/*****************************************************************************
- * set.c: h264 encoder library
+ * set.c: quantization init
*****************************************************************************
- * Copyright (C) 2005 x264 project
+ * Copyright (C) 2005-2012 x264 project
*
* Authors: Loren Merritt <lorenm@u.washington.edu>
*
*
* You should have received a copy of the GNU General Public License
* along with this program; if not, write to the Free Software
- * Foundation, Inc., 59 Temple Place - Suite 330, Boston, MA 02111, USA.
+ * Foundation, Inc., 51 Franklin Street, Fifth Floor, Boston, MA 02111, USA.
+ *
+ * This program is also available under a commercial proprietary license.
+ * For more information, contact us at licensing@x264.com.
*****************************************************************************/
+#define _ISOC99_SOURCE
#include "common.h"
-static const int dequant4_scale[6][3] =
+#define SHIFT(x,s) ((s)<=0 ? (x)<<-(s) : ((x)+(1<<((s)-1)))>>(s))
+#define DIV(n,d) (((n) + ((d)>>1)) / (d))
+
+static const uint8_t dequant4_scale[6][3] =
{
{ 10, 13, 16 },
{ 11, 14, 18 },
{ 16, 20, 25 },
{ 18, 23, 29 }
};
-static const int quant4_scale[6][3] =
+static const uint16_t quant4_scale[6][3] =
{
{ 13107, 8066, 5243 },
{ 11916, 7490, 4660 },
{ 7282, 4559, 2893 },
};
-static const int quant8_scan[16] =
+static const uint8_t quant8_scan[16] =
{
0,3,4,3, 3,1,5,1, 4,5,2,5, 3,1,5,1
};
-static const int dequant8_scale[6][6] =
+static const uint8_t dequant8_scale[6][6] =
{
{ 20, 18, 32, 19, 25, 24 },
{ 22, 19, 35, 21, 28, 26 },
{ 32, 28, 51, 30, 40, 38 },
{ 36, 32, 58, 34, 46, 43 },
};
-static const int quant8_scale[6][6] =
+static const uint16_t quant8_scale[6][6] =
{
{ 13107, 11428, 20972, 12222, 16777, 15481 },
{ 11916, 10826, 19174, 11058, 14980, 14290 },
{ 7282, 6428, 11570, 6830, 9118, 8640 }
};
-void x264_cqm_init( x264_t *h )
+int x264_cqm_init( x264_t *h )
{
int def_quant4[6][16];
int def_quant8[6][64];
int def_dequant4[6][16];
int def_dequant8[6][64];
- int q, i, i_list;
+ int quant4_mf[4][6][16];
+ int quant8_mf[4][6][64];
+ int deadzone[4] = { 32 - h->param.analyse.i_luma_deadzone[1],
+ 32 - h->param.analyse.i_luma_deadzone[0],
+ 32 - 11, 32 - 21 };
+ int max_qp_err = -1;
+ int max_chroma_qp_err = -1;
+ int min_qp_err = QP_MAX+1;
+ int num_8x8_lists = h->sps->i_chroma_format_idc == CHROMA_444 ? 4 : 2; /* Checkasm may segfault if optimized out by --chroma-format */
- for( q = 0; q < 6; q++ )
+ for( int i = 0; i < 4 + num_8x8_lists; i++ )
{
- for( i = 0; i < 16; i++ )
+ int size = i<4 ? 16 : 64;
+ int j;
+ for( j = (i<4 ? 0 : 4); j < i; j++ )
+ if( !memcmp( h->pps->scaling_list[i], h->pps->scaling_list[j], size*sizeof(uint8_t) ) )
+ break;
+ if( j < i )
+ {
+ h-> quant4_mf[i] = h-> quant4_mf[j];
+ h->dequant4_mf[i] = h->dequant4_mf[j];
+ h->unquant4_mf[i] = h->unquant4_mf[j];
+ }
+ else
+ {
+ CHECKED_MALLOC( h-> quant4_mf[i], (QP_MAX+1)*size*sizeof(udctcoef) );
+ CHECKED_MALLOC( h->dequant4_mf[i], 6*size*sizeof(int) );
+ CHECKED_MALLOC( h->unquant4_mf[i], (QP_MAX+1)*size*sizeof(int) );
+ }
+
+ for( j = (i<4 ? 0 : 4); j < i; j++ )
+ if( deadzone[j&3] == deadzone[i&3] &&
+ !memcmp( h->pps->scaling_list[i], h->pps->scaling_list[j], size*sizeof(uint8_t) ) )
+ break;
+ if( j < i )
+ {
+ h->quant4_bias[i] = h->quant4_bias[j];
+ h->quant4_bias0[i] = h->quant4_bias0[j];
+ }
+ else
+ {
+ CHECKED_MALLOC( h->quant4_bias[i], (QP_MAX+1)*size*sizeof(udctcoef) );
+ CHECKED_MALLOC( h->quant4_bias0[i], (QP_MAX+1)*size*sizeof(udctcoef) );
+ }
+ }
+
+ for( int q = 0; q < 6; q++ )
+ {
+ for( int i = 0; i < 16; i++ )
{
int j = (i&1) + ((i>>2)&1);
def_dequant4[q][i] = dequant4_scale[q][j];
def_quant4[q][i] = quant4_scale[q][j];
}
- for( i = 0; i < 64; i++ )
+ for( int i = 0; i < 64; i++ )
{
int j = quant8_scan[((i>>1)&12) | (i&3)];
def_dequant8[q][i] = dequant8_scale[q][j];
}
}
- for( q = 0; q < 6; q++ )
+ for( int q = 0; q < 6; q++ )
+ {
+ for( int i_list = 0; i_list < 4; i_list++ )
+ for( int i = 0; i < 16; i++ )
+ {
+ h->dequant4_mf[i_list][q][i] = def_dequant4[q][i] * h->pps->scaling_list[i_list][i];
+ quant4_mf[i_list][q][i] = DIV(def_quant4[q][i] * 16, h->pps->scaling_list[i_list][i]);
+ }
+ for( int i_list = 0; i_list < num_8x8_lists; i_list++ )
+ for( int i = 0; i < 64; i++ )
+ {
+ h->dequant8_mf[i_list][q][i] = def_dequant8[q][i] * h->pps->scaling_list[4+i_list][i];
+ quant8_mf[i_list][q][i] = DIV(def_quant8[q][i] * 16, h->pps->scaling_list[4+i_list][i]);
+ }
+ }
+ for( int q = 0; q < QP_MAX+1; q++ )
{
- for( i_list = 0; i_list < 4; i_list++ )
- for( i = 0; i < 16; i++ )
+ int j;
+ for( int i_list = 0; i_list < 4; i_list++ )
+ for( int i = 0; i < 16; i++ )
{
- h->dequant4_mf[i_list][q][0][i] = def_dequant4[q][i] * h->pps->scaling_list[i_list][i];
- h-> quant4_mf[i_list][q][0][i] = def_quant4[q][i] * 16 / h->pps->scaling_list[i_list][i];
+ h->unquant4_mf[i_list][q][i] = (1ULL << (q/6 + 15 + 8)) / quant4_mf[i_list][q%6][i];
+ h->quant4_mf[i_list][q][i] = j = SHIFT(quant4_mf[i_list][q%6][i], q/6 - 1);
+ if( !j )
+ {
+ min_qp_err = X264_MIN( min_qp_err, q );
+ continue;
+ }
+ // round to nearest, unless that would cause the deadzone to be negative
+ h->quant4_bias[i_list][q][i] = X264_MIN( DIV(deadzone[i_list]<<10, j), (1<<15)/j );
+ h->quant4_bias0[i_list][q][i] = (1<<15)/j;
+ if( j > 0xffff && q > max_qp_err && (i_list == CQM_4IY || i_list == CQM_4PY) )
+ max_qp_err = q;
+ if( j > 0xffff && q > max_chroma_qp_err && (i_list == CQM_4IC || i_list == CQM_4PC) )
+ max_chroma_qp_err = q;
}
- for( i_list = 0; i_list < 2; i_list++ )
- for( i = 0; i < 64; i++ )
+ if( h->param.analyse.b_transform_8x8 )
+ for( int i_list = 0; i_list < num_8x8_lists; i_list++ )
+ for( int i = 0; i < 64; i++ )
+ {
+ h->unquant8_mf[i_list][q][i] = (1ULL << (q/6 + 16 + 8)) / quant8_mf[i_list][q%6][i];
+ j = SHIFT(quant8_mf[i_list][q%6][i], q/6);
+ h->quant8_mf[i_list][q][i] = (uint16_t)j;
+
+ if( !j )
+ {
+ min_qp_err = X264_MIN( min_qp_err, q );
+ continue;
+ }
+ h->quant8_bias[i_list][q][i] = X264_MIN( DIV(deadzone[i_list]<<10, j), (1<<15)/j );
+ h->quant8_bias0[i_list][q][i] = (1<<15)/j;
+ if( j > 0xffff && q > max_qp_err && (i_list == CQM_8IY || i_list == CQM_8PY) )
+ max_qp_err = q;
+ if( j > 0xffff && q > max_chroma_qp_err && (i_list == CQM_8IC || i_list == CQM_8PC) )
+ max_chroma_qp_err = q;
+ }
+ }
+
+ /* Emergency mode denoising. */
+ x264_emms();
+ CHECKED_MALLOC( h->nr_offset_emergency, sizeof(*h->nr_offset_emergency)*(QP_MAX-QP_MAX_SPEC) );
+ for( int q = 0; q < QP_MAX - QP_MAX_SPEC; q++ )
+ for( int cat = 0; cat < 3 + CHROMA444; cat++ )
+ {
+ int dct8x8 = cat&1;
+ int size = dct8x8 ? 64 : 16;
+ udctcoef *nr_offset = h->nr_offset_emergency[q][cat];
+ /* Denoise chroma first (due to h264's chroma QP offset), then luma, then DC. */
+ int dc_threshold = (QP_MAX-QP_MAX_SPEC)*2/3;
+ int luma_threshold = (QP_MAX-QP_MAX_SPEC)*2/3;
+ int chroma_threshold = 0;
+
+ for( int i = 0; i < size; i++ )
{
- h->dequant8_mf[i_list][q][0][i] = def_dequant8[q][i] * h->pps->scaling_list[4+i_list][i];
- h-> quant8_mf[i_list][q][0][i] = def_quant8[q][i] * 16 / h->pps->scaling_list[4+i_list][i];
+ int max = (1 << (7 + BIT_DEPTH)) - 1;
+ /* True "emergency mode": remove all DCT coefficients */
+ if( q == QP_MAX - QP_MAX_SPEC - 1 )
+ {
+ nr_offset[i] = max;
+ continue;
+ }
+
+ int thresh = i == 0 ? dc_threshold : cat >= 2 ? chroma_threshold : luma_threshold;
+ if( q < thresh )
+ {
+ nr_offset[i] = 0;
+ continue;
+ }
+ double pos = (double)(q-thresh+1) / (QP_MAX - QP_MAX_SPEC - thresh);
+
+ /* XXX: this math is largely tuned for /dev/random input. */
+ double start = dct8x8 ? h->unquant8_mf[CQM_8PY][QP_MAX_SPEC][i]
+ : h->unquant4_mf[CQM_4PY][QP_MAX_SPEC][i];
+ /* Formula chosen as an exponential scale to vaguely mimic the effects
+ * of a higher quantizer. */
+ double bias = (pow( 2, pos*(QP_MAX - QP_MAX_SPEC)/10. )*0.003-0.003) * start;
+ nr_offset[i] = X264_MIN( bias + 0.5, max );
}
+ }
+
+ if( !h->mb.b_lossless )
+ {
+ while( h->chroma_qp_table[SPEC_QP(h->param.rc.i_qp_min)] <= max_chroma_qp_err )
+ h->param.rc.i_qp_min++;
+ if( min_qp_err <= h->param.rc.i_qp_max )
+ h->param.rc.i_qp_max = min_qp_err-1;
+ if( max_qp_err >= h->param.rc.i_qp_min )
+ h->param.rc.i_qp_min = max_qp_err+1;
+ /* If long level-codes aren't allowed, we need to allow QP high enough to avoid them. */
+ if( !h->param.b_cabac && h->sps->i_profile_idc < PROFILE_HIGH )
+ while( h->chroma_qp_table[SPEC_QP(h->param.rc.i_qp_max)] <= 12 || h->param.rc.i_qp_max <= 12 )
+ h->param.rc.i_qp_max++;
+ if( h->param.rc.i_qp_min > h->param.rc.i_qp_max )
+ {
+ x264_log( h, X264_LOG_ERROR, "Impossible QP constraints for CQM (min=%d, max=%d)\n", h->param.rc.i_qp_min, h->param.rc.i_qp_max );
+ return -1;
+ }
+ }
+ return 0;
+fail:
+ x264_cqm_delete( h );
+ return -1;
+}
+
+#define CQM_DELETE( n, max )\
+ for( int i = 0; i < (max); i++ )\
+ {\
+ int j;\
+ for( j = 0; j < i; j++ )\
+ if( h->quant##n##_mf[i] == h->quant##n##_mf[j] )\
+ break;\
+ if( j == i )\
+ {\
+ x264_free( h-> quant##n##_mf[i] );\
+ x264_free( h->dequant##n##_mf[i] );\
+ x264_free( h->unquant##n##_mf[i] );\
+ }\
+ for( j = 0; j < i; j++ )\
+ if( h->quant##n##_bias[i] == h->quant##n##_bias[j] )\
+ break;\
+ if( j == i )\
+ {\
+ x264_free( h->quant##n##_bias[i] );\
+ x264_free( h->quant##n##_bias0[i] );\
+ }\
+ }
+
+void x264_cqm_delete( x264_t *h )
+{
+ CQM_DELETE( 4, 4 );
+ CQM_DELETE( 8, CHROMA444 ? 4 : 2 );
+ x264_free( h->nr_offset_emergency );
+}
+
+static int x264_cqm_parse_jmlist( x264_t *h, const char *buf, const char *name,
+ uint8_t *cqm, const uint8_t *jvt, int length )
+{
+ int i;
+
+ char *p = strstr( buf, name );
+ if( !p )
+ {
+ memset( cqm, 16, length );
+ return 0;
+ }
+
+ p += strlen( name );
+ if( *p == 'U' || *p == 'V' )
+ p++;
+
+ char *nextvar = strstr( p, "INT" );
+
+ for( i = 0; i < length && (p = strpbrk( p, " \t\n," )) && (p = strpbrk( p, "0123456789" )); i++ )
+ {
+ int coef = -1;
+ sscanf( p, "%d", &coef );
+ if( i == 0 && coef == 0 )
+ {
+ memcpy( cqm, jvt, length );
+ return 0;
+ }
+ if( coef < 1 || coef > 255 )
+ {
+ x264_log( h, X264_LOG_ERROR, "bad coefficient in list '%s'\n", name );
+ return -1;
+ }
+ cqm[i] = coef;
}
+
+ if( (nextvar && p > nextvar) || i != length )
+ {
+ x264_log( h, X264_LOG_ERROR, "not enough coefficients in list '%s'\n", name );
+ return -1;
+ }
+
+ return 0;
+}
+
+int x264_cqm_parse_file( x264_t *h, const char *filename )
+{
+ char *p;
+ int b_error = 0;
+
+ h->param.i_cqm_preset = X264_CQM_CUSTOM;
+
+ char *buf = x264_slurp_file( filename );
+ if( !buf )
+ {
+ x264_log( h, X264_LOG_ERROR, "can't open file '%s'\n", filename );
+ return -1;
+ }
+
+ while( (p = strchr( buf, '#' )) != NULL )
+ memset( p, ' ', strcspn( p, "\n" ) );
+
+ b_error |= x264_cqm_parse_jmlist( h, buf, "INTRA4X4_LUMA", h->param.cqm_4iy, x264_cqm_jvt4i, 16 );
+ b_error |= x264_cqm_parse_jmlist( h, buf, "INTER4X4_LUMA", h->param.cqm_4py, x264_cqm_jvt4p, 16 );
+ b_error |= x264_cqm_parse_jmlist( h, buf, "INTRA4X4_CHROMA", h->param.cqm_4ic, x264_cqm_jvt4i, 16 );
+ b_error |= x264_cqm_parse_jmlist( h, buf, "INTER4X4_CHROMA", h->param.cqm_4pc, x264_cqm_jvt4p, 16 );
+ b_error |= x264_cqm_parse_jmlist( h, buf, "INTRA8X8_LUMA", h->param.cqm_8iy, x264_cqm_jvt8i, 64 );
+ b_error |= x264_cqm_parse_jmlist( h, buf, "INTER8X8_LUMA", h->param.cqm_8py, x264_cqm_jvt8p, 64 );
+ if( CHROMA444 )
+ {
+ b_error |= x264_cqm_parse_jmlist( h, buf, "INTRA8X8_CHROMA", h->param.cqm_8ic, x264_cqm_jvt8i, 64 );
+ b_error |= x264_cqm_parse_jmlist( h, buf, "INTER8X8_CHROMA", h->param.cqm_8pc, x264_cqm_jvt8p, 64 );
+ }
+
+ x264_free( buf );
+ return b_error;
}
projects / x264 / blobdiff