fast_me.c

H.264视频编解码的标准测试模型

/*!
 ************************************************************************
 *
 * \file fast_me.c
 *
 * \brief
 *   Fast integer pel motion estimation and fractional pel motion estimation
 *   algorithms are described in this file.
 *   1. get_mem_FME() and free_mem_FME() are functions for allocation and release
 *      of memories about motion estimation
 *   2. FME_BlockMotionSearch() is the function for fast integer pel motion 
 *      estimation and fractional pel motion estimation
 *   3. DefineThreshold() defined thresholds for early termination
 * \author 
 *    Main contributors: (see contributors.h for copyright, address and affiliation details)
 *    - Zhibo Chen         <[e-mail]>
 *    - JianFeng Xu        <[e-mail]>  
 *    - Wenfang Fu         <[e-mail]>
 * \date    
 *    2003.8
 ************************************************************************
 */

#include <stdlib.h>
#include <string.h>

#include "global.h"

#include "memalloc.h"
#include "fast_me.h"
#include "refbuf.h"

#define Q_BITS          15

extern  unsigned int*   byte_abs;
extern  int*   mvbits;
extern  short*   spiral_search_x;
extern  short*   spiral_search_y;


static pel_t *(*get_line) (pel_t**, int, int, int, int);

static const int quant_coef[6][4][4] = {
  {{13107, 8066,13107, 8066},{ 8066, 5243, 8066, 5243},{13107, 8066,13107, 8066},{ 8066, 5243, 8066, 5243}},
  {{11916, 7490,11916, 7490},{ 7490, 4660, 7490, 4660},{11916, 7490,11916, 7490},{ 7490, 4660, 7490, 4660}},
  {{10082, 6554,10082, 6554},{ 6554, 4194, 6554, 4194},{10082, 6554,10082, 6554},{ 6554, 4194, 6554, 4194}},
  {{ 9362, 5825, 9362, 5825},{ 5825, 3647, 5825, 3647},{ 9362, 5825, 9362, 5825},{ 5825, 3647, 5825, 3647}},
  {{ 8192, 5243, 8192, 5243},{ 5243, 3355, 5243, 3355},{ 8192, 5243, 8192, 5243},{ 5243, 3355, 5243, 3355}},
  {{ 7282, 4559, 7282, 4559},{ 4559, 2893, 4559, 2893},{ 7282, 4559, 7282, 4559},{ 4559, 2893, 4559, 2893}}
};


void DefineThreshold()
{
  AlphaSec[1] = 0.01f;
  AlphaSec[2] = 0.01f;
  AlphaSec[3] = 0.01f;
  AlphaSec[4] = 0.02f;
  AlphaSec[5] = 0.03f;
  AlphaSec[6] = 0.03f;
  AlphaSec[7] = 0.04f;

  AlphaThird[1] = 0.06f;
  AlphaThird[2] = 0.07f;
  AlphaThird[3] = 0.07f;
  AlphaThird[4] = 0.08f;
  AlphaThird[5] = 0.12f;
  AlphaThird[6] = 0.11f;
  AlphaThird[7] = 0.15f;

  DefineThresholdMB();
  return;
}

void DefineThresholdMB()
{
  int gb_qp_per    = (input->qpN-MIN_QP)/6;
  int gb_qp_rem    = (input->qpN-MIN_QP)%6;
  
  int gb_q_bits    = Q_BITS+gb_qp_per;
  int gb_qp_const,Thresh4x4;

  float Quantize_step;

  gb_qp_const=(1<<gb_q_bits)/6;
  
  Thresh4x4 =   ((1<<gb_q_bits) - gb_qp_const)/quant_coef[gb_qp_rem][0][0];
  Quantize_step = Thresh4x4/(4*5.61f);
  Bsize[7]=(16*16)*Quantize_step;

  Bsize[6]=Bsize[7]*4;
  Bsize[5]=Bsize[7]*4;
  Bsize[4]=Bsize[5]*4;
  Bsize[3]=Bsize[4]*4;
  Bsize[2]=Bsize[4]*4;
  Bsize[1]=Bsize[2]*4;
}


int get_mem_FME()
{
  int memory_size = 0;
  if (NULL==(flag_intra = calloc ((img->width>>4)+1,sizeof(byte)))) no_mem_exit("get_mem_FME: flag_intra"); //fwf 20050330

  memory_size += get_mem2D(&McostState, 2*input->search_range+1, 2*input->search_range+1); 
  memory_size += get_mem4Dint(&(fastme_ref_cost), img->max_num_references, 9, 4, 4);
  memory_size += get_mem3Dint(&(fastme_l0_cost), 9, img->height/4, img->width/4);
  memory_size += get_mem3Dint(&(fastme_l1_cost), 9, img->height/4, img->width/4);
  memory_size += get_mem2D(&SearchState,7,7);
  
  return memory_size;
}


void free_mem_FME()
{
  free_mem2D(McostState);
  free_mem4Dint(fastme_ref_cost, img->max_num_references, 9);
  free_mem3Dint(fastme_l0_cost, 9);
  free_mem3Dint(fastme_l1_cost, 9);
  free_mem2D(SearchState);

  free (flag_intra);
}


int PartCalMad(pel_t *ref_pic,pel_t** orig_pic,pel_t *(*get_ref_line)(int, pel_t*, int, int, int, int), int blocksize_y,int blocksize_x, int blocksize_x4,int mcost,int min_mcost,int cand_x,int cand_y)
{
  int y,x4;
  int height=((img->MbaffFrameFlag)&&(img->mb_data[img->current_mb_nr].mb_field))?img->height/2:img->height;
  pel_t *orig_line, *ref_line;
  for (y=0; y<blocksize_y; y++)
    {
    ref_line  = get_ref_line (blocksize_x, ref_pic, cand_y+y, cand_x, height, img->width);//2004.3.3
    orig_line = orig_pic [y];
    
    for (x4=0; x4<blocksize_x4; x4++)
    {
      mcost += byte_abs[ *orig_line++ - *ref_line++ ];
      mcost += byte_abs[ *orig_line++ - *ref_line++ ];
      mcost += byte_abs[ *orig_line++ - *ref_line++ ];
      mcost += byte_abs[ *orig_line++ - *ref_line++ ];
    }
    if (mcost >= min_mcost)
    {
      break;
    }
    }
    return mcost;
}

/*!
 ************************************************************************
 * \brief
 *    FastIntegerPelBlockMotionSearch: fast pixel block motion search 
 *    this algrithm is called UMHexagonS(see JVT-D016),which includes 
 *    four steps with different kinds of search patterns
 * \par Input:
 * pel_t**   orig_pic,     // <--  original picture
 * int       ref,          // <--  reference frame (0... or -1 (backward))
 * int       pic_pix_x,    // <--  absolute x-coordinate of regarded AxB block
 * int       pic_pix_y,    // <--  absolute y-coordinate of regarded AxB block
 * int       blocktype,    // <--  block type (1-16x16 ... 7-4x4)
 * int       pred_mv_x,    // <--  motion vector predictor (x) in sub-pel units
 * int       pred_mv_y,    // <--  motion vector predictor (y) in sub-pel units
 * int*      mv_x,         //  --> motion vector (x) - in pel units
 * int*      mv_y,         //  --> motion vector (y) - in pel units
 * int       search_range, // <--  1-d search range in pel units                         
 * int       min_mcost,    // <--  minimum motion cost (cost for center or huge value)
 * int       lambda_factor // <--  lagrangian parameter for determining motion cost
 * \par
 * Three macro definitions defined in this program:
 * 1. EARLY_TERMINATION: early termination algrithm, refer to JVT-D016.doc
 * 2. SEARCH_ONE_PIXEL: search one pixel in search range
 * 3. SEARCH_ONE_PIXEL1(value_iAbort): search one pixel in search range,
 *                                 but give a parameter to show if mincost refeshed
 * \author
 *   Main contributors: (see contributors.h for copyright, address and affiliation details)
 *   - Zhibo Chen         <[e-mail]>
 *   - JianFeng Xu        <[e-mail]>  
 * \date   :
 *   2003.8
 ************************************************************************
 */
int                                     //  ==> minimum motion cost after search
FastIntegerPelBlockMotionSearch  (pel_t**   orig_pic,     // <--  not used
                                  short     ref,          // <--  reference frame (0... or -1 (backward))
                                  int       list,
                                  int       pic_pix_x,    // <--  absolute x-coordinate of regarded AxB block
                                  int       pic_pix_y,    // <--  absolute y-coordinate of regarded AxB block
                                  int       blocktype,    // <--  block type (1-16x16 ... 7-4x4)
                                  short     pred_mv_x,    // <--  motion vector predictor (x) in sub-pel units
                                  short     pred_mv_y,    // <--  motion vector predictor (y) in sub-pel units
                                  short*    mv_x,         //  --> motion vector (x) - in pel units
                                  short*    mv_y,         //  --> motion vector (y) - in pel units
                                  int       search_range, // <--  1-d search range in pel units                         
                                  int       min_mcost,    // <--  minimum motion cost (cost for center or huge value)
                                  int       lambda_factor)       // <--  lagrangian parameter for determining motion cost
{
  static int Diamond_x[4] = {-1, 0, 1, 0};
  static int Diamond_y[4] = {0, 1, 0, -1};
  static int Hexagon_x[6] = {2, 1, -1, -2, -1, 1};
  static int Hexagon_y[6] = {0, -2, -2, 0,  2, 2};
  static int Big_Hexagon_x[16] = {0,-2, -4,-4,-4, -4, -4, -2,  0,  2,  4,  4, 4, 4, 4, 2};
  static int Big_Hexagon_y[16] = {4, 3, 2,  1, 0, -1, -2, -3, -4, -3, -2, -1, 0, 1, 2, 3};

  int   pos, cand_x, cand_y,  mcost;
  pel_t *(*get_ref_line)(int, pel_t*, int, int, int, int);
  int   list_offset   = ((img->MbaffFrameFlag)&&(img->mb_data[img->current_mb_nr].mb_field))? img->current_mb_nr%2 ? 4 : 2 : 0;

  int   mvshift       = 2;                  // motion vector shift for getting sub-pel units
  int   blocksize_y   = input->blc_size[blocktype][1];            // vertical block size
  int   blocksize_x   = input->blc_size[blocktype][0];            // horizontal block size
  int   blocksize_x4  = blocksize_x >> 2;                         // horizontal block size in 4-pel units
  int   pred_x        = (pic_pix_x << mvshift) + pred_mv_x;       // predicted position x (in sub-pel units)
  int   pred_y        = (pic_pix_y << mvshift) + pred_mv_y;       // predicted position y (in sub-pel units)
  int   center_x      = pic_pix_x + *mv_x;                        // center position x (in pel units)
  int   center_y      = pic_pix_y + *mv_y;                        // center position y (in pel units)
  int   best_x = 0, best_y = 0;
  int   search_step,iYMinNow, iXMinNow;
  int   i,m; 
  int   iAbort;
  float betaSec,betaThird;
  int height=((img->MbaffFrameFlag)&&(img->mb_data[img->current_mb_nr].mb_field))?img->height/2:img->height;
   
  //===== Use weighted Reference for ME ====
  pel_t*  ref_pic;
  int  apply_weights = ( (active_pps->weighted_pred_flag && (img->type == P_SLICE || img->type == SP_SLICE)) ||
                         (active_pps->weighted_bipred_idc && (img->type == B_SLICE)));  

  if (apply_weights && input->UseWeightedReferenceME)
    ref_pic       = listX[list+list_offset][ref]->imgY_11_w;
  else
    ref_pic       = listX[list+list_offset][ref]->imgY_11;
   

  //===== set function for getting reference picture lines =====
  if ((center_x > search_range) && (center_x < img->width -1-search_range-blocksize_x) &&
    (center_y > search_range) && (center_y < height-1-search_range-blocksize_y)   )
  {
    get_ref_line = FastLineX;
  }
  else
  {
    get_ref_line = UMVLineX;
  }
  
  //////allocate memory for search state//////////////////////////
  memset(McostState[0],0,(2*input->search_range+1)*(2*input->search_range+1));
  
   ///////Threshold defined for early termination///////////////////  
  if(list==0 && ref>0) 
  {
    if(pred_SAD_ref!=0)
    {
      betaSec = Bsize[blocktype]/(pred_SAD_ref*pred_SAD_ref)-AlphaSec[blocktype];
      betaThird = Bsize[blocktype]/(pred_SAD_ref*pred_SAD_ref)-AlphaThird[blocktype];
    }
    else
    {
      betaSec = 0;
      betaThird = 0;
    }
  }
  else 
  {
    if(blocktype==1)
    {
      if(pred_SAD_space !=0)
      {
        betaSec = Bsize[blocktype]/(pred_SAD_space*pred_SAD_space)-AlphaSec[blocktype];
        betaThird = Bsize[blocktype]/(pred_SAD_space*pred_SAD_space)-AlphaThird[blocktype];
      }
      else
      {
        betaSec = 0;
        betaThird = 0;
      }
    }
    else
    {
      if(pred_SAD_uplayer !=0)
      {
        betaSec = Bsize[blocktype]/(pred_SAD_uplayer*pred_SAD_uplayer)-AlphaSec[blocktype];
        betaThird = Bsize[blocktype]/(pred_SAD_uplayer*pred_SAD_uplayer)-AlphaThird[blocktype];
      }
      else
      {
        betaSec = 0;
        betaThird = 0;
      }
    }
  }
  /*****************************/

  //check the center median predictor
  cand_x = center_x ;
  cand_y = center_y ;
  mcost = MV_COST (lambda_factor, mvshift, cand_x, cand_y, pred_x, pred_y);
  mcost = PartCalMad(ref_pic, orig_pic, get_ref_line,blocksize_y,blocksize_x,blocksize_x4,mcost,min_mcost,cand_x,cand_y);
  McostState[search_range][search_range] = 1;
  if (mcost < min_mcost)
  {
    min_mcost = mcost;