--- sad.c 2003/01/04 06:14:33 1.7.2.3 +++ sad.c 2010/12/28 19:19:43 1.18 @@ -1,53 +1,47 @@ -/************************************************************************** +/***************************************************************************** * - * XVID MPEG-4 VIDEO CODEC - * sum of absolute difference + * XVID MPEG-4 VIDEO CODEC + * - Sum Of Absolute Difference related code - * - * This program is an implementation of a part of one or more MPEG-4 - * Video tools as specified in ISO/IEC 14496-2 standard. Those intending - * to use this software module in hardware or software products are - * advised that its use may infringe existing patents or copyrights, and - * any such use would be at such party's own risk. The original - * developer of this software module and his/her company, and subsequent - * editors and their companies, will have no liability for use of this - * software or modifications or derivatives thereof. + * Copyright(C) 2001-2010 Peter Ross + * 2010 Michael Militzer * - * This program is free software; you can redistribute it and/or modify - * it under the terms of the GNU General Public License as published by - * the Free Software Foundation; either version 2 of the License, or - * (at your option) any later version. + * This program is free software ; you can redistribute it and/or modify + * it under the terms of the GNU General Public License as published by + * the Free Software Foundation ; either version 2 of the License, or + * (at your option) any later version. * - * This program is distributed in the hope that it will be useful, - * but WITHOUT ANY WARRANTY; without even the implied warranty of - * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the - * GNU General Public License for more details. + * This program is distributed in the hope that it will be useful, + * but WITHOUT ANY WARRANTY ; without even the implied warranty of + * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the + * GNU General Public License for more details. * - * 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., 675 Mass Ave, Cambridge, MA 02139, USA. + * 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 02111-1307 USA * - *************************************************************************/ - -/************************************************************************** - * - * History: - * - * 14.02.2002 added sad16bi_c() - * 10.11.2001 initial version; (c)2001 peter ross + * $Id: sad.c,v 1.18 2010/12/28 19:19:43 Isibaar Exp $ * - *************************************************************************/ - + ****************************************************************************/ #include "../portab.h" #include "../global.h" #include "sad.h" +#include + sad16FuncPtr sad16; sad8FuncPtr sad8; sad16biFuncPtr sad16bi; -sad8biFuncPtr sad8bi; // not really sad16, but no difference in prototype +sad8biFuncPtr sad8bi; /* not really sad16, but no difference in prototype */ dev16FuncPtr dev16; sad16vFuncPtr sad16v; +sse8Func_16bitPtr sse8_16bit; +sse8Func_8bitPtr sse8_8bit; + +sseh8Func_16bitPtr sseh8_16bit; +coeff8_energyFunc_Ptr coeff8_energy; +blocksum8Func_Ptr blocksum8; sadInitFuncPtr sadInit; @@ -65,24 +59,24 @@ uint8_t const *ptr_ref = ref; for (j = 0; j < 16; j++) { - sad += ABS(ptr_cur[0] - ptr_ref[0]); - sad += ABS(ptr_cur[1] - ptr_ref[1]); - sad += ABS(ptr_cur[2] - ptr_ref[2]); - sad += ABS(ptr_cur[3] - ptr_ref[3]); - sad += ABS(ptr_cur[4] - ptr_ref[4]); - sad += ABS(ptr_cur[5] - ptr_ref[5]); - sad += ABS(ptr_cur[6] - ptr_ref[6]); - sad += ABS(ptr_cur[7] - ptr_ref[7]); - sad += ABS(ptr_cur[8] - ptr_ref[8]); - sad += ABS(ptr_cur[9] - ptr_ref[9]); - sad += ABS(ptr_cur[10] - ptr_ref[10]); - sad += ABS(ptr_cur[11] - ptr_ref[11]); - sad += ABS(ptr_cur[12] - ptr_ref[12]); - sad += ABS(ptr_cur[13] - ptr_ref[13]); - sad += ABS(ptr_cur[14] - ptr_ref[14]); - sad += ABS(ptr_cur[15] - ptr_ref[15]); + sad += abs(ptr_cur[0] - ptr_ref[0]); + sad += abs(ptr_cur[1] - ptr_ref[1]); + sad += abs(ptr_cur[2] - ptr_ref[2]); + sad += abs(ptr_cur[3] - ptr_ref[3]); + sad += abs(ptr_cur[4] - ptr_ref[4]); + sad += abs(ptr_cur[5] - ptr_ref[5]); + sad += abs(ptr_cur[6] - ptr_ref[6]); + sad += abs(ptr_cur[7] - ptr_ref[7]); + sad += abs(ptr_cur[8] - ptr_ref[8]); + sad += abs(ptr_cur[9] - ptr_ref[9]); + sad += abs(ptr_cur[10] - ptr_ref[10]); + sad += abs(ptr_cur[11] - ptr_ref[11]); + sad += abs(ptr_cur[12] - ptr_ref[12]); + sad += abs(ptr_cur[13] - ptr_ref[13]); + sad += abs(ptr_cur[14] - ptr_ref[14]); + sad += abs(ptr_cur[15] - ptr_ref[15]); - if (sad >= best_sad) + if (sad >= best_sad) return sad; ptr_cur += stride; @@ -111,14 +105,7 @@ for (i = 0; i < 16; i++) { int pixel = (ptr_ref1[i] + ptr_ref2[i] + 1) / 2; - - if (pixel < 0) { - pixel = 0; - } else if (pixel > 255) { - pixel = 255; - } - - sad += ABS(ptr_cur[i] - pixel); + sad += abs(ptr_cur[i] - pixel); } ptr_cur += stride; @@ -148,14 +135,7 @@ for (i = 0; i < 8; i++) { int pixel = (ptr_ref1[i] + ptr_ref2[i] + 1) / 2; - - if (pixel < 0) { - pixel = 0; - } else if (pixel > 255) { - pixel = 255; - } - - sad += ABS(ptr_cur[i] - pixel); + sad += abs(ptr_cur[i] - pixel); } ptr_cur += stride; @@ -182,15 +162,15 @@ for (j = 0; j < 8; j++) { - sad += ABS(ptr_cur[0] - ptr_ref[0]); - sad += ABS(ptr_cur[1] - ptr_ref[1]); - sad += ABS(ptr_cur[2] - ptr_ref[2]); - sad += ABS(ptr_cur[3] - ptr_ref[3]); - sad += ABS(ptr_cur[4] - ptr_ref[4]); - sad += ABS(ptr_cur[5] - ptr_ref[5]); - sad += ABS(ptr_cur[6] - ptr_ref[6]); - sad += ABS(ptr_cur[7] - ptr_ref[7]); - + sad += abs(ptr_cur[0] - ptr_ref[0]); + sad += abs(ptr_cur[1] - ptr_ref[1]); + sad += abs(ptr_cur[2] - ptr_ref[2]); + sad += abs(ptr_cur[3] - ptr_ref[3]); + sad += abs(ptr_cur[4] - ptr_ref[4]); + sad += abs(ptr_cur[5] - ptr_ref[5]); + sad += abs(ptr_cur[6] - ptr_ref[6]); + sad += abs(ptr_cur[7] - ptr_ref[7]); + ptr_cur += stride; ptr_ref += stride; @@ -227,7 +207,7 @@ for (j = 0; j < 16; j++) { for (i = 0; i < 16; i++) - dev += ABS(*(ptr_cur + i) - (int32_t) mean); + dev += abs(*(ptr_cur + i) - (int32_t) mean); ptr_cur += stride; @@ -236,29 +216,29 @@ return dev; } -uint32_t sad16v_c(const uint8_t * const cur, - const uint8_t * const ref, - const uint32_t stride, +uint32_t sad16v_c(const uint8_t * const cur, + const uint8_t * const ref, + const uint32_t stride, int32_t *sad) { sad[0] = sad8(cur, ref, stride); sad[1] = sad8(cur + 8, ref + 8, stride); sad[2] = sad8(cur + 8*stride, ref + 8*stride, stride); sad[3] = sad8(cur + 8*stride + 8, ref + 8*stride + 8, stride); - + return sad[0]+sad[1]+sad[2]+sad[3]; } -uint32_t sad32v_c(const uint8_t * const cur, - const uint8_t * const ref, - const uint32_t stride, +uint32_t sad32v_c(const uint8_t * const cur, + const uint8_t * const ref, + const uint32_t stride, int32_t *sad) { sad[0] = sad16(cur, ref, stride, 256*4096); - sad[1] = sad16(cur + 8, ref + 8, stride, 256*4096); - sad[2] = sad16(cur + 8*stride, ref + 8*stride, stride, 256*4096); - sad[3] = sad16(cur + 8*stride + 8, ref + 8*stride + 8, stride, 256*4096); - + sad[1] = sad16(cur + 16, ref + 16, stride, 256*4096); + sad[2] = sad16(cur + 16*stride, ref + 16*stride, stride, 256*4096); + sad[3] = sad16(cur + 16*stride + 16, ref + 16*stride + 16, stride, 256*4096); + return sad[0]+sad[1]+sad[2]+sad[3]; } @@ -295,7 +275,7 @@ for (i = 0; i < 16; i++) { - sad += ABS(*(ptr_cur + i) - *(ptr_ref + i) - mean); + sad += abs(*(ptr_cur + i) - *(ptr_ref + i) - mean); if (sad >= best_sad) { return MRSAD16_CORRFACTOR * sad; } @@ -303,7 +283,143 @@ } return MRSAD16_CORRFACTOR * sad; +} + +uint32_t +sse8_16bit_c(const int16_t * b1, + const int16_t * b2, + const uint32_t stride) +{ + int i; + int sse = 0; + + for (i=0; i<8; i++) { + sse += (b1[0] - b2[0])*(b1[0] - b2[0]); + sse += (b1[1] - b2[1])*(b1[1] - b2[1]); + sse += (b1[2] - b2[2])*(b1[2] - b2[2]); + sse += (b1[3] - b2[3])*(b1[3] - b2[3]); + sse += (b1[4] - b2[4])*(b1[4] - b2[4]); + sse += (b1[5] - b2[5])*(b1[5] - b2[5]); + sse += (b1[6] - b2[6])*(b1[6] - b2[6]); + sse += (b1[7] - b2[7])*(b1[7] - b2[7]); + + b1 = (const int16_t*)((int8_t*)b1+stride); + b2 = (const int16_t*)((int8_t*)b2+stride); + } + + return(sse); +} + +uint32_t +sse8_8bit_c(const uint8_t * b1, + const uint8_t * b2, + const uint32_t stride) +{ + int i; + int sse = 0; + + for (i=0; i<8; i++) { + sse += (b1[0] - b2[0])*(b1[0] - b2[0]); + sse += (b1[1] - b2[1])*(b1[1] - b2[1]); + sse += (b1[2] - b2[2])*(b1[2] - b2[2]); + sse += (b1[3] - b2[3])*(b1[3] - b2[3]); + sse += (b1[4] - b2[4])*(b1[4] - b2[4]); + sse += (b1[5] - b2[5])*(b1[5] - b2[5]); + sse += (b1[6] - b2[6])*(b1[6] - b2[6]); + sse += (b1[7] - b2[7])*(b1[7] - b2[7]); + + b1 = b1+stride; + b2 = b2+stride; + } + + return(sse); +} + +/* PSNR-HVS-M helper functions */ + +static const int16_t iMask_Coeff[64] = { + 0, 29788, 32767, 20479, 13653, 8192, 6425, 5372, + 27306, 27306, 23405, 17246, 12603, 5650, 5461, 5958, + 23405, 25205, 20479, 13653, 8192, 5749, 4749, 5851, + 23405, 19275, 14894, 11299, 6425, 3766, 4096, 5285, + 18204, 14894, 8856, 5851, 4819, 3006, 3181, 4255, + 13653, 9362, 5958, 5120, 4045, 3151, 2900, 3562, + 6687, 5120, 4201, 3766, 3181, 2708, 2730, 3244, + 4551, 3562, 3449, 3344, 2926, 3277, 3181, 3310 +}; + +/* Calculate CSF weighted energy of DCT coefficients */ + +uint32_t +coeff8_energy_c(const int16_t * dct) +{ + int x, y; + uint32_t sum_a = 0; + + for (y = 0; y < 8; y += 2) { + for (x = 0; x < 8; x += 2) { + int16_t a0 = ((dct[y*8+x]<<4) * iMask_Coeff[y*8+x]) >> 16; + int16_t a1 = ((dct[y*8+x+1]<<4) * iMask_Coeff[y*8+x+1]) >> 16; + int16_t a2 = ((dct[(y+1)*8+x]<<4) * iMask_Coeff[(y+1)*8+x]) >> 16; + int16_t a3 = ((dct[(y+1)*8+x+1]<<4) * iMask_Coeff[(y+1)*8+x+1]) >> 16; + + sum_a += ((a0*a0 + a1*a1 + a2*a2 + a3*a3) >> 3); + } + } + + return sum_a; } +/* Calculate MSE of DCT coeffs reduced by masking effect */ +uint32_t +sseh8_16bit_c(const int16_t * cur, const int16_t * ref, uint16_t mask) +{ + int j, i; + uint32_t mse_h = 0; + + for (j = 0; j < 8; j++) { + for (i = 0; i < 8; i++) { + uint32_t t = (mask * Inv_iMask_Coeff[j*8+i] + 32) >> 7; + uint16_t u = abs(cur[j*8+i] - ref[j*8+i]) << 4; + uint16_t thresh = (t < 65536) ? t : 65535; + + if (u < thresh) + u = 0; /* The error is not perceivable */ + else + u -= thresh; + + u = ((u + iCSF_Round[j*8 + i]) * iCSF_Coeff[j*8 + i]) >> 16; + + mse_h += (uint32_t) (u * u); + } + } + + return mse_h; +} + +/* Sums all pixels of 8x8 block */ + +uint32_t +blocksum8_c(const uint8_t * cur, int stride, uint16_t sums[4], uint32_t squares[4]) +{ + int i, j; + uint32_t sum = 0; + + sums[0] = sums[1] = sums[2] = sums[3] = 0; + squares[0] = squares[1] = squares[2] = squares[3] = 0; + + for (j = 0; j < 8; j++) { + for (i = 0; i < 8; i++) { + uint8_t p = cur[j*stride + i]; + + sums[(j>>2)*2 + (i>>2)] += p; + squares[(j>>2)*2 + (i>>2)] += p*p; + + sum += p; + } + } + + return sum; +}