Use downsized frame in VMAF RDO update
The RDO multiplier calculation time is largely reduced with little VMAF
quality loss. Compare to old tune=vmaf_without_preprocessing, VMAF
BD-rate changes:
midres 0.24%, hdres -0.18%, lowres_10bd -0.19%, midres_10bd 0.16%
Compression speed (instruction count) of different modes
--------------------------------------------------------
MODE 480P 1080P
PSNR (baseline) 1.0x 1.0x
VMAF (before) 2.2x 10.2x
VMAF (after) 1.5x 4.1x
VMAF_WO_PP (before) 1.9x 9.5x
VMAF_WO_PP (after) 1.2x 3.4x
VMAF_W_PP 1.2x 1.5x
Full test results
-----------------
Performance counter stats for './aomenc red_kayak_480p.y4m --limit=30
-o output_new --tune=psnr --cpu-used=1':
642,437,621,503 instructions:u
84.889688323 seconds time elapsed
84.588830000 seconds user
0.291975000 seconds sys
Performance counter stats for './aomenc red_kayak_480p.y4m --limit=30
-o output_new --tune=vmaf_with_preprocessing --cpu-used=1':
775,025,691,607 instructions:u
109.458205579 seconds time elapsed
107.989239000 seconds user
4.284554000 seconds sys
Performance counter stats for './aomenc red_kayak_480p.y4m --limit=30
-o output_new --tune=vmaf --cpu-used=1':
Before:
1,405,511,943,075 instructions:u
129.392291668 seconds time elapsed
254.063729000 seconds user
49.802219000 seconds sys
After:
964,828,402,519 instructions:u
123.807091424 seconds time elapsed
161.892974000 seconds user
42.540103000 seconds sys
Performance counter stats for './aomenc red_kayak_480p.y4m --limit=30
-o output_new --tune=vmaf_without_preprocessing --cpu-used=1':
Before:
1,218,021,966,132 instructions:u
83.938745902 seconds time elapsed
206.545182000 seconds user
5.033487000 seconds sys
After:
762,965,561,826 instructions:u
79.566220976 seconds time elapsed
118.461060000 seconds user
4.023039000 seconds sys
------------------
Performance counter stats for './aomenc basketballdrive_1080p50.y4m
--limit=30 -o output_new --tune=psnr --cpu-used=1':
1,582,885,665,086 instructions:u
204.687949181 seconds time elapsed
203.775777000 seconds user
0.811887000 seconds sys
Performance counter stats for './aomenc basketballdrive_1080p50.y4m
--limit=30 -o output_new --tune=vmaf_with_preprocessing
--cpu-used=1':
2,298,587,912,062 instructions:u
361.306249556 seconds time elapsed
326.457704000 seconds user
41.029086000 seconds sys
Performance counter stats for './aomenc basketballdrive_1080p50.y4m
--limit=30 -o output_new --tune=vmaf --cpu-used=1':
Before:
16,214,552,275,288 instructions:u
696.679505444 seconds time elapsed
3579.636014000 seconds user
1746.357289000 seconds sys
After:
6,543,459,539,854 instructions:u
553.099946250 seconds time elapsed
1257.052953000 seconds user
250.151702000 seconds sys
Performance counter stats for './aomenc basketballdrive_1080p50.y4m
--limit=30 -o output_new --tune=vmaf_without_preprocessing
--cpu-used=1':
Before:
15,040,376,209,757 instructions:u
398.965821306 seconds time elapsed
3205.411780000 seconds user
1157.351525000 seconds sys
After:
5,311,331,047,231 instructions:u
247.754365865 seconds time elapsed
984.609408000 seconds user
12.773107000 seconds sys
Change-Id: Id64d59c2ef0fc63e0549a648406615edd71e43bd
diff --git a/av1/encoder/tune_vmaf.c b/av1/encoder/tune_vmaf.c
index 79220a6..5aecb4e 100644
--- a/av1/encoder/tune_vmaf.c
+++ b/av1/encoder/tune_vmaf.c
@@ -77,16 +77,14 @@
// all co-efficients must be even.
DECLARE_ALIGNED(16, static const int16_t, gauss_filter[8]) = { 0, 8, 30, 52,
30, 8, 0, 0 };
-static AOM_INLINE void gaussian_blur(const AV1_COMP *const cpi,
+static AOM_INLINE void gaussian_blur(const int bit_depth,
const YV12_BUFFER_CONFIG *source,
const YV12_BUFFER_CONFIG *dst) {
- const AV1_COMMON *cm = &cpi->common;
- const int bit_depth = cpi->td.mb.e_mbd.bd;
const int block_size = BLOCK_128X128;
- const int num_mi_w = mi_size_wide[block_size];
- const int num_mi_h = mi_size_high[block_size];
- const int num_cols = (cm->mi_cols + num_mi_w - 1) / num_mi_w;
- const int num_rows = (cm->mi_rows + num_mi_h - 1) / num_mi_h;
+ const int block_w = mi_size_wide[block_size] * 4;
+ const int block_h = mi_size_high[block_size] * 4;
+ const int num_cols = (source->y_width + block_w - 1) / block_w;
+ const int num_rows = (source->y_height + block_h - 1) / block_h;
int row, col;
ConvolveParams conv_params = get_conv_params(0, 0, bit_depth);
@@ -97,11 +95,8 @@
for (row = 0; row < num_rows; ++row) {
for (col = 0; col < num_cols; ++col) {
- const int mi_row = row * num_mi_h;
- const int mi_col = col * num_mi_w;
-
- const int row_offset_y = mi_row << 2;
- const int col_offset_y = mi_col << 2;
+ const int row_offset_y = row * block_h;
+ const int col_offset_y = col * block_w;
uint8_t *src_buf =
source->y_buffer + row_offset_y * source->y_stride + col_offset_y;
@@ -111,11 +106,11 @@
if (bit_depth > 8) {
av1_highbd_convolve_2d_sr(
CONVERT_TO_SHORTPTR(src_buf), source->y_stride,
- CONVERT_TO_SHORTPTR(dst_buf), dst->y_stride, num_mi_w << 2,
- num_mi_h << 2, &filter, &filter, 0, 0, &conv_params, bit_depth);
+ CONVERT_TO_SHORTPTR(dst_buf), dst->y_stride, block_w, block_h,
+ &filter, &filter, 0, 0, &conv_params, bit_depth);
} else {
av1_convolve_2d_sr(src_buf, source->y_stride, dst_buf, dst->y_stride,
- num_mi_w << 2, num_mi_h << 2, &filter, &filter, 0, 0,
+ block_w, block_h, &filter, &filter, 0, 0,
&conv_params);
}
}
@@ -199,6 +194,7 @@
YV12_BUFFER_CONFIG *const source) {
aom_clear_system_state();
const AV1_COMMON *const cm = &cpi->common;
+ const int bit_depth = cpi->td.mb.e_mbd.bd;
const int width = source->y_width;
const int height = source->y_height;
@@ -213,7 +209,7 @@
cpi->oxcf.border_in_pixels, cm->byte_alignment);
av1_copy_and_extend_frame(source, &source_extended);
- gaussian_blur(cpi, &source_extended, &blurred);
+ gaussian_blur(bit_depth, &source_extended, &blurred);
aom_free_frame_buffer(&source_extended);
const double best_frame_unsharp_amount =
@@ -243,19 +239,17 @@
cpi->oxcf.border_in_pixels, cm->byte_alignment);
av1_copy_and_extend_frame(source, &source_extended);
- gaussian_blur(cpi, &source_extended, &blurred);
+ gaussian_blur(bit_depth, &source_extended, &blurred);
aom_free_frame_buffer(&source_extended);
const double best_frame_unsharp_amount =
find_best_frame_unsharp_amount(cpi, source, &blurred, 0.0, 0.05, 20);
const int block_size = BLOCK_64X64;
- const int num_mi_w = mi_size_wide[block_size];
- const int num_mi_h = mi_size_high[block_size];
- const int num_cols = (cm->mi_cols + num_mi_w - 1) / num_mi_w;
- const int num_rows = (cm->mi_rows + num_mi_h - 1) / num_mi_h;
- const int block_w = num_mi_w << 2;
- const int block_h = num_mi_h << 2;
+ const int block_w = mi_size_wide[block_size] * 4;
+ const int block_h = mi_size_high[block_size] * 4;
+ const int num_cols = (source->y_width + block_w - 1) / block_w;
+ const int num_rows = (source->y_height + block_h - 1) / block_h;
double *best_unsharp_amounts =
aom_malloc(sizeof(*best_unsharp_amounts) * num_cols * num_rows);
memset(best_unsharp_amounts, 0,
@@ -340,10 +334,8 @@
// Apply best blur amounts
for (int row = 0; row < num_rows; ++row) {
for (int col = 0; col < num_cols; ++col) {
- const int mi_row = row * num_mi_h;
- const int mi_col = col * num_mi_w;
- const int row_offset_y = mi_row << 2;
- const int col_offset_y = mi_col << 2;
+ const int row_offset_y = row * block_h;
+ const int col_offset_y = col * block_w;
const int block_width = AOMMIN(source->y_width - col_offset_y, block_w);
const int block_height = AOMMIN(source->y_height - row_offset_y, block_h);
const int index = col + row * num_cols;
@@ -463,65 +455,75 @@
void av1_set_mb_vmaf_rdmult_scaling(AV1_COMP *cpi) {
AV1_COMMON *cm = &cpi->common;
- uint8_t *const y_buffer = cpi->source->y_buffer;
- const int y_stride = cpi->source->y_stride;
const int y_width = cpi->source->y_width;
const int y_height = cpi->source->y_height;
- const int block_size = BLOCK_64X64;
-
- const int num_mi_w = mi_size_wide[block_size];
- const int num_mi_h = mi_size_high[block_size];
- const int num_cols = (cm->mi_cols + num_mi_w - 1) / num_mi_w;
- const int num_rows = (cm->mi_rows + num_mi_h - 1) / num_mi_h;
- const int block_w = num_mi_w << 2;
- const int block_h = num_mi_h << 2;
+ const int resized_block_size = BLOCK_32X32;
+ const int resize_factor = 2;
const int bit_depth = cpi->td.mb.e_mbd.bd;
aom_clear_system_state();
- YV12_BUFFER_CONFIG blurred;
- memset(&blurred, 0, sizeof(blurred));
- aom_alloc_frame_buffer(&blurred, y_width, y_height, 1, 1,
+ YV12_BUFFER_CONFIG resized_source;
+ memset(&resized_source, 0, sizeof(resized_source));
+ aom_alloc_frame_buffer(&resized_source, y_width / resize_factor,
+ y_height / resize_factor, 1, 1,
cm->seq_params.use_highbitdepth,
cpi->oxcf.border_in_pixels, cm->byte_alignment);
- gaussian_blur(cpi, cpi->source, &blurred);
+ av1_resize_and_extend_frame(cpi->source, &resized_source, bit_depth,
+ av1_num_planes(cm));
+
+ const int resized_y_width = resized_source.y_width;
+ const int resized_y_height = resized_source.y_height;
+ const int resized_block_w = mi_size_wide[resized_block_size] * 4;
+ const int resized_block_h = mi_size_high[resized_block_size] * 4;
+ const int num_cols =
+ (resized_y_width + resized_block_w - 1) / resized_block_w;
+ const int num_rows =
+ (resized_y_height + resized_block_h - 1) / resized_block_h;
+
+ YV12_BUFFER_CONFIG blurred;
+ memset(&blurred, 0, sizeof(blurred));
+ aom_alloc_frame_buffer(&blurred, resized_y_width, resized_y_height, 1, 1,
+ cm->seq_params.use_highbitdepth,
+ cpi->oxcf.border_in_pixels, cm->byte_alignment);
+ gaussian_blur(bit_depth, &resized_source, &blurred);
double *scores = aom_malloc(sizeof(*scores) * (num_rows * num_cols));
memset(scores, 0, sizeof(*scores) * (num_rows * num_cols));
FrameData frame_data;
- frame_data.source = cpi->source;
+ frame_data.source = &resized_source;
frame_data.blurred = &blurred;
- frame_data.block_w = block_w;
- frame_data.block_h = block_h;
+ frame_data.block_w = resized_block_w;
+ frame_data.block_h = resized_block_h;
frame_data.num_rows = num_rows;
frame_data.num_cols = num_cols;
frame_data.row = 0;
frame_data.col = 0;
frame_data.bit_depth = bit_depth;
aom_calc_vmaf_multi_frame(&frame_data, cpi->oxcf.vmaf_model_path,
- update_frame, y_width, y_height, bit_depth, scores);
+ update_frame, resized_y_width, resized_y_height,
+ bit_depth, scores);
// Loop through each 'block_size' block.
for (int row = 0; row < num_rows; ++row) {
for (int col = 0; col < num_cols; ++col) {
- const int mi_row = row * num_mi_h;
- const int mi_col = col * num_mi_w;
const int index = row * num_cols + col;
- const int row_offset_y = mi_row << 2;
- const int col_offset_y = mi_col << 2;
+ const int row_offset_y = row * resized_block_h;
+ const int col_offset_y = col * resized_block_w;
- uint8_t *const orig_buf =
- y_buffer + row_offset_y * y_stride + col_offset_y;
+ uint8_t *const orig_buf = resized_source.y_buffer +
+ row_offset_y * resized_source.y_stride +
+ col_offset_y;
uint8_t *const blurred_buf =
blurred.y_buffer + row_offset_y * blurred.y_stride + col_offset_y;
const double vmaf = scores[index];
const double dvmaf = kBaselineVmaf - vmaf;
unsigned int sse;
- cpi->fn_ptr[block_size].vf(orig_buf, y_stride, blurred_buf,
- blurred.y_stride, &sse);
+ cpi->fn_ptr[resized_block_size].vf(orig_buf, resized_source.y_stride,
+ blurred_buf, blurred.y_stride, &sse);
- const double mse = (double)sse / (double)(y_width * y_height);
-
+ const double mse =
+ (double)sse / (double)(resized_y_width * resized_y_height);
double weight;
const double eps = 0.01 / (num_rows * num_cols);
if (dvmaf < eps || mse < eps) {
@@ -536,6 +538,7 @@
}
}
+ aom_free_frame_buffer(&resized_source);
aom_free_frame_buffer(&blurred);
aom_free(scores);
aom_clear_system_state();
@@ -635,12 +638,11 @@
cm->seq_params.use_highbitdepth,
cpi->oxcf.border_in_pixels, cm->byte_alignment);
- gaussian_blur(cpi, cur, &blurred_cur);
- gaussian_blur(cpi, last, &blurred_last);
- if (next) gaussian_blur(cpi, next, &blurred_next);
+ gaussian_blur(bit_depth, cur, &blurred_cur);
+ gaussian_blur(bit_depth, last, &blurred_last);
+ if (next) gaussian_blur(bit_depth, next, &blurred_next);
double motion1, motion2 = 65536.0;
-
if (bit_depth > 8) {
const float scale_factor = 1.0f / (float)(1 << (bit_depth - 8));
motion1 = highbd_image_sad_c(CONVERT_TO_SHORTPTR(blurred_cur.y_buffer),
