongjie
if(pgMbHeader->cbp_4x4[b8]==0)
pgMbHeader->cbp &= (~(1<cbp |= (1<cbp &= (63 - (1<pred_residual[b8][b4][j][i] = 0;
}
// dongjie
else
{
for(b4=0;b4<4;b4++)
{
if(coeff_cost_4x4[b4] <= CHROMACOEFFCOST4x4)
{
pgMbHeader->cbp_4x4[b8] &= (15 - (1<pred_residual[b8][b4][j][i] = 0;
}
}
}
// end
for(b4=0; b4<4; b4++)
Dequant_B4(pgMbData->qp_mb, b8, b4);
//InvTransformChromaDC(b8);
for (b4=0; b4<4; b4++)
Inv_Transform_B4(b8, b4);
}
for (b8=0; b8<6; b8++)
{
for (b4=0; b4<4; b4++)
{
//Inv_Transform_B4(b8, b4);
Recon_B4(b8, b4, 1);
if(b8<4)
{
for(j=0; j<4; j++)
for(i=0; i<4; i++)
{
imgY[(pgMbData->mb_y<<4) + (b8/2)*8+(b4/2)*4+j][(pgMbData->mb_x<<4) + (b8%2)*8+(b4%2)*4+i] = imgY1[j][i];
}
}
else
{
for(j=0; j<4; j++)
for(i=0; i<4; i++)
{
imgUV[b8-4][(pgMbData->mb_y<<3) + (b4/2)*4+j][(pgMbData->mb_x<<3) +(b4%2)*4+ i] = imgUV1[b8-4][j][i];
}
}
}
}
pgMbHeader->abt_block_size_flag=abt_size_flag;
Save_RD_State();
old_skip_counter = pgImage->cod_counter;
rdcost = RD_Cost_Macrobock();
pgImage->cod_counter = old_skip_counter;
Restore_RD_State();
if(rdcost < min_cost)
{
best_mode = blocktype;
min_cost = (int) rdcost;
best_abt_size=abt_size_flag;
}
if(input.ABT && input.ABT_RDO)//WJP 040824N
blocktype=blocktype-abt_size_flag; //qhg for abt
}
pgMbHeader->mb_type = I4MB;
pgMbHeader->cbp = 0;//WJP 040824
// dongjie
for(i=0;i<6;i++)
pgMbHeader->cbp_4x4[i] = 0;
// end
Intra_Luma_Prediction(&dummy);
Intra_Chroma_Prediction(&dummy);
Save_RD_State();
intra_rdcost = (int)(4.0*floor(6.0 * pgMbData->lamda_mode + 0.4999));
old_skip_counter = pgImage->cod_counter;
intra_rdcost += RD_Cost_Macrobock();
pgImage->cod_counter = old_skip_counter;
Restore_RD_State();
Get_Skip_RD_Cost_MB();
//cbzhu 12-15
#ifdef _ISOLATED_REGION_
if (input.IREGEnable)
{
if (pCurrFrmIREGMap[pgImage->mb_y][pgImage->mb_x] == 1 )
{
if (iIntegerPixFlag == 0 || iSubPixFlag1 == 1)
{
pgMbData->skip_cost = 1 << 30;
min_cost = 1 << 30;
}
}
if(typemap[0][0][0][0]==0)
{
pgMbData->skip_cost = 1 << 30;
}
if (pCurrFrmIREGMap[pgImage->mb_y][pgImage->mb_x] == 2)
{
pgMbData->skip_cost = 1 << 30;
min_cost = 1 << 30;
}
}//if (input->IREGEnable)
#endif // _ISOLATED_REGION_
if(pgMbData->skip_costskip_costmb_type = 0;
//--- Update Intra line buffer ---
for (b8=0;b8<4;b8++)
for (b4=0;b4<4;b4++)
{
pic_block_x=(img_x/4 + 2*(b8%2)+(b4%2));
pic_block_y=(img_y/4 + 2*(b8/2)+(b4/2));
ipredmode[pic_block_x+1][pic_block_y+1] = (active_pps->constrained_intra_pred_flag ? -1:10);//MZ
/* if(active_pps->constrained_intra_pred_flag)
c_ipredmode[img_x/16 +1][img_y/16+1] = -1;//zhangnan
else
c_ipredmode[img_x/16 +1][img_y/16+1] = 0; //MZ 04-03-05
*/
}
write_one_macroblock();
return 0;
}
else if(intra_rdcost>min_cost)
{
Restore_Motion_Vector(best_mode);
pgMbData->best_mode = best_mode;
pgMbHeader->abt_block_size_flag=best_abt_size; //qhg for abt
//restore reference index
for(b8=0; b8<4; b8++)
for(b4=0; b4<4; b4++)
{
refFrArr[pgImage->block_y+((b8>>1)<<1) + (b4>>1)][pgImage->block_x +((b8 &1)<<1) + (b4 &1)] = sInter_Ref[best_mode][b8][b4];
}
return 1;
}
else//INTRA-coded
{
pgMbHeader->mb_type = I4MB;
pgMbHeader->cbp = 0;//WJP 040824
// dongjie
for(i=0;i<6;i++)
pgMbHeader->cbp_4x4[i] = 0;
// end
Intra_Luma_Prediction(&dummy);
Intra_Chroma_Prediction(&dummy);
pgMbData->left_ref = -1;
for (j = 0; j < 4; j++)
for (i = 0; i < 4; i++)
{
tmp_mv[0][pgMbData->mb_y*4+j][pgMbData->mb_x*4+i+4] = 0;
tmp_mv[1][pgMbData->mb_y*4+j][pgMbData->mb_x*4+i+4] = 0;
refFrArr [pgMbData->mb_y*4+j][pgMbData->mb_x*4+i ] = -1;
}
write_one_macroblock();
#ifdef _ISOLATED_REGION_
if(input.IREGEnable)
{
intrambcount++;
}
#endif
return 0;
}
}
/*
*************************************************************************
* Function:Calculate SA(T)D
* Input:
* Output:
* Return:
* Attention:
*************************************************************************
*/
int SATD (int* diff)
{
int k, satd = 0, m[16], dd, *d=diff;
/*===== hadamard transform =====*/
m[ 0] = d[ 0] + d[12];
m[ 4] = d[ 4] + d[ 8];
m[ 8] = d[ 4] - d[ 8];
m[12] = d[ 0] - d[12];
m[ 1] = d[ 1] + d[13];
m[ 5] = d[ 5] + d[ 9];
m[ 9] = d[ 5] - d[ 9];
m[13] = d[ 1] - d[13];
m[ 2] = d[ 2] + d[14];
m[ 6] = d[ 6] + d[10];
m[10] = d[ 6] - d[10];
m[14] = d[ 2] - d[14];
m[ 3] = d[ 3] + d[15];
m[ 7] = d[ 7] + d[11];
m[11] = d[ 7] - d[11];
m[15] = d[ 3] - d[15];
d[ 0] = m[ 0] + m[ 4];
d[ 8] = m[ 0] - m[ 4];
d[ 4] = m[ 8] + m[12];
d[12] = m[12] - m[ 8];
d[ 1] = m[ 1] + m[ 5];
d[ 9] = m[ 1] - m[ 5];
d[ 5] = m[ 9] + m[13];
d[13] = m[13] - m[ 9];
d[ 2] = m[ 2] + m[ 6];
d[10] = m[ 2] - m[ 6];
d[ 6] = m[10] + m[14];
d[14] = m[14] - m[10];
d[ 3] = m[ 3] + m[ 7];
d[11] = m[ 3] - m[ 7];
d[ 7] = m[11] + m[15];
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];
/*===== sum up =====*/
for (dd=diff[k=0]; k<16; dd=diff[++k])
{
satd += (dd < 0 ? -dd : dd);
}
satd >>= 1;
return satd;
}