SUPERRES_AUTO: Enable superres + TPL combo.
This includes TPL based modulation of both q index and rdmult.
BDRate baseline vs SUPERRES_AUTO mode for hdres2:
(1) VBR mode:
- before = +0.159 (per clip: -0.2 to +0.8)
- after = -0.057 (per clip: -1.7 to +0.5)
(2) Q mode:
- before = +0.112 (per clip: -0.40 to +0.70)
- after = +0.008 (per clip: -0.45 to +0.39)
BUG=aomedia:2844
Change-Id: I5ee1b7630e6765a701dfd3073e20c8ec7f7fadca
(cherry picked from commit 68af52828e39a4bcab73d5dc317ad86b84c035c8)
diff --git a/av1/encoder/encodeframe_utils.c b/av1/encoder/encodeframe_utils.c
index 7cc3bb7..eb9b3e0 100644
--- a/av1/encoder/encodeframe_utils.c
+++ b/av1/encoder/encodeframe_utils.c
@@ -62,6 +62,26 @@
aom_clear_system_state();
}
+// Return the end column for the current superblock, in unit of TPL blocks.
+static int get_superblock_tpl_column_end(const AV1_COMMON *const cm, int mi_col,
+ int num_mi_w) {
+ // Find the start column of this superblock.
+ const int sb_mi_col_start = (mi_col >> cm->seq_params.mib_size_log2)
+ << cm->seq_params.mib_size_log2;
+ // Same but in superres upscaled dimension.
+ const int sb_mi_col_start_sr =
+ coded_to_superres_mi(sb_mi_col_start, cm->superres_scale_denominator);
+ // Width of this superblock in mi units.
+ const int sb_mi_width = mi_size_wide[cm->seq_params.sb_size];
+ // Same but in superres upscaled dimension.
+ const int sb_mi_width_sr =
+ coded_to_superres_mi(sb_mi_width, cm->superres_scale_denominator);
+ // Superblock end in mi units.
+ const int sb_mi_end = sb_mi_col_start_sr + sb_mi_width_sr;
+ // Superblock end in TPL units.
+ return (sb_mi_end + num_mi_w - 1) / num_mi_w;
+}
+
int av1_get_hier_tpl_rdmult(const AV1_COMP *const cpi, MACROBLOCK *const x,
const BLOCK_SIZE bsize, const int mi_row,
const int mi_col, int orig_rdmult) {
@@ -75,24 +95,34 @@
if (tpl_frame->is_valid == 0) return deltaq_rdmult;
if (!is_frame_tpl_eligible(gf_group, gf_group->index)) return deltaq_rdmult;
if (tpl_idx >= MAX_TPL_FRAME_IDX) return deltaq_rdmult;
- if (cpi->superres_mode != AOM_SUPERRES_NONE) return deltaq_rdmult;
if (cpi->oxcf.q_cfg.aq_mode != NO_AQ) return deltaq_rdmult;
+ const int mi_col_sr =
+ coded_to_superres_mi(mi_col, cm->superres_scale_denominator);
+ const int mi_cols_sr = av1_pixels_to_mi(cm->superres_upscaled_width);
+ const int block_mi_width_sr =
+ coded_to_superres_mi(mi_size_wide[bsize], cm->superres_scale_denominator);
+
const int bsize_base = BLOCK_16X16;
const int num_mi_w = mi_size_wide[bsize_base];
const int num_mi_h = mi_size_high[bsize_base];
- const int num_cols = (cm->mi_params.mi_cols + num_mi_w - 1) / num_mi_w;
+ const int num_cols = (mi_cols_sr + num_mi_w - 1) / num_mi_w;
const int num_rows = (cm->mi_params.mi_rows + num_mi_h - 1) / num_mi_h;
- const int num_bcols = (mi_size_wide[bsize] + num_mi_w - 1) / num_mi_w;
+ const int num_bcols = (block_mi_width_sr + num_mi_w - 1) / num_mi_w;
const int num_brows = (mi_size_high[bsize] + num_mi_h - 1) / num_mi_h;
+ // This is required because the end col of superblock may be off by 1 in case
+ // of superres.
+ const int sb_bcol_end = get_superblock_tpl_column_end(cm, mi_col, num_mi_w);
int row, col;
double base_block_count = 0.0;
double geom_mean_of_scale = 0.0;
aom_clear_system_state();
for (row = mi_row / num_mi_w;
row < num_rows && row < mi_row / num_mi_w + num_brows; ++row) {
- for (col = mi_col / num_mi_h;
- col < num_cols && col < mi_col / num_mi_h + num_bcols; ++col) {
+ for (col = mi_col_sr / num_mi_h;
+ col < num_cols && col < mi_col_sr / num_mi_h + num_bcols &&
+ col < sb_bcol_end;
+ ++col) {
const int index = row * num_cols + col;
geom_mean_of_scale += log(cpi->tpl_sb_rdmult_scaling_factors[index]);
base_block_count += 1.0;
@@ -677,8 +707,11 @@
coded_to_superres_mi(mi_col + mi_wide, cm->superres_scale_denominator);
const int mi_cols_sr = av1_pixels_to_mi(cm->superres_upscaled_width);
const int step = 1 << block_mis_log2;
- for (int row = mi_row; row < mi_row + mi_high; row += step) {
- for (int col = mi_col_sr; col < mi_col_end_sr; col += step) {
+ const int row_step = step;
+ const int col_step_sr =
+ coded_to_superres_mi(step, cm->superres_scale_denominator);
+ for (int row = mi_row; row < mi_row + mi_high; row += row_step) {
+ for (int col = mi_col_sr; col < mi_col_end_sr; col += col_step_sr) {
if (row >= cm->mi_params.mi_rows || col >= mi_cols_sr) continue;
TplDepStats *this_stats =
&tpl_stats[av1_tpl_ptr_pos(row, col, tpl_stride, block_mis_log2)];
@@ -690,6 +723,7 @@
mi_count++;
}
}
+ assert(mi_count <= MAX_TPL_BLK_IN_SB * MAX_TPL_BLK_IN_SB);
aom_clear_system_state();
@@ -747,7 +781,6 @@
sb_enc->tpl_data_count = 0;
if (!cpi->oxcf.algo_cfg.enable_tpl_model) return;
- if (cpi->superres_mode != AOM_SUPERRES_NONE) return;
if (cpi->common.current_frame.frame_type == KEY_FRAME) return;
const FRAME_UPDATE_TYPE update_type = get_frame_update_type(&cpi->gf_group);
if (update_type == INTNL_OVERLAY_UPDATE || update_type == OVERLAY_UPDATE ||
@@ -781,15 +814,18 @@
// Here always use motion estimation size to avoid getting repetitive inter/
// intra cost.
const BLOCK_SIZE tpl_bsize = convert_length_to_bsize(tpl_data->tpl_bsize_1d);
- const int step = mi_size_wide[tpl_bsize];
assert(mi_size_wide[tpl_bsize] == mi_size_high[tpl_bsize]);
+ const int row_step = mi_size_high[tpl_bsize];
+ const int col_step_sr = coded_to_superres_mi(mi_size_wide[tpl_bsize],
+ cm->superres_scale_denominator);
// Stride is only based on SB size, and we fill in values for every 16x16
// block in a SB.
- sb_enc->tpl_stride = (mi_col_end_sr - mi_col_sr) / step;
+ sb_enc->tpl_stride = (mi_col_end_sr - mi_col_sr) / col_step_sr;
- for (int row = mi_row; row < mi_row + mi_high; row += step) {
- for (int col = mi_col_sr; col < mi_col_end_sr; col += step) {
+ for (int row = mi_row; row < mi_row + mi_high; row += row_step) {
+ for (int col = mi_col_sr; col < mi_col_end_sr; col += col_step_sr) {
+ assert(count < MAX_TPL_BLK_IN_SB * MAX_TPL_BLK_IN_SB);
// Handle partial SB, so that no invalid values are used later.
if (row >= cm->mi_params.mi_rows || col >= mi_cols_sr) {
sb_enc->tpl_inter_cost[count] = INT64_MAX;
@@ -811,6 +847,7 @@
}
}
+ assert(mi_count <= MAX_TPL_BLK_IN_SB * MAX_TPL_BLK_IN_SB);
sb_enc->tpl_data_count = mi_count;
}
@@ -849,8 +886,11 @@
coded_to_superres_mi(mi_col + mi_wide, cm->superres_scale_denominator);
const int mi_cols_sr = av1_pixels_to_mi(cm->superres_upscaled_width);
const int step = 1 << block_mis_log2;
- for (int row = mi_row; row < mi_row + mi_high; row += step) {
- for (int col = mi_col_sr; col < mi_col_end_sr; col += step) {
+ const int row_step = step;
+ const int col_step_sr =
+ coded_to_superres_mi(step, cm->superres_scale_denominator);
+ for (int row = mi_row; row < mi_row + mi_high; row += row_step) {
+ for (int col = mi_col_sr; col < mi_col_end_sr; col += col_step_sr) {
if (row >= cm->mi_params.mi_rows || col >= mi_cols_sr) continue;
TplDepStats *this_stats =
&tpl_stats[av1_tpl_ptr_pos(row, col, tpl_stride, block_mis_log2)];
@@ -862,6 +902,7 @@
mi_count++;
}
}
+ assert(mi_count <= MAX_TPL_BLK_IN_SB * MAX_TPL_BLK_IN_SB);
aom_clear_system_state();
