-
Ronald S. Bultje authored
I now see I didn't write a very long description, so let's do it here then. We took a pretty big quality hit (0.1-0.2%) from my recent fix of the inversion of arguments to vp8_cost_bit() in the RD reference frame costing. I looked into it and basically the costing prevented us from switching reference frames. This is of course silly, since each frame codes its own prob_intra_coded, so using last frame cost indications as a limiting factor can never be right. Here, I've rewritten that code to estimate costings based partially on statistics from progress on current frame encoding. Overall, this gives us a ~0.2%-0.3% improvement over what we had previously before my argument-inversion-fix, and thus about ~0.4% over current git (on derf-set), and a little more (0.5-1.0%) on HD/STD-HD/YT. Change-Id: I79ebd4ccec4d6edbf0e152d9590d103ba2747775
0b8a95a0
rdopt.c 137.28 KiB
/* * Copyright (c) 2010 The WebM project authors. All Rights Reserved. * * Use of this source code is governed by a BSD-style license * that can be found in the LICENSE file in the root of the source * tree. An additional intellectual property rights grant can be found * in the file PATENTS. All contributing project authors may * be found in the AUTHORS file in the root of the source tree. */#include <stdio.h>#include <math.h>#include <limits.h>#include <assert.h>#include "vp8/common/pragmas.h"#include "tokenize.h"#include "treewriter.h"#include "onyx_int.h"#include "modecosts.h"#include "encodeintra.h"#include "vp8/common/entropymode.h"#include "vp8/common/reconinter.h"#include "vp8/common/reconintra.h"#include "vp8/common/reconintra4x4.h"#include "vp8/common/findnearmv.h"#include "vp8/common/quant_common.h"#include "encodemb.h"#include "quantize.h"#include "vp8/common/idct.h"#include "vp8/common/g_common.h"#include "variance.h"#include "mcomp.h"#include "rdopt.h"#include "ratectrl.h"#include "vpx_mem/vpx_mem.h"#include "dct.h"#include "vp8/common/systemdependent.h"#include "vp8/common/seg_common.h"#include "vp8/common/pred_common.h"#if CONFIG_RUNTIME_CPU_DETECT#define IF_RTCD(x) (x)#else#define IF_RTCD(x) NULL#endifextern void vp8cx_mb_init_quantizer(VP8_COMP *cpi, MACROBLOCK *x);extern void vp8_update_zbin_extra(VP8_COMP *cpi, MACROBLOCK *x);#if CONFIG_HIGH_PRECISION_MV#define XMVCOST (x->e_mbd.allow_high_precision_mv?x->mvcost_hp:x->mvcost)#else#define XMVCOST (x->mvcost)#endif#define MAXF(a,b) (((a) > (b)) ? (a) : (b))#define INVALID_MV 0x80008000static const int auto_speed_thresh[17] ={ 1000, 200, 150, 130, 150, 125,7172737475767778798081828384858687888990919293949596979899100101102103104105106107108109110111112113114115116117118119120121122123124125126127128129130131132133134135136137138139140
120,
115,
115,
115,
115,
115,
115,
115,
115,
115,
105
};
const MB_PREDICTION_MODE vp8_mode_order[MAX_MODES] =
{
ZEROMV,
DC_PRED,
NEARESTMV,
NEARMV,
ZEROMV,
NEARESTMV,
ZEROMV,
NEARESTMV,
NEARMV,
NEARMV,
V_PRED,
H_PRED,
#if CONFIG_NEWINTRAMODES
D45_PRED,
D135_PRED,
D117_PRED,
D153_PRED,
D27_PRED,
D63_PRED,
#endif
TM_PRED,
NEWMV,
NEWMV,
NEWMV,
SPLITMV,
SPLITMV,
SPLITMV,
B_PRED,
I8X8_PRED,
/* compound prediction modes */
ZEROMV,
NEARESTMV,
NEARMV,
ZEROMV,
NEARESTMV,
NEARMV,
ZEROMV,
NEARESTMV,
NEARMV,
NEWMV,
NEWMV,
NEWMV,
141142143144145146147148149150151152153154155156157158159160161162163164165166167168169170171172173174175176177178179180181182183184185186187188189190191192193194195196197198199200201202203204205206207208209210
SPLITMV,
SPLITMV,
SPLITMV,
};
const MV_REFERENCE_FRAME vp8_ref_frame_order[MAX_MODES] =
{
LAST_FRAME,
INTRA_FRAME,
LAST_FRAME,
LAST_FRAME,
GOLDEN_FRAME,
GOLDEN_FRAME,
ALTREF_FRAME,
ALTREF_FRAME,
GOLDEN_FRAME,
ALTREF_FRAME,
INTRA_FRAME,
INTRA_FRAME,
#if CONFIG_NEWINTRAMODES
INTRA_FRAME,
INTRA_FRAME,
INTRA_FRAME,
INTRA_FRAME,
INTRA_FRAME,
INTRA_FRAME,
#endif
INTRA_FRAME,
LAST_FRAME,
GOLDEN_FRAME,
ALTREF_FRAME,
LAST_FRAME,
GOLDEN_FRAME,
ALTREF_FRAME,
INTRA_FRAME,
INTRA_FRAME,
/* compound prediction modes */
LAST_FRAME,
LAST_FRAME,
LAST_FRAME,
ALTREF_FRAME,
ALTREF_FRAME,
ALTREF_FRAME,
GOLDEN_FRAME,
GOLDEN_FRAME,
GOLDEN_FRAME,
LAST_FRAME,
ALTREF_FRAME,
GOLDEN_FRAME,
LAST_FRAME,
ALTREF_FRAME,
GOLDEN_FRAME,
};
const MV_REFERENCE_FRAME vp8_second_ref_frame_order[MAX_MODES] =
{
0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0,
211212213214215216217218219220221222223224225226227228229230231232233234235236237238239240241242243244245246247248249250251252253254255256257258259260261262263264265266267268269270271272273274275276277278279280
#if CONFIG_NEWINTRAMODES
0, 0,
0, 0,
0, 0,
#endif
0, 0, 0, 0, 0, 0, 0, 0, 0,
/* compound prediction modes */
GOLDEN_FRAME,
GOLDEN_FRAME,
GOLDEN_FRAME,
LAST_FRAME,
LAST_FRAME,
LAST_FRAME,
ALTREF_FRAME,
ALTREF_FRAME,
ALTREF_FRAME,
GOLDEN_FRAME,
LAST_FRAME,
ALTREF_FRAME,
GOLDEN_FRAME,
LAST_FRAME,
ALTREF_FRAME,
};
static void fill_token_costs(
unsigned int (*c)[COEF_BANDS] [PREV_COEF_CONTEXTS] [MAX_ENTROPY_TOKENS],
const vp8_prob (*p)[COEF_BANDS] [PREV_COEF_CONTEXTS] [ENTROPY_NODES],
int block_type_counts)
{
int i, j, k;
for (i = 0; i < block_type_counts; i++)
for (j = 0; j < COEF_BANDS; j++)
for (k = 0; k < PREV_COEF_CONTEXTS; k++)
{
if(k == 0 && ((j > 0 && i > 0) || (j > 1 && i == 0)))
vp8_cost_tokens_skip((int *)( c [i][j][k]),
p [i][j][k],
vp8_coef_tree);
else
vp8_cost_tokens((int *)(c [i][j][k]),
p [i][j][k],
vp8_coef_tree);
}
}
static int rd_iifactor [ 32 ] = { 4, 4, 3, 2, 1, 0, 0, 0,
0, 0, 0, 0, 0, 0, 0, 0,
0, 0, 0, 0, 0, 0, 0, 0,
0, 0, 0, 0, 0, 0, 0, 0,
};
// 3* dc_qlookup[Q]*dc_qlookup[Q];
/* values are now correlated to quantizer */
static int sad_per_bit16lut[QINDEX_RANGE];
static int sad_per_bit4lut[QINDEX_RANGE];
void vp8_init_me_luts()
{
int i;
// Initialize the sad lut tables using a formulaic calculation for now
// This is to make it easier to resolve the impact of experimental changes
281282283284285286287288289290291292293294295296297298299300301302303304305306307308309310311312313314315316317318319320321322323324325326327328329330331332333334335336337338339340341342343344345346347348349350
// to the quantizer tables.
for ( i = 0; i < QINDEX_RANGE; i++ )
{
sad_per_bit16lut[i] =
(int)((0.0418*vp8_convert_qindex_to_q(i)) + 2.4107);
sad_per_bit4lut[i] = (int)((0.063*vp8_convert_qindex_to_q(i)) + 2.742);
}
}
int compute_rd_mult( int qindex )
{
int q;
q = vp8_dc_quant(qindex,0);
return (11 * q * q) >> 6;
}
void vp8cx_initialize_me_consts(VP8_COMP *cpi, int QIndex)
{
cpi->mb.sadperbit16 = sad_per_bit16lut[QIndex];
cpi->mb.sadperbit4 = sad_per_bit4lut[QIndex];
}
void vp8_initialize_rd_consts(VP8_COMP *cpi, int QIndex)
{
int q;
int i;
vp8_clear_system_state(); //__asm emms;
// Further tests required to see if optimum is different
// for key frames, golden frames and arf frames.
// if (cpi->common.refresh_golden_frame ||
// cpi->common.refresh_alt_ref_frame)
QIndex=(QIndex<0)? 0 : ((QIndex>MAXQ)?MAXQ : QIndex);
cpi->RDMULT = compute_rd_mult(QIndex);
// Extend rate multiplier along side quantizer zbin increases
if (cpi->zbin_over_quant > 0)
{
double oq_factor;
// Experimental code using the same basic equation as used for Q above
// The units of cpi->zbin_over_quant are 1/128 of Q bin size
oq_factor = 1.0 + ((double)0.0015625 * cpi->zbin_over_quant);
cpi->RDMULT = (int)((double)cpi->RDMULT * oq_factor * oq_factor);
}
if (cpi->pass == 2 && (cpi->common.frame_type != KEY_FRAME))
{
if (cpi->twopass.next_iiratio > 31)
cpi->RDMULT += (cpi->RDMULT * rd_iifactor[31]) >> 4;
else
cpi->RDMULT +=
(cpi->RDMULT * rd_iifactor[cpi->twopass.next_iiratio]) >> 4;
}
if (cpi->RDMULT < 7)
cpi->RDMULT = 7;
cpi->mb.errorperbit = (cpi->RDMULT / 110);
cpi->mb.errorperbit += (cpi->mb.errorperbit==0);
vp8_set_speed_features(cpi);
q = (int)pow(vp8_dc_quant(QIndex,0)>>2, 1.25);
q = q << 2;
cpi->RDMULT = cpi->RDMULT << 4;
351352353354355356357358359360361362363364365366367368369370371372373374375376377378379380381382383384385386387388389390391392393394395396397398399400401402403404405406407408409410411412413414415416417418419420
if (q < 8)
q = 8;
if (cpi->RDMULT > 1000)
{
cpi->RDDIV = 1;
cpi->RDMULT /= 100;
for (i = 0; i < MAX_MODES; i++)
{
if (cpi->sf.thresh_mult[i] < INT_MAX)
{
cpi->rd_threshes[i] = cpi->sf.thresh_mult[i] * q / 100;
}
else
{
cpi->rd_threshes[i] = INT_MAX;
}
cpi->rd_baseline_thresh[i] = cpi->rd_threshes[i];
}
}
else
{
cpi->RDDIV = 100;
for (i = 0; i < MAX_MODES; i++)
{
if (cpi->sf.thresh_mult[i] < (INT_MAX / q))
{
cpi->rd_threshes[i] = cpi->sf.thresh_mult[i] * q;
}
else
{
cpi->rd_threshes[i] = INT_MAX;
}
cpi->rd_baseline_thresh[i] = cpi->rd_threshes[i];
}
}
fill_token_costs(
cpi->mb.token_costs,
(const vp8_prob( *)[8][PREV_COEF_CONTEXTS][11]) cpi->common.fc.coef_probs,
BLOCK_TYPES);
fill_token_costs(
cpi->mb.token_costs_8x8,
(const vp8_prob( *)[8][PREV_COEF_CONTEXTS][11]) cpi->common.fc.coef_probs_8x8,
BLOCK_TYPES_8X8);
/*rough estimate for costing*/
cpi->common.kf_ymode_probs_index = cpi->common.base_qindex>>4;
vp8_init_mode_costs(cpi);
}
void vp8_auto_select_speed(VP8_COMP *cpi)
{
int milliseconds_for_compress = (int)(1000000 / cpi->oxcf.frame_rate);
milliseconds_for_compress = milliseconds_for_compress * (16 - cpi->oxcf.cpu_used) / 16;
#if 0
if (0)
{
FILE *f;
421422423424425426427428429430431432433434435436437438439440441442443444445446447448449450451452453454455456457458459460461462463464465466467468469470471472473474475476477478479480481482483484485486487488489490
f = fopen("speed.stt", "a");
fprintf(f, " %8ld %10ld %10ld %10ld\n",
cpi->common.current_video_frame, cpi->Speed, milliseconds_for_compress, cpi->avg_pick_mode_time);
fclose(f);
}
#endif
/*
// this is done during parameter valid check
if( cpi->oxcf.cpu_used > 16)
cpi->oxcf.cpu_used = 16;
if( cpi->oxcf.cpu_used < -16)
cpi->oxcf.cpu_used = -16;
*/
if (cpi->avg_pick_mode_time < milliseconds_for_compress && (cpi->avg_encode_time - cpi->avg_pick_mode_time) < milliseconds_for_compress)
{
if (cpi->avg_pick_mode_time == 0)
{
cpi->Speed = 4;
}
else
{
if (milliseconds_for_compress * 100 < cpi->avg_encode_time * 95)
{
cpi->Speed += 2;
cpi->avg_pick_mode_time = 0;
cpi->avg_encode_time = 0;
if (cpi->Speed > 16)
{
cpi->Speed = 16;
}
}
if (milliseconds_for_compress * 100 > cpi->avg_encode_time * auto_speed_thresh[cpi->Speed])
{
cpi->Speed -= 1;
cpi->avg_pick_mode_time = 0;
cpi->avg_encode_time = 0;
// In real-time mode, cpi->speed is in [4, 16].
if (cpi->Speed < 4) //if ( cpi->Speed < 0 )
{
cpi->Speed = 4; //cpi->Speed = 0;
}
}
}
}
else
{
cpi->Speed += 4;
if (cpi->Speed > 16)
cpi->Speed = 16;
cpi->avg_pick_mode_time = 0;
cpi->avg_encode_time = 0;
}
}
int vp8_block_error_c(short *coeff, short *dqcoeff)
{
int i;
int error = 0;
for (i = 0; i < 16; i++)
{
491492493494495496497498499500501502503504505506507508509510511512513514515516517518519520521522523524525526527528529530531532533534535536537538539540541542543544545546547548549550551552553554555556557558559560
int this_diff = coeff[i] - dqcoeff[i];
error += this_diff * this_diff;
}
return error;
}
int vp8_mbblock_error_c(MACROBLOCK *mb, int dc)
{
BLOCK *be;
BLOCKD *bd;
int i, j;
int berror, error = 0;
for (i = 0; i < 16; i++)
{
be = &mb->block[i];
bd = &mb->e_mbd.block[i];
berror = 0;
for (j = dc; j < 16; j++)
{
int this_diff = be->coeff[j] - bd->dqcoeff[j];
berror += this_diff * this_diff;
}
error += berror;
}
return error;
}
int vp8_mbuverror_c(MACROBLOCK *mb)
{
BLOCK *be;
BLOCKD *bd;
int i;
int error = 0;
for (i = 16; i < 24; i++)
{
be = &mb->block[i];
bd = &mb->e_mbd.block[i];
error += vp8_block_error_c(be->coeff, bd->dqcoeff);
}
return error;
}
int VP8_UVSSE(MACROBLOCK *x, const vp8_variance_rtcd_vtable_t *rtcd)
{
unsigned char *uptr, *vptr;
unsigned char *upred_ptr = (*(x->block[16].base_src) + x->block[16].src);
unsigned char *vpred_ptr = (*(x->block[20].base_src) + x->block[20].src);
int uv_stride = x->block[16].src_stride;
unsigned int sse1 = 0;
unsigned int sse2 = 0;
int mv_row = x->e_mbd.mode_info_context->mbmi.mv.as_mv.row;
int mv_col = x->e_mbd.mode_info_context->mbmi.mv.as_mv.col;
int offset;
int pre_stride = x->e_mbd.block[16].pre_stride;
if (mv_row < 0)
mv_row -= 1;
561562563564565566567568569570571572573574575576577578579580581582583584585586587588589590591592593594595596597598599600601602603604605606607608609610611612613614615616617618619620621622623624625626627628629630
else
mv_row += 1;
if (mv_col < 0)
mv_col -= 1;
else
mv_col += 1;
mv_row /= 2;
mv_col /= 2;
offset = (mv_row >> 3) * pre_stride + (mv_col >> 3);
uptr = x->e_mbd.pre.u_buffer + offset;
vptr = x->e_mbd.pre.v_buffer + offset;
if ((mv_row | mv_col) & 7)
{
#if CONFIG_SIXTEENTH_SUBPEL_UV
VARIANCE_INVOKE(rtcd, subpixvar8x8)(uptr, pre_stride,
(mv_col & 7)<<1, (mv_row & 7)<<1, upred_ptr, uv_stride, &sse2);
VARIANCE_INVOKE(rtcd, subpixvar8x8)(vptr, pre_stride,
(mv_col & 7)<<1, (mv_row & 7)<<1, vpred_ptr, uv_stride, &sse1);
#else
VARIANCE_INVOKE(rtcd, subpixvar8x8)(uptr, pre_stride,
mv_col & 7, mv_row & 7, upred_ptr, uv_stride, &sse2);
VARIANCE_INVOKE(rtcd, subpixvar8x8)(vptr, pre_stride,
mv_col & 7, mv_row & 7, vpred_ptr, uv_stride, &sse1);
#endif
sse2 += sse1;
}
else
{
VARIANCE_INVOKE(rtcd, var8x8)(uptr, pre_stride,
upred_ptr, uv_stride, &sse2);
VARIANCE_INVOKE(rtcd, var8x8)(vptr, pre_stride,
vpred_ptr, uv_stride, &sse1);
sse2 += sse1;
}
return sse2;
}
static int cost_coeffs(MACROBLOCK *mb, BLOCKD *b, int type, ENTROPY_CONTEXT *a, ENTROPY_CONTEXT *l)
{
int c = !type; /* start at coef 0, unless Y with Y2 */
int eob = b->eob;
int pt ; /* surrounding block/prev coef predictor */
int cost = 0;
short *qcoeff_ptr = b->qcoeff;
VP8_COMBINEENTROPYCONTEXTS(pt, *a, *l);
# define QC( I) ( qcoeff_ptr [vp8_default_zig_zag1d[I]] )
for (; c < eob; c++)
{
int v = QC(c);
int t = vp8_dct_value_tokens_ptr[v].Token;
cost += mb->token_costs [type] [vp8_coef_bands[c]] [pt] [t];
cost += vp8_dct_value_cost_ptr[v];
pt = vp8_prev_token_class[t];
}
# undef QC
if (c < 16)
cost += mb->token_costs [type] [vp8_coef_bands[c]] [pt] [DCT_EOB_TOKEN];
pt = (c != !type); // is eob first coefficient;
*a = *l = pt;
631632633634635636637638639640641642643644645646647648649650651652653654655656657658659660661662663664665666667668669670671672673674675676677678679680681682683684685686687688689690691692693694695696697698699700
return cost;
}
static int vp8_rdcost_mby(MACROBLOCK *mb)
{
int cost = 0;
int b;
MACROBLOCKD *x = &mb->e_mbd;
ENTROPY_CONTEXT_PLANES t_above, t_left;
ENTROPY_CONTEXT *ta;
ENTROPY_CONTEXT *tl;
vpx_memcpy(&t_above, mb->e_mbd.above_context, sizeof(ENTROPY_CONTEXT_PLANES));
vpx_memcpy(&t_left, mb->e_mbd.left_context, sizeof(ENTROPY_CONTEXT_PLANES));
ta = (ENTROPY_CONTEXT *)&t_above;
tl = (ENTROPY_CONTEXT *)&t_left;
for (b = 0; b < 16; b++)
cost += cost_coeffs(mb, x->block + b, PLANE_TYPE_Y_NO_DC,
ta + vp8_block2above[b], tl + vp8_block2left[b]);
cost += cost_coeffs(mb, x->block + 24, PLANE_TYPE_Y2,
ta + vp8_block2above[24], tl + vp8_block2left[24]);
return cost;
}
static void macro_block_yrd( MACROBLOCK *mb,
int *Rate,
int *Distortion,
const VP8_ENCODER_RTCD *rtcd)
{
int b;
MACROBLOCKD *const x = &mb->e_mbd;
BLOCK *const mb_y2 = mb->block + 24;
BLOCKD *const x_y2 = x->block + 24;
short *Y2DCPtr = mb_y2->src_diff;
BLOCK *beptr;
int d;
ENCODEMB_INVOKE(&rtcd->encodemb, submby)(
mb->src_diff,
*(mb->block[0].base_src),
mb->e_mbd.predictor,
mb->block[0].src_stride );
// Fdct and building the 2nd order block
for (beptr = mb->block; beptr < mb->block + 16; beptr += 2)
{
mb->vp8_short_fdct8x4(beptr->src_diff, beptr->coeff, 32);
*Y2DCPtr++ = beptr->coeff[0];
*Y2DCPtr++ = beptr->coeff[16];
}
// 2nd order fdct
mb->short_walsh4x4(mb_y2->src_diff, mb_y2->coeff, 8);
// Quantization
for (b = 0; b < 16; b++)
{
mb->quantize_b(&mb->block[b], &mb->e_mbd.block[b]);
}
// DC predication and Quantization of 2nd Order block
mb->quantize_b(mb_y2, x_y2);
// Distortion
d = ENCODEMB_INVOKE(&rtcd->encodemb, mberr)(mb, 1);
701702703704705706707708709710711712713714715716717718719720721722723724725726727728729730731732733734735736737738739740741742743744745746747748749750751752753754755756757758759760761762763764765766767768769770
d += ENCODEMB_INVOKE(&rtcd->encodemb, berr)(mb_y2->coeff, x_y2->dqcoeff);
*Distortion = (d >> 2);
// rate
*Rate = vp8_rdcost_mby(mb);
}
static int cost_coeffs_2x2(MACROBLOCK *mb,
BLOCKD *b, int type,
ENTROPY_CONTEXT *a, ENTROPY_CONTEXT *l)
{
int c = !type; /* start at coef 0, unless Y with Y2 */
int eob = b->eob;
int pt ; /* surrounding block/prev coef predictor */
int cost = 0;
short *qcoeff_ptr = b->qcoeff;
VP8_COMBINEENTROPYCONTEXTS(pt, *a, *l);
assert(eob<=4);
# define QC2X2( I) ( qcoeff_ptr [vp8_default_zig_zag1d[I]] )
for (; c < eob; c++)
{
int v = QC2X2(c);
int t = vp8_dct_value_tokens_ptr[v].Token;
cost += mb->token_costs_8x8[type] [vp8_coef_bands[c]] [pt] [t];
cost += vp8_dct_value_cost_ptr[v];
pt = vp8_prev_token_class[t];
}
# undef QC2X2
if (c < 4)
cost += mb->token_costs_8x8 [type][vp8_coef_bands[c]]
[pt] [DCT_EOB_TOKEN];
pt = (c != !type); // is eob first coefficient;
*a = *l = pt;
return cost;
}
static int cost_coeffs_8x8(MACROBLOCK *mb,
BLOCKD *b, int type,
ENTROPY_CONTEXT *a, ENTROPY_CONTEXT *l)
{
int c = !type; /* start at coef 0, unless Y with Y2 */
int eob = b->eob;
int pt ; /* surrounding block/prev coef predictor */
int cost = 0;
short *qcoeff_ptr = b->qcoeff;
VP8_COMBINEENTROPYCONTEXTS(pt, *a, *l);
# define QC8X8( I) ( qcoeff_ptr [vp8_default_zig_zag1d_8x8[I]] )
for (; c < eob; c++)
{
int v = QC8X8(c);
int t = vp8_dct_value_tokens_ptr[v].Token;
cost += mb->token_costs_8x8[type] [vp8_coef_bands_8x8[c]] [pt] [t];
cost += vp8_dct_value_cost_ptr[v];
pt = vp8_prev_token_class[t];
}
# undef QC8X8
if (c < 64)
cost += mb->token_costs_8x8 [type][vp8_coef_bands_8x8[c]]
771772773774775776777778779780781782783784785786787788789790791792793794795796797798799800801802803804805806807808809810811812813814815816817818819820821822823824825826827828829830831832833834835836837838839840
[pt] [DCT_EOB_TOKEN];
pt = (c != !type); // is eob first coefficient;
*a = *l = pt;
return cost;
}
static int vp8_rdcost_mby_8x8(MACROBLOCK *mb)
{
int cost = 0;
int b;
MACROBLOCKD *x = &mb->e_mbd;
ENTROPY_CONTEXT_PLANES t_above, t_left;
ENTROPY_CONTEXT *ta;
ENTROPY_CONTEXT *tl;
vpx_memcpy(&t_above, mb->e_mbd.above_context, sizeof(ENTROPY_CONTEXT_PLANES));
vpx_memcpy(&t_left, mb->e_mbd.left_context, sizeof(ENTROPY_CONTEXT_PLANES));
ta = (ENTROPY_CONTEXT *)&t_above;
tl = (ENTROPY_CONTEXT *)&t_left;
for (b = 0; b < 16; b+=4)
cost += cost_coeffs_8x8(mb, x->block + b, PLANE_TYPE_Y_NO_DC,
ta + vp8_block2above_8x8[b], tl + vp8_block2left_8x8[b]);
cost += cost_coeffs_2x2(mb, x->block + 24, PLANE_TYPE_Y2,
ta + vp8_block2above[24], tl + vp8_block2left[24]);
return cost;
}
static void macro_block_yrd_8x8( MACROBLOCK *mb,
int *Rate,
int *Distortion,
const VP8_ENCODER_RTCD *rtcd)
{
MACROBLOCKD *const x = &mb->e_mbd;
BLOCK *const mb_y2 = mb->block + 24;
BLOCKD *const x_y2 = x->block + 24;
short *Y2DCPtr = mb_y2->src_diff;
int d;
ENCODEMB_INVOKE(&rtcd->encodemb, submby)(
mb->src_diff,
*(mb->block[0].base_src),
mb->e_mbd.predictor,
mb->block[0].src_stride );
vp8_transform_mby_8x8(mb);
vp8_quantize_mby_8x8(mb);
/* remove 1st order dc to properly combine 1st/2nd order distortion */
mb->coeff[0] = 0;
mb->coeff[64] = 0;
mb->coeff[128] = 0;
mb->coeff[192] = 0;
mb->e_mbd.dqcoeff[0] = 0;
mb->e_mbd.dqcoeff[64] = 0;
mb->e_mbd.dqcoeff[128] = 0;
mb->e_mbd.dqcoeff[192] = 0;
d = ENCODEMB_INVOKE(&rtcd->encodemb, mberr)(mb, 0);
d += ENCODEMB_INVOKE(&rtcd->encodemb, berr)(mb_y2->coeff, x_y2->dqcoeff);
*Distortion = (d >> 2);
// rate
*Rate = vp8_rdcost_mby_8x8(mb);
}
static void copy_predictor(unsigned char *dst, const unsigned char *predictor)
{
841842843844845846847848849850851852853854855856857858859860861862863864865866867868869870871872873874875876877878879880881882883884885886887888889890891892893894895896897898899900901902903904905906907908909910
const unsigned int *p = (const unsigned int *)predictor;
unsigned int *d = (unsigned int *)dst;
d[0] = p[0];
d[4] = p[4];
d[8] = p[8];
d[12] = p[12];
}
static void copy_predictor_8x8(unsigned char *dst, const unsigned char *predictor)
{
const unsigned int *p = (const unsigned int *)predictor;
unsigned int *d = (unsigned int *)dst;
d[0] = p[0];
d[1] = p[1];
d[4] = p[4];
d[5] = p[5];
d[8] = p[8];
d[9] = p[9];
d[12] = p[12];
d[13] = p[13];
d[16] = p[16];
d[17] = p[17];
d[20] = p[20];
d[21] = p[21];
d[24] = p[24];
d[25] = p[25];
d[28] = p[28];
d[29] = p[29];
}
static int rd_pick_intra4x4block(
VP8_COMP *cpi,
MACROBLOCK *x,
BLOCK *be,
BLOCKD *b,
B_PREDICTION_MODE *best_mode,
#if CONFIG_COMP_INTRA_PRED
B_PREDICTION_MODE *best_second_mode,
int allow_comp,
#endif
unsigned int *bmode_costs,
ENTROPY_CONTEXT *a,
ENTROPY_CONTEXT *l,
int *bestrate,
int *bestratey,
int *bestdistortion)
{
B_PREDICTION_MODE mode;
#if CONFIG_COMP_INTRA_PRED
B_PREDICTION_MODE mode2;
#endif
int best_rd = INT_MAX;
int rate = 0;
int distortion;
ENTROPY_CONTEXT ta = *a, tempa = *a;
ENTROPY_CONTEXT tl = *l, templ = *l;
/*
* The predictor buffer is a 2d buffer with a stride of 16. Create
* a temp buffer that meets the stride requirements, but we are only
* interested in the left 4x4 block
* */
DECLARE_ALIGNED_ARRAY(16, unsigned char, best_predictor, 16*4);
DECLARE_ALIGNED_ARRAY(16, short, best_dqcoeff, 16);
for (mode = B_DC_PRED; mode <= B_HU_PRED; mode++)
{
#if CONFIG_COMP_INTRA_PRED
for (mode2 = (allow_comp ? 0 : (B_DC_PRED - 1)); mode2 != (allow_comp ? (mode + 1) : 0); mode2++)
911912913914915916917918919920921922923924925926927928929930931932933934935936937938939940941942943944945946947948949950951952953954955956957958959960961962963964965966967968969970971972973974975976977978979980
{
#endif
int this_rd;
int ratey;
// TODO Temporarily ignore modes that need the above-right data. SB
// encoding means this data is not available for the bottom right MB
// Do we need to do this for mode2 also?
if (mode==B_LD_PRED || mode==B_VL_PRED)
continue;
rate = bmode_costs[mode];
#if CONFIG_COMP_INTRA_PRED
if (mode2 == (B_PREDICTION_MODE) (B_DC_PRED - 1))
{
#endif
RECON_INVOKE(&cpi->rtcd.common->recon, intra4x4_predict)
(b, mode, b->predictor);
#if CONFIG_COMP_INTRA_PRED
}
else
{
RECON_INVOKE(&cpi->rtcd.common->recon, comp_intra4x4_predict)
(b, mode, mode2, b->predictor);
rate += bmode_costs[mode2];
}
#endif
ENCODEMB_INVOKE(IF_RTCD(&cpi->rtcd.encodemb), subb)(be, b, 16);
x->vp8_short_fdct4x4(be->src_diff, be->coeff, 32);
x->quantize_b(be, b);
tempa = ta;
templ = tl;
ratey = cost_coeffs(x, b, PLANE_TYPE_Y_WITH_DC, &tempa, &templ);
rate += ratey;
distortion = ENCODEMB_INVOKE(IF_RTCD(&cpi->rtcd.encodemb), berr)(
be->coeff, b->dqcoeff) >> 2;
this_rd = RDCOST(x->rdmult, x->rddiv, rate, distortion);
if (this_rd < best_rd)
{
*bestrate = rate;
*bestratey = ratey;
*bestdistortion = distortion;
best_rd = this_rd;
*best_mode = mode;
#if CONFIG_COMP_INTRA_PRED
*best_second_mode = mode2;
#endif
*a = tempa;
*l = templ;
copy_predictor(best_predictor, b->predictor);
vpx_memcpy(best_dqcoeff, b->dqcoeff, 32);
#if CONFIG_COMP_INTRA_PRED
}
#endif
}
}
b->bmi.as_mode.first = (B_PREDICTION_MODE)(*best_mode);
#if CONFIG_COMP_INTRA_PRED
b->bmi.as_mode.second = (B_PREDICTION_MODE)(*best_second_mode);
#endif
IDCT_INVOKE(IF_RTCD(&cpi->rtcd.common->idct), idct16)(best_dqcoeff, b->diff, 32);
RECON_INVOKE(IF_RTCD(&cpi->rtcd.common->recon), recon)(best_predictor, b->diff, *(b->base_dst) + b->dst, b->dst_stride);
return best_rd;
}
981982983984985986987988989990991992993994995996997998999100010011002100310041005100610071008100910101011101210131014101510161017101810191020102110221023102410251026102710281029103010311032103310341035103610371038103910401041104210431044104510461047104810491050
static int rd_pick_intra4x4mby_modes(VP8_COMP *cpi, MACROBLOCK *mb, int *Rate,
int *rate_y, int *Distortion, int best_rd,
#if CONFIG_COMP_INTRA_PRED
int allow_comp,
#endif
int update_contexts)
{
int i;
MACROBLOCKD *const xd = &mb->e_mbd;
int cost = mb->mbmode_cost [xd->frame_type] [B_PRED];
int distortion = 0;
int tot_rate_y = 0;
int64_t total_rd = 0;
ENTROPY_CONTEXT_PLANES t_above, t_left;
ENTROPY_CONTEXT *ta;
ENTROPY_CONTEXT *tl;
unsigned int *bmode_costs;
if (update_contexts)
{
ta = (ENTROPY_CONTEXT *)mb->e_mbd.above_context;
tl = (ENTROPY_CONTEXT *)mb->e_mbd.left_context;
}
else
{
vpx_memcpy(&t_above, mb->e_mbd.above_context,
sizeof(ENTROPY_CONTEXT_PLANES));
vpx_memcpy(&t_left, mb->e_mbd.left_context,
sizeof(ENTROPY_CONTEXT_PLANES));
ta = (ENTROPY_CONTEXT *)&t_above;
tl = (ENTROPY_CONTEXT *)&t_left;
}
#if 0
vp8_intra_prediction_down_copy(xd);
#endif
bmode_costs = mb->inter_bmode_costs;
for (i = 0; i < 16; i++)
{
MODE_INFO *const mic = xd->mode_info_context;
const int mis = xd->mode_info_stride;
B_PREDICTION_MODE UNINITIALIZED_IS_SAFE(best_mode);
#if CONFIG_COMP_INTRA_PRED
B_PREDICTION_MODE UNINITIALIZED_IS_SAFE(best_second_mode);
#endif
int UNINITIALIZED_IS_SAFE(r), UNINITIALIZED_IS_SAFE(ry), UNINITIALIZED_IS_SAFE(d);
if (mb->e_mbd.frame_type == KEY_FRAME)
{
const B_PREDICTION_MODE A = above_block_mode(mic, i, mis);
const B_PREDICTION_MODE L = left_block_mode(mic, i);
bmode_costs = mb->bmode_costs[A][L];
}
total_rd += rd_pick_intra4x4block(
cpi, mb, mb->block + i, xd->block + i, &best_mode,
#if CONFIG_COMP_INTRA_PRED
&best_second_mode, allow_comp,
#endif
bmode_costs, ta + vp8_block2above[i],
tl + vp8_block2left[i], &r, &ry, &d);
cost += r;
distortion += d;
tot_rate_y += ry;
1051105210531054105510561057105810591060106110621063106410651066106710681069107010711072107310741075107610771078107910801081108210831084108510861087108810891090109110921093109410951096109710981099110011011102110311041105110611071108110911101111111211131114111511161117111811191120
mic->bmi[i].as_mode.first = best_mode;
#if CONFIG_COMP_INTRA_PRED
mic->bmi[i].as_mode.second = best_second_mode;
#endif
if(total_rd >= (int64_t)best_rd)
break;
}
if(total_rd >= (int64_t)best_rd)
return INT_MAX;
#if CONFIG_COMP_INTRA_PRED
cost += vp8_cost_bit(128, allow_comp);
#endif
*Rate = cost;
*rate_y += tot_rate_y;
*Distortion = distortion;
return RDCOST(mb->rdmult, mb->rddiv, cost, distortion);
}
static int rd_pick_intra16x16mby_mode(VP8_COMP *cpi,
MACROBLOCK *x,
int *Rate,
int *rate_y,
int *Distortion)
{
MB_PREDICTION_MODE mode;
MB_PREDICTION_MODE UNINITIALIZED_IS_SAFE(mode_selected);
#if CONFIG_COMP_INTRA_PRED
MB_PREDICTION_MODE mode2;
MB_PREDICTION_MODE UNINITIALIZED_IS_SAFE(mode2_selected);
#endif
int rate, ratey;
int distortion;
int best_rd = INT_MAX;
int this_rd;
//Y Search for 16x16 intra prediction mode
for (mode = DC_PRED; mode <= TM_PRED; mode++)
{
x->e_mbd.mode_info_context->mbmi.mode = mode;
#if CONFIG_COMP_INTRA_PRED
for (mode2 = DC_PRED - 1; mode2 != TM_PRED + 1; mode2++)
{
x->e_mbd.mode_info_context->mbmi.second_mode = mode2;
if (mode2 == (MB_PREDICTION_MODE) (DC_PRED - 1))
{
#endif
RECON_INVOKE(&cpi->common.rtcd.recon, build_intra_predictors_mby)
(&x->e_mbd);
#if CONFIG_COMP_INTRA_PRED
}
else
{
continue; // i.e. disable for now
RECON_INVOKE(&cpi->common.rtcd.recon, build_comp_intra_predictors_mby)(&x->e_mbd);
}
#endif
macro_block_yrd(x, &ratey, &distortion, IF_RTCD(&cpi->rtcd));
// FIXME add compoundmode cost
// FIXME add rate for mode2
rate = ratey + x->mbmode_cost[x->e_mbd.frame_type]
[x->e_mbd.mode_info_context->mbmi.mode];
this_rd = RDCOST(x->rdmult, x->rddiv, rate, distortion);
1121112211231124112511261127112811291130113111321133113411351136113711381139114011411142114311441145114611471148114911501151115211531154115511561157115811591160116111621163116411651166116711681169117011711172117311741175117611771178117911801181118211831184118511861187118811891190
if (this_rd < best_rd)
{
mode_selected = mode;
#if CONFIG_COMP_INTRA_PRED
mode2_selected = mode2;
#endif
best_rd = this_rd;
*Rate = rate;
*rate_y = ratey;
*Distortion = distortion;
}
#if CONFIG_COMP_INTRA_PRED
}
#endif
}
x->e_mbd.mode_info_context->mbmi.mode = mode_selected;
#if CONFIG_COMP_INTRA_PRED
x->e_mbd.mode_info_context->mbmi.second_mode = mode2_selected;
#endif
return best_rd;
}
static int rd_pick_intra8x8block(
VP8_COMP *cpi,
MACROBLOCK *x,
int ib,
B_PREDICTION_MODE *best_mode,
#if CONFIG_COMP_INTRA_PRED
B_PREDICTION_MODE *best_second_mode,
#endif
unsigned int *mode_costs,
ENTROPY_CONTEXT *a,
ENTROPY_CONTEXT *l,
int *bestrate,
int *bestratey,
int *bestdistortion)
{
MB_PREDICTION_MODE mode;
#if CONFIG_COMP_INTRA_PRED
MB_PREDICTION_MODE mode2;
#endif
MACROBLOCKD *xd = &x->e_mbd;
int best_rd = INT_MAX;
int rate = 0;
int distortion;
BLOCK *be=x->block + ib;
BLOCKD *b=x->e_mbd.block + ib;
ENTROPY_CONTEXT ta0, ta1, besta0, besta1;
ENTROPY_CONTEXT tl0, tl1, bestl0, bestl1;
/*
* The predictor buffer is a 2d buffer with a stride of 16. Create
* a temp buffer that meets the stride requirements, but we are only
* interested in the left 8x8 block
* */
DECLARE_ALIGNED_ARRAY(16, unsigned char, best_predictor, 16*8);
DECLARE_ALIGNED_ARRAY(16, short, best_dqcoeff, 16*4);
for (mode = DC_PRED; mode <= TM_PRED; mode++)
{
#if CONFIG_COMP_INTRA_PRED
for (mode2 = DC_PRED - 1; mode2 != TM_PRED + 1; mode2++)
{
#endif
int this_rd;
int rate_t;
1191119211931194119511961197119811991200120112021203120412051206120712081209121012111212121312141215121612171218121912201221122212231224122512261227122812291230123112321233123412351236123712381239124012411242124312441245124612471248124912501251125212531254125512561257125812591260
// FIXME rate for compound mode and second intrapred mode
rate = mode_costs[mode];
#if CONFIG_COMP_INTRA_PRED
if (mode2 == (MB_PREDICTION_MODE) (DC_PRED - 1))
{
#endif
RECON_INVOKE(&cpi->rtcd.common->recon, intra8x8_predict)
(b, mode, b->predictor);
#if CONFIG_COMP_INTRA_PRED
}
else
{
continue; // i.e. disable for now
RECON_INVOKE(&cpi->rtcd.common->recon, comp_intra8x8_predict)
(b, mode, mode2, b->predictor);
}
#endif
vp8_subtract_4b_c(be, b, 16);
x->vp8_short_fdct8x4(be->src_diff, be->coeff, 32);
x->vp8_short_fdct8x4(be->src_diff + 64, be->coeff + 64, 32);
x->quantize_b_pair(x->block+ib, x->block+ib+1,
xd->block+ib, xd->block+ib+1);
x->quantize_b_pair(x->block+ib+4, x->block+ib+5,
xd->block+ib+4, xd->block+ib+5);
distortion = ENCODEMB_INVOKE(IF_RTCD(&cpi->rtcd.encodemb), berr)
((x->block+ib)->coeff,(xd->block+ib)->dqcoeff)>>2;
distortion += ENCODEMB_INVOKE(IF_RTCD(&cpi->rtcd.encodemb), berr)
((x->block+ib+1)->coeff,(xd->block+ib+1)->dqcoeff)>>2;
distortion += ENCODEMB_INVOKE(IF_RTCD(&cpi->rtcd.encodemb), berr)
((x->block+ib+4)->coeff,(xd->block+ib+4)->dqcoeff)>>2;
distortion += ENCODEMB_INVOKE(IF_RTCD(&cpi->rtcd.encodemb), berr)
((x->block+ib+5)->coeff,(xd->block+ib+5)->dqcoeff)>>2;
ta0 = *(a + vp8_block2above[ib]);
ta1 = *(a + vp8_block2above[ib+1]);
tl0 = *(l + vp8_block2above[ib]);
tl1 = *(l + vp8_block2above[ib+4]);
rate_t = cost_coeffs(x, xd->block+ib, PLANE_TYPE_Y_WITH_DC,
&ta0, &tl0);
rate_t += cost_coeffs(x, xd->block+ib+1, PLANE_TYPE_Y_WITH_DC,
&ta1, &tl0);
rate_t += cost_coeffs(x, xd->block+ib+4, PLANE_TYPE_Y_WITH_DC,
&ta0, &tl1);
rate_t += cost_coeffs(x, xd->block+ib+5, PLANE_TYPE_Y_WITH_DC,
&ta1, &tl1);
rate += rate_t;
this_rd = RDCOST(x->rdmult, x->rddiv, rate, distortion);
if (this_rd < best_rd)
{
*bestrate = rate;
*bestratey = rate_t;
*bestdistortion = distortion;
besta0 = ta0;
besta1 = ta1;
bestl0 = tl0;
bestl1 = tl1;
best_rd = this_rd;
*best_mode = mode;
#if CONFIG_COMP_INTRA_PRED
*best_second_mode = mode2;
#endif
copy_predictor_8x8(best_predictor, b->predictor);
vpx_memcpy(best_dqcoeff, b->dqcoeff, 64);
vpx_memcpy(best_dqcoeff+32, b->dqcoeff+64, 64);
#if CONFIG_COMP_INTRA_PRED
1261126212631264126512661267126812691270127112721273127412751276127712781279128012811282128312841285128612871288128912901291129212931294129512961297129812991300130113021303130413051306130713081309131013111312131313141315131613171318131913201321132213231324132513261327132813291330
}
#endif
}
}
b->bmi.as_mode.first = (*best_mode);
#if CONFIG_COMP_INTRA_PRED
b->bmi.as_mode.second = (*best_second_mode);
#endif
vp8_encode_intra8x8 (IF_RTCD(&cpi->rtcd), x, ib);
*(a + vp8_block2above[ib]) = besta0;
*(a + vp8_block2above[ib+1]) = besta1;
*(l + vp8_block2above[ib]) = bestl0;
*(l + vp8_block2above[ib+4]) = bestl1;
return best_rd;
}
const int vp8_i8x8_block[4]={0, 2, 8, 10};
int rd_pick_intra8x8mby_modes(VP8_COMP *cpi,
MACROBLOCK *mb,
int *Rate,
int *rate_y,
int *Distortion,
int best_rd)
{
MACROBLOCKD *const xd = &mb->e_mbd;
int i,ib;
int cost = mb->mbmode_cost [xd->frame_type] [I8X8_PRED];
int distortion = 0;
int tot_rate_y = 0;
long long total_rd = 0;
ENTROPY_CONTEXT_PLANES t_above, t_left;
ENTROPY_CONTEXT *ta;
ENTROPY_CONTEXT *tl;
unsigned int *i8x8mode_costs;
vpx_memcpy(&t_above, mb->e_mbd.above_context, sizeof(ENTROPY_CONTEXT_PLANES));
vpx_memcpy(&t_left, mb->e_mbd.left_context, sizeof(ENTROPY_CONTEXT_PLANES));
ta = (ENTROPY_CONTEXT *)&t_above;
tl = (ENTROPY_CONTEXT *)&t_left;
i8x8mode_costs = mb->i8x8_mode_costs;
for (i = 0; i < 4; i++)
{
MODE_INFO *const mic = xd->mode_info_context;
B_PREDICTION_MODE UNINITIALIZED_IS_SAFE(best_mode);
#if CONFIG_COMP_INTRA_PRED
B_PREDICTION_MODE UNINITIALIZED_IS_SAFE(best_second_mode);
#endif
int UNINITIALIZED_IS_SAFE(r), UNINITIALIZED_IS_SAFE(ry), UNINITIALIZED_IS_SAFE(d);
ib = vp8_i8x8_block[i];
total_rd += rd_pick_intra8x8block(
cpi, mb, ib, &best_mode,
#if CONFIG_COMP_INTRA_PRED
&best_second_mode,
#endif
i8x8mode_costs, ta, tl, &r, &ry, &d);
cost += r;
distortion += d;
tot_rate_y += ry;
mic->bmi[ib].as_mode.first = best_mode;
#if CONFIG_COMP_INTRA_PRED
mic->bmi[ib].as_mode.second = best_second_mode;
#endif
}
*Rate = cost;
*rate_y += tot_rate_y;
*Distortion = distortion;
1331133213331334133513361337133813391340134113421343134413451346134713481349135013511352135313541355135613571358135913601361136213631364136513661367136813691370137113721373137413751376137713781379138013811382138313841385138613871388138913901391139213931394139513961397139813991400
return RDCOST(mb->rdmult, mb->rddiv, cost, distortion);
}
static int rd_cost_mbuv(MACROBLOCK *mb)
{
int b;
int cost = 0;
MACROBLOCKD *x = &mb->e_mbd;
ENTROPY_CONTEXT_PLANES t_above, t_left;
ENTROPY_CONTEXT *ta;
ENTROPY_CONTEXT *tl;
vpx_memcpy(&t_above, mb->e_mbd.above_context, sizeof(ENTROPY_CONTEXT_PLANES));
vpx_memcpy(&t_left, mb->e_mbd.left_context, sizeof(ENTROPY_CONTEXT_PLANES));
ta = (ENTROPY_CONTEXT *)&t_above;
tl = (ENTROPY_CONTEXT *)&t_left;
for (b = 16; b < 24; b++)
cost += cost_coeffs(mb, x->block + b, PLANE_TYPE_UV,
ta + vp8_block2above[b], tl + vp8_block2left[b]);
return cost;
}
static int rd_inter16x16_uv(VP8_COMP *cpi, MACROBLOCK *x, int *rate,
int *distortion, int fullpixel)
{
ENCODEMB_INVOKE(IF_RTCD(&cpi->rtcd.encodemb), submbuv)(x->src_diff,
x->src.u_buffer, x->src.v_buffer, x->e_mbd.predictor, x->src.uv_stride);
vp8_transform_mbuv(x);
vp8_quantize_mbuv(x);
*rate = rd_cost_mbuv(x);
*distortion = ENCODEMB_INVOKE(&cpi->rtcd.encodemb, mbuverr)(x) / 4;
return RDCOST(x->rdmult, x->rddiv, *rate, *distortion);
}
static int rd_cost_mbuv_8x8(MACROBLOCK *mb)
{
int b;
int cost = 0;
MACROBLOCKD *x = &mb->e_mbd;
ENTROPY_CONTEXT_PLANES t_above, t_left;
ENTROPY_CONTEXT *ta;
ENTROPY_CONTEXT *tl;
vpx_memcpy(&t_above, mb->e_mbd.above_context, sizeof(ENTROPY_CONTEXT_PLANES));
vpx_memcpy(&t_left, mb->e_mbd.left_context, sizeof(ENTROPY_CONTEXT_PLANES));
ta = (ENTROPY_CONTEXT *)&t_above;
tl = (ENTROPY_CONTEXT *)&t_left;
for (b = 16; b < 24; b+=4)
cost += cost_coeffs_8x8(mb, x->block + b, PLANE_TYPE_UV,
ta + vp8_block2above_8x8[b],
tl + vp8_block2left_8x8[b]);
return cost;
}
static int rd_inter16x16_uv_8x8(VP8_COMP *cpi, MACROBLOCK *x, int *rate,
int *distortion, int fullpixel)
{
ENCODEMB_INVOKE(IF_RTCD(&cpi->rtcd.encodemb), submbuv)(x->src_diff,
x->src.u_buffer, x->src.v_buffer, x->e_mbd.predictor, x->src.uv_stride);
1401140214031404140514061407140814091410141114121413141414151416141714181419142014211422142314241425142614271428142914301431143214331434143514361437143814391440144114421443144414451446144714481449145014511452145314541455145614571458145914601461146214631464146514661467146814691470
vp8_transform_mbuv_8x8(x);
vp8_quantize_mbuv_8x8(x);
*rate = rd_cost_mbuv_8x8(x);
*distortion = ENCODEMB_INVOKE(&cpi->rtcd.encodemb, mbuverr)(x) / 4;
return RDCOST(x->rdmult, x->rddiv, *rate, *distortion);
}
static int rd_inter4x4_uv(VP8_COMP *cpi, MACROBLOCK *x, int *rate,
int *distortion, int fullpixel)
{
vp8_build_inter4x4_predictors_mbuv(&x->e_mbd);
ENCODEMB_INVOKE(IF_RTCD(&cpi->rtcd.encodemb), submbuv)(x->src_diff,
x->src.u_buffer, x->src.v_buffer, x->e_mbd.predictor, x->src.uv_stride);
vp8_transform_mbuv(x);
vp8_quantize_mbuv(x);
*rate = rd_cost_mbuv(x);
*distortion = ENCODEMB_INVOKE(&cpi->rtcd.encodemb, mbuverr)(x) / 4;
return RDCOST(x->rdmult, x->rddiv, *rate, *distortion);
}
static void rd_pick_intra_mbuv_mode(VP8_COMP *cpi,
MACROBLOCK *x,
int *rate,
int *rate_tokenonly,
int *distortion)
{
MB_PREDICTION_MODE mode;
MB_PREDICTION_MODE UNINITIALIZED_IS_SAFE(mode_selected);
#if CONFIG_COMP_INTRA_PRED
MB_PREDICTION_MODE mode2;
MB_PREDICTION_MODE UNINITIALIZED_IS_SAFE(mode2_selected);
#endif
int best_rd = INT_MAX;
int UNINITIALIZED_IS_SAFE(d), UNINITIALIZED_IS_SAFE(r);
int rate_to;
for (mode = DC_PRED; mode <= TM_PRED; mode++)
{
#if CONFIG_COMP_INTRA_PRED
for (mode2 = DC_PRED - 1; mode2 != TM_PRED + 1; mode2++)
{
#endif
int rate;
int distortion;
int this_rd;
x->e_mbd.mode_info_context->mbmi.uv_mode = mode;
#if CONFIG_COMP_INTRA_PRED
x->e_mbd.mode_info_context->mbmi.second_uv_mode = mode2;
if (mode2 == (MB_PREDICTION_MODE) (DC_PRED - 1))
{
#endif
RECON_INVOKE(&cpi->rtcd.common->recon, build_intra_predictors_mbuv)
(&x->e_mbd);
#if CONFIG_COMP_INTRA_PRED
}
else
{
continue;
RECON_INVOKE(&cpi->rtcd.common->recon, build_comp_intra_predictors_mbuv)
(&x->e_mbd);
}
1471147214731474147514761477147814791480148114821483148414851486148714881489149014911492149314941495149614971498149915001501150215031504150515061507150815091510151115121513151415151516151715181519152015211522152315241525152615271528152915301531153215331534153515361537153815391540
#endif
ENCODEMB_INVOKE(IF_RTCD(&cpi->rtcd.encodemb), submbuv)(x->src_diff,
x->src.u_buffer, x->src.v_buffer, x->e_mbd.predictor,
x->src.uv_stride);
vp8_transform_mbuv(x);
vp8_quantize_mbuv(x);
rate_to = rd_cost_mbuv(x);
rate = rate_to
+ x->intra_uv_mode_cost[x->e_mbd.frame_type]
[x->e_mbd.mode_info_context->mbmi.uv_mode];
distortion = ENCODEMB_INVOKE(&cpi->rtcd.encodemb, mbuverr)(x) / 4;
this_rd = RDCOST(x->rdmult, x->rddiv, rate, distortion);
if (this_rd < best_rd)
{
best_rd = this_rd;
d = distortion;
r = rate;
*rate_tokenonly = rate_to;
mode_selected = mode;
#if CONFIG_COMP_INTRA_PRED
mode2_selected = mode2;
}
#endif
}
}
*rate = r;
*distortion = d;
x->e_mbd.mode_info_context->mbmi.uv_mode = mode_selected;
#if CONFIG_COMP_INTRA_PRED
x->e_mbd.mode_info_context->mbmi.second_uv_mode = mode2_selected;
#endif
}
static void rd_pick_intra_mbuv_mode_8x8(VP8_COMP *cpi,
MACROBLOCK *x,
int *rate,
int *rate_tokenonly,
int *distortion)
{
MB_PREDICTION_MODE mode;
MB_PREDICTION_MODE UNINITIALIZED_IS_SAFE(mode_selected);
int best_rd = INT_MAX;
int UNINITIALIZED_IS_SAFE(d), UNINITIALIZED_IS_SAFE(r);
int rate_to;
for (mode = DC_PRED; mode <= TM_PRED; mode++)
{
int rate;
int distortion;
int this_rd;
x->e_mbd.mode_info_context->mbmi.uv_mode = mode;
RECON_INVOKE(&cpi->rtcd.common->recon, build_intra_predictors_mbuv)
(&x->e_mbd);
ENCODEMB_INVOKE(IF_RTCD(&cpi->rtcd.encodemb), submbuv)(x->src_diff,
x->src.u_buffer, x->src.v_buffer, x->e_mbd.predictor,
x->src.uv_stride);
vp8_transform_mbuv_8x8(x);
vp8_quantize_mbuv_8x8(x);
rate_to = rd_cost_mbuv_8x8(x);
rate = rate_to + x->intra_uv_mode_cost[x->e_mbd.frame_type]
1541154215431544154515461547154815491550155115521553155415551556155715581559156015611562156315641565156615671568156915701571157215731574157515761577157815791580158115821583158415851586158715881589159015911592159315941595159615971598159916001601160216031604160516061607160816091610
[x->e_mbd.mode_info_context->mbmi.uv_mode];
distortion = ENCODEMB_INVOKE(&cpi->rtcd.encodemb, mbuverr)(x) / 4;
this_rd = RDCOST(x->rdmult, x->rddiv, rate, distortion);
if (this_rd < best_rd)
{
best_rd = this_rd;
d = distortion;
r = rate;
*rate_tokenonly = rate_to;
mode_selected = mode;
}
}
*rate = r;
*distortion = d;
x->e_mbd.mode_info_context->mbmi.uv_mode = mode_selected;
}
int vp8_cost_mv_ref(VP8_COMP *cpi,
MB_PREDICTION_MODE m,
const int near_mv_ref_ct[4])
{
MACROBLOCKD *xd = &cpi->mb.e_mbd;
int segment_id = xd->mode_info_context->mbmi.segment_id;
// If the mode coding is done entirely at the segment level
// we should not account for it at the per mb level in rd code.
// Note that if the segment level coding is expanded from single mode
// to multiple mode masks as per reference frame coding we will need
// to do something different here.
if ( !segfeature_active( xd, segment_id, SEG_LVL_MODE) )
{
VP8_COMMON *pc = &cpi->common;
vp8_prob p [VP8_MVREFS-1];
assert(NEARESTMV <= m && m <= SPLITMV);
vp8_mv_ref_probs(pc, p, near_mv_ref_ct);
return vp8_cost_token(vp8_mv_ref_tree, p,
vp8_mv_ref_encoding_array - NEARESTMV + m);
}
else
return 0;
}
void vp8_set_mbmode_and_mvs(MACROBLOCK *x, MB_PREDICTION_MODE mb, int_mv *mv)
{
x->e_mbd.mode_info_context->mbmi.mode = mb;
x->e_mbd.mode_info_context->mbmi.mv.as_int = mv->as_int;
}
static int labels2mode(
MACROBLOCK *x,
int const *labelings, int which_label,
B_PREDICTION_MODE this_mode,
int_mv *this_mv, int_mv *this_second_mv,
int_mv seg_mvs[MAX_REF_FRAMES - 1],
int_mv *best_ref_mv,
int_mv *second_best_ref_mv,
int *mvcost[2]
)
{
MACROBLOCKD *const xd = & x->e_mbd;
MODE_INFO *const mic = xd->mode_info_context;
const int mis = xd->mode_info_stride;
int cost = 0;
int thismvcost = 0;
/* We have to be careful retrieving previously-encoded motion vectors.
1611161216131614161516161617161816191620162116221623162416251626162716281629163016311632163316341635163616371638163916401641164216431644164516461647164816491650165116521653165416551656165716581659166016611662166316641665166616671668166916701671167216731674167516761677167816791680
Ones from this macroblock have to be pulled from the BLOCKD array
as they have not yet made it to the bmi array in our MB_MODE_INFO. */
int i = 0;
do
{
BLOCKD *const d = xd->block + i;
const int row = i >> 2, col = i & 3;
B_PREDICTION_MODE m;
if (labelings[i] != which_label)
continue;
if (col && labelings[i] == labelings[i-1])
m = LEFT4X4;
else if (row && labelings[i] == labelings[i-4])
m = ABOVE4X4;
else
{
// the only time we should do costing for new motion vector or mode
// is when we are on a new label (jbb May 08, 2007)
switch (m = this_mode)
{
case NEW4X4 :
if (xd->mode_info_context->mbmi.second_ref_frame)
{
this_mv->as_int = seg_mvs[xd->mode_info_context->mbmi.ref_frame - 1].as_int;
this_second_mv->as_int = seg_mvs[xd->mode_info_context->mbmi.second_ref_frame - 1].as_int;
}
#if CONFIG_HIGH_PRECISION_MV
thismvcost = vp8_mv_bit_cost(this_mv, best_ref_mv, mvcost,
102, xd->allow_high_precision_mv);
if (xd->mode_info_context->mbmi.second_ref_frame)
{
thismvcost += vp8_mv_bit_cost(this_second_mv, second_best_ref_mv, mvcost,
102, xd->allow_high_precision_mv);
}
#else
thismvcost = vp8_mv_bit_cost(this_mv, best_ref_mv, mvcost, 102);
if (xd->mode_info_context->mbmi.second_ref_frame)
{
thismvcost += vp8_mv_bit_cost(this_second_mv, second_best_ref_mv, mvcost, 102);
}
#endif
break;
case LEFT4X4:
this_mv->as_int = col ? d[-1].bmi.as_mv.first.as_int : left_block_mv(mic, i);
if (xd->mode_info_context->mbmi.second_ref_frame)
this_second_mv->as_int = col ? d[-1].bmi.as_mv.second.as_int : left_block_second_mv(mic, i);
break;
case ABOVE4X4:
this_mv->as_int = row ? d[-4].bmi.as_mv.first.as_int : above_block_mv(mic, i, mis);
if (xd->mode_info_context->mbmi.second_ref_frame)
this_second_mv->as_int = row ? d[-4].bmi.as_mv.second.as_int : above_block_second_mv(mic, i, mis);
break;
case ZERO4X4:
this_mv->as_int = 0;
if (xd->mode_info_context->mbmi.second_ref_frame)
this_second_mv->as_int = 0;
break;
default:
break;
}
if (m == ABOVE4X4) // replace above with left if same
{
int_mv left_mv, left_second_mv;
1681168216831684168516861687168816891690169116921693169416951696169716981699170017011702170317041705170617071708170917101711171217131714171517161717171817191720172117221723172417251726172717281729173017311732173317341735173617371738173917401741174217431744174517461747174817491750
left_mv.as_int = col ? d[-1].bmi.as_mv.first.as_int :
left_block_mv(mic, i);
if (xd->mode_info_context->mbmi.second_ref_frame)
left_second_mv.as_int = col ? d[-1].bmi.as_mv.second.as_int :
left_block_second_mv(mic, i);
if (left_mv.as_int == this_mv->as_int &&
(!xd->mode_info_context->mbmi.second_ref_frame ||
left_second_mv.as_int == this_second_mv->as_int))
m = LEFT4X4;
}
cost = x->inter_bmode_costs[ m];
}
d->bmi.as_mv.first.as_int = this_mv->as_int;
if (xd->mode_info_context->mbmi.second_ref_frame)
d->bmi.as_mv.second.as_int = this_second_mv->as_int;
x->partition_info->bmi[i].mode = m;
x->partition_info->bmi[i].mv.as_int = this_mv->as_int;
if (xd->mode_info_context->mbmi.second_ref_frame)
x->partition_info->bmi[i].second_mv.as_int = this_second_mv->as_int;
}
while (++i < 16);
cost += thismvcost ;
return cost;
}
static int rdcost_mbsegment_y(MACROBLOCK *mb, const int *labels,
int which_label, ENTROPY_CONTEXT *ta,
ENTROPY_CONTEXT *tl)
{
int cost = 0;
int b;
MACROBLOCKD *x = &mb->e_mbd;
for (b = 0; b < 16; b++)
if (labels[ b] == which_label)
cost += cost_coeffs(mb, x->block + b, PLANE_TYPE_Y_WITH_DC,
ta + vp8_block2above[b],
tl + vp8_block2left[b]);
return cost;
}
static unsigned int vp8_encode_inter_mb_segment(
MACROBLOCK *x,
int const *labels,
int which_label,
const VP8_ENCODER_RTCD *rtcd)
{
int i;
unsigned int distortion = 0;
for (i = 0; i < 16; i++)
{
if (labels[i] == which_label)
{
BLOCKD *bd = &x->e_mbd.block[i];
BLOCK *be = &x->block[i];
int thisdistortion;
vp8_build_inter_predictors_b(bd, 16, x->e_mbd.subpixel_predict);
if (x->e_mbd.mode_info_context->mbmi.second_ref_frame)
vp8_build_2nd_inter_predictors_b(bd, 16, x->e_mbd.subpixel_predict_avg);
ENCODEMB_INVOKE(&rtcd->encodemb, subb)(be, bd, 16);
x->vp8_short_fdct4x4(be->src_diff, be->coeff, 32);
1751175217531754175517561757175817591760176117621763176417651766176717681769177017711772177317741775177617771778177917801781178217831784178517861787178817891790179117921793179417951796179717981799180018011802180318041805180618071808180918101811181218131814181518161817181818191820
// set to 0 no way to account for 2nd order DC so discount
//be->coeff[0] = 0;
x->quantize_b(be, bd);
thisdistortion = ENCODEMB_INVOKE(&rtcd->encodemb,berr)(
be->coeff,
bd->dqcoeff)/4;
distortion += thisdistortion;
}
}
return distortion;
}
static const unsigned int segmentation_to_sseshift[4] = {3, 3, 2, 0};
typedef struct
{
int_mv *ref_mv, *second_ref_mv;
int_mv mvp;
int segment_rd;
int segment_num;
int r;
int d;
int segment_yrate;
B_PREDICTION_MODE modes[16];
int_mv mvs[16], second_mvs[16];
unsigned char eobs[16];
int mvthresh;
int *mdcounts;
int_mv sv_mvp[4]; // save 4 mvp from 8x8
int sv_istep[2]; // save 2 initial step_param for 16x8/8x16
} BEST_SEG_INFO;
static void rd_check_segment(VP8_COMP *cpi, MACROBLOCK *x,
BEST_SEG_INFO *bsi, unsigned int segmentation,
int_mv seg_mvs[16 /* n_blocks */][MAX_REF_FRAMES - 1])
{
int i;
int const *labels;
int br = 0;
int bd = 0;
B_PREDICTION_MODE this_mode;
int label_count;
int this_segment_rd = 0;
int label_mv_thresh;
int rate = 0;
int sbr = 0;
int sbd = 0;
int segmentyrate = 0;
vp8_variance_fn_ptr_t *v_fn_ptr;
ENTROPY_CONTEXT_PLANES t_above, t_left;
ENTROPY_CONTEXT *ta;
ENTROPY_CONTEXT *tl;
ENTROPY_CONTEXT_PLANES t_above_b, t_left_b;
ENTROPY_CONTEXT *ta_b;
ENTROPY_CONTEXT *tl_b;
vpx_memcpy(&t_above, x->e_mbd.above_context, sizeof(ENTROPY_CONTEXT_PLANES));
vpx_memcpy(&t_left, x->e_mbd.left_context, sizeof(ENTROPY_CONTEXT_PLANES));
1821182218231824182518261827182818291830183118321833183418351836183718381839184018411842184318441845184618471848184918501851185218531854185518561857185818591860186118621863186418651866186718681869187018711872187318741875187618771878187918801881188218831884188518861887188818891890
ta = (ENTROPY_CONTEXT *)&t_above;
tl = (ENTROPY_CONTEXT *)&t_left;
ta_b = (ENTROPY_CONTEXT *)&t_above_b;
tl_b = (ENTROPY_CONTEXT *)&t_left_b;
br = 0;
bd = 0;
v_fn_ptr = &cpi->fn_ptr[segmentation];
labels = vp8_mbsplits[segmentation];
label_count = vp8_mbsplit_count[segmentation];
// 64 makes this threshold really big effectively
// making it so that we very rarely check mvs on
// segments. setting this to 1 would make mv thresh
// roughly equal to what it is for macroblocks
label_mv_thresh = 1 * bsi->mvthresh / label_count ;
// Segmentation method overheads
rate = vp8_cost_token(vp8_mbsplit_tree, vp8_mbsplit_probs, vp8_mbsplit_encodings + segmentation);
rate += vp8_cost_mv_ref(cpi, SPLITMV, bsi->mdcounts);
this_segment_rd += RDCOST(x->rdmult, x->rddiv, rate, 0);
br += rate;
for (i = 0; i < label_count; i++)
{
int_mv mode_mv[B_MODE_COUNT], second_mode_mv[B_MODE_COUNT];
int best_label_rd = INT_MAX;
B_PREDICTION_MODE mode_selected = ZERO4X4;
int bestlabelyrate = 0;
// search for the best motion vector on this segment
for (this_mode = LEFT4X4; this_mode <= NEW4X4 ; this_mode ++)
{
int this_rd;
int distortion;
int labelyrate;
ENTROPY_CONTEXT_PLANES t_above_s, t_left_s;
ENTROPY_CONTEXT *ta_s;
ENTROPY_CONTEXT *tl_s;
vpx_memcpy(&t_above_s, &t_above, sizeof(ENTROPY_CONTEXT_PLANES));
vpx_memcpy(&t_left_s, &t_left, sizeof(ENTROPY_CONTEXT_PLANES));
ta_s = (ENTROPY_CONTEXT *)&t_above_s;
tl_s = (ENTROPY_CONTEXT *)&t_left_s;
// motion search for newmv (single predictor case only)
if (!x->e_mbd.mode_info_context->mbmi.second_ref_frame && this_mode == NEW4X4)
{
int sseshift;
int num00;
int step_param = 0;
int further_steps;
int n;
int thissme;
int bestsme = INT_MAX;
int_mv temp_mv;
BLOCK *c;
BLOCKD *e;
// Is the best so far sufficiently good that we cant justify doing and new motion search.
if (best_label_rd < label_mv_thresh)
break;
if(cpi->compressor_speed)
{
if (segmentation == BLOCK_8X16 || segmentation == BLOCK_16X8)
{
1891189218931894189518961897189818991900190119021903190419051906190719081909191019111912191319141915191619171918191919201921192219231924192519261927192819291930193119321933193419351936193719381939194019411942194319441945194619471948194919501951195219531954195519561957195819591960
bsi->mvp.as_int = bsi->sv_mvp[i].as_int;
if (i==1 && segmentation == BLOCK_16X8)
bsi->mvp.as_int = bsi->sv_mvp[2].as_int;
step_param = bsi->sv_istep[i];
}
// use previous block's result as next block's MV predictor.
if (segmentation == BLOCK_4X4 && i>0)
{
bsi->mvp.as_int = x->e_mbd.block[i-1].bmi.as_mv.first.as_int;
if (i==4 || i==8 || i==12)
bsi->mvp.as_int = x->e_mbd.block[i-4].bmi.as_mv.first.as_int;
step_param = 2;
}
}
further_steps = (MAX_MVSEARCH_STEPS - 1) - step_param;
{
int sadpb = x->sadperbit4;
int_mv mvp_full;
mvp_full.as_mv.row = bsi->mvp.as_mv.row >>3;
mvp_full.as_mv.col = bsi->mvp.as_mv.col >>3;
// find first label
n = vp8_mbsplit_offset[segmentation][i];
c = &x->block[n];
e = &x->e_mbd.block[n];
{
bestsme = cpi->diamond_search_sad(x, c, e, &mvp_full,
&mode_mv[NEW4X4], step_param,
sadpb, &num00, v_fn_ptr,
XMVCOST,
bsi->ref_mv);
n = num00;
num00 = 0;
while (n < further_steps)
{
n++;
if (num00)
num00--;
else
{
thissme = cpi->diamond_search_sad(x, c, e,
&mvp_full, &temp_mv,
step_param + n, sadpb,
&num00, v_fn_ptr,
XMVCOST,
bsi->ref_mv);
if (thissme < bestsme)
{
bestsme = thissme;
mode_mv[NEW4X4].as_int = temp_mv.as_int;
}
}
}
}
sseshift = segmentation_to_sseshift[segmentation];
// Should we do a full search (best quality only)
if ((cpi->compressor_speed == 0) && (bestsme >> sseshift) > 4000)
1961196219631964196519661967196819691970197119721973197419751976197719781979198019811982198319841985198619871988198919901991199219931994199519961997199819992000200120022003200420052006200720082009201020112012201320142015201620172018201920202021202220232024202520262027202820292030
{
/* Check if mvp_full is within the range. */
vp8_clamp_mv(&mvp_full, x->mv_col_min, x->mv_col_max, x->mv_row_min, x->mv_row_max);
thissme = cpi->full_search_sad(x, c, e, &mvp_full,
sadpb, 16, v_fn_ptr,
XMVCOST, bsi->ref_mv);
if (thissme < bestsme)
{
bestsme = thissme;
mode_mv[NEW4X4].as_int = e->bmi.as_mv.first.as_int;
}
else
{
// The full search result is actually worse so re-instate the previous best vector
e->bmi.as_mv.first.as_int = mode_mv[NEW4X4].as_int;
}
}
}
if (bestsme < INT_MAX)
{
int distortion;
unsigned int sse;
cpi->find_fractional_mv_step(x, c, e, &mode_mv[NEW4X4],
bsi->ref_mv, x->errorperbit, v_fn_ptr, XMVCOST,
&distortion, &sse);
// safe motion search result for use in compound prediction
seg_mvs[i][x->e_mbd.mode_info_context->mbmi.ref_frame - 1].as_int = mode_mv[NEW4X4].as_int;
}
} /* NEW4X4 */
else if (x->e_mbd.mode_info_context->mbmi.second_ref_frame && this_mode == NEW4X4)
{
// motion search not completed? Then skip newmv for this block with comppred
if (seg_mvs[i][x->e_mbd.mode_info_context->mbmi.second_ref_frame - 1].as_int == INVALID_MV ||
seg_mvs[i][x->e_mbd.mode_info_context->mbmi.ref_frame - 1].as_int == INVALID_MV)
{
continue;
}
}
rate = labels2mode(x, labels, i, this_mode, &mode_mv[this_mode],
&second_mode_mv[this_mode], seg_mvs[i], bsi->ref_mv, bsi->second_ref_mv, XMVCOST);
// Trap vectors that reach beyond the UMV borders
if (((mode_mv[this_mode].as_mv.row >> 3) < x->mv_row_min) || ((mode_mv[this_mode].as_mv.row >> 3) > x->mv_row_max) ||
((mode_mv[this_mode].as_mv.col >> 3) < x->mv_col_min) || ((mode_mv[this_mode].as_mv.col >> 3) > x->mv_col_max))
{
continue;
}
if (x->e_mbd.mode_info_context->mbmi.second_ref_frame)
{
if (((second_mode_mv[this_mode].as_mv.row >> 3) < x->mv_row_min) ||
((second_mode_mv[this_mode].as_mv.row >> 3) > x->mv_row_max) ||
((second_mode_mv[this_mode].as_mv.col >> 3) < x->mv_col_min) ||
((second_mode_mv[this_mode].as_mv.col >> 3) > x->mv_col_max))
{
continue;
}
}
distortion = vp8_encode_inter_mb_segment(
x, labels, i,
IF_RTCD(&cpi->rtcd));
labelyrate = rdcost_mbsegment_y(x, labels, i, ta_s, tl_s);
rate += labelyrate;
2031203220332034203520362037203820392040204120422043204420452046204720482049205020512052205320542055205620572058205920602061206220632064206520662067206820692070207120722073207420752076207720782079208020812082208320842085208620872088208920902091209220932094209520962097209820992100
this_rd = RDCOST(x->rdmult, x->rddiv, rate, distortion);
if (this_rd < best_label_rd)
{
sbr = rate;
sbd = distortion;
bestlabelyrate = labelyrate;
mode_selected = this_mode;
best_label_rd = this_rd;
vpx_memcpy(ta_b, ta_s, sizeof(ENTROPY_CONTEXT_PLANES));
vpx_memcpy(tl_b, tl_s, sizeof(ENTROPY_CONTEXT_PLANES));
}
} /*for each 4x4 mode*/
vpx_memcpy(ta, ta_b, sizeof(ENTROPY_CONTEXT_PLANES));
vpx_memcpy(tl, tl_b, sizeof(ENTROPY_CONTEXT_PLANES));
labels2mode(x, labels, i, mode_selected, &mode_mv[mode_selected],
&second_mode_mv[mode_selected], seg_mvs[i], bsi->ref_mv, bsi->second_ref_mv, XMVCOST);
br += sbr;
bd += sbd;
segmentyrate += bestlabelyrate;
this_segment_rd += best_label_rd;
if (this_segment_rd >= bsi->segment_rd) {
break;
}
} /* for each label */
if (this_segment_rd < bsi->segment_rd)
{
bsi->r = br;
bsi->d = bd;
bsi->segment_yrate = segmentyrate;
bsi->segment_rd = this_segment_rd;
bsi->segment_num = segmentation;
// store everything needed to come back to this!!
for (i = 0; i < 16; i++)
{
BLOCKD *bd = &x->e_mbd.block[i];
bsi->mvs[i].as_mv = x->partition_info->bmi[i].mv.as_mv;
if (x->e_mbd.mode_info_context->mbmi.second_ref_frame)
bsi->second_mvs[i].as_mv = x->partition_info->bmi[i].second_mv.as_mv;
bsi->modes[i] = x->partition_info->bmi[i].mode;
bsi->eobs[i] = bd->eob;
}
}
}
static __inline
void vp8_cal_step_param(int sr, int *sp)
{
int step = 0;
if (sr > MAX_FIRST_STEP) sr = MAX_FIRST_STEP;
else if (sr < 1) sr = 1;
while (sr>>=1)
step++;
*sp = MAX_MVSEARCH_STEPS - 1 - step;
}
2101210221032104210521062107210821092110211121122113211421152116211721182119212021212122212321242125212621272128212921302131213221332134213521362137213821392140214121422143214421452146214721482149215021512152215321542155215621572158215921602161216221632164216521662167216821692170
static int vp8_rd_pick_best_mbsegmentation(VP8_COMP *cpi, MACROBLOCK *x,
int_mv *best_ref_mv, int_mv *second_best_ref_mv, int best_rd,
int *mdcounts, int *returntotrate,
int *returnyrate, int *returndistortion,
int mvthresh,
int_mv seg_mvs[BLOCK_MAX_SEGMENTS - 1][16 /* n_blocks */][MAX_REF_FRAMES - 1])
{
int i;
BEST_SEG_INFO bsi;
vpx_memset(&bsi, 0, sizeof(bsi));
bsi.segment_rd = best_rd;
bsi.ref_mv = best_ref_mv;
bsi.second_ref_mv = second_best_ref_mv;
bsi.mvp.as_int = best_ref_mv->as_int;
bsi.mvthresh = mvthresh;
bsi.mdcounts = mdcounts;
for(i = 0; i < 16; i++)
{
bsi.modes[i] = ZERO4X4;
}
if(cpi->compressor_speed == 0)
{
/* for now, we will keep the original segmentation order
when in best quality mode */
rd_check_segment(cpi, x, &bsi, BLOCK_16X8, seg_mvs[BLOCK_16X8]);
rd_check_segment(cpi, x, &bsi, BLOCK_8X16, seg_mvs[BLOCK_8X16]);
rd_check_segment(cpi, x, &bsi, BLOCK_8X8, seg_mvs[BLOCK_8X8]);
rd_check_segment(cpi, x, &bsi, BLOCK_4X4, seg_mvs[BLOCK_4X4]);
}
else
{
int sr;
rd_check_segment(cpi, x, &bsi, BLOCK_8X8, seg_mvs[BLOCK_8X8]);
if (bsi.segment_rd < best_rd)
{
int col_min = (best_ref_mv->as_mv.col>>3) - MAX_FULL_PEL_VAL + ((best_ref_mv->as_mv.col & 7)?1:0);
int row_min = (best_ref_mv->as_mv.row>>3) - MAX_FULL_PEL_VAL + ((best_ref_mv->as_mv.row & 7)?1:0);
int col_max = (best_ref_mv->as_mv.col>>3) + MAX_FULL_PEL_VAL;
int row_max = (best_ref_mv->as_mv.row>>3) + MAX_FULL_PEL_VAL;
int tmp_col_min = x->mv_col_min;
int tmp_col_max = x->mv_col_max;
int tmp_row_min = x->mv_row_min;
int tmp_row_max = x->mv_row_max;
/* Get intersection of UMV window and valid MV window to reduce # of checks in diamond search. */
if (x->mv_col_min < col_min )
x->mv_col_min = col_min;
if (x->mv_col_max > col_max )
x->mv_col_max = col_max;
if (x->mv_row_min < row_min )
x->mv_row_min = row_min;
if (x->mv_row_max > row_max )
x->mv_row_max = row_max;
/* Get 8x8 result */
bsi.sv_mvp[0].as_int = bsi.mvs[0].as_int;
bsi.sv_mvp[1].as_int = bsi.mvs[2].as_int;
bsi.sv_mvp[2].as_int = bsi.mvs[8].as_int;
bsi.sv_mvp[3].as_int = bsi.mvs[10].as_int;
/* Use 8x8 result as 16x8/8x16's predictor MV. Adjust search range according to the closeness of 2 MV. */
/* block 8X16 */
2171217221732174217521762177217821792180218121822183218421852186218721882189219021912192219321942195219621972198219922002201220222032204220522062207220822092210221122122213221422152216221722182219222022212222222322242225222622272228222922302231223222332234223522362237223822392240
{
sr = MAXF((abs(bsi.sv_mvp[0].as_mv.row - bsi.sv_mvp[2].as_mv.row))>>3, (abs(bsi.sv_mvp[0].as_mv.col - bsi.sv_mvp[2].as_mv.col))>>3);
vp8_cal_step_param(sr, &bsi.sv_istep[0]);
sr = MAXF((abs(bsi.sv_mvp[1].as_mv.row - bsi.sv_mvp[3].as_mv.row))>>3, (abs(bsi.sv_mvp[1].as_mv.col - bsi.sv_mvp[3].as_mv.col))>>3);
vp8_cal_step_param(sr, &bsi.sv_istep[1]);
rd_check_segment(cpi, x, &bsi, BLOCK_8X16, seg_mvs[BLOCK_8X16]);
}
/* block 16X8 */
{
sr = MAXF((abs(bsi.sv_mvp[0].as_mv.row - bsi.sv_mvp[1].as_mv.row))>>3, (abs(bsi.sv_mvp[0].as_mv.col - bsi.sv_mvp[1].as_mv.col))>>3);
vp8_cal_step_param(sr, &bsi.sv_istep[0]);
sr = MAXF((abs(bsi.sv_mvp[2].as_mv.row - bsi.sv_mvp[3].as_mv.row))>>3, (abs(bsi.sv_mvp[2].as_mv.col - bsi.sv_mvp[3].as_mv.col))>>3);
vp8_cal_step_param(sr, &bsi.sv_istep[1]);
rd_check_segment(cpi, x, &bsi, BLOCK_16X8, seg_mvs[BLOCK_16X8]);
}
/* If 8x8 is better than 16x8/8x16, then do 4x4 search */
/* Not skip 4x4 if speed=0 (good quality) */
if (cpi->sf.no_skip_block4x4_search || bsi.segment_num == BLOCK_8X8) /* || (sv_segment_rd8x8-bsi.segment_rd) < sv_segment_rd8x8>>5) */
{
bsi.mvp.as_int = bsi.sv_mvp[0].as_int;
rd_check_segment(cpi, x, &bsi, BLOCK_4X4, seg_mvs[BLOCK_4X4]);
}
/* restore UMV window */
x->mv_col_min = tmp_col_min;
x->mv_col_max = tmp_col_max;
x->mv_row_min = tmp_row_min;
x->mv_row_max = tmp_row_max;
}
}
/* set it to the best */
for (i = 0; i < 16; i++)
{
BLOCKD *bd = &x->e_mbd.block[i];
bd->bmi.as_mv.first.as_int = bsi.mvs[i].as_int;
if (x->e_mbd.mode_info_context->mbmi.second_ref_frame)
bd->bmi.as_mv.second.as_int = bsi.second_mvs[i].as_int;
bd->eob = bsi.eobs[i];
}
*returntotrate = bsi.r;
*returndistortion = bsi.d;
*returnyrate = bsi.segment_yrate;
/* save partitions */
x->e_mbd.mode_info_context->mbmi.partitioning = bsi.segment_num;
x->partition_info->count = vp8_mbsplit_count[bsi.segment_num];
for (i = 0; i < x->partition_info->count; i++)
{
int j;
j = vp8_mbsplit_offset[bsi.segment_num][i];
x->partition_info->bmi[i].mode = bsi.modes[j];
x->partition_info->bmi[i].mv.as_mv = bsi.mvs[j].as_mv;
if (x->e_mbd.mode_info_context->mbmi.second_ref_frame)
x->partition_info->bmi[i].second_mv.as_mv = bsi.second_mvs[j].as_mv;
}
/*
* used to set x->e_mbd.mode_info_context->mbmi.mv.as_int
*/
2241224222432244224522462247224822492250225122522253225422552256225722582259226022612262226322642265226622672268226922702271227222732274227522762277227822792280228122822283228422852286228722882289229022912292229322942295229622972298229923002301230223032304230523062307230823092310
x->partition_info->bmi[15].mv.as_int = bsi.mvs[15].as_int;
if (x->e_mbd.mode_info_context->mbmi.second_ref_frame)
x->partition_info->bmi[15].second_mv.as_int = bsi.second_mvs[15].as_int;
return bsi.segment_rd;
}
/* Order arr in increasing order, original position stored in idx */
static void insertsortmv(int arr[], int len)
{
int i, j, k;
for ( i = 1 ; i <= len-1 ; i++ )
{
for ( j = 0 ; j < i ; j++ )
{
if ( arr[j] > arr[i] )
{
int temp;
temp = arr[i];
for ( k = i; k >j; k--)
arr[k] = arr[k - 1] ;
arr[j] = temp ;
}
}
}
}
static void insertsortsad(int arr[],int idx[], int len)
{
int i, j, k;
for ( i = 1 ; i <= len-1 ; i++ )
{
for ( j = 0 ; j < i ; j++ )
{
if ( arr[j] > arr[i] )
{
int temp, tempi;
temp = arr[i];
tempi = idx[i];
for ( k = i; k >j; k--)
{
arr[k] = arr[k - 1] ;
idx[k] = idx[k - 1];
}
arr[j] = temp ;
idx[j] = tempi;
}
}
}
}
//The improved MV prediction
void vp8_mv_pred
(
VP8_COMP *cpi,
MACROBLOCKD *xd,
const MODE_INFO *here,
int_mv *mvp,
int refframe,
int *ref_frame_sign_bias,
int *sr,
int near_sadidx[]
2311231223132314231523162317231823192320232123222323232423252326232723282329233023312332233323342335233623372338233923402341234223432344234523462347234823492350235123522353235423552356235723582359236023612362236323642365236623672368236923702371237223732374237523762377237823792380
)
{
const MODE_INFO *above = here - xd->mode_info_stride;
const MODE_INFO *left = here - 1;
const MODE_INFO *aboveleft = above - 1;
int_mv near_mvs[8];
int near_ref[8];
int_mv mv;
int vcnt=0;
int find=0;
int mb_offset;
int mvx[8];
int mvy[8];
int i;
mv.as_int = 0;
if(here->mbmi.ref_frame != INTRA_FRAME)
{
near_mvs[0].as_int = near_mvs[1].as_int = near_mvs[2].as_int = near_mvs[3].as_int = near_mvs[4].as_int = near_mvs[5].as_int = near_mvs[6].as_int = near_mvs[7].as_int = 0;
near_ref[0] = near_ref[1] = near_ref[2] = near_ref[3] = near_ref[4] = near_ref[5] = near_ref[6] = near_ref[7] = 0;
// read in 3 nearby block's MVs from current frame as prediction candidates.
if (above->mbmi.ref_frame != INTRA_FRAME)
{
near_mvs[vcnt].as_int = above->mbmi.mv.as_int;
mv_bias(ref_frame_sign_bias[above->mbmi.ref_frame], refframe, &near_mvs[vcnt], ref_frame_sign_bias);
near_ref[vcnt] = above->mbmi.ref_frame;
}
vcnt++;
if (left->mbmi.ref_frame != INTRA_FRAME)
{
near_mvs[vcnt].as_int = left->mbmi.mv.as_int;
mv_bias(ref_frame_sign_bias[left->mbmi.ref_frame], refframe, &near_mvs[vcnt], ref_frame_sign_bias);
near_ref[vcnt] = left->mbmi.ref_frame;
}
vcnt++;
if (aboveleft->mbmi.ref_frame != INTRA_FRAME)
{
near_mvs[vcnt].as_int = aboveleft->mbmi.mv.as_int;
mv_bias(ref_frame_sign_bias[aboveleft->mbmi.ref_frame], refframe, &near_mvs[vcnt], ref_frame_sign_bias);
near_ref[vcnt] = aboveleft->mbmi.ref_frame;
}
vcnt++;
// read in 5 nearby block's MVs from last frame.
if(cpi->common.last_frame_type != KEY_FRAME)
{
mb_offset = (-xd->mb_to_top_edge/128 + 1) * (xd->mode_info_stride +1) + (-xd->mb_to_left_edge/128 +1) ;
// current in last frame
if (cpi->lf_ref_frame[mb_offset] != INTRA_FRAME)
{
near_mvs[vcnt].as_int = cpi->lfmv[mb_offset].as_int;
mv_bias(cpi->lf_ref_frame_sign_bias[mb_offset], refframe, &near_mvs[vcnt], ref_frame_sign_bias);
near_ref[vcnt] = cpi->lf_ref_frame[mb_offset];
}
vcnt++;
// above in last frame
if (cpi->lf_ref_frame[mb_offset - xd->mode_info_stride-1] != INTRA_FRAME)
{
near_mvs[vcnt].as_int = cpi->lfmv[mb_offset - xd->mode_info_stride-1].as_int;
mv_bias(cpi->lf_ref_frame_sign_bias[mb_offset - xd->mode_info_stride-1], refframe, &near_mvs[vcnt], ref_frame_sign_bias);
near_ref[vcnt] = cpi->lf_ref_frame[mb_offset - xd->mode_info_stride-1];
}
vcnt++;
// left in last frame
2381238223832384238523862387238823892390239123922393239423952396239723982399240024012402240324042405240624072408240924102411241224132414241524162417241824192420242124222423242424252426242724282429243024312432243324342435243624372438243924402441244224432444244524462447244824492450
if (cpi->lf_ref_frame[mb_offset-1] != INTRA_FRAME)
{
near_mvs[vcnt].as_int = cpi->lfmv[mb_offset -1].as_int;
mv_bias(cpi->lf_ref_frame_sign_bias[mb_offset -1], refframe, &near_mvs[vcnt], ref_frame_sign_bias);
near_ref[vcnt] = cpi->lf_ref_frame[mb_offset - 1];
}
vcnt++;
// right in last frame
if (cpi->lf_ref_frame[mb_offset +1] != INTRA_FRAME)
{
near_mvs[vcnt].as_int = cpi->lfmv[mb_offset +1].as_int;
mv_bias(cpi->lf_ref_frame_sign_bias[mb_offset +1], refframe, &near_mvs[vcnt], ref_frame_sign_bias);
near_ref[vcnt] = cpi->lf_ref_frame[mb_offset +1];
}
vcnt++;
// below in last frame
if (cpi->lf_ref_frame[mb_offset + xd->mode_info_stride +1] != INTRA_FRAME)
{
near_mvs[vcnt].as_int = cpi->lfmv[mb_offset + xd->mode_info_stride +1].as_int;
mv_bias(cpi->lf_ref_frame_sign_bias[mb_offset + xd->mode_info_stride +1], refframe, &near_mvs[vcnt], ref_frame_sign_bias);
near_ref[vcnt] = cpi->lf_ref_frame[mb_offset + xd->mode_info_stride +1];
}
vcnt++;
}
for(i=0; i< vcnt; i++)
{
if(near_ref[near_sadidx[i]] != INTRA_FRAME)
{
if(here->mbmi.ref_frame == near_ref[near_sadidx[i]])
{
mv.as_int = near_mvs[near_sadidx[i]].as_int;
find = 1;
if (i < 3)
*sr = 3;
else
*sr = 2;
break;
}
}
}
if(!find)
{
for(i=0; i<vcnt; i++)
{
mvx[i] = near_mvs[i].as_mv.row;
mvy[i] = near_mvs[i].as_mv.col;
}
insertsortmv(mvx, vcnt);
insertsortmv(mvy, vcnt);
mv.as_mv.row = mvx[vcnt/2];
mv.as_mv.col = mvy[vcnt/2];
find = 1;
//sr is set to 0 to allow calling function to decide the search range.
*sr = 0;
}
}
/* Set up return values */
mvp->as_int = mv.as_int;
vp8_clamp_mv2(mvp, xd);
}
void vp8_cal_sad(VP8_COMP *cpi, MACROBLOCKD *xd, MACROBLOCK *x, int recon_yoffset, int near_sadidx[])
{
2451245224532454245524562457245824592460246124622463246424652466246724682469247024712472247324742475247624772478247924802481248224832484248524862487248824892490249124922493249424952496249724982499250025012502250325042505250625072508250925102511251225132514251525162517251825192520
int near_sad[8] = {0}; // 0-cf above, 1-cf left, 2-cf aboveleft, 3-lf current, 4-lf above, 5-lf left, 6-lf right, 7-lf below
BLOCK *b = &x->block[0];
unsigned char *src_y_ptr = *(b->base_src);
//calculate sad for current frame 3 nearby MBs.
if( xd->mb_to_top_edge==0 && xd->mb_to_left_edge ==0)
{
near_sad[0] = near_sad[1] = near_sad[2] = INT_MAX;
}else if(xd->mb_to_top_edge==0)
{ //only has left MB for sad calculation.
near_sad[0] = near_sad[2] = INT_MAX;
near_sad[1] = cpi->fn_ptr[BLOCK_16X16].sdf(src_y_ptr, b->src_stride, xd->dst.y_buffer - 16,xd->dst.y_stride, 0x7fffffff);
}else if(xd->mb_to_left_edge ==0)
{ //only has left MB for sad calculation.
near_sad[1] = near_sad[2] = INT_MAX;
near_sad[0] = cpi->fn_ptr[BLOCK_16X16].sdf(src_y_ptr, b->src_stride, xd->dst.y_buffer - xd->dst.y_stride *16,xd->dst.y_stride, 0x7fffffff);
}else
{
near_sad[0] = cpi->fn_ptr[BLOCK_16X16].sdf(src_y_ptr, b->src_stride, xd->dst.y_buffer - xd->dst.y_stride *16,xd->dst.y_stride, 0x7fffffff);
near_sad[1] = cpi->fn_ptr[BLOCK_16X16].sdf(src_y_ptr, b->src_stride, xd->dst.y_buffer - 16,xd->dst.y_stride, 0x7fffffff);
near_sad[2] = cpi->fn_ptr[BLOCK_16X16].sdf(src_y_ptr, b->src_stride, xd->dst.y_buffer - xd->dst.y_stride *16 -16,xd->dst.y_stride, 0x7fffffff);
}
if(cpi->common.last_frame_type != KEY_FRAME)
{
//calculate sad for last frame 5 nearby MBs.
unsigned char *pre_y_buffer = cpi->common.yv12_fb[cpi->common.lst_fb_idx].y_buffer + recon_yoffset;
int pre_y_stride = cpi->common.yv12_fb[cpi->common.lst_fb_idx].y_stride;
if(xd->mb_to_top_edge==0) near_sad[4] = INT_MAX;
if(xd->mb_to_left_edge ==0) near_sad[5] = INT_MAX;
if(xd->mb_to_right_edge ==0) near_sad[6] = INT_MAX;
if(xd->mb_to_bottom_edge==0) near_sad[7] = INT_MAX;
if(near_sad[4] != INT_MAX)
near_sad[4] = cpi->fn_ptr[BLOCK_16X16].sdf(src_y_ptr, b->src_stride, pre_y_buffer - pre_y_stride *16, pre_y_stride, 0x7fffffff);
if(near_sad[5] != INT_MAX)
near_sad[5] = cpi->fn_ptr[BLOCK_16X16].sdf(src_y_ptr, b->src_stride, pre_y_buffer - 16, pre_y_stride, 0x7fffffff);
near_sad[3] = cpi->fn_ptr[BLOCK_16X16].sdf(src_y_ptr, b->src_stride, pre_y_buffer, pre_y_stride, 0x7fffffff);
if(near_sad[6] != INT_MAX)
near_sad[6] = cpi->fn_ptr[BLOCK_16X16].sdf(src_y_ptr, b->src_stride, pre_y_buffer + 16, pre_y_stride, 0x7fffffff);
if(near_sad[7] != INT_MAX)
near_sad[7] = cpi->fn_ptr[BLOCK_16X16].sdf(src_y_ptr, b->src_stride, pre_y_buffer + pre_y_stride *16, pre_y_stride, 0x7fffffff);
}
if(cpi->common.last_frame_type != KEY_FRAME)
{
insertsortsad(near_sad, near_sadidx, 8);
}else
{
insertsortsad(near_sad, near_sadidx, 3);
}
}
/*static */void rd_update_mvcount(VP8_COMP *cpi, MACROBLOCK *x, int_mv *best_ref_mv)
{
if (x->e_mbd.mode_info_context->mbmi.mode == SPLITMV)
{
int i;
for (i = 0; i < x->partition_info->count; i++)
{
if (x->partition_info->bmi[i].mode == NEW4X4)
{
#if CONFIG_HIGH_PRECISION_MV
if (x->e_mbd.allow_high_precision_mv)
{
cpi->MVcount_hp[0][mv_max_hp+(x->partition_info->bmi[i].mv.as_mv.row
- best_ref_mv->as_mv.row)]++;
2521252225232524252525262527252825292530253125322533253425352536253725382539254025412542254325442545254625472548254925502551255225532554255525562557255825592560256125622563256425652566256725682569257025712572257325742575257625772578257925802581258225832584258525862587258825892590
cpi->MVcount_hp[1][mv_max_hp+(x->partition_info->bmi[i].mv.as_mv.col
- best_ref_mv->as_mv.col)]++;
}
else
#endif
{
cpi->MVcount[0][mv_max+((x->partition_info->bmi[i].mv.as_mv.row
- best_ref_mv->as_mv.row) >> 1)]++;
cpi->MVcount[1][mv_max+((x->partition_info->bmi[i].mv.as_mv.col
- best_ref_mv->as_mv.col) >> 1)]++;
}
}
}
}
else if (x->e_mbd.mode_info_context->mbmi.mode == NEWMV)
{
#if CONFIG_HIGH_PRECISION_MV
if (x->e_mbd.allow_high_precision_mv)
{
cpi->MVcount_hp[0][mv_max_hp+(x->e_mbd.mode_info_context->mbmi.mv.as_mv.row
- best_ref_mv->as_mv.row)]++;
cpi->MVcount_hp[1][mv_max_hp+(x->e_mbd.mode_info_context->mbmi.mv.as_mv.col
- best_ref_mv->as_mv.col)]++;
}
else
#endif
{
cpi->MVcount[0][mv_max+((x->e_mbd.mode_info_context->mbmi.mv.as_mv.row
- best_ref_mv->as_mv.row) >> 1)]++;
cpi->MVcount[1][mv_max+((x->e_mbd.mode_info_context->mbmi.mv.as_mv.col
- best_ref_mv->as_mv.col) >> 1)]++;
}
}
}
static void set_i8x8_block_modes(MACROBLOCK *x, int modes[2][4])
{
int i;
MACROBLOCKD *xd = &x->e_mbd;
for(i=0;i<4;i++)
{
int ib = vp8_i8x8_block[i];
x->e_mbd.mode_info_context->bmi[ib+0].as_mode.first= modes[0][i];
x->e_mbd.mode_info_context->bmi[ib+1].as_mode.first= modes[0][i];
x->e_mbd.mode_info_context->bmi[ib+4].as_mode.first= modes[0][i];
x->e_mbd.mode_info_context->bmi[ib+5].as_mode.first= modes[0][i];
#if CONFIG_COMP_INTRA_PRED
x->e_mbd.mode_info_context->bmi[ib+0].as_mode.second= modes[1][i];
x->e_mbd.mode_info_context->bmi[ib+1].as_mode.second= modes[1][i];
x->e_mbd.mode_info_context->bmi[ib+4].as_mode.second= modes[1][i];
x->e_mbd.mode_info_context->bmi[ib+5].as_mode.second= modes[1][i];
#endif
//printf("%d,%d,%d,%d %d,%d,%d,%d\n",
// modes[0][0], modes[0][1], modes[0][2], modes[0][3],
// modes[1][0], modes[1][1], modes[1][2], modes[1][3]);
}
for (i = 0; i < 16; i++)
{
xd->block[i].bmi = xd->mode_info_context->bmi[i];
}
}
extern void calc_ref_probs( int * count, vp8_prob * probs );
static void estimate_curframe_refprobs(VP8_COMP *cpi, vp8_prob mod_refprobs[3], int pred_ref)
{
int norm_cnt[MAX_REF_FRAMES];
const int *const rfct = cpi->count_mb_ref_frame_usage;
int intra_count = rfct[INTRA_FRAME];
int last_count = rfct[LAST_FRAME];
2591259225932594259525962597259825992600260126022603260426052606260726082609261026112612261326142615261626172618261926202621262226232624262526262627262826292630263126322633263426352636263726382639264026412642264326442645264626472648264926502651265226532654265526562657265826592660
int gf_count = rfct[GOLDEN_FRAME];
int arf_count = rfct[ALTREF_FRAME];
// Work out modified reference frame probabilities to use where prediction
// of the reference frame fails
if (pred_ref == INTRA_FRAME)
{
norm_cnt[0] = 0;
norm_cnt[1] = last_count;
norm_cnt[2] = gf_count;
norm_cnt[3] = arf_count;
calc_ref_probs( norm_cnt, mod_refprobs );
mod_refprobs[0] = 0; // This branch implicit
}
else if (pred_ref == LAST_FRAME)
{
norm_cnt[0] = intra_count;
norm_cnt[1] = 0;
norm_cnt[2] = gf_count;
norm_cnt[3] = arf_count;
calc_ref_probs( norm_cnt, mod_refprobs);
mod_refprobs[1] = 0; // This branch implicit
}
else if (pred_ref == GOLDEN_FRAME)
{
norm_cnt[0] = intra_count;
norm_cnt[1] = last_count;
norm_cnt[2] = 0;
norm_cnt[3] = arf_count;
calc_ref_probs( norm_cnt, mod_refprobs );
mod_refprobs[2] = 0; // This branch implicit
}
else
{
norm_cnt[0] = intra_count;
norm_cnt[1] = last_count;
norm_cnt[2] = gf_count;
norm_cnt[3] = 0;
calc_ref_probs( norm_cnt, mod_refprobs );
mod_refprobs[2] = 0; // This branch implicit
}
}
static __inline unsigned weighted_cost(vp8_prob *tab0, vp8_prob *tab1, int idx, int val, int weight)
{
unsigned cost0 = tab0[idx] ? vp8_cost_bit(tab0[idx], val) : 0;
unsigned cost1 = tab1[idx] ? vp8_cost_bit(tab1[idx], val) : 0;
// weight is 16-bit fixed point, so this basically calculates:
// 0.5 + weight * cost1 + (1.0 - weight) * cost0
return (0x8000 + weight * cost1 + (0x10000 - weight) * cost0) >> 16;
}
static void vp8_estimate_ref_frame_costs(VP8_COMP *cpi, int segment_id, unsigned int * ref_costs )
{
VP8_COMMON *cm = &cpi->common;
MACROBLOCKD *xd = &cpi->mb.e_mbd;
vp8_prob * mod_refprobs;
unsigned int cost;
int pred_ref ;
int pred_flag;
int pred_ctx ;
int i;
int tot_count;
vp8_prob pred_prob, new_pred_prob;
int seg_ref_active;
int seg_ref_count = 0;
seg_ref_active = segfeature_active( xd,
segment_id,
2661266226632664266526662667266826692670267126722673267426752676267726782679268026812682268326842685268626872688268926902691269226932694269526962697269826992700270127022703270427052706270727082709271027112712271327142715271627172718271927202721272227232724272527262727272827292730
SEG_LVL_REF_FRAME );
if ( seg_ref_active )
{
seg_ref_count = check_segref( xd, segment_id, INTRA_FRAME ) +
check_segref( xd, segment_id, LAST_FRAME ) +
check_segref( xd, segment_id, GOLDEN_FRAME ) +
check_segref( xd, segment_id, ALTREF_FRAME );
}
// Get the predicted reference for this mb
pred_ref = get_pred_ref( cm, xd );
// Get the context probability for the prediction flag (based on last frame)
pred_prob = get_pred_prob( cm, xd, PRED_REF );
// Predict probability for current frame based on stats so far
pred_ctx = get_pred_context(cm, xd, PRED_REF);
tot_count = cpi->ref_pred_count[pred_ctx][0] + cpi->ref_pred_count[pred_ctx][1];
if ( tot_count )
{
new_pred_prob =
( cpi->ref_pred_count[pred_ctx][0] * 255 + (tot_count >> 1)) / tot_count;
new_pred_prob += !new_pred_prob;
}
else
new_pred_prob = 128;
// Get the set of probabilities to use if prediction fails
mod_refprobs = cm->mod_refprobs[pred_ref];
// For each possible selected reference frame work out a cost.
for ( i = 0; i < MAX_REF_FRAMES; i++ )
{
if (seg_ref_active && seg_ref_count == 1)
{
cost = 0;
}
else
{
pred_flag = (i == pred_ref);
// Get the prediction for the current mb
cost = weighted_cost(&pred_prob, &new_pred_prob, 0,
pred_flag, cpi->seg0_progress);
if (cost > 1024) cost = 768; // i.e. account for 4 bits max.
// for incorrectly predicted cases
if ( ! pred_flag )
{
vp8_prob curframe_mod_refprobs[3];
if (cpi->seg0_progress)
{
estimate_curframe_refprobs(cpi, curframe_mod_refprobs, pred_ref);
}
else
{
vpx_memset(curframe_mod_refprobs, 0, sizeof(curframe_mod_refprobs));
}
cost += weighted_cost(mod_refprobs, curframe_mod_refprobs, 0,
(i != INTRA_FRAME), cpi->seg0_progress);
if (i != INTRA_FRAME)
{
cost += weighted_cost(mod_refprobs, curframe_mod_refprobs, 1,
(i != LAST_FRAME), cpi->seg0_progress);
if (i != LAST_FRAME)
{
cost += weighted_cost(mod_refprobs, curframe_mod_refprobs, 2,
2731273227332734273527362737273827392740274127422743274427452746274727482749275027512752275327542755275627572758275927602761276227632764276527662767276827692770277127722773277427752776277727782779278027812782278327842785278627872788278927902791279227932794279527962797279827992800
(i != GOLDEN_FRAME), cpi->seg0_progress);
}
}
}
}
ref_costs[i] = cost;
}
}
static void store_coding_context (MACROBLOCK *x, int mb_index,
int mode_index,
PARTITION_INFO *partition,
int_mv *ref_mv)
{
MACROBLOCKD *xd = &x->e_mbd;
// Take a snapshot of the coding context so it can be
// restored if we decide to encode this way
x->mb_context[mb_index].best_mode_index = mode_index;
vpx_memcpy(&x->mb_context[mb_index].mic, xd->mode_info_context,
sizeof(MODE_INFO));
vpx_memcpy(&x->mb_context[mb_index].partition_info, partition,
sizeof(PARTITION_INFO));
x->mb_context[mb_index].best_ref_mv.as_int = ref_mv->as_int;
//x->mb_context[mb_index].rddiv = x->rddiv;
//x->mb_context[mb_index].rdmult = x->rdmult;
}
void vp8_rd_pick_inter_mode(VP8_COMP *cpi, MACROBLOCK *x, int recon_yoffset, int recon_uvoffset,
int *returnrate, int *returndistortion, int *returnintra,
int *best_single_rd_diff, int *best_comp_rd_diff,
int *best_hybrid_rd_diff)
{
VP8_COMMON *cm = &cpi->common;
BLOCK *b = &x->block[0];
BLOCKD *d = &x->e_mbd.block[0];
MACROBLOCKD *xd = &x->e_mbd;
union b_mode_info best_bmodes[16];
MB_MODE_INFO best_mbmode;
PARTITION_INFO best_partition;
int_mv best_ref_mv, second_best_ref_mv;
int_mv mode_mv[MB_MODE_COUNT];
MB_PREDICTION_MODE this_mode;
int num00;
int best_mode_index = 0;
int mode8x8[2][4];
unsigned char segment_id = xd->mode_info_context->mbmi.segment_id;
int mb_index = xd->mb_index;
int i;
int mode_index;
int mdcounts[4];
int rate;
int distortion;
int best_rd = INT_MAX;
int best_intra_rd = INT_MAX;
int best_comp_rd = INT_MAX;
int best_single_rd = INT_MAX;
int best_hybrid_rd = INT_MAX;
int rate2, distortion2;
int uv_intra_rate, uv_intra_distortion, uv_intra_rate_tokenonly;
int uv_intra_skippable = 0;
int uv_intra_rate_8x8, uv_intra_distortion_8x8, uv_intra_rate_tokenonly_8x8;
int uv_intra_skippable_8x8=0;
int rate_y, UNINITIALIZED_IS_SAFE(rate_uv);
int distortion_uv;
int best_yrd = INT_MAX;
2801280228032804280528062807280828092810281128122813281428152816281728182819282028212822282328242825282628272828282928302831283228332834283528362837283828392840284128422843284428452846284728482849285028512852285328542855285628572858285928602861286228632864286528662867286828692870
//int all_rds[MAX_MODES]; // Experimental debug code.
//int all_rates[MAX_MODES];
//int all_dist[MAX_MODES];
//int intermodecost[MAX_MODES];
MB_PREDICTION_MODE uv_intra_mode;
MB_PREDICTION_MODE uv_intra_mode_8x8;
int_mv mvp;
int near_sadidx[8] = {0, 1, 2, 3, 4, 5, 6, 7};
int saddone=0;
int sr=0; //search range got from mv_pred(). It uses step_param levels. (0-7)
int_mv frame_nearest_mv[4];
int_mv frame_near_mv[4];
int_mv frame_best_ref_mv[4];
int_mv mc_search_result[4];
int frame_mdcounts[4][4];
unsigned char *y_buffer[4];
unsigned char *u_buffer[4];
unsigned char *v_buffer[4];
unsigned int ref_costs[MAX_REF_FRAMES];
int_mv seg_mvs[BLOCK_MAX_SEGMENTS - 1][16 /* n_blocks */][MAX_REF_FRAMES - 1];
vpx_memset(&best_mbmode, 0, sizeof(best_mbmode));
vpx_memset(&best_bmodes, 0, sizeof(best_bmodes));
vpx_memset(&x->mb_context[mb_index], 0, sizeof(PICK_MODE_CONTEXT));
for (i = 0; i < 4; i++)
{
mc_search_result[i].as_int = INVALID_MV;
}
for (i = 0; i < BLOCK_MAX_SEGMENTS - 1; i++)
{
int j;
for (j = 0; j < 16; j++)
{
int k;
for (k = 0; k < MAX_REF_FRAMES - 1; k++)
{
seg_mvs[i][j][k].as_int = INVALID_MV;
}
}
}
if (cpi->ref_frame_flags & VP8_LAST_FLAG)
{
YV12_BUFFER_CONFIG *lst_yv12 = &cpi->common.yv12_fb[cpi->common.lst_fb_idx];
vp8_find_near_mvs(&x->e_mbd, x->e_mbd.mode_info_context,
x->e_mbd.prev_mode_info_context,
&frame_nearest_mv[LAST_FRAME], &frame_near_mv[LAST_FRAME],
&frame_best_ref_mv[LAST_FRAME], frame_mdcounts[LAST_FRAME], LAST_FRAME, cpi->common.ref_frame_sign_bias);
y_buffer[LAST_FRAME] = lst_yv12->y_buffer + recon_yoffset;
u_buffer[LAST_FRAME] = lst_yv12->u_buffer + recon_uvoffset;
v_buffer[LAST_FRAME] = lst_yv12->v_buffer + recon_uvoffset;
}
if (cpi->ref_frame_flags & VP8_GOLD_FLAG)
{
YV12_BUFFER_CONFIG *gld_yv12 = &cpi->common.yv12_fb[cpi->common.gld_fb_idx];
vp8_find_near_mvs(&x->e_mbd, x->e_mbd.mode_info_context,
x->e_mbd.prev_mode_info_context,
&frame_nearest_mv[GOLDEN_FRAME], &frame_near_mv[GOLDEN_FRAME],
2871287228732874287528762877287828792880288128822883288428852886288728882889289028912892289328942895289628972898289929002901290229032904290529062907290829092910291129122913291429152916291729182919292029212922292329242925292629272928292929302931293229332934293529362937293829392940
&frame_best_ref_mv[GOLDEN_FRAME], frame_mdcounts[GOLDEN_FRAME], GOLDEN_FRAME, cpi->common.ref_frame_sign_bias);
y_buffer[GOLDEN_FRAME] = gld_yv12->y_buffer + recon_yoffset;
u_buffer[GOLDEN_FRAME] = gld_yv12->u_buffer + recon_uvoffset;
v_buffer[GOLDEN_FRAME] = gld_yv12->v_buffer + recon_uvoffset;
}
if (cpi->ref_frame_flags & VP8_ALT_FLAG)
{
YV12_BUFFER_CONFIG *alt_yv12 = &cpi->common.yv12_fb[cpi->common.alt_fb_idx];
vp8_find_near_mvs(&x->e_mbd, x->e_mbd.mode_info_context,
x->e_mbd.prev_mode_info_context,
&frame_nearest_mv[ALTREF_FRAME], &frame_near_mv[ALTREF_FRAME],
&frame_best_ref_mv[ALTREF_FRAME], frame_mdcounts[ALTREF_FRAME], ALTREF_FRAME, cpi->common.ref_frame_sign_bias);
y_buffer[ALTREF_FRAME] = alt_yv12->y_buffer + recon_yoffset;
u_buffer[ALTREF_FRAME] = alt_yv12->u_buffer + recon_uvoffset;
v_buffer[ALTREF_FRAME] = alt_yv12->v_buffer + recon_uvoffset;
}
*returnintra = INT_MAX;
x->skip = 0;
vpx_memset(mode_mv, 0, sizeof(mode_mv));
x->e_mbd.mode_info_context->mbmi.ref_frame = INTRA_FRAME;
/* Initialize zbin mode boost for uv costing */
cpi->zbin_mode_boost = 0;
vp8_update_zbin_extra(cpi, x);
rd_pick_intra_mbuv_mode(cpi, x, &uv_intra_rate,
&uv_intra_rate_tokenonly, &uv_intra_distortion);
uv_intra_mode = x->e_mbd.mode_info_context->mbmi.uv_mode;
uv_intra_skippable = mbuv_is_skippable(&x->e_mbd);
/* rough estimate for now */
if(cpi->common.txfm_mode==ALLOW_8X8)
{
rd_pick_intra_mbuv_mode_8x8(cpi, x, &uv_intra_rate_8x8,
&uv_intra_rate_tokenonly_8x8,
&uv_intra_distortion_8x8);
uv_intra_mode_8x8 = x->e_mbd.mode_info_context->mbmi.uv_mode;
uv_intra_skippable_8x8 = mbuv_is_skippable_8x8(&x->e_mbd);
}
// Get estimates of reference frame costs for each reference frame
// that depend on the current prediction etc.
vp8_estimate_ref_frame_costs( cpi, segment_id, ref_costs );
for (mode_index = 0; mode_index < MAX_MODES; mode_index++)
{
int this_rd = INT_MAX;
int disable_skip = 0;
int other_cost = 0;
int compmode_cost = 0;
int mode_excluded = 0;
// Test best rd so far against threshold for trying this mode.
if (best_rd <= cpi->rd_threshes[mode_index])
{
continue;
}
// These variables hold are rolling total cost and distortion for this mode
rate2 = 0;
distortion2 = 0;
rate_y = 0;
2941294229432944294529462947294829492950295129522953295429552956295729582959296029612962296329642965296629672968296929702971297229732974297529762977297829792980298129822983298429852986298729882989299029912992299329942995299629972998299930003001300230033004300530063007300830093010
rate_uv =0;
this_mode = vp8_mode_order[mode_index];
x->e_mbd.mode_info_context->mbmi.mode = this_mode;
#if CONFIG_COMP_INTRA_PRED
x->e_mbd.mode_info_context->mbmi.second_mode = (MB_PREDICTION_MODE) (DC_PRED - 1);
x->e_mbd.mode_info_context->mbmi.second_uv_mode = (MB_PREDICTION_MODE) (DC_PRED - 1);
#endif
x->e_mbd.mode_info_context->mbmi.uv_mode = DC_PRED;
x->e_mbd.mode_info_context->mbmi.ref_frame = vp8_ref_frame_order[mode_index];
x->e_mbd.mode_info_context->mbmi.second_ref_frame = vp8_second_ref_frame_order[mode_index];
// If the segment reference frame feature is enabled....
// then do nothing if the current ref frame is not allowed..
if ( segfeature_active( xd, segment_id, SEG_LVL_REF_FRAME ) &&
!check_segref( xd, segment_id,
xd->mode_info_context->mbmi.ref_frame ) )
{
continue;
}
// If the segment mode feature is enabled....
// then do nothing if the current mode is not allowed..
else if ( segfeature_active( xd, segment_id, SEG_LVL_MODE ) &&
( this_mode !=
get_segdata( xd, segment_id, SEG_LVL_MODE ) ) )
{
continue;
}
// Disable this drop out case if either the mode or ref frame
// segment level feature is enabled for this segment. This is to
// prevent the possibility that the we end up unable to pick any mode.
else if ( !segfeature_active( xd, segment_id, SEG_LVL_REF_FRAME ) &&
!segfeature_active( xd, segment_id, SEG_LVL_MODE ) )
{
// Only consider ZEROMV/ALTREF_FRAME for alt ref frame,
// unless ARNR filtering is enabled in which case we want
// an unfiltered alternative
if (cpi->is_src_frame_alt_ref && (cpi->oxcf.arnr_max_frames == 0))
{
if (this_mode != ZEROMV ||
x->e_mbd.mode_info_context->mbmi.ref_frame != ALTREF_FRAME)
{
continue;
}
}
}
/* everything but intra */
if (x->e_mbd.mode_info_context->mbmi.ref_frame)
{
int ref = x->e_mbd.mode_info_context->mbmi.ref_frame;
x->e_mbd.pre.y_buffer = y_buffer[ref];
x->e_mbd.pre.u_buffer = u_buffer[ref];
x->e_mbd.pre.v_buffer = v_buffer[ref];
mode_mv[NEARESTMV] = frame_nearest_mv[ref];
mode_mv[NEARMV] = frame_near_mv[ref];
best_ref_mv = frame_best_ref_mv[ref];
vpx_memcpy(mdcounts, frame_mdcounts[ref], sizeof(mdcounts));
}
if (x->e_mbd.mode_info_context->mbmi.second_ref_frame)
{
int ref = x->e_mbd.mode_info_context->mbmi.second_ref_frame;
x->e_mbd.second_pre.y_buffer = y_buffer[ref];
x->e_mbd.second_pre.u_buffer = u_buffer[ref];
x->e_mbd.second_pre.v_buffer = v_buffer[ref];
3011301230133014301530163017301830193020302130223023302430253026302730283029303030313032303330343035303630373038303930403041304230433044304530463047304830493050305130523053305430553056305730583059306030613062306330643065306630673068306930703071307230733074307530763077307830793080
second_best_ref_mv = frame_best_ref_mv[ref];
}
// Experimental code. Special case for gf and arf zeromv modes.
// Increase zbin size to suppress noise
if (cpi->zbin_mode_boost_enabled)
{
if ( vp8_ref_frame_order[mode_index] == INTRA_FRAME )
cpi->zbin_mode_boost = 0;
else
{
if (vp8_mode_order[mode_index] == ZEROMV)
{
if (vp8_ref_frame_order[mode_index] != LAST_FRAME)
cpi->zbin_mode_boost = GF_ZEROMV_ZBIN_BOOST;
else
cpi->zbin_mode_boost = LF_ZEROMV_ZBIN_BOOST;
}
else if (vp8_mode_order[mode_index] == SPLITMV)
cpi->zbin_mode_boost = 0;
else
cpi->zbin_mode_boost = MV_ZBIN_BOOST;
}
vp8_update_zbin_extra(cpi, x);
}
if (!x->e_mbd.mode_info_context->mbmi.second_ref_frame)
switch (this_mode)
{
case B_PRED:
{
int tmp_rd;
// Note the rate value returned here includes the cost of coding the BPRED mode : x->mbmode_cost[x->e_mbd.frame_type][BPRED];
tmp_rd = rd_pick_intra4x4mby_modes(cpi, x, &rate, &rate_y, &distortion, best_yrd,
#if CONFIG_COMP_INTRA_PRED
0,
#endif
0);
rate2 += rate;
distortion2 += distortion;
if(tmp_rd < best_yrd)
{
rate2 += uv_intra_rate;
rate_uv = uv_intra_rate_tokenonly;
distortion2 += uv_intra_distortion;
distortion_uv = uv_intra_distortion;
}
else
{
this_rd = INT_MAX;
disable_skip = 1;
}
}
break;
case I8X8_PRED:
{
int tmp_rd;
tmp_rd = rd_pick_intra8x8mby_modes(cpi,
x, &rate, &rate_y, &distortion, best_yrd);
rate2 += rate;
distortion2 += distortion;
mode8x8[0][0]= x->e_mbd.mode_info_context->bmi[0].as_mode.first;
mode8x8[0][1]= x->e_mbd.mode_info_context->bmi[2].as_mode.first;
mode8x8[0][2]= x->e_mbd.mode_info_context->bmi[8].as_mode.first;
mode8x8[0][3]= x->e_mbd.mode_info_context->bmi[10].as_mode.first;
3081308230833084308530863087308830893090309130923093309430953096309730983099310031013102310331043105310631073108310931103111311231133114311531163117311831193120312131223123312431253126312731283129313031313132313331343135313631373138313931403141314231433144314531463147314831493150
#if CONFIG_COMP_INTRA_PRED
mode8x8[1][0]= x->e_mbd.mode_info_context->bmi[0].as_mode.second;
mode8x8[1][1]= x->e_mbd.mode_info_context->bmi[2].as_mode.second;
mode8x8[1][2]= x->e_mbd.mode_info_context->bmi[8].as_mode.second;
mode8x8[1][3]= x->e_mbd.mode_info_context->bmi[10].as_mode.second;
#endif
/* TODO: uv rate maybe over-estimated here since there is UV intra
mode coded in I8X8_PRED prediction */
if(tmp_rd < best_yrd)
{
rate2 += uv_intra_rate;
rate_uv = uv_intra_rate_tokenonly;
distortion2 += uv_intra_distortion;
distortion_uv = uv_intra_distortion;
}
else
{
this_rd = INT_MAX;
disable_skip = 1;
}
}
break;
case SPLITMV:
{
int tmp_rd;
int this_rd_thresh;
this_rd_thresh = (x->e_mbd.mode_info_context->mbmi.ref_frame == LAST_FRAME) ? cpi->rd_threshes[THR_NEWMV] : cpi->rd_threshes[THR_NEWA];
this_rd_thresh = (x->e_mbd.mode_info_context->mbmi.ref_frame == GOLDEN_FRAME) ? cpi->rd_threshes[THR_NEWG]: this_rd_thresh;
tmp_rd = vp8_rd_pick_best_mbsegmentation(cpi, x, &best_ref_mv, NULL,
best_yrd, mdcounts,
&rate, &rate_y, &distortion, this_rd_thresh, seg_mvs) ;
rate2 += rate;
distortion2 += distortion;
// If even the 'Y' rd value of split is higher than best so far then dont bother looking at UV
if (tmp_rd < best_yrd)
{
// Now work out UV cost and add it in
rd_inter4x4_uv(cpi, x, &rate_uv, &distortion_uv, cpi->common.full_pixel);
rate2 += rate_uv;
distortion2 += distortion_uv;
}
else
{
this_rd = INT_MAX;
disable_skip = 1;
}
mode_excluded = cpi->common.comp_pred_mode == COMP_PREDICTION_ONLY;
compmode_cost =
vp8_cost_bit( get_pred_prob( cm, xd, PRED_COMP ), 0 );
}
break;
#if CONFIG_NEWINTRAMODES
case D45_PRED:
case D135_PRED:
case D117_PRED:
case D153_PRED:
case D27_PRED:
case D63_PRED:
#endif
case DC_PRED:
case V_PRED:
case H_PRED:
case TM_PRED:
x->e_mbd.mode_info_context->mbmi.ref_frame = INTRA_FRAME;
3151315231533154315531563157315831593160316131623163316431653166316731683169317031713172317331743175317631773178317931803181318231833184318531863187318831893190319131923193319431953196319731983199320032013202320332043205320632073208320932103211321232133214321532163217321832193220
// FIXME compound intra prediction
RECON_INVOKE(&cpi->common.rtcd.recon, build_intra_predictors_mby)
(&x->e_mbd);
if(cpi->common.txfm_mode == ALLOW_8X8)
macro_block_yrd_8x8(x, &rate_y, &distortion,
IF_RTCD(&cpi->rtcd)) ;
else
macro_block_yrd(x, &rate_y, &distortion,
IF_RTCD(&cpi->rtcd)) ;
rate2 += rate_y;
distortion2 += distortion;
rate2 += x->mbmode_cost[x->e_mbd.frame_type][x->e_mbd.mode_info_context->mbmi.mode];
if(cpi->common.txfm_mode == ALLOW_8X8)
{
rate2 += uv_intra_rate_8x8;
rate_uv = uv_intra_rate_tokenonly_8x8;
distortion2 += uv_intra_distortion_8x8;
distortion_uv = uv_intra_distortion_8x8;
}
else
{
rate2 += uv_intra_rate;
rate_uv = uv_intra_rate_tokenonly;
distortion2 += uv_intra_distortion;
distortion_uv = uv_intra_distortion;
}
break;
case NEWMV:
{
int thissme;
int bestsme = INT_MAX;
int step_param = cpi->sf.first_step;
int further_steps;
int n;
int do_refine=1; /* If last step (1-away) of n-step search doesn't pick the center point as the best match,
we will do a final 1-away diamond refining search */
int sadpb = x->sadperbit16;
int_mv mvp_full;
int col_min = (best_ref_mv.as_mv.col>>3) - MAX_FULL_PEL_VAL + ((best_ref_mv.as_mv.col & 7)?1:0);
int row_min = (best_ref_mv.as_mv.row>>3) - MAX_FULL_PEL_VAL + ((best_ref_mv.as_mv.row & 7)?1:0);
int col_max = (best_ref_mv.as_mv.col>>3) + MAX_FULL_PEL_VAL;
int row_max = (best_ref_mv.as_mv.row>>3) + MAX_FULL_PEL_VAL;
int tmp_col_min = x->mv_col_min;
int tmp_col_max = x->mv_col_max;
int tmp_row_min = x->mv_row_min;
int tmp_row_max = x->mv_row_max;
if(!saddone)
{
vp8_cal_sad(cpi,xd,x, recon_yoffset ,&near_sadidx[0] );
saddone = 1;
}
vp8_mv_pred(cpi, &x->e_mbd, x->e_mbd.mode_info_context, &mvp,
x->e_mbd.mode_info_context->mbmi.ref_frame, cpi->common.ref_frame_sign_bias, &sr, &near_sadidx[0]);
mvp_full.as_mv.col = mvp.as_mv.col>>3;
mvp_full.as_mv.row = mvp.as_mv.row>>3;
// Get intersection of UMV window and valid MV window to reduce # of checks in diamond search.
if (x->mv_col_min < col_min )
x->mv_col_min = col_min;
if (x->mv_col_max > col_max )
x->mv_col_max = col_max;
if (x->mv_row_min < row_min )
x->mv_row_min = row_min;
3221322232233224322532263227322832293230323132323233323432353236323732383239324032413242324332443245324632473248324932503251325232533254325532563257325832593260326132623263326432653266326732683269327032713272327332743275327632773278327932803281328232833284328532863287328832893290
if (x->mv_row_max > row_max )
x->mv_row_max = row_max;
//adjust search range according to sr from mv prediction
if(sr > step_param)
step_param = sr;
// Initial step/diamond search
{
bestsme = cpi->diamond_search_sad(x, b, d, &mvp_full, &d->bmi.as_mv.first,
step_param, sadpb, &num00,
&cpi->fn_ptr[BLOCK_16X16],
XMVCOST, &best_ref_mv);
mode_mv[NEWMV].as_int = d->bmi.as_mv.first.as_int;
// Further step/diamond searches as necessary
n = 0;
further_steps = (cpi->sf.max_step_search_steps - 1) - step_param;
n = num00;
num00 = 0;
/* If there won't be more n-step search, check to see if refining search is needed. */
if (n > further_steps)
do_refine = 0;
while (n < further_steps)
{
n++;
if (num00)
num00--;
else
{
thissme = cpi->diamond_search_sad(x, b, d, &mvp_full,
&d->bmi.as_mv.first, step_param + n, sadpb, &num00,
&cpi->fn_ptr[BLOCK_16X16],
XMVCOST, &best_ref_mv);
/* check to see if refining search is needed. */
if (num00 > (further_steps-n))
do_refine = 0;
if (thissme < bestsme)
{
bestsme = thissme;
mode_mv[NEWMV].as_int = d->bmi.as_mv.first.as_int;
}
else
{
d->bmi.as_mv.first.as_int = mode_mv[NEWMV].as_int;
}
}
}
}
/* final 1-away diamond refining search */
if (do_refine == 1)
{
int search_range;
//It seems not a good way to set search_range. Need further investigation.
//search_range = MAXF(abs((mvp.row>>3) - d->bmi.mv.as_mv.row), abs((mvp.col>>3) - d->bmi.mv.as_mv.col));
search_range = 8;
//thissme = cpi->full_search_sad(x, b, d, &d->bmi.mv.as_mv, sadpb, search_range, &cpi->fn_ptr[BLOCK_16X16], x->mvcost, &best_ref_mv);
thissme = cpi->refining_search_sad(x, b, d, &d->bmi.as_mv.first, sadpb,
search_range, &cpi->fn_ptr[BLOCK_16X16],
XMVCOST, &best_ref_mv);
3291329232933294329532963297329832993300330133023303330433053306330733083309331033113312331333143315331633173318331933203321332233233324332533263327332833293330333133323333333433353336333733383339334033413342334333443345334633473348334933503351335233533354335533563357335833593360
if (thissme < bestsme)
{
bestsme = thissme;
mode_mv[NEWMV].as_int = d->bmi.as_mv.first.as_int;
}
else
{
d->bmi.as_mv.first.as_int = mode_mv[NEWMV].as_int;
}
}
x->mv_col_min = tmp_col_min;
x->mv_col_max = tmp_col_max;
x->mv_row_min = tmp_row_min;
x->mv_row_max = tmp_row_max;
if (bestsme < INT_MAX)
{
int dis; /* TODO: use dis in distortion calculation later. */
unsigned int sse;
cpi->find_fractional_mv_step(x, b, d, &d->bmi.as_mv.first, &best_ref_mv,
x->errorperbit,
&cpi->fn_ptr[BLOCK_16X16],
XMVCOST, &dis, &sse);
}
mc_search_result[x->e_mbd.mode_info_context->mbmi.ref_frame].as_int = d->bmi.as_mv.first.as_int;
mode_mv[NEWMV].as_int = d->bmi.as_mv.first.as_int;
// Add the new motion vector cost to our rolling cost variable
#if CONFIG_HIGH_PRECISION_MV
rate2 += vp8_mv_bit_cost(&mode_mv[NEWMV], &best_ref_mv,
XMVCOST, 96,
x->e_mbd.allow_high_precision_mv);
#else
rate2 += vp8_mv_bit_cost(&mode_mv[NEWMV], &best_ref_mv,
XMVCOST, 96);
#endif
}
case NEARESTMV:
case NEARMV:
// Clip "next_nearest" so that it does not extend to far out of image
vp8_clamp_mv2(&mode_mv[this_mode], xd);
// Do not bother proceeding if the vector (from newmv,nearest or near) is 0,0 as this should then be coded using the zeromv mode.
if (((this_mode == NEARMV) || (this_mode == NEARESTMV)) && (mode_mv[this_mode].as_int == 0))
{
continue;
}
case ZEROMV:
// Trap vectors that reach beyond the UMV borders
// Note that ALL New MV, Nearest MV Near MV and Zero MV code drops through to this point
// because of the lack of break statements in the previous two cases.
if (((mode_mv[this_mode].as_mv.row >> 3) < x->mv_row_min) || ((mode_mv[this_mode].as_mv.row >> 3) > x->mv_row_max) ||
((mode_mv[this_mode].as_mv.col >> 3) < x->mv_col_min) || ((mode_mv[this_mode].as_mv.col >> 3) > x->mv_col_max))
{
continue;
}
vp8_set_mbmode_and_mvs(x, this_mode, &mode_mv[this_mode]);
vp8_build_inter16x16_predictors_mby(&x->e_mbd);
compmode_cost =
vp8_cost_bit( get_pred_prob( cm, xd, PRED_COMP ), 0 );
if (cpi->active_map_enabled && x->active_ptr[0] == 0) {
x->skip = 1;
3361336233633364336533663367336833693370337133723373337433753376337733783379338033813382338333843385338633873388338933903391339233933394339533963397339833993400340134023403340434053406340734083409341034113412341334143415341634173418341934203421342234233424342534263427342834293430
}
else if (x->encode_breakout)
{
unsigned int sse;
unsigned int var;
int threshold = (xd->block[0].dequant[1]
* xd->block[0].dequant[1] >>4);
if(threshold < x->encode_breakout)
threshold = x->encode_breakout;
var = VARIANCE_INVOKE(&cpi->rtcd.variance, var16x16)
(*(b->base_src), b->src_stride,
x->e_mbd.predictor, 16, &sse);
if (sse < threshold)
{
unsigned int q2dc = xd->block[24].dequant[0];
/* If there is no codeable 2nd order dc
or a very small uniform pixel change change */
if ((sse - var < q2dc * q2dc >>4) ||
(sse /2 > var && sse-var < 64))
{
// Check u and v to make sure skip is ok
int sse2= VP8_UVSSE(x, IF_RTCD(&cpi->rtcd.variance));
if (sse2 * 2 < threshold)
{
x->skip = 1;
distortion2 = sse + sse2;
rate2 = 500;
/* for best_yrd calculation */
rate_uv = 0;
distortion_uv = sse2;
disable_skip = 1;
this_rd = RDCOST(x->rdmult, x->rddiv, rate2, distortion2);
break;
}
}
}
}
//intermodecost[mode_index] = vp8_cost_mv_ref(cpi, this_mode, mdcounts); // Experimental debug code
// Add in the Mv/mode cost
rate2 += vp8_cost_mv_ref(cpi, this_mode, mdcounts);
// Y cost and distortion
if(cpi->common.txfm_mode == ALLOW_8X8)
macro_block_yrd_8x8(x, &rate_y, &distortion,
IF_RTCD(&cpi->rtcd));
else
macro_block_yrd(x, &rate_y, &distortion,
IF_RTCD(&cpi->rtcd));
rate2 += rate_y;
distortion2 += distortion;
// UV cost and distortion
vp8_build_inter16x16_predictors_mbuv(&x->e_mbd);
if(cpi->common.txfm_mode == ALLOW_8X8)
rd_inter16x16_uv_8x8(cpi, x, &rate_uv,
&distortion_uv,
cpi->common.full_pixel);
else
rd_inter16x16_uv(cpi, x, &rate_uv,
&distortion_uv,
cpi->common.full_pixel);
3431343234333434343534363437343834393440344134423443344434453446344734483449345034513452345334543455345634573458345934603461346234633464346534663467346834693470347134723473347434753476347734783479348034813482348334843485348634873488348934903491349234933494349534963497349834993500
rate2 += rate_uv;
distortion2 += distortion_uv;
mode_excluded = cpi->common.comp_pred_mode == COMP_PREDICTION_ONLY;
break;
default:
break;
}
else /* x->e_mbd.mode_info_context->mbmi.second_ref_frame != 0 */
{
int ref1 = x->e_mbd.mode_info_context->mbmi.ref_frame;
int ref2 = x->e_mbd.mode_info_context->mbmi.second_ref_frame;
mode_excluded = cpi->common.comp_pred_mode == SINGLE_PREDICTION_ONLY;
switch (this_mode)
{
case NEWMV:
if (mc_search_result[ref1].as_int == INVALID_MV ||
mc_search_result[ref2].as_int == INVALID_MV)
continue;
x->e_mbd.mode_info_context->mbmi.mv.as_int = mc_search_result[ref1].as_int;
x->e_mbd.mode_info_context->mbmi.second_mv.as_int = mc_search_result[ref2].as_int;
#if CONFIG_HIGH_PRECISION_MV
rate2 += vp8_mv_bit_cost(&mc_search_result[ref1],
&frame_best_ref_mv[ref1],
XMVCOST, 96,
x->e_mbd.allow_high_precision_mv);
rate2 += vp8_mv_bit_cost(&mc_search_result[ref2],
&frame_best_ref_mv[ref2],
XMVCOST, 96,
x->e_mbd.allow_high_precision_mv);
#else
rate2 += vp8_mv_bit_cost(&mc_search_result[ref1],
&frame_best_ref_mv[ref1],
XMVCOST, 96);
rate2 += vp8_mv_bit_cost(&mc_search_result[ref2],
&frame_best_ref_mv[ref2],
XMVCOST, 96);
#endif
break;
case ZEROMV:
x->e_mbd.mode_info_context->mbmi.mv.as_int = 0;
x->e_mbd.mode_info_context->mbmi.second_mv.as_int = 0;
break;
case NEARMV:
if (frame_near_mv[ref1].as_int == 0 || frame_near_mv[ref2].as_int == 0)
{
continue;
}
x->e_mbd.mode_info_context->mbmi.mv.as_int = frame_near_mv[ref1].as_int;
x->e_mbd.mode_info_context->mbmi.second_mv.as_int = frame_near_mv[ref2].as_int;
break;
case NEARESTMV:
if (frame_nearest_mv[ref1].as_int == 0 || frame_nearest_mv[ref2].as_int == 0)
{
continue;
}
x->e_mbd.mode_info_context->mbmi.mv.as_int = frame_nearest_mv[ref1].as_int;
x->e_mbd.mode_info_context->mbmi.second_mv.as_int = frame_nearest_mv[ref2].as_int;
break;
case SPLITMV:
{
int tmp_rd;
int this_rd_thresh;
this_rd_thresh = (x->e_mbd.mode_info_context->mbmi.ref_frame == LAST_FRAME) ? cpi->rd_threshes[THR_NEWMV] : cpi->rd_threshes[THR_NEWA];
this_rd_thresh = (x->e_mbd.mode_info_context->mbmi.ref_frame == GOLDEN_FRAME) ? cpi->rd_threshes[THR_NEWG]: this_rd_thresh;
tmp_rd = vp8_rd_pick_best_mbsegmentation(cpi, x, &best_ref_mv, &second_best_ref_mv,
best_yrd, mdcounts,
3501350235033504350535063507350835093510351135123513351435153516351735183519352035213522352335243525352635273528352935303531353235333534353535363537353835393540354135423543354435453546354735483549355035513552355335543555355635573558355935603561356235633564356535663567356835693570
&rate, &rate_y, &distortion, this_rd_thresh, seg_mvs) ;
rate2 += rate;
distortion2 += distortion;
// If even the 'Y' rd value of split is higher than best so far then dont bother looking at UV
if (tmp_rd < best_yrd)
{
// Now work out UV cost and add it in
rd_inter4x4_uv(cpi, x, &rate_uv, &distortion_uv, cpi->common.full_pixel);
rate2 += rate_uv;
distortion2 += distortion_uv;
}
else
{
this_rd = INT_MAX;
disable_skip = 1;
}
}
break;
default:
break;
}
if (this_mode != SPLITMV)
{
/* Add in the Mv/mode cost */
rate2 += vp8_cost_mv_ref(cpi, this_mode, mdcounts);
vp8_clamp_mv2(&x->e_mbd.mode_info_context->mbmi.mv, xd);
vp8_clamp_mv2(&x->e_mbd.mode_info_context->mbmi.second_mv, xd);
if (((x->e_mbd.mode_info_context->mbmi.mv.as_mv.row >> 3) < x->mv_row_min) ||
((x->e_mbd.mode_info_context->mbmi.mv.as_mv.row >> 3) > x->mv_row_max) ||
((x->e_mbd.mode_info_context->mbmi.mv.as_mv.col >> 3) < x->mv_col_min) ||
((x->e_mbd.mode_info_context->mbmi.mv.as_mv.col >> 3) > x->mv_col_max) ||
((x->e_mbd.mode_info_context->mbmi.second_mv.as_mv.row >> 3) < x->mv_row_min) ||
((x->e_mbd.mode_info_context->mbmi.second_mv.as_mv.row >> 3) > x->mv_row_max) ||
((x->e_mbd.mode_info_context->mbmi.second_mv.as_mv.col >> 3) < x->mv_col_min) ||
((x->e_mbd.mode_info_context->mbmi.second_mv.as_mv.col >> 3) > x->mv_col_max))
{
continue;
}
/* build first and second prediction */
vp8_build_inter16x16_predictors_mby(&x->e_mbd);
vp8_build_inter16x16_predictors_mbuv(&x->e_mbd);
/* do second round and average the results */
vp8_build_2nd_inter16x16_predictors_mb(&x->e_mbd, x->e_mbd.predictor,
&x->e_mbd.predictor[256],
&x->e_mbd.predictor[320], 16, 8);
/* Y cost and distortion */
if (cpi->common.txfm_mode == ALLOW_8X8)
macro_block_yrd_8x8(x, &rate_y, &distortion,
IF_RTCD(&cpi->rtcd));
else
macro_block_yrd(x, &rate_y, &distortion,
IF_RTCD(&cpi->rtcd));
rate2 += rate_y;
distortion2 += distortion;
/* UV cost and distortion */
if(cpi->common.txfm_mode == ALLOW_8X8)
rd_inter16x16_uv_8x8(cpi, x, &rate_uv,
&distortion_uv,
cpi->common.full_pixel);
else
rd_inter16x16_uv(cpi, x, &rate_uv,
&distortion_uv,
3571357235733574357535763577357835793580358135823583358435853586358735883589359035913592359335943595359635973598359936003601360236033604360536063607360836093610361136123613361436153616361736183619362036213622362336243625362636273628362936303631363236333634363536363637363836393640
cpi->common.full_pixel);
rate2 += rate_uv;
distortion2 += distortion_uv;
}
/* don't bother w/ skip, we would never have come here if skip were enabled */
x->e_mbd.mode_info_context->mbmi.mode = this_mode;
/* We don't include the cost of the second reference here, because there are only
* three options: Last/Golden, ARF/Last or Golden/ARF, or in other words if you
* present them in that order, the second one is always known if the first is known */
compmode_cost =
vp8_cost_bit( get_pred_prob( cm, xd, PRED_COMP ), 1 );
}
if (cpi->common.comp_pred_mode == HYBRID_PREDICTION)
{
rate2 += compmode_cost;
}
// Estimate the reference frame signaling cost and add it
// to the rolling cost variable.
rate2 += ref_costs[x->e_mbd.mode_info_context->mbmi.ref_frame];
if (!disable_skip)
{
// Test for the condition where skip block will be activated
// because there are no non zero coefficients and make any
// necessary adjustment for rate. Ignore if skip is coded at
// segment level as the cost wont have been added in.
if ( cpi->common.mb_no_coeff_skip )
{
int mb_skippable;
int mb_skip_allowed;
int has_y2 = ( this_mode!=SPLITMV
&&this_mode!=B_PRED
&&this_mode!=I8X8_PRED);
if((cpi->common.txfm_mode == ALLOW_8X8) && has_y2)
{
if(x->e_mbd.mode_info_context->mbmi.ref_frame!=INTRA_FRAME)
mb_skippable = mb_is_skippable_8x8(&x->e_mbd);
else
mb_skippable = uv_intra_skippable_8x8
& mby_is_skippable_8x8(&x->e_mbd);
}
else
{
if(x->e_mbd.mode_info_context->mbmi.ref_frame!=INTRA_FRAME)
mb_skippable = mb_is_skippable(&x->e_mbd, has_y2);
else
mb_skippable = uv_intra_skippable
& mby_is_skippable(&x->e_mbd, has_y2);
}
// Is Mb level skip allowed for this mb.
mb_skip_allowed =
!segfeature_active( xd, segment_id, SEG_LVL_EOB ) ||
get_segdata( xd, segment_id, SEG_LVL_EOB );
if (mb_skippable)
{
// Back out the coefficient coding costs
rate2 -= (rate_y + rate_uv);
//for best_yrd calculation
rate_uv = 0;
if ( mb_skip_allowed )
{
3641364236433644364536463647364836493650365136523653365436553656365736583659366036613662366336643665366636673668366936703671367236733674367536763677367836793680368136823683368436853686368736883689369036913692369336943695369636973698369937003701370237033704370537063707370837093710
int prob_skip_cost;
// Cost the skip mb case
#if CONFIG_NEWENTROPY
vp8_prob skip_prob =
get_pred_prob(cm, &x->e_mbd, PRED_MBSKIP);
if (skip_prob)
{
prob_skip_cost = vp8_cost_bit(skip_prob, 1);
rate2 += prob_skip_cost;
other_cost += prob_skip_cost;
}
#else
if (cpi->prob_skip_false)
{
prob_skip_cost =
vp8_cost_bit(cpi->prob_skip_false, 1);
rate2 += prob_skip_cost;
other_cost += prob_skip_cost;
}
#endif
}
}
// Add in the cost of the no skip flag.
else if ( mb_skip_allowed )
{
#if CONFIG_NEWENTROPY
int prob_skip_cost = vp8_cost_bit(
get_pred_prob(cm, &x->e_mbd, PRED_MBSKIP), 0);
#else
int prob_skip_cost = vp8_cost_bit(cpi->prob_skip_false, 0);
#endif
rate2 += prob_skip_cost;
other_cost += prob_skip_cost;
}
}
// Calculate the final RD estimate for this mode.
this_rd = RDCOST(x->rdmult, x->rddiv, rate2, distortion2);
}
// Experimental debug code.
//all_rds[mode_index] = this_rd;
//all_rates[mode_index] = rate2;
//all_dist[mode_index] = distortion2;
// Keep record of best intra distortion
if ((x->e_mbd.mode_info_context->mbmi.ref_frame == INTRA_FRAME) &&
(this_rd < best_intra_rd) )
{
best_intra_rd = this_rd;
*returnintra = distortion2 ;
}
if (!disable_skip && x->e_mbd.mode_info_context->mbmi.ref_frame == INTRA_FRAME)
{
if (this_rd < best_comp_rd)
best_comp_rd = this_rd;
if (this_rd < best_single_rd)
best_single_rd = this_rd;
if (this_rd < best_hybrid_rd)
best_hybrid_rd = this_rd;
}
// Did this mode help.. i.e. is it the new best mode
if (this_rd < best_rd || x->skip)
{
if (!mode_excluded)
{
3711371237133714371537163717371837193720372137223723372437253726372737283729373037313732373337343735373637373738373937403741374237433744374537463747374837493750375137523753375437553756375737583759376037613762376337643765376637673768376937703771377237733774377537763777377837793780
// Note index of best mode so far
best_mode_index = mode_index;
if (this_mode <= B_PRED)
{
if( cpi->common.txfm_mode == ALLOW_8X8
&& this_mode != B_PRED
&& this_mode != I8X8_PRED)
x->e_mbd.mode_info_context->mbmi.uv_mode = uv_intra_mode_8x8;
else
x->e_mbd.mode_info_context->mbmi.uv_mode = uv_intra_mode;
/* required for left and above block mv */
x->e_mbd.mode_info_context->mbmi.mv.as_int = 0;
}
other_cost +=
ref_costs[x->e_mbd.mode_info_context->mbmi.ref_frame];
/* Calculate the final y RD estimate for this mode */
best_yrd = RDCOST(x->rdmult, x->rddiv, (rate2-rate_uv-other_cost),
(distortion2-distortion_uv));
*returnrate = rate2;
*returndistortion = distortion2;
best_rd = this_rd;
vpx_memcpy(&best_mbmode, &x->e_mbd.mode_info_context->mbmi, sizeof(MB_MODE_INFO));
vpx_memcpy(&best_partition, x->partition_info, sizeof(PARTITION_INFO));
if ((this_mode == B_PRED)
||(this_mode == I8X8_PRED)
|| (this_mode == SPLITMV))
for (i = 0; i < 16; i++)
{
best_bmodes[i] = x->e_mbd.block[i].bmi;
}
}
// Testing this mode gave rise to an improvement in best error score. Lower threshold a bit for next time
cpi->rd_thresh_mult[mode_index] = (cpi->rd_thresh_mult[mode_index] >= (MIN_THRESHMULT + 2)) ? cpi->rd_thresh_mult[mode_index] - 2 : MIN_THRESHMULT;
cpi->rd_threshes[mode_index] = (cpi->rd_baseline_thresh[mode_index] >> 7) * cpi->rd_thresh_mult[mode_index];
}
// If the mode did not help improve the best error case then raise the threshold for testing that mode next time around.
else
{
cpi->rd_thresh_mult[mode_index] += 4;
if (cpi->rd_thresh_mult[mode_index] > MAX_THRESHMULT)
cpi->rd_thresh_mult[mode_index] = MAX_THRESHMULT;
cpi->rd_threshes[mode_index] = (cpi->rd_baseline_thresh[mode_index] >> 7) * cpi->rd_thresh_mult[mode_index];
}
/* keep record of best compound/single-only prediction */
if (!disable_skip &&
x->e_mbd.mode_info_context->mbmi.ref_frame != INTRA_FRAME)
{
int single_rd, hybrid_rd, single_rate, hybrid_rate;
if (cpi->common.comp_pred_mode == HYBRID_PREDICTION)
{
single_rate = rate2 - compmode_cost;
hybrid_rate = rate2;
}
else
{
single_rate = rate2;
hybrid_rate = rate2 + compmode_cost;
}
single_rd = RDCOST(x->rdmult, x->rddiv, single_rate, distortion2);
3781378237833784378537863787378837893790379137923793379437953796379737983799380038013802380338043805380638073808380938103811381238133814381538163817381838193820382138223823382438253826382738283829383038313832383338343835383638373838383938403841384238433844384538463847384838493850
hybrid_rd = RDCOST(x->rdmult, x->rddiv, hybrid_rate, distortion2);
if (x->e_mbd.mode_info_context->mbmi.second_ref_frame == INTRA_FRAME &&
single_rd < best_single_rd)
{
best_single_rd = single_rd;
}
else if (x->e_mbd.mode_info_context->mbmi.second_ref_frame != INTRA_FRAME &&
single_rd < best_comp_rd)
{
best_comp_rd = single_rd;
}
if (hybrid_rd < best_hybrid_rd)
{
best_hybrid_rd = hybrid_rd;
}
}
if (x->skip)
break;
}
// Reduce the activation RD thresholds for the best choice mode
if ((cpi->rd_baseline_thresh[best_mode_index] > 0) && (cpi->rd_baseline_thresh[best_mode_index] < (INT_MAX >> 2)))
{
int best_adjustment = (cpi->rd_thresh_mult[best_mode_index] >> 2);
cpi->rd_thresh_mult[best_mode_index] = (cpi->rd_thresh_mult[best_mode_index] >= (MIN_THRESHMULT + best_adjustment)) ? cpi->rd_thresh_mult[best_mode_index] - best_adjustment : MIN_THRESHMULT;
cpi->rd_threshes[best_mode_index] = (cpi->rd_baseline_thresh[best_mode_index] >> 7) * cpi->rd_thresh_mult[best_mode_index];
// If we chose a split mode then reset the new MV thresholds as well
/*if ( vp8_mode_order[best_mode_index] == SPLITMV )
{
best_adjustment = 4; //(cpi->rd_thresh_mult[THR_NEWMV] >> 4);
cpi->rd_thresh_mult[THR_NEWMV] = (cpi->rd_thresh_mult[THR_NEWMV] >= (MIN_THRESHMULT+best_adjustment)) ? cpi->rd_thresh_mult[THR_NEWMV]-best_adjustment: MIN_THRESHMULT;
cpi->rd_threshes[THR_NEWMV] = (cpi->rd_baseline_thresh[THR_NEWMV] >> 7) * cpi->rd_thresh_mult[THR_NEWMV];
best_adjustment = 4; //(cpi->rd_thresh_mult[THR_NEWG] >> 4);
cpi->rd_thresh_mult[THR_NEWG] = (cpi->rd_thresh_mult[THR_NEWG] >= (MIN_THRESHMULT+best_adjustment)) ? cpi->rd_thresh_mult[THR_NEWG]-best_adjustment: MIN_THRESHMULT;
cpi->rd_threshes[THR_NEWG] = (cpi->rd_baseline_thresh[THR_NEWG] >> 7) * cpi->rd_thresh_mult[THR_NEWG];
best_adjustment = 4; //(cpi->rd_thresh_mult[THR_NEWA] >> 4);
cpi->rd_thresh_mult[THR_NEWA] = (cpi->rd_thresh_mult[THR_NEWA] >= (MIN_THRESHMULT+best_adjustment)) ? cpi->rd_thresh_mult[THR_NEWA]-best_adjustment: MIN_THRESHMULT;
cpi->rd_threshes[THR_NEWA] = (cpi->rd_baseline_thresh[THR_NEWA] >> 7) * cpi->rd_thresh_mult[THR_NEWA];
}*/
}
// This code force Altref,0,0 and skip for the frame that overlays a
// an alrtef unless Altref is filtered. However, this is unsafe if
// segment level coding of ref frame or mode is enabled for this
// segment.
if (!segfeature_active( xd, segment_id, SEG_LVL_REF_FRAME ) &&
!segfeature_active( xd, segment_id, SEG_LVL_MODE ) &&
cpi->is_src_frame_alt_ref &&
(cpi->oxcf.arnr_max_frames == 0) &&
(best_mbmode.mode != ZEROMV || best_mbmode.ref_frame != ALTREF_FRAME))
{
x->e_mbd.mode_info_context->mbmi.mode = ZEROMV;
x->e_mbd.mode_info_context->mbmi.ref_frame = ALTREF_FRAME;
x->e_mbd.mode_info_context->mbmi.mv.as_int = 0;
x->e_mbd.mode_info_context->mbmi.uv_mode = DC_PRED;
x->e_mbd.mode_info_context->mbmi.mb_skip_coeff =
(cpi->common.mb_no_coeff_skip) ? 1 : 0;
x->e_mbd.mode_info_context->mbmi.partitioning = 0;
*best_single_rd_diff = *best_comp_rd_diff = *best_hybrid_rd_diff = 0;
store_coding_context (x, mb_index, best_mode_index, &best_partition,
3851385238533854385538563857385838593860386138623863386438653866386738683869387038713872387338743875387638773878387938803881388238833884388538863887388838893890389138923893389438953896389738983899390039013902390339043905390639073908390939103911391239133914391539163917391839193920
&frame_best_ref_mv[xd->mode_info_context->mbmi.ref_frame] );
return;
}
// macroblock modes
vpx_memcpy(&x->e_mbd.mode_info_context->mbmi, &best_mbmode, sizeof(MB_MODE_INFO));
if (best_mbmode.mode == B_PRED)
{
for (i = 0; i < 16; i++)
{
xd->mode_info_context->bmi[i].as_mode = best_bmodes[i].as_mode;
xd->block[i].bmi.as_mode = xd->mode_info_context->bmi[i].as_mode;
}
}
if (best_mbmode.mode == I8X8_PRED)
{
set_i8x8_block_modes(x, mode8x8);
}
if (best_mbmode.mode == SPLITMV)
{
for (i = 0; i < 16; i++)
xd->mode_info_context->bmi[i].as_mv.first.as_int = best_bmodes[i].as_mv.first.as_int;
if (xd->mode_info_context->mbmi.second_ref_frame)
for (i = 0; i < 16; i++)
xd->mode_info_context->bmi[i].as_mv.second.as_int = best_bmodes[i].as_mv.second.as_int;
vpx_memcpy(x->partition_info, &best_partition, sizeof(PARTITION_INFO));
x->e_mbd.mode_info_context->mbmi.mv.as_int =
x->partition_info->bmi[15].mv.as_int;
x->e_mbd.mode_info_context->mbmi.second_mv.as_int =
x->partition_info->bmi[15].second_mv.as_int;
}
if (best_single_rd == INT_MAX)
*best_single_rd_diff = INT_MIN;
else
*best_single_rd_diff = best_rd - best_single_rd;
if (best_comp_rd == INT_MAX)
*best_comp_rd_diff = INT_MIN;
else
*best_comp_rd_diff = best_rd - best_comp_rd;
if (best_hybrid_rd == INT_MAX)
*best_hybrid_rd_diff = INT_MIN;
else
*best_hybrid_rd_diff = best_rd - best_hybrid_rd;
store_coding_context (x, mb_index, best_mode_index, &best_partition,
&frame_best_ref_mv[xd->mode_info_context->mbmi.ref_frame] );
}
int vp8_rd_pick_intra_mode(VP8_COMP *cpi, MACROBLOCK *x)
{
MACROBLOCKD *xd = &x->e_mbd;
int error4x4, error16x16;
#if CONFIG_COMP_INTRA_PRED
int error4x4d, rate4x4d, dist4x4d;
#endif
int rate4x4, rate16x16 = 0, rateuv;
int dist4x4, dist16x16, distuv;
int rate;
int rate4x4_tokenonly = 0;
int rate16x16_tokenonly = 0;
int rateuv_tokenonly = 0;
int error8x8, rate8x8_tokenonly=0;
int rate8x8, dist8x8;
int mode16x16;
3921392239233924392539263927392839293930393139323933393439353936393739383939394039413942394339443945394639473948394939503951395239533954395539563957395839593960396139623963396439653966396739683969397039713972397339743975397639773978397939803981398239833984398539863987398839893990
int mode8x8[2][4];
xd->mode_info_context->mbmi.ref_frame = INTRA_FRAME;
rd_pick_intra_mbuv_mode(cpi, x, &rateuv, &rateuv_tokenonly, &distuv);
rate = rateuv;
error16x16 = rd_pick_intra16x16mby_mode(cpi, x,
&rate16x16, &rate16x16_tokenonly,
&dist16x16);
mode16x16 = xd->mode_info_context->mbmi.mode;
error8x8 = rd_pick_intra8x8mby_modes(cpi, x,
&rate8x8, &rate8x8_tokenonly,
&dist8x8, error16x16);
mode8x8[0][0]= xd->mode_info_context->bmi[0].as_mode.first;
mode8x8[0][1]= xd->mode_info_context->bmi[2].as_mode.first;
mode8x8[0][2]= xd->mode_info_context->bmi[8].as_mode.first;
mode8x8[0][3]= xd->mode_info_context->bmi[10].as_mode.first;
#if CONFIG_COMP_INTRA_PRED
mode8x8[1][0]= xd->mode_info_context->bmi[0].as_mode.second;
mode8x8[1][1]= xd->mode_info_context->bmi[2].as_mode.second;
mode8x8[1][2]= xd->mode_info_context->bmi[8].as_mode.second;
mode8x8[1][3]= xd->mode_info_context->bmi[10].as_mode.second;
#endif
error4x4 = rd_pick_intra4x4mby_modes(cpi, x,
&rate4x4, &rate4x4_tokenonly,
&dist4x4, error16x16,
#if CONFIG_COMP_INTRA_PRED
0,
#endif
0);
#if CONFIG_COMP_INTRA_PRED
error4x4d = rd_pick_intra4x4mby_modes(cpi, x,
&rate4x4d, &rate4x4_tokenonly,
&dist4x4d, error16x16, 1, 0);
#endif
if(error8x8> error16x16)
{
if (error4x4 < error16x16)
{
#if CONFIG_COMP_INTRA_PRED
rate += (error4x4d < error4x4) ? rate4x4d : rate4x4;
if (error4x4d >= error4x4) // FIXME save original modes etc.
error4x4 = rd_pick_intra4x4mby_modes(cpi, x, &rate4x4,
&rate4x4_tokenonly,
&dist4x4, error16x16, 0,
cpi->update_context);
#else
rate += rate4x4;
#endif
xd->mode_info_context->mbmi.mode = B_PRED;
}
else
{
xd->mode_info_context->mbmi.mode = mode16x16;
rate += rate16x16;
}
}
else
{
if (error4x4 < error8x8)
{
#if CONFIG_COMP_INTRA_PRED
rate += (error4x4d < error4x4) ? rate4x4d : rate4x4;
if (error4x4d >= error4x4) // FIXME save original modes etc.
error4x4 = rd_pick_intra4x4mby_modes(cpi, x, &rate4x4,
3991399239933994399539963997399839994000400140024003400440054006400740084009401040114012401340144015401640174018401940204021402240234024402540264027402840294030403140324033403440354036403740384039404040414042404340444045404640474048404940504051405240534054405540564057405840594060
&rate4x4_tokenonly,
&dist4x4, error16x16, 0,
cpi->update_context);
#else
rate += rate4x4;
#endif
xd->mode_info_context->mbmi.mode = B_PRED;
}
else
{
xd->mode_info_context->mbmi.mode = I8X8_PRED;
set_i8x8_block_modes(x, mode8x8);
rate += rate8x8;
}
}
return rate;
}
int vp8cx_pick_mode_inter_macroblock
(
VP8_COMP *cpi, MACROBLOCK *x,
int recon_yoffset, int recon_uvoffset
)
{
VP8_COMMON *cm = &cpi->common;
MACROBLOCKD *const xd = &x->e_mbd;
int rate;
int distortion;
int intra_error = 0;
unsigned char *segment_id = &xd->mode_info_context->mbmi.segment_id;
#if CONFIG_COMPRED
unsigned char ref_pred_flag;
#endif
if (xd->segmentation_enabled)
x->encode_breakout = cpi->segment_encode_breakout[*segment_id];
else
x->encode_breakout = cpi->oxcf.encode_breakout;
//if (cpi->sf.RD)
// For now this codebase is limited to a single rd encode path
{
int zbin_mode_boost_enabled = cpi->zbin_mode_boost_enabled;
int single, compound, hybrid;
vp8_rd_pick_inter_mode(cpi, x, recon_yoffset, recon_uvoffset, &rate,
&distortion, &intra_error, &single, &compound,
&hybrid);
// TODO Save these to add in only if MB coding mode is selected?
cpi->rd_single_diff += single;
cpi->rd_comp_diff += compound;
cpi->rd_hybrid_diff += hybrid;
if (xd->mode_info_context->mbmi.ref_frame)
{
unsigned char pred_context;
pred_context = get_pred_context( cm, xd, PRED_COMP );
if (xd->mode_info_context->mbmi.second_ref_frame == INTRA_FRAME)
cpi->single_pred_count[pred_context]++;
else
cpi->comp_pred_count[pred_context]++;
}
/* test code: set transform size based on mode selection */
if( cpi->common.txfm_mode == ALLOW_8X8
&& xd->mode_info_context->mbmi.mode != I8X8_PRED
&& xd->mode_info_context->mbmi.mode != B_PRED
40614062406340644065406640674068406940704071407240734074407540764077407840794080408140824083408440854086
&& xd->mode_info_context->mbmi.mode != SPLITMV)
{
xd->mode_info_context->mbmi.txfm_size = TX_8X8;
cpi->t8x8_count ++;
}
else
{
xd->mode_info_context->mbmi.txfm_size = TX_4X4;
cpi->t4x4_count++;
}
/* restore cpi->zbin_mode_boost_enabled */
cpi->zbin_mode_boost_enabled = zbin_mode_boost_enabled;
}
//else
// The non rd encode path has been deleted from this code base
// to simplify development
// vp8_pick_inter_mode
// Store metrics so they can be added in to totals if this mode is picked
x->mb_context[xd->mb_index].distortion = distortion;
x->mb_context[xd->mb_index].intra_error = intra_error;
return rate;
}