blender
/
blender
136
406
Fork 590
Code Issues 6.2k Pull Requests 760 Projects 19 Wiki Activity

ImBuf: Limit transform region to pixels that are affected by the transformation. 

Only the area where the source buffer will be drawn on the destination buffer
will be evaluated. This will improve performance in sequencer and image editors as
less calcuations needs to happen.
This commit is contained in:
Jeroen Bakker 2023-01-27 09:56:19 +01:00
parent b3fd169259
commit 71d7c919c0
Notes: blender-bot 2023-04-27 04:37:01 +02:00 
Referenced by issue #107390, Regression: Image transformation in VSE is very slow
Referenced by commit be53da087d, Fix #107390: Image transformation in VSE is very slow
1 changed files with 64 additions and 42 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

106
source/blender/imbuf/intern/transform.cc
View File

@ -8,10 +8,12 @@
#include <array>
#include <type_traits>
#include "BLI_float4x4.hh"
#include "BLI_math.h"
#include "BLI_math_color_blend.h"
#include "BLI_math_vector.hh"
#include "BLI_rect.h"
#include "BLI_task.hh"
#include "BLI_vector.hh"
#include "IMB_imbuf.h"
@ -46,6 +48,11 @@ struct TransformUserData {
Vector<double2, 9> delta_uvs;
} subsampling;
struct {
IndexRange x_range;
IndexRange y_range;
} destination_region;
/**
* \brief Cropping region in source image pixel space.
*/
@ -54,12 +61,15 @@ struct TransformUserData {
/**
* \brief Initialize the start_uv, add_x and add_y fields based on the given transform matrix.
*/
void init(const float transform_matrix[4][4], const int num_subsamples)
void init(const float transform_matrix[4][4],
const int num_subsamples,
const bool do_crop_destination_region)
{
init_start_uv(transform_matrix);
init_add_x(transform_matrix);
init_add_y(transform_matrix);
init_subsampling(num_subsamples);
init_destination_region(transform_matrix, do_crop_destination_region);
}
private:
@ -110,6 +120,39 @@ struct TransformUserData {
}
}
}
void init_destination_region(const float4x4 &transform_matrix,
const bool do_crop_destination_region)
{
if (!do_crop_destination_region) {
destination_region.x_range = IndexRange(dst->x);
destination_region.y_range = IndexRange(dst->y);
return;
}
/* Transform the src_crop to the destination buffer with a margin.*/
const int2 margin(2);
rcti rect;
BLI_rcti_init_minmax(&rect);
float4x4 inverse = transform_matrix.inverted();
for (const int2 &src_coords : {
int2(src_crop.xmin, src_crop.ymin),
int2(src_crop.xmax, src_crop.ymin),
int2(src_crop.xmin, src_crop.ymax),
int2(src_crop.xmax, src_crop.ymax),
}) {
float3 dst_co = inverse * float3(src_coords.x, src_coords.y, 0.0f);
BLI_rcti_do_minmax_v(&rect, int2(dst_co) + margin);
BLI_rcti_do_minmax_v(&rect, int2(dst_co) - margin);
}
/* Clamp rect to fit inside the image buffer.*/
rcti dest_rect;
BLI_rcti_init(&dest_rect, 0, dst->x, 0, dst->y);
BLI_rcti_isect(&rect, &dest_rect, &rect);
destination_region.x_range = IndexRange(rect.xmin, BLI_rcti_size_x(&rect));
destination_region.y_range = IndexRange(rect.ymin, BLI_rcti_size_y(&rect));
}
};
/**
@ -545,22 +588,14 @@ class ScanlineProcessor {
private:
void process_one_sample_per_pixel(const TransformUserData *user_data, int scanline)
{
const int width = user_data->dst->x;
double2 uv = user_data->start_uv + user_data->add_y * scanline;
double2 uv = user_data->start_uv +
user_data->destination_region.x_range.first() * user_data->add_x +
user_data->add_y * scanline;
output.init_pixel_pointer(user_data->dst, int2(0, scanline));
int xi = 0;
while (xi < width) {
const bool discard_pixel = discarder.should_discard(*user_data, uv);
if (!discard_pixel) {
break;
}
uv += user_data->add_x;
output.increase_pixel_pointer();
xi += 1;
}
for (; xi < width; xi++) {
output.init_pixel_pointer(user_data->dst,
int2(user_data->destination_region.x_range.first(), scanline));
for (int xi : user_data->destination_region.x_range) {
UNUSED_VARS(xi);
if (!discarder.should_discard(*user_data, uv)) {
typename Sampler::SampleType sample;
sampler.sample(user_data->src, uv, sample);
@ -574,31 +609,14 @@ class ScanlineProcessor {
void process_with_subsampling(const TransformUserData *user_data, int scanline)
{
const int width = user_data->dst->x;
double2 uv = user_data->start_uv + user_data->add_y * scanline;
double2 uv = user_data->start_uv +
user_data->destination_region.x_range.first() * user_data->add_x +
user_data->add_y * scanline;
output.init_pixel_pointer(user_data->dst, int2(0, scanline));
int xi = 0;
/*
* Skip leading pixels that would be fully discarded.
*
* NOTE: This could be improved by intersection between an ray and the image bounds.
*/
while (xi < width) {
const bool discard_pixel = discarder.should_discard(*user_data, uv) &&
discarder.should_discard(*user_data, uv + user_data->add_x) &&
discarder.should_discard(*user_data, uv + user_data->add_y) &&
discarder.should_discard(
*user_data, uv + user_data->add_x + user_data->add_y);
if (!discard_pixel) {
break;
}
uv += user_data->add_x;
output.increase_pixel_pointer();
xi += 1;
}
for (; xi < width; xi++) {
output.init_pixel_pointer(user_data->dst,
int2(user_data->destination_region.x_range.first(), scanline));
for (int xi : user_data->destination_region.x_range) {
UNUSED_VARS(xi);
typename Sampler::SampleType sample;
sample.clear();
int num_subsamples_added = 0;
@ -699,7 +717,11 @@ static void transform_threaded(TransformUserData *user_data, const eIMBTransform
}
if (scanline_func != nullptr) {
IMB_processor_apply_threaded_scanlines(user_data->dst->y, scanline_func, user_data);
threading::parallel_for(user_data->destination_region.y_range, 8, [&](IndexRange range) {
for (int scanline : range) {
scanline_func(user_data, scanline);
}
});
}
}
@ -727,7 +749,7 @@ void IMB_transform(const struct ImBuf *src,
if (mode == IMB_TRANSFORM_MODE_CROP_SRC) {
user_data.src_crop = *src_crop;
}
user_data.init(transform_matrix, num_subsamples);
user_data.init(transform_matrix, num_subsamples, ELEM(mode, IMB_TRANSFORM_MODE_CROP_SRC));
if (filter == IMB_FILTER_NEAREST) {
transform_threaded<IMB_FILTER_NEAREST>(&user_data, mode);