pickrst.c 57.3 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

Yaowu Xu's avatar
Yaowu Xu committed
19
#include "aom_dsp/aom_dsp_common.h"
20 21
#include "aom_dsp/binary_codes_writer.h"
#include "aom_dsp/psnr.h"
Yaowu Xu's avatar
Yaowu Xu committed
22
#include "aom_mem/aom_mem.h"
23
#include "aom_ports/mem.h"
24
#include "aom_ports/system_state.h"
25

26 27
#include "av1/common/onyxc_int.h"
#include "av1/common/quant_common.h"
28
#include "av1/common/restoration.h"
29

30
#include "av1/encoder/av1_quantize.h"
31 32 33
#include "av1/encoder/encoder.h"
#include "av1/encoder/picklpf.h"
#include "av1/encoder/pickrst.h"
34
#include "av1/encoder/mathutils.h"
35

36 37 38
// When set to RESTORE_WIENER or RESTORE_SGRPROJ only those are allowed.
// When set to RESTORE_NONE (0) we allow switchable.
const RestorationType force_restore_type = RESTORE_NONE;
39 40

// Number of Wiener iterations
41
#define NUM_WIENER_ITERS 5
42

43
typedef double (*search_restore_type)(const YV12_BUFFER_CONFIG *src,
44
                                      AV1_COMP *cpi, int partial_frame,
45
                                      int plane, RestorationInfo *info,
46
                                      RestorationType *rest_level,
47 48
                                      double *best_tile_cost,
                                      YV12_BUFFER_CONFIG *dst_frame);
49

50
const int frame_level_restore_bits[RESTORE_TYPES] = { 2, 2, 2, 2 };
51 52

static int64_t sse_restoration_tile(const YV12_BUFFER_CONFIG *src,
53 54
                                    const YV12_BUFFER_CONFIG *dst,
                                    const AV1_COMMON *cm, int h_start,
55
                                    int width, int v_start, int height,
56 57
                                    int components_pattern) {
  int64_t filt_err = 0;
58 59 60 61
  (void)cm;
  // Y and UV components cannot be mixed
  assert(components_pattern == 1 || components_pattern == 2 ||
         components_pattern == 4 || components_pattern == 6);
62
#if CONFIG_HIGHBITDEPTH
63
  if (cm->use_highbitdepth) {
64 65 66 67 68
    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) {
69 70
      filt_err +=
          aom_highbd_get_u_sse_part(src, dst, h_start, width, v_start, height);
71 72
    }
    if ((components_pattern >> AOM_PLANE_V) & 1) {
73 74
      filt_err +=
          aom_highbd_get_v_sse_part(src, dst, h_start, width, v_start, height);
75 76
    }
    return filt_err;
77
  }
78
#endif  // CONFIG_HIGHBITDEPTH
79 80 81 82
  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) {
83
    filt_err += aom_get_u_sse_part(src, dst, h_start, width, v_start, height);
84 85
  }
  if ((components_pattern >> AOM_PLANE_V) & 1) {
86
    filt_err += aom_get_v_sse_part(src, dst, h_start, width, v_start, height);
87
  }
88 89 90
  return filt_err;
}

91 92
static int64_t sse_restoration_frame(AV1_COMMON *const cm,
                                     const YV12_BUFFER_CONFIG *src,
93 94 95
                                     const YV12_BUFFER_CONFIG *dst,
                                     int components_pattern) {
  int64_t filt_err = 0;
96
#if CONFIG_HIGHBITDEPTH
97 98 99 100 101 102 103 104 105 106 107 108
  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;
  }
109 110
#else
  (void)cm;
111
#endif  // CONFIG_HIGHBITDEPTH
112 113 114 115 116 117 118 119 120 121 122 123
  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;
}

124 125
static int64_t try_restoration_tile(const YV12_BUFFER_CONFIG *src,
                                    AV1_COMP *const cpi, RestorationInfo *rsi,
126 127 128
                                    int components_pattern, int partial_frame,
                                    int tile_idx, int subtile_idx,
                                    int subtile_bits,
129
                                    YV12_BUFFER_CONFIG *dst_frame) {
130 131 132 133
  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;
134 135 136 137 138 139 140 141 142 143 144 145 146
  int ntiles, width, height;

  // Y and UV components cannot be mixed
  assert(components_pattern == 1 || components_pattern == 2 ||
         components_pattern == 4 || components_pattern == 6);

  if (components_pattern == 1) {  // Y only
    width = src->y_crop_width;
    height = src->y_crop_height;
  } else {  // Color
    width = src->uv_crop_width;
    height = src->uv_crop_height;
  }
147 148 149
  ntiles = av1_get_rest_ntiles(
      width, height, cm->rst_info[components_pattern > 1].restoration_tilesize,
      &tile_width, &tile_height, &nhtiles, &nvtiles);
150 151
  (void)ntiles;

152 153
  av1_loop_restoration_frame(cm->frame_to_show, cm, rsi, components_pattern,
                             partial_frame, dst_frame);
154
  av1_get_rest_tile_limits(tile_idx, subtile_idx, subtile_bits, nhtiles,
155 156
                           nvtiles, tile_width, tile_height, width, height, 0,
                           0, &h_start, &h_end, &v_start, &v_end);
157
  filt_err = sse_restoration_tile(src, dst_frame, cm, h_start, h_end - h_start,
158
                                  v_start, v_end - v_start, components_pattern);
159 160 161 162 163

  return filt_err;
}

static int64_t try_restoration_frame(const YV12_BUFFER_CONFIG *src,
Yaowu Xu's avatar
Yaowu Xu committed
164
                                     AV1_COMP *const cpi, RestorationInfo *rsi,
165
                                     int components_pattern, int partial_frame,
166
                                     YV12_BUFFER_CONFIG *dst_frame) {
Yaowu Xu's avatar
Yaowu Xu committed
167
  AV1_COMMON *const cm = &cpi->common;
168
  int64_t filt_err;
169 170
  av1_loop_restoration_frame(cm->frame_to_show, cm, rsi, components_pattern,
                             partial_frame, dst_frame);
171
  filt_err = sse_restoration_frame(cm, src, dst_frame, components_pattern);
172 173 174
  return filt_err;
}

175 176
static int64_t get_pixel_proj_error(const uint8_t *src8, int width, int height,
                                    int src_stride, const uint8_t *dat8,
177
                                    int dat_stride, int use_highbitdepth,
178 179
                                    int32_t *flt1, int flt1_stride,
                                    int32_t *flt2, int flt2_stride, int *xqd) {
180 181 182 183
  int i, j;
  int64_t err = 0;
  int xq[2];
  decode_xq(xqd, xq);
184
  if (!use_highbitdepth) {
185 186 187 188 189 190 191 192
    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;
David Barker's avatar
David Barker committed
193
        const int32_t v = xq[0] * f1 + xq[1] * f2 + (u << SGRPROJ_PRJ_BITS);
194 195 196 197 198 199 200 201 202 203 204 205 206 207 208
        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;
David Barker's avatar
David Barker committed
209
        const int32_t v = xq[0] * f1 + xq[1] * f2 + (u << SGRPROJ_PRJ_BITS);
210 211 212 213 214
        const int32_t e =
            ROUND_POWER_OF_TWO(v, SGRPROJ_RST_BITS + SGRPROJ_PRJ_BITS) -
            src[i * src_stride + j];
        err += e * e;
      }
215 216 217 218 219
    }
  }
  return err;
}

220 221
#define USE_SGRPROJ_REFINEMENT_SEARCH 1
static int64_t finer_search_pixel_proj_error(
222
    const uint8_t *src8, int width, int height, int src_stride,
223
    const uint8_t *dat8, int dat_stride, int use_highbitdepth, int32_t *flt1,
224
    int flt1_stride, int32_t *flt2, int flt2_stride, int start_step, int *xqd) {
225
  int64_t err = get_pixel_proj_error(src8, width, height, src_stride, dat8,
226 227
                                     dat_stride, use_highbitdepth, flt1,
                                     flt1_stride, flt2, flt2_stride, xqd);
228 229 230 231 232 233 234 235 236 237 238 239
  (void)start_step;
#if USE_SGRPROJ_REFINEMENT_SEARCH
  int64_t err2;
  int tap_min[] = { SGRPROJ_PRJ_MIN0, SGRPROJ_PRJ_MIN1 };
  int tap_max[] = { SGRPROJ_PRJ_MAX0, SGRPROJ_PRJ_MAX1 };
  for (int s = start_step; s >= 1; s >>= 1) {
    for (int p = 0; p < 2; ++p) {
      int skip = 0;
      do {
        if (xqd[p] - s >= tap_min[p]) {
          xqd[p] -= s;
          err2 = get_pixel_proj_error(src8, width, height, src_stride, dat8,
240 241
                                      dat_stride, use_highbitdepth, flt1,
                                      flt1_stride, flt2, flt2_stride, xqd);
242 243 244 245 246 247 248 249 250 251 252 253 254 255 256 257
          if (err2 > err) {
            xqd[p] += s;
          } else {
            err = err2;
            skip = 1;
            // At the highest step size continue moving in the same direction
            if (s == start_step) continue;
          }
        }
        break;
      } while (1);
      if (skip) break;
      do {
        if (xqd[p] + s <= tap_max[p]) {
          xqd[p] += s;
          err2 = get_pixel_proj_error(src8, width, height, src_stride, dat8,
258 259
                                      dat_stride, use_highbitdepth, flt1,
                                      flt1_stride, flt2, flt2_stride, xqd);
260 261 262 263 264 265 266 267 268 269 270 271 272 273 274 275
          if (err2 > err) {
            xqd[p] -= s;
          } else {
            err = err2;
            // At the highest step size continue moving in the same direction
            if (s == start_step) continue;
          }
        }
        break;
      } while (1);
    }
  }
#endif  // USE_SGRPROJ_REFINEMENT_SEARCH
  return err;
}

276
static void get_proj_subspace(const uint8_t *src8, int width, int height,
277
                              int src_stride, uint8_t *dat8, int dat_stride,
278 279 280
                              int use_highbitdepth, int32_t *flt1,
                              int flt1_stride, int32_t *flt2, int flt2_stride,
                              int *xq) {
281 282 283 284 285 286 287
  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;

288 289
  aom_clear_system_state();

290 291 292
  // Default
  xq[0] = 0;
  xq[1] = 0;
293
  if (!use_highbitdepth) {
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
    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;
      }
326 327 328 329 330 331 332 333 334 335 336 337 338 339 340 341 342
    }
  }
  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) {
343
  xqd[0] = xq[0];
344
  xqd[0] = clamp(xqd[0], SGRPROJ_PRJ_MIN0, SGRPROJ_PRJ_MAX0);
345
  xqd[1] = (1 << SGRPROJ_PRJ_BITS) - xqd[0] - xq[1];
346 347 348 349
  xqd[1] = clamp(xqd[1], SGRPROJ_PRJ_MIN1, SGRPROJ_PRJ_MAX1);
}

static void search_selfguided_restoration(uint8_t *dat8, int width, int height,
350
                                          int dat_stride, const uint8_t *src8,
351 352 353
                                          int src_stride, int use_highbitdepth,
                                          int bit_depth, int *eps, int *xqd,
                                          int32_t *rstbuf) {
354
  int32_t *flt1 = rstbuf;
355
  int32_t *flt2 = flt1 + RESTORATION_TILEPELS_MAX;
356
  int32_t *tmpbuf2 = flt2 + RESTORATION_TILEPELS_MAX;
357
  int ep, bestep = 0;
358 359
  int64_t err, besterr = -1;
  int exqd[2], bestxqd[2] = { 0, 0 };
360

361 362
  for (ep = 0; ep < SGRPROJ_PARAMS; ep++) {
    int exq[2];
363
#if CONFIG_HIGHBITDEPTH
364
    if (use_highbitdepth) {
365
      uint16_t *dat = CONVERT_TO_SHORTPTR(dat8);
366 367
#if USE_HIGHPASS_IN_SGRPROJ
      av1_highpass_filter_highbd(dat, width, height, dat_stride, flt1, width,
368
                                 sgr_params[ep].corner, sgr_params[ep].edge);
369
#else
370 371 372
      av1_selfguided_restoration_highbd(dat, width, height, dat_stride, flt1,
                                        width, bit_depth, sgr_params[ep].r1,
                                        sgr_params[ep].e1, tmpbuf2);
373
#endif  // USE_HIGHPASS_IN_SGRPROJ
374 375 376
      av1_selfguided_restoration_highbd(dat, width, height, dat_stride, flt2,
                                        width, bit_depth, sgr_params[ep].r2,
                                        sgr_params[ep].e2, tmpbuf2);
377
    } else {
378
#endif
379 380
#if USE_HIGHPASS_IN_SGRPROJ
      av1_highpass_filter(dat8, width, height, dat_stride, flt1, width,
381
                          sgr_params[ep].corner, sgr_params[ep].edge);
382 383 384 385
#else
    av1_selfguided_restoration(dat8, width, height, dat_stride, flt1, width,
                               sgr_params[ep].r1, sgr_params[ep].e1, tmpbuf2);
#endif  // USE_HIGHPASS_IN_SGRPROJ
386
      av1_selfguided_restoration(dat8, width, height, dat_stride, flt2, width,
387
                                 sgr_params[ep].r2, sgr_params[ep].e2, tmpbuf2);
388
#if CONFIG_HIGHBITDEPTH
389
    }
390
#endif
391
    aom_clear_system_state();
392
    get_proj_subspace(src8, width, height, src_stride, dat8, dat_stride,
393
                      use_highbitdepth, flt1, width, flt2, width, exq);
394
    aom_clear_system_state();
395
    encode_xq(exq, exqd);
396
    err = finer_search_pixel_proj_error(src8, width, height, src_stride, dat8,
397 398
                                        dat_stride, use_highbitdepth, flt1,
                                        width, flt2, width, 2, exqd);
399 400 401 402 403 404 405 406 407 408 409 410
    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];
}

411 412 413 414 415 416 417 418 419 420 421 422 423 424
static int count_sgrproj_bits(SgrprojInfo *sgrproj_info,
                              SgrprojInfo *ref_sgrproj_info) {
  int bits = SGRPROJ_PARAMS_BITS;
  bits += aom_count_primitive_refsubexpfin(
      SGRPROJ_PRJ_MAX0 - SGRPROJ_PRJ_MIN0 + 1, SGRPROJ_PRJ_SUBEXP_K,
      ref_sgrproj_info->xqd[0] - SGRPROJ_PRJ_MIN0,
      sgrproj_info->xqd[0] - SGRPROJ_PRJ_MIN0);
  bits += aom_count_primitive_refsubexpfin(
      SGRPROJ_PRJ_MAX1 - SGRPROJ_PRJ_MIN1 + 1, SGRPROJ_PRJ_SUBEXP_K,
      ref_sgrproj_info->xqd[1] - SGRPROJ_PRJ_MIN1,
      sgrproj_info->xqd[1] - SGRPROJ_PRJ_MIN1);
  return bits;
}

425 426 427 428 429 430 431 432 433 434 435 436 437 438 439 440 441 442 443 444 445 446 447 448
struct rest_search_ctxt {
  const YV12_BUFFER_CONFIG *src;
  AV1_COMP *cpi;
  uint8_t *dgd_buffer;
  const uint8_t *src_buffer;
  int dgd_stride;
  int src_stride;
  int partial_frame;
  RestorationInfo *info;
  RestorationType *type;
  double *best_tile_cost;
  int plane;
  int plane_width;
  int plane_height;
  int nrtiles_x;
  int nrtiles_y;
  YV12_BUFFER_CONFIG *dst_frame;
};

// Fill in ctxt. Returns the number of restoration tiles for this plane
static INLINE int init_rest_search_ctxt(
    const YV12_BUFFER_CONFIG *src, AV1_COMP *cpi, int partial_frame, int plane,
    RestorationInfo *info, RestorationType *type, double *best_tile_cost,
    YV12_BUFFER_CONFIG *dst_frame, struct rest_search_ctxt *ctxt) {
449
  AV1_COMMON *const cm = &cpi->common;
450 451 452 453 454 455 456 457 458
  ctxt->src = src;
  ctxt->cpi = cpi;
  ctxt->partial_frame = partial_frame;
  ctxt->info = info;
  ctxt->type = type;
  ctxt->best_tile_cost = best_tile_cost;
  ctxt->plane = plane;
  ctxt->dst_frame = dst_frame;

459
  const YV12_BUFFER_CONFIG *dgd = cm->frame_to_show;
460
  if (plane == AOM_PLANE_Y) {
461 462 463 464 465 466 467 468 469 470
    ctxt->plane_width = src->y_crop_width;
    ctxt->plane_height = src->y_crop_height;
    ctxt->src_buffer = src->y_buffer;
    ctxt->src_stride = src->y_stride;
    ctxt->dgd_buffer = dgd->y_buffer;
    ctxt->dgd_stride = dgd->y_stride;
    assert(ctxt->plane_width == dgd->y_crop_width);
    assert(ctxt->plane_height == dgd->y_crop_height);
    assert(ctxt->plane_width == src->y_crop_width);
    assert(ctxt->plane_height == src->y_crop_height);
471
  } else {
472 473 474 475 476 477 478 479
    ctxt->plane_width = src->uv_crop_width;
    ctxt->plane_height = src->uv_crop_height;
    ctxt->src_stride = src->uv_stride;
    ctxt->dgd_stride = dgd->uv_stride;
    ctxt->src_buffer = plane == AOM_PLANE_U ? src->u_buffer : src->v_buffer;
    ctxt->dgd_buffer = plane == AOM_PLANE_U ? dgd->u_buffer : dgd->v_buffer;
    assert(ctxt->plane_width == dgd->uv_crop_width);
    assert(ctxt->plane_height == dgd->uv_crop_height);
480
  }
481

482 483 484 485
  return av1_get_rest_ntiles(ctxt->plane_width, ctxt->plane_height,
                             cm->rst_info[plane].restoration_tilesize, NULL,
                             NULL, &ctxt->nrtiles_x, &ctxt->nrtiles_y);
}
486

487 488 489 490 491 492 493 494 495 496 497 498 499 500 501 502 503 504 505 506 507 508 509 510 511 512 513 514 515 516 517 518 519 520 521 522 523 524 525 526 527 528 529 530 531 532
typedef void (*rtile_visitor_t)(const struct rest_search_ctxt *search_ctxt,
                                int rtile_idx, int h_start, int h_end,
                                int v_start, int v_end, void *arg);

static void foreach_rtile_in_tile(const struct rest_search_ctxt *ctxt,
                                  int tile_row, int tile_col,
                                  rtile_visitor_t fun, void *arg) {
  const AV1_COMMON *const cm = &ctxt->cpi->common;
  const RestorationInfo *rsi = ctxt->cpi->rst_search;

  const int tile_width_y = cm->tile_width * MI_SIZE;
  const int tile_height_y = cm->tile_height * MI_SIZE;

  const int tile_width =
      (ctxt->plane > 0) ? ROUND_POWER_OF_TWO(tile_width_y, cm->subsampling_x)
                        : tile_width_y;
  const int tile_height =
      (ctxt->plane > 0) ? ROUND_POWER_OF_TWO(tile_height_y, cm->subsampling_y)
                        : tile_height_y;

  const int rtile_size = rsi->restoration_tilesize;
  const int rtiles_per_tile_x = tile_width * MI_SIZE / rtile_size;
  const int rtiles_per_tile_y = tile_height * MI_SIZE / rtile_size;

  const int rtile_row0 = rtiles_per_tile_y * tile_row;
  const int rtile_row1 =
      AOMMIN(rtile_row0 + rtiles_per_tile_y, ctxt->nrtiles_y);

  const int rtile_col0 = rtiles_per_tile_x * tile_col;
  const int rtile_col1 =
      AOMMIN(rtile_col0 + rtiles_per_tile_x, ctxt->nrtiles_x);

  const int rtile_width = AOMMIN(tile_width, rtile_size);
  const int rtile_height = AOMMIN(tile_height, rtile_size);

  for (int rtile_row = rtile_row0; rtile_row < rtile_row1; ++rtile_row) {
    for (int rtile_col = rtile_col0; rtile_col < rtile_col1; ++rtile_col) {
      const int rtile_idx = rtile_row * ctxt->nrtiles_x + rtile_col;
      int h_start, h_end, v_start, v_end;
      av1_get_rest_tile_limits(rtile_idx, 0, 0, ctxt->nrtiles_x,
                               ctxt->nrtiles_y, rtile_width, rtile_height,
                               ctxt->plane_width, ctxt->plane_height, 0, 0,
                               &h_start, &h_end, &v_start, &v_end);

      fun(ctxt, rtile_idx, h_start, h_end, v_start, v_end, arg);
    }
533
  }
534 535 536 537 538 539 540 541 542 543 544 545 546 547 548 549 550 551 552 553 554 555 556 557 558 559 560 561
}

static void search_sgrproj_for_rtile(const struct rest_search_ctxt *ctxt,
                                     int rtile_idx, int h_start, int h_end,
                                     int v_start, int v_end, void *arg) {
  const MACROBLOCK *const x = &ctxt->cpi->td.mb;
  const AV1_COMMON *const cm = &ctxt->cpi->common;
  RestorationInfo *rsi = ctxt->cpi->rst_search;
  SgrprojInfo *sgrproj_info = ctxt->info->sgrproj_info;

  SgrprojInfo *ref_sgrproj_info = (SgrprojInfo *)arg;

  int64_t err = sse_restoration_tile(ctxt->src, cm->frame_to_show, cm, h_start,
                                     h_end - h_start, v_start, v_end - v_start,
                                     (1 << ctxt->plane));
  // #bits when a tile is not restored
  int bits = av1_cost_bit(RESTORE_NONE_SGRPROJ_PROB, 0);
  double cost_norestore = RDCOST_DBL(x->rdmult, (bits >> 4), err);
  ctxt->best_tile_cost[rtile_idx] = DBL_MAX;

  RestorationInfo *plane_rsi = &rsi[ctxt->plane];
  SgrprojInfo *rtile_sgrproj_info = &plane_rsi->sgrproj_info[rtile_idx];
  uint8_t *dgd_start = ctxt->dgd_buffer + v_start * ctxt->dgd_stride + h_start;
  const uint8_t *src_start =
      ctxt->src_buffer + v_start * ctxt->src_stride + h_start;

  search_selfguided_restoration(dgd_start, h_end - h_start, v_end - v_start,
                                ctxt->dgd_stride, src_start, ctxt->src_stride,
562
#if CONFIG_HIGHBITDEPTH
563
                                cm->use_highbitdepth, cm->bit_depth,
564
#else
565
                                0, 8,
566
#endif  // CONFIG_HIGHBITDEPTH
567 568 569 570 571 572 573 574 575 576 577 578 579 580 581 582 583 584 585 586 587 588 589 590 591 592 593 594 595 596 597 598 599 600 601 602 603 604 605 606 607 608 609 610 611 612 613 614 615
                                &rtile_sgrproj_info->ep,
                                rtile_sgrproj_info->xqd,
                                cm->rst_internal.tmpbuf);
  plane_rsi->restoration_type[rtile_idx] = RESTORE_SGRPROJ;
  err = try_restoration_tile(ctxt->src, ctxt->cpi, rsi, (1 << ctxt->plane),
                             ctxt->partial_frame, rtile_idx, 0, 0,
                             ctxt->dst_frame);
  bits =
      count_sgrproj_bits(&plane_rsi->sgrproj_info[rtile_idx], ref_sgrproj_info)
      << AV1_PROB_COST_SHIFT;
  bits += av1_cost_bit(RESTORE_NONE_SGRPROJ_PROB, 1);
  double cost_sgrproj = RDCOST_DBL(x->rdmult, (bits >> 4), err);
  if (cost_sgrproj >= cost_norestore) {
    ctxt->type[rtile_idx] = RESTORE_NONE;
  } else {
    ctxt->type[rtile_idx] = RESTORE_SGRPROJ;
    *ref_sgrproj_info = sgrproj_info[rtile_idx] =
        plane_rsi->sgrproj_info[rtile_idx];
    ctxt->best_tile_cost[rtile_idx] = err;
  }
  plane_rsi->restoration_type[rtile_idx] = RESTORE_NONE;
}

static double search_sgrproj(const YV12_BUFFER_CONFIG *src, AV1_COMP *cpi,
                             int partial_frame, int plane,
                             RestorationInfo *info, RestorationType *type,
                             double *best_tile_cost,
                             YV12_BUFFER_CONFIG *dst_frame) {
  struct rest_search_ctxt ctxt;
  const int nrtiles =
      init_rest_search_ctxt(src, cpi, partial_frame, plane, info, type,
                            best_tile_cost, dst_frame, &ctxt);

  RestorationInfo *plane_rsi = &cpi->rst_search[plane];
  plane_rsi->frame_restoration_type = RESTORE_SGRPROJ;
  for (int rtile_idx = 0; rtile_idx < nrtiles; ++rtile_idx) {
    plane_rsi->restoration_type[rtile_idx] = RESTORE_NONE;
  }

  // Compute best Sgrproj filters for each rtile, one (encoder/decoder)
  // tile at a time.
  const AV1_COMMON *const cm = &cpi->common;
  for (int tile_row = 0; tile_row < cm->tile_rows; ++tile_row) {
    for (int tile_col = 0; tile_col < cm->tile_cols; ++tile_col) {
      SgrprojInfo ref_sgrproj_info;
      set_default_sgrproj(&ref_sgrproj_info);

      foreach_rtile_in_tile(&ctxt, tile_row, tile_col, search_sgrproj_for_rtile,
                            &ref_sgrproj_info);
616 617
    }
  }
618

619
  // Cost for Sgrproj filtering
620
  SgrprojInfo ref_sgrproj_info;
621
  set_default_sgrproj(&ref_sgrproj_info);
622 623 624 625 626 627 628 629 630 631
  SgrprojInfo *sgrproj_info = info->sgrproj_info;

  int bits = frame_level_restore_bits[plane_rsi->frame_restoration_type]
             << AV1_PROB_COST_SHIFT;
  for (int rtile_idx = 0; rtile_idx < nrtiles; ++rtile_idx) {
    bits += av1_cost_bit(RESTORE_NONE_SGRPROJ_PROB,
                         type[rtile_idx] != RESTORE_NONE);
    plane_rsi->sgrproj_info[rtile_idx] = sgrproj_info[rtile_idx];
    if (type[rtile_idx] == RESTORE_SGRPROJ) {
      bits += count_sgrproj_bits(&plane_rsi->sgrproj_info[rtile_idx],
632 633
                                 &ref_sgrproj_info)
              << AV1_PROB_COST_SHIFT;
634
      ref_sgrproj_info = plane_rsi->sgrproj_info[rtile_idx];
635
    }
636
    plane_rsi->restoration_type[rtile_idx] = type[rtile_idx];
637
  }
638 639 640
  double err = try_restoration_frame(src, cpi, cpi->rst_search, (1 << plane),
                                     partial_frame, dst_frame);
  double cost_sgrproj = RDCOST_DBL(cpi->td.mb.rdmult, (bits >> 4), err);
641 642 643
  return cost_sgrproj;
}

644 645
static double find_average(const uint8_t *src, int h_start, int h_end,
                           int v_start, int v_end, int stride) {
646 647 648
  uint64_t sum = 0;
  double avg = 0;
  int i, j;
649
  aom_clear_system_state();
650 651 652
  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));
653 654 655
  return avg;
}

656 657 658 659
static void compute_stats(int wiener_win, const uint8_t *dgd,
                          const 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) {
660
  int i, j, k, l;
661
  double Y[WIENER_WIN2];
662 663
  const int wiener_win2 = wiener_win * wiener_win;
  const int wiener_halfwin = (wiener_win >> 1);
664 665
  const double avg =
      find_average(dgd, h_start, h_end, v_start, v_end, dgd_stride);
666

667 668
  memset(M, 0, sizeof(*M) * wiener_win2);
  memset(H, 0, sizeof(*H) * wiener_win2 * wiener_win2);
669 670
  for (i = v_start; i < v_end; i++) {
    for (j = h_start; j < h_end; j++) {
671 672
      const double X = (double)src[i * src_stride + j] - avg;
      int idx = 0;
673 674
      for (k = -wiener_halfwin; k <= wiener_halfwin; k++) {
        for (l = -wiener_halfwin; l <= wiener_halfwin; l++) {
675 676 677 678
          Y[idx] = (double)dgd[(i + l) * dgd_stride + (j + k)] - avg;
          idx++;
        }
      }
679
      for (k = 0; k < wiener_win2; ++k) {
680
        M[k] += Y[k] * X;
681 682
        H[k * wiener_win2 + k] += Y[k] * Y[k];
        for (l = k + 1; l < wiener_win2; ++l) {
683 684 685
          // 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.
686
          H[k * wiener_win2 + l] += Y[k] * Y[l];
687 688 689 690
        }
      }
    }
  }
691 692 693
  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];
694 695
    }
  }
696 697
}

698
#if CONFIG_HIGHBITDEPTH
699
static double find_average_highbd(const uint16_t *src, int h_start, int h_end,
700
                                  int v_start, int v_end, int stride) {
701 702 703
  uint64_t sum = 0;
  double avg = 0;
  int i, j;
704
  aom_clear_system_state();
705 706 707
  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));
708 709 710
  return avg;
}

711 712 713 714
static void compute_stats_highbd(int wiener_win, const uint8_t *dgd8,
                                 const 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) {
715
  int i, j, k, l;
716
  double Y[WIENER_WIN2];
717 718
  const int wiener_win2 = wiener_win * wiener_win;
  const int wiener_halfwin = (wiener_win >> 1);
719 720
  const uint16_t *src = CONVERT_TO_SHORTPTR(src8);
  const uint16_t *dgd = CONVERT_TO_SHORTPTR(dgd8);
721 722
  const double avg =
      find_average_highbd(dgd, h_start, h_end, v_start, v_end, dgd_stride);
723

724 725
  memset(M, 0, sizeof(*M) * wiener_win2);
  memset(H, 0, sizeof(*H) * wiener_win2 * wiener_win2);
726 727
  for (i = v_start; i < v_end; i++) {
    for (j = h_start; j < h_end; j++) {
728 729
      const double X = (double)src[i * src_stride + j] - avg;
      int idx = 0;
730 731
      for (k = -wiener_halfwin; k <= wiener_halfwin; k++) {
        for (l = -wiener_halfwin; l <= wiener_halfwin; l++) {
732 733 734 735
          Y[idx] = (double)dgd[(i + l) * dgd_stride + (j + k)] - avg;
          idx++;
        }
      }
736
      for (k = 0; k < wiener_win2; ++k) {
737
        M[k] += Y[k] * X;
738 739
        H[k * wiener_win2 + k] += Y[k] * Y[k];
        for (l = k + 1; l < wiener_win2; ++l) {
740 741 742
          // 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.
743
          H[k * wiener_win2 + l] += Y[k] * Y[l];
744 745 746 747
        }
      }
    }
  }
748 749 750
  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];
751 752
    }
  }
753
}
754
#endif  // CONFIG_HIGHBITDEPTH
755

756 757 758
static INLINE int wrap_index(int i, int wiener_win) {
  const int wiener_halfwin1 = (wiener_win >> 1) + 1;
  return (i >= wiener_halfwin1 ? wiener_win - 1 - i : i);
759 760 761
}

// Fix vector b, update vector a
762 763
static void update_a_sep_sym(int wiener_win, double **Mc, double **Hc,
                             double *a, double *b) {
764
  int i, j;
765
  double S[WIENER_WIN];
766
  double A[WIENER_HALFWIN1], B[WIENER_HALFWIN1 * WIENER_HALFWIN1];
767 768
  const int wiener_win2 = wiener_win * wiener_win;
  const int wiener_halfwin1 = (wiener_win >> 1) + 1;
769 770
  memset(A, 0, sizeof(A));
  memset(B, 0, sizeof(B));
771 772 773
  for (i = 0; i < wiener_win; i++) {
    for (j = 0; j < wiener_win; ++j) {
      const int jj = wrap_index(j, wiener_win);
774 775 776
      A[jj] += Mc[i][j] * b[i];
    }
  }
777 778
  for (i = 0; i < wiener_win; i++) {
    for (j = 0; j < wiener_win; j++) {
779
      int k, l;
780 781 782 783 784 785
      for (k = 0; k < wiener_win; ++k)
        for (l = 0; l < wiener_win; ++l) {
          const int kk = wrap_index(k, wiener_win);
          const int ll = wrap_index(l, wiener_win);
          B[ll * wiener_halfwin1 + kk] +=
              Hc[j * wiener_win + i][k * wiener_win2 + l] * b[i] * b[j];
786 787 788
        }
    }
  }
Aamir Anis's avatar
Aamir Anis committed
789
  // Normalization enforcement in the system of equations itself
790
  for (i = 0; i < wiener_halfwin1 - 1; ++i)
791
    A[i] -=
792 793 794 795 796 797 798 799 800 801 802 803 804 805 806
        A[wiener_halfwin1 - 1] * 2 +
        B[i * wiener_halfwin1 + wiener_halfwin1 - 1] -
        2 * B[(wiener_halfwin1 - 1) * wiener_halfwin1 + (wiener_halfwin1 - 1)];
  for (i = 0; i < wiener_halfwin1 - 1; ++i)
    for (j = 0; j < wiener_halfwin1 - 1; ++j)
      B[i * wiener_halfwin1 + j] -=
          2 * (B[i * wiener_halfwin1 + (wiener_halfwin1 - 1)] +
               B[(wiener_halfwin1 - 1) * wiener_halfwin1 + j] -
               2 * B[(wiener_halfwin1 - 1) * wiener_halfwin1 +
                     (wiener_halfwin1 - 1)]);
  if (linsolve(wiener_halfwin1 - 1, B, wiener_halfwin1, A, S)) {
    S[wiener_halfwin1 - 1] = 1.0;
    for (i = wiener_halfwin1; i < wiener_win; ++i) {
      S[i] = S[wiener_win - 1 - i];
      S[wiener_halfwin1 - 1] -= 2 * S[i];
807
    }
808
    memcpy(a, S, wiener_win * sizeof(*a));
809 810 811 812
  }
}

// Fix vector a, update vector b
813 814
static void update_b_sep_sym(int wiener_win, double **Mc, double **Hc,
                             double *a, double *b) {
815
  int i, j;
816
  double S[WIENER_WIN];
817
  double A[WIENER_HALFWIN1], B[WIENER_HALFWIN1 * WIENER_HALFWIN1];
818 819
  const int wiener_win2 = wiener_win * wiener_win;
  const int wiener_halfwin1 = (wiener_win >> 1) + 1;
820 821
  memset(A, 0, sizeof(A));
  memset(B, 0, sizeof(B));
822 823 824
  for (i = 0; i < wiener_win; i++) {
    const int ii = wrap_index(i, wiener_win);
    for (j = 0; j < wiener_win; j++) A[ii] += Mc[i][j] * a[j];
825 826
  }

827 828 829 830
  for (i = 0; i < wiener_win; i++) {
    for (j = 0; j < wiener_win; j++) {
      const int ii = wrap_index(i, wiener_win);
      const int jj = wrap_index(j, wiener_win);
831
      int k, l;
832 833 834 835
      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];
836 837
    }
  }
Aamir Anis's avatar
Aamir Anis committed
838
  // Normalization enforcement in the system of equations itself
839
  for (i = 0; i < wiener_halfwin1 - 1; ++i)
840
    A[i] -=
841 842 843 844 845 846 847 848 849 850 851 852 853 854 855
        A[wiener_halfwin1 - 1] * 2 +
        B[i * wiener_halfwin1 + wiener_halfwin1 - 1] -
        2 * B[(wiener_halfwin1 - 1) * wiener_halfwin1 + (wiener_halfwin1 - 1)];
  for (i = 0; i < wiener_halfwin1 - 1; ++i)
    for (j = 0; j < wiener_halfwin1 - 1; ++j)
      B[i * wiener_halfwin1 + j] -=
          2 * (B[i * wiener_halfwin1 + (wiener_halfwin1 - 1)] +
               B[(wiener_halfwin1 - 1) * wiener_halfwin1 + j] -
               2 * B[(wiener_halfwin1 - 1) * wiener_halfwin1 +
                     (wiener_halfwin1 - 1)]);
  if (linsolve(wiener_halfwin1 - 1, B, wiener_halfwin1, A, S)) {
    S[wiener_halfwin1 - 1] = 1.0;
    for (i = wiener_halfwin1; i < wiener_win; ++i) {
      S[i] = S[wiener_win - 1 - i];
      S[wiener_halfwin1 - 1] -= 2 * S[i];
856
    }
857
    memcpy(b, S, wiener_win * sizeof(*b));
858 859 860
  }
}

861 862
static int wiener_decompose_sep_sym(int wiener_win, double *M, double *H,
                                    double *a, double *b) {
863 864 865 866
  static const int init_filt[WIENER_WIN] = {
    WIENER_FILT_TAP0_MIDV, WIENER_FILT_TAP1_MIDV, WIENER_FILT_TAP2_MIDV,
    WIENER_FILT_TAP3_MIDV, WIENER_FILT_TAP2_MIDV, WIENER_FILT_TAP1_MIDV,
    WIENER_FILT_TAP0_MIDV,
867
  };
868 869
  double *Hc[WIENER_WIN2];
  double *Mc[WIENER_WIN];
870 871 872 873 874
  int i, j, iter;
  const int plane_off = (WIENER_WIN - wiener_win) >> 1;
  const int wiener_win2 = wiener_win * wiener_win;
  for (i = 0; i < wiener_win; i++) {
    a[i] = b[i] = (double)init_filt[i + plane_off] / WIENER_FILT_STEP;
875
  }
876 877 878 879 880 881
  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;
    }
882
  }
883 884

  iter = 1;
885
  while (iter < NUM_WIENER_ITERS) {
886 887
    update_a_sep_sym(wiener_win, Mc, Hc, a, b);
    update_b_sep_sym(wiener_win, Mc, Hc, a, b);
888 889
    iter++;
  }
890
  return 1;
891 892
}

893
// Computes the function x'*H*x - x'*M for the learned 2D filter x, and compares
Aamir Anis's avatar
Aamir Anis committed
894 895
// against identity filters; Final score is defined as the difference between
// the function values
896 897
static double compute_score(int wiener_win, double *M, double *H,
                            InterpKernel vfilt, InterpKernel hfilt) {
898
  double ab[WIENER_WIN * WIENER_WIN];
Aamir Anis's avatar
Aamir Anis committed
899 900 901 902
  int i, k, l;
  double P = 0, Q = 0;
  double iP = 0, iQ = 0;
  double Score, iScore;
903
  double a[WIENER_WIN], b[WIENER_WIN];
904 905
  const int plane_off = (WIENER_WIN - wiener_win) >> 1;
  const int wiener_win2 = wiener_win * wiener_win;
906 907 908

  aom_clear_system_state();

909 910 911 912 913 914
  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
915
  }
916 917 918
  for (k = 0; k < wiener_win; ++k) {
    for (l = 0; l < wiener_win; ++l)
      ab[k * wiener_win + l] = a[l + plane_off] * b[k + plane_off];
Aamir Anis's avatar
Aamir Anis committed
919
  }
920
  for (k = 0; k < wiener_win2; ++k) {
Aamir Anis's avatar
Aamir Anis committed
921
    P += ab[k] * M[k];
922 923
    for (l = 0; l < wiener_win2; ++l)
      Q += ab[k] * H[k * wiener_win2 + l] * ab[l];
Aamir Anis's avatar
Aamir Anis committed
924 925 926
  }
  Score = Q - 2 * P;

927 928
  iP = M[wiener_win2 >> 1];
  iQ = H[(wiener_win2 >> 1) * wiener_win2 + (wiener_win2 >> 1)];
Aamir Anis's avatar
Aamir Anis committed
929 930 931 932 933
  iScore = iQ - 2 * iP;

  return Score - iScore;
}

934
static void quantize_sym_filter(int wiener_win, double *f, InterpKernel fi) {