1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
222
223
224
225
226
227
228
229
230
231
232
233
234
235
236
237
238
239
240
241
242
243
244
245
246
247
248
249
250
251
252
253
254
|
/*
* Copyright (c) 2010 The WebM project authors. All Rights Reserved.
*
* Use of this source code is governed by a BSD-style license
* that can be found in the LICENSE file in the root of the source
* tree. An additional intellectual property rights grant can be found
* in the file PATENTS. All contributing project authors may
* be found in the AUTHORS file in the root of the source tree.
*/
#include "./vp9_rtcd.h"
#include "vp9/encoder/vp9_ssim.h"
void vp9_ssim_parms_16x16_c(uint8_t *s, int sp, uint8_t *r,
int rp, unsigned long *sum_s, unsigned long *sum_r,
unsigned long *sum_sq_s, unsigned long *sum_sq_r,
unsigned long *sum_sxr) {
int i, j;
for (i = 0; i < 16; i++, s += sp, r += rp) {
for (j = 0; j < 16; j++) {
*sum_s += s[j];
*sum_r += r[j];
*sum_sq_s += s[j] * s[j];
*sum_sq_r += r[j] * r[j];
*sum_sxr += s[j] * r[j];
}
}
}
void vp9_ssim_parms_8x8_c(uint8_t *s, int sp, uint8_t *r, int rp,
unsigned long *sum_s, unsigned long *sum_r,
unsigned long *sum_sq_s, unsigned long *sum_sq_r,
unsigned long *sum_sxr) {
int i, j;
for (i = 0; i < 8; i++, s += sp, r += rp) {
for (j = 0; j < 8; j++) {
*sum_s += s[j];
*sum_r += r[j];
*sum_sq_s += s[j] * s[j];
*sum_sq_r += r[j] * r[j];
*sum_sxr += s[j] * r[j];
}
}
}
#if CONFIG_VP9_HIGHBITDEPTH
void vp9_highbd_ssim_parms_8x8_c(uint16_t *s, int sp, uint16_t *r, int rp,
uint32_t *sum_s, uint32_t *sum_r,
uint32_t *sum_sq_s, uint32_t *sum_sq_r,
uint32_t *sum_sxr) {
int i, j;
for (i = 0; i < 8; i++, s += sp, r += rp) {
for (j = 0; j < 8; j++) {
*sum_s += s[j];
*sum_r += r[j];
*sum_sq_s += s[j] * s[j];
*sum_sq_r += r[j] * r[j];
*sum_sxr += s[j] * r[j];
}
}
}
#endif // CONFIG_VP9_HIGHBITDEPTH
static const int64_t cc1 = 26634; // (64^2*(.01*255)^2
static const int64_t cc2 = 239708; // (64^2*(.03*255)^2
static double similarity(unsigned long sum_s, unsigned long sum_r,
unsigned long sum_sq_s, unsigned long sum_sq_r,
unsigned long sum_sxr, int count) {
int64_t ssim_n, ssim_d;
int64_t c1, c2;
// scale the constants by number of pixels
c1 = (cc1 * count * count) >> 12;
c2 = (cc2 * count * count) >> 12;
ssim_n = (2 * sum_s * sum_r + c1) * ((int64_t) 2 * count * sum_sxr -
(int64_t) 2 * sum_s * sum_r + c2);
ssim_d = (sum_s * sum_s + sum_r * sum_r + c1) *
((int64_t)count * sum_sq_s - (int64_t)sum_s * sum_s +
(int64_t)count * sum_sq_r - (int64_t) sum_r * sum_r + c2);
return ssim_n * 1.0 / ssim_d;
}
static double ssim_8x8(uint8_t *s, int sp, uint8_t *r, int rp) {
unsigned long sum_s = 0, sum_r = 0, sum_sq_s = 0, sum_sq_r = 0, sum_sxr = 0;
vp9_ssim_parms_8x8(s, sp, r, rp, &sum_s, &sum_r, &sum_sq_s, &sum_sq_r,
&sum_sxr);
return similarity(sum_s, sum_r, sum_sq_s, sum_sq_r, sum_sxr, 64);
}
#if CONFIG_VP9_HIGHBITDEPTH
static double highbd_ssim_8x8(uint16_t *s, int sp, uint16_t *r, int rp,
unsigned int bd) {
uint32_t sum_s = 0, sum_r = 0, sum_sq_s = 0, sum_sq_r = 0, sum_sxr = 0;
const int oshift = bd - 8;
vp9_highbd_ssim_parms_8x8(s, sp, r, rp, &sum_s, &sum_r, &sum_sq_s, &sum_sq_r,
&sum_sxr);
return similarity(sum_s >> oshift,
sum_r >> oshift,
sum_sq_s >> (2 * oshift),
sum_sq_r >> (2 * oshift),
sum_sxr >> (2 * oshift),
64);
}
#endif // CONFIG_VP9_HIGHBITDEPTH
// We are using a 8x8 moving window with starting location of each 8x8 window
// on the 4x4 pixel grid. Such arrangement allows the windows to overlap
// block boundaries to penalize blocking artifacts.
double vp9_ssim2(uint8_t *img1, uint8_t *img2, int stride_img1,
int stride_img2, int width, int height) {
int i, j;
int samples = 0;
double ssim_total = 0;
// sample point start with each 4x4 location
for (i = 0; i <= height - 8;
i += 4, img1 += stride_img1 * 4, img2 += stride_img2 * 4) {
for (j = 0; j <= width - 8; j += 4) {
double v = ssim_8x8(img1 + j, stride_img1, img2 + j, stride_img2);
ssim_total += v;
samples++;
}
}
ssim_total /= samples;
return ssim_total;
}
#if CONFIG_VP9_HIGHBITDEPTH
double vp9_highbd_ssim2(uint8_t *img1, uint8_t *img2, int stride_img1,
int stride_img2, int width, int height,
unsigned int bd) {
int i, j;
int samples = 0;
double ssim_total = 0;
// sample point start with each 4x4 location
for (i = 0; i <= height - 8;
i += 4, img1 += stride_img1 * 4, img2 += stride_img2 * 4) {
for (j = 0; j <= width - 8; j += 4) {
double v = highbd_ssim_8x8(CONVERT_TO_SHORTPTR(img1 + j), stride_img1,
CONVERT_TO_SHORTPTR(img2 + j), stride_img2,
bd);
ssim_total += v;
samples++;
}
}
ssim_total /= samples;
return ssim_total;
}
#endif // CONFIG_VP9_HIGHBITDEPTH
double vp9_calc_ssim(YV12_BUFFER_CONFIG *source, YV12_BUFFER_CONFIG *dest,
double *weight) {
double a, b, c;
double ssimv;
a = vp9_ssim2(source->y_buffer, dest->y_buffer,
source->y_stride, dest->y_stride,
source->y_crop_width, source->y_crop_height);
b = vp9_ssim2(source->u_buffer, dest->u_buffer,
source->uv_stride, dest->uv_stride,
source->uv_crop_width, source->uv_crop_height);
c = vp9_ssim2(source->v_buffer, dest->v_buffer,
source->uv_stride, dest->uv_stride,
source->uv_crop_width, source->uv_crop_height);
ssimv = a * .8 + .1 * (b + c);
*weight = 1;
return ssimv;
}
double vp9_calc_ssimg(YV12_BUFFER_CONFIG *source, YV12_BUFFER_CONFIG *dest,
double *ssim_y, double *ssim_u, double *ssim_v) {
double ssim_all = 0;
double a, b, c;
a = vp9_ssim2(source->y_buffer, dest->y_buffer,
source->y_stride, dest->y_stride,
source->y_crop_width, source->y_crop_height);
b = vp9_ssim2(source->u_buffer, dest->u_buffer,
source->uv_stride, dest->uv_stride,
source->uv_crop_width, source->uv_crop_height);
c = vp9_ssim2(source->v_buffer, dest->v_buffer,
source->uv_stride, dest->uv_stride,
source->uv_crop_width, source->uv_crop_height);
*ssim_y = a;
*ssim_u = b;
*ssim_v = c;
ssim_all = (a * 4 + b + c) / 6;
return ssim_all;
}
#if CONFIG_VP9_HIGHBITDEPTH
double vp9_highbd_calc_ssim(YV12_BUFFER_CONFIG *source,
YV12_BUFFER_CONFIG *dest,
double *weight, unsigned int bd) {
double a, b, c;
double ssimv;
a = vp9_highbd_ssim2(source->y_buffer, dest->y_buffer,
source->y_stride, dest->y_stride,
source->y_crop_width, source->y_crop_height, bd);
b = vp9_highbd_ssim2(source->u_buffer, dest->u_buffer,
source->uv_stride, dest->uv_stride,
source->uv_crop_width, source->uv_crop_height, bd);
c = vp9_highbd_ssim2(source->v_buffer, dest->v_buffer,
source->uv_stride, dest->uv_stride,
source->uv_crop_width, source->uv_crop_height, bd);
ssimv = a * .8 + .1 * (b + c);
*weight = 1;
return ssimv;
}
double vp9_highbd_calc_ssimg(YV12_BUFFER_CONFIG *source,
YV12_BUFFER_CONFIG *dest, double *ssim_y,
double *ssim_u, double *ssim_v, unsigned int bd) {
double ssim_all = 0;
double a, b, c;
a = vp9_highbd_ssim2(source->y_buffer, dest->y_buffer,
source->y_stride, dest->y_stride,
source->y_crop_width, source->y_crop_height, bd);
b = vp9_highbd_ssim2(source->u_buffer, dest->u_buffer,
source->uv_stride, dest->uv_stride,
source->uv_crop_width, source->uv_crop_height, bd);
c = vp9_highbd_ssim2(source->v_buffer, dest->v_buffer,
source->uv_stride, dest->uv_stride,
source->uv_crop_width, source->uv_crop_height, bd);
*ssim_y = a;
*ssim_u = b;
*ssim_v = c;
ssim_all = (a * 4 + b + c) / 6;
return ssim_all;
}
#endif // CONFIG_VP9_HIGHBITDEPTH
|
