pickrst.c 52.6 KB
Newer Older
1
/*
2
 * Copyright (c) 2016, Alliance for Open Media. All rights reserved
3
 *
4
5
6
7
8
9
 * This source code is subject to the terms of the BSD 2 Clause License and
 * the Alliance for Open Media Patent License 1.0. If the BSD 2 Clause License
 * was not distributed with this source code in the LICENSE file, you can
 * obtain it at www.aomedia.org/license/software. If the Alliance for Open
 * Media Patent License 1.0 was not distributed with this source code in the
 * PATENTS file, you can obtain it at www.aomedia.org/license/patent.
10
11
12
 */

#include <assert.h>
13
#include <float.h>
14
15
16
#include <limits.h>
#include <math.h>

Yaowu Xu's avatar
Yaowu Xu committed
17
#include "./aom_scale_rtcd.h"
18

19
#include "aom_dsp/psnr.h"
Yaowu Xu's avatar
Yaowu Xu committed
20
21
#include "aom_dsp/aom_dsp_common.h"
#include "aom_mem/aom_mem.h"
22
#include "aom_ports/mem.h"
23

24
25
#include "av1/common/onyxc_int.h"
#include "av1/common/quant_common.h"
26
#include "av1/common/restoration.h"
27

28
29
30
31
#include "av1/encoder/encoder.h"
#include "av1/encoder/picklpf.h"
#include "av1/encoder/pickrst.h"
#include "av1/encoder/quantize.h"
32

33
34
35
typedef double (*search_restore_type)(const YV12_BUFFER_CONFIG *src,
                                      AV1_COMP *cpi, int filter_level,
                                      int partial_frame, RestorationInfo *info,
36
37
                                      double *best_tile_cost,
                                      YV12_BUFFER_CONFIG *dst_frame);
38

39
const int frame_level_restore_bits[RESTORE_TYPES] = { 2, 2, 3, 3, 2 };
40
41

static int64_t sse_restoration_tile(const YV12_BUFFER_CONFIG *src,
42
43
                                    const YV12_BUFFER_CONFIG *dst,
                                    const AV1_COMMON *cm, int h_start,
44
                                    int width, int v_start, int height,
45
46
                                    int components_pattern) {
  int64_t filt_err = 0;
47
48
#if CONFIG_AOM_HIGHBITDEPTH
  if (cm->use_highbitdepth) {
49
50
51
52
53
    if ((components_pattern >> AOM_PLANE_Y) & 1) {
      filt_err +=
          aom_highbd_get_y_sse_part(src, dst, h_start, width, v_start, height);
    }
    if ((components_pattern >> AOM_PLANE_U) & 1) {
54
55
56
      filt_err += aom_highbd_get_u_sse_part(
          src, dst, h_start >> cm->subsampling_x, width >> cm->subsampling_x,
          v_start >> cm->subsampling_y, height >> cm->subsampling_y);
57
58
    }
    if ((components_pattern >> AOM_PLANE_V) & 1) {
59
60
61
62
63
      filt_err += aom_highbd_get_v_sse_part(
          src, dst, h_start >> cm->subsampling_x, width >> cm->subsampling_x,
          v_start >> cm->subsampling_y, height >> cm->subsampling_y);
    }
    return filt_err;
64
65
  }
#endif  // CONFIG_AOM_HIGHBITDEPTH
66
67
68
69
  if ((components_pattern >> AOM_PLANE_Y) & 1) {
    filt_err += aom_get_y_sse_part(src, dst, h_start, width, v_start, height);
  }
  if ((components_pattern >> AOM_PLANE_U) & 1) {
70
71
72
    filt_err += aom_get_u_sse_part(
        src, dst, h_start >> cm->subsampling_x, width >> cm->subsampling_x,
        v_start >> cm->subsampling_y, height >> cm->subsampling_y);
73
74
  }
  if ((components_pattern >> AOM_PLANE_V) & 1) {
75
76
77
78
    filt_err += aom_get_u_sse_part(
        src, dst, h_start >> cm->subsampling_x, width >> cm->subsampling_x,
        v_start >> cm->subsampling_y, height >> cm->subsampling_y);
  }
79
80
81
  return filt_err;
}

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
static int64_t sse_restoration_frame(const YV12_BUFFER_CONFIG *src,
                                     const YV12_BUFFER_CONFIG *dst,
                                     int components_pattern) {
  int64_t filt_err = 0;
#if CONFIG_AOM_HIGHBITDEPTH
  if (cm->use_highbitdepth) {
    if ((components_pattern >> AOM_PLANE_Y) & 1) {
      filt_err += aom_highbd_get_y_sse(src, dst);
    }
    if ((components_pattern >> AOM_PLANE_U) & 1) {
      filt_err += aom_highbd_get_u_sse(src, dst);
    }
    if ((components_pattern >> AOM_PLANE_V) & 1) {
      filt_err += aom_highbd_get_v_sse(src, dst);
    }
    return filt_err;
  }
#endif  // CONFIG_AOM_HIGHBITDEPTH
  if ((components_pattern >> AOM_PLANE_Y) & 1) {
    filt_err = aom_get_y_sse(src, dst);
  }
  if ((components_pattern >> AOM_PLANE_U) & 1) {
    filt_err += aom_get_u_sse(src, dst);
  }
  if ((components_pattern >> AOM_PLANE_V) & 1) {
    filt_err += aom_get_v_sse(src, dst);
  }
  return filt_err;
}

112
113
static int64_t try_restoration_tile(const YV12_BUFFER_CONFIG *src,
                                    AV1_COMP *const cpi, RestorationInfo *rsi,
114
115
116
                                    int components_pattern, int partial_frame,
                                    int tile_idx, int subtile_idx,
                                    int subtile_bits,
117
                                    YV12_BUFFER_CONFIG *dst_frame) {
118
119
120
121
122
123
124
125
  AV1_COMMON *const cm = &cpi->common;
  int64_t filt_err;
  int tile_width, tile_height, nhtiles, nvtiles;
  int h_start, h_end, v_start, v_end;
  const int ntiles = av1_get_rest_ntiles(cm->width, cm->height, &tile_width,
                                         &tile_height, &nhtiles, &nvtiles);
  (void)ntiles;

126
127
  av1_loop_restoration_frame(cm->frame_to_show, cm, rsi, components_pattern,
                             partial_frame, dst_frame);
128
129
130
131
  av1_get_rest_tile_limits(tile_idx, subtile_idx, subtile_bits, nhtiles,
                           nvtiles, tile_width, tile_height, cm->width,
                           cm->height, 0, 0, &h_start, &h_end, &v_start,
                           &v_end);
132
  filt_err = sse_restoration_tile(src, dst_frame, cm, h_start, h_end - h_start,
133
                                  v_start, v_end - v_start, components_pattern);
134
135
136
137
138

  return filt_err;
}

static int64_t try_restoration_frame(const YV12_BUFFER_CONFIG *src,
Yaowu Xu's avatar
Yaowu Xu committed
139
                                     AV1_COMP *const cpi, RestorationInfo *rsi,
140
                                     int components_pattern, int partial_frame,
141
                                     YV12_BUFFER_CONFIG *dst_frame) {
Yaowu Xu's avatar
Yaowu Xu committed
142
  AV1_COMMON *const cm = &cpi->common;
143
  int64_t filt_err;
144
145
146
  av1_loop_restoration_frame(cm->frame_to_show, cm, rsi, components_pattern,
                             partial_frame, dst_frame);
  filt_err = sse_restoration_frame(src, dst_frame, components_pattern);
147
148
149
  return filt_err;
}

150
151
152
153
154
static int64_t get_pixel_proj_error(uint8_t *src8, int width, int height,
                                    int src_stride, uint8_t *dat8,
                                    int dat_stride, int bit_depth,
                                    int32_t *flt1, int flt1_stride,
                                    int32_t *flt2, int flt2_stride, int *xqd) {
155
156
157
158
  int i, j;
  int64_t err = 0;
  int xq[2];
  decode_xq(xqd, xq);
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
  if (bit_depth == 8) {
    const uint8_t *src = src8;
    const uint8_t *dat = dat8;
    for (i = 0; i < height; ++i) {
      for (j = 0; j < width; ++j) {
        const int32_t u =
            (int32_t)(dat[i * dat_stride + j] << SGRPROJ_RST_BITS);
        const int32_t f1 = (int32_t)flt1[i * flt1_stride + j] - u;
        const int32_t f2 = (int32_t)flt2[i * flt2_stride + j] - u;
        const int64_t v = xq[0] * f1 + xq[1] * f2 + (u << SGRPROJ_PRJ_BITS);
        const int32_t e =
            ROUND_POWER_OF_TWO(v, SGRPROJ_RST_BITS + SGRPROJ_PRJ_BITS) -
            src[i * src_stride + j];
        err += e * e;
      }
    }
  } else {
    const uint16_t *src = CONVERT_TO_SHORTPTR(src8);
    const uint16_t *dat = CONVERT_TO_SHORTPTR(dat8);
    for (i = 0; i < height; ++i) {
      for (j = 0; j < width; ++j) {
        const int32_t u =
            (int32_t)(dat[i * dat_stride + j] << SGRPROJ_RST_BITS);
        const int32_t f1 = (int32_t)flt1[i * flt1_stride + j] - u;
        const int32_t f2 = (int32_t)flt2[i * flt2_stride + j] - u;
        const int64_t v = xq[0] * f1 + xq[1] * f2 + (u << SGRPROJ_PRJ_BITS);
        const int32_t e =
            ROUND_POWER_OF_TWO(v, SGRPROJ_RST_BITS + SGRPROJ_PRJ_BITS) -
            src[i * src_stride + j];
        err += e * e;
      }
190
191
192
193
194
    }
  }
  return err;
}

195
196
197
198
static void get_proj_subspace(uint8_t *src8, int width, int height,
                              int src_stride, uint8_t *dat8, int dat_stride,
                              int bit_depth, int32_t *flt1, int flt1_stride,
                              int32_t *flt2, int flt2_stride, int *xq) {
199
200
201
202
203
204
205
206
207
  int i, j;
  double H[2][2] = { { 0, 0 }, { 0, 0 } };
  double C[2] = { 0, 0 };
  double Det;
  double x[2];
  const int size = width * height;

  xq[0] = -(1 << SGRPROJ_PRJ_BITS) / 4;
  xq[1] = (1 << SGRPROJ_PRJ_BITS) - xq[0];
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
  if (bit_depth == 8) {
    const uint8_t *src = src8;
    const uint8_t *dat = dat8;
    for (i = 0; i < height; ++i) {
      for (j = 0; j < width; ++j) {
        const double u = (double)(dat[i * dat_stride + j] << SGRPROJ_RST_BITS);
        const double s =
            (double)(src[i * src_stride + j] << SGRPROJ_RST_BITS) - u;
        const double f1 = (double)flt1[i * flt1_stride + j] - u;
        const double f2 = (double)flt2[i * flt2_stride + j] - u;
        H[0][0] += f1 * f1;
        H[1][1] += f2 * f2;
        H[0][1] += f1 * f2;
        C[0] += f1 * s;
        C[1] += f2 * s;
      }
    }
  } else {
    const uint16_t *src = CONVERT_TO_SHORTPTR(src8);
    const uint16_t *dat = CONVERT_TO_SHORTPTR(dat8);
    for (i = 0; i < height; ++i) {
      for (j = 0; j < width; ++j) {
        const double u = (double)(dat[i * dat_stride + j] << SGRPROJ_RST_BITS);
        const double s =
            (double)(src[i * src_stride + j] << SGRPROJ_RST_BITS) - u;
        const double f1 = (double)flt1[i * flt1_stride + j] - u;
        const double f2 = (double)flt2[i * flt2_stride + j] - u;
        H[0][0] += f1 * f1;
        H[1][1] += f2 * f2;
        H[0][1] += f1 * f2;
        C[0] += f1 * s;
        C[1] += f2 * s;
      }
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
    }
  }
  H[0][0] /= size;
  H[0][1] /= size;
  H[1][1] /= size;
  H[1][0] = H[0][1];
  C[0] /= size;
  C[1] /= size;
  Det = (H[0][0] * H[1][1] - H[0][1] * H[1][0]);
  if (Det < 1e-8) return;  // ill-posed, return default values
  x[0] = (H[1][1] * C[0] - H[0][1] * C[1]) / Det;
  x[1] = (H[0][0] * C[1] - H[1][0] * C[0]) / Det;
  xq[0] = (int)rint(x[0] * (1 << SGRPROJ_PRJ_BITS));
  xq[1] = (int)rint(x[1] * (1 << SGRPROJ_PRJ_BITS));
}

void encode_xq(int *xq, int *xqd) {
  xqd[0] = -xq[0];
  xqd[0] = clamp(xqd[0], SGRPROJ_PRJ_MIN0, SGRPROJ_PRJ_MAX0);
  xqd[1] = (1 << SGRPROJ_PRJ_BITS) + xqd[0] - xq[1];
  xqd[1] = clamp(xqd[1], SGRPROJ_PRJ_MIN1, SGRPROJ_PRJ_MAX1);
}

static void search_selfguided_restoration(uint8_t *dat8, int width, int height,
                                          int dat_stride, uint8_t *src8,
                                          int src_stride, int bit_depth,
267
268
                                          int *eps, int *xqd, int32_t *rstbuf) {
  int32_t *flt1 = rstbuf;
269
  int32_t *flt2 = flt1 + RESTORATION_TILEPELS_MAX;
270
  int32_t *tmpbuf2 = flt2 + RESTORATION_TILEPELS_MAX;
271
272
273
  int i, j, ep, bestep = 0;
  int64_t err, besterr = -1;
  int exqd[2], bestxqd[2] = { 0, 0 };
274

275
276
277
278
279
280
  for (ep = 0; ep < SGRPROJ_PARAMS; ep++) {
    int exq[2];
    if (bit_depth > 8) {
      uint16_t *dat = CONVERT_TO_SHORTPTR(dat8);
      for (i = 0; i < height; ++i) {
        for (j = 0; j < width; ++j) {
281
282
          flt1[i * width + j] = (int32_t)dat[i * dat_stride + j];
          flt2[i * width + j] = (int32_t)dat[i * dat_stride + j];
283
284
285
286
287
288
289
290
        }
      }
    } else {
      uint8_t *dat = dat8;
      for (i = 0; i < height; ++i) {
        for (j = 0; j < width; ++j) {
          const int k = i * width + j;
          const int l = i * dat_stride + j;
291
292
          flt1[k] = (int32_t)dat[l];
          flt2[k] = (int32_t)dat[l];
293
294
295
296
297
298
299
        }
      }
    }
    av1_selfguided_restoration(flt1, width, height, width, bit_depth,
                               sgr_params[ep].r1, sgr_params[ep].e1, tmpbuf2);
    av1_selfguided_restoration(flt2, width, height, width, bit_depth,
                               sgr_params[ep].r2, sgr_params[ep].e2, tmpbuf2);
300
301
    get_proj_subspace(src8, width, height, src_stride, dat8, dat_stride,
                      bit_depth, flt1, width, flt2, width, exq);
302
    encode_xq(exq, exqd);
303
304
305
    err =
        get_pixel_proj_error(src8, width, height, src_stride, dat8, dat_stride,
                             bit_depth, flt1, width, flt2, width, exqd);
306
307
308
309
310
311
312
313
314
315
316
317
318
319
    if (besterr == -1 || err < besterr) {
      bestep = ep;
      besterr = err;
      bestxqd[0] = exqd[0];
      bestxqd[1] = exqd[1];
    }
  }
  *eps = bestep;
  xqd[0] = bestxqd[0];
  xqd[1] = bestxqd[1];
}

static double search_sgrproj(const YV12_BUFFER_CONFIG *src, AV1_COMP *cpi,
                             int filter_level, int partial_frame,
320
321
                             RestorationInfo *info, double *best_tile_cost,
                             YV12_BUFFER_CONFIG *dst_frame) {
322
323
324
325
326
327
  SgrprojInfo *sgrproj_info = info->sgrproj_info;
  double err, cost_norestore, cost_sgrproj;
  int bits;
  MACROBLOCK *x = &cpi->td.mb;
  AV1_COMMON *const cm = &cpi->common;
  const YV12_BUFFER_CONFIG *dgd = cm->frame_to_show;
328
  RestorationInfo *rsi = &cpi->rst_search[0];
329
330
  int tile_idx, tile_width, tile_height, nhtiles, nvtiles;
  int h_start, h_end, v_start, v_end;
331
  // Allocate for the src buffer at high precision
332
333
334
335
336
337
338
339
  const int ntiles = av1_get_rest_ntiles(cm->width, cm->height, &tile_width,
                                         &tile_height, &nhtiles, &nvtiles);
  //  Make a copy of the unfiltered / processed recon buffer
  aom_yv12_copy_y(cm->frame_to_show, &cpi->last_frame_uf);
  av1_loop_filter_frame(cm->frame_to_show, cm, &cpi->td.mb.e_mbd, filter_level,
                        1, partial_frame);
  aom_yv12_copy_y(cm->frame_to_show, &cpi->last_frame_db);

340
  rsi->frame_restoration_type = RESTORE_SGRPROJ;
341
342

  for (tile_idx = 0; tile_idx < ntiles; ++tile_idx)
343
    rsi->sgrproj_info[tile_idx].level = 0;
344
345
346
347
348
  // Compute best Sgrproj filters for each tile
  for (tile_idx = 0; tile_idx < ntiles; ++tile_idx) {
    av1_get_rest_tile_limits(tile_idx, 0, 0, nhtiles, nvtiles, tile_width,
                             tile_height, cm->width, cm->height, 0, 0, &h_start,
                             &h_end, &v_start, &v_end);
349
    err = sse_restoration_tile(src, cm->frame_to_show, cm, h_start,
350
                               h_end - h_start, v_start, v_end - v_start, 1);
351
352
353
354
355
356
357
358
359
360
361
362
363
    // #bits when a tile is not restored
    bits = av1_cost_bit(RESTORE_NONE_SGRPROJ_PROB, 0);
    cost_norestore = RDCOST_DBL(x->rdmult, x->rddiv, (bits >> 4), err);
    best_tile_cost[tile_idx] = DBL_MAX;
    search_selfguided_restoration(
        dgd->y_buffer + v_start * dgd->y_stride + h_start, h_end - h_start,
        v_end - v_start, dgd->y_stride,
        src->y_buffer + v_start * src->y_stride + h_start, src->y_stride,
#if CONFIG_AOM_HIGHBITDEPTH
        cm->bit_depth,
#else
        8,
#endif  // CONFIG_AOM_HIGHBITDEPTH
364
        &rsi->sgrproj_info[tile_idx].ep, rsi->sgrproj_info[tile_idx].xqd,
365
        cm->rst_internal.tmpbuf);
366
367
    rsi->sgrproj_info[tile_idx].level = 1;
    err = try_restoration_tile(src, cpi, rsi, 1, partial_frame, tile_idx, 0, 0,
368
                               dst_frame);
369
370
371
372
373
374
    bits = SGRPROJ_BITS << AV1_PROB_COST_SHIFT;
    bits += av1_cost_bit(RESTORE_NONE_SGRPROJ_PROB, 1);
    cost_sgrproj = RDCOST_DBL(x->rdmult, x->rddiv, (bits >> 4), err);
    if (cost_sgrproj >= cost_norestore) {
      sgrproj_info[tile_idx].level = 0;
    } else {
375
      memcpy(&sgrproj_info[tile_idx], &rsi->sgrproj_info[tile_idx],
376
377
378
379
380
381
             sizeof(sgrproj_info[tile_idx]));
      bits = SGRPROJ_BITS << AV1_PROB_COST_SHIFT;
      best_tile_cost[tile_idx] = RDCOST_DBL(
          x->rdmult, x->rddiv,
          (bits + cpi->switchable_restore_cost[RESTORE_SGRPROJ]) >> 4, err);
    }
382
    rsi->sgrproj_info[tile_idx].level = 0;
383
384
  }
  // Cost for Sgrproj filtering
385
  bits = frame_level_restore_bits[rsi->frame_restoration_type]
386
387
388
389
         << AV1_PROB_COST_SHIFT;
  for (tile_idx = 0; tile_idx < ntiles; ++tile_idx) {
    bits +=
        av1_cost_bit(RESTORE_NONE_SGRPROJ_PROB, sgrproj_info[tile_idx].level);
390
    memcpy(&rsi->sgrproj_info[tile_idx], &sgrproj_info[tile_idx],
391
392
393
394
395
           sizeof(sgrproj_info[tile_idx]));
    if (sgrproj_info[tile_idx].level) {
      bits += (SGRPROJ_BITS << AV1_PROB_COST_SHIFT);
    }
  }
396
  err = try_restoration_frame(src, cpi, rsi, 1, partial_frame, dst_frame);
397
398
399
400
401
402
  cost_sgrproj = RDCOST_DBL(x->rdmult, x->rddiv, (bits >> 4), err);

  aom_yv12_copy_y(&cpi->last_frame_uf, cm->frame_to_show);
  return cost_sgrproj;
}

403
404
405
406
407
408
409
410
411
412
413
414
415
416
417
418
419
420
421
422
423
424
425
426
427
428
429
430
431
432
433
434
435
436
static int64_t compute_sse(uint8_t *dgd, int width, int height, int dgd_stride,
                           uint8_t *src, int src_stride) {
  int64_t sse = 0;
  int i, j;
  for (i = 0; i < height; ++i) {
    for (j = 0; j < width; ++j) {
      const int diff =
          (int)dgd[i * dgd_stride + j] - (int)src[i * src_stride + j];
      sse += diff * diff;
    }
  }
  return sse;
}

#if CONFIG_AOM_HIGHBITDEPTH
static int64_t compute_sse_highbd(uint16_t *dgd, int width, int height,
                                  int dgd_stride, uint16_t *src,
                                  int src_stride) {
  int64_t sse = 0;
  int i, j;
  for (i = 0; i < height; ++i) {
    for (j = 0; j < width; ++j) {
      const int diff =
          (int)dgd[i * dgd_stride + j] - (int)src[i * src_stride + j];
      sse += diff * diff;
    }
  }
  return sse;
}
#endif  // CONFIG_AOM_HIGHBITDEPTH

static void search_domaintxfmrf_restoration(uint8_t *dgd8, int width,
                                            int height, int dgd_stride,
                                            uint8_t *src8, int src_stride,
437
                                            int bit_depth, int *sigma_r,
438
                                            uint8_t *fltbuf, int32_t *tmpbuf) {
439
440
441
442
443
  const int first_p_step = 8;
  const int second_p_range = first_p_step >> 1;
  const int second_p_step = 2;
  const int third_p_range = second_p_step >> 1;
  const int third_p_step = 1;
444
  int p, best_p0, best_p = -1;
445
446
  int64_t best_sse = INT64_MAX, sse;
  if (bit_depth == 8) {
447
    uint8_t *flt = fltbuf;
448
449
450
451
    uint8_t *dgd = dgd8;
    uint8_t *src = src8;
    // First phase
    for (p = first_p_step / 2; p < DOMAINTXFMRF_PARAMS; p += first_p_step) {
452
      av1_domaintxfmrf_restoration(dgd, width, height, dgd_stride, p, flt,
453
                                   width, tmpbuf);
454
      sse = compute_sse(flt, width, height, width, src, src_stride);
455
456
457
458
459
460
461
462
463
464
      if (sse < best_sse || best_p == -1) {
        best_p = p;
        best_sse = sse;
      }
    }
    // Second Phase
    best_p0 = best_p;
    for (p = best_p0 - second_p_range; p <= best_p0 + second_p_range;
         p += second_p_step) {
      if (p < 0 || p == best_p || p >= DOMAINTXFMRF_PARAMS) continue;
465
      av1_domaintxfmrf_restoration(dgd, width, height, dgd_stride, p, flt,
466
                                   width, tmpbuf);
467
      sse = compute_sse(flt, width, height, width, src, src_stride);
468
469
470
471
472
473
474
475
476
477
      if (sse < best_sse) {
        best_p = p;
        best_sse = sse;
      }
    }
    // Third Phase
    best_p0 = best_p;
    for (p = best_p0 - third_p_range; p <= best_p0 + third_p_range;
         p += third_p_step) {
      if (p < 0 || p == best_p || p >= DOMAINTXFMRF_PARAMS) continue;
478
      av1_domaintxfmrf_restoration(dgd, width, height, dgd_stride, p, flt,
479
                                   width, tmpbuf);
480
      sse = compute_sse(flt, width, height, width, src, src_stride);
481
482
483
484
485
486
487
      if (sse < best_sse) {
        best_p = p;
        best_sse = sse;
      }
    }
  } else {
#if CONFIG_AOM_HIGHBITDEPTH
488
    uint16_t *flt = (uint16_t *)fltbuf;
489
490
491
492
    uint16_t *dgd = CONVERT_TO_SHORTPTR(dgd8);
    uint16_t *src = CONVERT_TO_SHORTPTR(src8);
    // First phase
    for (p = first_p_step / 2; p < DOMAINTXFMRF_PARAMS; p += first_p_step) {
493
      av1_domaintxfmrf_restoration_highbd(dgd, width, height, dgd_stride, p,
494
495
                                          bit_depth, flt, width, tmpbuf);
      sse = compute_sse_highbd(flt, width, height, width, src, src_stride);
496
497
498
499
500
501
502
503
504
505
      if (sse < best_sse || best_p == -1) {
        best_p = p;
        best_sse = sse;
      }
    }
    // Second Phase
    best_p0 = best_p;
    for (p = best_p0 - second_p_range; p <= best_p0 + second_p_range;
         p += second_p_step) {
      if (p < 0 || p == best_p || p >= DOMAINTXFMRF_PARAMS) continue;
506
      av1_domaintxfmrf_restoration_highbd(dgd, width, height, dgd_stride, p,
507
508
                                          bit_depth, flt, width, tmpbuf);
      sse = compute_sse_highbd(flt, width, height, width, src, src_stride);
509
510
511
512
513
514
515
516
517
518
      if (sse < best_sse) {
        best_p = p;
        best_sse = sse;
      }
    }
    // Third Phase
    best_p0 = best_p;
    for (p = best_p0 - third_p_range; p <= best_p0 + third_p_range;
         p += third_p_step) {
      if (p < 0 || p == best_p || p >= DOMAINTXFMRF_PARAMS) continue;
519
      av1_domaintxfmrf_restoration_highbd(dgd, width, height, dgd_stride, p,
520
521
                                          bit_depth, flt, width, tmpbuf);
      sse = compute_sse_highbd(flt, width, height, width, src, src_stride);
522
523
524
525
526
527
528
529
530
531
532
533
534
535
      if (sse < best_sse) {
        best_p = p;
        best_sse = sse;
      }
    }
#else
    assert(0);
#endif  // CONFIG_AOM_HIGHBITDEPTH
  }
  *sigma_r = best_p;
}

static double search_domaintxfmrf(const YV12_BUFFER_CONFIG *src, AV1_COMP *cpi,
                                  int filter_level, int partial_frame,
536
537
                                  RestorationInfo *info, double *best_tile_cost,
                                  YV12_BUFFER_CONFIG *dst_frame) {
538
  DomaintxfmrfInfo *domaintxfmrf_info = info->domaintxfmrf_info;
539
540
  double cost_norestore, cost_domaintxfmrf;
  int64_t err;
541
542
543
544
  int bits;
  MACROBLOCK *x = &cpi->td.mb;
  AV1_COMMON *const cm = &cpi->common;
  const YV12_BUFFER_CONFIG *dgd = cm->frame_to_show;
545
  RestorationInfo *rsi = &cpi->rst_search[0];
546
547
548
549
550
551
552
553
554
555
  int tile_idx, tile_width, tile_height, nhtiles, nvtiles;
  int h_start, h_end, v_start, v_end;
  const int ntiles = av1_get_rest_ntiles(cm->width, cm->height, &tile_width,
                                         &tile_height, &nhtiles, &nvtiles);
  //  Make a copy of the unfiltered / processed recon buffer
  aom_yv12_copy_y(cm->frame_to_show, &cpi->last_frame_uf);
  av1_loop_filter_frame(cm->frame_to_show, cm, &cpi->td.mb.e_mbd, filter_level,
                        1, partial_frame);
  aom_yv12_copy_y(cm->frame_to_show, &cpi->last_frame_db);

556
  rsi->frame_restoration_type = RESTORE_DOMAINTXFMRF;
557
558

  for (tile_idx = 0; tile_idx < ntiles; ++tile_idx)
559
    rsi->domaintxfmrf_info[tile_idx].level = 0;
560
561
562
563
564
  // Compute best Domaintxfm filters for each tile
  for (tile_idx = 0; tile_idx < ntiles; ++tile_idx) {
    av1_get_rest_tile_limits(tile_idx, 0, 0, nhtiles, nvtiles, tile_width,
                             tile_height, cm->width, cm->height, 0, 0, &h_start,
                             &h_end, &v_start, &v_end);
565
    err = sse_restoration_tile(src, cm->frame_to_show, cm, h_start,
566
                               h_end - h_start, v_start, v_end - v_start, 1);
567
568
569
570
571
572
573
574
575
576
577
578
579
580
    // #bits when a tile is not restored
    bits = av1_cost_bit(RESTORE_NONE_DOMAINTXFMRF_PROB, 0);
    cost_norestore = RDCOST_DBL(x->rdmult, x->rddiv, (bits >> 4), err);
    best_tile_cost[tile_idx] = DBL_MAX;

    search_domaintxfmrf_restoration(
        dgd->y_buffer + v_start * dgd->y_stride + h_start, h_end - h_start,
        v_end - v_start, dgd->y_stride,
        src->y_buffer + v_start * src->y_stride + h_start, src->y_stride,
#if CONFIG_AOM_HIGHBITDEPTH
        cm->bit_depth,
#else
        8,
#endif  // CONFIG_AOM_HIGHBITDEPTH
581
582
        &rsi->domaintxfmrf_info[tile_idx].sigma_r, cpi->extra_rstbuf,
        cm->rst_internal.tmpbuf);
583

584
585
    rsi->domaintxfmrf_info[tile_idx].level = 1;
    err = try_restoration_tile(src, cpi, rsi, 1, partial_frame, tile_idx, 0, 0,
586
                               dst_frame);
587
588
589
590
591
592
    bits = DOMAINTXFMRF_PARAMS_BITS << AV1_PROB_COST_SHIFT;
    bits += av1_cost_bit(RESTORE_NONE_DOMAINTXFMRF_PROB, 1);
    cost_domaintxfmrf = RDCOST_DBL(x->rdmult, x->rddiv, (bits >> 4), err);
    if (cost_domaintxfmrf >= cost_norestore) {
      domaintxfmrf_info[tile_idx].level = 0;
    } else {
593
      memcpy(&domaintxfmrf_info[tile_idx], &rsi->domaintxfmrf_info[tile_idx],
594
595
596
597
598
599
600
             sizeof(domaintxfmrf_info[tile_idx]));
      bits = DOMAINTXFMRF_PARAMS_BITS << AV1_PROB_COST_SHIFT;
      best_tile_cost[tile_idx] = RDCOST_DBL(
          x->rdmult, x->rddiv,
          (bits + cpi->switchable_restore_cost[RESTORE_DOMAINTXFMRF]) >> 4,
          err);
    }
601
    rsi->domaintxfmrf_info[tile_idx].level = 0;
602
603
  }
  // Cost for Domaintxfmrf filtering
604
  bits = frame_level_restore_bits[rsi->frame_restoration_type]
605
606
607
608
         << AV1_PROB_COST_SHIFT;
  for (tile_idx = 0; tile_idx < ntiles; ++tile_idx) {
    bits += av1_cost_bit(RESTORE_NONE_DOMAINTXFMRF_PROB,
                         domaintxfmrf_info[tile_idx].level);
609
    memcpy(&rsi->domaintxfmrf_info[tile_idx], &domaintxfmrf_info[tile_idx],
610
611
612
613
614
           sizeof(domaintxfmrf_info[tile_idx]));
    if (domaintxfmrf_info[tile_idx].level) {
      bits += (DOMAINTXFMRF_PARAMS_BITS << AV1_PROB_COST_SHIFT);
    }
  }
615
  err = try_restoration_frame(src, cpi, rsi, 1, partial_frame, dst_frame);
616
617
618
619
620
621
  cost_domaintxfmrf = RDCOST_DBL(x->rdmult, x->rddiv, (bits >> 4), err);

  aom_yv12_copy_y(&cpi->last_frame_uf, cm->frame_to_show);
  return cost_domaintxfmrf;
}

622
623
static double find_average(uint8_t *src, int h_start, int h_end, int v_start,
                           int v_end, int stride) {
624
625
626
  uint64_t sum = 0;
  double avg = 0;
  int i, j;
627
628
629
  for (i = v_start; i < v_end; i++)
    for (j = h_start; j < h_end; j++) sum += src[i * stride + j];
  avg = (double)sum / ((v_end - v_start) * (h_end - h_start));
630
631
632
  return avg;
}

633
634
635
static void compute_stats(uint8_t *dgd, uint8_t *src, int h_start, int h_end,
                          int v_start, int v_end, int dgd_stride,
                          int src_stride, double *M, double *H) {
636
  int i, j, k, l;
637
  double Y[WIENER_WIN2];
638
639
  const double avg =
      find_average(dgd, h_start, h_end, v_start, v_end, dgd_stride);
640

641
642
  memset(M, 0, sizeof(*M) * WIENER_WIN2);
  memset(H, 0, sizeof(*H) * WIENER_WIN2 * WIENER_WIN2);
643
644
  for (i = v_start; i < v_end; i++) {
    for (j = h_start; j < h_end; j++) {
645
646
      const double X = (double)src[i * src_stride + j] - avg;
      int idx = 0;
647
648
      for (k = -WIENER_HALFWIN; k <= WIENER_HALFWIN; k++) {
        for (l = -WIENER_HALFWIN; l <= WIENER_HALFWIN; l++) {
649
650
651
652
          Y[idx] = (double)dgd[(i + l) * dgd_stride + (j + k)] - avg;
          idx++;
        }
      }
653
      for (k = 0; k < WIENER_WIN2; ++k) {
654
        M[k] += Y[k] * X;
655
656
        H[k * WIENER_WIN2 + k] += Y[k] * Y[k];
        for (l = k + 1; l < WIENER_WIN2; ++l) {
657
658
659
660
          // H is a symmetric matrix, so we only need to fill out the upper
          // triangle here. We can copy it down to the lower triangle outside
          // the (i, j) loops.
          H[k * WIENER_WIN2 + l] += Y[k] * Y[l];
661
662
663
664
        }
      }
    }
  }
665
666
667
668
669
  for (k = 0; k < WIENER_WIN2; ++k) {
    for (l = k + 1; l < WIENER_WIN2; ++l) {
      H[l * WIENER_WIN2 + k] = H[k * WIENER_WIN2 + l];
    }
  }
670
671
}

Yaowu Xu's avatar
Yaowu Xu committed
672
#if CONFIG_AOM_HIGHBITDEPTH
673
674
static double find_average_highbd(uint16_t *src, int h_start, int h_end,
                                  int v_start, int v_end, int stride) {
675
676
677
  uint64_t sum = 0;
  double avg = 0;
  int i, j;
678
679
680
  for (i = v_start; i < v_end; i++)
    for (j = h_start; j < h_end; j++) sum += src[i * stride + j];
  avg = (double)sum / ((v_end - v_start) * (h_end - h_start));
681
682
683
  return avg;
}

684
685
686
687
static void compute_stats_highbd(uint8_t *dgd8, uint8_t *src8, int h_start,
                                 int h_end, int v_start, int v_end,
                                 int dgd_stride, int src_stride, double *M,
                                 double *H) {
688
  int i, j, k, l;
689
  double Y[WIENER_WIN2];
690
691
  uint16_t *src = CONVERT_TO_SHORTPTR(src8);
  uint16_t *dgd = CONVERT_TO_SHORTPTR(dgd8);
692
693
  const double avg =
      find_average_highbd(dgd, h_start, h_end, v_start, v_end, dgd_stride);
694

695
696
  memset(M, 0, sizeof(*M) * WIENER_WIN2);
  memset(H, 0, sizeof(*H) * WIENER_WIN2 * WIENER_WIN2);
697
698
  for (i = v_start; i < v_end; i++) {
    for (j = h_start; j < h_end; j++) {
699
700
      const double X = (double)src[i * src_stride + j] - avg;
      int idx = 0;
701
702
      for (k = -WIENER_HALFWIN; k <= WIENER_HALFWIN; k++) {
        for (l = -WIENER_HALFWIN; l <= WIENER_HALFWIN; l++) {
703
704
705
706
          Y[idx] = (double)dgd[(i + l) * dgd_stride + (j + k)] - avg;
          idx++;
        }
      }
707
      for (k = 0; k < WIENER_WIN2; ++k) {
708
        M[k] += Y[k] * X;
709
710
        H[k * WIENER_WIN2 + k] += Y[k] * Y[k];
        for (l = k + 1; l < WIENER_WIN2; ++l) {
711
712
713
714
          // H is a symmetric matrix, so we only need to fill out the upper
          // triangle here. We can copy it down to the lower triangle outside
          // the (i, j) loops.
          H[k * WIENER_WIN2 + l] += Y[k] * Y[l];
715
716
717
718
        }
      }
    }
  }
719
720
721
722
723
  for (k = 0; k < WIENER_WIN2; ++k) {
    for (l = k + 1; l < WIENER_WIN2; ++l) {
      H[l * WIENER_WIN2 + k] = H[k * WIENER_WIN2 + l];
    }
  }
724
}
Yaowu Xu's avatar
Yaowu Xu committed
725
#endif  // CONFIG_AOM_HIGHBITDEPTH
726
727
728
729
730
731
732
733
734
735
736
737
738
739
740
741
742
743
744
745
746
747

// Solves Ax = b, where x and b are column vectors
static int linsolve(int n, double *A, int stride, double *b, double *x) {
  int i, j, k;
  double c;
  // Partial pivoting
  for (i = n - 1; i > 0; i--) {
    if (A[(i - 1) * stride] < A[i * stride]) {
      for (j = 0; j < n; j++) {
        c = A[i * stride + j];
        A[i * stride + j] = A[(i - 1) * stride + j];
        A[(i - 1) * stride + j] = c;
      }
      c = b[i];
      b[i] = b[i - 1];
      b[i - 1] = c;
    }
  }
  // Forward elimination
  for (k = 0; k < n - 1; k++) {
    for (i = k; i < n - 1; i++) {
      c = A[(i + 1) * stride + k] / A[k * stride + k];
748
      for (j = 0; j < n; j++) A[(i + 1) * stride + j] -= c * A[k * stride + j];
749
750
751
752
753
      b[i + 1] -= c * b[k];
    }
  }
  // Backward substitution
  for (i = n - 1; i >= 0; i--) {
754
    if (fabs(A[i * stride + i]) < 1e-10) return 0;
755
    c = 0;
756
    for (j = i + 1; j <= n - 1; j++) c += A[i * stride + j] * x[j];
757
758
759
760
761
762
    x[i] = (b[i] - c) / A[i * stride + i];
  }
  return 1;
}

static INLINE int wrap_index(int i) {
763
  return (i >= WIENER_HALFWIN1 ? WIENER_WIN - 1 - i : i);
764
765
766
767
768
}

// Fix vector b, update vector a
static void update_a_sep_sym(double **Mc, double **Hc, double *a, double *b) {
  int i, j;
769
770
  double S[WIENER_WIN];
  double A[WIENER_WIN], B[WIENER_WIN2];
Aamir Anis's avatar
Aamir Anis committed
771
  int w, w2;
772
773
  memset(A, 0, sizeof(A));
  memset(B, 0, sizeof(B));
774
775
  for (i = 0; i < WIENER_WIN; i++) {
    for (j = 0; j < WIENER_WIN; ++j) {
776
777
778
779
      const int jj = wrap_index(j);
      A[jj] += Mc[i][j] * b[i];
    }
  }
780
781
  for (i = 0; i < WIENER_WIN; i++) {
    for (j = 0; j < WIENER_WIN; j++) {
782
      int k, l;
783
784
      for (k = 0; k < WIENER_WIN; ++k)
        for (l = 0; l < WIENER_WIN; ++l) {
785
786
          const int kk = wrap_index(k);
          const int ll = wrap_index(l);
787
788
          B[ll * WIENER_HALFWIN1 + kk] +=
              Hc[j * WIENER_WIN + i][k * WIENER_WIN2 + l] * b[i] * b[j];
789
790
791
        }
    }
  }
Aamir Anis's avatar
Aamir Anis committed
792
  // Normalization enforcement in the system of equations itself
793
  w = WIENER_WIN;
Aamir Anis's avatar
Aamir Anis committed
794
795
  w2 = (w >> 1) + 1;
  for (i = 0; i < w2 - 1; ++i)
796
797
    A[i] -=
        A[w2 - 1] * 2 + B[i * w2 + w2 - 1] - 2 * B[(w2 - 1) * w2 + (w2 - 1)];
Aamir Anis's avatar
Aamir Anis committed
798
799
800
801
802
803
804
805
806
  for (i = 0; i < w2 - 1; ++i)
    for (j = 0; j < w2 - 1; ++j)
      B[i * w2 + j] -= 2 * (B[i * w2 + (w2 - 1)] + B[(w2 - 1) * w2 + j] -
                            2 * B[(w2 - 1) * w2 + (w2 - 1)]);
  if (linsolve(w2 - 1, B, w2, A, S)) {
    S[w2 - 1] = 1.0;
    for (i = w2; i < w; ++i) {
      S[i] = S[w - 1 - i];
      S[w2 - 1] -= 2 * S[i];
807
    }
Aamir Anis's avatar
Aamir Anis committed
808
    memcpy(a, S, w * sizeof(*a));
809
810
811
812
813
814
  }
}

// Fix vector a, update vector b
static void update_b_sep_sym(double **Mc, double **Hc, double *a, double *b) {
  int i, j;
815
816
  double S[WIENER_WIN];
  double A[WIENER_WIN], B[WIENER_WIN2];
Aamir Anis's avatar
Aamir Anis committed
817
  int w, w2;
818
819
  memset(A, 0, sizeof(A));
  memset(B, 0, sizeof(B));
820
  for (i = 0; i < WIENER_WIN; i++) {
821
    const int ii = wrap_index(i);
822
    for (j = 0; j < WIENER_WIN; j++) A[ii] += Mc[i][j] * a[j];
823
824
  }

825
826
  for (i = 0; i < WIENER_WIN; i++) {
    for (j = 0; j < WIENER_WIN; j++) {
827
828
829
      const int ii = wrap_index(i);
      const int jj = wrap_index(j);
      int k, l;
830
831
832
833
      for (k = 0; k < WIENER_WIN; ++k)
        for (l = 0; l < WIENER_WIN; ++l)
          B[jj * WIENER_HALFWIN1 + ii] +=
              Hc[i * WIENER_WIN + j][k * WIENER_WIN2 + l] * a[k] * a[l];
834
835
    }
  }
Aamir Anis's avatar
Aamir Anis committed
836
  // Normalization enforcement in the system of equations itself
837
838
  w = WIENER_WIN;
  w2 = WIENER_HALFWIN1;
Aamir Anis's avatar
Aamir Anis committed
839
  for (i = 0; i < w2 - 1; ++i)
840
841
    A[i] -=
        A[w2 - 1] * 2 + B[i * w2 + w2 - 1] - 2 * B[(w2 - 1) * w2 + (w2 - 1)];
Aamir Anis's avatar
Aamir Anis committed
842
843
844
845
846
847
848
849
850
  for (i = 0; i < w2 - 1; ++i)
    for (j = 0; j < w2 - 1; ++j)
      B[i * w2 + j] -= 2 * (B[i * w2 + (w2 - 1)] + B[(w2 - 1) * w2 + j] -
                            2 * B[(w2 - 1) * w2 + (w2 - 1)]);
  if (linsolve(w2 - 1, B, w2, A, S)) {
    S[w2 - 1] = 1.0;
    for (i = w2; i < w; ++i) {
      S[i] = S[w - 1 - i];
      S[w2 - 1] -= 2 * S[i];
851
    }
Aamir Anis's avatar
Aamir Anis committed
852
    memcpy(b, S, w * sizeof(*b));
853
854
855
  }
}

856
857
static int wiener_decompose_sep_sym(double *M, double *H, double *a,
                                    double *b) {
858
  static const double init_filt[WIENER_WIN] = {
859
    0.035623, -0.127154, 0.211436, 0.760190, 0.211436, -0.127154, 0.035623,
860
861
  };
  int i, j, iter;
862
863
864
865
866
867
868
  double *Hc[WIENER_WIN2];
  double *Mc[WIENER_WIN];
  for (i = 0; i < WIENER_WIN; i++) {
    Mc[i] = M + i * WIENER_WIN;
    for (j = 0; j < WIENER_WIN; j++) {
      Hc[i * WIENER_WIN + j] =
          H + i * WIENER_WIN * WIENER_WIN2 + j * WIENER_WIN;
869
870
    }
  }
871
872
  memcpy(a, init_filt, sizeof(*a) * WIENER_WIN);
  memcpy(b, init_filt, sizeof(*b) * WIENER_WIN);
873
874
875
876
877
878
879

  iter = 1;
  while (iter < 10) {
    update_a_sep_sym(Mc, Hc, a, b);
    update_b_sep_sym(Mc, Hc, a, b);
    iter++;
  }
880
  return 1;
881
882
}

883
// Computes the function x'*H*x - x'*M for the learned 2D filter x, and compares
Aamir Anis's avatar
Aamir Anis committed
884
885
// against identity filters; Final score is defined as the difference between
// the function values
886
static double compute_score(double *M, double *H, int *vfilt, int *hfilt) {
887
  double ab[WIENER_WIN * WIENER_WIN];
Aamir Anis's avatar
Aamir Anis committed
888
889
890
891
  int i, k, l;
  double P = 0, Q = 0;
  double iP = 0, iQ = 0;
  double Score, iScore;
892
893
894
895
896
897
898
  double a[WIENER_WIN], b[WIENER_WIN];
  a[WIENER_HALFWIN] = b[WIENER_HALFWIN] = 1.0;
  for (i = 0; i < WIENER_HALFWIN; ++i) {
    a[i] = a[WIENER_WIN - i - 1] = (double)vfilt[i] / WIENER_FILT_STEP;
    b[i] = b[WIENER_WIN - i - 1] = (double)hfilt[i] / WIENER_FILT_STEP;
    a[WIENER_HALFWIN] -= 2 * a[i];
    b[WIENER_HALFWIN] -= 2 * b[i];
Aamir Anis's avatar
Aamir Anis committed
899
  }
900
901
  for (k = 0; k < WIENER_WIN; ++k) {
    for (l = 0; l < WIENER_WIN; ++l) ab[k * WIENER_WIN + l] = a[l] * b[k];
Aamir Anis's avatar
Aamir Anis committed
902
  }
903
  for (k = 0; k < WIENER_WIN2; ++k) {
Aamir Anis's avatar
Aamir Anis committed
904
    P += ab[k] * M[k];
905
906
    for (l = 0; l < WIENER_WIN2; ++l)
      Q += ab[k] * H[k * WIENER_WIN2 + l] * ab[l];
Aamir Anis's avatar
Aamir Anis committed
907
908
909
  }
  Score = Q - 2 * P;

910
911
  iP = M[WIENER_WIN2 >> 1];
  iQ = H[(WIENER_WIN2 >> 1) * WIENER_WIN2 + (WIENER_WIN2 >> 1)];
Aamir Anis's avatar
Aamir Anis committed
912
913
914
915
916
  iScore = iQ - 2 * iP;

  return Score - iScore;
}

917
918
static void quantize_sym_filter(double *f, int *fi) {
  int i;
919
920
  for (i = 0; i < WIENER_HALFWIN; ++i) {
    fi[i] = RINT(f[i] * WIENER_FILT_STEP);
921
922
923
924
925
  }
  // Specialize for 7-tap filter
  fi[0] = CLIP(fi[0], WIENER_FILT_TAP0_MINV, WIENER_FILT_TAP0_MAXV);
  fi[1] = CLIP(fi[1], WIENER_FILT_TAP1_MINV, WIENER_FILT_TAP1_MAXV);
  fi[2] = CLIP(fi[2], WIENER_FILT_TAP2_MINV, WIENER_FILT_TAP2_MAXV);
926
927
928
929
930
931
932
933
934
935
936
937
938
939
940
941
942
943
944
945
946
947
948
949
950
951
952
953
954
955
956
957
958
959
960
  // Satisfy filter constraints
  fi[WIENER_WIN - 1] = fi[0];
  fi[WIENER_WIN - 2] = fi[1];
  fi[WIENER_WIN - 3] = fi[2];
  fi[3] = WIENER_FILT_STEP - 2 * (fi[0] + fi[1] + fi[2]);
}

static double search_wiener_uv(const YV12_BUFFER_CONFIG *src, AV1_COMP *cpi,
                               int filter_level, int partial_frame, int plane,
                               RestorationInfo *info,
                               YV12_BUFFER_CONFIG *dst_frame) {
  WienerInfo *wiener_info = info->wiener_info;
  AV1_COMMON *const cm = &cpi->common;
  RestorationInfo *rsi = cpi->rst_search;
  int64_t err;
  int bits;
  double cost_wiener = 0, cost_norestore = 0;
  MACROBLOCK *x = &cpi->td.mb;
  double M[WIENER_WIN2];
  double H[WIENER_WIN2 * WIENER_WIN2];
  double vfilterd[WIENER_WIN], hfilterd[WIENER_WIN];
  const YV12_BUFFER_CONFIG *dgd = cm->frame_to_show;
  const int width = src->uv_crop_width;
  const int height = src->uv_crop_height;
  const int src_stride = src->uv_stride;
  const int dgd_stride = dgd->uv_stride;
  double score;
  int tile_idx, tile_width, tile_height, nhtiles, nvtiles;
  const int ntiles = av1_get_rest_ntiles(cm->width, cm->height, &tile_width,
                                         &tile_height, &nhtiles, &nvtiles);

  assert(width == dgd->uv_crop_width);
  assert(height == dgd->uv_crop_height);

  //  Make a copy of the unfiltered / processed recon buffer
961
962
963
  aom_yv12_copy_y(cm->frame_to_show, &cpi->last_frame_uf);
  aom_yv12_copy_u(cm->frame_to_show, &cpi->last_frame_uf);
  aom_yv12_copy_v(cm->frame_to_show, &cpi->last_frame_uf);
964
965
  av1_loop_filter_frame(cm->frame_to_show, cm, &cpi->td.mb.e_mbd, filter_level,
                        0, partial_frame);
966
967
968
  aom_yv12_copy_y(cm->frame_to_show, &cpi->last_frame_db);
  aom_yv12_copy_u(cm->frame_to_show, &cpi->last_frame_db);
  aom_yv12_copy_v(cm->frame_to_show, &cpi->last_frame_db);
969
970
971

  rsi[plane].frame_restoration_type = RESTORE_NONE;

972
  err = sse_restoration_frame(src, cm->frame_to_show, (1 << plane));
973
974
975
976
  bits = 0;
  cost_norestore = RDCOST_DBL(x->rdmult, x->rddiv, (bits >> 4), err);

  rsi[plane].frame_restoration_type = RESTORE_WIENER;
977

978
#if CONFIG_AOM_HIGHBITDEPTH
979
980
981
982
983
984
985
986
987
988
989
990
991
992
  if (cm->use_highbitdepth) {
    if (plane == AOM_PLANE_U) {
      extend_frame_highbd(CONVERT_TO_SHORTPTR(dgd->u_buffer), width, height,
                          dgd_stride);
      compute_stats_highbd(dgd->u_buffer, src->u_buffer, 0, width, 0, height,
                           dgd_stride, src_stride, M, H);
    } else if (plane == AOM_PLANE_V) {
      extend_frame_highbd(CONVERT_TO_SHORTPTR(dgd->v_buffer), width, height,
                          dgd_stride);
      compute_stats_highbd(dgd->v_buffer, src->v_buffer, 0, width, 0, height,
                           dgd_stride, src_stride, M, H);
    } else {
      assert(0);
    }
993
  } else {
994
995
996
997
998
999
1000
1001
1002
1003
1004
1005
1006
#endif
    if (plane == AOM_PLANE_U) {
      extend_frame(dgd->u_buffer, width, height, dgd_stride);
      compute_stats(dgd->u_buffer, src->u_buffer, 0, width, 0, height,
                    dgd_stride, src_stride, M, H);
    } else if (plane == AOM_PLANE_V) {
      extend_frame(dgd->v_buffer, width, height, dgd_stride);
      compute_stats(dgd->v_buffer, src->v_buffer, 0, width, 0, height,
                    dgd_stride, src_stride, M, H);
    } else {
      assert(0);
    }
#if CONFIG_AOM_HIGHBITDEPTH
1007
  }
1008
1009
#endif

1010
1011
  if (!wiener_decompose_sep_sym(M, H, vfilterd, hfilterd)) {
    info->frame_restoration_type = RESTORE_NONE;
1012
1013
1014
    aom_yv12_copy_y(&cpi->last_frame_uf, cm->frame_to_show);
    aom_yv12_copy_u(&cpi->last_frame_uf, cm->frame_to_show);
    aom_yv12_copy_v(&cpi->last_frame_uf, cm->frame_to_show);
1015
1016
1017
1018
1019
1020
1021
1022
1023
1024
1025
1026
    return cost_norestore;
  }
  quantize_sym_filter(vfilterd, rsi[plane].wiener_info[0].vfilter);
  quantize_sym_filter(hfilterd, rsi[plane].wiener_info[0].hfilter);

  // Filter score computes the value of the function x'*A*x - x'*b for the
  // learned filter and compares it against identity filer. If there is no
  // reduction in the function, the filter is reverted back to identity
  score = compute_score(M, H, rsi[plane].wiener_info[0].vfilter,
                        rsi[plane].wiener_info[0].hfilter);
  if (score > 0.0) {
    info->frame_restoration_type = RESTORE_NONE;
1027
1028
1029
    aom_yv12_copy_y(&cpi->last_frame_uf, cm->frame_to_show);
    aom_yv12_copy_u(&cpi->last_frame_uf, cm->frame_to_show);
    aom_yv12_copy_v(&cpi->last_frame_uf, cm->frame_to_show);
1030
1031
1032
1033
1034
1035
1036
1037
1038
1039
1040
1041
1042
1043
1044
1045
1046
1047
1048
1049
    return cost_norestore;
  }

  info->frame_restoration_type = RESTORE_WIENER;
  rsi[plane].restoration_type[0] = info->restoration_type[0] = RESTORE_WIENER;
  rsi[plane].wiener_info[0].level = 1;
  memcpy(&wiener_info[0], &rsi[plane].wiener_info[0], sizeof(wiener_info[0]));
  for (tile_idx = 1; tile_idx < ntiles; ++tile_idx) {
    info->restoration_type[tile_idx] = RESTORE_WIENER;
    memcpy(&rsi[plane].wiener_info[tile_idx], &rsi[plane].wiener_info[0],
           sizeof(rsi[plane].wiener_info[0]));
    memcpy(&wiener_info[tile_idx], &rsi[plane].wiener_info[0],
           sizeof(rsi[plane].wiener_info[0]));
  }
  err = try_restoration_frame(src, cpi, rsi, (1 << plane), partial_frame,
                              dst_frame);
  bits = WIENER_FILT_BITS << AV1_PROB_COST_SHIFT;
  cost_wiener = RDCOST_DBL(x->rdmult, x->rddiv, (bits >> 4), err);
  if (cost_wiener > cost_norestore) {
    info->frame_restoration_type = RESTORE_NONE;
1050
1051
1052
    aom_yv12_copy_y(&cpi->last_frame_uf, cm->frame_to_show);
    aom_yv12_copy_u(&cpi->last_frame_uf, cm->frame_to_show);
    aom_yv12_copy_v(&cpi->last_frame_uf, cm->frame_to_show);
1053
1054
1055
    return cost_norestore;
  }

1056
1057
1058
  aom_yv12_copy_y(&cpi->last_frame_uf, cm->frame_to_show);
  aom_yv12_copy_u(&cpi->last_frame_uf, cm->frame_to_show);
  aom_yv12_copy_v(&cpi->last_frame_uf, cm->frame_to_show);
1059
  return cost_wiener;
1060
1061
}

1062
1063
static double search_wiener(const YV12_BUFFER_CONFIG *src, AV1_COMP *cpi,
                            int filter_level, int partial_frame,
1064
1065
                            RestorationInfo *info, double *best_tile_cost,
                            YV12_BUFFER_CONFIG *dst_frame) {
1066
  WienerInfo *wiener_info = info->wiener_info;
Yaowu Xu's avatar
Yaowu Xu committed
1067
  AV1_COMMON *const cm = &cpi->common;
1068
  RestorationInfo *rsi = cpi->rst_search;
1069
1070
  int64_t err;
  int bits;
1071
  double cost_wiener, cost_norestore;
1072
  MACROBLOCK *x = &cpi->td.mb;
1073
1074
1075
  double M[WIENER_WIN2];
  double H[WIENER_WIN2 * WIENER_WIN2];
  double vfilterd[WIENER_WIN], hfilterd[WIENER_WIN];
1076
1077
1078
1079
1080
  const YV12_BUFFER_CONFIG *dgd = cm->frame_to_show;
  const int width = cm->width;
  const int height = cm->height;
  const int src_stride = src->y_stride;
  const int dgd_stride = dgd->y_stride;
Aamir Anis's avatar
Aamir Anis committed
1081
  double score;
1082
  int tile_idx, tile_width, tile_height, nhtiles, nvtiles;
1083
  int h_start, h_end, v_start, v_end;
1084
  int i;
1085
1086
  const int ntiles = av1_get_rest_ntiles(width, height, &tile_width,
                                         &tile_height, &nhtiles, &nvtiles);
1087
1088
1089
1090
1091
1092
  assert(width == dgd->y_crop_width);
  assert(height == dgd->y_crop_height);
  assert(width == src->y_crop_width);
  assert(height == src->y_crop_height);

  //  Make a copy of the unfiltered / processed recon buffer
Yaowu Xu's avatar
Yaowu Xu committed
1093
1094
1095
1096
  aom_yv12_copy_y(cm->frame_to_show, &cpi->last_frame_uf);
  av1_loop_filter_frame(cm->frame_to_show, cm, &cpi->td.mb.e_mbd, filter_level,
                        1, partial_frame);
  aom_yv12_copy_y(cm->frame_to_show, &cpi->last_frame_db);
1097

1098
  rsi->frame_restoration_type = RESTORE_WIENER;
1099

1100
  for (tile_idx = 0; tile_idx < ntiles; ++tile_idx)
1101
    rsi->wiener_info[tile_idx].level = 0;
1102

1103
1104
1105
1106
1107
1108
1109
1110
1111
// Construct a (WIENER_HALFWIN)-pixel border around the frame
#if CONFIG_AOM_HIGHBITDEPTH
  if (cm->use_highbitdepth)
    extend_frame_highbd(CONVERT_TO_SHORTPTR(dgd->y_buffer), width, height,
                        dgd_stride);
  else
#endif
    extend_frame(dgd->y_buffer, width, height, dgd_stride);

1112
1113
  // Compute best Wiener filters for each tile
  for (tile_idx = 0; tile_idx < ntiles; ++tile_idx) {
1114
1115
1116
    av1_get_rest_tile_limits(tile_idx, 0, 0, nhtiles, nvtiles, tile_width,
                             tile_height, width, height, 0, 0, &h_start, &h_end,
                             &v_start, &v_end);
1117
    err = sse_restoration_tile(src, cm->frame_to_show, cm, h_start,
1118
                               h_end - h_start, v_start, v_end - v_start, 1);
1119
1120
    // #bits when a tile is not restored
    bits = av1_cost_bit(RESTORE_NONE_WIENER_PROB, 0);
1121
    cost_norestore = RDCOST_DBL(x->rdmult, x->rddiv, (bits >> 4), err);
1122
    best_tile_cost[tile_idx] = DBL_MAX;
1123
1124

    av1_get_rest_tile_limits(tile_idx, 0, 0, nhtiles, nvtiles, tile_width,
1125
1126
                             tile_height, width, height, 0, 0, &h_start, &h_end,
                             &v_start, &v_end);
Yaowu Xu's avatar
Yaowu Xu committed
1127
#if CONFIG_AOM_HIGHBITDEPTH
1128
1129
1130
1131
    if (cm->use_highbitdepth)
      compute_stats_highbd(dgd->y_buffer, src->y_buffer, h_start, h_end,
                           v_start, v_end, dgd_stride, src_stride, M, H);
    else
Yaowu Xu's avatar
Yaowu Xu committed
1132
#endif  // CONFIG_AOM_HIGHBITDEPTH
1133
1134
1135
      compute_stats(dgd