/src/aom/av1/decoder/grain_synthesis.c
Line | Count | Source (jump to first uncovered line) |
1 | | /* |
2 | | * Copyright (c) 2016, Alliance for Open Media. All rights reserved |
3 | | * |
4 | | * This source code is subject to the terms of the BSD 2 Clause License and |
5 | | * the Alliance for Open Media Patent License 1.0. If the BSD 2 Clause License |
6 | | * was not distributed with this source code in the LICENSE file, you can |
7 | | * obtain it at www.aomedia.org/license/software. If the Alliance for Open |
8 | | * Media Patent License 1.0 was not distributed with this source code in the |
9 | | * PATENTS file, you can obtain it at www.aomedia.org/license/patent. |
10 | | */ |
11 | | |
12 | | /*!\file |
13 | | * \brief Describes film grain parameters and film grain synthesis |
14 | | * |
15 | | */ |
16 | | |
17 | | #include <stdbool.h> |
18 | | #include <stdio.h> |
19 | | #include <string.h> |
20 | | #include <stdlib.h> |
21 | | #include <assert.h> |
22 | | #include "aom_dsp/aom_dsp_common.h" |
23 | | #include "aom_mem/aom_mem.h" |
24 | | #include "av1/decoder/grain_synthesis.h" |
25 | | |
26 | | // Samples with Gaussian distribution in the range of [-2048, 2047] (12 bits) |
27 | | // with zero mean and standard deviation of about 512. |
28 | | // should be divided by 4 for 10-bit range and 16 for 8-bit range. |
29 | | static const int gaussian_sequence[2048] = { |
30 | | 56, 568, -180, 172, 124, -84, 172, -64, -900, 24, 820, |
31 | | 224, 1248, 996, 272, -8, -916, -388, -732, -104, -188, 800, |
32 | | 112, -652, -320, -376, 140, -252, 492, -168, 44, -788, 588, |
33 | | -584, 500, -228, 12, 680, 272, -476, 972, -100, 652, 368, |
34 | | 432, -196, -720, -192, 1000, -332, 652, -136, -552, -604, -4, |
35 | | 192, -220, -136, 1000, -52, 372, -96, -624, 124, -24, 396, |
36 | | 540, -12, -104, 640, 464, 244, -208, -84, 368, -528, -740, |
37 | | 248, -968, -848, 608, 376, -60, -292, -40, -156, 252, -292, |
38 | | 248, 224, -280, 400, -244, 244, -60, 76, -80, 212, 532, |
39 | | 340, 128, -36, 824, -352, -60, -264, -96, -612, 416, -704, |
40 | | 220, -204, 640, -160, 1220, -408, 900, 336, 20, -336, -96, |
41 | | -792, 304, 48, -28, -1232, -1172, -448, 104, -292, -520, 244, |
42 | | 60, -948, 0, -708, 268, 108, 356, -548, 488, -344, -136, |
43 | | 488, -196, -224, 656, -236, -1128, 60, 4, 140, 276, -676, |
44 | | -376, 168, -108, 464, 8, 564, 64, 240, 308, -300, -400, |
45 | | -456, -136, 56, 120, -408, -116, 436, 504, -232, 328, 844, |
46 | | -164, -84, 784, -168, 232, -224, 348, -376, 128, 568, 96, |
47 | | -1244, -288, 276, 848, 832, -360, 656, 464, -384, -332, -356, |
48 | | 728, -388, 160, -192, 468, 296, 224, 140, -776, -100, 280, |
49 | | 4, 196, 44, -36, -648, 932, 16, 1428, 28, 528, 808, |
50 | | 772, 20, 268, 88, -332, -284, 124, -384, -448, 208, -228, |
51 | | -1044, -328, 660, 380, -148, -300, 588, 240, 540, 28, 136, |
52 | | -88, -436, 256, 296, -1000, 1400, 0, -48, 1056, -136, 264, |
53 | | -528, -1108, 632, -484, -592, -344, 796, 124, -668, -768, 388, |
54 | | 1296, -232, -188, -200, -288, -4, 308, 100, -168, 256, -500, |
55 | | 204, -508, 648, -136, 372, -272, -120, -1004, -552, -548, -384, |
56 | | 548, -296, 428, -108, -8, -912, -324, -224, -88, -112, -220, |
57 | | -100, 996, -796, 548, 360, -216, 180, 428, -200, -212, 148, |
58 | | 96, 148, 284, 216, -412, -320, 120, -300, -384, -604, -572, |
59 | | -332, -8, -180, -176, 696, 116, -88, 628, 76, 44, -516, |
60 | | 240, -208, -40, 100, -592, 344, -308, -452, -228, 20, 916, |
61 | | -1752, -136, -340, -804, 140, 40, 512, 340, 248, 184, -492, |
62 | | 896, -156, 932, -628, 328, -688, -448, -616, -752, -100, 560, |
63 | | -1020, 180, -800, -64, 76, 576, 1068, 396, 660, 552, -108, |
64 | | -28, 320, -628, 312, -92, -92, -472, 268, 16, 560, 516, |
65 | | -672, -52, 492, -100, 260, 384, 284, 292, 304, -148, 88, |
66 | | -152, 1012, 1064, -228, 164, -376, -684, 592, -392, 156, 196, |
67 | | -524, -64, -884, 160, -176, 636, 648, 404, -396, -436, 864, |
68 | | 424, -728, 988, -604, 904, -592, 296, -224, 536, -176, -920, |
69 | | 436, -48, 1176, -884, 416, -776, -824, -884, 524, -548, -564, |
70 | | -68, -164, -96, 692, 364, -692, -1012, -68, 260, -480, 876, |
71 | | -1116, 452, -332, -352, 892, -1088, 1220, -676, 12, -292, 244, |
72 | | 496, 372, -32, 280, 200, 112, -440, -96, 24, -644, -184, |
73 | | 56, -432, 224, -980, 272, -260, 144, -436, 420, 356, 364, |
74 | | -528, 76, 172, -744, -368, 404, -752, -416, 684, -688, 72, |
75 | | 540, 416, 92, 444, 480, -72, -1416, 164, -1172, -68, 24, |
76 | | 424, 264, 1040, 128, -912, -524, -356, 64, 876, -12, 4, |
77 | | -88, 532, 272, -524, 320, 276, -508, 940, 24, -400, -120, |
78 | | 756, 60, 236, -412, 100, 376, -484, 400, -100, -740, -108, |
79 | | -260, 328, -268, 224, -200, -416, 184, -604, -564, -20, 296, |
80 | | 60, 892, -888, 60, 164, 68, -760, 216, -296, 904, -336, |
81 | | -28, 404, -356, -568, -208, -1480, -512, 296, 328, -360, -164, |
82 | | -1560, -776, 1156, -428, 164, -504, -112, 120, -216, -148, -264, |
83 | | 308, 32, 64, -72, 72, 116, 176, -64, -272, 460, -536, |
84 | | -784, -280, 348, 108, -752, -132, 524, -540, -776, 116, -296, |
85 | | -1196, -288, -560, 1040, -472, 116, -848, -1116, 116, 636, 696, |
86 | | 284, -176, 1016, 204, -864, -648, -248, 356, 972, -584, -204, |
87 | | 264, 880, 528, -24, -184, 116, 448, -144, 828, 524, 212, |
88 | | -212, 52, 12, 200, 268, -488, -404, -880, 824, -672, -40, |
89 | | 908, -248, 500, 716, -576, 492, -576, 16, 720, -108, 384, |
90 | | 124, 344, 280, 576, -500, 252, 104, -308, 196, -188, -8, |
91 | | 1268, 296, 1032, -1196, 436, 316, 372, -432, -200, -660, 704, |
92 | | -224, 596, -132, 268, 32, -452, 884, 104, -1008, 424, -1348, |
93 | | -280, 4, -1168, 368, 476, 696, 300, -8, 24, 180, -592, |
94 | | -196, 388, 304, 500, 724, -160, 244, -84, 272, -256, -420, |
95 | | 320, 208, -144, -156, 156, 364, 452, 28, 540, 316, 220, |
96 | | -644, -248, 464, 72, 360, 32, -388, 496, -680, -48, 208, |
97 | | -116, -408, 60, -604, -392, 548, -840, 784, -460, 656, -544, |
98 | | -388, -264, 908, -800, -628, -612, -568, 572, -220, 164, 288, |
99 | | -16, -308, 308, -112, -636, -760, 280, -668, 432, 364, 240, |
100 | | -196, 604, 340, 384, 196, 592, -44, -500, 432, -580, -132, |
101 | | 636, -76, 392, 4, -412, 540, 508, 328, -356, -36, 16, |
102 | | -220, -64, -248, -60, 24, -192, 368, 1040, 92, -24, -1044, |
103 | | -32, 40, 104, 148, 192, -136, -520, 56, -816, -224, 732, |
104 | | 392, 356, 212, -80, -424, -1008, -324, 588, -1496, 576, 460, |
105 | | -816, -848, 56, -580, -92, -1372, -112, -496, 200, 364, 52, |
106 | | -140, 48, -48, -60, 84, 72, 40, 132, -356, -268, -104, |
107 | | -284, -404, 732, -520, 164, -304, -540, 120, 328, -76, -460, |
108 | | 756, 388, 588, 236, -436, -72, -176, -404, -316, -148, 716, |
109 | | -604, 404, -72, -88, -888, -68, 944, 88, -220, -344, 960, |
110 | | 472, 460, -232, 704, 120, 832, -228, 692, -508, 132, -476, |
111 | | 844, -748, -364, -44, 1116, -1104, -1056, 76, 428, 552, -692, |
112 | | 60, 356, 96, -384, -188, -612, -576, 736, 508, 892, 352, |
113 | | -1132, 504, -24, -352, 324, 332, -600, -312, 292, 508, -144, |
114 | | -8, 484, 48, 284, -260, -240, 256, -100, -292, -204, -44, |
115 | | 472, -204, 908, -188, -1000, -256, 92, 1164, -392, 564, 356, |
116 | | 652, -28, -884, 256, 484, -192, 760, -176, 376, -524, -452, |
117 | | -436, 860, -736, 212, 124, 504, -476, 468, 76, -472, 552, |
118 | | -692, -944, -620, 740, -240, 400, 132, 20, 192, -196, 264, |
119 | | -668, -1012, -60, 296, -316, -828, 76, -156, 284, -768, -448, |
120 | | -832, 148, 248, 652, 616, 1236, 288, -328, -400, -124, 588, |
121 | | 220, 520, -696, 1032, 768, -740, -92, -272, 296, 448, -464, |
122 | | 412, -200, 392, 440, -200, 264, -152, -260, 320, 1032, 216, |
123 | | 320, -8, -64, 156, -1016, 1084, 1172, 536, 484, -432, 132, |
124 | | 372, -52, -256, 84, 116, -352, 48, 116, 304, -384, 412, |
125 | | 924, -300, 528, 628, 180, 648, 44, -980, -220, 1320, 48, |
126 | | 332, 748, 524, -268, -720, 540, -276, 564, -344, -208, -196, |
127 | | 436, 896, 88, -392, 132, 80, -964, -288, 568, 56, -48, |
128 | | -456, 888, 8, 552, -156, -292, 948, 288, 128, -716, -292, |
129 | | 1192, -152, 876, 352, -600, -260, -812, -468, -28, -120, -32, |
130 | | -44, 1284, 496, 192, 464, 312, -76, -516, -380, -456, -1012, |
131 | | -48, 308, -156, 36, 492, -156, -808, 188, 1652, 68, -120, |
132 | | -116, 316, 160, -140, 352, 808, -416, 592, 316, -480, 56, |
133 | | 528, -204, -568, 372, -232, 752, -344, 744, -4, 324, -416, |
134 | | -600, 768, 268, -248, -88, -132, -420, -432, 80, -288, 404, |
135 | | -316, -1216, -588, 520, -108, 92, -320, 368, -480, -216, -92, |
136 | | 1688, -300, 180, 1020, -176, 820, -68, -228, -260, 436, -904, |
137 | | 20, 40, -508, 440, -736, 312, 332, 204, 760, -372, 728, |
138 | | 96, -20, -632, -520, -560, 336, 1076, -64, -532, 776, 584, |
139 | | 192, 396, -728, -520, 276, -188, 80, -52, -612, -252, -48, |
140 | | 648, 212, -688, 228, -52, -260, 428, -412, -272, -404, 180, |
141 | | 816, -796, 48, 152, 484, -88, -216, 988, 696, 188, -528, |
142 | | 648, -116, -180, 316, 476, 12, -564, 96, 476, -252, -364, |
143 | | -376, -392, 556, -256, -576, 260, -352, 120, -16, -136, -260, |
144 | | -492, 72, 556, 660, 580, 616, 772, 436, 424, -32, -324, |
145 | | -1268, 416, -324, -80, 920, 160, 228, 724, 32, -516, 64, |
146 | | 384, 68, -128, 136, 240, 248, -204, -68, 252, -932, -120, |
147 | | -480, -628, -84, 192, 852, -404, -288, -132, 204, 100, 168, |
148 | | -68, -196, -868, 460, 1080, 380, -80, 244, 0, 484, -888, |
149 | | 64, 184, 352, 600, 460, 164, 604, -196, 320, -64, 588, |
150 | | -184, 228, 12, 372, 48, -848, -344, 224, 208, -200, 484, |
151 | | 128, -20, 272, -468, -840, 384, 256, -720, -520, -464, -580, |
152 | | 112, -120, 644, -356, -208, -608, -528, 704, 560, -424, 392, |
153 | | 828, 40, 84, 200, -152, 0, -144, 584, 280, -120, 80, |
154 | | -556, -972, -196, -472, 724, 80, 168, -32, 88, 160, -688, |
155 | | 0, 160, 356, 372, -776, 740, -128, 676, -248, -480, 4, |
156 | | -364, 96, 544, 232, -1032, 956, 236, 356, 20, -40, 300, |
157 | | 24, -676, -596, 132, 1120, -104, 532, -1096, 568, 648, 444, |
158 | | 508, 380, 188, -376, -604, 1488, 424, 24, 756, -220, -192, |
159 | | 716, 120, 920, 688, 168, 44, -460, 568, 284, 1144, 1160, |
160 | | 600, 424, 888, 656, -356, -320, 220, 316, -176, -724, -188, |
161 | | -816, -628, -348, -228, -380, 1012, -452, -660, 736, 928, 404, |
162 | | -696, -72, -268, -892, 128, 184, -344, -780, 360, 336, 400, |
163 | | 344, 428, 548, -112, 136, -228, -216, -820, -516, 340, 92, |
164 | | -136, 116, -300, 376, -244, 100, -316, -520, -284, -12, 824, |
165 | | 164, -548, -180, -128, 116, -924, -828, 268, -368, -580, 620, |
166 | | 192, 160, 0, -1676, 1068, 424, -56, -360, 468, -156, 720, |
167 | | 288, -528, 556, -364, 548, -148, 504, 316, 152, -648, -620, |
168 | | -684, -24, -376, -384, -108, -920, -1032, 768, 180, -264, -508, |
169 | | -1268, -260, -60, 300, -240, 988, 724, -376, -576, -212, -736, |
170 | | 556, 192, 1092, -620, -880, 376, -56, -4, -216, -32, 836, |
171 | | 268, 396, 1332, 864, -600, 100, 56, -412, -92, 356, 180, |
172 | | 884, -468, -436, 292, -388, -804, -704, -840, 368, -348, 140, |
173 | | -724, 1536, 940, 372, 112, -372, 436, -480, 1136, 296, -32, |
174 | | -228, 132, -48, -220, 868, -1016, -60, -1044, -464, 328, 916, |
175 | | 244, 12, -736, -296, 360, 468, -376, -108, -92, 788, 368, |
176 | | -56, 544, 400, -672, -420, 728, 16, 320, 44, -284, -380, |
177 | | -796, 488, 132, 204, -596, -372, 88, -152, -908, -636, -572, |
178 | | -624, -116, -692, -200, -56, 276, -88, 484, -324, 948, 864, |
179 | | 1000, -456, -184, -276, 292, -296, 156, 676, 320, 160, 908, |
180 | | -84, -1236, -288, -116, 260, -372, -644, 732, -756, -96, 84, |
181 | | 344, -520, 348, -688, 240, -84, 216, -1044, -136, -676, -396, |
182 | | -1500, 960, -40, 176, 168, 1516, 420, -504, -344, -364, -360, |
183 | | 1216, -940, -380, -212, 252, -660, -708, 484, -444, -152, 928, |
184 | | -120, 1112, 476, -260, 560, -148, -344, 108, -196, 228, -288, |
185 | | 504, 560, -328, -88, 288, -1008, 460, -228, 468, -836, -196, |
186 | | 76, 388, 232, 412, -1168, -716, -644, 756, -172, -356, -504, |
187 | | 116, 432, 528, 48, 476, -168, -608, 448, 160, -532, -272, |
188 | | 28, -676, -12, 828, 980, 456, 520, 104, -104, 256, -344, |
189 | | -4, -28, -368, -52, -524, -572, -556, -200, 768, 1124, -208, |
190 | | -512, 176, 232, 248, -148, -888, 604, -600, -304, 804, -156, |
191 | | -212, 488, -192, -804, -256, 368, -360, -916, -328, 228, -240, |
192 | | -448, -472, 856, -556, -364, 572, -12, -156, -368, -340, 432, |
193 | | 252, -752, -152, 288, 268, -580, -848, -592, 108, -76, 244, |
194 | | 312, -716, 592, -80, 436, 360, 4, -248, 160, 516, 584, |
195 | | 732, 44, -468, -280, -292, -156, -588, 28, 308, 912, 24, |
196 | | 124, 156, 180, -252, 944, -924, -772, -520, -428, -624, 300, |
197 | | -212, -1144, 32, -724, 800, -1128, -212, -1288, -848, 180, -416, |
198 | | 440, 192, -576, -792, -76, -1080, 80, -532, -352, -132, 380, |
199 | | -820, 148, 1112, 128, 164, 456, 700, -924, 144, -668, -384, |
200 | | 648, -832, 508, 552, -52, -100, -656, 208, -568, 748, -88, |
201 | | 680, 232, 300, 192, -408, -1012, -152, -252, -268, 272, -876, |
202 | | -664, -648, -332, -136, 16, 12, 1152, -28, 332, -536, 320, |
203 | | -672, -460, -316, 532, -260, 228, -40, 1052, -816, 180, 88, |
204 | | -496, -556, -672, -368, 428, 92, 356, 404, -408, 252, 196, |
205 | | -176, -556, 792, 268, 32, 372, 40, 96, -332, 328, 120, |
206 | | 372, -900, -40, 472, -264, -592, 952, 128, 656, 112, 664, |
207 | | -232, 420, 4, -344, -464, 556, 244, -416, -32, 252, 0, |
208 | | -412, 188, -696, 508, -476, 324, -1096, 656, -312, 560, 264, |
209 | | -136, 304, 160, -64, -580, 248, 336, -720, 560, -348, -288, |
210 | | -276, -196, -500, 852, -544, -236, -1128, -992, -776, 116, 56, |
211 | | 52, 860, 884, 212, -12, 168, 1020, 512, -552, 924, -148, |
212 | | 716, 188, 164, -340, -520, -184, 880, -152, -680, -208, -1156, |
213 | | -300, -528, -472, 364, 100, -744, -1056, -32, 540, 280, 144, |
214 | | -676, -32, -232, -280, -224, 96, 568, -76, 172, 148, 148, |
215 | | 104, 32, -296, -32, 788, -80, 32, -16, 280, 288, 944, |
216 | | 428, -484 |
217 | | }; |
218 | | |
219 | | static const int gauss_bits = 11; |
220 | | |
221 | | static int luma_subblock_size_y = 32; |
222 | | static int luma_subblock_size_x = 32; |
223 | | |
224 | | static int chroma_subblock_size_y = 16; |
225 | | static int chroma_subblock_size_x = 16; |
226 | | |
227 | | static const int min_luma_legal_range = 16; |
228 | | static const int max_luma_legal_range = 235; |
229 | | |
230 | | static const int min_chroma_legal_range = 16; |
231 | | static const int max_chroma_legal_range = 240; |
232 | | |
233 | | static int scaling_lut_y[256]; |
234 | | static int scaling_lut_cb[256]; |
235 | | static int scaling_lut_cr[256]; |
236 | | |
237 | | static int grain_min; |
238 | | static int grain_max; |
239 | | |
240 | | static uint16_t random_register = 0; // random number generator register |
241 | | |
242 | | static void dealloc_arrays(const aom_film_grain_t *params, int ***pred_pos_luma, |
243 | | int ***pred_pos_chroma, int **luma_grain_block, |
244 | | int **cb_grain_block, int **cr_grain_block, |
245 | | int **y_line_buf, int **cb_line_buf, |
246 | | int **cr_line_buf, int **y_col_buf, int **cb_col_buf, |
247 | 10.6k | int **cr_col_buf) { |
248 | 10.6k | int num_pos_luma = 2 * params->ar_coeff_lag * (params->ar_coeff_lag + 1); |
249 | 10.6k | int num_pos_chroma = num_pos_luma; |
250 | 10.6k | if (params->num_y_points > 0) ++num_pos_chroma; |
251 | | |
252 | 10.6k | if (*pred_pos_luma) { |
253 | 32.7k | for (int row = 0; row < num_pos_luma; row++) { |
254 | 22.1k | aom_free((*pred_pos_luma)[row]); |
255 | 22.1k | } |
256 | 10.6k | aom_free(*pred_pos_luma); |
257 | 10.6k | *pred_pos_luma = NULL; |
258 | 10.6k | } |
259 | | |
260 | 10.6k | if (*pred_pos_chroma) { |
261 | 42.2k | for (int row = 0; row < num_pos_chroma; row++) { |
262 | 31.6k | aom_free((*pred_pos_chroma)[row]); |
263 | 31.6k | } |
264 | 10.6k | aom_free(*pred_pos_chroma); |
265 | 10.6k | *pred_pos_chroma = NULL; |
266 | 10.6k | } |
267 | | |
268 | 10.6k | aom_free(*y_line_buf); |
269 | 10.6k | *y_line_buf = NULL; |
270 | | |
271 | 10.6k | aom_free(*cb_line_buf); |
272 | 10.6k | *cb_line_buf = NULL; |
273 | | |
274 | 10.6k | aom_free(*cr_line_buf); |
275 | 10.6k | *cr_line_buf = NULL; |
276 | | |
277 | 10.6k | aom_free(*y_col_buf); |
278 | 10.6k | *y_col_buf = NULL; |
279 | | |
280 | 10.6k | aom_free(*cb_col_buf); |
281 | 10.6k | *cb_col_buf = NULL; |
282 | | |
283 | 10.6k | aom_free(*cr_col_buf); |
284 | 10.6k | *cr_col_buf = NULL; |
285 | | |
286 | 10.6k | aom_free(*luma_grain_block); |
287 | 10.6k | *luma_grain_block = NULL; |
288 | | |
289 | 10.6k | aom_free(*cb_grain_block); |
290 | 10.6k | *cb_grain_block = NULL; |
291 | | |
292 | 10.6k | aom_free(*cr_grain_block); |
293 | 10.6k | *cr_grain_block = NULL; |
294 | 10.6k | } |
295 | | |
296 | | static bool init_arrays(const aom_film_grain_t *params, int luma_stride, |
297 | | int chroma_stride, int ***pred_pos_luma_p, |
298 | | int ***pred_pos_chroma_p, int **luma_grain_block, |
299 | | int **cb_grain_block, int **cr_grain_block, |
300 | | int **y_line_buf, int **cb_line_buf, int **cr_line_buf, |
301 | | int **y_col_buf, int **cb_col_buf, int **cr_col_buf, |
302 | | int luma_grain_samples, int chroma_grain_samples, |
303 | 10.6k | int chroma_subsamp_y, int chroma_subsamp_x) { |
304 | 10.6k | *pred_pos_luma_p = NULL; |
305 | 10.6k | *pred_pos_chroma_p = NULL; |
306 | 10.6k | *luma_grain_block = NULL; |
307 | 10.6k | *cb_grain_block = NULL; |
308 | 10.6k | *cr_grain_block = NULL; |
309 | 10.6k | *y_line_buf = NULL; |
310 | 10.6k | *cb_line_buf = NULL; |
311 | 10.6k | *cr_line_buf = NULL; |
312 | 10.6k | *y_col_buf = NULL; |
313 | 10.6k | *cb_col_buf = NULL; |
314 | 10.6k | *cr_col_buf = NULL; |
315 | | |
316 | 10.6k | memset(scaling_lut_y, 0, sizeof(*scaling_lut_y) * 256); |
317 | 10.6k | memset(scaling_lut_cb, 0, sizeof(*scaling_lut_cb) * 256); |
318 | 10.6k | memset(scaling_lut_cr, 0, sizeof(*scaling_lut_cr) * 256); |
319 | | |
320 | 10.6k | int num_pos_luma = 2 * params->ar_coeff_lag * (params->ar_coeff_lag + 1); |
321 | 10.6k | int num_pos_chroma = num_pos_luma; |
322 | 10.6k | if (params->num_y_points > 0) ++num_pos_chroma; |
323 | | |
324 | 10.6k | int **pred_pos_luma; |
325 | 10.6k | int **pred_pos_chroma; |
326 | | |
327 | 10.6k | pred_pos_luma = (int **)aom_calloc(num_pos_luma, sizeof(*pred_pos_luma)); |
328 | 10.6k | if (!pred_pos_luma) return false; |
329 | | |
330 | 32.7k | for (int row = 0; row < num_pos_luma; row++) { |
331 | 22.1k | pred_pos_luma[row] = (int *)aom_malloc(sizeof(**pred_pos_luma) * 3); |
332 | 22.1k | if (!pred_pos_luma[row]) { |
333 | 0 | dealloc_arrays(params, pred_pos_luma_p, pred_pos_chroma_p, |
334 | 0 | luma_grain_block, cb_grain_block, cr_grain_block, |
335 | 0 | y_line_buf, cb_line_buf, cr_line_buf, y_col_buf, |
336 | 0 | cb_col_buf, cr_col_buf); |
337 | 0 | return false; |
338 | 0 | } |
339 | 22.1k | } |
340 | | |
341 | 10.6k | pred_pos_chroma = |
342 | 10.6k | (int **)aom_calloc(num_pos_chroma, sizeof(*pred_pos_chroma)); |
343 | 10.6k | if (!pred_pos_chroma) { |
344 | 0 | dealloc_arrays(params, pred_pos_luma_p, pred_pos_chroma_p, luma_grain_block, |
345 | 0 | cb_grain_block, cr_grain_block, y_line_buf, cb_line_buf, |
346 | 0 | cr_line_buf, y_col_buf, cb_col_buf, cr_col_buf); |
347 | 0 | return false; |
348 | 0 | } |
349 | | |
350 | 42.2k | for (int row = 0; row < num_pos_chroma; row++) { |
351 | 31.6k | pred_pos_chroma[row] = (int *)aom_malloc(sizeof(**pred_pos_chroma) * 3); |
352 | 31.6k | if (!pred_pos_chroma[row]) { |
353 | 0 | dealloc_arrays(params, pred_pos_luma_p, pred_pos_chroma_p, |
354 | 0 | luma_grain_block, cb_grain_block, cr_grain_block, |
355 | 0 | y_line_buf, cb_line_buf, cr_line_buf, y_col_buf, |
356 | 0 | cb_col_buf, cr_col_buf); |
357 | 0 | return false; |
358 | 0 | } |
359 | 31.6k | } |
360 | | |
361 | 10.6k | int pos_ar_index = 0; |
362 | | |
363 | 15.9k | for (int row = -params->ar_coeff_lag; row < 0; row++) { |
364 | 22.1k | for (int col = -params->ar_coeff_lag; col < params->ar_coeff_lag + 1; |
365 | 16.8k | col++) { |
366 | 16.8k | pred_pos_luma[pos_ar_index][0] = row; |
367 | 16.8k | pred_pos_luma[pos_ar_index][1] = col; |
368 | 16.8k | pred_pos_luma[pos_ar_index][2] = 0; |
369 | | |
370 | 16.8k | pred_pos_chroma[pos_ar_index][0] = row; |
371 | 16.8k | pred_pos_chroma[pos_ar_index][1] = col; |
372 | 16.8k | pred_pos_chroma[pos_ar_index][2] = 0; |
373 | 16.8k | ++pos_ar_index; |
374 | 16.8k | } |
375 | 5.22k | } |
376 | | |
377 | 15.9k | for (int col = -params->ar_coeff_lag; col < 0; col++) { |
378 | 5.22k | pred_pos_luma[pos_ar_index][0] = 0; |
379 | 5.22k | pred_pos_luma[pos_ar_index][1] = col; |
380 | 5.22k | pred_pos_luma[pos_ar_index][2] = 0; |
381 | | |
382 | 5.22k | pred_pos_chroma[pos_ar_index][0] = 0; |
383 | 5.22k | pred_pos_chroma[pos_ar_index][1] = col; |
384 | 5.22k | pred_pos_chroma[pos_ar_index][2] = 0; |
385 | | |
386 | 5.22k | ++pos_ar_index; |
387 | 5.22k | } |
388 | | |
389 | 10.6k | if (params->num_y_points > 0) { |
390 | 9.50k | pred_pos_chroma[pos_ar_index][0] = 0; |
391 | 9.50k | pred_pos_chroma[pos_ar_index][1] = 0; |
392 | 9.50k | pred_pos_chroma[pos_ar_index][2] = 1; |
393 | 9.50k | } |
394 | | |
395 | 10.6k | *pred_pos_luma_p = pred_pos_luma; |
396 | 10.6k | *pred_pos_chroma_p = pred_pos_chroma; |
397 | | |
398 | 10.6k | *y_line_buf = (int *)aom_malloc(sizeof(**y_line_buf) * luma_stride * 2); |
399 | 10.6k | *cb_line_buf = (int *)aom_malloc(sizeof(**cb_line_buf) * chroma_stride * |
400 | 10.6k | (2 >> chroma_subsamp_y)); |
401 | 10.6k | *cr_line_buf = (int *)aom_malloc(sizeof(**cr_line_buf) * chroma_stride * |
402 | 10.6k | (2 >> chroma_subsamp_y)); |
403 | | |
404 | 10.6k | *y_col_buf = |
405 | 10.6k | (int *)aom_malloc(sizeof(**y_col_buf) * (luma_subblock_size_y + 2) * 2); |
406 | 10.6k | *cb_col_buf = |
407 | 10.6k | (int *)aom_malloc(sizeof(**cb_col_buf) * |
408 | 10.6k | (chroma_subblock_size_y + (2 >> chroma_subsamp_y)) * |
409 | 10.6k | (2 >> chroma_subsamp_x)); |
410 | 10.6k | *cr_col_buf = |
411 | 10.6k | (int *)aom_malloc(sizeof(**cr_col_buf) * |
412 | 10.6k | (chroma_subblock_size_y + (2 >> chroma_subsamp_y)) * |
413 | 10.6k | (2 >> chroma_subsamp_x)); |
414 | | |
415 | 10.6k | *luma_grain_block = |
416 | 10.6k | (int *)aom_malloc(sizeof(**luma_grain_block) * luma_grain_samples); |
417 | 10.6k | *cb_grain_block = |
418 | 10.6k | (int *)aom_malloc(sizeof(**cb_grain_block) * chroma_grain_samples); |
419 | 10.6k | *cr_grain_block = |
420 | 10.6k | (int *)aom_malloc(sizeof(**cr_grain_block) * chroma_grain_samples); |
421 | 10.6k | if (!(*pred_pos_luma_p && *pred_pos_chroma_p && *y_line_buf && *cb_line_buf && |
422 | 10.6k | *cr_line_buf && *y_col_buf && *cb_col_buf && *cr_col_buf && |
423 | 10.6k | *luma_grain_block && *cb_grain_block && *cr_grain_block)) { |
424 | 0 | dealloc_arrays(params, pred_pos_luma_p, pred_pos_chroma_p, luma_grain_block, |
425 | 0 | cb_grain_block, cr_grain_block, y_line_buf, cb_line_buf, |
426 | 0 | cr_line_buf, y_col_buf, cb_col_buf, cr_col_buf); |
427 | 0 | return false; |
428 | 0 | } |
429 | 10.6k | return true; |
430 | 10.6k | } |
431 | | |
432 | | // get a number between 0 and 2^bits - 1 |
433 | 98.9M | static INLINE int get_random_number(int bits) { |
434 | 98.9M | uint16_t bit; |
435 | 98.9M | bit = ((random_register >> 0) ^ (random_register >> 1) ^ |
436 | 98.9M | (random_register >> 3) ^ (random_register >> 12)) & |
437 | 98.9M | 1; |
438 | 98.9M | random_register = (random_register >> 1) | (bit << 15); |
439 | 98.9M | return (random_register >> (16 - bits)) & ((1 << bits) - 1); |
440 | 98.9M | } |
441 | | |
442 | 31.3k | static void init_random_generator(int luma_line, uint16_t seed) { |
443 | | // same for the picture |
444 | | |
445 | 31.3k | uint16_t msb = (seed >> 8) & 255; |
446 | 31.3k | uint16_t lsb = seed & 255; |
447 | | |
448 | 31.3k | random_register = (msb << 8) + lsb; |
449 | | |
450 | | // changes for each row |
451 | 31.3k | int luma_num = luma_line >> 5; |
452 | | |
453 | 31.3k | random_register ^= ((luma_num * 37 + 178) & 255) << 8; |
454 | 31.3k | random_register ^= ((luma_num * 173 + 105) & 255); |
455 | 31.3k | } |
456 | | |
457 | | static void generate_luma_grain_block( |
458 | | const aom_film_grain_t *params, int **pred_pos_luma, int *luma_grain_block, |
459 | | int luma_block_size_y, int luma_block_size_x, int luma_grain_stride, |
460 | 10.6k | int left_pad, int top_pad, int right_pad, int bottom_pad) { |
461 | 10.6k | if (params->num_y_points == 0) { |
462 | 1.18k | memset(luma_grain_block, 0, |
463 | 1.18k | sizeof(*luma_grain_block) * luma_block_size_y * luma_grain_stride); |
464 | 1.18k | return; |
465 | 1.18k | } |
466 | | |
467 | 9.50k | int bit_depth = params->bit_depth; |
468 | 9.50k | int gauss_sec_shift = 12 - bit_depth + params->grain_scale_shift; |
469 | | |
470 | 9.50k | int num_pos_luma = 2 * params->ar_coeff_lag * (params->ar_coeff_lag + 1); |
471 | 9.50k | int rounding_offset = (1 << (params->ar_coeff_shift - 1)); |
472 | | |
473 | 703k | for (int i = 0; i < luma_block_size_y; i++) |
474 | 57.5M | for (int j = 0; j < luma_block_size_x; j++) |
475 | 56.8M | luma_grain_block[i * luma_grain_stride + j] = |
476 | 56.8M | (gaussian_sequence[get_random_number(gauss_bits)] + |
477 | 56.8M | ((1 << gauss_sec_shift) >> 1)) >> |
478 | 56.8M | gauss_sec_shift; |
479 | | |
480 | 674k | for (int i = top_pad; i < luma_block_size_y - bottom_pad; i++) |
481 | 51.2M | for (int j = left_pad; j < luma_block_size_x - right_pad; j++) { |
482 | 50.5M | int wsum = 0; |
483 | 150M | for (int pos = 0; pos < num_pos_luma; pos++) { |
484 | 100M | wsum = wsum + params->ar_coeffs_y[pos] * |
485 | 100M | luma_grain_block[(i + pred_pos_luma[pos][0]) * |
486 | 100M | luma_grain_stride + |
487 | 100M | j + pred_pos_luma[pos][1]]; |
488 | 100M | } |
489 | 50.5M | luma_grain_block[i * luma_grain_stride + j] = |
490 | 50.5M | clamp(luma_grain_block[i * luma_grain_stride + j] + |
491 | 50.5M | ((wsum + rounding_offset) >> params->ar_coeff_shift), |
492 | 50.5M | grain_min, grain_max); |
493 | 50.5M | } |
494 | 9.50k | } |
495 | | |
496 | | static bool generate_chroma_grain_blocks( |
497 | | const aom_film_grain_t *params, int **pred_pos_chroma, |
498 | | int *luma_grain_block, int *cb_grain_block, int *cr_grain_block, |
499 | | int luma_grain_stride, int chroma_block_size_y, int chroma_block_size_x, |
500 | | int chroma_grain_stride, int left_pad, int top_pad, int right_pad, |
501 | 10.6k | int bottom_pad, int chroma_subsamp_y, int chroma_subsamp_x) { |
502 | 10.6k | int bit_depth = params->bit_depth; |
503 | 10.6k | int gauss_sec_shift = 12 - bit_depth + params->grain_scale_shift; |
504 | | |
505 | 10.6k | int num_pos_chroma = 2 * params->ar_coeff_lag * (params->ar_coeff_lag + 1); |
506 | 10.6k | if (params->num_y_points > 0) ++num_pos_chroma; |
507 | 10.6k | int rounding_offset = (1 << (params->ar_coeff_shift - 1)); |
508 | 10.6k | int chroma_grain_block_size = chroma_block_size_y * chroma_grain_stride; |
509 | | |
510 | 10.6k | if (params->num_cb_points || params->chroma_scaling_from_luma) { |
511 | 8.41k | init_random_generator(7 << 5, params->random_seed); |
512 | | |
513 | 425k | for (int i = 0; i < chroma_block_size_y; i++) |
514 | 25.5M | for (int j = 0; j < chroma_block_size_x; j++) |
515 | 25.1M | cb_grain_block[i * chroma_grain_stride + j] = |
516 | 25.1M | (gaussian_sequence[get_random_number(gauss_bits)] + |
517 | 25.1M | ((1 << gauss_sec_shift) >> 1)) >> |
518 | 25.1M | gauss_sec_shift; |
519 | 8.41k | } else { |
520 | 2.27k | memset(cb_grain_block, 0, |
521 | 2.27k | sizeof(*cb_grain_block) * chroma_grain_block_size); |
522 | 2.27k | } |
523 | | |
524 | 10.6k | if (params->num_cr_points || params->chroma_scaling_from_luma) { |
525 | 7.08k | init_random_generator(11 << 5, params->random_seed); |
526 | | |
527 | 327k | for (int i = 0; i < chroma_block_size_y; i++) |
528 | 17.2M | for (int j = 0; j < chroma_block_size_x; j++) |
529 | 16.8M | cr_grain_block[i * chroma_grain_stride + j] = |
530 | 16.8M | (gaussian_sequence[get_random_number(gauss_bits)] + |
531 | 16.8M | ((1 << gauss_sec_shift) >> 1)) >> |
532 | 16.8M | gauss_sec_shift; |
533 | 7.08k | } else { |
534 | 3.59k | memset(cr_grain_block, 0, |
535 | 3.59k | sizeof(*cr_grain_block) * chroma_grain_block_size); |
536 | 3.59k | } |
537 | | |
538 | 550k | for (int i = top_pad; i < chroma_block_size_y - bottom_pad; i++) |
539 | 32.0M | for (int j = left_pad; j < chroma_block_size_x - right_pad; j++) { |
540 | 31.5M | int wsum_cb = 0; |
541 | 31.5M | int wsum_cr = 0; |
542 | 104M | for (int pos = 0; pos < num_pos_chroma; pos++) { |
543 | 73.2M | if (pred_pos_chroma[pos][2] == 0) { |
544 | 43.5M | wsum_cb = wsum_cb + params->ar_coeffs_cb[pos] * |
545 | 43.5M | cb_grain_block[(i + pred_pos_chroma[pos][0]) * |
546 | 43.5M | chroma_grain_stride + |
547 | 43.5M | j + pred_pos_chroma[pos][1]]; |
548 | 43.5M | wsum_cr = wsum_cr + params->ar_coeffs_cr[pos] * |
549 | 43.5M | cr_grain_block[(i + pred_pos_chroma[pos][0]) * |
550 | 43.5M | chroma_grain_stride + |
551 | 43.5M | j + pred_pos_chroma[pos][1]]; |
552 | 43.5M | } else if (pred_pos_chroma[pos][2] == 1) { |
553 | 29.6M | int av_luma = 0; |
554 | 29.6M | int luma_coord_y = ((i - top_pad) << chroma_subsamp_y) + top_pad; |
555 | 29.6M | int luma_coord_x = ((j - left_pad) << chroma_subsamp_x) + left_pad; |
556 | | |
557 | 65.9M | for (int k = luma_coord_y; k < luma_coord_y + chroma_subsamp_y + 1; |
558 | 36.2M | k++) |
559 | 86.7M | for (int l = luma_coord_x; l < luma_coord_x + chroma_subsamp_x + 1; |
560 | 50.5M | l++) |
561 | 50.5M | av_luma += luma_grain_block[k * luma_grain_stride + l]; |
562 | | |
563 | 29.6M | av_luma = |
564 | 29.6M | (av_luma + ((1 << (chroma_subsamp_y + chroma_subsamp_x)) >> 1)) >> |
565 | 29.6M | (chroma_subsamp_y + chroma_subsamp_x); |
566 | | |
567 | 29.6M | wsum_cb = wsum_cb + params->ar_coeffs_cb[pos] * av_luma; |
568 | 29.6M | wsum_cr = wsum_cr + params->ar_coeffs_cr[pos] * av_luma; |
569 | 29.6M | } else { |
570 | 0 | fprintf( |
571 | 0 | stderr, |
572 | 0 | "Grain synthesis: prediction between two chroma components is " |
573 | 0 | "not supported!"); |
574 | 0 | return false; |
575 | 0 | } |
576 | 73.2M | } |
577 | 31.5M | if (params->num_cb_points || params->chroma_scaling_from_luma) |
578 | 21.4M | cb_grain_block[i * chroma_grain_stride + j] = |
579 | 21.4M | clamp(cb_grain_block[i * chroma_grain_stride + j] + |
580 | 21.4M | ((wsum_cb + rounding_offset) >> params->ar_coeff_shift), |
581 | 21.4M | grain_min, grain_max); |
582 | 31.5M | if (params->num_cr_points || params->chroma_scaling_from_luma) |
583 | 14.0M | cr_grain_block[i * chroma_grain_stride + j] = |
584 | 14.0M | clamp(cr_grain_block[i * chroma_grain_stride + j] + |
585 | 14.0M | ((wsum_cr + rounding_offset) >> params->ar_coeff_shift), |
586 | 14.0M | grain_min, grain_max); |
587 | 31.5M | } |
588 | 10.6k | return true; |
589 | 10.6k | } |
590 | | |
591 | | static void init_scaling_function(const int scaling_points[][2], int num_points, |
592 | 20.9k | int scaling_lut[]) { |
593 | 20.9k | if (num_points == 0) return; |
594 | | |
595 | 667k | for (int i = 0; i < scaling_points[0][0]; i++) |
596 | 653k | scaling_lut[i] = scaling_points[0][1]; |
597 | | |
598 | 30.1k | for (int point = 0; point < num_points - 1; point++) { |
599 | 16.2k | int delta_y = scaling_points[point + 1][1] - scaling_points[point][1]; |
600 | 16.2k | int delta_x = scaling_points[point + 1][0] - scaling_points[point][0]; |
601 | | |
602 | 16.2k | int64_t delta = delta_y * ((65536 + (delta_x >> 1)) / delta_x); |
603 | | |
604 | 1.50M | for (int x = 0; x < delta_x; x++) { |
605 | 1.48M | scaling_lut[scaling_points[point][0] + x] = |
606 | 1.48M | scaling_points[point][1] + (int)((x * delta + 32768) >> 16); |
607 | 1.48M | } |
608 | 16.2k | } |
609 | | |
610 | 1.42M | for (int i = scaling_points[num_points - 1][0]; i < 256; i++) |
611 | 1.40M | scaling_lut[i] = scaling_points[num_points - 1][1]; |
612 | 13.8k | } |
613 | | |
614 | | // function that extracts samples from a LUT (and interpolates intemediate |
615 | | // frames for 10- and 12-bit video) |
616 | 34.7M | static int scale_LUT(int *scaling_lut, int index, int bit_depth) { |
617 | 34.7M | int x = index >> (bit_depth - 8); |
618 | | |
619 | 34.7M | if (!(bit_depth - 8) || x == 255) |
620 | 6.72M | return scaling_lut[x]; |
621 | 28.0M | else |
622 | 28.0M | return scaling_lut[x] + (((scaling_lut[x + 1] - scaling_lut[x]) * |
623 | 28.0M | (index & ((1 << (bit_depth - 8)) - 1)) + |
624 | 28.0M | (1 << (bit_depth - 9))) >> |
625 | 28.0M | (bit_depth - 8)); |
626 | 34.7M | } |
627 | | |
628 | | static void add_noise_to_block(const aom_film_grain_t *params, uint8_t *luma, |
629 | | uint8_t *cb, uint8_t *cr, int luma_stride, |
630 | | int chroma_stride, int *luma_grain, |
631 | | int *cb_grain, int *cr_grain, |
632 | | int luma_grain_stride, int chroma_grain_stride, |
633 | | int half_luma_height, int half_luma_width, |
634 | | int bit_depth, int chroma_subsamp_y, |
635 | 28.9k | int chroma_subsamp_x, int mc_identity) { |
636 | 28.9k | int cb_mult = params->cb_mult - 128; // fixed scale |
637 | 28.9k | int cb_luma_mult = params->cb_luma_mult - 128; // fixed scale |
638 | 28.9k | int cb_offset = params->cb_offset - 256; |
639 | | |
640 | 28.9k | int cr_mult = params->cr_mult - 128; // fixed scale |
641 | 28.9k | int cr_luma_mult = params->cr_luma_mult - 128; // fixed scale |
642 | 28.9k | int cr_offset = params->cr_offset - 256; |
643 | | |
644 | 28.9k | int rounding_offset = (1 << (params->scaling_shift - 1)); |
645 | | |
646 | 28.9k | int apply_y = params->num_y_points > 0 ? 1 : 0; |
647 | 28.9k | int apply_cb = |
648 | 28.9k | (params->num_cb_points > 0 || params->chroma_scaling_from_luma) ? 1 : 0; |
649 | 28.9k | int apply_cr = |
650 | 28.9k | (params->num_cr_points > 0 || params->chroma_scaling_from_luma) ? 1 : 0; |
651 | | |
652 | 28.9k | if (params->chroma_scaling_from_luma) { |
653 | 16.4k | cb_mult = 0; // fixed scale |
654 | 16.4k | cb_luma_mult = 64; // fixed scale |
655 | 16.4k | cb_offset = 0; |
656 | | |
657 | 16.4k | cr_mult = 0; // fixed scale |
658 | 16.4k | cr_luma_mult = 64; // fixed scale |
659 | 16.4k | cr_offset = 0; |
660 | 16.4k | } |
661 | | |
662 | 28.9k | int min_luma, max_luma, min_chroma, max_chroma; |
663 | | |
664 | 28.9k | if (params->clip_to_restricted_range) { |
665 | 6.28k | min_luma = min_luma_legal_range; |
666 | 6.28k | max_luma = max_luma_legal_range; |
667 | | |
668 | 6.28k | if (mc_identity) { |
669 | 0 | min_chroma = min_luma_legal_range; |
670 | 0 | max_chroma = max_luma_legal_range; |
671 | 6.28k | } else { |
672 | 6.28k | min_chroma = min_chroma_legal_range; |
673 | 6.28k | max_chroma = max_chroma_legal_range; |
674 | 6.28k | } |
675 | 22.7k | } else { |
676 | 22.7k | min_luma = min_chroma = 0; |
677 | 22.7k | max_luma = max_chroma = 255; |
678 | 22.7k | } |
679 | | |
680 | 306k | for (int i = 0; i < (half_luma_height << (1 - chroma_subsamp_y)); i++) { |
681 | 3.04M | for (int j = 0; j < (half_luma_width << (1 - chroma_subsamp_x)); j++) { |
682 | 2.76M | int average_luma = 0; |
683 | 2.76M | if (chroma_subsamp_x) { |
684 | 2.76M | average_luma = (luma[(i << chroma_subsamp_y) * luma_stride + |
685 | 2.76M | (j << chroma_subsamp_x)] + |
686 | 2.76M | luma[(i << chroma_subsamp_y) * luma_stride + |
687 | 2.76M | (j << chroma_subsamp_x) + 1] + |
688 | 2.76M | 1) >> |
689 | 2.76M | 1; |
690 | 2.76M | } else { |
691 | 5.29k | average_luma = luma[(i << chroma_subsamp_y) * luma_stride + j]; |
692 | 5.29k | } |
693 | | |
694 | 2.76M | if (apply_cb) { |
695 | 1.74M | cb[i * chroma_stride + j] = clamp( |
696 | 1.74M | cb[i * chroma_stride + j] + |
697 | 1.74M | ((scale_LUT(scaling_lut_cb, |
698 | 1.74M | clamp(((average_luma * cb_luma_mult + |
699 | 1.74M | cb_mult * cb[i * chroma_stride + j]) >> |
700 | 1.74M | 6) + |
701 | 1.74M | cb_offset, |
702 | 1.74M | 0, (256 << (bit_depth - 8)) - 1), |
703 | 1.74M | 8) * |
704 | 1.74M | cb_grain[i * chroma_grain_stride + j] + |
705 | 1.74M | rounding_offset) >> |
706 | 1.74M | params->scaling_shift), |
707 | 1.74M | min_chroma, max_chroma); |
708 | 1.74M | } |
709 | | |
710 | 2.76M | if (apply_cr) { |
711 | 1.74M | cr[i * chroma_stride + j] = clamp( |
712 | 1.74M | cr[i * chroma_stride + j] + |
713 | 1.74M | ((scale_LUT(scaling_lut_cr, |
714 | 1.74M | clamp(((average_luma * cr_luma_mult + |
715 | 1.74M | cr_mult * cr[i * chroma_stride + j]) >> |
716 | 1.74M | 6) + |
717 | 1.74M | cr_offset, |
718 | 1.74M | 0, (256 << (bit_depth - 8)) - 1), |
719 | 1.74M | 8) * |
720 | 1.74M | cr_grain[i * chroma_grain_stride + j] + |
721 | 1.74M | rounding_offset) >> |
722 | 1.74M | params->scaling_shift), |
723 | 1.74M | min_chroma, max_chroma); |
724 | 1.74M | } |
725 | 2.76M | } |
726 | 277k | } |
727 | | |
728 | 28.9k | if (apply_y) { |
729 | 150k | for (int i = 0; i < (half_luma_height << 1); i++) { |
730 | 2.61M | for (int j = 0; j < (half_luma_width << 1); j++) { |
731 | 2.47M | luma[i * luma_stride + j] = |
732 | 2.47M | clamp(luma[i * luma_stride + j] + |
733 | 2.47M | ((scale_LUT(scaling_lut_y, luma[i * luma_stride + j], 8) * |
734 | 2.47M | luma_grain[i * luma_grain_stride + j] + |
735 | 2.47M | rounding_offset) >> |
736 | 2.47M | params->scaling_shift), |
737 | 2.47M | min_luma, max_luma); |
738 | 2.47M | } |
739 | 139k | } |
740 | 10.4k | } |
741 | 28.9k | } |
742 | | |
743 | | static void add_noise_to_block_hbd( |
744 | | const aom_film_grain_t *params, uint16_t *luma, uint16_t *cb, uint16_t *cr, |
745 | | int luma_stride, int chroma_stride, int *luma_grain, int *cb_grain, |
746 | | int *cr_grain, int luma_grain_stride, int chroma_grain_stride, |
747 | | int half_luma_height, int half_luma_width, int bit_depth, |
748 | 76.1k | int chroma_subsamp_y, int chroma_subsamp_x, int mc_identity) { |
749 | 76.1k | int cb_mult = params->cb_mult - 128; // fixed scale |
750 | 76.1k | int cb_luma_mult = params->cb_luma_mult - 128; // fixed scale |
751 | | // offset value depends on the bit depth |
752 | 76.1k | int cb_offset = (params->cb_offset << (bit_depth - 8)) - (1 << bit_depth); |
753 | | |
754 | 76.1k | int cr_mult = params->cr_mult - 128; // fixed scale |
755 | 76.1k | int cr_luma_mult = params->cr_luma_mult - 128; // fixed scale |
756 | | // offset value depends on the bit depth |
757 | 76.1k | int cr_offset = (params->cr_offset << (bit_depth - 8)) - (1 << bit_depth); |
758 | | |
759 | 76.1k | int rounding_offset = (1 << (params->scaling_shift - 1)); |
760 | | |
761 | 76.1k | int apply_y = params->num_y_points > 0 ? 1 : 0; |
762 | 76.1k | int apply_cb = |
763 | 76.1k | (params->num_cb_points > 0 || params->chroma_scaling_from_luma) > 0 ? 1 |
764 | 76.1k | : 0; |
765 | 76.1k | int apply_cr = |
766 | 76.1k | (params->num_cr_points > 0 || params->chroma_scaling_from_luma) > 0 ? 1 |
767 | 76.1k | : 0; |
768 | | |
769 | 76.1k | if (params->chroma_scaling_from_luma) { |
770 | 56.2k | cb_mult = 0; // fixed scale |
771 | 56.2k | cb_luma_mult = 64; // fixed scale |
772 | 56.2k | cb_offset = 0; |
773 | | |
774 | 56.2k | cr_mult = 0; // fixed scale |
775 | 56.2k | cr_luma_mult = 64; // fixed scale |
776 | 56.2k | cr_offset = 0; |
777 | 56.2k | } |
778 | | |
779 | 76.1k | int min_luma, max_luma, min_chroma, max_chroma; |
780 | | |
781 | 76.1k | if (params->clip_to_restricted_range) { |
782 | 5.31k | min_luma = min_luma_legal_range << (bit_depth - 8); |
783 | 5.31k | max_luma = max_luma_legal_range << (bit_depth - 8); |
784 | | |
785 | 5.31k | if (mc_identity) { |
786 | 0 | min_chroma = min_luma_legal_range << (bit_depth - 8); |
787 | 0 | max_chroma = max_luma_legal_range << (bit_depth - 8); |
788 | 5.31k | } else { |
789 | 5.31k | min_chroma = min_chroma_legal_range << (bit_depth - 8); |
790 | 5.31k | max_chroma = max_chroma_legal_range << (bit_depth - 8); |
791 | 5.31k | } |
792 | 70.8k | } else { |
793 | 70.8k | min_luma = min_chroma = 0; |
794 | 70.8k | max_luma = max_chroma = (256 << (bit_depth - 8)) - 1; |
795 | 70.8k | } |
796 | | |
797 | 1.00M | for (int i = 0; i < (half_luma_height << (1 - chroma_subsamp_y)); i++) { |
798 | 12.0M | for (int j = 0; j < (half_luma_width << (1 - chroma_subsamp_x)); j++) { |
799 | 11.0M | int average_luma = 0; |
800 | 11.0M | if (chroma_subsamp_x) { |
801 | 3.23M | average_luma = (luma[(i << chroma_subsamp_y) * luma_stride + |
802 | 3.23M | (j << chroma_subsamp_x)] + |
803 | 3.23M | luma[(i << chroma_subsamp_y) * luma_stride + |
804 | 3.23M | (j << chroma_subsamp_x) + 1] + |
805 | 3.23M | 1) >> |
806 | 3.23M | 1; |
807 | 7.84M | } else { |
808 | 7.84M | average_luma = luma[(i << chroma_subsamp_y) * luma_stride + j]; |
809 | 7.84M | } |
810 | | |
811 | 11.0M | if (apply_cb) { |
812 | 10.5M | cb[i * chroma_stride + j] = clamp( |
813 | 10.5M | cb[i * chroma_stride + j] + |
814 | 10.5M | ((scale_LUT(scaling_lut_cb, |
815 | 10.5M | clamp(((average_luma * cb_luma_mult + |
816 | 10.5M | cb_mult * cb[i * chroma_stride + j]) >> |
817 | 10.5M | 6) + |
818 | 10.5M | cb_offset, |
819 | 10.5M | 0, (256 << (bit_depth - 8)) - 1), |
820 | 10.5M | bit_depth) * |
821 | 10.5M | cb_grain[i * chroma_grain_stride + j] + |
822 | 10.5M | rounding_offset) >> |
823 | 10.5M | params->scaling_shift), |
824 | 10.5M | min_chroma, max_chroma); |
825 | 10.5M | } |
826 | 11.0M | if (apply_cr) { |
827 | 8.31M | cr[i * chroma_stride + j] = clamp( |
828 | 8.31M | cr[i * chroma_stride + j] + |
829 | 8.31M | ((scale_LUT(scaling_lut_cr, |
830 | 8.31M | clamp(((average_luma * cr_luma_mult + |
831 | 8.31M | cr_mult * cr[i * chroma_stride + j]) >> |
832 | 8.31M | 6) + |
833 | 8.31M | cr_offset, |
834 | 8.31M | 0, (256 << (bit_depth - 8)) - 1), |
835 | 8.31M | bit_depth) * |
836 | 8.31M | cr_grain[i * chroma_grain_stride + j] + |
837 | 8.31M | rounding_offset) >> |
838 | 8.31M | params->scaling_shift), |
839 | 8.31M | min_chroma, max_chroma); |
840 | 8.31M | } |
841 | 11.0M | } |
842 | 930k | } |
843 | | |
844 | 76.1k | if (apply_y) { |
845 | 696k | for (int i = 0; i < (half_luma_height << 1); i++) { |
846 | 10.5M | for (int j = 0; j < (half_luma_width << 1); j++) { |
847 | 9.94M | luma[i * luma_stride + j] = |
848 | 9.94M | clamp(luma[i * luma_stride + j] + |
849 | 9.94M | ((scale_LUT(scaling_lut_y, luma[i * luma_stride + j], |
850 | 9.94M | bit_depth) * |
851 | 9.94M | luma_grain[i * luma_grain_stride + j] + |
852 | 9.94M | rounding_offset) >> |
853 | 9.94M | params->scaling_shift), |
854 | 9.94M | min_luma, max_luma); |
855 | 9.94M | } |
856 | 649k | } |
857 | 46.8k | } |
858 | 76.1k | } |
859 | | |
860 | | static void copy_rect(uint8_t *src, int src_stride, uint8_t *dst, |
861 | | int dst_stride, int width, int height, |
862 | 31.9k | int use_high_bit_depth) { |
863 | 31.9k | int hbd_coeff = use_high_bit_depth ? 2 : 1; |
864 | 635k | while (height) { |
865 | 603k | memcpy(dst, src, width * sizeof(uint8_t) * hbd_coeff); |
866 | 603k | src += src_stride; |
867 | 603k | dst += dst_stride; |
868 | 603k | --height; |
869 | 603k | } |
870 | 31.9k | return; |
871 | 31.9k | } |
872 | | |
873 | | static void copy_area(int *src, int src_stride, int *dst, int dst_stride, |
874 | 338k | int width, int height) { |
875 | 3.01M | while (height) { |
876 | 2.67M | memcpy(dst, src, width * sizeof(*src)); |
877 | 2.67M | src += src_stride; |
878 | 2.67M | dst += dst_stride; |
879 | 2.67M | --height; |
880 | 2.67M | } |
881 | 338k | return; |
882 | 338k | } |
883 | | |
884 | | static void extend_even(uint8_t *dst, int dst_stride, int width, int height, |
885 | 10.6k | int use_high_bit_depth) { |
886 | 10.6k | if ((width & 1) == 0 && (height & 1) == 0) return; |
887 | 9.13k | if (use_high_bit_depth) { |
888 | 6.81k | uint16_t *dst16 = (uint16_t *)dst; |
889 | 6.81k | int dst16_stride = dst_stride / 2; |
890 | 6.81k | if (width & 1) { |
891 | 130k | for (int i = 0; i < height; ++i) |
892 | 124k | dst16[i * dst16_stride + width] = dst16[i * dst16_stride + width - 1]; |
893 | 6.25k | } |
894 | 6.81k | width = (width + 1) & (~1); |
895 | 6.81k | if (height & 1) { |
896 | 851 | memcpy(&dst16[height * dst16_stride], &dst16[(height - 1) * dst16_stride], |
897 | 851 | sizeof(*dst16) * width); |
898 | 851 | } |
899 | 6.81k | } else { |
900 | 2.32k | if (width & 1) { |
901 | 27.3k | for (int i = 0; i < height; ++i) |
902 | 25.3k | dst[i * dst_stride + width] = dst[i * dst_stride + width - 1]; |
903 | 2.00k | } |
904 | 2.32k | width = (width + 1) & (~1); |
905 | 2.32k | if (height & 1) { |
906 | 897 | memcpy(&dst[height * dst_stride], &dst[(height - 1) * dst_stride], |
907 | 897 | sizeof(*dst) * width); |
908 | 897 | } |
909 | 2.32k | } |
910 | 9.13k | } |
911 | | |
912 | | static void ver_boundary_overlap(int *left_block, int left_stride, |
913 | | int *right_block, int right_stride, |
914 | | int *dst_block, int dst_stride, int width, |
915 | 90.6k | int height) { |
916 | 90.6k | if (width == 1) { |
917 | 613k | while (height) { |
918 | 571k | *dst_block = clamp((*left_block * 23 + *right_block * 22 + 16) >> 5, |
919 | 571k | grain_min, grain_max); |
920 | 571k | left_block += left_stride; |
921 | 571k | right_block += right_stride; |
922 | 571k | dst_block += dst_stride; |
923 | 571k | --height; |
924 | 571k | } |
925 | 42.5k | return; |
926 | 48.0k | } else if (width == 2) { |
927 | 1.26M | while (height) { |
928 | 1.21M | dst_block[0] = clamp((27 * left_block[0] + 17 * right_block[0] + 16) >> 5, |
929 | 1.21M | grain_min, grain_max); |
930 | 1.21M | dst_block[1] = clamp((17 * left_block[1] + 27 * right_block[1] + 16) >> 5, |
931 | 1.21M | grain_min, grain_max); |
932 | 1.21M | left_block += left_stride; |
933 | 1.21M | right_block += right_stride; |
934 | 1.21M | dst_block += dst_stride; |
935 | 1.21M | --height; |
936 | 1.21M | } |
937 | 48.0k | return; |
938 | 48.0k | } |
939 | 90.6k | } |
940 | | |
941 | | static void hor_boundary_overlap(int *top_block, int top_stride, |
942 | | int *bottom_block, int bottom_stride, |
943 | | int *dst_block, int dst_stride, int width, |
944 | 111k | int height) { |
945 | 111k | if (height == 1) { |
946 | 372k | while (width) { |
947 | 331k | *dst_block = clamp((*top_block * 23 + *bottom_block * 22 + 16) >> 5, |
948 | 331k | grain_min, grain_max); |
949 | 331k | ++top_block; |
950 | 331k | ++bottom_block; |
951 | 331k | ++dst_block; |
952 | 331k | --width; |
953 | 331k | } |
954 | 41.7k | return; |
955 | 69.4k | } else if (height == 2) { |
956 | 1.10M | while (width) { |
957 | 1.03M | dst_block[0] = clamp((27 * top_block[0] + 17 * bottom_block[0] + 16) >> 5, |
958 | 1.03M | grain_min, grain_max); |
959 | 1.03M | dst_block[dst_stride] = clamp((17 * top_block[top_stride] + |
960 | 1.03M | 27 * bottom_block[bottom_stride] + 16) >> |
961 | 1.03M | 5, |
962 | 1.03M | grain_min, grain_max); |
963 | 1.03M | ++top_block; |
964 | 1.03M | ++bottom_block; |
965 | 1.03M | ++dst_block; |
966 | 1.03M | --width; |
967 | 1.03M | } |
968 | 69.4k | return; |
969 | 69.4k | } |
970 | 111k | } |
971 | | |
972 | | int av1_add_film_grain(const aom_film_grain_t *params, const aom_image_t *src, |
973 | 10.6k | aom_image_t *dst) { |
974 | 10.6k | uint8_t *luma, *cb, *cr; |
975 | 10.6k | int height, width, luma_stride, chroma_stride; |
976 | 10.6k | int use_high_bit_depth = 0; |
977 | 10.6k | int chroma_subsamp_x = 0; |
978 | 10.6k | int chroma_subsamp_y = 0; |
979 | 10.6k | int mc_identity = src->mc == AOM_CICP_MC_IDENTITY ? 1 : 0; |
980 | | |
981 | 10.6k | switch (src->fmt) { |
982 | 0 | case AOM_IMG_FMT_AOMI420: |
983 | 1.40k | case AOM_IMG_FMT_I420: |
984 | 1.40k | use_high_bit_depth = 0; |
985 | 1.40k | chroma_subsamp_x = 1; |
986 | 1.40k | chroma_subsamp_y = 1; |
987 | 1.40k | break; |
988 | 4.53k | case AOM_IMG_FMT_I42016: |
989 | 4.53k | use_high_bit_depth = 1; |
990 | 4.53k | chroma_subsamp_x = 1; |
991 | 4.53k | chroma_subsamp_y = 1; |
992 | 4.53k | break; |
993 | | // case AOM_IMG_FMT_444A: |
994 | 1.29k | case AOM_IMG_FMT_I444: |
995 | 1.29k | use_high_bit_depth = 0; |
996 | 1.29k | chroma_subsamp_x = 0; |
997 | 1.29k | chroma_subsamp_y = 0; |
998 | 1.29k | break; |
999 | 2.83k | case AOM_IMG_FMT_I44416: |
1000 | 2.83k | use_high_bit_depth = 1; |
1001 | 2.83k | chroma_subsamp_x = 0; |
1002 | 2.83k | chroma_subsamp_y = 0; |
1003 | 2.83k | break; |
1004 | 134 | case AOM_IMG_FMT_I422: |
1005 | 134 | use_high_bit_depth = 0; |
1006 | 134 | chroma_subsamp_x = 1; |
1007 | 134 | chroma_subsamp_y = 0; |
1008 | 134 | break; |
1009 | 482 | case AOM_IMG_FMT_I42216: |
1010 | 482 | use_high_bit_depth = 1; |
1011 | 482 | chroma_subsamp_x = 1; |
1012 | 482 | chroma_subsamp_y = 0; |
1013 | 482 | break; |
1014 | 0 | default: // unknown input format |
1015 | 0 | fprintf(stderr, "Film grain error: input format is not supported!"); |
1016 | 0 | return -1; |
1017 | 10.6k | } |
1018 | | |
1019 | 10.6k | assert(params->bit_depth == src->bit_depth); |
1020 | | |
1021 | 0 | dst->fmt = src->fmt; |
1022 | 10.6k | dst->bit_depth = src->bit_depth; |
1023 | | |
1024 | 10.6k | dst->r_w = src->r_w; |
1025 | 10.6k | dst->r_h = src->r_h; |
1026 | 10.6k | dst->d_w = src->d_w; |
1027 | 10.6k | dst->d_h = src->d_h; |
1028 | | |
1029 | 10.6k | dst->cp = src->cp; |
1030 | 10.6k | dst->tc = src->tc; |
1031 | 10.6k | dst->mc = src->mc; |
1032 | | |
1033 | 10.6k | dst->monochrome = src->monochrome; |
1034 | 10.6k | dst->csp = src->csp; |
1035 | 10.6k | dst->range = src->range; |
1036 | | |
1037 | 10.6k | dst->x_chroma_shift = src->x_chroma_shift; |
1038 | 10.6k | dst->y_chroma_shift = src->y_chroma_shift; |
1039 | | |
1040 | 10.6k | dst->temporal_id = src->temporal_id; |
1041 | 10.6k | dst->spatial_id = src->spatial_id; |
1042 | | |
1043 | 10.6k | width = src->d_w % 2 ? src->d_w + 1 : src->d_w; |
1044 | 10.6k | height = src->d_h % 2 ? src->d_h + 1 : src->d_h; |
1045 | | |
1046 | 10.6k | copy_rect(src->planes[AOM_PLANE_Y], src->stride[AOM_PLANE_Y], |
1047 | 10.6k | dst->planes[AOM_PLANE_Y], dst->stride[AOM_PLANE_Y], src->d_w, |
1048 | 10.6k | src->d_h, use_high_bit_depth); |
1049 | | // Note that dst is already assumed to be aligned to even. |
1050 | 10.6k | extend_even(dst->planes[AOM_PLANE_Y], dst->stride[AOM_PLANE_Y], src->d_w, |
1051 | 10.6k | src->d_h, use_high_bit_depth); |
1052 | | |
1053 | 10.6k | if (!src->monochrome) { |
1054 | 10.6k | copy_rect(src->planes[AOM_PLANE_U], src->stride[AOM_PLANE_U], |
1055 | 10.6k | dst->planes[AOM_PLANE_U], dst->stride[AOM_PLANE_U], |
1056 | 10.6k | width >> chroma_subsamp_x, height >> chroma_subsamp_y, |
1057 | 10.6k | use_high_bit_depth); |
1058 | | |
1059 | 10.6k | copy_rect(src->planes[AOM_PLANE_V], src->stride[AOM_PLANE_V], |
1060 | 10.6k | dst->planes[AOM_PLANE_V], dst->stride[AOM_PLANE_V], |
1061 | 10.6k | width >> chroma_subsamp_x, height >> chroma_subsamp_y, |
1062 | 10.6k | use_high_bit_depth); |
1063 | 10.6k | } |
1064 | | |
1065 | 10.6k | luma = dst->planes[AOM_PLANE_Y]; |
1066 | 10.6k | cb = dst->planes[AOM_PLANE_U]; |
1067 | 10.6k | cr = dst->planes[AOM_PLANE_V]; |
1068 | | |
1069 | | // luma and chroma strides in samples |
1070 | 10.6k | luma_stride = dst->stride[AOM_PLANE_Y] >> use_high_bit_depth; |
1071 | 10.6k | chroma_stride = dst->stride[AOM_PLANE_U] >> use_high_bit_depth; |
1072 | | |
1073 | 10.6k | return av1_add_film_grain_run( |
1074 | 10.6k | params, luma, cb, cr, height, width, luma_stride, chroma_stride, |
1075 | 10.6k | use_high_bit_depth, chroma_subsamp_y, chroma_subsamp_x, mc_identity); |
1076 | 10.6k | } |
1077 | | |
1078 | | int av1_add_film_grain_run(const aom_film_grain_t *params, uint8_t *luma, |
1079 | | uint8_t *cb, uint8_t *cr, int height, int width, |
1080 | | int luma_stride, int chroma_stride, |
1081 | | int use_high_bit_depth, int chroma_subsamp_y, |
1082 | 10.6k | int chroma_subsamp_x, int mc_identity) { |
1083 | 10.6k | int **pred_pos_luma; |
1084 | 10.6k | int **pred_pos_chroma; |
1085 | 10.6k | int *luma_grain_block; |
1086 | 10.6k | int *cb_grain_block; |
1087 | 10.6k | int *cr_grain_block; |
1088 | | |
1089 | 10.6k | int *y_line_buf; |
1090 | 10.6k | int *cb_line_buf; |
1091 | 10.6k | int *cr_line_buf; |
1092 | | |
1093 | 10.6k | int *y_col_buf; |
1094 | 10.6k | int *cb_col_buf; |
1095 | 10.6k | int *cr_col_buf; |
1096 | | |
1097 | 10.6k | random_register = params->random_seed; |
1098 | | |
1099 | 10.6k | int left_pad = 3; |
1100 | 10.6k | int right_pad = 3; // padding to offset for AR coefficients |
1101 | 10.6k | int top_pad = 3; |
1102 | 10.6k | int bottom_pad = 0; |
1103 | | |
1104 | 10.6k | int ar_padding = 3; // maximum lag used for stabilization of AR coefficients |
1105 | | |
1106 | 10.6k | luma_subblock_size_y = 32; |
1107 | 10.6k | luma_subblock_size_x = 32; |
1108 | | |
1109 | 10.6k | chroma_subblock_size_y = luma_subblock_size_y >> chroma_subsamp_y; |
1110 | 10.6k | chroma_subblock_size_x = luma_subblock_size_x >> chroma_subsamp_x; |
1111 | | |
1112 | | // Initial padding is only needed for generation of |
1113 | | // film grain templates (to stabilize the AR process) |
1114 | | // Only a 64x64 luma and 32x32 chroma part of a template |
1115 | | // is used later for adding grain, padding can be discarded |
1116 | | |
1117 | 10.6k | int luma_block_size_y = |
1118 | 10.6k | top_pad + 2 * ar_padding + luma_subblock_size_y * 2 + bottom_pad; |
1119 | 10.6k | int luma_block_size_x = left_pad + 2 * ar_padding + luma_subblock_size_x * 2 + |
1120 | 10.6k | 2 * ar_padding + right_pad; |
1121 | | |
1122 | 10.6k | int chroma_block_size_y = top_pad + (2 >> chroma_subsamp_y) * ar_padding + |
1123 | 10.6k | chroma_subblock_size_y * 2 + bottom_pad; |
1124 | 10.6k | int chroma_block_size_x = left_pad + (2 >> chroma_subsamp_x) * ar_padding + |
1125 | 10.6k | chroma_subblock_size_x * 2 + |
1126 | 10.6k | (2 >> chroma_subsamp_x) * ar_padding + right_pad; |
1127 | | |
1128 | 10.6k | int luma_grain_stride = luma_block_size_x; |
1129 | 10.6k | int chroma_grain_stride = chroma_block_size_x; |
1130 | | |
1131 | 10.6k | int overlap = params->overlap_flag; |
1132 | 10.6k | int bit_depth = params->bit_depth; |
1133 | | |
1134 | 10.6k | const int grain_center = 128 << (bit_depth - 8); |
1135 | 10.6k | grain_min = 0 - grain_center; |
1136 | 10.6k | grain_max = grain_center - 1; |
1137 | | |
1138 | 10.6k | if (!init_arrays(params, luma_stride, chroma_stride, &pred_pos_luma, |
1139 | 10.6k | &pred_pos_chroma, &luma_grain_block, &cb_grain_block, |
1140 | 10.6k | &cr_grain_block, &y_line_buf, &cb_line_buf, &cr_line_buf, |
1141 | 10.6k | &y_col_buf, &cb_col_buf, &cr_col_buf, |
1142 | 10.6k | luma_block_size_y * luma_block_size_x, |
1143 | 10.6k | chroma_block_size_y * chroma_block_size_x, chroma_subsamp_y, |
1144 | 10.6k | chroma_subsamp_x)) |
1145 | 0 | return -1; |
1146 | | |
1147 | 10.6k | generate_luma_grain_block(params, pred_pos_luma, luma_grain_block, |
1148 | 10.6k | luma_block_size_y, luma_block_size_x, |
1149 | 10.6k | luma_grain_stride, left_pad, top_pad, right_pad, |
1150 | 10.6k | bottom_pad); |
1151 | | |
1152 | 10.6k | if (!generate_chroma_grain_blocks( |
1153 | 10.6k | params, pred_pos_chroma, luma_grain_block, cb_grain_block, |
1154 | 10.6k | cr_grain_block, luma_grain_stride, chroma_block_size_y, |
1155 | 10.6k | chroma_block_size_x, chroma_grain_stride, left_pad, top_pad, |
1156 | 10.6k | right_pad, bottom_pad, chroma_subsamp_y, chroma_subsamp_x)) |
1157 | 0 | return -1; |
1158 | | |
1159 | 10.6k | init_scaling_function(params->scaling_points_y, params->num_y_points, |
1160 | 10.6k | scaling_lut_y); |
1161 | | |
1162 | 10.6k | if (params->chroma_scaling_from_luma) { |
1163 | 5.57k | memcpy(scaling_lut_cb, scaling_lut_y, sizeof(*scaling_lut_y) * 256); |
1164 | 5.57k | memcpy(scaling_lut_cr, scaling_lut_y, sizeof(*scaling_lut_y) * 256); |
1165 | 5.57k | } else { |
1166 | 5.11k | init_scaling_function(params->scaling_points_cb, params->num_cb_points, |
1167 | 5.11k | scaling_lut_cb); |
1168 | 5.11k | init_scaling_function(params->scaling_points_cr, params->num_cr_points, |
1169 | 5.11k | scaling_lut_cr); |
1170 | 5.11k | } |
1171 | 26.5k | for (int y = 0; y < height / 2; y += (luma_subblock_size_y >> 1)) { |
1172 | 15.8k | init_random_generator(y * 2, params->random_seed); |
1173 | | |
1174 | 69.8k | for (int x = 0; x < width / 2; x += (luma_subblock_size_x >> 1)) { |
1175 | 53.9k | int offset_y = get_random_number(8); |
1176 | 53.9k | int offset_x = (offset_y >> 4) & 15; |
1177 | 53.9k | offset_y &= 15; |
1178 | | |
1179 | 53.9k | int luma_offset_y = left_pad + 2 * ar_padding + (offset_y << 1); |
1180 | 53.9k | int luma_offset_x = top_pad + 2 * ar_padding + (offset_x << 1); |
1181 | | |
1182 | 53.9k | int chroma_offset_y = top_pad + (2 >> chroma_subsamp_y) * ar_padding + |
1183 | 53.9k | offset_y * (2 >> chroma_subsamp_y); |
1184 | 53.9k | int chroma_offset_x = left_pad + (2 >> chroma_subsamp_x) * ar_padding + |
1185 | 53.9k | offset_x * (2 >> chroma_subsamp_x); |
1186 | | |
1187 | 53.9k | if (overlap && x) { |
1188 | 30.2k | ver_boundary_overlap( |
1189 | 30.2k | y_col_buf, 2, |
1190 | 30.2k | luma_grain_block + luma_offset_y * luma_grain_stride + |
1191 | 30.2k | luma_offset_x, |
1192 | 30.2k | luma_grain_stride, y_col_buf, 2, 2, |
1193 | 30.2k | AOMMIN(luma_subblock_size_y + 2, height - (y << 1))); |
1194 | | |
1195 | 30.2k | ver_boundary_overlap( |
1196 | 30.2k | cb_col_buf, 2 >> chroma_subsamp_x, |
1197 | 30.2k | cb_grain_block + chroma_offset_y * chroma_grain_stride + |
1198 | 30.2k | chroma_offset_x, |
1199 | 30.2k | chroma_grain_stride, cb_col_buf, 2 >> chroma_subsamp_x, |
1200 | 30.2k | 2 >> chroma_subsamp_x, |
1201 | 30.2k | AOMMIN(chroma_subblock_size_y + (2 >> chroma_subsamp_y), |
1202 | 30.2k | (height - (y << 1)) >> chroma_subsamp_y)); |
1203 | | |
1204 | 30.2k | ver_boundary_overlap( |
1205 | 30.2k | cr_col_buf, 2 >> chroma_subsamp_x, |
1206 | 30.2k | cr_grain_block + chroma_offset_y * chroma_grain_stride + |
1207 | 30.2k | chroma_offset_x, |
1208 | 30.2k | chroma_grain_stride, cr_col_buf, 2 >> chroma_subsamp_x, |
1209 | 30.2k | 2 >> chroma_subsamp_x, |
1210 | 30.2k | AOMMIN(chroma_subblock_size_y + (2 >> chroma_subsamp_y), |
1211 | 30.2k | (height - (y << 1)) >> chroma_subsamp_y)); |
1212 | | |
1213 | 30.2k | int i = y ? 1 : 0; |
1214 | | |
1215 | 30.2k | if (use_high_bit_depth) { |
1216 | 22.8k | add_noise_to_block_hbd( |
1217 | 22.8k | params, |
1218 | 22.8k | (uint16_t *)luma + ((y + i) << 1) * luma_stride + (x << 1), |
1219 | 22.8k | (uint16_t *)cb + |
1220 | 22.8k | ((y + i) << (1 - chroma_subsamp_y)) * chroma_stride + |
1221 | 22.8k | (x << (1 - chroma_subsamp_x)), |
1222 | 22.8k | (uint16_t *)cr + |
1223 | 22.8k | ((y + i) << (1 - chroma_subsamp_y)) * chroma_stride + |
1224 | 22.8k | (x << (1 - chroma_subsamp_x)), |
1225 | 22.8k | luma_stride, chroma_stride, y_col_buf + i * 4, |
1226 | 22.8k | cb_col_buf + i * (2 - chroma_subsamp_y) * (2 - chroma_subsamp_x), |
1227 | 22.8k | cr_col_buf + i * (2 - chroma_subsamp_y) * (2 - chroma_subsamp_x), |
1228 | 22.8k | 2, (2 - chroma_subsamp_x), |
1229 | 22.8k | AOMMIN(luma_subblock_size_y >> 1, height / 2 - y) - i, 1, |
1230 | 22.8k | bit_depth, chroma_subsamp_y, chroma_subsamp_x, mc_identity); |
1231 | 22.8k | } else { |
1232 | 7.33k | add_noise_to_block( |
1233 | 7.33k | params, luma + ((y + i) << 1) * luma_stride + (x << 1), |
1234 | 7.33k | cb + ((y + i) << (1 - chroma_subsamp_y)) * chroma_stride + |
1235 | 7.33k | (x << (1 - chroma_subsamp_x)), |
1236 | 7.33k | cr + ((y + i) << (1 - chroma_subsamp_y)) * chroma_stride + |
1237 | 7.33k | (x << (1 - chroma_subsamp_x)), |
1238 | 7.33k | luma_stride, chroma_stride, y_col_buf + i * 4, |
1239 | 7.33k | cb_col_buf + i * (2 - chroma_subsamp_y) * (2 - chroma_subsamp_x), |
1240 | 7.33k | cr_col_buf + i * (2 - chroma_subsamp_y) * (2 - chroma_subsamp_x), |
1241 | 7.33k | 2, (2 - chroma_subsamp_x), |
1242 | 7.33k | AOMMIN(luma_subblock_size_y >> 1, height / 2 - y) - i, 1, |
1243 | 7.33k | bit_depth, chroma_subsamp_y, chroma_subsamp_x, mc_identity); |
1244 | 7.33k | } |
1245 | 30.2k | } |
1246 | | |
1247 | 53.9k | if (overlap && y) { |
1248 | 20.9k | if (x) { |
1249 | 16.1k | hor_boundary_overlap(y_line_buf + (x << 1), luma_stride, y_col_buf, 2, |
1250 | 16.1k | y_line_buf + (x << 1), luma_stride, 2, 2); |
1251 | | |
1252 | 16.1k | hor_boundary_overlap(cb_line_buf + x * (2 >> chroma_subsamp_x), |
1253 | 16.1k | chroma_stride, cb_col_buf, 2 >> chroma_subsamp_x, |
1254 | 16.1k | cb_line_buf + x * (2 >> chroma_subsamp_x), |
1255 | 16.1k | chroma_stride, 2 >> chroma_subsamp_x, |
1256 | 16.1k | 2 >> chroma_subsamp_y); |
1257 | | |
1258 | 16.1k | hor_boundary_overlap(cr_line_buf + x * (2 >> chroma_subsamp_x), |
1259 | 16.1k | chroma_stride, cr_col_buf, 2 >> chroma_subsamp_x, |
1260 | 16.1k | cr_line_buf + x * (2 >> chroma_subsamp_x), |
1261 | 16.1k | chroma_stride, 2 >> chroma_subsamp_x, |
1262 | 16.1k | 2 >> chroma_subsamp_y); |
1263 | 16.1k | } |
1264 | | |
1265 | 20.9k | hor_boundary_overlap( |
1266 | 20.9k | y_line_buf + ((x ? x + 1 : 0) << 1), luma_stride, |
1267 | 20.9k | luma_grain_block + luma_offset_y * luma_grain_stride + |
1268 | 20.9k | luma_offset_x + (x ? 2 : 0), |
1269 | 20.9k | luma_grain_stride, y_line_buf + ((x ? x + 1 : 0) << 1), luma_stride, |
1270 | 20.9k | AOMMIN(luma_subblock_size_x - ((x ? 1 : 0) << 1), |
1271 | 20.9k | width - ((x ? x + 1 : 0) << 1)), |
1272 | 20.9k | 2); |
1273 | | |
1274 | 20.9k | hor_boundary_overlap( |
1275 | 20.9k | cb_line_buf + ((x ? x + 1 : 0) << (1 - chroma_subsamp_x)), |
1276 | 20.9k | chroma_stride, |
1277 | 20.9k | cb_grain_block + chroma_offset_y * chroma_grain_stride + |
1278 | 20.9k | chroma_offset_x + ((x ? 1 : 0) << (1 - chroma_subsamp_x)), |
1279 | 20.9k | chroma_grain_stride, |
1280 | 20.9k | cb_line_buf + ((x ? x + 1 : 0) << (1 - chroma_subsamp_x)), |
1281 | 20.9k | chroma_stride, |
1282 | 20.9k | AOMMIN(chroma_subblock_size_x - |
1283 | 20.9k | ((x ? 1 : 0) << (1 - chroma_subsamp_x)), |
1284 | 20.9k | (width - ((x ? x + 1 : 0) << 1)) >> chroma_subsamp_x), |
1285 | 20.9k | 2 >> chroma_subsamp_y); |
1286 | | |
1287 | 20.9k | hor_boundary_overlap( |
1288 | 20.9k | cr_line_buf + ((x ? x + 1 : 0) << (1 - chroma_subsamp_x)), |
1289 | 20.9k | chroma_stride, |
1290 | 20.9k | cr_grain_block + chroma_offset_y * chroma_grain_stride + |
1291 | 20.9k | chroma_offset_x + ((x ? 1 : 0) << (1 - chroma_subsamp_x)), |
1292 | 20.9k | chroma_grain_stride, |
1293 | 20.9k | cr_line_buf + ((x ? x + 1 : 0) << (1 - chroma_subsamp_x)), |
1294 | 20.9k | chroma_stride, |
1295 | 20.9k | AOMMIN(chroma_subblock_size_x - |
1296 | 20.9k | ((x ? 1 : 0) << (1 - chroma_subsamp_x)), |
1297 | 20.9k | (width - ((x ? x + 1 : 0) << 1)) >> chroma_subsamp_x), |
1298 | 20.9k | 2 >> chroma_subsamp_y); |
1299 | | |
1300 | 20.9k | if (use_high_bit_depth) { |
1301 | 16.0k | add_noise_to_block_hbd( |
1302 | 16.0k | params, (uint16_t *)luma + (y << 1) * luma_stride + (x << 1), |
1303 | 16.0k | (uint16_t *)cb + (y << (1 - chroma_subsamp_y)) * chroma_stride + |
1304 | 16.0k | (x << ((1 - chroma_subsamp_x))), |
1305 | 16.0k | (uint16_t *)cr + (y << (1 - chroma_subsamp_y)) * chroma_stride + |
1306 | 16.0k | (x << ((1 - chroma_subsamp_x))), |
1307 | 16.0k | luma_stride, chroma_stride, y_line_buf + (x << 1), |
1308 | 16.0k | cb_line_buf + (x << (1 - chroma_subsamp_x)), |
1309 | 16.0k | cr_line_buf + (x << (1 - chroma_subsamp_x)), luma_stride, |
1310 | 16.0k | chroma_stride, 1, |
1311 | 16.0k | AOMMIN(luma_subblock_size_x >> 1, width / 2 - x), bit_depth, |
1312 | 16.0k | chroma_subsamp_y, chroma_subsamp_x, mc_identity); |
1313 | 16.0k | } else { |
1314 | 4.85k | add_noise_to_block( |
1315 | 4.85k | params, luma + (y << 1) * luma_stride + (x << 1), |
1316 | 4.85k | cb + (y << (1 - chroma_subsamp_y)) * chroma_stride + |
1317 | 4.85k | (x << ((1 - chroma_subsamp_x))), |
1318 | 4.85k | cr + (y << (1 - chroma_subsamp_y)) * chroma_stride + |
1319 | 4.85k | (x << ((1 - chroma_subsamp_x))), |
1320 | 4.85k | luma_stride, chroma_stride, y_line_buf + (x << 1), |
1321 | 4.85k | cb_line_buf + (x << (1 - chroma_subsamp_x)), |
1322 | 4.85k | cr_line_buf + (x << (1 - chroma_subsamp_x)), luma_stride, |
1323 | 4.85k | chroma_stride, 1, |
1324 | 4.85k | AOMMIN(luma_subblock_size_x >> 1, width / 2 - x), bit_depth, |
1325 | 4.85k | chroma_subsamp_y, chroma_subsamp_x, mc_identity); |
1326 | 4.85k | } |
1327 | 20.9k | } |
1328 | | |
1329 | 53.9k | int i = overlap && y ? 1 : 0; |
1330 | 53.9k | int j = overlap && x ? 1 : 0; |
1331 | | |
1332 | 53.9k | if (use_high_bit_depth) { |
1333 | 37.1k | add_noise_to_block_hbd( |
1334 | 37.1k | params, |
1335 | 37.1k | (uint16_t *)luma + ((y + i) << 1) * luma_stride + ((x + j) << 1), |
1336 | 37.1k | (uint16_t *)cb + |
1337 | 37.1k | ((y + i) << (1 - chroma_subsamp_y)) * chroma_stride + |
1338 | 37.1k | ((x + j) << (1 - chroma_subsamp_x)), |
1339 | 37.1k | (uint16_t *)cr + |
1340 | 37.1k | ((y + i) << (1 - chroma_subsamp_y)) * chroma_stride + |
1341 | 37.1k | ((x + j) << (1 - chroma_subsamp_x)), |
1342 | 37.1k | luma_stride, chroma_stride, |
1343 | 37.1k | luma_grain_block + (luma_offset_y + (i << 1)) * luma_grain_stride + |
1344 | 37.1k | luma_offset_x + (j << 1), |
1345 | 37.1k | cb_grain_block + |
1346 | 37.1k | (chroma_offset_y + (i << (1 - chroma_subsamp_y))) * |
1347 | 37.1k | chroma_grain_stride + |
1348 | 37.1k | chroma_offset_x + (j << (1 - chroma_subsamp_x)), |
1349 | 37.1k | cr_grain_block + |
1350 | 37.1k | (chroma_offset_y + (i << (1 - chroma_subsamp_y))) * |
1351 | 37.1k | chroma_grain_stride + |
1352 | 37.1k | chroma_offset_x + (j << (1 - chroma_subsamp_x)), |
1353 | 37.1k | luma_grain_stride, chroma_grain_stride, |
1354 | 37.1k | AOMMIN(luma_subblock_size_y >> 1, height / 2 - y) - i, |
1355 | 37.1k | AOMMIN(luma_subblock_size_x >> 1, width / 2 - x) - j, bit_depth, |
1356 | 37.1k | chroma_subsamp_y, chroma_subsamp_x, mc_identity); |
1357 | 37.1k | } else { |
1358 | 16.8k | add_noise_to_block( |
1359 | 16.8k | params, luma + ((y + i) << 1) * luma_stride + ((x + j) << 1), |
1360 | 16.8k | cb + ((y + i) << (1 - chroma_subsamp_y)) * chroma_stride + |
1361 | 16.8k | ((x + j) << (1 - chroma_subsamp_x)), |
1362 | 16.8k | cr + ((y + i) << (1 - chroma_subsamp_y)) * chroma_stride + |
1363 | 16.8k | ((x + j) << (1 - chroma_subsamp_x)), |
1364 | 16.8k | luma_stride, chroma_stride, |
1365 | 16.8k | luma_grain_block + (luma_offset_y + (i << 1)) * luma_grain_stride + |
1366 | 16.8k | luma_offset_x + (j << 1), |
1367 | 16.8k | cb_grain_block + |
1368 | 16.8k | (chroma_offset_y + (i << (1 - chroma_subsamp_y))) * |
1369 | 16.8k | chroma_grain_stride + |
1370 | 16.8k | chroma_offset_x + (j << (1 - chroma_subsamp_x)), |
1371 | 16.8k | cr_grain_block + |
1372 | 16.8k | (chroma_offset_y + (i << (1 - chroma_subsamp_y))) * |
1373 | 16.8k | chroma_grain_stride + |
1374 | 16.8k | chroma_offset_x + (j << (1 - chroma_subsamp_x)), |
1375 | 16.8k | luma_grain_stride, chroma_grain_stride, |
1376 | 16.8k | AOMMIN(luma_subblock_size_y >> 1, height / 2 - y) - i, |
1377 | 16.8k | AOMMIN(luma_subblock_size_x >> 1, width / 2 - x) - j, bit_depth, |
1378 | 16.8k | chroma_subsamp_y, chroma_subsamp_x, mc_identity); |
1379 | 16.8k | } |
1380 | | |
1381 | 53.9k | if (overlap) { |
1382 | 41.3k | if (x) { |
1383 | | // Copy overlapped column bufer to line buffer |
1384 | 30.2k | copy_area(y_col_buf + (luma_subblock_size_y << 1), 2, |
1385 | 30.2k | y_line_buf + (x << 1), luma_stride, 2, 2); |
1386 | | |
1387 | 30.2k | copy_area( |
1388 | 30.2k | cb_col_buf + (chroma_subblock_size_y << (1 - chroma_subsamp_x)), |
1389 | 30.2k | 2 >> chroma_subsamp_x, |
1390 | 30.2k | cb_line_buf + (x << (1 - chroma_subsamp_x)), chroma_stride, |
1391 | 30.2k | 2 >> chroma_subsamp_x, 2 >> chroma_subsamp_y); |
1392 | | |
1393 | 30.2k | copy_area( |
1394 | 30.2k | cr_col_buf + (chroma_subblock_size_y << (1 - chroma_subsamp_x)), |
1395 | 30.2k | 2 >> chroma_subsamp_x, |
1396 | 30.2k | cr_line_buf + (x << (1 - chroma_subsamp_x)), chroma_stride, |
1397 | 30.2k | 2 >> chroma_subsamp_x, 2 >> chroma_subsamp_y); |
1398 | 30.2k | } |
1399 | | |
1400 | | // Copy grain to the line buffer for overlap with a bottom block |
1401 | 41.3k | copy_area( |
1402 | 41.3k | luma_grain_block + |
1403 | 41.3k | (luma_offset_y + luma_subblock_size_y) * luma_grain_stride + |
1404 | 41.3k | luma_offset_x + ((x ? 2 : 0)), |
1405 | 41.3k | luma_grain_stride, y_line_buf + ((x ? x + 1 : 0) << 1), luma_stride, |
1406 | 41.3k | AOMMIN(luma_subblock_size_x, width - (x << 1)) - (x ? 2 : 0), 2); |
1407 | | |
1408 | 41.3k | copy_area(cb_grain_block + |
1409 | 41.3k | (chroma_offset_y + chroma_subblock_size_y) * |
1410 | 41.3k | chroma_grain_stride + |
1411 | 41.3k | chroma_offset_x + (x ? 2 >> chroma_subsamp_x : 0), |
1412 | 41.3k | chroma_grain_stride, |
1413 | 41.3k | cb_line_buf + ((x ? x + 1 : 0) << (1 - chroma_subsamp_x)), |
1414 | 41.3k | chroma_stride, |
1415 | 41.3k | AOMMIN(chroma_subblock_size_x, |
1416 | 41.3k | ((width - (x << 1)) >> chroma_subsamp_x)) - |
1417 | 41.3k | (x ? 2 >> chroma_subsamp_x : 0), |
1418 | 41.3k | 2 >> chroma_subsamp_y); |
1419 | | |
1420 | 41.3k | copy_area(cr_grain_block + |
1421 | 41.3k | (chroma_offset_y + chroma_subblock_size_y) * |
1422 | 41.3k | chroma_grain_stride + |
1423 | 41.3k | chroma_offset_x + (x ? 2 >> chroma_subsamp_x : 0), |
1424 | 41.3k | chroma_grain_stride, |
1425 | 41.3k | cr_line_buf + ((x ? x + 1 : 0) << (1 - chroma_subsamp_x)), |
1426 | 41.3k | chroma_stride, |
1427 | 41.3k | AOMMIN(chroma_subblock_size_x, |
1428 | 41.3k | ((width - (x << 1)) >> chroma_subsamp_x)) - |
1429 | 41.3k | (x ? 2 >> chroma_subsamp_x : 0), |
1430 | 41.3k | 2 >> chroma_subsamp_y); |
1431 | | |
1432 | | // Copy grain to the column buffer for overlap with the next block to |
1433 | | // the right |
1434 | | |
1435 | 41.3k | copy_area(luma_grain_block + luma_offset_y * luma_grain_stride + |
1436 | 41.3k | luma_offset_x + luma_subblock_size_x, |
1437 | 41.3k | luma_grain_stride, y_col_buf, 2, 2, |
1438 | 41.3k | AOMMIN(luma_subblock_size_y + 2, height - (y << 1))); |
1439 | | |
1440 | 41.3k | copy_area(cb_grain_block + chroma_offset_y * chroma_grain_stride + |
1441 | 41.3k | chroma_offset_x + chroma_subblock_size_x, |
1442 | 41.3k | chroma_grain_stride, cb_col_buf, 2 >> chroma_subsamp_x, |
1443 | 41.3k | 2 >> chroma_subsamp_x, |
1444 | 41.3k | AOMMIN(chroma_subblock_size_y + (2 >> chroma_subsamp_y), |
1445 | 41.3k | (height - (y << 1)) >> chroma_subsamp_y)); |
1446 | | |
1447 | 41.3k | copy_area(cr_grain_block + chroma_offset_y * chroma_grain_stride + |
1448 | 41.3k | chroma_offset_x + chroma_subblock_size_x, |
1449 | 41.3k | chroma_grain_stride, cr_col_buf, 2 >> chroma_subsamp_x, |
1450 | 41.3k | 2 >> chroma_subsamp_x, |
1451 | 41.3k | AOMMIN(chroma_subblock_size_y + (2 >> chroma_subsamp_y), |
1452 | 41.3k | (height - (y << 1)) >> chroma_subsamp_y)); |
1453 | 41.3k | } |
1454 | 53.9k | } |
1455 | 15.8k | } |
1456 | | |
1457 | 10.6k | dealloc_arrays(params, &pred_pos_luma, &pred_pos_chroma, &luma_grain_block, |
1458 | 10.6k | &cb_grain_block, &cr_grain_block, &y_line_buf, &cb_line_buf, |
1459 | 10.6k | &cr_line_buf, &y_col_buf, &cb_col_buf, &cr_col_buf); |
1460 | 10.6k | return 0; |
1461 | 10.6k | } |