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
255
256
257
258
259
260
261
262
263
264
265
266
267
268
269
270
271
272
273
274
275
276
277
278
279
280
281
282
283
284
285
286
287
288
289
290
291
292
293
294
295
296
297
298
299
300
301
302
303
304
305
306
307
308
309
310
311
312
313
314
315
316
317
318
319
320
321
322
323
324
325
326
327
328
329
330
331
332
333
334
335
336
337
|
/*
* 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 "./vp8_rtcd.h"
#include "filter.h"
#include "variance.h"
/* This is a bad idea.
* ctz = count trailing zeros */
static int ctz(int a) {
int b = 0;
while (a != 1) {
a >>= 1;
b++;
}
return b;
}
static unsigned int variance(
const unsigned char *src_ptr,
int source_stride,
const unsigned char *ref_ptr,
int recon_stride,
int w,
int h,
unsigned int *sse)
{
int i, j;
int diff, sum;
sum = 0;
*sse = 0;
for (i = 0; i < h; i++)
{
for (j = 0; j < w; j++)
{
diff = src_ptr[j] - ref_ptr[j];
sum += diff;
*sse += diff * diff;
}
src_ptr += source_stride;
ref_ptr += recon_stride;
}
return (*sse - (((unsigned int)sum * sum) >> (int)((ctz(w) + ctz(h)))));
}
/****************************************************************************
*
* ROUTINE : filter_block2d_bil_first_pass
*
* INPUTS : UINT8 *src_ptr : Pointer to source block.
* UINT32 src_pixels_per_line : Stride of input block.
* UINT32 pixel_step : Offset between filter input samples (see notes).
* UINT32 output_height : Input block height.
* UINT32 output_width : Input block width.
* INT32 *vp8_filter : Array of 2 bi-linear filter taps.
*
* OUTPUTS : INT32 *output_ptr : Pointer to filtered block.
*
* RETURNS : void
*
* FUNCTION : Applies a 1-D 2-tap bi-linear filter to the source block in
* either horizontal or vertical direction to produce the
* filtered output block. Used to implement first-pass
* of 2-D separable filter.
*
* SPECIAL NOTES : Produces INT32 output to retain precision for next pass.
* Two filter taps should sum to VP8_FILTER_WEIGHT.
* pixel_step defines whether the filter is applied
* horizontally (pixel_step=1) or vertically (pixel_step=stride).
* It defines the offset required to move from one input
* to the next.
*
****************************************************************************/
static void var_filter_block2d_bil_first_pass
(
const unsigned char *src_ptr,
unsigned short *output_ptr,
unsigned int src_pixels_per_line,
int pixel_step,
unsigned int output_height,
unsigned int output_width,
const short *vp8_filter
)
{
unsigned int i, j;
for (i = 0; i < output_height; i++)
{
for (j = 0; j < output_width; j++)
{
/* Apply bilinear filter */
output_ptr[j] = (((int)src_ptr[0] * vp8_filter[0]) +
((int)src_ptr[pixel_step] * vp8_filter[1]) +
(VP8_FILTER_WEIGHT / 2)) >> VP8_FILTER_SHIFT;
src_ptr++;
}
/* Next row... */
src_ptr += src_pixels_per_line - output_width;
output_ptr += output_width;
}
}
/****************************************************************************
*
* ROUTINE : filter_block2d_bil_second_pass
*
* INPUTS : INT32 *src_ptr : Pointer to source block.
* UINT32 src_pixels_per_line : Stride of input block.
* UINT32 pixel_step : Offset between filter input samples (see notes).
* UINT32 output_height : Input block height.
* UINT32 output_width : Input block width.
* INT32 *vp8_filter : Array of 2 bi-linear filter taps.
*
* OUTPUTS : UINT16 *output_ptr : Pointer to filtered block.
*
* RETURNS : void
*
* FUNCTION : Applies a 1-D 2-tap bi-linear filter to the source block in
* either horizontal or vertical direction to produce the
* filtered output block. Used to implement second-pass
* of 2-D separable filter.
*
* SPECIAL NOTES : Requires 32-bit input as produced by filter_block2d_bil_first_pass.
* Two filter taps should sum to VP8_FILTER_WEIGHT.
* pixel_step defines whether the filter is applied
* horizontally (pixel_step=1) or vertically (pixel_step=stride).
* It defines the offset required to move from one input
* to the next.
*
****************************************************************************/
static void var_filter_block2d_bil_second_pass
(
const unsigned short *src_ptr,
unsigned char *output_ptr,
unsigned int src_pixels_per_line,
unsigned int pixel_step,
unsigned int output_height,
unsigned int output_width,
const short *vp8_filter
)
{
unsigned int i, j;
int Temp;
for (i = 0; i < output_height; i++)
{
for (j = 0; j < output_width; j++)
{
/* Apply filter */
Temp = ((int)src_ptr[0] * vp8_filter[0]) +
((int)src_ptr[pixel_step] * vp8_filter[1]) +
(VP8_FILTER_WEIGHT / 2);
output_ptr[j] = (unsigned int)(Temp >> VP8_FILTER_SHIFT);
src_ptr++;
}
/* Next row... */
src_ptr += src_pixels_per_line - output_width;
output_ptr += output_width;
}
}
unsigned int vp8_sub_pixel_variance4x4_c
(
const unsigned char *src_ptr,
int src_pixels_per_line,
int xoffset,
int yoffset,
const unsigned char *dst_ptr,
int dst_pixels_per_line,
unsigned int *sse
)
{
unsigned char temp2[20*16];
const short *HFilter, *VFilter;
unsigned short FData3[5*4]; /* Temp data bufffer used in filtering */
HFilter = vp8_bilinear_filters[xoffset];
VFilter = vp8_bilinear_filters[yoffset];
/* First filter 1d Horizontal */
var_filter_block2d_bil_first_pass(src_ptr, FData3, src_pixels_per_line, 1, 5, 4, HFilter);
/* Now filter Verticaly */
var_filter_block2d_bil_second_pass(FData3, temp2, 4, 4, 4, 4, VFilter);
return variance(temp2, 4, dst_ptr, dst_pixels_per_line, 4, 4, sse);
}
unsigned int vp8_sub_pixel_variance8x8_c
(
const unsigned char *src_ptr,
int src_pixels_per_line,
int xoffset,
int yoffset,
const unsigned char *dst_ptr,
int dst_pixels_per_line,
unsigned int *sse
)
{
unsigned short FData3[9*8]; /* Temp data bufffer used in filtering */
unsigned char temp2[20*16];
const short *HFilter, *VFilter;
HFilter = vp8_bilinear_filters[xoffset];
VFilter = vp8_bilinear_filters[yoffset];
var_filter_block2d_bil_first_pass(src_ptr, FData3, src_pixels_per_line, 1, 9, 8, HFilter);
var_filter_block2d_bil_second_pass(FData3, temp2, 8, 8, 8, 8, VFilter);
return variance(temp2, 8, dst_ptr, dst_pixels_per_line, 8, 8, sse);
}
unsigned int vp8_sub_pixel_variance16x16_c
(
const unsigned char *src_ptr,
int src_pixels_per_line,
int xoffset,
int yoffset,
const unsigned char *dst_ptr,
int dst_pixels_per_line,
unsigned int *sse
)
{
unsigned short FData3[17*16]; /* Temp data bufffer used in filtering */
unsigned char temp2[20*16];
const short *HFilter, *VFilter;
HFilter = vp8_bilinear_filters[xoffset];
VFilter = vp8_bilinear_filters[yoffset];
var_filter_block2d_bil_first_pass(src_ptr, FData3, src_pixels_per_line, 1, 17, 16, HFilter);
var_filter_block2d_bil_second_pass(FData3, temp2, 16, 16, 16, 16, VFilter);
return variance(temp2, 16, dst_ptr, dst_pixels_per_line, 16, 16, sse);
}
unsigned int vp8_variance_halfpixvar16x16_h_c(
const unsigned char *src_ptr,
int source_stride,
const unsigned char *ref_ptr,
int recon_stride,
unsigned int *sse)
{
return vp8_sub_pixel_variance16x16_c(src_ptr, source_stride, 4, 0,
ref_ptr, recon_stride, sse);
}
unsigned int vp8_variance_halfpixvar16x16_v_c(
const unsigned char *src_ptr,
int source_stride,
const unsigned char *ref_ptr,
int recon_stride,
unsigned int *sse)
{
return vp8_sub_pixel_variance16x16_c(src_ptr, source_stride, 0, 4,
ref_ptr, recon_stride, sse);
}
unsigned int vp8_variance_halfpixvar16x16_hv_c(
const unsigned char *src_ptr,
int source_stride,
const unsigned char *ref_ptr,
int recon_stride,
unsigned int *sse)
{
return vp8_sub_pixel_variance16x16_c(src_ptr, source_stride, 4, 4,
ref_ptr, recon_stride, sse);
}
unsigned int vp8_sub_pixel_variance16x8_c
(
const unsigned char *src_ptr,
int src_pixels_per_line,
int xoffset,
int yoffset,
const unsigned char *dst_ptr,
int dst_pixels_per_line,
unsigned int *sse
)
{
unsigned short FData3[16*9]; /* Temp data bufffer used in filtering */
unsigned char temp2[20*16];
const short *HFilter, *VFilter;
HFilter = vp8_bilinear_filters[xoffset];
VFilter = vp8_bilinear_filters[yoffset];
var_filter_block2d_bil_first_pass(src_ptr, FData3, src_pixels_per_line, 1, 9, 16, HFilter);
var_filter_block2d_bil_second_pass(FData3, temp2, 16, 16, 8, 16, VFilter);
return variance(temp2, 16, dst_ptr, dst_pixels_per_line, 16, 8, sse);
}
unsigned int vp8_sub_pixel_variance8x16_c
(
const unsigned char *src_ptr,
int src_pixels_per_line,
int xoffset,
int yoffset,
const unsigned char *dst_ptr,
int dst_pixels_per_line,
unsigned int *sse
)
{
unsigned short FData3[9*16]; /* Temp data bufffer used in filtering */
unsigned char temp2[20*16];
const short *HFilter, *VFilter;
HFilter = vp8_bilinear_filters[xoffset];
VFilter = vp8_bilinear_filters[yoffset];
var_filter_block2d_bil_first_pass(src_ptr, FData3, src_pixels_per_line, 1, 17, 8, HFilter);
var_filter_block2d_bil_second_pass(FData3, temp2, 8, 8, 16, 8, VFilter);
return variance(temp2, 8, dst_ptr, dst_pixels_per_line, 8, 16, sse);
}
|
