vpx/vp9/common/vp9_entropy.c

1657 lines
62 KiB
C
Raw Normal View History

/*
* Copyright (c) 2010 The WebM project authors. All Rights Reserved.
2010-05-18 17:58:33 +02:00
*
* 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.
2010-05-18 17:58:33 +02:00
*/
#include <stdio.h>
#include "vp9/common/vp9_entropy.h"
2010-05-18 17:58:33 +02:00
#include "string.h"
#include "vp9/common/vp9_blockd.h"
#include "vp9/common/vp9_onyxc_int.h"
#include "vp9/common/vp9_entropymode.h"
#include "vpx_mem/vpx_mem.h"
#include "vpx/vpx_integer.h"
#include "vp9/common/vp9_coefupdateprobs.h"
2010-05-18 17:58:33 +02:00
const int vp9_i8x8_block[4] = {0, 2, 8, 10};
DECLARE_ALIGNED(16, const uint8_t, vp9_norm[256]) = {
0, 7, 6, 6, 5, 5, 5, 5, 4, 4, 4, 4, 4, 4, 4, 4,
3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3,
2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2,
2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2,
1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1,
1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1,
1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1,
1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 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, 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, 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, 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, 0, 0, 0, 0,
0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0
};
// Unified coefficient band structure used by all block sizes
DECLARE_ALIGNED(16, const int, vp9_coef_bands[32]) = {
0, 1, 2, 3, 3, 3, 4, 4,
4, 4, 4, 4, 4, 4, 4, 5,
5, 5, 5, 5, 5, 5, 5, 5,
5, 5, 5, 5, 5, 5, 5, 5
};
DECLARE_ALIGNED(16, const int, vp9_coef_bands4x4[16]) = {
0, 1, 2, 3, 3, 3, 4, 4,
4, 4, 5, 5, 5, 5, 5, 5
};
DECLARE_ALIGNED(16, const uint8_t, vp9_pt_energy_class[MAX_ENTROPY_TOKENS]) = {
0, 1, 2, 3, 3, 4, 4, 5, 5, 5, 5, 5
};
DECLARE_ALIGNED(16, const int, vp9_default_zig_zag1d_4x4[16]) = {
0, 1, 4, 8,
5, 2, 3, 6,
9, 12, 13, 10,
7, 11, 14, 15,
2010-05-18 17:58:33 +02:00
};
DECLARE_ALIGNED(16, const int, vp9_col_scan_4x4[16]) = {
0, 4, 8, 12,
1, 5, 9, 13,
2, 6, 10, 14,
3, 7, 11, 15
};
DECLARE_ALIGNED(16, const int, vp9_row_scan_4x4[16]) = {
0, 1, 2, 3,
4, 5, 6, 7,
8, 9, 10, 11,
12, 13, 14, 15
};
DECLARE_ALIGNED(64, const int, vp9_default_zig_zag1d_8x8[64]) = {
0, 1, 8, 16, 9, 2, 3, 10, 17, 24, 32, 25, 18, 11, 4, 5,
12, 19, 26, 33, 40, 48, 41, 34, 27, 20, 13, 6, 7, 14, 21, 28,
35, 42, 49, 56, 57, 50, 43, 36, 29, 22, 15, 23, 30, 37, 44, 51,
58, 59, 52, 45, 38, 31, 39, 46, 53, 60, 61, 54, 47, 55, 62, 63,
};
2010-05-18 17:58:33 +02:00
DECLARE_ALIGNED(16, const int, vp9_default_zig_zag1d_16x16[256]) = {
0, 1, 16, 32, 17, 2, 3, 18,
33, 48, 64, 49, 34, 19, 4, 5,
20, 35, 50, 65, 80, 96, 81, 66,
51, 36, 21, 6, 7, 22, 37, 52,
67, 82, 97, 112, 128, 113, 98, 83,
68, 53, 38, 23, 8, 9, 24, 39,
54, 69, 84, 99, 114, 129, 144, 160,
145, 130, 115, 100, 85, 70, 55, 40,
25, 10, 11, 26, 41, 56, 71, 86,
101, 116, 131, 146, 161, 176, 192, 177,
162, 147, 132, 117, 102, 87, 72, 57,
42, 27, 12, 13, 28, 43, 58, 73,
88, 103, 118, 133, 148, 163, 178, 193,
208, 224, 209, 194, 179, 164, 149, 134,
119, 104, 89, 74, 59, 44, 29, 14,
15, 30, 45, 60, 75, 90, 105, 120,
135, 150, 165, 180, 195, 210, 225, 240,
241, 226, 211, 196, 181, 166, 151, 136,
121, 106, 91, 76, 61, 46, 31, 47,
62, 77, 92, 107, 122, 137, 152, 167,
182, 197, 212, 227, 242, 243, 228, 213,
198, 183, 168, 153, 138, 123, 108, 93,
78, 63, 79, 94, 109, 124, 139, 154,
169, 184, 199, 214, 229, 244, 245, 230,
215, 200, 185, 170, 155, 140, 125, 110,
95, 111, 126, 141, 156, 171, 186, 201,
216, 231, 246, 247, 232, 217, 202, 187,
172, 157, 142, 127, 143, 158, 173, 188,
203, 218, 233, 248, 249, 234, 219, 204,
189, 174, 159, 175, 190, 205, 220, 235,
250, 251, 236, 221, 206, 191, 207, 222,
237, 252, 253, 238, 223, 239, 254, 255,
};
32x32 transform for superblocks. This adds Debargha's DCT/DWT hybrid and a regular 32x32 DCT, and adds code all over the place to wrap that in the bitstream/encoder/decoder/RD. Some implementation notes (these probably need careful review): - token range is extended by 1 bit, since the value range out of this transform is [-16384,16383]. - the coefficients coming out of the FDCT are manually scaled back by 1 bit, or else they won't fit in int16_t (they are 17 bits). Because of this, the RD error scoring does not right-shift the MSE score by two (unlike for 4x4/8x8/16x16). - to compensate for this loss in precision, the quantizer is halved also. This is currently a little hacky. - FDCT and IDCT is double-only right now. Needs a fixed-point impl. - There are no default probabilities for the 32x32 transform yet; I'm simply using the 16x16 luma ones. A future commit will add newly generated probabilities for all transforms. - No ADST version. I don't think we'll add one for this level; if an ADST is desired, transform-size selection can scale back to 16x16 or lower, and use an ADST at that level. Additional notes specific to Debargha's DWT/DCT hybrid: - coefficient scale is different for the top/left 16x16 (DCT-over-DWT) block than for the rest (DWT pixel differences) of the block. Therefore, RD error scoring isn't easily scalable between coefficient and pixel domain. Thus, unfortunately, we need to compute the RD distortion in the pixel domain until we figure out how to scale these appropriately. Change-Id: I00386f20f35d7fabb19aba94c8162f8aee64ef2b
2012-12-07 23:45:05 +01:00
DECLARE_ALIGNED(16, const int, vp9_default_zig_zag1d_32x32[1024]) = {
0, 1, 32, 64, 33, 2, 3, 34, 65, 96, 128, 97, 66, 35, 4, 5, 36, 67, 98, 129, 160, 192, 161, 130, 99, 68, 37, 6, 7, 38, 69, 100,
131, 162, 193, 224, 256, 225, 194, 163, 132, 101, 70, 39, 8, 9, 40, 71, 102, 133, 164, 195, 226, 257, 288, 320, 289, 258, 227, 196, 165, 134, 103, 72,
41, 10, 11, 42, 73, 104, 135, 166, 197, 228, 259, 290, 321, 352, 384, 353, 322, 291, 260, 229, 198, 167, 136, 105, 74, 43, 12, 13, 44, 75, 106, 137,
168, 199, 230, 261, 292, 323, 354, 385, 416, 448, 417, 386, 355, 324, 293, 262, 231, 200, 169, 138, 107, 76, 45, 14, 15, 46, 77, 108, 139, 170, 201, 232,
263, 294, 325, 356, 387, 418, 449, 480, 512, 481, 450, 419, 388, 357, 326, 295, 264, 233, 202, 171, 140, 109, 78, 47, 16, 17, 48, 79, 110, 141, 172, 203,
234, 265, 296, 327, 358, 389, 420, 451, 482, 513, 544, 576, 545, 514, 483, 452, 421, 390, 359, 328, 297, 266, 235, 204, 173, 142, 111, 80, 49, 18, 19, 50,
81, 112, 143, 174, 205, 236, 267, 298, 329, 360, 391, 422, 453, 484, 515, 546, 577, 608, 640, 609, 578, 547, 516, 485, 454, 423, 392, 361, 330, 299, 268, 237,
206, 175, 144, 113, 82, 51, 20, 21, 52, 83, 114, 145, 176, 207, 238, 269, 300, 331, 362, 393, 424, 455, 486, 517, 548, 579, 610, 641, 672, 704, 673, 642,
611, 580, 549, 518, 487, 456, 425, 394, 363, 332, 301, 270, 239, 208, 177, 146, 115, 84, 53, 22, 23, 54, 85, 116, 147, 178, 209, 240, 271, 302, 333, 364,
395, 426, 457, 488, 519, 550, 581, 612, 643, 674, 705, 736, 768, 737, 706, 675, 644, 613, 582, 551, 520, 489, 458, 427, 396, 365, 334, 303, 272, 241, 210, 179,
148, 117, 86, 55, 24, 25, 56, 87, 118, 149, 180, 211, 242, 273, 304, 335, 366, 397, 428, 459, 490, 521, 552, 583, 614, 645, 676, 707, 738, 769, 800, 832,
801, 770, 739, 708, 677, 646, 615, 584, 553, 522, 491, 460, 429, 398, 367, 336, 305, 274, 243, 212, 181, 150, 119, 88, 57, 26, 27, 58, 89, 120, 151, 182,
213, 244, 275, 306, 337, 368, 399, 430, 461, 492, 523, 554, 585, 616, 647, 678, 709, 740, 771, 802, 833, 864, 896, 865, 834, 803, 772, 741, 710, 679, 648, 617,
586, 555, 524, 493, 462, 431, 400, 369, 338, 307, 276, 245, 214, 183, 152, 121, 90, 59, 28, 29, 60, 91, 122, 153, 184, 215, 246, 277, 308, 339, 370, 401,
432, 463, 494, 525, 556, 587, 618, 649, 680, 711, 742, 773, 804, 835, 866, 897, 928, 960, 929, 898, 867, 836, 805, 774, 743, 712, 681, 650, 619, 588, 557, 526,
495, 464, 433, 402, 371, 340, 309, 278, 247, 216, 185, 154, 123, 92, 61, 30, 31, 62, 93, 124, 155, 186, 217, 248, 279, 310, 341, 372, 403, 434, 465, 496,
527, 558, 589, 620, 651, 682, 713, 744, 775, 806, 837, 868, 899, 930, 961, 992, 993, 962, 931, 900, 869, 838, 807, 776, 745, 714, 683, 652, 621, 590, 559, 528,
497, 466, 435, 404, 373, 342, 311, 280, 249, 218, 187, 156, 125, 94, 63, 95, 126, 157, 188, 219, 250, 281, 312, 343, 374, 405, 436, 467, 498, 529, 560, 591,
622, 653, 684, 715, 746, 777, 808, 839, 870, 901, 932, 963, 994, 995, 964, 933, 902, 871, 840, 809, 778, 747, 716, 685, 654, 623, 592, 561, 530, 499, 468, 437,
406, 375, 344, 313, 282, 251, 220, 189, 158, 127, 159, 190, 221, 252, 283, 314, 345, 376, 407, 438, 469, 500, 531, 562, 593, 624, 655, 686, 717, 748, 779, 810,
841, 872, 903, 934, 965, 996, 997, 966, 935, 904, 873, 842, 811, 780, 749, 718, 687, 656, 625, 594, 563, 532, 501, 470, 439, 408, 377, 346, 315, 284, 253, 222,
191, 223, 254, 285, 316, 347, 378, 409, 440, 471, 502, 533, 564, 595, 626, 657, 688, 719, 750, 781, 812, 843, 874, 905, 936, 967, 998, 999, 968, 937, 906, 875,
844, 813, 782, 751, 720, 689, 658, 627, 596, 565, 534, 503, 472, 441, 410, 379, 348, 317, 286, 255, 287, 318, 349, 380, 411, 442, 473, 504, 535, 566, 597, 628,
659, 690, 721, 752, 783, 814, 845, 876, 907, 938, 969, 1000, 1001, 970, 939, 908, 877, 846, 815, 784, 753, 722, 691, 660, 629, 598, 567, 536, 505, 474, 443, 412,
381, 350, 319, 351, 382, 413, 444, 475, 506, 537, 568, 599, 630, 661, 692, 723, 754, 785, 816, 847, 878, 909, 940, 971, 1002, 1003, 972, 941, 910, 879, 848, 817,
786, 755, 724, 693, 662, 631, 600, 569, 538, 507, 476, 445, 414, 383, 415, 446, 477, 508, 539, 570, 601, 632, 663, 694, 725, 756, 787, 818, 849, 880, 911, 942,
973, 1004, 1005, 974, 943, 912, 881, 850, 819, 788, 757, 726, 695, 664, 633, 602, 571, 540, 509, 478, 447, 479, 510, 541, 572, 603, 634, 665, 696, 727, 758, 789,
820, 851, 882, 913, 944, 975, 1006, 1007, 976, 945, 914, 883, 852, 821, 790, 759, 728, 697, 666, 635, 604, 573, 542, 511, 543, 574, 605, 636, 667, 698, 729, 760,
791, 822, 853, 884, 915, 946, 977, 1008, 1009, 978, 947, 916, 885, 854, 823, 792, 761, 730, 699, 668, 637, 606, 575, 607, 638, 669, 700, 731, 762, 793, 824, 855,
886, 917, 948, 979, 1010, 1011, 980, 949, 918, 887, 856, 825, 794, 763, 732, 701, 670, 639, 671, 702, 733, 764, 795, 826, 857, 888, 919, 950, 981, 1012, 1013, 982,
951, 920, 889, 858, 827, 796, 765, 734, 703, 735, 766, 797, 828, 859, 890, 921, 952, 983, 1014, 1015, 984, 953, 922, 891, 860, 829, 798, 767, 799, 830, 861, 892,
923, 954, 985, 1016, 1017, 986, 955, 924, 893, 862, 831, 863, 894, 925, 956, 987, 1018, 1019, 988, 957, 926, 895, 927, 958, 989, 1020, 1021, 990, 959, 991, 1022, 1023,
};
2010-05-18 17:58:33 +02:00
/* Array indices are identical to previously-existing CONTEXT_NODE indices */
const vp9_tree_index vp9_coef_tree[ 22] = /* corresponding _CONTEXT_NODEs */
2010-05-18 17:58:33 +02:00
{
-DCT_EOB_TOKEN, 2, /* 0 = EOB */
-ZERO_TOKEN, 4, /* 1 = ZERO */
-ONE_TOKEN, 6, /* 2 = ONE */
8, 12, /* 3 = LOW_VAL */
-TWO_TOKEN, 10, /* 4 = TWO */
-THREE_TOKEN, -FOUR_TOKEN, /* 5 = THREE */
14, 16, /* 6 = HIGH_LOW */
-DCT_VAL_CATEGORY1, -DCT_VAL_CATEGORY2, /* 7 = CAT_ONE */
18, 20, /* 8 = CAT_THREEFOUR */
-DCT_VAL_CATEGORY3, -DCT_VAL_CATEGORY4, /* 9 = CAT_THREE */
-DCT_VAL_CATEGORY5, -DCT_VAL_CATEGORY6 /* 10 = CAT_FIVE */
2010-05-18 17:58:33 +02:00
};
struct vp9_token_struct vp9_coef_encodings[MAX_ENTROPY_TOKENS];
2010-05-18 17:58:33 +02:00
/* Trees for extra bits. Probabilities are constant and
do not depend on previously encoded bits */
static const vp9_prob Pcat1[] = { 159};
static const vp9_prob Pcat2[] = { 165, 145};
static const vp9_prob Pcat3[] = { 173, 148, 140};
static const vp9_prob Pcat4[] = { 176, 155, 140, 135};
static const vp9_prob Pcat5[] = { 180, 157, 141, 134, 130};
static const vp9_prob Pcat6[] = {
32x32 transform for superblocks. This adds Debargha's DCT/DWT hybrid and a regular 32x32 DCT, and adds code all over the place to wrap that in the bitstream/encoder/decoder/RD. Some implementation notes (these probably need careful review): - token range is extended by 1 bit, since the value range out of this transform is [-16384,16383]. - the coefficients coming out of the FDCT are manually scaled back by 1 bit, or else they won't fit in int16_t (they are 17 bits). Because of this, the RD error scoring does not right-shift the MSE score by two (unlike for 4x4/8x8/16x16). - to compensate for this loss in precision, the quantizer is halved also. This is currently a little hacky. - FDCT and IDCT is double-only right now. Needs a fixed-point impl. - There are no default probabilities for the 32x32 transform yet; I'm simply using the 16x16 luma ones. A future commit will add newly generated probabilities for all transforms. - No ADST version. I don't think we'll add one for this level; if an ADST is desired, transform-size selection can scale back to 16x16 or lower, and use an ADST at that level. Additional notes specific to Debargha's DWT/DCT hybrid: - coefficient scale is different for the top/left 16x16 (DCT-over-DWT) block than for the rest (DWT pixel differences) of the block. Therefore, RD error scoring isn't easily scalable between coefficient and pixel domain. Thus, unfortunately, we need to compute the RD distortion in the pixel domain until we figure out how to scale these appropriately. Change-Id: I00386f20f35d7fabb19aba94c8162f8aee64ef2b
2012-12-07 23:45:05 +01:00
254, 254, 254, 252, 249, 243, 230, 196, 177, 153, 140, 133, 130, 129
};
2010-05-18 17:58:33 +02:00
#if CONFIG_CODE_NONZEROCOUNT
const vp9_tree_index vp9_nzc4x4_tree[2 * NZC4X4_NODES] = {
-NZC_0, 2,
4, 6,
-NZC_1, -NZC_2,
-NZC_3TO4, 8,
-NZC_5TO8, -NZC_9TO16,
};
struct vp9_token_struct vp9_nzc4x4_encodings[NZC4X4_TOKENS];
const vp9_tree_index vp9_nzc8x8_tree[2 * NZC8X8_NODES] = {
-NZC_0, 2,
4, 6,
-NZC_1, -NZC_2,
8, 10,
-NZC_3TO4, -NZC_5TO8,
-NZC_9TO16, 12,
-NZC_17TO32, -NZC_33TO64,
};
struct vp9_token_struct vp9_nzc8x8_encodings[NZC8X8_TOKENS];
const vp9_tree_index vp9_nzc16x16_tree[2 * NZC16X16_NODES] = {
-NZC_0, 2,
4, 6,
-NZC_1, -NZC_2,
8, 10,
-NZC_3TO4, -NZC_5TO8,
12, 14,
-NZC_9TO16, -NZC_17TO32,
-NZC_33TO64, 16,
-NZC_65TO128, -NZC_129TO256,
};
struct vp9_token_struct vp9_nzc16x16_encodings[NZC16X16_TOKENS];
const vp9_tree_index vp9_nzc32x32_tree[2 * NZC32X32_NODES] = {
-NZC_0, 2,
4, 6,
-NZC_1, -NZC_2,
8, 10,
-NZC_3TO4, -NZC_5TO8,
12, 14,
-NZC_9TO16, -NZC_17TO32,
16, 18,
-NZC_33TO64, -NZC_65TO128,
-NZC_129TO256, 20,
-NZC_257TO512, -NZC_513TO1024,
};
struct vp9_token_struct vp9_nzc32x32_encodings[NZC32X32_TOKENS];
const int vp9_extranzcbits[NZC32X32_TOKENS] = {
0, 0, 0, 1, 2, 3, 4, 5, 6, 7, 8, 9
};
const int vp9_basenzcvalue[NZC32X32_TOKENS] = {
0, 1, 2, 3, 5, 9, 17, 33, 65, 129, 257, 513
};
#endif // CONFIG_CODE_NONZEROCOUNT
32x32 transform for superblocks. This adds Debargha's DCT/DWT hybrid and a regular 32x32 DCT, and adds code all over the place to wrap that in the bitstream/encoder/decoder/RD. Some implementation notes (these probably need careful review): - token range is extended by 1 bit, since the value range out of this transform is [-16384,16383]. - the coefficients coming out of the FDCT are manually scaled back by 1 bit, or else they won't fit in int16_t (they are 17 bits). Because of this, the RD error scoring does not right-shift the MSE score by two (unlike for 4x4/8x8/16x16). - to compensate for this loss in precision, the quantizer is halved also. This is currently a little hacky. - FDCT and IDCT is double-only right now. Needs a fixed-point impl. - There are no default probabilities for the 32x32 transform yet; I'm simply using the 16x16 luma ones. A future commit will add newly generated probabilities for all transforms. - No ADST version. I don't think we'll add one for this level; if an ADST is desired, transform-size selection can scale back to 16x16 or lower, and use an ADST at that level. Additional notes specific to Debargha's DWT/DCT hybrid: - coefficient scale is different for the top/left 16x16 (DCT-over-DWT) block than for the rest (DWT pixel differences) of the block. Therefore, RD error scoring isn't easily scalable between coefficient and pixel domain. Thus, unfortunately, we need to compute the RD distortion in the pixel domain until we figure out how to scale these appropriately. Change-Id: I00386f20f35d7fabb19aba94c8162f8aee64ef2b
2012-12-07 23:45:05 +01:00
static vp9_tree_index cat1[2], cat2[4], cat3[6], cat4[8], cat5[10], cat6[28];
2010-05-18 17:58:33 +02:00
static void init_bit_tree(vp9_tree_index *p, int n) {
int i = 0;
2010-05-18 17:58:33 +02:00
while (++i < n) {
p[0] = p[1] = i << 1;
p += 2;
}
2010-05-18 17:58:33 +02:00
p[0] = p[1] = 0;
2010-05-18 17:58:33 +02:00
}
static void init_bit_trees() {
init_bit_tree(cat1, 1);
init_bit_tree(cat2, 2);
init_bit_tree(cat3, 3);
init_bit_tree(cat4, 4);
init_bit_tree(cat5, 5);
32x32 transform for superblocks. This adds Debargha's DCT/DWT hybrid and a regular 32x32 DCT, and adds code all over the place to wrap that in the bitstream/encoder/decoder/RD. Some implementation notes (these probably need careful review): - token range is extended by 1 bit, since the value range out of this transform is [-16384,16383]. - the coefficients coming out of the FDCT are manually scaled back by 1 bit, or else they won't fit in int16_t (they are 17 bits). Because of this, the RD error scoring does not right-shift the MSE score by two (unlike for 4x4/8x8/16x16). - to compensate for this loss in precision, the quantizer is halved also. This is currently a little hacky. - FDCT and IDCT is double-only right now. Needs a fixed-point impl. - There are no default probabilities for the 32x32 transform yet; I'm simply using the 16x16 luma ones. A future commit will add newly generated probabilities for all transforms. - No ADST version. I don't think we'll add one for this level; if an ADST is desired, transform-size selection can scale back to 16x16 or lower, and use an ADST at that level. Additional notes specific to Debargha's DWT/DCT hybrid: - coefficient scale is different for the top/left 16x16 (DCT-over-DWT) block than for the rest (DWT pixel differences) of the block. Therefore, RD error scoring isn't easily scalable between coefficient and pixel domain. Thus, unfortunately, we need to compute the RD distortion in the pixel domain until we figure out how to scale these appropriately. Change-Id: I00386f20f35d7fabb19aba94c8162f8aee64ef2b
2012-12-07 23:45:05 +01:00
init_bit_tree(cat6, 14);
2010-05-18 17:58:33 +02:00
}
vp9_extra_bit_struct vp9_extra_bits[12] = {
{ 0, 0, 0, 0},
{ 0, 0, 0, 1},
{ 0, 0, 0, 2},
{ 0, 0, 0, 3},
{ 0, 0, 0, 4},
{ cat1, Pcat1, 1, 5},
{ cat2, Pcat2, 2, 7},
{ cat3, Pcat3, 3, 11},
{ cat4, Pcat4, 4, 19},
{ cat5, Pcat5, 5, 35},
32x32 transform for superblocks. This adds Debargha's DCT/DWT hybrid and a regular 32x32 DCT, and adds code all over the place to wrap that in the bitstream/encoder/decoder/RD. Some implementation notes (these probably need careful review): - token range is extended by 1 bit, since the value range out of this transform is [-16384,16383]. - the coefficients coming out of the FDCT are manually scaled back by 1 bit, or else they won't fit in int16_t (they are 17 bits). Because of this, the RD error scoring does not right-shift the MSE score by two (unlike for 4x4/8x8/16x16). - to compensate for this loss in precision, the quantizer is halved also. This is currently a little hacky. - FDCT and IDCT is double-only right now. Needs a fixed-point impl. - There are no default probabilities for the 32x32 transform yet; I'm simply using the 16x16 luma ones. A future commit will add newly generated probabilities for all transforms. - No ADST version. I don't think we'll add one for this level; if an ADST is desired, transform-size selection can scale back to 16x16 or lower, and use an ADST at that level. Additional notes specific to Debargha's DWT/DCT hybrid: - coefficient scale is different for the top/left 16x16 (DCT-over-DWT) block than for the rest (DWT pixel differences) of the block. Therefore, RD error scoring isn't easily scalable between coefficient and pixel domain. Thus, unfortunately, we need to compute the RD distortion in the pixel domain until we figure out how to scale these appropriately. Change-Id: I00386f20f35d7fabb19aba94c8162f8aee64ef2b
2012-12-07 23:45:05 +01:00
{ cat6, Pcat6, 14, 67},
{ 0, 0, 0, 0}
2010-05-18 17:58:33 +02:00
};
#include "vp9/common/vp9_default_coef_probs.h"
2010-05-18 17:58:33 +02:00
// This function updates and then returns n AC coefficient context
// This is currently a placeholder function to allow experimentation
// using various context models based on the energy earlier tokens
// within the current block.
//
// For now it just returns the previously used context.
int vp9_get_coef_context(int * recent_energy, int token) {
// int token_energy;
// int av_energy;
/*token_energy = ((token != DCT_EOB_TOKEN) ? token : 0);
if (!token_energy) {
if (!(*recent_energy)) {
av_energy = 0;
} else {
av_energy = 1;
}
} else {
av_energy = ((token_energy + *recent_energy + 1) >> 1) + 1;
if (av_energy > DCT_VAL_CATEGORY6)
av_energy = DCT_VAL_CATEGORY6;
}
*recent_energy = token_energy;*/
return vp9_pt_energy_class[token];
};
void vp9_default_coef_probs(VP9_COMMON *pc) {
#if CONFIG_CODE_NONZEROCOUNT
#ifdef NZC_DEFAULT_COUNTS
int h, g;
for (h = 0; h < MAX_NZC_CONTEXTS; ++h) {
for (g = 0; g < REF_TYPES; ++g) {
int i;
unsigned int branch_ct4x4[NZC4X4_NODES][2];
unsigned int branch_ct8x8[NZC8X8_NODES][2];
unsigned int branch_ct16x16[NZC16X16_NODES][2];
unsigned int branch_ct32x32[NZC32X32_NODES][2];
for (i = 0; i < BLOCK_TYPES; ++i) {
vp9_tree_probs_from_distribution(
vp9_nzc4x4_tree,
pc->fc.nzc_probs_4x4[h][g][i], branch_ct4x4,
default_nzc_counts_4x4[h][g][i], 0);
}
for (i = 0; i < BLOCK_TYPES; ++i) {
vp9_tree_probs_from_distribution(
vp9_nzc8x8_tree,
pc->fc.nzc_probs_8x8[h][g][i], branch_ct8x8,
default_nzc_counts_8x8[h][g][i], 0);
}
for (i = 0; i < BLOCK_TYPES; ++i) {
vp9_tree_probs_from_distribution(
vp9_nzc16x16_tree,
pc->fc.nzc_probs_16x16[h][g][i], branch_ct16x16,
default_nzc_counts_16x16[h][g][i], 0);
}
for (i = 0; i < BLOCK_TYPES; ++i) {
vp9_tree_probs_from_distribution(
vp9_nzc32x32_tree,
pc->fc.nzc_probs_32x32[h][g][i], branch_ct32x32,
default_nzc_counts_32x32[h][g][i], 0);
}
}
}
#else
vpx_memcpy(pc->fc.nzc_probs_4x4, default_nzc_probs_4x4,
sizeof(pc->fc.nzc_probs_4x4));
vpx_memcpy(pc->fc.nzc_probs_8x8, default_nzc_probs_8x8,
sizeof(pc->fc.nzc_probs_8x8));
vpx_memcpy(pc->fc.nzc_probs_16x16, default_nzc_probs_16x16,
sizeof(pc->fc.nzc_probs_16x16));
vpx_memcpy(pc->fc.nzc_probs_32x32, default_nzc_probs_32x32,
sizeof(pc->fc.nzc_probs_32x32));
#endif
vpx_memcpy(pc->fc.nzc_pcat_probs, default_nzc_pcat_probs,
sizeof(pc->fc.nzc_pcat_probs));
#endif // CONFIG_CODE_NONZEROCOUNTyy
vpx_memcpy(pc->fc.coef_probs_4x4, default_coef_probs_4x4,
sizeof(pc->fc.coef_probs_4x4));
vpx_memcpy(pc->fc.coef_probs_8x8, default_coef_probs_8x8,
sizeof(pc->fc.coef_probs_8x8));
vpx_memcpy(pc->fc.coef_probs_16x16, default_coef_probs_16x16,
sizeof(pc->fc.coef_probs_16x16));
vpx_memcpy(pc->fc.coef_probs_32x32, default_coef_probs_32x32,
sizeof(pc->fc.coef_probs_32x32));
}
2010-05-18 17:58:33 +02:00
void vp9_coef_tree_initialize() {
init_bit_trees();
vp9_tokens_from_tree(vp9_coef_encodings, vp9_coef_tree);
#if CONFIG_CODE_NONZEROCOUNT
vp9_tokens_from_tree(vp9_nzc4x4_encodings, vp9_nzc4x4_tree);
vp9_tokens_from_tree(vp9_nzc8x8_encodings, vp9_nzc8x8_tree);
vp9_tokens_from_tree(vp9_nzc16x16_encodings, vp9_nzc16x16_tree);
vp9_tokens_from_tree(vp9_nzc32x32_encodings, vp9_nzc32x32_tree);
#endif
}
#if CONFIG_CODE_NONZEROCOUNT
#define mb_in_cur_tile(cm, mb_row, mb_col) \
((mb_col) >= (cm)->cur_tile_mb_col_start && \
(mb_col) <= (cm)->cur_tile_mb_col_end && \
(mb_row) >= 0)
#define choose_nzc_context(nzc_exp, t2, t1) \
((nzc_exp) >= (t2) ? 2 : (nzc_exp) >= (t1) ? 1 : 0)
#define NZC_T2_32X32 (16 << 6)
#define NZC_T1_32X32 (4 << 6)
#define NZC_T2_16X16 (12 << 6)
#define NZC_T1_16X16 (3 << 6)
#define NZC_T2_8X8 (8 << 6)
#define NZC_T1_8X8 (2 << 6)
#define NZC_T2_4X4 (4 << 6)
#define NZC_T1_4X4 (1 << 6)
// Transforms a mb16 block index to a sb64 block index
static inline int mb16_to_sb64_index(int mb_row, int mb_col, int block) {
int r = (mb_row & 3);
int c = (mb_col & 3);
int b;
if (block < 16) { // Y
int ib = block >> 2;
int jb = block & 3;
ib += r * 4;
jb += c * 4;
b = ib * 16 + jb;
assert(b < 256);
return b;
} else { // UV
int base = block - (block & 3);
int ib = (block - base) >> 1;
int jb = (block - base) & 1;
ib += r * 2;
jb += c * 2;
b = base * 16 + ib * 8 + jb;
assert(b >= 256 && b < 384);
return b;
}
}
// Transforms a mb16 block index to a sb32 block index
static inline int mb16_to_sb32_index(int mb_row, int mb_col, int block) {
int r = (mb_row & 1);
int c = (mb_col & 1);
int b;
if (block < 16) { // Y
int ib = block >> 2;
int jb = block & 3;
ib += r * 4;
jb += c * 4;
b = ib * 8 + jb;
assert(b < 64);
return b;
} else { // UV
int base = block - (block & 3);
int ib = (block - base) >> 1;
int jb = (block - base) & 1;
ib += r * 2;
jb += c * 2;
b = base * 4 + ib * 4 + jb;
assert(b >= 64 && b < 96);
return b;
}
}
static inline int block_to_txfm_index(int block, TX_SIZE tx_size, int s) {
// s is the log of the number of 4x4 blocks in each row/col of larger block
int b, ib, jb, nb;
ib = block >> s;
jb = block - (ib << s);
ib >>= tx_size;
jb >>= tx_size;
nb = 1 << (s - tx_size);
b = (ib * nb + jb) << (2 * tx_size);
return b;
}
/* BEGIN - Helper functions to get the y nzcs */
static unsigned int get_nzc_4x4_y_sb64(MB_MODE_INFO *mi, int block) {
int b;
assert(block < 256);
b = block_to_txfm_index(block, mi->txfm_size, 4);
assert(b < 256);
return mi->nzcs[b] << (6 - 2 * mi->txfm_size);
}
static unsigned int get_nzc_4x4_y_sb32(MB_MODE_INFO *mi, int block) {
int b;
assert(block < 64);
b = block_to_txfm_index(block, mi->txfm_size, 3);
assert(b < 64);
return mi->nzcs[b] << (6 - 2 * mi->txfm_size);
}
static unsigned int get_nzc_4x4_y_mb16(MB_MODE_INFO *mi, int block) {
int b;
assert(block < 16);
b = block_to_txfm_index(block, mi->txfm_size, 2);
assert(b < 16);
return mi->nzcs[b] << (6 - 2 * mi->txfm_size);
}
/* END - Helper functions to get the y nzcs */
/* Function to get y nzc where block index is in mb16 terms */
static unsigned int get_nzc_4x4_y(VP9_COMMON *cm, MODE_INFO *m,
int mb_row, int mb_col, int block) {
// NOTE: All values returned are at 64 times the true value at 4x4 scale
MB_MODE_INFO *const mi = &m->mbmi;
const int mis = cm->mode_info_stride;
if (mi->mb_skip_coeff || !mb_in_cur_tile(cm, mb_row, mb_col))
return 0;
if (mi->sb_type == BLOCK_SIZE_SB64X64) {
int r = mb_row & 3;
int c = mb_col & 3;
m -= c + r * mis;
if (m->mbmi.mb_skip_coeff || !mb_in_cur_tile(cm, mb_row - r, mb_col - c))
return 0;
else
return get_nzc_4x4_y_sb64(
&m->mbmi, mb16_to_sb64_index(mb_row, mb_col, block));
} else if (mi->sb_type == BLOCK_SIZE_SB32X32) {
int r = mb_row & 1;
int c = mb_col & 1;
m -= c + r * mis;
if (m->mbmi.mb_skip_coeff || !mb_in_cur_tile(cm, mb_row - r, mb_col - c))
return 0;
else
return get_nzc_4x4_y_sb32(
&m->mbmi, mb16_to_sb32_index(mb_row, mb_col, block));
} else {
if (m->mbmi.mb_skip_coeff || !mb_in_cur_tile(cm, mb_row, mb_col))
return 0;
return get_nzc_4x4_y_mb16(mi, block);
}
}
/* BEGIN - Helper functions to get the uv nzcs */
static unsigned int get_nzc_4x4_uv_sb64(MB_MODE_INFO *mi, int block) {
int b;
int base, uvtxfm_size;
assert(block >= 256 && block < 384);
uvtxfm_size = mi->txfm_size;
base = 256 + (block & 64);
block -= base;
b = base + block_to_txfm_index(block, uvtxfm_size, 3);
assert(b >= 256 && b < 384);
return mi->nzcs[b] << (6 - 2 * uvtxfm_size);
}
static unsigned int get_nzc_4x4_uv_sb32(MB_MODE_INFO *mi, int block) {
int b;
int base, uvtxfm_size;
assert(block >= 64 && block < 96);
if (mi->txfm_size == TX_32X32)
uvtxfm_size = TX_16X16;
else
uvtxfm_size = mi->txfm_size;
base = 64 + (block & 16);
block -= base;
b = base + block_to_txfm_index(block, uvtxfm_size, 2);
assert(b >= 64 && b < 96);
return mi->nzcs[b] << (6 - 2 * uvtxfm_size);
}
static unsigned int get_nzc_4x4_uv_mb16(MB_MODE_INFO *mi, int block) {
int b;
int base, uvtxfm_size;
assert(block >= 16 && block < 24);
if (mi->txfm_size == TX_8X8 &&
(mi->mode == SPLITMV || mi->mode == I8X8_PRED))
uvtxfm_size = TX_4X4;
else if (mi->txfm_size == TX_16X16)
uvtxfm_size = TX_8X8;
else
uvtxfm_size = mi->txfm_size;
base = 16 + (block & 4);
block -= base;
b = base + block_to_txfm_index(block, uvtxfm_size, 1);
assert(b >= 16 && b < 24);
return mi->nzcs[b] << (6 - 2 * uvtxfm_size);
}
/* END - Helper functions to get the uv nzcs */
/* Function to get uv nzc where block index is in mb16 terms */
static unsigned int get_nzc_4x4_uv(VP9_COMMON *cm, MODE_INFO *m,
int mb_row, int mb_col, int block) {
// NOTE: All values returned are at 64 times the true value at 4x4 scale
MB_MODE_INFO *const mi = &m->mbmi;
const int mis = cm->mode_info_stride;
if (mi->mb_skip_coeff || !mb_in_cur_tile(cm, mb_row, mb_col))
return 0;
if (mi->sb_type == BLOCK_SIZE_SB64X64) {
int r = mb_row & 3;
int c = mb_col & 3;
m -= c + r * mis;
if (m->mbmi.mb_skip_coeff || !mb_in_cur_tile(cm, mb_row - r, mb_col - c))
return 0;
else
return get_nzc_4x4_uv_sb64(
&m->mbmi, mb16_to_sb64_index(mb_row, mb_col, block));
} else if (mi->sb_type == BLOCK_SIZE_SB32X32) {
int r = mb_row & 1;
int c = mb_col & 1;
m -= c + r * mis;
if (m->mbmi.mb_skip_coeff || !mb_in_cur_tile(cm, mb_row - r, mb_col - c))
return 0;
else
return get_nzc_4x4_uv_sb32(
&m->mbmi, mb16_to_sb32_index(mb_row, mb_col, block));
} else {
return get_nzc_4x4_uv_mb16(mi, block);
}
}
int vp9_get_nzc_context_y_sb64(VP9_COMMON *cm, MODE_INFO *cur,
int mb_row, int mb_col, int block) {
// returns an index in [0, MAX_NZC_CONTEXTS - 1] to reflect how busy
// neighboring blocks are
int mis = cm->mode_info_stride;
int nzc_exp = 0;
TX_SIZE txfm_size = cur->mbmi.txfm_size;
assert(block < 256);
switch (txfm_size) {
case TX_32X32:
assert((block & 63) == 0);
if (block < 128) {
int o = (block >> 6) * 2;
nzc_exp =
get_nzc_4x4_y(cm, cur - mis + o, mb_row - 1, mb_col + o, 12) +
get_nzc_4x4_y(cm, cur - mis + o, mb_row - 1, mb_col + o, 13) +
get_nzc_4x4_y(cm, cur - mis + o, mb_row - 1, mb_col + o, 14) +
get_nzc_4x4_y(cm, cur - mis + o, mb_row - 1, mb_col + o, 15) +
get_nzc_4x4_y(cm, cur - mis + o + 1,
mb_row - 1, mb_col + o + 1, 12) +
get_nzc_4x4_y(cm, cur - mis + o + 1,
mb_row - 1, mb_col + o + 1, 13) +
get_nzc_4x4_y(cm, cur - mis + o + 1,
mb_row - 1, mb_col + o + 1, 14) +
get_nzc_4x4_y(cm, cur - mis + o + 1,
mb_row - 1, mb_col + o + 1, 15);
} else {
nzc_exp = cur->mbmi.nzcs[block - 128] << 3;
}
if ((block & 127) == 0) {
int o = (block >> 7) * 2;
nzc_exp +=
get_nzc_4x4_y(cm, cur - 1 + o * mis, mb_row + o, mb_col - 1, 3) +
get_nzc_4x4_y(cm, cur - 1 + o * mis, mb_row + o, mb_col - 1, 7) +
get_nzc_4x4_y(cm, cur - 1 + o * mis, mb_row + o, mb_col - 1, 11) +
get_nzc_4x4_y(cm, cur - 1 + o * mis, mb_row + o, mb_col - 1, 15) +
get_nzc_4x4_y(cm, cur - 1 + o * mis + mis,
mb_row + o + 1, mb_col - 1, 3) +
get_nzc_4x4_y(cm, cur - 1 + o * mis + mis,
mb_row + o + 1, mb_col - 1, 7) +
get_nzc_4x4_y(cm, cur - 1 + o * mis + mis,
mb_row + o + 1, mb_col - 1, 11) +
get_nzc_4x4_y(cm, cur - 1 + o * mis + mis,
mb_row + o + 1, mb_col - 1, 15);
} else {
nzc_exp += cur->mbmi.nzcs[block - 64] << 3;
}
nzc_exp <<= 2;
// Note nzc_exp is 64 times the average value expected at 32x32 scale
return choose_nzc_context(nzc_exp, NZC_T2_32X32, NZC_T1_32X32);
break;
case TX_16X16:
assert((block & 15) == 0);
if (block < 64) {
int o = block >> 4;
nzc_exp =
get_nzc_4x4_y(cm, cur - mis + o, mb_row - 1, mb_col + o, 12) +
get_nzc_4x4_y(cm, cur - mis + o, mb_row - 1, mb_col + o, 13) +
get_nzc_4x4_y(cm, cur - mis + o, mb_row - 1, mb_col + o, 14) +
get_nzc_4x4_y(cm, cur - mis + o, mb_row - 1, mb_col + o, 15);
} else {
nzc_exp = cur->mbmi.nzcs[block - 64] << 4;
}
if ((block & 63) == 0) {
int o = block >> 6;
nzc_exp +=
get_nzc_4x4_y(cm, cur - 1 + o * mis, mb_row + o, mb_col - 1, 3) +
get_nzc_4x4_y(cm, cur - 1 + o * mis, mb_row + o, mb_col - 1, 7) +
get_nzc_4x4_y(cm, cur - 1 + o * mis, mb_row + o, mb_col - 1, 11) +
get_nzc_4x4_y(cm, cur - 1 + o * mis, mb_row + o, mb_col - 1, 15);
} else {
nzc_exp += cur->mbmi.nzcs[block - 16] << 4;
}
nzc_exp <<= 1;
// Note nzc_exp is 64 times the average value expected at 16x16 scale
return choose_nzc_context(nzc_exp, NZC_T2_16X16, NZC_T1_16X16);
break;
case TX_8X8:
assert((block & 3) == 0);
if (block < 32) {
int o = block >> 3;
int p = ((block >> 2) & 1) ? 14 : 12;
nzc_exp =
get_nzc_4x4_y(cm, cur - mis + o, mb_row - 1, mb_col + o, p) +
get_nzc_4x4_y(cm, cur - mis + o, mb_row - 1, mb_col + o, p + 1);
} else {
nzc_exp = cur->mbmi.nzcs[block - 32] << 5;
}
if ((block & 31) == 0) {
int o = block >> 6;
int p = ((block >> 5) & 1) ? 11 : 3;
nzc_exp +=
get_nzc_4x4_y(cm, cur - 1 + o * mis, mb_row + o, mb_col - 1, p) +
get_nzc_4x4_y(cm, cur - 1 + o * mis, mb_row + o, mb_col - 1, p + 4);
} else {
nzc_exp += cur->mbmi.nzcs[block - 4] << 5;
}
// Note nzc_exp is 64 times the average value expected at 8x8 scale
return choose_nzc_context(nzc_exp, NZC_T2_8X8, NZC_T1_8X8);
break;
case TX_4X4:
if (block < 16) {
int o = block >> 2;
int p = block & 3;
nzc_exp = get_nzc_4x4_y(cm, cur - mis + o, mb_row - 1, mb_col + o,
12 + p);
} else {
nzc_exp = (cur->mbmi.nzcs[block - 16] << 6);
}
if ((block & 15) == 0) {
int o = block >> 6;
int p = (block >> 4) & 3;
nzc_exp += get_nzc_4x4_y(cm, cur - 1 + o * mis, mb_row + o, mb_col - 1,
3 + 4 * p);
} else {
nzc_exp += (cur->mbmi.nzcs[block - 1] << 6);
}
nzc_exp >>= 1;
// Note nzc_exp is 64 times the average value expected at 4x4 scale
return choose_nzc_context(nzc_exp, NZC_T2_4X4, NZC_T1_4X4);
break;
default:
return 0;
}
}
int vp9_get_nzc_context_y_sb32(VP9_COMMON *cm, MODE_INFO *cur,
int mb_row, int mb_col, int block) {
// returns an index in [0, MAX_NZC_CONTEXTS - 1] to reflect how busy
// neighboring blocks are
int mis = cm->mode_info_stride;
int nzc_exp = 0;
TX_SIZE txfm_size = cur->mbmi.txfm_size;
assert(block < 64);
switch (txfm_size) {
case TX_32X32:
assert(block == 0);
nzc_exp =
(get_nzc_4x4_y(cm, cur - mis, mb_row - 1, mb_col, 12) +
get_nzc_4x4_y(cm, cur - mis, mb_row - 1, mb_col, 13) +
get_nzc_4x4_y(cm, cur - mis, mb_row - 1, mb_col, 14) +
get_nzc_4x4_y(cm, cur - mis, mb_row - 1, mb_col, 15) +
get_nzc_4x4_y(cm, cur - mis + 1, mb_row - 1, mb_col + 1, 12) +
get_nzc_4x4_y(cm, cur - mis + 1, mb_row - 1, mb_col + 1, 13) +
get_nzc_4x4_y(cm, cur - mis + 1, mb_row - 1, mb_col + 1, 14) +
get_nzc_4x4_y(cm, cur - mis + 1, mb_row - 1, mb_col + 1, 15) +
get_nzc_4x4_y(cm, cur - 1, mb_row, mb_col - 1, 3) +
get_nzc_4x4_y(cm, cur - 1, mb_row, mb_col - 1, 7) +
get_nzc_4x4_y(cm, cur - 1, mb_row, mb_col - 1, 11) +
get_nzc_4x4_y(cm, cur - 1, mb_row, mb_col - 1, 15) +
get_nzc_4x4_y(cm, cur - 1 + mis, mb_row + 1, mb_col - 1, 3) +
get_nzc_4x4_y(cm, cur - 1 + mis, mb_row + 1, mb_col - 1, 7) +
get_nzc_4x4_y(cm, cur - 1 + mis, mb_row + 1, mb_col - 1, 11) +
get_nzc_4x4_y(cm, cur - 1 + mis, mb_row + 1, mb_col - 1, 15)) << 2;
// Note nzc_exp is 64 times the average value expected at 32x32 scale
return choose_nzc_context(nzc_exp, NZC_T2_32X32, NZC_T1_32X32);
break;
case TX_16X16:
assert((block & 15) == 0);
if (block < 32) {
int o = (block >> 4) & 1;
nzc_exp =
get_nzc_4x4_y(cm, cur - mis + o, mb_row - 1, mb_col + o, 12) +
get_nzc_4x4_y(cm, cur - mis + o, mb_row - 1, mb_col + o, 13) +
get_nzc_4x4_y(cm, cur - mis + o, mb_row - 1, mb_col + o, 14) +
get_nzc_4x4_y(cm, cur - mis + o, mb_row - 1, mb_col + o, 15);
} else {
nzc_exp = cur->mbmi.nzcs[block - 32] << 4;
}
if ((block & 31) == 0) {
int o = block >> 5;
nzc_exp +=
get_nzc_4x4_y(cm, cur - 1 + o * mis, mb_row + o, mb_col - 1, 3) +
get_nzc_4x4_y(cm, cur - 1 + o * mis, mb_row + o, mb_col - 1, 7) +
get_nzc_4x4_y(cm, cur - 1 + o * mis, mb_row + o, mb_col - 1, 11) +
get_nzc_4x4_y(cm, cur - 1 + o * mis, mb_row + o, mb_col - 1, 15);
} else {
nzc_exp += cur->mbmi.nzcs[block - 16] << 4;
}
nzc_exp <<= 1;
// Note nzc_exp is 64 times the average value expected at 16x16 scale
return choose_nzc_context(nzc_exp, NZC_T2_16X16, NZC_T1_16X16);
break;
case TX_8X8:
assert((block & 3) == 0);
if (block < 16) {
int o = block >> 3;
int p = ((block >> 2) & 1) ? 14 : 12;
nzc_exp =
get_nzc_4x4_y(cm, cur - mis + o, mb_row - 1, mb_col + o, p) +
get_nzc_4x4_y(cm, cur - mis + o, mb_row - 1, mb_col + o, p + 1);
} else {
nzc_exp = cur->mbmi.nzcs[block - 16] << 5;
}
if ((block & 15) == 0) {
int o = block >> 5;
int p = ((block >> 4) & 1) ? 11 : 3;
nzc_exp +=
get_nzc_4x4_y(cm, cur - 1 + o * mis, mb_row + o, mb_col - 1, p) +
get_nzc_4x4_y(cm, cur - 1 + o * mis, mb_row + o, mb_col - 1, p + 4);
} else {
nzc_exp += cur->mbmi.nzcs[block - 4] << 5;
}
// Note nzc_exp is 64 times the average value expected at 8x8 scale
return choose_nzc_context(nzc_exp, NZC_T2_8X8, NZC_T1_8X8);
break;
case TX_4X4:
if (block < 8) {
int o = block >> 2;
int p = block & 3;
nzc_exp = get_nzc_4x4_y(cm, cur - mis + o, mb_row - 1, mb_col + o,
12 + p);
} else {
nzc_exp = (cur->mbmi.nzcs[block - 8] << 6);
}
if ((block & 7) == 0) {
int o = block >> 5;
int p = (block >> 3) & 3;
nzc_exp += get_nzc_4x4_y(cm, cur - 1 + o * mis, mb_row + o, mb_col - 1,
3 + 4 * p);
} else {
nzc_exp += (cur->mbmi.nzcs[block - 1] << 6);
}
nzc_exp >>= 1;
// Note nzc_exp is 64 times the average value expected at 4x4 scale
return choose_nzc_context(nzc_exp, NZC_T2_4X4, NZC_T1_4X4);
break;
default:
return 0;
break;
}
}
int vp9_get_nzc_context_y_mb16(VP9_COMMON *cm, MODE_INFO *cur,
int mb_row, int mb_col, int block) {
// returns an index in [0, MAX_NZC_CONTEXTS - 1] to reflect how busy
// neighboring blocks are
int mis = cm->mode_info_stride;
int nzc_exp = 0;
TX_SIZE txfm_size = cur->mbmi.txfm_size;
assert(block < 16);
switch (txfm_size) {
case TX_16X16:
assert(block == 0);
nzc_exp =
get_nzc_4x4_y(cm, cur - mis, mb_row - 1, mb_col, 12) +
get_nzc_4x4_y(cm, cur - mis, mb_row - 1, mb_col, 13) +
get_nzc_4x4_y(cm, cur - mis, mb_row - 1, mb_col, 14) +
get_nzc_4x4_y(cm, cur - mis, mb_row - 1, mb_col, 15) +
get_nzc_4x4_y(cm, cur - 1, mb_row, mb_col - 1, 3) +
get_nzc_4x4_y(cm, cur - 1, mb_row, mb_col - 1, 7) +
get_nzc_4x4_y(cm, cur - 1, mb_row, mb_col - 1, 11) +
get_nzc_4x4_y(cm, cur - 1, mb_row, mb_col - 1, 15);
nzc_exp <<= 1;
// Note nzc_exp is 64 times the average value expected at 16x16 scale
return choose_nzc_context(nzc_exp, NZC_T2_16X16, NZC_T1_16X16);
case TX_8X8:
assert((block & 3) == 0);
if (block < 8) {
int p = ((block >> 2) & 1) ? 14 : 12;
nzc_exp =
get_nzc_4x4_y(cm, cur - mis, mb_row - 1, mb_col, p) +
get_nzc_4x4_y(cm, cur - mis, mb_row - 1, mb_col, p + 1);
} else {
nzc_exp = cur->mbmi.nzcs[block - 8] << 5;
}
if ((block & 7) == 0) {
int p = ((block >> 3) & 1) ? 11 : 3;
nzc_exp +=
get_nzc_4x4_y(cm, cur - 1, mb_row, mb_col - 1, p) +
get_nzc_4x4_y(cm, cur - 1, mb_row, mb_col - 1, p + 4);
} else {
nzc_exp += cur->mbmi.nzcs[block - 4] << 5;
}
// Note nzc_exp is 64 times the average value expected at 8x8 scale
return choose_nzc_context(nzc_exp, NZC_T2_8X8, NZC_T1_8X8);
case TX_4X4:
if (block < 4) {
int p = block & 3;
nzc_exp = get_nzc_4x4_y(cm, cur - mis, mb_row - 1, mb_col,
12 + p);
} else {
nzc_exp = (cur->mbmi.nzcs[block - 4] << 6);
}
if ((block & 3) == 0) {
int p = (block >> 2) & 3;
nzc_exp += get_nzc_4x4_y(cm, cur - 1, mb_row, mb_col - 1,
3 + 4 * p);
} else {
nzc_exp += (cur->mbmi.nzcs[block - 1] << 6);
}
nzc_exp >>= 1;
// Note nzc_exp is 64 times the average value expected at 4x4 scale
return choose_nzc_context(nzc_exp, NZC_T2_4X4, NZC_T1_4X4);
default:
return 0;
break;
}
}
int vp9_get_nzc_context_uv_sb64(VP9_COMMON *cm, MODE_INFO *cur,
int mb_row, int mb_col, int block) {
// returns an index in [0, MAX_NZC_CONTEXTS - 1] to reflect how busy
// neighboring blocks are
int mis = cm->mode_info_stride;
int nzc_exp = 0;
const int base = block - (block & 63);
const int boff = (block & 63);
const int base_mb16 = base >> 4;
TX_SIZE txfm_size = cur->mbmi.txfm_size;
TX_SIZE txfm_size_uv;
assert(block >= 256 && block < 384);
txfm_size_uv = txfm_size;
switch (txfm_size_uv) {
case TX_32X32:
assert(block == 256 || block == 320);
nzc_exp =
get_nzc_4x4_uv(cm, cur - mis, mb_row - 1, mb_col,
base_mb16 + 2) +
get_nzc_4x4_uv(cm, cur - mis, mb_row - 1, mb_col,
base_mb16 + 3) +
get_nzc_4x4_uv(cm, cur - mis + 1, mb_row - 1, mb_col + 1,
base_mb16 + 2) +
get_nzc_4x4_uv(cm, cur - mis + 1, mb_row - 1, mb_col + 1,
base_mb16 + 3) +
get_nzc_4x4_uv(cm, cur - mis + 2, mb_row - 1, mb_col + 2,
base_mb16 + 2) +
get_nzc_4x4_uv(cm, cur - mis + 2, mb_row - 1, mb_col + 2,
base_mb16 + 3) +
get_nzc_4x4_uv(cm, cur - mis + 3, mb_row - 1, mb_col + 3,
base_mb16 + 2) +
get_nzc_4x4_uv(cm, cur - mis + 3, mb_row - 1, mb_col + 3,
base_mb16 + 3) +
get_nzc_4x4_uv(cm, cur - 1, mb_row, mb_col - 1,
base_mb16 + 1) +
get_nzc_4x4_uv(cm, cur - 1, mb_row, mb_col - 1,
base_mb16 + 3) +
get_nzc_4x4_uv(cm, cur - 1 + mis, mb_row + 1, mb_col - 1,
base_mb16 + 1) +
get_nzc_4x4_uv(cm, cur - 1 + mis, mb_row + 1, mb_col - 1,
base_mb16 + 3) +
get_nzc_4x4_uv(cm, cur - 1 + 2 * mis, mb_row + 2, mb_col - 1,
base_mb16 + 1) +
get_nzc_4x4_uv(cm, cur - 1 + 2 * mis, mb_row + 2, mb_col - 1,
base_mb16 + 3) +
get_nzc_4x4_uv(cm, cur - 1 + 3 * mis, mb_row + 3, mb_col - 1,
base_mb16 + 1) +
get_nzc_4x4_uv(cm, cur - 1 + 3 * mis, mb_row + 3, mb_col - 1,
base_mb16 + 3);
nzc_exp <<= 2;
// Note nzc_exp is 64 times the average value expected at 32x32 scale
return choose_nzc_context(nzc_exp, NZC_T2_32X32, NZC_T1_32X32);
case TX_16X16:
// uv txfm_size 16x16
assert((block & 15) == 0);
if (boff < 32) {
int o = (boff >> 4) & 1;
nzc_exp =
get_nzc_4x4_uv(cm, cur - mis + o, mb_row - 1, mb_col + o,
base_mb16 + 2) +
get_nzc_4x4_uv(cm, cur - mis + o, mb_row - 1, mb_col + o,
base_mb16 + 3) +
get_nzc_4x4_uv(cm, cur - mis + o + 1, mb_row - 1, mb_col + o + 1,
base_mb16 + 2) +
get_nzc_4x4_uv(cm, cur - mis + o + 1, mb_row - 1, mb_col + o + 1,
base_mb16 + 3);
} else {
nzc_exp = cur->mbmi.nzcs[block - 32] << 4;
}
if ((boff & 31) == 0) {
int o = boff >> 5;
nzc_exp +=
get_nzc_4x4_uv(cm, cur - 1 + o * mis,
mb_row + o, mb_col - 1, base_mb16 + 1) +
get_nzc_4x4_uv(cm, cur - 1 + o * mis,
mb_row + o, mb_col - 1, base_mb16 + 3) +
get_nzc_4x4_uv(cm, cur - 1 + o * mis + mis,
mb_row + o + 1, mb_col - 1, base_mb16 + 1) +
get_nzc_4x4_uv(cm, cur - 1 + o * mis + mis,
mb_row + o + 1, mb_col - 1, base_mb16 + 3);
} else {
nzc_exp += cur->mbmi.nzcs[block - 16] << 4;
}
nzc_exp <<= 1;
// Note nzc_exp is 64 times the average value expected at 16x16 scale
return choose_nzc_context(nzc_exp, NZC_T2_16X16, NZC_T1_16X16);
case TX_8X8:
assert((block & 3) == 0);
if (boff < 16) {
int o = boff >> 2;
nzc_exp =
get_nzc_4x4_uv(cm, cur - mis + o, mb_row - 1, mb_col + o,
base_mb16 + 2) +
get_nzc_4x4_uv(cm, cur - mis + o, mb_row - 1, mb_col + o,
base_mb16 + 3);
} else {
nzc_exp = cur->mbmi.nzcs[block - 16] << 5;
}
if ((boff & 15) == 0) {
int o = boff >> 4;
nzc_exp +=
get_nzc_4x4_uv(cm, cur - 1 + o * mis, mb_row + o, mb_col - 1,
base_mb16 + 1) +
get_nzc_4x4_uv(cm, cur - 1 + o * mis, mb_row + o, mb_col - 1,
base_mb16 + 3);
} else {
nzc_exp += cur->mbmi.nzcs[block - 4] << 5;
}
// Note nzc_exp is 64 times the average value expected at 8x8 scale
return choose_nzc_context(nzc_exp, NZC_T2_8X8, NZC_T1_8X8);
case TX_4X4:
if (boff < 8) {
int o = boff >> 1;
int p = boff & 1;
nzc_exp = get_nzc_4x4_uv(cm, cur - mis + o, mb_row - 1, mb_col + o,
base_mb16 + 2 + p);
} else {
nzc_exp = (cur->mbmi.nzcs[block - 8] << 6);
}
if ((boff & 7) == 0) {
int o = boff >> 4;
int p = (boff >> 3) & 1;
nzc_exp += get_nzc_4x4_uv(cm, cur - 1 + o * mis, mb_row + o, mb_col - 1,
base_mb16 + 1 + 2 * p);
} else {
nzc_exp += (cur->mbmi.nzcs[block - 1] << 6);
}
nzc_exp >>= 1;
// Note nzc_exp is 64 times the average value expected at 4x4 scale
return choose_nzc_context(nzc_exp, NZC_T2_4X4, NZC_T1_4X4);
default:
return 0;
}
}
int vp9_get_nzc_context_uv_sb32(VP9_COMMON *cm, MODE_INFO *cur,
int mb_row, int mb_col, int block) {
// returns an index in [0, MAX_NZC_CONTEXTS - 1] to reflect how busy
// neighboring blocks are
int mis = cm->mode_info_stride;
int nzc_exp = 0;
const int base = block - (block & 15);
const int boff = (block & 15);
const int base_mb16 = base >> 2;
TX_SIZE txfm_size = cur->mbmi.txfm_size;
TX_SIZE txfm_size_uv;
assert(block >= 64 && block < 96);
if (txfm_size == TX_32X32)
txfm_size_uv = TX_16X16;
else
txfm_size_uv = txfm_size;
switch (txfm_size_uv) {
case TX_16X16:
// uv txfm_size 16x16
assert(block == 64 || block == 80);
nzc_exp =
get_nzc_4x4_uv(cm, cur - mis, mb_row - 1, mb_col,
base_mb16 + 2) +
get_nzc_4x4_uv(cm, cur - mis, mb_row - 1, mb_col,
base_mb16 + 3) +
get_nzc_4x4_uv(cm, cur - mis + 1, mb_row - 1, mb_col + 1,
base_mb16 + 2) +
get_nzc_4x4_uv(cm, cur - mis + 1, mb_row - 1, mb_col + 1,
base_mb16 + 3) +
get_nzc_4x4_uv(cm, cur - 1 + mis, mb_row, mb_col - 1,
base_mb16 + 1) +
get_nzc_4x4_uv(cm, cur - 1 + mis, mb_row, mb_col - 1,
base_mb16 + 3) +
get_nzc_4x4_uv(cm, cur - 1 + mis, mb_row + 1, mb_col - 1,
base_mb16 + 1) +
get_nzc_4x4_uv(cm, cur - 1 + mis, mb_row + 1, mb_col - 1,
base_mb16 + 3);
nzc_exp <<= 1;
// Note nzc_exp is 64 times the average value expected at 16x16 scale
return choose_nzc_context(nzc_exp, NZC_T2_16X16, NZC_T1_16X16);
break;
case TX_8X8:
assert((block & 3) == 0);
if (boff < 8) {
int o = boff >> 2;
nzc_exp =
get_nzc_4x4_uv(cm, cur - mis + o, mb_row - 1, mb_col + o,
base_mb16 + 2) +
get_nzc_4x4_uv(cm, cur - mis + o, mb_row - 1, mb_col + o,
base_mb16 + 3);
} else {
nzc_exp = cur->mbmi.nzcs[block - 8] << 5;
}
if ((boff & 7) == 0) {
int o = boff >> 3;
nzc_exp +=
get_nzc_4x4_uv(cm, cur - 1 + o * mis, mb_row + o, mb_col - 1,
base_mb16 + 1) +
get_nzc_4x4_uv(cm, cur - 1 + o * mis, mb_row + o, mb_col - 1,
base_mb16 + 3);
} else {
nzc_exp += cur->mbmi.nzcs[block - 4] << 5;
}
// Note nzc_exp is 64 times the average value expected at 8x8 scale
return choose_nzc_context(nzc_exp, NZC_T2_8X8, NZC_T1_8X8);
case TX_4X4:
if (boff < 4) {
int o = boff >> 1;
int p = boff & 1;
nzc_exp = get_nzc_4x4_uv(cm, cur - mis + o, mb_row - 1, mb_col + o,
base_mb16 + 2 + p);
} else {
nzc_exp = (cur->mbmi.nzcs[block - 4] << 6);
}
if ((boff & 3) == 0) {
int o = boff >> 3;
int p = (boff >> 2) & 1;
nzc_exp += get_nzc_4x4_uv(cm, cur - 1 + o * mis, mb_row + o, mb_col - 1,
base_mb16 + 1 + 2 * p);
} else {
nzc_exp += (cur->mbmi.nzcs[block - 1] << 6);
}
nzc_exp >>= 1;
// Note nzc_exp is 64 times the average value expected at 4x4 scale
return choose_nzc_context(nzc_exp, NZC_T2_4X4, NZC_T1_4X4);
default:
return 0;
}
}
int vp9_get_nzc_context_uv_mb16(VP9_COMMON *cm, MODE_INFO *cur,
int mb_row, int mb_col, int block) {
// returns an index in [0, MAX_NZC_CONTEXTS - 1] to reflect how busy
// neighboring blocks are
int mis = cm->mode_info_stride;
int nzc_exp = 0;
const int base = block - (block & 3);
const int boff = (block & 3);
const int base_mb16 = base;
TX_SIZE txfm_size = cur->mbmi.txfm_size;
TX_SIZE txfm_size_uv;
assert(block >= 16 && block < 24);
if (txfm_size == TX_16X16)
txfm_size_uv = TX_8X8;
else if (txfm_size == TX_8X8 &&
(cur->mbmi.mode == I8X8_PRED || cur->mbmi.mode == SPLITMV))
txfm_size_uv = TX_4X4;
else
txfm_size_uv = txfm_size;
switch (txfm_size_uv) {
case TX_8X8:
assert((block & 3) == 0);
nzc_exp =
get_nzc_4x4_uv(cm, cur - mis, mb_row - 1, mb_col, base_mb16 + 2) +
get_nzc_4x4_uv(cm, cur - mis, mb_row - 1, mb_col, base_mb16 + 3) +
get_nzc_4x4_uv(cm, cur - 1, mb_row, mb_col - 1, base_mb16 + 1) +
get_nzc_4x4_uv(cm, cur - 1, mb_row, mb_col - 1, base_mb16 + 3);
// Note nzc_exp is 64 times the average value expected at 8x8 scale
return choose_nzc_context(nzc_exp, NZC_T2_8X8, NZC_T1_8X8);
case TX_4X4:
if (boff < 2) {
int p = boff & 1;
nzc_exp = get_nzc_4x4_uv(cm, cur - mis, mb_row - 1, mb_col,
base_mb16 + 2 + p);
} else {
nzc_exp = (cur->mbmi.nzcs[block - 2] << 6);
}
if ((boff & 1) == 0) {
int p = (boff >> 1) & 1;
nzc_exp += get_nzc_4x4_uv(cm, cur - 1, mb_row, mb_col - 1,
base_mb16 + 1 + 2 * p);
} else {
nzc_exp += (cur->mbmi.nzcs[block - 1] << 6);
}
nzc_exp >>= 1;
// Note nzc_exp is 64 times the average value expected at 4x4 scale
return choose_nzc_context(nzc_exp, NZC_T2_4X4, NZC_T1_4X4);
default:
return 0;
}
}
int vp9_get_nzc_context(VP9_COMMON *cm, MACROBLOCKD *xd, int block) {
if (xd->mode_info_context->mbmi.sb_type == BLOCK_SIZE_SB64X64) {
assert(block < 384);
if (block < 256)
return vp9_get_nzc_context_y_sb64(cm, xd->mode_info_context,
get_mb_row(xd), get_mb_col(xd), block);
else
return vp9_get_nzc_context_uv_sb64(cm, xd->mode_info_context,
get_mb_row(xd), get_mb_col(xd), block);
} else if (xd->mode_info_context->mbmi.sb_type == BLOCK_SIZE_SB32X32) {
assert(block < 96);
if (block < 64)
return vp9_get_nzc_context_y_sb32(cm, xd->mode_info_context,
get_mb_row(xd), get_mb_col(xd), block);
else
return vp9_get_nzc_context_uv_sb32(cm, xd->mode_info_context,
get_mb_row(xd), get_mb_col(xd), block);
} else {
assert(block < 64);
if (block < 16)
return vp9_get_nzc_context_y_mb16(cm, xd->mode_info_context,
get_mb_row(xd), get_mb_col(xd), block);
else
return vp9_get_nzc_context_uv_mb16(cm, xd->mode_info_context,
get_mb_row(xd), get_mb_col(xd), block);
}
}
static void update_nzc(VP9_COMMON *cm,
uint16_t nzc,
int nzc_context,
TX_SIZE tx_size,
int ref,
int type) {
int e, c;
c = codenzc(nzc);
if (tx_size == TX_32X32)
cm->fc.nzc_counts_32x32[nzc_context][ref][type][c]++;
else if (tx_size == TX_16X16)
cm->fc.nzc_counts_16x16[nzc_context][ref][type][c]++;
else if (tx_size == TX_8X8)
cm->fc.nzc_counts_8x8[nzc_context][ref][type][c]++;
else if (tx_size == TX_4X4)
cm->fc.nzc_counts_4x4[nzc_context][ref][type][c]++;
else
assert(0);
if ((e = vp9_extranzcbits[c])) {
int x = nzc - vp9_basenzcvalue[c];
while (e--) {
int b = (x >> e) & 1;
cm->fc.nzc_pcat_counts[nzc_context][c - NZC_TOKENS_NOEXTRA][e][b]++;
}
}
}
static void update_nzcs_sb64(VP9_COMMON *cm,
MACROBLOCKD *xd,
int mb_row,
int mb_col) {
MODE_INFO *m = xd->mode_info_context;
MB_MODE_INFO *const mi = &m->mbmi;
int j, nzc_context;
const int ref = m->mbmi.ref_frame != INTRA_FRAME;
assert(mb_col == get_mb_col(xd));
assert(mb_row == get_mb_row(xd));
if (mi->mb_skip_coeff)
return;
switch (mi->txfm_size) {
case TX_32X32:
for (j = 0; j < 256; j += 64) {
nzc_context = vp9_get_nzc_context_y_sb64(cm, m, mb_row, mb_col, j);
update_nzc(cm, m->mbmi.nzcs[j], nzc_context, TX_32X32, ref, 0);
}
for (j = 256; j < 384; j += 64) {
nzc_context = vp9_get_nzc_context_uv_sb64(cm, m, mb_row, mb_col, j);
update_nzc(cm, m->mbmi.nzcs[j], nzc_context, TX_32X32, ref, 1);
}
break;
case TX_16X16:
for (j = 0; j < 256; j += 16) {
nzc_context = vp9_get_nzc_context_y_sb64(cm, m, mb_row, mb_col, j);
update_nzc(cm, m->mbmi.nzcs[j], nzc_context, TX_16X16, ref, 0);
}
for (j = 256; j < 384; j += 16) {
nzc_context = vp9_get_nzc_context_uv_sb64(cm, m, mb_row, mb_col, j);
update_nzc(cm, m->mbmi.nzcs[j], nzc_context, TX_16X16, ref, 1);
}
break;
case TX_8X8:
for (j = 0; j < 256; j += 4) {
nzc_context = vp9_get_nzc_context_y_sb64(cm, m, mb_row, mb_col, j);
update_nzc(cm, m->mbmi.nzcs[j], nzc_context, TX_8X8, ref, 0);
}
for (j = 256; j < 384; j += 4) {
nzc_context = vp9_get_nzc_context_uv_sb64(cm, m, mb_row, mb_col, j);
update_nzc(cm, m->mbmi.nzcs[j], nzc_context, TX_8X8, ref, 1);
}
break;
case TX_4X4:
for (j = 0; j < 256; ++j) {
nzc_context = vp9_get_nzc_context_y_sb64(cm, m, mb_row, mb_col, j);
update_nzc(cm, m->mbmi.nzcs[j], nzc_context, TX_4X4, ref, 0);
}
for (j = 256; j < 384; ++j) {
nzc_context = vp9_get_nzc_context_uv_sb64(cm, m, mb_row, mb_col, j);
update_nzc(cm, m->mbmi.nzcs[j], nzc_context, TX_4X4, ref, 1);
}
break;
default:
break;
}
}
static void update_nzcs_sb32(VP9_COMMON *cm,
MACROBLOCKD *xd,
int mb_row,
int mb_col) {
MODE_INFO *m = xd->mode_info_context;
MB_MODE_INFO *const mi = &m->mbmi;
int j, nzc_context;
const int ref = m->mbmi.ref_frame != INTRA_FRAME;
assert(mb_col == get_mb_col(xd));
assert(mb_row == get_mb_row(xd));
if (mi->mb_skip_coeff)
return;
switch (mi->txfm_size) {
case TX_32X32:
for (j = 0; j < 64; j += 64) {
nzc_context = vp9_get_nzc_context_y_sb32(cm, m, mb_row, mb_col, j);
update_nzc(cm, m->mbmi.nzcs[j], nzc_context, TX_32X32, ref, 0);
}
for (j = 64; j < 96; j += 16) {
nzc_context = vp9_get_nzc_context_uv_sb32(cm, m, mb_row, mb_col, j);
update_nzc(cm, m->mbmi.nzcs[j], nzc_context, TX_16X16, ref, 1);
}
break;
case TX_16X16:
for (j = 0; j < 64; j += 16) {
nzc_context = vp9_get_nzc_context_y_sb32(cm, m, mb_row, mb_col, j);
update_nzc(cm, m->mbmi.nzcs[j], nzc_context, TX_16X16, ref, 0);
}
for (j = 64; j < 96; j += 16) {
nzc_context = vp9_get_nzc_context_uv_sb32(cm, m, mb_row, mb_col, j);
update_nzc(cm, m->mbmi.nzcs[j], nzc_context, TX_16X16, ref, 1);
}
break;
case TX_8X8:
for (j = 0; j < 64; j += 4) {
nzc_context = vp9_get_nzc_context_y_sb32(cm, m, mb_row, mb_col, j);
update_nzc(cm, m->mbmi.nzcs[j], nzc_context, TX_8X8, ref, 0);
}
for (j = 64; j < 96; j += 4) {
nzc_context = vp9_get_nzc_context_uv_sb32(cm, m, mb_row, mb_col, j);
update_nzc(cm, m->mbmi.nzcs[j], nzc_context, TX_8X8, ref, 1);
}
break;
case TX_4X4:
for (j = 0; j < 64; ++j) {
nzc_context = vp9_get_nzc_context_y_sb32(cm, m, mb_row, mb_col, j);
update_nzc(cm, m->mbmi.nzcs[j], nzc_context, TX_4X4, ref, 0);
}
for (j = 64; j < 96; ++j) {
nzc_context = vp9_get_nzc_context_uv_sb32(cm, m, mb_row, mb_col, j);
update_nzc(cm, m->mbmi.nzcs[j], nzc_context, TX_4X4, ref, 1);
}
break;
default:
break;
}
2010-05-18 17:58:33 +02:00
}
static void update_nzcs_mb16(VP9_COMMON *cm,
MACROBLOCKD *xd,
int mb_row,
int mb_col) {
MODE_INFO *m = xd->mode_info_context;
MB_MODE_INFO *const mi = &m->mbmi;
int j, nzc_context;
const int ref = m->mbmi.ref_frame != INTRA_FRAME;
assert(mb_col == get_mb_col(xd));
assert(mb_row == get_mb_row(xd));
if (mi->mb_skip_coeff)
return;
switch (mi->txfm_size) {
case TX_16X16:
for (j = 0; j < 16; j += 16) {
nzc_context = vp9_get_nzc_context_y_mb16(cm, m, mb_row, mb_col, j);
update_nzc(cm, m->mbmi.nzcs[j], nzc_context, TX_16X16, ref, 0);
}
for (j = 16; j < 24; j += 4) {
nzc_context = vp9_get_nzc_context_uv_mb16(cm, m, mb_row, mb_col, j);
update_nzc(cm, m->mbmi.nzcs[j], nzc_context, TX_8X8, ref, 1);
}
break;
case TX_8X8:
for (j = 0; j < 16; j += 4) {
nzc_context = vp9_get_nzc_context_y_mb16(cm, m, mb_row, mb_col, j);
update_nzc(cm, m->mbmi.nzcs[j], nzc_context, TX_8X8, ref, 0);
}
if (mi->mode == I8X8_PRED || mi->mode == SPLITMV) {
for (j = 16; j < 24; ++j) {
nzc_context = vp9_get_nzc_context_uv_mb16(cm, m, mb_row, mb_col, j);
update_nzc(cm, m->mbmi.nzcs[j], nzc_context, TX_4X4, ref, 1);
}
} else {
for (j = 16; j < 24; j += 4) {
nzc_context = vp9_get_nzc_context_uv_mb16(cm, m, mb_row, mb_col, j);
update_nzc(cm, m->mbmi.nzcs[j], nzc_context, TX_8X8, ref, 1);
}
}
break;
case TX_4X4:
for (j = 0; j < 16; ++j) {
nzc_context = vp9_get_nzc_context_y_mb16(cm, m, mb_row, mb_col, j);
update_nzc(cm, m->mbmi.nzcs[j], nzc_context, TX_4X4, ref, 0);
}
for (j = 16; j < 24; ++j) {
nzc_context = vp9_get_nzc_context_uv_mb16(cm, m, mb_row, mb_col, j);
update_nzc(cm, m->mbmi.nzcs[j], nzc_context, TX_4X4, ref, 1);
}
break;
default:
break;
}
}
void vp9_update_nzc_counts(VP9_COMMON *cm,
MACROBLOCKD *xd,
int mb_row,
int mb_col) {
if (xd->mode_info_context->mbmi.sb_type == BLOCK_SIZE_SB64X64)
update_nzcs_sb64(cm, xd, mb_row, mb_col);
else if (xd->mode_info_context->mbmi.sb_type == BLOCK_SIZE_SB32X32)
update_nzcs_sb32(cm, xd, mb_row, mb_col);
else
update_nzcs_mb16(cm, xd, mb_row, mb_col);
}
#endif // CONFIG_CODE_NONZEROCOUNT
// #define COEF_COUNT_TESTING
#define COEF_COUNT_SAT 24
#define COEF_MAX_UPDATE_FACTOR 112
#define COEF_COUNT_SAT_KEY 24
#define COEF_MAX_UPDATE_FACTOR_KEY 112
#define COEF_COUNT_SAT_AFTER_KEY 24
#define COEF_MAX_UPDATE_FACTOR_AFTER_KEY 128
static void adapt_coef_probs(vp9_coeff_probs *dst_coef_probs,
vp9_coeff_probs *pre_coef_probs,
int block_types, vp9_coeff_count *coef_counts,
int count_sat, int update_factor) {
int t, i, j, k, l, count;
unsigned int branch_ct[ENTROPY_NODES][2];
vp9_prob coef_probs[ENTROPY_NODES];
int factor;
for (i = 0; i < block_types; ++i)
for (j = 0; j < REF_TYPES; ++j)
for (k = 0; k < COEF_BANDS; ++k)
for (l = 0; l < PREV_COEF_CONTEXTS; ++l) {
if (l >= 3 && k == 0)
continue;
vp9_tree_probs_from_distribution(vp9_coef_tree,
coef_probs, branch_ct,
coef_counts[i][j][k][l], 0);
for (t = 0; t < ENTROPY_NODES; ++t) {
count = branch_ct[t][0] + branch_ct[t][1];
count = count > count_sat ? count_sat : count;
factor = (update_factor * count / count_sat);
dst_coef_probs[i][j][k][l][t] =
weighted_prob(pre_coef_probs[i][j][k][l][t],
coef_probs[t], factor);
}
}
}
void vp9_adapt_coef_probs(VP9_COMMON *cm) {
int count_sat;
int update_factor; /* denominator 256 */
// printf("Frame type: %d\n", cm->frame_type);
if (cm->frame_type == KEY_FRAME) {
update_factor = COEF_MAX_UPDATE_FACTOR_KEY;
count_sat = COEF_COUNT_SAT_KEY;
} else if (cm->last_frame_type == KEY_FRAME) {
update_factor = COEF_MAX_UPDATE_FACTOR_AFTER_KEY; /* adapt quickly */
count_sat = COEF_COUNT_SAT_AFTER_KEY;
} else {
update_factor = COEF_MAX_UPDATE_FACTOR;
count_sat = COEF_COUNT_SAT;
}
adapt_coef_probs(cm->fc.coef_probs_4x4, cm->fc.pre_coef_probs_4x4,
BLOCK_TYPES, cm->fc.coef_counts_4x4,
count_sat, update_factor);
adapt_coef_probs(cm->fc.coef_probs_8x8, cm->fc.pre_coef_probs_8x8,
BLOCK_TYPES, cm->fc.coef_counts_8x8,
count_sat, update_factor);
adapt_coef_probs(cm->fc.coef_probs_16x16, cm->fc.pre_coef_probs_16x16,
BLOCK_TYPES, cm->fc.coef_counts_16x16,
count_sat, update_factor);
adapt_coef_probs(cm->fc.coef_probs_32x32, cm->fc.pre_coef_probs_32x32,
BLOCK_TYPES, cm->fc.coef_counts_32x32,
count_sat, update_factor);
}
#if CONFIG_CODE_NONZEROCOUNT
static void adapt_nzc_probs(VP9_COMMON *cm,
int block_size,
int count_sat,
int update_factor) {
int c, r, b, n;
int count, factor;
unsigned int nzc_branch_ct[NZC32X32_NODES][2];
vp9_prob nzc_probs[NZC32X32_NODES];
int tokens, nodes;
const vp9_tree_index *nzc_tree;
vp9_prob *dst_nzc_probs;
vp9_prob *pre_nzc_probs;
unsigned int *nzc_counts;
if (block_size == 32) {
tokens = NZC32X32_TOKENS;
nzc_tree = vp9_nzc32x32_tree;
dst_nzc_probs = cm->fc.nzc_probs_32x32[0][0][0];
pre_nzc_probs = cm->fc.pre_nzc_probs_32x32[0][0][0];
nzc_counts = cm->fc.nzc_counts_32x32[0][0][0];
} else if (block_size == 16) {
tokens = NZC16X16_TOKENS;
nzc_tree = vp9_nzc16x16_tree;
dst_nzc_probs = cm->fc.nzc_probs_16x16[0][0][0];
pre_nzc_probs = cm->fc.pre_nzc_probs_16x16[0][0][0];
nzc_counts = cm->fc.nzc_counts_16x16[0][0][0];
} else if (block_size == 8) {
tokens = NZC8X8_TOKENS;
nzc_tree = vp9_nzc8x8_tree;
dst_nzc_probs = cm->fc.nzc_probs_8x8[0][0][0];
pre_nzc_probs = cm->fc.pre_nzc_probs_8x8[0][0][0];
nzc_counts = cm->fc.nzc_counts_8x8[0][0][0];
} else {
nzc_tree = vp9_nzc4x4_tree;
tokens = NZC4X4_TOKENS;
dst_nzc_probs = cm->fc.nzc_probs_4x4[0][0][0];
pre_nzc_probs = cm->fc.pre_nzc_probs_4x4[0][0][0];
nzc_counts = cm->fc.nzc_counts_4x4[0][0][0];
}
nodes = tokens - 1;
for (c = 0; c < MAX_NZC_CONTEXTS; ++c)
for (r = 0; r < REF_TYPES; ++r)
for (b = 0; b < BLOCK_TYPES; ++b) {
int offset = c * REF_TYPES * BLOCK_TYPES + r * BLOCK_TYPES + b;
int offset_nodes = offset * nodes;
int offset_tokens = offset * tokens;
vp9_tree_probs_from_distribution(nzc_tree,
nzc_probs, nzc_branch_ct,
nzc_counts + offset_tokens, 0);
for (n = 0; n < nodes; ++n) {
count = nzc_branch_ct[n][0] + nzc_branch_ct[n][1];
count = count > count_sat ? count_sat : count;
factor = (update_factor * count / count_sat);
dst_nzc_probs[offset_nodes + n] =
weighted_prob(pre_nzc_probs[offset_nodes + n],
nzc_probs[n], factor);
}
}
}
static void adapt_nzc_pcat(VP9_COMMON *cm, int count_sat, int update_factor) {
int c, t;
int count, factor;
for (c = 0; c < MAX_NZC_CONTEXTS; ++c) {
for (t = 0; t < NZC_TOKENS_EXTRA; ++t) {
int bits = vp9_extranzcbits[t + NZC_TOKENS_NOEXTRA];
int b;
for (b = 0; b < bits; ++b) {
vp9_prob prob = get_binary_prob(cm->fc.nzc_pcat_counts[c][t][b][0],
cm->fc.nzc_pcat_counts[c][t][b][1]);
count = cm->fc.nzc_pcat_counts[c][t][b][0] +
cm->fc.nzc_pcat_counts[c][t][b][1];
count = count > count_sat ? count_sat : count;
factor = (update_factor * count / count_sat);
cm->fc.nzc_pcat_probs[c][t][b] = weighted_prob(
cm->fc.pre_nzc_pcat_probs[c][t][b], prob, factor);
}
}
}
}
// #define NZC_COUNT_TESTING
void vp9_adapt_nzc_probs(VP9_COMMON *cm) {
int count_sat;
int update_factor; /* denominator 256 */
#ifdef NZC_COUNT_TESTING
int c, r, b, t;
printf("\n");
for (c = 0; c < MAX_NZC_CONTEXTS; ++c)
for (r = 0; r < REF_TYPES; ++r) {
for (b = 0; b < BLOCK_TYPES; ++b) {
printf(" {");
for (t = 0; t < NZC4X4_TOKENS; ++t) {
printf(" %d,", cm->fc.nzc_counts_4x4[c][r][b][t]);
}
printf("}\n");
}
printf("\n");
}
#endif
if (cm->frame_type == KEY_FRAME) {
update_factor = COEF_MAX_UPDATE_FACTOR_KEY;
count_sat = COEF_COUNT_SAT_KEY;
} else if (cm->last_frame_type == KEY_FRAME) {
update_factor = COEF_MAX_UPDATE_FACTOR_AFTER_KEY; /* adapt quickly */
count_sat = COEF_COUNT_SAT_AFTER_KEY;
} else {
update_factor = COEF_MAX_UPDATE_FACTOR;
count_sat = COEF_COUNT_SAT;
}
adapt_nzc_probs(cm, 4, count_sat, update_factor);
adapt_nzc_probs(cm, 8, count_sat, update_factor);
adapt_nzc_probs(cm, 16, count_sat, update_factor);
adapt_nzc_probs(cm, 32, count_sat, update_factor);
adapt_nzc_pcat(cm, count_sat, update_factor);
}
#endif // CONFIG_CODE_NONZEROCOUNT