This repository has been archived on 2023-10-09. You can view files and clone it, but cannot push or open issues or pull requests.
Files
blender-archive/source/blender/compositor/operations/COM_ScaleOperation.cpp
Sergey Sharybin f81e30a41f Solved issue with distorted compositor results in some cases
Originally issue was discovered when using stabilization and movie distortion
nodes, but in fact issue was caused by render layer node always doing nearest
interpolation. Now made it so this node will respect sampler passed to it's
executePixel function and do an interpolation.

Added two new functions to do bilinear/bicubic interpolation in float buffer
with variable number of components per element, so it could interpolate 1, 3
and 4 component vectors. This functions currently mostly duplicates the same
functions from imageprocess.c and it should actually be de-duplicated. Think
it's ok to leave a bit of time with such duplication, since functions should
be generalized one more time to support byte buffers, which could backfire on
readability.

Also removed mark as complex from stabilization node, which isn't needed sine
int fact this node is not complex.
2012-11-10 19:11:25 +00:00

291 lines
9.0 KiB
C++

/*
* Copyright 2011, Blender Foundation.
*
* This program is free software; you can redistribute it and/or
* modify it under the terms of the GNU General Public License
* as published by the Free Software Foundation; either version 2
* of the License, or (at your option) any later version.
*
* This program is distributed in the hope that it will be useful,
* but WITHOUT ANY WARRANTY; without even the implied warranty of
* MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
* GNU General Public License for more details.
*
* You should have received a copy of the GNU General Public License
* along with this program; if not, write to the Free Software Foundation,
* Inc., 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301, USA.
*
* Contributor:
* Jeroen Bakker
* Monique Dewanchand
*/
#include "COM_ScaleOperation.h"
#define USE_FORCE_BILINEAR
/* XXX - ignore input and use default from old compositor,
* could become an option like the transform node - campbell
*
* note: use bilinear because bicubic makes fuzzy even when not scaling at all (1:1)
*/
BaseScaleOperation::BaseScaleOperation()
{
#ifdef USE_FORCE_BILINEAR
m_sampler = (int) COM_PS_BILINEAR;
#else
m_sampler = -1;
#endif
}
ScaleOperation::ScaleOperation() : BaseScaleOperation()
{
this->addInputSocket(COM_DT_COLOR);
this->addInputSocket(COM_DT_VALUE);
this->addInputSocket(COM_DT_VALUE);
this->addOutputSocket(COM_DT_COLOR);
this->setResolutionInputSocketIndex(0);
this->m_inputOperation = NULL;
this->m_inputXOperation = NULL;
this->m_inputYOperation = NULL;
}
void ScaleOperation::initExecution()
{
this->m_inputOperation = this->getInputSocketReader(0);
this->m_inputXOperation = this->getInputSocketReader(1);
this->m_inputYOperation = this->getInputSocketReader(2);
this->m_centerX = this->getWidth() / 2.0;
this->m_centerY = this->getHeight() / 2.0;
}
void ScaleOperation::deinitExecution()
{
this->m_inputOperation = NULL;
this->m_inputXOperation = NULL;
this->m_inputYOperation = NULL;
}
void ScaleOperation::executePixel(float output[4], float x, float y, PixelSampler sampler)
{
PixelSampler effective_sampler = getEffectiveSampler(sampler);
float scaleX[4];
float scaleY[4];
this->m_inputXOperation->read(scaleX, x, y, effective_sampler);
this->m_inputYOperation->read(scaleY, x, y, effective_sampler);
const float scx = scaleX[0];
const float scy = scaleY[0];
float nx = this->m_centerX + (x - this->m_centerX) / scx;
float ny = this->m_centerY + (y - this->m_centerY) / scy;
this->m_inputOperation->read(output, nx, ny, effective_sampler);
}
bool ScaleOperation::determineDependingAreaOfInterest(rcti *input, ReadBufferOperation *readOperation, rcti *output)
{
rcti newInput;
float scaleX[4];
float scaleY[4];
this->m_inputXOperation->read(scaleX, 0, 0, COM_PS_NEAREST);
this->m_inputYOperation->read(scaleY, 0, 0, COM_PS_NEAREST);
const float scx = scaleX[0];
const float scy = scaleY[0];
newInput.xmax = this->m_centerX + (input->xmax - this->m_centerX) / scx;
newInput.xmin = this->m_centerX + (input->xmin - this->m_centerX) / scx;
newInput.ymax = this->m_centerY + (input->ymax - this->m_centerY) / scy;
newInput.ymin = this->m_centerY + (input->ymin - this->m_centerY) / scy;
return BaseScaleOperation::determineDependingAreaOfInterest(&newInput, readOperation, output);
}
// SCALE ABSOLUTE
ScaleAbsoluteOperation::ScaleAbsoluteOperation() : BaseScaleOperation()
{
this->addInputSocket(COM_DT_COLOR);
this->addInputSocket(COM_DT_VALUE);
this->addInputSocket(COM_DT_VALUE);
this->addOutputSocket(COM_DT_COLOR);
this->setResolutionInputSocketIndex(0);
this->m_inputOperation = NULL;
this->m_inputXOperation = NULL;
this->m_inputYOperation = NULL;
}
void ScaleAbsoluteOperation::initExecution()
{
this->m_inputOperation = this->getInputSocketReader(0);
this->m_inputXOperation = this->getInputSocketReader(1);
this->m_inputYOperation = this->getInputSocketReader(2);
this->m_centerX = this->getWidth() / 2.0;
this->m_centerY = this->getHeight() / 2.0;
}
void ScaleAbsoluteOperation::deinitExecution()
{
this->m_inputOperation = NULL;
this->m_inputXOperation = NULL;
this->m_inputYOperation = NULL;
}
void ScaleAbsoluteOperation::executePixel(float output[4], float x, float y, PixelSampler sampler)
{
PixelSampler effective_sampler = getEffectiveSampler(sampler);
float scaleX[4];
float scaleY[4];
this->m_inputXOperation->read(scaleX, x, y, effective_sampler);
this->m_inputYOperation->read(scaleY, x, y, effective_sampler);
const float scx = scaleX[0]; // target absolute scale
const float scy = scaleY[0]; // target absolute scale
const float width = this->getWidth();
const float height = this->getHeight();
//div
float relativeXScale = scx / width;
float relativeYScale = scy / height;
float nx = this->m_centerX + (x - this->m_centerX) / relativeXScale;
float ny = this->m_centerY + (y - this->m_centerY) / relativeYScale;
this->m_inputOperation->read(output, nx, ny, effective_sampler);
}
bool ScaleAbsoluteOperation::determineDependingAreaOfInterest(rcti *input, ReadBufferOperation *readOperation, rcti *output)
{
rcti newInput;
float scaleX[4];
float scaleY[4];
this->m_inputXOperation->read(scaleX, 0, 0, COM_PS_NEAREST);
this->m_inputYOperation->read(scaleY, 0, 0, COM_PS_NEAREST);
const float scx = scaleX[0];
const float scy = scaleY[0];
const float width = this->getWidth();
const float height = this->getHeight();
//div
float relateveXScale = scx / width;
float relateveYScale = scy / height;
newInput.xmax = this->m_centerX + (input->xmax - this->m_centerX) / relateveXScale;
newInput.xmin = this->m_centerX + (input->xmin - this->m_centerX) / relateveXScale;
newInput.ymax = this->m_centerY + (input->ymax - this->m_centerY) / relateveYScale;
newInput.ymin = this->m_centerY + (input->ymin - this->m_centerY) / relateveYScale;
return BaseScaleOperation::determineDependingAreaOfInterest(&newInput, readOperation, output);
}
// Absolute fixed siez
ScaleFixedSizeOperation::ScaleFixedSizeOperation() : BaseScaleOperation()
{
this->addInputSocket(COM_DT_COLOR, COM_SC_NO_RESIZE);
this->addOutputSocket(COM_DT_COLOR);
this->setResolutionInputSocketIndex(0);
this->m_inputOperation = NULL;
this->m_is_offset = false;
}
void ScaleFixedSizeOperation::initExecution()
{
this->m_inputOperation = this->getInputSocketReader(0);
this->m_relX = this->m_inputOperation->getWidth() / (float)this->m_newWidth;
this->m_relY = this->m_inputOperation->getHeight() / (float)this->m_newHeight;
/* *** all the options below are for a fairly special case - camera framing *** */
if (this->m_offsetX != 0.0f || this->m_offsetY != 0.0f) {
this->m_is_offset = true;
if (this->m_newWidth > this->m_newHeight) {
this->m_offsetX *= this->m_newWidth;
this->m_offsetY *= this->m_newWidth;
}
else {
this->m_offsetX *= this->m_newHeight;
this->m_offsetY *= this->m_newHeight;
}
}
if (this->m_is_aspect) {
/* apply aspect from clip */
const float w_src = this->m_inputOperation->getWidth();
const float h_src = this->m_inputOperation->getHeight();
/* destination aspect is already applied from the camera frame */
const float w_dst = this->m_newWidth;
const float h_dst = this->m_newHeight;
const float asp_src = w_src / h_src;
const float asp_dst = w_dst / h_dst;
if (fabsf(asp_src - asp_dst) >= FLT_EPSILON) {
if ((asp_src > asp_dst) == (this->m_is_crop == true)) {
/* fit X */
const float div = asp_src / asp_dst;
this->m_relX /= div;
this->m_offsetX += ((w_src - (w_src * div)) / (w_src / w_dst)) / 2.0f;
}
else {
/* fit Y */
const float div = asp_dst / asp_src;
this->m_relY /= div;
this->m_offsetY += ((h_src - (h_src * div)) / (h_src / h_dst)) / 2.0f;
}
this->m_is_offset = true;
}
}
/* *** end framing options *** */
}
void ScaleFixedSizeOperation::deinitExecution()
{
this->m_inputOperation = NULL;
}
void ScaleFixedSizeOperation::executePixel(float output[4], float x, float y, PixelSampler sampler)
{
PixelSampler effective_sampler = getEffectiveSampler(sampler);
if (this->m_is_offset) {
float nx = ((x - this->m_offsetX) * this->m_relX);
float ny = ((y - this->m_offsetY) * this->m_relY);
this->m_inputOperation->read(output, nx, ny, effective_sampler);
}
else {
this->m_inputOperation->read(output, x * this->m_relX, y * this->m_relY, effective_sampler);
}
}
bool ScaleFixedSizeOperation::determineDependingAreaOfInterest(rcti *input, ReadBufferOperation *readOperation, rcti *output)
{
rcti newInput;
newInput.xmax = input->xmax * this->m_relX;
newInput.xmin = input->xmin * this->m_relX;
newInput.ymax = input->ymax * this->m_relY;
newInput.ymin = input->ymin * this->m_relY;
return BaseScaleOperation::determineDependingAreaOfInterest(&newInput, readOperation, output);
}
void ScaleFixedSizeOperation::determineResolution(unsigned int resolution[2], unsigned int preferredResolution[2])
{
unsigned int nr[2];
nr[0] = this->m_newWidth;
nr[1] = this->m_newHeight;
BaseScaleOperation::determineResolution(resolution, nr);
resolution[0] = this->m_newWidth;
resolution[1] = this->m_newHeight;
}